id
stringlengths 13
16
| title
stringclasses 225
values | content
stringlengths 1.09k
2.04k
| contents
stringlengths 1.12k
2.05k
| ACRID
stringclasses 225
values |
---|---|---|---|---|
acrac_69402_7 | Suspected Acute Aortic Syndrome | In contrast to MRA, MRI does not contain sequences designed to specifically produce images of the aorta, such as gating, 3-D data sets, thin sections, or double-oblique planes. MRI contains sequences that produce images in orthogonal planes, with greater slice thickness. MRI Chest, Abdomen, and Pelvis Without and With IV Contrast There is no relevant literature supporting the use of MRI chest, abdomen, and pelvis without and with IV contrast for the diagnosis of AAS. In contrast to MRA, MRI does not contain sequences designed to specifically produce images of the aorta, such as gating, 3-D data sets, thin sections, or double-oblique planes. MRI contains sequences that produce images in orthogonal planes, with greater slice thickness. Radiography Chest Chest radiographs can identify an enlarged aorta and or mediastinum and can identify complications such as pleural fluid. A substantial portion of patients with AAS may have a normal chest radiograph, and further imaging should be pursued despite a normal chest radiograph in cases of suspected AAS. In most cases when patients with an AAS have an abnormal radiograph, the findings are nonspecific [41]. The most recent report from the International Registry of Acute Aortic Dissection showed that in the last 10 years, the incidence of abnormal radiographic findings has decreased in patients with AD [10]. In that report, radiographs were reported as abnormal in 52% of patients with type A AD, whereas previously it was 61%. Abnormal findings were present in 39% of patients with type B AD, whereas previously it was 56%. Radiographs were reported to be completely normal in 36% to 38% of patients with AD. In another study that measured the diagnostic performance of radiography for all diagnoses of AAS, including IMH, PAU, and ruptured aneurysm, chest radiography had a sensitivity of 70.8% and a specificity of 82.5% [42]. | Suspected Acute Aortic Syndrome. In contrast to MRA, MRI does not contain sequences designed to specifically produce images of the aorta, such as gating, 3-D data sets, thin sections, or double-oblique planes. MRI contains sequences that produce images in orthogonal planes, with greater slice thickness. MRI Chest, Abdomen, and Pelvis Without and With IV Contrast There is no relevant literature supporting the use of MRI chest, abdomen, and pelvis without and with IV contrast for the diagnosis of AAS. In contrast to MRA, MRI does not contain sequences designed to specifically produce images of the aorta, such as gating, 3-D data sets, thin sections, or double-oblique planes. MRI contains sequences that produce images in orthogonal planes, with greater slice thickness. Radiography Chest Chest radiographs can identify an enlarged aorta and or mediastinum and can identify complications such as pleural fluid. A substantial portion of patients with AAS may have a normal chest radiograph, and further imaging should be pursued despite a normal chest radiograph in cases of suspected AAS. In most cases when patients with an AAS have an abnormal radiograph, the findings are nonspecific [41]. The most recent report from the International Registry of Acute Aortic Dissection showed that in the last 10 years, the incidence of abnormal radiographic findings has decreased in patients with AD [10]. In that report, radiographs were reported as abnormal in 52% of patients with type A AD, whereas previously it was 61%. Abnormal findings were present in 39% of patients with type B AD, whereas previously it was 56%. Radiographs were reported to be completely normal in 36% to 38% of patients with AD. In another study that measured the diagnostic performance of radiography for all diagnoses of AAS, including IMH, PAU, and ruptured aneurysm, chest radiography had a sensitivity of 70.8% and a specificity of 82.5% [42]. | 69402 |
acrac_69411_0 | Lower Extremity Arterial Claudication Imaging Assessment for Revascularization | Estimates of the prevalence of claudication in the general population range from 1.6% to almost 8%, depending on age, sex, the geographic location of the population, and the diagnostic criteria used [2,3]. In most studies, fewer than 10% of patients with intermittent claudication progress to chronic limb-threatening ischemia in 5 years [4,5]. However, one large meta-analysis of 16,440 patients demonstrated that 21% of patients with intermittent claudication progressed to chronic limb-threatening ischemia [6]. The presence of vascular disease in patients with symptoms of claudication is reliably established by a variety of noninvasive hemodynamic tests. In the absence of demonstrable arterial disease, imaging studies of other systems, such as the lumbar spine or soft tissues of the pelvis, may be indicated. If peripheral vascular disease is confirmed, additional studies may be indicated to screen the heart and carotid arteries for involvement [2]. Initial Imaging Definition Initial imaging is defined as imaging at the beginning of the care episode for the medical condition defined by the variant. More than one procedure can be considered usually appropriate in the initial imaging evaluation when: The American College of Radiology seeks and encourages collaboration with other organizations on the development of the ACR Appropriateness Criteria through representation of such organizations on expert panels. Participation on the expert panel does not necessarily imply endorsement of the final document by individual contributors or their respective organization. Reprint requests to: [email protected] OR Special Imaging Considerations CT angiography (CTA) in the tibial arteries is limited by the difficulty in accurately timing the image acquisition with respect to arrival of the iodine bolus: images acquired too late will have problematic venous contamination; images acquired too early will not have adequate contrast enhancement. | Lower Extremity Arterial Claudication Imaging Assessment for Revascularization. Estimates of the prevalence of claudication in the general population range from 1.6% to almost 8%, depending on age, sex, the geographic location of the population, and the diagnostic criteria used [2,3]. In most studies, fewer than 10% of patients with intermittent claudication progress to chronic limb-threatening ischemia in 5 years [4,5]. However, one large meta-analysis of 16,440 patients demonstrated that 21% of patients with intermittent claudication progressed to chronic limb-threatening ischemia [6]. The presence of vascular disease in patients with symptoms of claudication is reliably established by a variety of noninvasive hemodynamic tests. In the absence of demonstrable arterial disease, imaging studies of other systems, such as the lumbar spine or soft tissues of the pelvis, may be indicated. If peripheral vascular disease is confirmed, additional studies may be indicated to screen the heart and carotid arteries for involvement [2]. Initial Imaging Definition Initial imaging is defined as imaging at the beginning of the care episode for the medical condition defined by the variant. More than one procedure can be considered usually appropriate in the initial imaging evaluation when: The American College of Radiology seeks and encourages collaboration with other organizations on the development of the ACR Appropriateness Criteria through representation of such organizations on expert panels. Participation on the expert panel does not necessarily imply endorsement of the final document by individual contributors or their respective organization. Reprint requests to: [email protected] OR Special Imaging Considerations CT angiography (CTA) in the tibial arteries is limited by the difficulty in accurately timing the image acquisition with respect to arrival of the iodine bolus: images acquired too late will have problematic venous contamination; images acquired too early will not have adequate contrast enhancement. | 69411 |
acrac_69411_1 | Lower Extremity Arterial Claudication Imaging Assessment for Revascularization | More common with newer CT technologies is imaging too early either for one leg with slow flow from outflow disease or for both legs secondary to the very fast scanning protocols. Delayed images of the calves can be obtained to catch the bolus in these patients. The presence of diffusely diseased arteries can present challenges during angiography, because stenosis severity can be difficult to determine in the absence of normal arterial segments for comparison. In addition, serial lesions, luminal irregularity, and the degree of collateral development may produce effects on the blood flow that are difficult to quantify angiographically. The main drawbacks of arteriography in patients with claudication are its invasive nature and the known complications from catheterization [2,10]. These difficulties can be avoided by using examinations such as duplex ultrasound (US), MR angiography (MRA), or CTA to accurately triage patients with confirmed PAD for percutaneous or surgical treatments. For the latter, preoperative arteriography may not be needed. Finally, arteriography has inconsistent correlation between the hemodynamic or functional effects and the morphology of the arterial lesions [12]. Several studies have reported this problem, but in some of them the problem may be accentuated by less-than-optimal angiographic technique (eg, single-projection, nonselective injections). CTA Abdomen and Pelvis with Bilateral Lower Extremity Runoff With IV Contrast CTA is commonly used for imaging peripheral vascular disease. Multidetector CT scanners, including helical and multistation axial acquisitions, enable rapid scanning of the entire arterial system [13]. When compared with catheter arteriography, CTA offers volumetric as opposed to planar images. The volumetric acquisition enables extensive image postprocessing, including multiplanar reformatted and maximum-intensity projection images to create an arterial road map [13]. | Lower Extremity Arterial Claudication Imaging Assessment for Revascularization. More common with newer CT technologies is imaging too early either for one leg with slow flow from outflow disease or for both legs secondary to the very fast scanning protocols. Delayed images of the calves can be obtained to catch the bolus in these patients. The presence of diffusely diseased arteries can present challenges during angiography, because stenosis severity can be difficult to determine in the absence of normal arterial segments for comparison. In addition, serial lesions, luminal irregularity, and the degree of collateral development may produce effects on the blood flow that are difficult to quantify angiographically. The main drawbacks of arteriography in patients with claudication are its invasive nature and the known complications from catheterization [2,10]. These difficulties can be avoided by using examinations such as duplex ultrasound (US), MR angiography (MRA), or CTA to accurately triage patients with confirmed PAD for percutaneous or surgical treatments. For the latter, preoperative arteriography may not be needed. Finally, arteriography has inconsistent correlation between the hemodynamic or functional effects and the morphology of the arterial lesions [12]. Several studies have reported this problem, but in some of them the problem may be accentuated by less-than-optimal angiographic technique (eg, single-projection, nonselective injections). CTA Abdomen and Pelvis with Bilateral Lower Extremity Runoff With IV Contrast CTA is commonly used for imaging peripheral vascular disease. Multidetector CT scanners, including helical and multistation axial acquisitions, enable rapid scanning of the entire arterial system [13]. When compared with catheter arteriography, CTA offers volumetric as opposed to planar images. The volumetric acquisition enables extensive image postprocessing, including multiplanar reformatted and maximum-intensity projection images to create an arterial road map [13]. | 69411 |
acrac_69411_2 | Lower Extremity Arterial Claudication Imaging Assessment for Revascularization | With optimized timing of the acquisition, CT images include collaterals and arteries distal to occlusions that may not appear on catheter angiography images. Like MRA, CTA is a cross-sectional technique, which shows nonvascular findings, as well as vascular lesions associated with aneurysms and cystic adventitial disease that are not detected with the projectional technique of catheter arteriography. Unlike imaging of the peripheral arteries, CTA is considered to have replaced catheter angiography as the reference standard for imaging of the aorta [13,14]. CTA can readily detect stenosis caused by plaque or thrombus in the aorta and iliac arteries that may be contributing to symptoms of claudication. CTA alone can be used to plan treatment, including assessment of the length, severity, and number of stenoses [15,16]. Compared with catheter angiography, the sensitivity and specificity of CTA for detection of stenoses >50% diameter are 90% to 100% [10,17-20]. Accuracy in patients with bypass grafts is excellent compared with duplex Lower Extremity Arterial Claudication US [21]. CTA is also more clinically useful than duplex US [21]. However, heavily calcified atheromatous disease can limit the ability to interpret CT images. This drawback is usually more pronounced in tibial arteries. Identification of patients who may be unsuitable candidates for CTA of the tibial arteries (eg, >80 years of age, diabetic, on dialysis) will reduce the number of nondiagnostic studies [22]. Dual-energy CTA can reduce blooming and beam-hardening artifact created by heavily calcified atheromatous disease and metallic stents [23]. Compared with MRA, CT has the advantages of more rapid acquisition, better safety in patients with pacemakers or defibrillators, and generally less severe artifacts from metal. Finally, claustrophobia is far less of a problem. | Lower Extremity Arterial Claudication Imaging Assessment for Revascularization. With optimized timing of the acquisition, CT images include collaterals and arteries distal to occlusions that may not appear on catheter angiography images. Like MRA, CTA is a cross-sectional technique, which shows nonvascular findings, as well as vascular lesions associated with aneurysms and cystic adventitial disease that are not detected with the projectional technique of catheter arteriography. Unlike imaging of the peripheral arteries, CTA is considered to have replaced catheter angiography as the reference standard for imaging of the aorta [13,14]. CTA can readily detect stenosis caused by plaque or thrombus in the aorta and iliac arteries that may be contributing to symptoms of claudication. CTA alone can be used to plan treatment, including assessment of the length, severity, and number of stenoses [15,16]. Compared with catheter angiography, the sensitivity and specificity of CTA for detection of stenoses >50% diameter are 90% to 100% [10,17-20]. Accuracy in patients with bypass grafts is excellent compared with duplex Lower Extremity Arterial Claudication US [21]. CTA is also more clinically useful than duplex US [21]. However, heavily calcified atheromatous disease can limit the ability to interpret CT images. This drawback is usually more pronounced in tibial arteries. Identification of patients who may be unsuitable candidates for CTA of the tibial arteries (eg, >80 years of age, diabetic, on dialysis) will reduce the number of nondiagnostic studies [22]. Dual-energy CTA can reduce blooming and beam-hardening artifact created by heavily calcified atheromatous disease and metallic stents [23]. Compared with MRA, CT has the advantages of more rapid acquisition, better safety in patients with pacemakers or defibrillators, and generally less severe artifacts from metal. Finally, claustrophobia is far less of a problem. | 69411 |
acrac_69411_3 | Lower Extremity Arterial Claudication Imaging Assessment for Revascularization | CTA Abdomen and Pelvis with Bilateral Lower Extremity Runoff Without and With IV Contrast CTA is commonly used for imaging peripheral vascular disease. Multidetector CT scanners, including helical and multistation axial acquisitions, enable rapid scanning of the entire arterial system [13]. When compared with catheter arteriography, CTA offers volumetric as opposed to planar images. The volumetric acquisition enables extensive image postprocessing, including multiplanar reformatted and maximum-intensity projection images to create an arterial road map [13]. With optimized timing of the acquisition, CT images include collaterals and arteries distal to occlusions that may not appear on catheter angiography images. Like MRA, CTA is a cross-sectional technique, which shows nonvascular findings as well as vascular lesions associated with aneurysms and cystic adventitial disease that are not detected with the projectional technique of catheter arteriography. Unlike imaging of the peripheral arteries, CTA is considered to have replaced catheter angiography as the reference standard for imaging of the aorta [13,14]. CTA can readily detect stenosis caused by plaque or thrombus in the aorta and iliac arteries that may be contributing to symptoms of claudication. CTA alone can be used to plan treatment, including assessment of the length, severity, and number of stenoses [15,16]. Compared with catheter angiography, the sensitivity and specificity of CTA for detection of stenoses >50% diameter are 90% to 100% [10,17-20]. Accuracy in patients with bypass grafts is excellent compared with duplex US [21]. CTA is also more clinically useful than duplex US [21]. However, heavily calcified atheromatous disease can limit the ability to interpret CT images. This drawback is usually more pronounced in tibial arteries. | Lower Extremity Arterial Claudication Imaging Assessment for Revascularization. CTA Abdomen and Pelvis with Bilateral Lower Extremity Runoff Without and With IV Contrast CTA is commonly used for imaging peripheral vascular disease. Multidetector CT scanners, including helical and multistation axial acquisitions, enable rapid scanning of the entire arterial system [13]. When compared with catheter arteriography, CTA offers volumetric as opposed to planar images. The volumetric acquisition enables extensive image postprocessing, including multiplanar reformatted and maximum-intensity projection images to create an arterial road map [13]. With optimized timing of the acquisition, CT images include collaterals and arteries distal to occlusions that may not appear on catheter angiography images. Like MRA, CTA is a cross-sectional technique, which shows nonvascular findings as well as vascular lesions associated with aneurysms and cystic adventitial disease that are not detected with the projectional technique of catheter arteriography. Unlike imaging of the peripheral arteries, CTA is considered to have replaced catheter angiography as the reference standard for imaging of the aorta [13,14]. CTA can readily detect stenosis caused by plaque or thrombus in the aorta and iliac arteries that may be contributing to symptoms of claudication. CTA alone can be used to plan treatment, including assessment of the length, severity, and number of stenoses [15,16]. Compared with catheter angiography, the sensitivity and specificity of CTA for detection of stenoses >50% diameter are 90% to 100% [10,17-20]. Accuracy in patients with bypass grafts is excellent compared with duplex US [21]. CTA is also more clinically useful than duplex US [21]. However, heavily calcified atheromatous disease can limit the ability to interpret CT images. This drawback is usually more pronounced in tibial arteries. | 69411 |
acrac_69411_4 | Lower Extremity Arterial Claudication Imaging Assessment for Revascularization | Identification of patients who may be unsuitable candidates for CTA of the tibial arteries (eg, >80 years of age, diabetic, on dialysis) will reduce the number of nondiagnostic studies [22]. Dual-energy CTA can reduce blooming and beam-hardening artifact created by heavily calcified atheromatous disease and metallic stents [23]. CTA without intravenous (IV) contrast can be performed before CTA with IV contrast to fine-tune the scan range and identify calcified plaque and calcium within thrombus [24]. Compared with MRA, CT has the advantages of more rapid acquisition, better safety in patients with pacemakers or defibrillators, and generally less severe artifacts from metal. Finally, claustrophobia is far less of a problem. CTA Lower Extremity With IV Contrast CTA is commonly used for imaging peripheral vascular disease. Multidetector CT scanners, including helical and multistation axial acquisitions, enable rapid scanning of the entire arterial system [13]. When compared with catheter arteriography, CTA offers volumetric as opposed to planar images. The volumetric acquisition enables extensive image postprocessing, including multiplanar reformatted and maximum-intensity projection images to create an arterial road map [13]. With optimized timing of the acquisition, CT images include collaterals and arteries distal to occlusions that may not appear on catheter angiography images. Like MRA, CTA is a cross-sectional technique, which shows nonvascular findings, as well as vascular lesions associated with aneurysms and cystic adventitial disease that are not detected with the projectional technique of catheter arteriography. CTA alone can be used to plan treatment, including assessment of the length, severity, and number of stenoses [15,16]. Compared with catheter angiography, the sensitivity and specificity of CTA for detection of stenoses >50% diameter are 90% to 100% [10,17-20]. | Lower Extremity Arterial Claudication Imaging Assessment for Revascularization. Identification of patients who may be unsuitable candidates for CTA of the tibial arteries (eg, >80 years of age, diabetic, on dialysis) will reduce the number of nondiagnostic studies [22]. Dual-energy CTA can reduce blooming and beam-hardening artifact created by heavily calcified atheromatous disease and metallic stents [23]. CTA without intravenous (IV) contrast can be performed before CTA with IV contrast to fine-tune the scan range and identify calcified plaque and calcium within thrombus [24]. Compared with MRA, CT has the advantages of more rapid acquisition, better safety in patients with pacemakers or defibrillators, and generally less severe artifacts from metal. Finally, claustrophobia is far less of a problem. CTA Lower Extremity With IV Contrast CTA is commonly used for imaging peripheral vascular disease. Multidetector CT scanners, including helical and multistation axial acquisitions, enable rapid scanning of the entire arterial system [13]. When compared with catheter arteriography, CTA offers volumetric as opposed to planar images. The volumetric acquisition enables extensive image postprocessing, including multiplanar reformatted and maximum-intensity projection images to create an arterial road map [13]. With optimized timing of the acquisition, CT images include collaterals and arteries distal to occlusions that may not appear on catheter angiography images. Like MRA, CTA is a cross-sectional technique, which shows nonvascular findings, as well as vascular lesions associated with aneurysms and cystic adventitial disease that are not detected with the projectional technique of catheter arteriography. CTA alone can be used to plan treatment, including assessment of the length, severity, and number of stenoses [15,16]. Compared with catheter angiography, the sensitivity and specificity of CTA for detection of stenoses >50% diameter are 90% to 100% [10,17-20]. | 69411 |
acrac_69411_5 | Lower Extremity Arterial Claudication Imaging Assessment for Revascularization | Accuracy in patients with bypass grafts is excellent compared with duplex US [21]. CTA is also more clinically useful than duplex US [21]. However, heavily calcified atheromatous disease can limit the ability to interpret CT images. This drawback is usually more pronounced in tibial arteries. Identification of patients who may be unsuitable candidates for CTA of the tibial arteries (eg, >80 years of age, diabetic, on dialysis) will reduce the number of nondiagnostic studies [22]. Dual-energy CTA can reduce blooming and beam-hardening artifact created by heavily calcified atheromatous disease and metallic stents [23]. Compared with MRA, CT has the advantages of more rapid acquisition, better safety in patients with pacemakers or defibrillators, and generally less severe artifacts from metal. Finally, claustrophobia is far less of a problem. Lower Extremity Arterial Claudication CTA of one or both lower extremities can be performed without imaging the abdomen and pelvis when aortoiliac disease is not a concern or the state of the aorta and iliac arteries is already known. CTA Lower Extremity Without and With IV Contrast CTA is commonly used for imaging peripheral vascular disease. Multidetector CT scanners, including helical and multistation axial acquisitions, enable rapid scanning of the entire arterial system [13]. When compared with catheter arteriography, CTA offers volumetric as opposed to planar images. The volumetric acquisition enables extensive image postprocessing, including multiplanar reformatted and maximum-intensity projection images to create an arterial road map [13]. With optimized timing of the acquisition, CT images include collaterals and arteries distal to occlusions that may not appear on catheter angiography images. | Lower Extremity Arterial Claudication Imaging Assessment for Revascularization. Accuracy in patients with bypass grafts is excellent compared with duplex US [21]. CTA is also more clinically useful than duplex US [21]. However, heavily calcified atheromatous disease can limit the ability to interpret CT images. This drawback is usually more pronounced in tibial arteries. Identification of patients who may be unsuitable candidates for CTA of the tibial arteries (eg, >80 years of age, diabetic, on dialysis) will reduce the number of nondiagnostic studies [22]. Dual-energy CTA can reduce blooming and beam-hardening artifact created by heavily calcified atheromatous disease and metallic stents [23]. Compared with MRA, CT has the advantages of more rapid acquisition, better safety in patients with pacemakers or defibrillators, and generally less severe artifacts from metal. Finally, claustrophobia is far less of a problem. Lower Extremity Arterial Claudication CTA of one or both lower extremities can be performed without imaging the abdomen and pelvis when aortoiliac disease is not a concern or the state of the aorta and iliac arteries is already known. CTA Lower Extremity Without and With IV Contrast CTA is commonly used for imaging peripheral vascular disease. Multidetector CT scanners, including helical and multistation axial acquisitions, enable rapid scanning of the entire arterial system [13]. When compared with catheter arteriography, CTA offers volumetric as opposed to planar images. The volumetric acquisition enables extensive image postprocessing, including multiplanar reformatted and maximum-intensity projection images to create an arterial road map [13]. With optimized timing of the acquisition, CT images include collaterals and arteries distal to occlusions that may not appear on catheter angiography images. | 69411 |
acrac_69411_6 | Lower Extremity Arterial Claudication Imaging Assessment for Revascularization | Like MRA, CTA is a cross-sectional technique, which shows nonvascular findings as well as vascular lesions associated with aneurysms and cystic adventitial disease that are not detected with the projectional technique of catheter arteriography. CTA alone can be used to plan treatment, including assessment of the length, severity, and number of stenoses [15,16]. Compared with catheter angiography, the sensitivity and specificity of CTA for detection of stenoses >50% diameter are 90% to 100% [10,17-20]. Accuracy in patients with bypass grafts is excellent compared with duplex US [21]. CTA is also more clinically useful than duplex US [21]. However, heavily calcified atheromatous disease can limit the ability to interpret CT images. This drawback is usually more pronounced in tibial arteries. Identification of patients who may be unsuitable candidates for CTA of the tibial arteries (eg, >80 years of age, diabetic, on dialysis) will reduce the number of nondiagnostic studies [22]. Dual-energy CTA can reduce blooming and beam-hardening artifact created by heavily calcified atheromatous disease and metallic stents [23]. CTA without IV contrast can be performed before CTA with IV contrast to fine-tune the scan range and identify calcified plaque and calcium within thrombus [24]. Compared with MRA, CT has the advantages of more rapid acquisition, better safety in patients with pacemakers or defibrillators, and generally less severe artifacts from metal. Finally, claustrophobia is far less of a problem. CTA of one or both lower extremities can be performed without imaging the abdomen and pelvis when aortoiliac disease is not a concern or the state of the aorta and iliac arteries is already known. MRA Abdomen and Pelvis with Bilateral Lower Extremity Runoff With IV Contrast Contrast-enhanced MRA (CE-MRA) techniques continue to evolve and improve. | Lower Extremity Arterial Claudication Imaging Assessment for Revascularization. Like MRA, CTA is a cross-sectional technique, which shows nonvascular findings as well as vascular lesions associated with aneurysms and cystic adventitial disease that are not detected with the projectional technique of catheter arteriography. CTA alone can be used to plan treatment, including assessment of the length, severity, and number of stenoses [15,16]. Compared with catheter angiography, the sensitivity and specificity of CTA for detection of stenoses >50% diameter are 90% to 100% [10,17-20]. Accuracy in patients with bypass grafts is excellent compared with duplex US [21]. CTA is also more clinically useful than duplex US [21]. However, heavily calcified atheromatous disease can limit the ability to interpret CT images. This drawback is usually more pronounced in tibial arteries. Identification of patients who may be unsuitable candidates for CTA of the tibial arteries (eg, >80 years of age, diabetic, on dialysis) will reduce the number of nondiagnostic studies [22]. Dual-energy CTA can reduce blooming and beam-hardening artifact created by heavily calcified atheromatous disease and metallic stents [23]. CTA without IV contrast can be performed before CTA with IV contrast to fine-tune the scan range and identify calcified plaque and calcium within thrombus [24]. Compared with MRA, CT has the advantages of more rapid acquisition, better safety in patients with pacemakers or defibrillators, and generally less severe artifacts from metal. Finally, claustrophobia is far less of a problem. CTA of one or both lower extremities can be performed without imaging the abdomen and pelvis when aortoiliac disease is not a concern or the state of the aorta and iliac arteries is already known. MRA Abdomen and Pelvis with Bilateral Lower Extremity Runoff With IV Contrast Contrast-enhanced MRA (CE-MRA) techniques continue to evolve and improve. | 69411 |
acrac_69411_7 | Lower Extremity Arterial Claudication Imaging Assessment for Revascularization | Three-dimensional imaging, contrast enhancement with gadolinium, subtraction, cardiac gating, bolus chase, parallel imaging, optimized K- space filling, 3T magnet strength, and improved coil technology have led to improved temporal resolution, spatial resolution, and signal-to-noise ratio in CE-MRA. Its sensitivity and specificity for detection of stenoses >50% are now in the 90% to 100% range [25-27]. Unlike with CTA, the presence of calcium in small vessels does not result in a CE-MRA artifact [28]. Furthermore, dedicated time-resolved CE-MRA of the tibial and pedal arteries significantly increases diagnostic accuracy of tibial and pedal lesions compared with standard multistation CE- MRA. Time-resolved CE-MRA also reduces insufficient arterial filling and venous contamination, which are common limitations of standard multistation CE-MRA [28]. Although CE-MRA has not supplanted angiography as a reference standard, one small study demonstrated that 3T CE-MRA with calf compression (to prevent venous contamination) resulted in better visualization of tibial arteries than DSA [29]. For these reasons, CE-MRA is ideally suited for patients at high risk for calcification of the tibial and pedal arteries, particularly patients with diabetes and patients >80 years of age [28,30]. In comparison with duplex US, CE-MRA is more accurate for detecting and quantifying significant stenoses and for preoperative planning [31]. In a randomized controlled trial comparison with duplex US, CE-MRA for the initial imaging workup of patients with PAD reduced the need for additional imaging [32]. In a meta-analysis comparison with CTA, CE-MRA had equivalent sensitivity and specificity for detecting arterial lesions from the aorta to the tibial arteries in patients with intermittent claudication [26]. Some technical problems limit the utility of CE-MRA for imaging PAD. | Lower Extremity Arterial Claudication Imaging Assessment for Revascularization. Three-dimensional imaging, contrast enhancement with gadolinium, subtraction, cardiac gating, bolus chase, parallel imaging, optimized K- space filling, 3T magnet strength, and improved coil technology have led to improved temporal resolution, spatial resolution, and signal-to-noise ratio in CE-MRA. Its sensitivity and specificity for detection of stenoses >50% are now in the 90% to 100% range [25-27]. Unlike with CTA, the presence of calcium in small vessels does not result in a CE-MRA artifact [28]. Furthermore, dedicated time-resolved CE-MRA of the tibial and pedal arteries significantly increases diagnostic accuracy of tibial and pedal lesions compared with standard multistation CE- MRA. Time-resolved CE-MRA also reduces insufficient arterial filling and venous contamination, which are common limitations of standard multistation CE-MRA [28]. Although CE-MRA has not supplanted angiography as a reference standard, one small study demonstrated that 3T CE-MRA with calf compression (to prevent venous contamination) resulted in better visualization of tibial arteries than DSA [29]. For these reasons, CE-MRA is ideally suited for patients at high risk for calcification of the tibial and pedal arteries, particularly patients with diabetes and patients >80 years of age [28,30]. In comparison with duplex US, CE-MRA is more accurate for detecting and quantifying significant stenoses and for preoperative planning [31]. In a randomized controlled trial comparison with duplex US, CE-MRA for the initial imaging workup of patients with PAD reduced the need for additional imaging [32]. In a meta-analysis comparison with CTA, CE-MRA had equivalent sensitivity and specificity for detecting arterial lesions from the aorta to the tibial arteries in patients with intermittent claudication [26]. Some technical problems limit the utility of CE-MRA for imaging PAD. | 69411 |
acrac_69411_8 | Lower Extremity Arterial Claudication Imaging Assessment for Revascularization | Challenges may include image quality related to low signal-to-noise ratio, limited spatial resolution, motion artifacts, long acquisition times, and loss of signal in arterial segments within metal stents or adjacent to metallic clips or prosthetic joints. Some of these problems have been addressed successfully with the use of newer imaging sequences and newer stent designs. MRA Abdomen and Pelvis with Bilateral Lower Extremity Runoff Without IV Contrast Noncontrast MRA techniques predate CE-MRA in their development. However, long acquisition time and motion artifacts limit the use of these techniques in the abdomen and peripheral arteries [33]. To that end, recent Lower Extremity Arterial Claudication advancements in noncontrast MRA techniques for imaging PAD have expanded the sequence options from time- of-flight and phase-contrast imaging to include electrocardiogram-gated fresh-blood partial Fourier fast spin echo, balanced steady-state free precession, and arterial spin labeling [33]. Two alternative approaches using balanced steady state for peripheral noncontrast MRA applications include flow-sensitive dephasing and quiescent-interval single shot [30,33-35]. When compared with bolus-chase and time-resolved gadolinium-enhanced MRA, initial studies of fresh-blood imaging of the tibial and pedal arteries have provided accurate imaging when technically successful. Overall, these methods are being increasingly adopted for patients with severe renal insufficiency at risk of developing nephrogenic systemic fibrosis. Some technical problems limit the utility of noncontrast MRA for imaging PAD. | Lower Extremity Arterial Claudication Imaging Assessment for Revascularization. Challenges may include image quality related to low signal-to-noise ratio, limited spatial resolution, motion artifacts, long acquisition times, and loss of signal in arterial segments within metal stents or adjacent to metallic clips or prosthetic joints. Some of these problems have been addressed successfully with the use of newer imaging sequences and newer stent designs. MRA Abdomen and Pelvis with Bilateral Lower Extremity Runoff Without IV Contrast Noncontrast MRA techniques predate CE-MRA in their development. However, long acquisition time and motion artifacts limit the use of these techniques in the abdomen and peripheral arteries [33]. To that end, recent Lower Extremity Arterial Claudication advancements in noncontrast MRA techniques for imaging PAD have expanded the sequence options from time- of-flight and phase-contrast imaging to include electrocardiogram-gated fresh-blood partial Fourier fast spin echo, balanced steady-state free precession, and arterial spin labeling [33]. Two alternative approaches using balanced steady state for peripheral noncontrast MRA applications include flow-sensitive dephasing and quiescent-interval single shot [30,33-35]. When compared with bolus-chase and time-resolved gadolinium-enhanced MRA, initial studies of fresh-blood imaging of the tibial and pedal arteries have provided accurate imaging when technically successful. Overall, these methods are being increasingly adopted for patients with severe renal insufficiency at risk of developing nephrogenic systemic fibrosis. Some technical problems limit the utility of noncontrast MRA for imaging PAD. | 69411 |
acrac_69411_9 | Lower Extremity Arterial Claudication Imaging Assessment for Revascularization | Challenges may include image quality related to low signal-to-noise ratio, limited spatial resolution, motion artifacts, long acquisition times, unreliable visualization of lesions with high flow and turbulence (excessive signal loss at regions of high-grade stenoses), nonvisualization of patent vessel segments with reversed blood flow, and loss of signal in arterial segments within metal stents or adjacent to metallic clips or prosthetic joints. Some of these problems have been addressed successfully with the use of newer imaging sequences. With the newer noncontrast techniques, cardiac arrhythmia can impair image quality, limiting evaluation of the distal tibial and pedal arteries. Although useful tools to improve image quality have been suggested, larger-scale trials are required for evaluation of small-vessel PAD with noncontrast MRA [29,36]. MRA Lower Extremity Without and With IV Contrast MRA of one or both lower extremities can be performed without imaging the abdomen and pelvis when aortoiliac disease is not a concern or the state of the aorta and iliac arteries is already known. CE-MRA techniques continue to evolve and improve. Three-dimensional imaging, contrast enhancement with gadolinium, subtraction, cardiac gating, bolus chase, parallel imaging, optimized K-space filling, 3T magnet strength, and improved coil technology have led to improved temporal resolution, spatial resolution, and signal-to- noise ratio in CE-MRA. Its sensitivity and specificity for detection of stenoses >50% are now in the 90% to 100% range [25-27]. Unlike with CTA, the presence of calcium in small vessels does not result in a CE-MRA artifact [28]. Furthermore, dedicated time-resolved CE-MRA of the tibial and pedal arteries significantly increases diagnostic accuracy of tibial and pedal lesions compared with standard multistation CE-MRA. | Lower Extremity Arterial Claudication Imaging Assessment for Revascularization. Challenges may include image quality related to low signal-to-noise ratio, limited spatial resolution, motion artifacts, long acquisition times, unreliable visualization of lesions with high flow and turbulence (excessive signal loss at regions of high-grade stenoses), nonvisualization of patent vessel segments with reversed blood flow, and loss of signal in arterial segments within metal stents or adjacent to metallic clips or prosthetic joints. Some of these problems have been addressed successfully with the use of newer imaging sequences. With the newer noncontrast techniques, cardiac arrhythmia can impair image quality, limiting evaluation of the distal tibial and pedal arteries. Although useful tools to improve image quality have been suggested, larger-scale trials are required for evaluation of small-vessel PAD with noncontrast MRA [29,36]. MRA Lower Extremity Without and With IV Contrast MRA of one or both lower extremities can be performed without imaging the abdomen and pelvis when aortoiliac disease is not a concern or the state of the aorta and iliac arteries is already known. CE-MRA techniques continue to evolve and improve. Three-dimensional imaging, contrast enhancement with gadolinium, subtraction, cardiac gating, bolus chase, parallel imaging, optimized K-space filling, 3T magnet strength, and improved coil technology have led to improved temporal resolution, spatial resolution, and signal-to- noise ratio in CE-MRA. Its sensitivity and specificity for detection of stenoses >50% are now in the 90% to 100% range [25-27]. Unlike with CTA, the presence of calcium in small vessels does not result in a CE-MRA artifact [28]. Furthermore, dedicated time-resolved CE-MRA of the tibial and pedal arteries significantly increases diagnostic accuracy of tibial and pedal lesions compared with standard multistation CE-MRA. | 69411 |
acrac_69411_10 | Lower Extremity Arterial Claudication Imaging Assessment for Revascularization | Time-resolved CE- MRA also reduces insufficient arterial filling and venous contamination, which are common limitations of standard multistation CE-MRA [28]. Although CE-MRA has not supplanted angiography as a reference standard, one small study demonstrated that 3T CE-MRA with calf compression (to prevent venous contamination) resulted in better visualization of tibial arteries than DSA [29]. For these reasons, CE-MRA is ideally suited for patients at high risk for calcification of the tibial and pedal arteries, particularly patients with diabetes and patients >80 years of age [28,30]. In comparison with duplex US, CE-MRA is more accurate for detecting and quantifying significant stenoses and for preoperative planning [31]. In a randomized controlled trial comparison with duplex US, CE-MRA for the initial imaging workup of patients with PAD reduced the need for additional imaging [32]. In a meta-analysis comparison to CTA, CE-MRA had equivalent sensitivity and specificity for detecting arterial lesions from the aorta to the tibial arteries in patients with intermittent claudication [26]. Some technical problems limit the utility of CE-MRA for imaging PAD. Challenges may include image quality related to low signal-to-noise ratio, limited spatial resolution, motion artifacts, long acquisition times, and loss of signal in arterial segments within metal stents or adjacent to metallic clips or prosthetic joints. Some of these problems have been addressed successfully with the use of newer imaging sequences and newer stent designs. Noncontrast MRA techniques predate CE-MRA in their development. However, long acquisition time and motion artifacts limit the use of these techniques in the abdomen and peripheral arteries [33]. | Lower Extremity Arterial Claudication Imaging Assessment for Revascularization. Time-resolved CE- MRA also reduces insufficient arterial filling and venous contamination, which are common limitations of standard multistation CE-MRA [28]. Although CE-MRA has not supplanted angiography as a reference standard, one small study demonstrated that 3T CE-MRA with calf compression (to prevent venous contamination) resulted in better visualization of tibial arteries than DSA [29]. For these reasons, CE-MRA is ideally suited for patients at high risk for calcification of the tibial and pedal arteries, particularly patients with diabetes and patients >80 years of age [28,30]. In comparison with duplex US, CE-MRA is more accurate for detecting and quantifying significant stenoses and for preoperative planning [31]. In a randomized controlled trial comparison with duplex US, CE-MRA for the initial imaging workup of patients with PAD reduced the need for additional imaging [32]. In a meta-analysis comparison to CTA, CE-MRA had equivalent sensitivity and specificity for detecting arterial lesions from the aorta to the tibial arteries in patients with intermittent claudication [26]. Some technical problems limit the utility of CE-MRA for imaging PAD. Challenges may include image quality related to low signal-to-noise ratio, limited spatial resolution, motion artifacts, long acquisition times, and loss of signal in arterial segments within metal stents or adjacent to metallic clips or prosthetic joints. Some of these problems have been addressed successfully with the use of newer imaging sequences and newer stent designs. Noncontrast MRA techniques predate CE-MRA in their development. However, long acquisition time and motion artifacts limit the use of these techniques in the abdomen and peripheral arteries [33]. | 69411 |
acrac_69411_11 | Lower Extremity Arterial Claudication Imaging Assessment for Revascularization | To that end, recent advancements in noncontrast MRA techniques for imaging PAD have expanded the sequence options from time- of-flight and phase-contrast imaging to include electrocardiogram-gated fresh-blood partial Fourier fast spin echo, balanced steady-state free precession, and arterial spin labeling [33]. Two alternative approaches using balanced steady state for peripheral noncontrast MRA applications include flow-sensitive dephasing and quiescent-interval single shot [30,33-35]. When compared with bolus-chase and time-resolved gadolinium-enhanced MRA, initial studies of fresh-blood imaging of the tibial and pedal arteries have provided accurate imaging when technically successful. Overall, these methods are being increasingly adopted for patients with severe renal insufficiency at risk of developing nephrogenic systemic fibrosis. Lower Extremity Arterial Claudication Some technical problems limit the utility of noncontrast MRA for imaging PAD. Challenges may include image quality related to low signal-to-noise ratio, limited spatial resolution, motion artifacts, long acquisition times, unreliable visualization of lesions with high flow and turbulence (excessive signal loss at regions of high-grade stenoses), nonvisualization of patent vessel segments with reversed blood flow, and loss of signal in arterial segments within metal stents or adjacent to metallic clips or prosthetic joints. Some of these problems have been addressed successfully with the use of newer imaging sequences. With the newer noncontrast techniques, cardiac arrhythmia can impair image quality, limiting evaluation of the distal tibial and pedal arteries. Although useful tools to improve image quality have been suggested, larger-scale trials are required for evaluation of small-vessel PAD with noncontrast MRA [29,36]. | Lower Extremity Arterial Claudication Imaging Assessment for Revascularization. To that end, recent advancements in noncontrast MRA techniques for imaging PAD have expanded the sequence options from time- of-flight and phase-contrast imaging to include electrocardiogram-gated fresh-blood partial Fourier fast spin echo, balanced steady-state free precession, and arterial spin labeling [33]. Two alternative approaches using balanced steady state for peripheral noncontrast MRA applications include flow-sensitive dephasing and quiescent-interval single shot [30,33-35]. When compared with bolus-chase and time-resolved gadolinium-enhanced MRA, initial studies of fresh-blood imaging of the tibial and pedal arteries have provided accurate imaging when technically successful. Overall, these methods are being increasingly adopted for patients with severe renal insufficiency at risk of developing nephrogenic systemic fibrosis. Lower Extremity Arterial Claudication Some technical problems limit the utility of noncontrast MRA for imaging PAD. Challenges may include image quality related to low signal-to-noise ratio, limited spatial resolution, motion artifacts, long acquisition times, unreliable visualization of lesions with high flow and turbulence (excessive signal loss at regions of high-grade stenoses), nonvisualization of patent vessel segments with reversed blood flow, and loss of signal in arterial segments within metal stents or adjacent to metallic clips or prosthetic joints. Some of these problems have been addressed successfully with the use of newer imaging sequences. With the newer noncontrast techniques, cardiac arrhythmia can impair image quality, limiting evaluation of the distal tibial and pedal arteries. Although useful tools to improve image quality have been suggested, larger-scale trials are required for evaluation of small-vessel PAD with noncontrast MRA [29,36]. | 69411 |
acrac_69411_12 | Lower Extremity Arterial Claudication Imaging Assessment for Revascularization | MRA Lower Extremity Without IV Contrast MRA of one or both lower extremities can be performed without imaging the abdomen and pelvis when aortoiliac disease is not a concern or the state of the aorta and iliac arteries is already known. Noncontrast MRA techniques predate CE-MRA in their development. However, long acquisition time and motion artifacts limit the use of these techniques in the abdomen and peripheral arteries [33]. To that end, recent advancements in noncontrast MRA techniques for imaging PAD have expanded the sequence options from time- of-flight and phase-contrast imaging to include electrocardiogram-gated fresh-blood partial Fourier fast spin echo, balanced steady-state free precession, and arterial spin labeling [33]. Two alternative approaches using balanced steady state for peripheral noncontrast MRA applications include flow-sensitive dephasing and quiescent-interval single shot [30,33-35]. When compared with bolus-chase and time-resolved gadolinium-enhanced MRA, initial studies of fresh-blood imaging of the tibial and pedal arteries have provided accurate imaging when technically successful. Overall, these methods are being increasingly adopted for patients with severe renal insufficiency at risk of developing nephrogenic systemic fibrosis. Some technical problems limit the utility of noncontrast MRA for imaging PAD. Challenges may include image quality related to low signal-to-noise ratio, limited spatial resolution, motion artifacts, long acquisition times, unreliable visualization of lesions with high flow and turbulence (excessive signal loss at regions of high-grade stenoses), nonvisualization of patent vessel segments with reversed blood flow, and loss of signal in arterial segments within metal stents or adjacent to metallic clips or prosthetic joints. Some of these problems have been addressed successfully with the use of newer imaging sequences. | Lower Extremity Arterial Claudication Imaging Assessment for Revascularization. MRA Lower Extremity Without IV Contrast MRA of one or both lower extremities can be performed without imaging the abdomen and pelvis when aortoiliac disease is not a concern or the state of the aorta and iliac arteries is already known. Noncontrast MRA techniques predate CE-MRA in their development. However, long acquisition time and motion artifacts limit the use of these techniques in the abdomen and peripheral arteries [33]. To that end, recent advancements in noncontrast MRA techniques for imaging PAD have expanded the sequence options from time- of-flight and phase-contrast imaging to include electrocardiogram-gated fresh-blood partial Fourier fast spin echo, balanced steady-state free precession, and arterial spin labeling [33]. Two alternative approaches using balanced steady state for peripheral noncontrast MRA applications include flow-sensitive dephasing and quiescent-interval single shot [30,33-35]. When compared with bolus-chase and time-resolved gadolinium-enhanced MRA, initial studies of fresh-blood imaging of the tibial and pedal arteries have provided accurate imaging when technically successful. Overall, these methods are being increasingly adopted for patients with severe renal insufficiency at risk of developing nephrogenic systemic fibrosis. Some technical problems limit the utility of noncontrast MRA for imaging PAD. Challenges may include image quality related to low signal-to-noise ratio, limited spatial resolution, motion artifacts, long acquisition times, unreliable visualization of lesions with high flow and turbulence (excessive signal loss at regions of high-grade stenoses), nonvisualization of patent vessel segments with reversed blood flow, and loss of signal in arterial segments within metal stents or adjacent to metallic clips or prosthetic joints. Some of these problems have been addressed successfully with the use of newer imaging sequences. | 69411 |
acrac_69411_13 | Lower Extremity Arterial Claudication Imaging Assessment for Revascularization | With the newer noncontrast techniques, cardiac arrhythmia can impair image quality, limiting evaluation of the distal tibial and pedal arteries. Although useful tools to improve image quality have been suggested, larger-scale trials are required for evaluation of small-vessel PAD with noncontrast MRA [29,36]. US Duplex Doppler Lower Extremity Duplex US of the extremities can be used to identify the location, degree, and extent of stenosis to the level of the knee [37]. Although duplex US includes images in grayscale or in color or power Doppler, the clinically relevant information derived from duplex studies has been validated from analysis of the velocity of blood flow. The sensitivity and specificity for the diagnosis of stenoses >50% in diameter from the iliac arteries to the popliteal arteries are each approximately 90% to 95% [37-39]. Accuracy of the duplex examination depends on the ability of the technique to visualize the vessel adequately. The use of color improves accuracy [40]. Accuracy is diminished in examinations of the iliac arteries if bowel gas or tortuosity obscures the iliac vessels. Dense calcification can also obscure flow, particularly if flow is slow. Accuracy of duplex US is also decreased in the setting of multiple sequential lesions [41]. Duplex US has been established as a useful surveillance tool for arterial bypass grafts with established criteria for graft stenosis and thresholds for reintervention [42]. However, evidence and standards for duplex US surveillance after endovascular treatment are lacking, although duplex US is commonly used for this indication [42]. In comparison with CE-MRA, duplex US is less accurate for detecting significant stenoses and for preoperative planning [31]. In a randomized controlled trial comparison with CE-MRA, duplex US for the initial imaging workup of patients with PAD increased the need for additional imaging [32]. CTA is more clinically useful than duplex US [21]. | Lower Extremity Arterial Claudication Imaging Assessment for Revascularization. With the newer noncontrast techniques, cardiac arrhythmia can impair image quality, limiting evaluation of the distal tibial and pedal arteries. Although useful tools to improve image quality have been suggested, larger-scale trials are required for evaluation of small-vessel PAD with noncontrast MRA [29,36]. US Duplex Doppler Lower Extremity Duplex US of the extremities can be used to identify the location, degree, and extent of stenosis to the level of the knee [37]. Although duplex US includes images in grayscale or in color or power Doppler, the clinically relevant information derived from duplex studies has been validated from analysis of the velocity of blood flow. The sensitivity and specificity for the diagnosis of stenoses >50% in diameter from the iliac arteries to the popliteal arteries are each approximately 90% to 95% [37-39]. Accuracy of the duplex examination depends on the ability of the technique to visualize the vessel adequately. The use of color improves accuracy [40]. Accuracy is diminished in examinations of the iliac arteries if bowel gas or tortuosity obscures the iliac vessels. Dense calcification can also obscure flow, particularly if flow is slow. Accuracy of duplex US is also decreased in the setting of multiple sequential lesions [41]. Duplex US has been established as a useful surveillance tool for arterial bypass grafts with established criteria for graft stenosis and thresholds for reintervention [42]. However, evidence and standards for duplex US surveillance after endovascular treatment are lacking, although duplex US is commonly used for this indication [42]. In comparison with CE-MRA, duplex US is less accurate for detecting significant stenoses and for preoperative planning [31]. In a randomized controlled trial comparison with CE-MRA, duplex US for the initial imaging workup of patients with PAD increased the need for additional imaging [32]. CTA is more clinically useful than duplex US [21]. | 69411 |
acrac_69377_0 | Growth Disturbances Risk of Fetal Growth Restriction | Introduction/Background Fetal growth restriction (FGR) is an important complication of pregnancy and is associated with significant risks of perinatal morbidity and mortality. A small-for-gestational-age (SGA) fetus is defined as a fetus whose estimated fetal weight (EFW) is below the 10th percentile for gestational age [1,2]. FGR implies an SGA fetus that has not reached its growth potential as measured by EFW. Although some SGA fetuses are constitutionally small and not at risk for perinatal morbidity and mortality [3], others are affected by a variety of maternal or placental conditions that lead to uteroplacental insufficiency and potential adverse outcomes, such as neurodevelopmental delay [4]. Given the difficulty in differentiating fetuses who are constitutionally SGA from those with FGR, it is typical for all fetuses with an EFW below the 10th percentile to be treated comparably. Finally, some fetuses may have an EFW at or above the 10th percentile but may also be at risk for adverse outcomes because of their suboptimal rate of growth [5,6]. The diagnosis of FGR is based upon an accurate assessment of gestational age. If gestational age is uncertain, then repeat evaluation to ensure appropriate growth velocity is recommended. Many different fetal biometric growth curves exist, and it is preferable to use the growth curve that most approximates the population being studied. The American College of Radiology seeks and encourages collaboration with other organizations on the development of the ACR Appropriateness Criteria through society representation on expert panels. Participation by representatives from collaborating societies on the expert panel does not necessarily imply individual or society endorsement of the final document. Reprint requests to: [email protected] | Growth Disturbances Risk of Fetal Growth Restriction. Introduction/Background Fetal growth restriction (FGR) is an important complication of pregnancy and is associated with significant risks of perinatal morbidity and mortality. A small-for-gestational-age (SGA) fetus is defined as a fetus whose estimated fetal weight (EFW) is below the 10th percentile for gestational age [1,2]. FGR implies an SGA fetus that has not reached its growth potential as measured by EFW. Although some SGA fetuses are constitutionally small and not at risk for perinatal morbidity and mortality [3], others are affected by a variety of maternal or placental conditions that lead to uteroplacental insufficiency and potential adverse outcomes, such as neurodevelopmental delay [4]. Given the difficulty in differentiating fetuses who are constitutionally SGA from those with FGR, it is typical for all fetuses with an EFW below the 10th percentile to be treated comparably. Finally, some fetuses may have an EFW at or above the 10th percentile but may also be at risk for adverse outcomes because of their suboptimal rate of growth [5,6]. The diagnosis of FGR is based upon an accurate assessment of gestational age. If gestational age is uncertain, then repeat evaluation to ensure appropriate growth velocity is recommended. Many different fetal biometric growth curves exist, and it is preferable to use the growth curve that most approximates the population being studied. The American College of Radiology seeks and encourages collaboration with other organizations on the development of the ACR Appropriateness Criteria through society representation on expert panels. Participation by representatives from collaborating societies on the expert panel does not necessarily imply individual or society endorsement of the final document. Reprint requests to: [email protected] | 69377 |
acrac_69377_1 | Growth Disturbances Risk of Fetal Growth Restriction | US Pregnant Uterus Biophysical Profile The BPP is the mainstay of fetal well-being evaluation and consists of four parameters variably sensitive to the acute exposure of the fetus to hypoxemia: fetal breathing movements, fetal limb and body movements, fetal tone, and amniotic fluid volume (which is thought to be more of an indicator of chronic hypoxemia). The NST, which is sometimes included with the BPP as a fifth component, can be used alone as a test of acute fetal well-being status, but it is often coupled with amniotic fluid measurement, a valuable reflection of fetal hypoxemic exposure over the previous week. Each of the four (or five) components of the BPP receives a score of 0 or 2, leading to a maximum score of 8 (or 10). Scores of 8 (or 10) are strong indicators of a well-compensated fetus [27]. For those at risk for fetal demise, testing strategies usually evaluate one or more of the fetal well-being parameters at least weekly. For the well-being of those fetuses at highest risk for fetal demise, testing can often occur twice weekly or even daily, from the point of postnatal viability until delivery is indicated. Amniotic fluid volume is usually assessed at least weekly, but may be evaluated more often if it is approaching severely low levels. Daily or even more frequent testing by BPP or NST may be indicated in critical situations. To our knowledge, there are no trials evaluating the use of the BPP as an initial procedure among those at low risk for FGR. US Duplex Doppler Velocimetry Fetal Umbilical Artery Although much research has centered on the use of Doppler velocimetry of the fetal umbilical artery among fetuses with known FGR, little work has been done among fetuses in women who are felt to be at low risk for FGR. In fact, umbilical artery Doppler velocimetry has not been shown to be a useful screening tool for FGR [31]. | Growth Disturbances Risk of Fetal Growth Restriction. US Pregnant Uterus Biophysical Profile The BPP is the mainstay of fetal well-being evaluation and consists of four parameters variably sensitive to the acute exposure of the fetus to hypoxemia: fetal breathing movements, fetal limb and body movements, fetal tone, and amniotic fluid volume (which is thought to be more of an indicator of chronic hypoxemia). The NST, which is sometimes included with the BPP as a fifth component, can be used alone as a test of acute fetal well-being status, but it is often coupled with amniotic fluid measurement, a valuable reflection of fetal hypoxemic exposure over the previous week. Each of the four (or five) components of the BPP receives a score of 0 or 2, leading to a maximum score of 8 (or 10). Scores of 8 (or 10) are strong indicators of a well-compensated fetus [27]. For those at risk for fetal demise, testing strategies usually evaluate one or more of the fetal well-being parameters at least weekly. For the well-being of those fetuses at highest risk for fetal demise, testing can often occur twice weekly or even daily, from the point of postnatal viability until delivery is indicated. Amniotic fluid volume is usually assessed at least weekly, but may be evaluated more often if it is approaching severely low levels. Daily or even more frequent testing by BPP or NST may be indicated in critical situations. To our knowledge, there are no trials evaluating the use of the BPP as an initial procedure among those at low risk for FGR. US Duplex Doppler Velocimetry Fetal Umbilical Artery Although much research has centered on the use of Doppler velocimetry of the fetal umbilical artery among fetuses with known FGR, little work has been done among fetuses in women who are felt to be at low risk for FGR. In fact, umbilical artery Doppler velocimetry has not been shown to be a useful screening tool for FGR [31]. | 69377 |
acrac_69377_2 | Growth Disturbances Risk of Fetal Growth Restriction | US Duplex Doppler Velocimetry Fetal Middle Cerebral Artery To our knowledge, there are no trials evaluating the use of Doppler velocimetry of the fetal middle cerebral artery as an FGR screening tool in low-risk women. US Duplex Doppler Velocimetry Ductus Venosus To our knowledge, there are no trials evaluating the use of Doppler velocimetry of the fetal ductus venosus as a FGR screening tool in low-risk women. Variant 2: Growth disturbance. High risk for fetal growth restriction. Initial evaluation. As discussed above, the mainstay for evaluation of a fetus for FGR is assessment of fetal biometry. This typically occurs during transabdominal US of the uterus. If a fetus is identified as having FGR, confirmation of fetal well- being, such as with a BPP, is necessary. Doppler interrogation of the umbilical artery is a useful tool for timing of delivery for those with FGR. US Pregnant Uterus Transabdominal As opposed to those at low-risk for FGR, those at high-risk for FGR are especially important to identify. As discussed above, US is currently the primary method of identification of fetuses with FGR. Many various historical and clinical factors can suggest FGR, but US-derived fetal biometry is the only current means to confirm a clinical suspicion for FGR. US Duplex Doppler Velocimetry Fetal Umbilical Artery Fetal umbilical artery Doppler velocimetry has garnered the most research for those with known or suspected FGR. Alfirevic et al [35] in a meta-analysis of 20 controlled trials of umbilical artery Doppler US found that management incorporating umbilical artery duplex Doppler was associated with improved perinatal outcome in high-risk pregnancies, reduced antenatal admissions, inductions of labor, and Cesarean delivery for fetal distress, and reduced odds of perinatal death by 38%. Variant 3: Established fetal growth restriction. Follow-up evaluation. | Growth Disturbances Risk of Fetal Growth Restriction. US Duplex Doppler Velocimetry Fetal Middle Cerebral Artery To our knowledge, there are no trials evaluating the use of Doppler velocimetry of the fetal middle cerebral artery as an FGR screening tool in low-risk women. US Duplex Doppler Velocimetry Ductus Venosus To our knowledge, there are no trials evaluating the use of Doppler velocimetry of the fetal ductus venosus as a FGR screening tool in low-risk women. Variant 2: Growth disturbance. High risk for fetal growth restriction. Initial evaluation. As discussed above, the mainstay for evaluation of a fetus for FGR is assessment of fetal biometry. This typically occurs during transabdominal US of the uterus. If a fetus is identified as having FGR, confirmation of fetal well- being, such as with a BPP, is necessary. Doppler interrogation of the umbilical artery is a useful tool for timing of delivery for those with FGR. US Pregnant Uterus Transabdominal As opposed to those at low-risk for FGR, those at high-risk for FGR are especially important to identify. As discussed above, US is currently the primary method of identification of fetuses with FGR. Many various historical and clinical factors can suggest FGR, but US-derived fetal biometry is the only current means to confirm a clinical suspicion for FGR. US Duplex Doppler Velocimetry Fetal Umbilical Artery Fetal umbilical artery Doppler velocimetry has garnered the most research for those with known or suspected FGR. Alfirevic et al [35] in a meta-analysis of 20 controlled trials of umbilical artery Doppler US found that management incorporating umbilical artery duplex Doppler was associated with improved perinatal outcome in high-risk pregnancies, reduced antenatal admissions, inductions of labor, and Cesarean delivery for fetal distress, and reduced odds of perinatal death by 38%. Variant 3: Established fetal growth restriction. Follow-up evaluation. | 69377 |
acrac_69377_3 | Growth Disturbances Risk of Fetal Growth Restriction | US Pregnant Uterus Transabdominal Once a diagnosis of FGR has been considered and/or made, serial evaluation of fetal growth is performed to assess the degree to which the fetus is compensating within this abnormal milieu [1,8,11]. Decreasing percentile growth would suggest the need for more intensive monitoring for assessment of fetal well-being or perhaps delivery depending upon clinical parameters, including gestational age. US Duplex Doppler Velocimetry Fetal Middle Cerebral Artery In a systematic review, middle cerebral artery Doppler indexes, which were thought to indicate cerebral redistribution, suggested that SGA fetuses with abnormal middle cerebral artery Doppler velocimetry were associated with neurodevelopmental problems at follow-up of SGA fetuses. The authors called for more adequately controlled studies with long-term follow-up before clear conclusions could be drawn [46]. Among Summary of Recommendations Variant 1: US pregnant uterus transabdominal is usually appropriate for the initial imaging of pregnant women who are at low risk for FGR. Variant 2: US pregnant uterus transabdominal is usually appropriate for the initial imaging of pregnant women who are at high risk for FGR. For those who are found to have growth restriction, US duplex Doppler velocimetry fetal umbilical artery and US pregnant uterus BPP are usually appropriate. Variant 3: US pregnant uterus BPP, US pregnant uterus transabdominal, US duplex Doppler velocimetry fetal umbilical artery, US duplex Doppler velocimetry ductus venosus, and US duplex Doppler velocimetry fetal middle cerebral artery are usually appropriate for follow-up evaluation in pregnant women with established FGR. These procedures are complementary (ie, all tests should be performed). Supporting Documents The evidence table, literature search, and appendix for this topic are available at https://acsearch. acr.org/list. | Growth Disturbances Risk of Fetal Growth Restriction. US Pregnant Uterus Transabdominal Once a diagnosis of FGR has been considered and/or made, serial evaluation of fetal growth is performed to assess the degree to which the fetus is compensating within this abnormal milieu [1,8,11]. Decreasing percentile growth would suggest the need for more intensive monitoring for assessment of fetal well-being or perhaps delivery depending upon clinical parameters, including gestational age. US Duplex Doppler Velocimetry Fetal Middle Cerebral Artery In a systematic review, middle cerebral artery Doppler indexes, which were thought to indicate cerebral redistribution, suggested that SGA fetuses with abnormal middle cerebral artery Doppler velocimetry were associated with neurodevelopmental problems at follow-up of SGA fetuses. The authors called for more adequately controlled studies with long-term follow-up before clear conclusions could be drawn [46]. Among Summary of Recommendations Variant 1: US pregnant uterus transabdominal is usually appropriate for the initial imaging of pregnant women who are at low risk for FGR. Variant 2: US pregnant uterus transabdominal is usually appropriate for the initial imaging of pregnant women who are at high risk for FGR. For those who are found to have growth restriction, US duplex Doppler velocimetry fetal umbilical artery and US pregnant uterus BPP are usually appropriate. Variant 3: US pregnant uterus BPP, US pregnant uterus transabdominal, US duplex Doppler velocimetry fetal umbilical artery, US duplex Doppler velocimetry ductus venosus, and US duplex Doppler velocimetry fetal middle cerebral artery are usually appropriate for follow-up evaluation in pregnant women with established FGR. These procedures are complementary (ie, all tests should be performed). Supporting Documents The evidence table, literature search, and appendix for this topic are available at https://acsearch. acr.org/list. | 69377 |
acrac_3194787_0 | Preprocedural Planning for Left Atrial Procedures in Atrial Fibrillation | Introduction/Background Atrial fibrillation (AF) is the most common cardiac arrhythmia, affecting 3% of the population [1,2], predisposing to circulatory stasis and formation of left atrial (LA) thrombi, mostly in the LA appendage (LAA) [1,3,4]. Endovascular treatment of AF continues to evolve, with favorable outcomes. Catheter ablation eliminates the electrical connection between the left atrium and ectopic pulmonary venous arrhythmogenic foci [5] and can be used to eliminate substrate for reentry within or adjacent to areas of LA fibrosis [6]. Imaging for preprocedural planning for LA ablation has been shown to improve success rate of procedures, minimize complications, exclude contraindications for the procedure, and minimize the risk of AF recurrence [7]. Endovascular device occlusion/closure of the LAA is used to reduce LAA thrombi from embolizing into the systemic circulation [8-10]. Cardioversion aims to restore a patient to normal sinus rhythm, and in such patients, precardioversion imaging is used to exclude the presence of LAA thrombus because up to 7% of patients have atrial thrombi despite anticoagulation regimens [7]. The document discusses imaging guidelines for the following preprocedural clinical tasks in patients needing endovascular treatment or cardioversion of AF: 1) before catheter ablation, to determine the anatomy of the pulmonary veins, the size of their ostia if using balloon technology, and to determine the relationship of the left atrium to adjacent structures (such as the esophagus) to minimize complications and exclude thrombi; 2) before LAA occlusion to assess LAA morphology and dimensions for device selection, exclude LAA thrombi, assess LAA relationship to adjacent cardiac, and extracardiac structures to aid in determining feasibility and preprocedural planning; and 3) before cardioversion, to exclude LA/LAA thrombi. aUniversity of Michigan, Ann Arbor, Michigan. | Preprocedural Planning for Left Atrial Procedures in Atrial Fibrillation. Introduction/Background Atrial fibrillation (AF) is the most common cardiac arrhythmia, affecting 3% of the population [1,2], predisposing to circulatory stasis and formation of left atrial (LA) thrombi, mostly in the LA appendage (LAA) [1,3,4]. Endovascular treatment of AF continues to evolve, with favorable outcomes. Catheter ablation eliminates the electrical connection between the left atrium and ectopic pulmonary venous arrhythmogenic foci [5] and can be used to eliminate substrate for reentry within or adjacent to areas of LA fibrosis [6]. Imaging for preprocedural planning for LA ablation has been shown to improve success rate of procedures, minimize complications, exclude contraindications for the procedure, and minimize the risk of AF recurrence [7]. Endovascular device occlusion/closure of the LAA is used to reduce LAA thrombi from embolizing into the systemic circulation [8-10]. Cardioversion aims to restore a patient to normal sinus rhythm, and in such patients, precardioversion imaging is used to exclude the presence of LAA thrombus because up to 7% of patients have atrial thrombi despite anticoagulation regimens [7]. The document discusses imaging guidelines for the following preprocedural clinical tasks in patients needing endovascular treatment or cardioversion of AF: 1) before catheter ablation, to determine the anatomy of the pulmonary veins, the size of their ostia if using balloon technology, and to determine the relationship of the left atrium to adjacent structures (such as the esophagus) to minimize complications and exclude thrombi; 2) before LAA occlusion to assess LAA morphology and dimensions for device selection, exclude LAA thrombi, assess LAA relationship to adjacent cardiac, and extracardiac structures to aid in determining feasibility and preprocedural planning; and 3) before cardioversion, to exclude LA/LAA thrombi. aUniversity of Michigan, Ann Arbor, Michigan. | 3194787 |
acrac_3194787_1 | Preprocedural Planning for Left Atrial Procedures in Atrial Fibrillation | bResearch Author, Allegheny Health Network Imaging Institute, Pittsburgh, Pennsylvania. cPanel Chair, Massachusetts General Hospital, Boston, Massachusetts. dUniversity of Utah, Department of Radiology and Imaging Sciences, Salt Lake City, Utah; Commission on Nuclear Medicine and Molecular Imaging. eCleveland Clinic, Cleveland, Ohio; Cardiology expert. fEmory University, Atlanta, Georgia. gOregon Health & Science University, Portland, Oregon. hChristianaCare Health System, Newark, Delaware; Society of General Internal Medicine. iUniversity of California San Francisco, San Francisco, California. jVA Palo Alto Health Care System, Palo Alto, California and Stanford University, Stanford, California. kInnovation Health Services, Norfolk, Virginia. lMassachusetts General Hospital, Boston, Massachusetts. mHospital of the University of Pennsylvania, Philadelphia, Pennsylvania; Heart Rhythm Society. nSanger Heart and Vascular Institute, Charlotte, North Carolina; Society of Cardiovascular Computed Tomography. oIndiana University School of Medicine, Indianapolis, Indiana; American College of Physicians. pDuke University Medical Center, Durham, North Carolina. qOhio State University, Columbus, Ohio; Society for Cardiovascular Magnetic Resonance. rSpecialty Chair, Duke University Medical Center, Durham, North Carolina. The American College of Radiology seeks and encourages collaboration with other organizations on the development of the ACR Appropriateness Criteria through representation of such organizations on expert panels. Participation on the expert panel does not necessarily imply endorsement of the final document by individual contributors or their respective organization. Reprint requests to: [email protected] All elements are essential: 1) timing, 2) reconstructions/reformats, and 3) 3-D renderings. Standard CTs with contrast also include timing issues and reconstructions/reformats. Only in CTA, however, is 3-D rendering a required element. | Preprocedural Planning for Left Atrial Procedures in Atrial Fibrillation. bResearch Author, Allegheny Health Network Imaging Institute, Pittsburgh, Pennsylvania. cPanel Chair, Massachusetts General Hospital, Boston, Massachusetts. dUniversity of Utah, Department of Radiology and Imaging Sciences, Salt Lake City, Utah; Commission on Nuclear Medicine and Molecular Imaging. eCleveland Clinic, Cleveland, Ohio; Cardiology expert. fEmory University, Atlanta, Georgia. gOregon Health & Science University, Portland, Oregon. hChristianaCare Health System, Newark, Delaware; Society of General Internal Medicine. iUniversity of California San Francisco, San Francisco, California. jVA Palo Alto Health Care System, Palo Alto, California and Stanford University, Stanford, California. kInnovation Health Services, Norfolk, Virginia. lMassachusetts General Hospital, Boston, Massachusetts. mHospital of the University of Pennsylvania, Philadelphia, Pennsylvania; Heart Rhythm Society. nSanger Heart and Vascular Institute, Charlotte, North Carolina; Society of Cardiovascular Computed Tomography. oIndiana University School of Medicine, Indianapolis, Indiana; American College of Physicians. pDuke University Medical Center, Durham, North Carolina. qOhio State University, Columbus, Ohio; Society for Cardiovascular Magnetic Resonance. rSpecialty Chair, Duke University Medical Center, Durham, North Carolina. The American College of Radiology seeks and encourages collaboration with other organizations on the development of the ACR Appropriateness Criteria through representation of such organizations on expert panels. Participation on the expert panel does not necessarily imply endorsement of the final document by individual contributors or their respective organization. Reprint requests to: [email protected] All elements are essential: 1) timing, 2) reconstructions/reformats, and 3) 3-D renderings. Standard CTs with contrast also include timing issues and reconstructions/reformats. Only in CTA, however, is 3-D rendering a required element. | 3194787 |
acrac_3194787_2 | Preprocedural Planning for Left Atrial Procedures in Atrial Fibrillation | This corresponds to the definitions that the CMS has applied to the Current Procedural Terminology codes. For the purposes of the document, CTA chest may be performed with electrocardiogram (ECG) gating, triggering, high pitch dual source acquisition, or nongated acquisition depending on site technology and experience. This is in distinction to CT heart, which is routinely performed with ECG gating and is often limited in z-axis to only cover the heart. Similarly, MR angiography (MRA) chest may be an ECG-gated or nongated acquisition and may be performed with or without contrast. The 3-D reconstructions and renderings are an essential element for MRA. Discussion of Procedures by Variant Variant 1: Adult. Atrial fibrillation, atrial tachycardia, or atypical atrial flutter. Preprocedural planning prior to left atrial ablation. Arteriography Coronary Invasive coronary arteriography provides assessment of coronary artery disease. There is no relevant literature to support the use of routine coronary arteriography to guide preprocedural planning before LA ablation procedure. Catheter Venography Pulmonary Catheter venography provides assessment of the pulmonary veins. There is little evidence in the literature supporting the use of catheter venography for routine preprocedural planning for atrial arrhythmia ablation. One retrospective review evaluated periprocedural 3-D rotational angiography in 547 consecutive patients undergoing ablation via direct contrast injection into either the right or left atrium, with or without simultaneous oral contrast administration for esophageal mapping [15]. With the availability of electroanatomic mapping systems, there is no added benefit or literature supporting the use of catheter venography during atrial arrhythmia ablation. CT Chest With IV Contrast There is no relevant literature to support the use of CT chest with intravenous (IV) contrast to guide preprocedural planning before LA ablation. | Preprocedural Planning for Left Atrial Procedures in Atrial Fibrillation. This corresponds to the definitions that the CMS has applied to the Current Procedural Terminology codes. For the purposes of the document, CTA chest may be performed with electrocardiogram (ECG) gating, triggering, high pitch dual source acquisition, or nongated acquisition depending on site technology and experience. This is in distinction to CT heart, which is routinely performed with ECG gating and is often limited in z-axis to only cover the heart. Similarly, MR angiography (MRA) chest may be an ECG-gated or nongated acquisition and may be performed with or without contrast. The 3-D reconstructions and renderings are an essential element for MRA. Discussion of Procedures by Variant Variant 1: Adult. Atrial fibrillation, atrial tachycardia, or atypical atrial flutter. Preprocedural planning prior to left atrial ablation. Arteriography Coronary Invasive coronary arteriography provides assessment of coronary artery disease. There is no relevant literature to support the use of routine coronary arteriography to guide preprocedural planning before LA ablation procedure. Catheter Venography Pulmonary Catheter venography provides assessment of the pulmonary veins. There is little evidence in the literature supporting the use of catheter venography for routine preprocedural planning for atrial arrhythmia ablation. One retrospective review evaluated periprocedural 3-D rotational angiography in 547 consecutive patients undergoing ablation via direct contrast injection into either the right or left atrium, with or without simultaneous oral contrast administration for esophageal mapping [15]. With the availability of electroanatomic mapping systems, there is no added benefit or literature supporting the use of catheter venography during atrial arrhythmia ablation. CT Chest With IV Contrast There is no relevant literature to support the use of CT chest with intravenous (IV) contrast to guide preprocedural planning before LA ablation. | 3194787 |
acrac_3194787_3 | Preprocedural Planning for Left Atrial Procedures in Atrial Fibrillation | CT Chest Without and With IV Contrast There is no relevant literature to support the use of CT chest without and with IV contrast to guide preprocedural planning before LA ablation. CT Chest Without IV Contrast A single prospective study of 50 patients evaluated with unenhanced chest CT found no statistically significant difference in procedural times and ablation success rate [16]. This is limited by the population size and paucity of additional supporting data. Integration of unenhanced CT with electroanatomic mapping cannot be adequately performed [16], and evaluation for contraindications such as thrombus is not possible. CT Heart Function and Morphology With IV Contrast ECG-gated cardiac multidetector CT (MDCT) can assess left atrium, LAA, and pulmonary venous anatomy. In a study including 1,040 consecutive patients who underwent pulmonary vein catheter ablation, cardiovascular anatomic variants were identified on gated-CT in 18.7% of cases [5]. For patients undergoing repeat ablation procedures, CT can assess sites of pulmonary vein stenosis or occlusion caused by a prior procedure. It also provides information on adjacent thoracic structures, including the esophagus, and facilitates electroanatomic mapping for preprocedural planning for ablation. Left Atrial Procedures in Atrial Fibrillation CT heart has also been evaluated in assessing thrombi before various LA procedures. With specific regard to the preablation setting, the specificity and negative predictive value (NPV) of a single acquisition CT (without delayed or prone imaging) have been reported to be 88% and 100%, respectively, based on a study of 51 patients when compared with transesophageal echocardiography (TEE) [17]. The diagnostic accuracy of CT for thrombus can be further improved by using a delayed phase or prone position [18]. | Preprocedural Planning for Left Atrial Procedures in Atrial Fibrillation. CT Chest Without and With IV Contrast There is no relevant literature to support the use of CT chest without and with IV contrast to guide preprocedural planning before LA ablation. CT Chest Without IV Contrast A single prospective study of 50 patients evaluated with unenhanced chest CT found no statistically significant difference in procedural times and ablation success rate [16]. This is limited by the population size and paucity of additional supporting data. Integration of unenhanced CT with electroanatomic mapping cannot be adequately performed [16], and evaluation for contraindications such as thrombus is not possible. CT Heart Function and Morphology With IV Contrast ECG-gated cardiac multidetector CT (MDCT) can assess left atrium, LAA, and pulmonary venous anatomy. In a study including 1,040 consecutive patients who underwent pulmonary vein catheter ablation, cardiovascular anatomic variants were identified on gated-CT in 18.7% of cases [5]. For patients undergoing repeat ablation procedures, CT can assess sites of pulmonary vein stenosis or occlusion caused by a prior procedure. It also provides information on adjacent thoracic structures, including the esophagus, and facilitates electroanatomic mapping for preprocedural planning for ablation. Left Atrial Procedures in Atrial Fibrillation CT heart has also been evaluated in assessing thrombi before various LA procedures. With specific regard to the preablation setting, the specificity and negative predictive value (NPV) of a single acquisition CT (without delayed or prone imaging) have been reported to be 88% and 100%, respectively, based on a study of 51 patients when compared with transesophageal echocardiography (TEE) [17]. The diagnostic accuracy of CT for thrombus can be further improved by using a delayed phase or prone position [18]. | 3194787 |
acrac_3194787_4 | Preprocedural Planning for Left Atrial Procedures in Atrial Fibrillation | In a small study of 70 patients, the specificity and positive predictive value of CT increased from 84% and 15% to 100% and 100%, respectively, by adding a delayed phase [19]. The high NPV can be used to decrease the usefulness of TEE, so that it is selectively used only when CT is positive. Hong et al [20] found that patients who underwent cardiac MDCT followed by routine TEE had no significant difference in periprocedural stroke incidence when compared with selective TEE (0.2% incidence in each group, P > . 99); the sensitivity and NPV of CT were 100% and 100%, respectively, in this study. This strategy of selective use of TEE was also assessed during the COVID-19 pandemic, and the study showed no significant difference in periprocedural thromboembolic events between routine TEE versus selective use of TEE guided by CT (0% versus 0.4%, P = . 33, n = 637) [21]. Cardiac CT can assess LA size, which may help guide patient management as increased LA volume is an independent predictor of arrhythmia recurrence (n = 103; hazard ratio [HR], 1.011/mL; 95% confidence interval [CI], 1.003-1.020; P = . 002) [22]. CTA Coronary Arteries With IV Contrast CTA coronary arteries is used to assess coronary arterial disease. There is no relevant literature supporting the routine use of CTA coronary arteries with IV contrast to guide preprocedural planning before LA ablation. This can also be technically challenging due to irregular heart rhythm in these patients. Although not routinely recommended, if the specific clinical scenario requires coronary arterial evaluation in addition to preablation planning, the study may be performed as a CTA coronary examination and left atrium and pulmonary veins can be assessed simultaneously. MRA Chest Without and With IV Contrast MRA without and with IV contrast can assess LA size, pulmonary venous anatomy, and thrombus. | Preprocedural Planning for Left Atrial Procedures in Atrial Fibrillation. In a small study of 70 patients, the specificity and positive predictive value of CT increased from 84% and 15% to 100% and 100%, respectively, by adding a delayed phase [19]. The high NPV can be used to decrease the usefulness of TEE, so that it is selectively used only when CT is positive. Hong et al [20] found that patients who underwent cardiac MDCT followed by routine TEE had no significant difference in periprocedural stroke incidence when compared with selective TEE (0.2% incidence in each group, P > . 99); the sensitivity and NPV of CT were 100% and 100%, respectively, in this study. This strategy of selective use of TEE was also assessed during the COVID-19 pandemic, and the study showed no significant difference in periprocedural thromboembolic events between routine TEE versus selective use of TEE guided by CT (0% versus 0.4%, P = . 33, n = 637) [21]. Cardiac CT can assess LA size, which may help guide patient management as increased LA volume is an independent predictor of arrhythmia recurrence (n = 103; hazard ratio [HR], 1.011/mL; 95% confidence interval [CI], 1.003-1.020; P = . 002) [22]. CTA Coronary Arteries With IV Contrast CTA coronary arteries is used to assess coronary arterial disease. There is no relevant literature supporting the routine use of CTA coronary arteries with IV contrast to guide preprocedural planning before LA ablation. This can also be technically challenging due to irregular heart rhythm in these patients. Although not routinely recommended, if the specific clinical scenario requires coronary arterial evaluation in addition to preablation planning, the study may be performed as a CTA coronary examination and left atrium and pulmonary veins can be assessed simultaneously. MRA Chest Without and With IV Contrast MRA without and with IV contrast can assess LA size, pulmonary venous anatomy, and thrombus. | 3194787 |
acrac_3194787_5 | Preprocedural Planning for Left Atrial Procedures in Atrial Fibrillation | Both standard contrast-enhanced and dynamic time-resolved contrast-enhanced MRA (the latter performed with a lower gadolinium agent contrast dose) can assess the LA and pulmonary venous anatomy [26,27]. A recent comparison of these 2 acquisitions in 50 patients showed that time-resolved MRA has significantly higher overall image quality (P < . 0001) [27]. MRA with IV contrast can identify LA/LAA thrombi with an accuracy of 94.3%, a sensitivity of 66.7%, and a specificity of 95.2% compared with TEE [28]. MRA Chest Without IV Contrast Noncontrast enhanced MRA using 2-D and 3-D balanced steady-state free-precession (b-SSFP) can assess LA size and pulmonary venous anatomy, although it cannot reliably assess for thrombus. It may be a feasible alternative for preablation planning in patients who cannot receive IV contrast [29,30]. Despite an overall lower image quality, 2- D b-SSFP MRA yielded adequate LA and pulmonary vein anatomical information and adequate electroanatomic Left Atrial Procedures in Atrial Fibrillation integration when compared with contrast-enhanced CT in 54 patients, 27 scanned with each modality [29]. Further supporting data are scarce. MRA Coronary Arteries Without and With IV Contrast MRA coronary artery examination can assess proximal coronary artery anatomy. There is no relevant literature to support the use of MRA coronary arteries without and with IV contrast to guide preprocedural planning before LA ablation. MRA Coronary Arteries Without IV Contrast MRA coronary artery examination can assess proximal coronary artery anatomy. There is no relevant literature to support the use of MRA coronary arteries without IV contrast to guide preprocedural planning before LA ablation. MRI Heart Function and Morphology Without and With IV Contrast MRI heart with IV contrast can assess chamber size, presence of LA/LAA thrombi, and atrial fibrosis. | Preprocedural Planning for Left Atrial Procedures in Atrial Fibrillation. Both standard contrast-enhanced and dynamic time-resolved contrast-enhanced MRA (the latter performed with a lower gadolinium agent contrast dose) can assess the LA and pulmonary venous anatomy [26,27]. A recent comparison of these 2 acquisitions in 50 patients showed that time-resolved MRA has significantly higher overall image quality (P < . 0001) [27]. MRA with IV contrast can identify LA/LAA thrombi with an accuracy of 94.3%, a sensitivity of 66.7%, and a specificity of 95.2% compared with TEE [28]. MRA Chest Without IV Contrast Noncontrast enhanced MRA using 2-D and 3-D balanced steady-state free-precession (b-SSFP) can assess LA size and pulmonary venous anatomy, although it cannot reliably assess for thrombus. It may be a feasible alternative for preablation planning in patients who cannot receive IV contrast [29,30]. Despite an overall lower image quality, 2- D b-SSFP MRA yielded adequate LA and pulmonary vein anatomical information and adequate electroanatomic Left Atrial Procedures in Atrial Fibrillation integration when compared with contrast-enhanced CT in 54 patients, 27 scanned with each modality [29]. Further supporting data are scarce. MRA Coronary Arteries Without and With IV Contrast MRA coronary artery examination can assess proximal coronary artery anatomy. There is no relevant literature to support the use of MRA coronary arteries without and with IV contrast to guide preprocedural planning before LA ablation. MRA Coronary Arteries Without IV Contrast MRA coronary artery examination can assess proximal coronary artery anatomy. There is no relevant literature to support the use of MRA coronary arteries without IV contrast to guide preprocedural planning before LA ablation. MRI Heart Function and Morphology Without and With IV Contrast MRI heart with IV contrast can assess chamber size, presence of LA/LAA thrombi, and atrial fibrosis. | 3194787 |
acrac_3194787_6 | Preprocedural Planning for Left Atrial Procedures in Atrial Fibrillation | However, for pulmonary vein anatomy and measurements, it requires a complementary procedure (ie, a simultaneous 3-D acquisition [MRA]). In a study that matched 400 patients to either preprocedural cardiac CT or cardiac MR (CMR), procedural characteristics (fluoroscopy time, procedure duration, pulmonary veins identified, targeted, and isolated) and arrhythmia recurrence rates were not significantly different [31]. In 2 studies, MRI heart was shown to have more accurate and reproducible measurements of LA volume, which, when increased, is an independent predictor or postablation recurrence [31,32]. Among various MR sequences, long inversion time-delayed enhanced CMR has the highest reported diagnostic accuracy (99.2%), sensitivity (100%), and specificity (99.2%) for thrombus assessment compared with TEE [28]. Multiple studies have shown that the identification of presence of LA fibrosis on late gadolinium enhancement imaging is an independent predictor of postablation arrhythmia recurrence [33- 36]. This has led to the investigation of LA substrate-based ablation, in which these atrial fibrotic lesions are targeted in addition to the pulmonary veins to curb recurrence rates [37]. The prospective, multicenter Efficacy of Delayed Enhancement MRI-Guided Ablation vs Conventional Catheter Ablation of Atrial Fibrillation (DECAAF II) trial showed that among patients with persistent AF, MRI-guided fibrosis ablation plus pulmonary vein isolation, compared with pulmonary vein isolation catheter ablation only, resulted in no significant difference in atrial arrhythmia recurrence [6]. Other LA parameters such as LA volume, sphericity, LA volume index, LA emptying function, peak strain, and LA T1 relaxation time have also been assessed with MRI heart [32]. | Preprocedural Planning for Left Atrial Procedures in Atrial Fibrillation. However, for pulmonary vein anatomy and measurements, it requires a complementary procedure (ie, a simultaneous 3-D acquisition [MRA]). In a study that matched 400 patients to either preprocedural cardiac CT or cardiac MR (CMR), procedural characteristics (fluoroscopy time, procedure duration, pulmonary veins identified, targeted, and isolated) and arrhythmia recurrence rates were not significantly different [31]. In 2 studies, MRI heart was shown to have more accurate and reproducible measurements of LA volume, which, when increased, is an independent predictor or postablation recurrence [31,32]. Among various MR sequences, long inversion time-delayed enhanced CMR has the highest reported diagnostic accuracy (99.2%), sensitivity (100%), and specificity (99.2%) for thrombus assessment compared with TEE [28]. Multiple studies have shown that the identification of presence of LA fibrosis on late gadolinium enhancement imaging is an independent predictor of postablation arrhythmia recurrence [33- 36]. This has led to the investigation of LA substrate-based ablation, in which these atrial fibrotic lesions are targeted in addition to the pulmonary veins to curb recurrence rates [37]. The prospective, multicenter Efficacy of Delayed Enhancement MRI-Guided Ablation vs Conventional Catheter Ablation of Atrial Fibrillation (DECAAF II) trial showed that among patients with persistent AF, MRI-guided fibrosis ablation plus pulmonary vein isolation, compared with pulmonary vein isolation catheter ablation only, resulted in no significant difference in atrial arrhythmia recurrence [6]. Other LA parameters such as LA volume, sphericity, LA volume index, LA emptying function, peak strain, and LA T1 relaxation time have also been assessed with MRI heart [32]. | 3194787 |
acrac_3194787_7 | Preprocedural Planning for Left Atrial Procedures in Atrial Fibrillation | MRI Heart Function and Morphology Without IV Contrast MRI heart without IV contrast can provide cardiac function and chamber sizes, but the role in preprocedural planning is limited and there is scant literature on this topic. Also, unlike contrast-enhanced MRI heart, it is not sensitive for thrombus and cannot be used for scar assessment. SPECT or SPECT/CT MPI Rest and Stress There is no relevant literature to support the use of single-photon emission computed tomography (SPECT) or SPECT/CT myocardial perfusion imaging (MPI) rest and stress for to guide preprocedural planning before LA ablation. However, many operators use isoproterenol infusions to assess the presence of nonpulmonary vein triggers. Prior to using high doses of isoproterenol, assessment of MPI may be advisable in patients with symptoms consistent with angina or those with limited functional capacity. US Echocardiography Transesophageal TEE is a useful imaging technique for LA/LAA thrombus detection [38-40] and has a sensitivity of 93.3% to 100% and a specificity of 99% to 100%. TEE cannot reliably assess pulmonary venous anatomy, pulmonary venous measurements, and extracardiac anatomic relations. Hence, alternative imaging techniques like CT and MR are being assessed to allow a more targeted use of TEE in patients with suspected thrombus. US Echocardiography Transthoracic Resting Although TTE is commonly used in the initial workup of patients with AF and parameters such as LA volume/volume index, left ventricular ejection fraction [41], and LA size [42] can be assessed, there is no relevant literature that describes the specific use of TTE to guide preprocedural planning before LA ablation procedure. Left Atrial Procedures in Atrial Fibrillation Variant 2: Adult. Atrial fibrillation. Preprocedural planning prior to left atrial appendage endovascular occlusion. | Preprocedural Planning for Left Atrial Procedures in Atrial Fibrillation. MRI Heart Function and Morphology Without IV Contrast MRI heart without IV contrast can provide cardiac function and chamber sizes, but the role in preprocedural planning is limited and there is scant literature on this topic. Also, unlike contrast-enhanced MRI heart, it is not sensitive for thrombus and cannot be used for scar assessment. SPECT or SPECT/CT MPI Rest and Stress There is no relevant literature to support the use of single-photon emission computed tomography (SPECT) or SPECT/CT myocardial perfusion imaging (MPI) rest and stress for to guide preprocedural planning before LA ablation. However, many operators use isoproterenol infusions to assess the presence of nonpulmonary vein triggers. Prior to using high doses of isoproterenol, assessment of MPI may be advisable in patients with symptoms consistent with angina or those with limited functional capacity. US Echocardiography Transesophageal TEE is a useful imaging technique for LA/LAA thrombus detection [38-40] and has a sensitivity of 93.3% to 100% and a specificity of 99% to 100%. TEE cannot reliably assess pulmonary venous anatomy, pulmonary venous measurements, and extracardiac anatomic relations. Hence, alternative imaging techniques like CT and MR are being assessed to allow a more targeted use of TEE in patients with suspected thrombus. US Echocardiography Transthoracic Resting Although TTE is commonly used in the initial workup of patients with AF and parameters such as LA volume/volume index, left ventricular ejection fraction [41], and LA size [42] can be assessed, there is no relevant literature that describes the specific use of TTE to guide preprocedural planning before LA ablation procedure. Left Atrial Procedures in Atrial Fibrillation Variant 2: Adult. Atrial fibrillation. Preprocedural planning prior to left atrial appendage endovascular occlusion. | 3194787 |
acrac_3194787_8 | Preprocedural Planning for Left Atrial Procedures in Atrial Fibrillation | Arteriography Coronary There is no relevant literature to support the use of routine coronary arteriography to guide preprocedural planning before LAA occlusion planning. Catheter Venography Pulmonary There is no relevant literature to support the use of catheter pulmonary venography to guide preprocedural planning before LAA occlusion. CT Chest With IV Contrast There is no relevant literature to support the use of CT chest with IV contrast to guide preprocedural planning before LAA occlusion. CT Chest Without and With IV Contrast There is no relevant literature to support the use of CT chest without and with IV contrast to guide preprocedural planning before LAA occlusion. CT Chest Without IV Contrast There is no relevant literature to support the use of CT chest without IV contrast to guide preprocedural planning before LAA occlusion. An important role of CT is excluding contraindications to endovascular LAA occlusion procedures such as presence of thrombi. In a meta-analysis comparing CT and TEE for thrombus assessment, CT had a sensitivity and specificity of 96% and 92%, which increased to 100% and 99% if delayed imaging was performed [48]. CTA Chest With IV Contrast CTA of the chest can allow for assessment of the left atrium and LAA and can be performed with various acquisition parameters (with or without ECG gating) depending on scanner and local protocols. For device sizing, ECG-gated CTA is useful, and studies have shown high accuracy for gated CTA as outlined in the preceding section on CT heart [43,49]. Depending on the exact parameters, CTA chest with IV contrast is also useful in excluding thrombi especially when combined with delayed imaging. When coupled with delayed phase it has a 100% sensitivity [48]. CTA Coronary Arteries With IV Contrast There is no relevant literature to support the routine use of CTA coronary arteries with IV contrast to guide preprocedural planning before LAA occlusion procedures. | Preprocedural Planning for Left Atrial Procedures in Atrial Fibrillation. Arteriography Coronary There is no relevant literature to support the use of routine coronary arteriography to guide preprocedural planning before LAA occlusion planning. Catheter Venography Pulmonary There is no relevant literature to support the use of catheter pulmonary venography to guide preprocedural planning before LAA occlusion. CT Chest With IV Contrast There is no relevant literature to support the use of CT chest with IV contrast to guide preprocedural planning before LAA occlusion. CT Chest Without and With IV Contrast There is no relevant literature to support the use of CT chest without and with IV contrast to guide preprocedural planning before LAA occlusion. CT Chest Without IV Contrast There is no relevant literature to support the use of CT chest without IV contrast to guide preprocedural planning before LAA occlusion. An important role of CT is excluding contraindications to endovascular LAA occlusion procedures such as presence of thrombi. In a meta-analysis comparing CT and TEE for thrombus assessment, CT had a sensitivity and specificity of 96% and 92%, which increased to 100% and 99% if delayed imaging was performed [48]. CTA Chest With IV Contrast CTA of the chest can allow for assessment of the left atrium and LAA and can be performed with various acquisition parameters (with or without ECG gating) depending on scanner and local protocols. For device sizing, ECG-gated CTA is useful, and studies have shown high accuracy for gated CTA as outlined in the preceding section on CT heart [43,49]. Depending on the exact parameters, CTA chest with IV contrast is also useful in excluding thrombi especially when combined with delayed imaging. When coupled with delayed phase it has a 100% sensitivity [48]. CTA Coronary Arteries With IV Contrast There is no relevant literature to support the routine use of CTA coronary arteries with IV contrast to guide preprocedural planning before LAA occlusion procedures. | 3194787 |
acrac_3194787_9 | Preprocedural Planning for Left Atrial Procedures in Atrial Fibrillation | MRA Chest Without and With IV Contrast MRA chest with contrast can assess LA and LAA size and patency and identify LA/LAA thrombi with an accuracy of 94.3%, a sensitivity of 66.7%, and a specificity of 95.2% compared with TEE [28]. However, there are no large studies evaluating the use of MRA chest without and with IV contrast to guide preprocedural planning before LAA occlusion. Left Atrial Procedures in Atrial Fibrillation MRA Chest Without IV Contrast There is no relevant literature to support the use of MRA chest without IV contrast to guide preprocedural planning for LAA occlusion procedures. MRA Coronary Arteries Without and With IV Contrast There is no relevant literature to support the use of MRA coronary arteries without and with IV contrast to guide preprocedural planning for LAA occlusion procedures. MRA Coronary Arteries Without IV Contrast There is no relevant literature to support the use of MRA coronary arteries without IV contrast to guide preprocedural planning for LAA occlusion procedures. MRI Heart Function and Morphology Without and With IV Contrast MRI heart can assess the patency of the left atrium and LAA. There is a small amount of data assessing contrast- enhanced CMR for detection of LA/LAA thrombus. A retrospective analysis of 97 patients showed 100% concordance in LAA thrombus detection between 3-D early postgadolinium enhancement CMR and TEE [50]. Another study comparing CMR with TEE in 261 patients showed a 99.2% diagnostic accuracy for long T1 delayed enhancement CMR in detecting LAA thrombi [28]. This diagnostic accuracy was not affected by the presence of arrythmia during image acquisition. For the measurement of LAA, it needs to be complemented with a simultaneous 3D acquisition (ie, MRA). There are; however, no large studies to support the use of MRI in quantifying LAA size and morphology before endovascular occlusion [7]. | Preprocedural Planning for Left Atrial Procedures in Atrial Fibrillation. MRA Chest Without and With IV Contrast MRA chest with contrast can assess LA and LAA size and patency and identify LA/LAA thrombi with an accuracy of 94.3%, a sensitivity of 66.7%, and a specificity of 95.2% compared with TEE [28]. However, there are no large studies evaluating the use of MRA chest without and with IV contrast to guide preprocedural planning before LAA occlusion. Left Atrial Procedures in Atrial Fibrillation MRA Chest Without IV Contrast There is no relevant literature to support the use of MRA chest without IV contrast to guide preprocedural planning for LAA occlusion procedures. MRA Coronary Arteries Without and With IV Contrast There is no relevant literature to support the use of MRA coronary arteries without and with IV contrast to guide preprocedural planning for LAA occlusion procedures. MRA Coronary Arteries Without IV Contrast There is no relevant literature to support the use of MRA coronary arteries without IV contrast to guide preprocedural planning for LAA occlusion procedures. MRI Heart Function and Morphology Without and With IV Contrast MRI heart can assess the patency of the left atrium and LAA. There is a small amount of data assessing contrast- enhanced CMR for detection of LA/LAA thrombus. A retrospective analysis of 97 patients showed 100% concordance in LAA thrombus detection between 3-D early postgadolinium enhancement CMR and TEE [50]. Another study comparing CMR with TEE in 261 patients showed a 99.2% diagnostic accuracy for long T1 delayed enhancement CMR in detecting LAA thrombi [28]. This diagnostic accuracy was not affected by the presence of arrythmia during image acquisition. For the measurement of LAA, it needs to be complemented with a simultaneous 3D acquisition (ie, MRA). There are; however, no large studies to support the use of MRI in quantifying LAA size and morphology before endovascular occlusion [7]. | 3194787 |
acrac_3194787_10 | Preprocedural Planning for Left Atrial Procedures in Atrial Fibrillation | A limitation of CMR is its lower spatial resolution and longer acquisition times when compared with CT. MRI Heart Function and Morphology Without IV Contrast MRI heart without IV contrast can assess the size of left atrium. There is scant data in the literature supporting the use of noncontrast CMR for detecting LAA thrombi. Kitkungvan et al [28] compared CMR with TEE for detecting LAA thrombi. Steady-state free-precession (SSFP) cine MRI had a 91.6% diagnostic accuracy, 66.7% sensitivity, and 92.5% specificity. There is no relevant literature to support the use of nonenhanced CMR in quantifying LAA size and morphology to guide preprocedural planning for LAA occlusion procedures. SPECT or SPECT/CT MPI Rest and Stress There is no relevant literature to support the use of SPECT or SPECT/CT MPI rest and stress to guide preprocedural planning for LAA endovascular occlusion procedures. US Echocardiography Transthoracic Resting Transthoracic echocardiography TTE assesses LA parameters that can help with patient risk stratification for thromboembolic events such as LA volume/volume index, left ventricular ejection fraction [41], and LA size [42]. However, there is no relevant literature to support the use of TTE specifically for preprocedural planning to guide LAA endovascular occlusion. Variant 3: Adult. Atrial fibrillation. Preprocedural planning prior to electrical or pharmacologic cardioversion. Arteriography Coronary There is no relevant literature to support the use of coronary arteriography to guide preprocedural planning for cardioversion. Catheter Venography Pulmonary There is no relevant literature to support the use of catheter pulmonary venography to guide preprocedural planning for cardioversion. Left Atrial Procedures in Atrial Fibrillation CT Chest With IV Contrast CT chest can, in some cases, demonstrate thrombi; however, it is not optimized for LA assessment. | Preprocedural Planning for Left Atrial Procedures in Atrial Fibrillation. A limitation of CMR is its lower spatial resolution and longer acquisition times when compared with CT. MRI Heart Function and Morphology Without IV Contrast MRI heart without IV contrast can assess the size of left atrium. There is scant data in the literature supporting the use of noncontrast CMR for detecting LAA thrombi. Kitkungvan et al [28] compared CMR with TEE for detecting LAA thrombi. Steady-state free-precession (SSFP) cine MRI had a 91.6% diagnostic accuracy, 66.7% sensitivity, and 92.5% specificity. There is no relevant literature to support the use of nonenhanced CMR in quantifying LAA size and morphology to guide preprocedural planning for LAA occlusion procedures. SPECT or SPECT/CT MPI Rest and Stress There is no relevant literature to support the use of SPECT or SPECT/CT MPI rest and stress to guide preprocedural planning for LAA endovascular occlusion procedures. US Echocardiography Transthoracic Resting Transthoracic echocardiography TTE assesses LA parameters that can help with patient risk stratification for thromboembolic events such as LA volume/volume index, left ventricular ejection fraction [41], and LA size [42]. However, there is no relevant literature to support the use of TTE specifically for preprocedural planning to guide LAA endovascular occlusion. Variant 3: Adult. Atrial fibrillation. Preprocedural planning prior to electrical or pharmacologic cardioversion. Arteriography Coronary There is no relevant literature to support the use of coronary arteriography to guide preprocedural planning for cardioversion. Catheter Venography Pulmonary There is no relevant literature to support the use of catheter pulmonary venography to guide preprocedural planning for cardioversion. Left Atrial Procedures in Atrial Fibrillation CT Chest With IV Contrast CT chest can, in some cases, demonstrate thrombi; however, it is not optimized for LA assessment. | 3194787 |
acrac_3194787_11 | Preprocedural Planning for Left Atrial Procedures in Atrial Fibrillation | There is no relevant literature to support the use of CT chest with IV contrast to guide preprocedural planning for cardioversion. CT Chest Without and With IV Contrast There is no relevant literature to support the use of CT chest without and with IV contrast to guide preprocedural planning for cardioversion. CT Chest Without IV Contrast There is no relevant literature to support the use of CT chest without IV contrast to guide preprocedural planning for cardioversion. CT Heart Function and Morphology With IV Contrast There is increasing use of CT to exclude thrombus before LA procedures, especially when CT is already being performed for measurements and mapping (ie, before ablation and appendage occlusion). However, there are no large studies assessing the role of CT before cardioversion. In terms of diagnostic accuracy, CT has shown good performance for this indication (exclusion of thrombus), with a meta-analysis showing CT to have a sensitivity and specificity of 96% and 92%, respectively, which increased to 100% and 99%, respectively, if delayed imaging was performed [48]. Other studies have shown similar results [18,19,25]. CT has the advantage of noninvasive nature with 3-D capability. CTA Chest With IV Contrast Depending on the exact parameters, CTA with IV contrast is useful in excluding thrombi especially when combined with delayed imaging. When coupled with delayed phase it has a 100% sensitivity and NPV [48]. CTA Coronary Arteries With IV Contrast There is no relevant literature to support the use of CTA coronary arteries with IV contrast to guide preprocedural planning for cardioversion. MRA Chest Without and With IV Contrast MRA with IV contrast can identify LA/LAA thrombi with an accuracy of 94.3%, a sensitivity of 66.7%, and a specificity of 95.2% compared with TEE [28]. | Preprocedural Planning for Left Atrial Procedures in Atrial Fibrillation. There is no relevant literature to support the use of CT chest with IV contrast to guide preprocedural planning for cardioversion. CT Chest Without and With IV Contrast There is no relevant literature to support the use of CT chest without and with IV contrast to guide preprocedural planning for cardioversion. CT Chest Without IV Contrast There is no relevant literature to support the use of CT chest without IV contrast to guide preprocedural planning for cardioversion. CT Heart Function and Morphology With IV Contrast There is increasing use of CT to exclude thrombus before LA procedures, especially when CT is already being performed for measurements and mapping (ie, before ablation and appendage occlusion). However, there are no large studies assessing the role of CT before cardioversion. In terms of diagnostic accuracy, CT has shown good performance for this indication (exclusion of thrombus), with a meta-analysis showing CT to have a sensitivity and specificity of 96% and 92%, respectively, which increased to 100% and 99%, respectively, if delayed imaging was performed [48]. Other studies have shown similar results [18,19,25]. CT has the advantage of noninvasive nature with 3-D capability. CTA Chest With IV Contrast Depending on the exact parameters, CTA with IV contrast is useful in excluding thrombi especially when combined with delayed imaging. When coupled with delayed phase it has a 100% sensitivity and NPV [48]. CTA Coronary Arteries With IV Contrast There is no relevant literature to support the use of CTA coronary arteries with IV contrast to guide preprocedural planning for cardioversion. MRA Chest Without and With IV Contrast MRA with IV contrast can identify LA/LAA thrombi with an accuracy of 94.3%, a sensitivity of 66.7%, and a specificity of 95.2% compared with TEE [28]. | 3194787 |
acrac_3194787_12 | Preprocedural Planning for Left Atrial Procedures in Atrial Fibrillation | However, there are no large studies or relevant literature specifically evaluating the use of MRA chest without and with IV contrast for preprocedural planning before cardioversion. MRA Chest Without IV Contrast There is no relevant literature to support the use of MRA chest without IV contrast to guide preprocedural planning for cardioversion. MRA Coronary Arteries Without and With IV Contrast There is no relevant literature to support the use of MRA coronary arteries without and with IV contrast to guide preprocedural planning for cardioversion. MRA Coronary Arteries Without IV Contrast There is no relevant literature to support the use of MRA coronary arteries without IV contrast to guide preprocedural planning for cardioversion. MRI Heart Function and Morphology Without and With IV Contrast There is emerging data that contrast-enhanced CMR is a reliable imaging modality for the detection of LA/LAA thrombus. A retrospective analysis of 97 patients showed 100% concordance in LAA thrombus detection between 3-D early postgadolinium enhancement CMR and TEE [50]. Another study comparing CMR with TEE in 261 patients showed a 99.2% diagnostic accuracy for long T1 delayed enhancement CMR in detecting LAA thrombi [28]. MRI Heart Function and Morphology Without IV Contrast There is limited data in the literature supporting the use of noncontrast CMR for detecting LAA thrombi. Kitkungvan et al [28] compared CMR with TEE for detecting LAA thrombi. SSFP cine MRI had a 91.6% diagnostic accuracy, a 66.7% sensitivity, and a 92.5% specificity. Flow-related breathing and inhomogeneity artifacts were the main causes for this lower yield. Left Atrial Procedures in Atrial Fibrillation SPECT or SPECT/CT MPI Rest and Stress There is no relevant literature to support the use of SPECT or SPECT/CT MPI rest and stress to guide preprocedural planning for cardioversion. | Preprocedural Planning for Left Atrial Procedures in Atrial Fibrillation. However, there are no large studies or relevant literature specifically evaluating the use of MRA chest without and with IV contrast for preprocedural planning before cardioversion. MRA Chest Without IV Contrast There is no relevant literature to support the use of MRA chest without IV contrast to guide preprocedural planning for cardioversion. MRA Coronary Arteries Without and With IV Contrast There is no relevant literature to support the use of MRA coronary arteries without and with IV contrast to guide preprocedural planning for cardioversion. MRA Coronary Arteries Without IV Contrast There is no relevant literature to support the use of MRA coronary arteries without IV contrast to guide preprocedural planning for cardioversion. MRI Heart Function and Morphology Without and With IV Contrast There is emerging data that contrast-enhanced CMR is a reliable imaging modality for the detection of LA/LAA thrombus. A retrospective analysis of 97 patients showed 100% concordance in LAA thrombus detection between 3-D early postgadolinium enhancement CMR and TEE [50]. Another study comparing CMR with TEE in 261 patients showed a 99.2% diagnostic accuracy for long T1 delayed enhancement CMR in detecting LAA thrombi [28]. MRI Heart Function and Morphology Without IV Contrast There is limited data in the literature supporting the use of noncontrast CMR for detecting LAA thrombi. Kitkungvan et al [28] compared CMR with TEE for detecting LAA thrombi. SSFP cine MRI had a 91.6% diagnostic accuracy, a 66.7% sensitivity, and a 92.5% specificity. Flow-related breathing and inhomogeneity artifacts were the main causes for this lower yield. Left Atrial Procedures in Atrial Fibrillation SPECT or SPECT/CT MPI Rest and Stress There is no relevant literature to support the use of SPECT or SPECT/CT MPI rest and stress to guide preprocedural planning for cardioversion. | 3194787 |
acrac_69413_0 | Nonvariceal Upper Gastrointestinal Bleeding | Introduction/Background Upper gastrointestinal (GI) bleeding (UGIB) refers to bleeding occurring proximal to the ligament of Treitz, from the esophagus, stomach, or duodenum [1,2]. The incidence of nonvariceal UGIB is almost 5 times higher than that of variceal UGIB [1]. Peptic ulcer disease caused by Helicobacter pylori infection or nonsteroidal anti-inflammatory drug use is the most common cause of nonvariceal UGIB. The other causes of nonvariceal UGIB include Mallory- Weiss tears, esophagitis, pancreatitis, trauma, iatrogenic, or neoplastic [1]. Some rare causes of nonvariceal UGIB include hemobilia, hemosuccus pancreaticus, and aortoenteric fistula [3,4]. UGIB frequently presents with hematemesis or melena. However, a minority of patients can present with hematochezia [2]. GI bleeding (GIB) is either overt or occult. Patients with overt GIB present with signs of visible bleeding such as hematemesis, hematochezia, or melena. Patients with occult GIB have guaiac-positive stools or iron deficiency anemia, without visible blood loss. Obscure GIB refers to bleeding with unknown source despite complete GI tract imaging and endoscopic evaluation [2,5]. Clinically, obscure GIB can be overt (manifests as continued passage of visible blood) or occult (no visible blood). Special Imaging Considerations Multiphase CT is used in the evaluation of patients with overt GIB, including noncontrast, late arterial, and venous phases of contrast administration [5]. With dual-energy CT, virtual noncontrast images can replace a true noncontrast acquisition [11]. Because the virtual noncontrast images are derived from the contrast-enhanced data set, these images are perfectly aligned, which can aid interpretation. The conspicuity of active GIB can also be increased by the use of low-keV virtual monoenergetic or iodine-only images [12]. In a study on the use of dual- energy CT imaging for patients presenting with clinical overt GIB, dual-energy CT images improved the radiologist | Nonvariceal Upper Gastrointestinal Bleeding. Introduction/Background Upper gastrointestinal (GI) bleeding (UGIB) refers to bleeding occurring proximal to the ligament of Treitz, from the esophagus, stomach, or duodenum [1,2]. The incidence of nonvariceal UGIB is almost 5 times higher than that of variceal UGIB [1]. Peptic ulcer disease caused by Helicobacter pylori infection or nonsteroidal anti-inflammatory drug use is the most common cause of nonvariceal UGIB. The other causes of nonvariceal UGIB include Mallory- Weiss tears, esophagitis, pancreatitis, trauma, iatrogenic, or neoplastic [1]. Some rare causes of nonvariceal UGIB include hemobilia, hemosuccus pancreaticus, and aortoenteric fistula [3,4]. UGIB frequently presents with hematemesis or melena. However, a minority of patients can present with hematochezia [2]. GI bleeding (GIB) is either overt or occult. Patients with overt GIB present with signs of visible bleeding such as hematemesis, hematochezia, or melena. Patients with occult GIB have guaiac-positive stools or iron deficiency anemia, without visible blood loss. Obscure GIB refers to bleeding with unknown source despite complete GI tract imaging and endoscopic evaluation [2,5]. Clinically, obscure GIB can be overt (manifests as continued passage of visible blood) or occult (no visible blood). Special Imaging Considerations Multiphase CT is used in the evaluation of patients with overt GIB, including noncontrast, late arterial, and venous phases of contrast administration [5]. With dual-energy CT, virtual noncontrast images can replace a true noncontrast acquisition [11]. Because the virtual noncontrast images are derived from the contrast-enhanced data set, these images are perfectly aligned, which can aid interpretation. The conspicuity of active GIB can also be increased by the use of low-keV virtual monoenergetic or iodine-only images [12]. In a study on the use of dual- energy CT imaging for patients presenting with clinical overt GIB, dual-energy CT images improved the radiologist | 69413 |
acrac_69413_1 | Nonvariceal Upper Gastrointestinal Bleeding | aUniversity of Wisconsin School of Medicine and Public Health, Madison, Wisconsin. bNYU Grossman School of Medicine, New York, New York. cPanel Chair, Brigham & Women's Hospital, Boston, Massachusetts. dPanel Chair, University of California San Diego, San Diego, California. ePanel Vice-Chair, Massachusetts General Hospital and Harvard Medical School, Boston, Massachusetts. fGlobal Advanced Imaging, PLLC, Little Rock, Arkansas; Commission on Nuclear Medicine and Molecular Imaging. gUniversity of Texas Health Science Center at Houston and McGovern Medical School, Houston, Texas; American Gastroenterological Association. hMayo Clinic, Rochester, Minnesota; Society for Cardiovascular Magnetic Resonance. iOchsner Hospital, Baton Rouge, Louisiana. jUMass Memorial Health and UMass Chan Medical School, Worcester, Massachusetts; Committee on Emergency Radiology- GSER. kUniversity of North Carolina at Chapel Hill, Chapel Hill, North Carolina. lVA Puget Sound Health Care System and University of Washington, Seattle, Washington. mUniversity of California San Diego, San Diego, California. nSpecialty Chair, University of Wisconsin School of Medicine and Public Health, Madison, Wisconsin. oSpecialty Chair, Brigham & Women's Hospital, Boston, Massachusetts. Reprint requests to: [email protected] Nonvariceal Upper Gastrointestinal Bleeding confidence in appropriate diagnosis, especially in patients without findings of bleeding on CT [12]. Additionally, low-keV images can be used to reduce the volume of intravenous (IV) contrast used. However, the use of these virtual unenhanced images in place of true unenhanced images is still limited to certain sites and remains a user- specific preference [5]. Hence, considerations specific to virtual noncontrast images are not discussed in this document. All elements are essential: 1) timing, 2) reconstructions/reformats, and 3) 3-D renderings. Standard CTs with contrast also include timing issues and reconstructions/reformats. | Nonvariceal Upper Gastrointestinal Bleeding. aUniversity of Wisconsin School of Medicine and Public Health, Madison, Wisconsin. bNYU Grossman School of Medicine, New York, New York. cPanel Chair, Brigham & Women's Hospital, Boston, Massachusetts. dPanel Chair, University of California San Diego, San Diego, California. ePanel Vice-Chair, Massachusetts General Hospital and Harvard Medical School, Boston, Massachusetts. fGlobal Advanced Imaging, PLLC, Little Rock, Arkansas; Commission on Nuclear Medicine and Molecular Imaging. gUniversity of Texas Health Science Center at Houston and McGovern Medical School, Houston, Texas; American Gastroenterological Association. hMayo Clinic, Rochester, Minnesota; Society for Cardiovascular Magnetic Resonance. iOchsner Hospital, Baton Rouge, Louisiana. jUMass Memorial Health and UMass Chan Medical School, Worcester, Massachusetts; Committee on Emergency Radiology- GSER. kUniversity of North Carolina at Chapel Hill, Chapel Hill, North Carolina. lVA Puget Sound Health Care System and University of Washington, Seattle, Washington. mUniversity of California San Diego, San Diego, California. nSpecialty Chair, University of Wisconsin School of Medicine and Public Health, Madison, Wisconsin. oSpecialty Chair, Brigham & Women's Hospital, Boston, Massachusetts. Reprint requests to: [email protected] Nonvariceal Upper Gastrointestinal Bleeding confidence in appropriate diagnosis, especially in patients without findings of bleeding on CT [12]. Additionally, low-keV images can be used to reduce the volume of intravenous (IV) contrast used. However, the use of these virtual unenhanced images in place of true unenhanced images is still limited to certain sites and remains a user- specific preference [5]. Hence, considerations specific to virtual noncontrast images are not discussed in this document. All elements are essential: 1) timing, 2) reconstructions/reformats, and 3) 3-D renderings. Standard CTs with contrast also include timing issues and reconstructions/reformats. | 69413 |
acrac_69413_2 | Nonvariceal Upper Gastrointestinal Bleeding | Only in CTA, however, is 3-D rendering a required element. This corresponds to the definitions that the CMS has applied to the Current Procedural Terminology codes. OR Discussion of Procedures by Variant Variant 1: Adult. Suspected nonvariceal upper gastrointestinal bleeding; no endoscopy performed. Initial imaging. Endoscopy is the usual first test in patients presenting with overt or occult UGIB [6]. This variant is applicable to a clinical scenario in which a patient presents clinically with overt UGIB and initial endoscopy was not performed due to large volume bleeding or clinical instability. CT Abdomen and Pelvis With IV Contrast There is no significant literature supporting the use of CT abdomen and pelvis with IV contrast. (Note: CTA is a separate procedure distinct from CT abdomen and pelvis with IV contrast. ) CT Abdomen and Pelvis Without and With IV Contrast There is no significant literature supporting the use of CT abdomen and pelvis without and with IV contrast. (Note: CTA is a separate procedure distinct from CT abdomen and pelvis with IV contrast. ) Nonvariceal Upper Gastrointestinal Bleeding CT Abdomen and Pelvis Without IV Contrast There is no significant literature supporting the use of CT abdomen and pelvis without IV contrast. (Note: CTA is a separate procedure distinct from CT abdomen and pelvis with IV contrast. ) CT Abdomen With IV Contrast There is no significant literature supporting the use of CT abdomen with IV contrast. (Note: CTA is a separate procedure distinct from CT abdomen and pelvis with IV contrast. ) CT Abdomen Without and With IV Contrast There is no significant literature supporting the use of CT abdomen without and with IV contrast. (Note: CTA is a separate procedure distinct from CT abdomen and pelvis with IV contrast. ) CT Abdomen Without IV Contrast There is no significant literature supporting the use of CT abdomen without IV contrast. (Note: CTA is a separate procedure distinct from CT abdomen and pelvis with IV contrast. | Nonvariceal Upper Gastrointestinal Bleeding. Only in CTA, however, is 3-D rendering a required element. This corresponds to the definitions that the CMS has applied to the Current Procedural Terminology codes. OR Discussion of Procedures by Variant Variant 1: Adult. Suspected nonvariceal upper gastrointestinal bleeding; no endoscopy performed. Initial imaging. Endoscopy is the usual first test in patients presenting with overt or occult UGIB [6]. This variant is applicable to a clinical scenario in which a patient presents clinically with overt UGIB and initial endoscopy was not performed due to large volume bleeding or clinical instability. CT Abdomen and Pelvis With IV Contrast There is no significant literature supporting the use of CT abdomen and pelvis with IV contrast. (Note: CTA is a separate procedure distinct from CT abdomen and pelvis with IV contrast. ) CT Abdomen and Pelvis Without and With IV Contrast There is no significant literature supporting the use of CT abdomen and pelvis without and with IV contrast. (Note: CTA is a separate procedure distinct from CT abdomen and pelvis with IV contrast. ) Nonvariceal Upper Gastrointestinal Bleeding CT Abdomen and Pelvis Without IV Contrast There is no significant literature supporting the use of CT abdomen and pelvis without IV contrast. (Note: CTA is a separate procedure distinct from CT abdomen and pelvis with IV contrast. ) CT Abdomen With IV Contrast There is no significant literature supporting the use of CT abdomen with IV contrast. (Note: CTA is a separate procedure distinct from CT abdomen and pelvis with IV contrast. ) CT Abdomen Without and With IV Contrast There is no significant literature supporting the use of CT abdomen without and with IV contrast. (Note: CTA is a separate procedure distinct from CT abdomen and pelvis with IV contrast. ) CT Abdomen Without IV Contrast There is no significant literature supporting the use of CT abdomen without IV contrast. (Note: CTA is a separate procedure distinct from CT abdomen and pelvis with IV contrast. | 69413 |
acrac_69413_3 | Nonvariceal Upper Gastrointestinal Bleeding | ) CT Enterography There is no significant literature supporting the use of CT enterography in patients with overt UGIB. CT enterography requires the administration of large volumes of neutral oral contrast, which can mask GIB by dilution. Additionally, a large volume of oral contrast is often not tolerated by acutely ill patients. CT enterography is recommended by the Society of Abdominal Radiology Gastrointestinal Bleeding Disease-Focused Panel for patients with occult GIB or suspected small bowel bleeding, which often uses a multiphase technique [5]. CTA Abdomen and Pelvis With IV Contrast There is no significant literature supporting the use of CTA abdomen and pelvis with IV contrast only. The literature primarily reflects studies with CTA examinations without and with IV contrast. Dual-energy CT allows for the generation of virtual noncontrast images from a CTA data set. However, the use of these virtual unenhanced images in place of true unenhanced images is still limited to certain sites and remains a user-specific preference. So, they are not discussed in this document. CTA Abdomen and Pelvis Without and With IV Contrast CTA without and with IV contrast can help with the detection of a source for GIB [7,18]. CTA has been shown to be able to detect bleeding rates as slow as 0.3 mL/min, compared with 0.5 to 1.0 mL/min for conventional angiography and 0.2 mL/min for Tc-99m-labeled red blood cell (RBC) scintigraphy [19]. Faster acquisition, thin collimation, and greater availability have led to greater utilization of this study. The noncontrast images are useful for the detection of intraluminal high-attenuation material that may mimic intraluminal blood on contrast-enhanced images and may be necessary for the identification of sentinel clot [7]. The Society of Abdominal Radiology Gastrointestinal Bleeding Disease-Focused Panel published a consensus on CT imaging protocols for the detection of overt GIB [20]. | Nonvariceal Upper Gastrointestinal Bleeding. ) CT Enterography There is no significant literature supporting the use of CT enterography in patients with overt UGIB. CT enterography requires the administration of large volumes of neutral oral contrast, which can mask GIB by dilution. Additionally, a large volume of oral contrast is often not tolerated by acutely ill patients. CT enterography is recommended by the Society of Abdominal Radiology Gastrointestinal Bleeding Disease-Focused Panel for patients with occult GIB or suspected small bowel bleeding, which often uses a multiphase technique [5]. CTA Abdomen and Pelvis With IV Contrast There is no significant literature supporting the use of CTA abdomen and pelvis with IV contrast only. The literature primarily reflects studies with CTA examinations without and with IV contrast. Dual-energy CT allows for the generation of virtual noncontrast images from a CTA data set. However, the use of these virtual unenhanced images in place of true unenhanced images is still limited to certain sites and remains a user-specific preference. So, they are not discussed in this document. CTA Abdomen and Pelvis Without and With IV Contrast CTA without and with IV contrast can help with the detection of a source for GIB [7,18]. CTA has been shown to be able to detect bleeding rates as slow as 0.3 mL/min, compared with 0.5 to 1.0 mL/min for conventional angiography and 0.2 mL/min for Tc-99m-labeled red blood cell (RBC) scintigraphy [19]. Faster acquisition, thin collimation, and greater availability have led to greater utilization of this study. The noncontrast images are useful for the detection of intraluminal high-attenuation material that may mimic intraluminal blood on contrast-enhanced images and may be necessary for the identification of sentinel clot [7]. The Society of Abdominal Radiology Gastrointestinal Bleeding Disease-Focused Panel published a consensus on CT imaging protocols for the detection of overt GIB [20]. | 69413 |
acrac_69413_4 | Nonvariceal Upper Gastrointestinal Bleeding | With a 100% consensus, the experts recommended noncontrast images for CTA performed on single-energy CT. This can be replaced by virtual noncontrast reconstructions with dual-energy CT [20,21]. In a meta-analysis of 22 studies evaluating accuracy of CTA for the diagnosis of active GIB (total of 672 patients), CTA had a sensitivity and specificity of 85% and 92%, respectively. Frequently, multiphase acquisition is performed, with a portal venous or a delayed phase (typically acquired during 70-90 sec window after the initiation of the contrast bolus injection) in addition to the angiographic phase [7,20]. In a study evaluating different CT imaging protocols, multiphasic CT protocols (unenhanced + arterial + portal venous phase) had the highest sensitivity of 92% for the detection of GIB compared with 83% for unenhanced phase with arterial or portal venous phase alone [18]. Oral contrast is usually not given for GIB studies because a positive oral contrast will render the examination nondiagnostic, and oral water can dilute intraluminal hemorrhage [7]. In an 8-year follow-up study after initial negative CT for suspected GIB, nearly 60% of patients with suspected UGIB, and nearly 77% patients with suspected lower GIB did not rebleed, suggesting relatively higher odds of rebleeding for UGIB compared with lower GIB despite a negative CTA [22]. In a retrospective study among patients with positive CTA for GIB, greater contrast extravasation volume on CT was significantly correlated with use of hemostatic therapy, intraprocedural active bleeding, and massive transfusion. The extravasation volume, however, did not correlate with patient mortality [23]. CTA may still be underused for the diagnosis of GIB [24]. Compared with Tc-99m-labeled RBC scintigraphy, CTA can lead to faster triage of patients toward definitive treatment by angiography [25]. | Nonvariceal Upper Gastrointestinal Bleeding. With a 100% consensus, the experts recommended noncontrast images for CTA performed on single-energy CT. This can be replaced by virtual noncontrast reconstructions with dual-energy CT [20,21]. In a meta-analysis of 22 studies evaluating accuracy of CTA for the diagnosis of active GIB (total of 672 patients), CTA had a sensitivity and specificity of 85% and 92%, respectively. Frequently, multiphase acquisition is performed, with a portal venous or a delayed phase (typically acquired during 70-90 sec window after the initiation of the contrast bolus injection) in addition to the angiographic phase [7,20]. In a study evaluating different CT imaging protocols, multiphasic CT protocols (unenhanced + arterial + portal venous phase) had the highest sensitivity of 92% for the detection of GIB compared with 83% for unenhanced phase with arterial or portal venous phase alone [18]. Oral contrast is usually not given for GIB studies because a positive oral contrast will render the examination nondiagnostic, and oral water can dilute intraluminal hemorrhage [7]. In an 8-year follow-up study after initial negative CT for suspected GIB, nearly 60% of patients with suspected UGIB, and nearly 77% patients with suspected lower GIB did not rebleed, suggesting relatively higher odds of rebleeding for UGIB compared with lower GIB despite a negative CTA [22]. In a retrospective study among patients with positive CTA for GIB, greater contrast extravasation volume on CT was significantly correlated with use of hemostatic therapy, intraprocedural active bleeding, and massive transfusion. The extravasation volume, however, did not correlate with patient mortality [23]. CTA may still be underused for the diagnosis of GIB [24]. Compared with Tc-99m-labeled RBC scintigraphy, CTA can lead to faster triage of patients toward definitive treatment by angiography [25]. | 69413 |
acrac_69413_5 | Nonvariceal Upper Gastrointestinal Bleeding | Likewise, the use of CTA as the first test leads to faster triage of patients in the emergency room when compared with endoscopy for GI bleed [24]. Nonvariceal Upper Gastrointestinal Bleeding CTA Abdomen With IV Contrast Although CTA can detect GIB, typically both the abdomen and pelvis are imaged, because the site of bleeding is unclear without endoscopy. There is no significant literature supporting the use of CTA abdomen with IV contrast only. CTA Abdomen Without and With IV Contrast Although CTA can detect GIB, typically both the abdomen and pelvis are imaged, because the site of bleeding is unclear without endoscopy. There is no significant literature supporting the use of CTA abdomen without and with IV contrast only. CTA Chest With IV Contrast Although esophageal bleed (which can be localized on CTA chest) can present as UGIB, the literature supports CTA of abdomen and pelvis if the site of bleeding is unclear without endoscopy. There is no significant literature supporting the use of CTA chest with IV contrast. CTA Chest Without and With IV Contrast Although esophageal bleed (which can be localized to CTA chest) can present as UGIB, the literature supports CTA of abdomen and pelvis without and with IV contrast, if the site of bleeding is unclear without endoscopy. There is no significant literature supporting the use of CTA chest without and with IV contrast. Fluoroscopy Upper GI Series Barium or iodine upper GI series has no role in the diagnosis of acute UGIB in a modern-day practice. MR Enterography There is no significant literature supporting the use of MR enterography in patients with overt UGIB. RBC Scan Abdomen and Pelvis There is no significant literature to support the use of a Tc-99m-labeled RBC scan of the abdomen and pelvis for the initial imaging of suspected overt UGIB (without endoscopy performed). Variant 2: Adult. | Nonvariceal Upper Gastrointestinal Bleeding. Likewise, the use of CTA as the first test leads to faster triage of patients in the emergency room when compared with endoscopy for GI bleed [24]. Nonvariceal Upper Gastrointestinal Bleeding CTA Abdomen With IV Contrast Although CTA can detect GIB, typically both the abdomen and pelvis are imaged, because the site of bleeding is unclear without endoscopy. There is no significant literature supporting the use of CTA abdomen with IV contrast only. CTA Abdomen Without and With IV Contrast Although CTA can detect GIB, typically both the abdomen and pelvis are imaged, because the site of bleeding is unclear without endoscopy. There is no significant literature supporting the use of CTA abdomen without and with IV contrast only. CTA Chest With IV Contrast Although esophageal bleed (which can be localized on CTA chest) can present as UGIB, the literature supports CTA of abdomen and pelvis if the site of bleeding is unclear without endoscopy. There is no significant literature supporting the use of CTA chest with IV contrast. CTA Chest Without and With IV Contrast Although esophageal bleed (which can be localized to CTA chest) can present as UGIB, the literature supports CTA of abdomen and pelvis without and with IV contrast, if the site of bleeding is unclear without endoscopy. There is no significant literature supporting the use of CTA chest without and with IV contrast. Fluoroscopy Upper GI Series Barium or iodine upper GI series has no role in the diagnosis of acute UGIB in a modern-day practice. MR Enterography There is no significant literature supporting the use of MR enterography in patients with overt UGIB. RBC Scan Abdomen and Pelvis There is no significant literature to support the use of a Tc-99m-labeled RBC scan of the abdomen and pelvis for the initial imaging of suspected overt UGIB (without endoscopy performed). Variant 2: Adult. | 69413 |
acrac_69413_6 | Nonvariceal Upper Gastrointestinal Bleeding | Endoscopy confirms nonvariceal upper gastrointestinal bleeding with a clear source, but treatment not possible or continued bleeding after endoscopic treatment. Initial imaging. This variant is applicable in a clinical scenario when the patient had endoscopy performed, which diagnosed the upper GI tract as the source of bleed, but definitive treatment of the bleeding was not possible or there is continued bleeding after treatment. Arteriography Visceral There is an 88% to 100% success rate of VA for diagnosis of endoscopically refractory bleeding from the esophagus due to inflammatory or neoplastic pathology [26,27] or periesophageal pseudoaneurysm [28]. In a study among patients with postesophagectomy gastric conduit hemorrhage [29], VA identified the source of active bleeding in 85% of patients. In a study among patients with gastric cancer refractory to endoscopy, angiography could diagnose active bleeding in 22.4% (13/58) of patients [30]. In other studies in patients with bleeding GI tumors, VA showed active bleeding in 25% to 55% of patients [31,32]. Among patients with pancreatic cancer presenting with endoscopically refractory UGIB, angiography detected active bleeding in 81% of patients [33,34]. Although there is more literature on clinical success of angiographic embolization for treatment of gastric and duodenal ulcers refractory to endoscopic treatment [34-37], duodenal fistula [38], and iatrogenic bleeding following laparoscopic cholecystectomy or endoscopic pancreaticobiliary drainage [4,39,40], the literature on diagnostic accuracy for bleeding is limited in these studies. CT Abdomen and Pelvis With IV Contrast There is no significant literature supporting the use of CT abdomen and pelvis with IV contrast. (Note: CTA is a separate procedure distinct from CT abdomen and pelvis with IV contrast). CT Abdomen and Pelvis Without and With IV Contrast There is no significant literature supporting the use of CT abdomen and pelvis without and with IV contrast. | Nonvariceal Upper Gastrointestinal Bleeding. Endoscopy confirms nonvariceal upper gastrointestinal bleeding with a clear source, but treatment not possible or continued bleeding after endoscopic treatment. Initial imaging. This variant is applicable in a clinical scenario when the patient had endoscopy performed, which diagnosed the upper GI tract as the source of bleed, but definitive treatment of the bleeding was not possible or there is continued bleeding after treatment. Arteriography Visceral There is an 88% to 100% success rate of VA for diagnosis of endoscopically refractory bleeding from the esophagus due to inflammatory or neoplastic pathology [26,27] or periesophageal pseudoaneurysm [28]. In a study among patients with postesophagectomy gastric conduit hemorrhage [29], VA identified the source of active bleeding in 85% of patients. In a study among patients with gastric cancer refractory to endoscopy, angiography could diagnose active bleeding in 22.4% (13/58) of patients [30]. In other studies in patients with bleeding GI tumors, VA showed active bleeding in 25% to 55% of patients [31,32]. Among patients with pancreatic cancer presenting with endoscopically refractory UGIB, angiography detected active bleeding in 81% of patients [33,34]. Although there is more literature on clinical success of angiographic embolization for treatment of gastric and duodenal ulcers refractory to endoscopic treatment [34-37], duodenal fistula [38], and iatrogenic bleeding following laparoscopic cholecystectomy or endoscopic pancreaticobiliary drainage [4,39,40], the literature on diagnostic accuracy for bleeding is limited in these studies. CT Abdomen and Pelvis With IV Contrast There is no significant literature supporting the use of CT abdomen and pelvis with IV contrast. (Note: CTA is a separate procedure distinct from CT abdomen and pelvis with IV contrast). CT Abdomen and Pelvis Without and With IV Contrast There is no significant literature supporting the use of CT abdomen and pelvis without and with IV contrast. | 69413 |
acrac_69413_7 | Nonvariceal Upper Gastrointestinal Bleeding | (Note: CTA is a separate procedure distinct from CT abdomen and pelvis without and with IV contrast). CT Abdomen and Pelvis Without IV Contrast There is no significant literature supporting the use of CT abdomen and pelvis without IV contrast. (Note: CTA is a separate procedure distinct from CT abdomen and pelvis without IV contrast). Nonvariceal Upper Gastrointestinal Bleeding CT Abdomen With IV Contrast There is no significant literature supporting the use of CT abdomen with IV contrast. (Note: CTA is a separate procedure distinct from CT abdomen with IV contrast). CT Abdomen Without and With IV Contrast There is no significant literature supporting the use of CT abdomen without and with IV contrast. (Note: CTA is a separate procedure distinct from CT abdomen without and with IV contrast). CT Abdomen Without IV Contrast There is no significant literature supporting the use of CT abdomen without IV contrast. (Note: CTA is a separate procedure distinct from CT abdomen without IV contrast). CT Enterography There is no significant literature supporting the use of CT enterography as an imaging test for overt UGIB, which is untreatable by endoscopy. Although multiphasic CT enterography can detect bleeding because of technique parameters that can mirror CTA, the primary use of this protocol is directed more at finding a potential bleeding source when bleeding is of a slow rate [41] in patients with occult GI or suspected small bowel bleeding [5]. CTA Abdomen and Pelvis With IV Contrast Although CTA can detect overt GIB, performance with IV contrast only may not be as helpful. There is no significant literature supporting the use of CTA abdomen and pelvis with IV contrast. Noncontrast images can aid in the identification of a sentinel clot. Dual-energy CT allows for the generation of virtual noncontrast images from a CTA data set. | Nonvariceal Upper Gastrointestinal Bleeding. (Note: CTA is a separate procedure distinct from CT abdomen and pelvis without and with IV contrast). CT Abdomen and Pelvis Without IV Contrast There is no significant literature supporting the use of CT abdomen and pelvis without IV contrast. (Note: CTA is a separate procedure distinct from CT abdomen and pelvis without IV contrast). Nonvariceal Upper Gastrointestinal Bleeding CT Abdomen With IV Contrast There is no significant literature supporting the use of CT abdomen with IV contrast. (Note: CTA is a separate procedure distinct from CT abdomen with IV contrast). CT Abdomen Without and With IV Contrast There is no significant literature supporting the use of CT abdomen without and with IV contrast. (Note: CTA is a separate procedure distinct from CT abdomen without and with IV contrast). CT Abdomen Without IV Contrast There is no significant literature supporting the use of CT abdomen without IV contrast. (Note: CTA is a separate procedure distinct from CT abdomen without IV contrast). CT Enterography There is no significant literature supporting the use of CT enterography as an imaging test for overt UGIB, which is untreatable by endoscopy. Although multiphasic CT enterography can detect bleeding because of technique parameters that can mirror CTA, the primary use of this protocol is directed more at finding a potential bleeding source when bleeding is of a slow rate [41] in patients with occult GI or suspected small bowel bleeding [5]. CTA Abdomen and Pelvis With IV Contrast Although CTA can detect overt GIB, performance with IV contrast only may not be as helpful. There is no significant literature supporting the use of CTA abdomen and pelvis with IV contrast. Noncontrast images can aid in the identification of a sentinel clot. Dual-energy CT allows for the generation of virtual noncontrast images from a CTA data set. | 69413 |
acrac_69413_8 | Nonvariceal Upper Gastrointestinal Bleeding | However, the use of these virtual unenhanced images in place of true unenhanced images is still limited to certain sites and remains a user-specific preference. So, they are not discussed in this document. CTA Abdomen and Pelvis Without and With IV Contrast Although there is no literature to discuss the difference between diagnostic accuracy of CTA abdomen and pelvis without and with IV contrast versus CTA abdomen without and with IV contrast or CTA chest without and with IV contrast, in patients with a known source of bleeding that cannot be controlled endoscopically, the imaging should be tailored to include the site of the bleeding. Overall, CTA has good accuracy for the detection of tumor bleeding, vascular malformations, or diverticular bleeding [7,42,43]. CTA Abdomen With IV Contrast There is no significant literature supporting the use of CTA abdomen with IV contrast. CTA Abdomen Without and With IV Contrast There is no significant literature comparing the use of CTA abdomen and pelvis without and with IV contrast for UGIB. However, as highlighted earlier in this document, if the known source of bleeding if localized to abdomen, this test may be helpful. CTA pelvis may be ordered simultaneously; however, the primary role would be for vascular access mapping. CTA Chest With IV Contrast There is no significant literature supporting the use of CTA chest with IV contrast. CTA Chest Without and With IV Contrast There is no significant literature comparing the use of CTA chest without and with IV contrast for UGIB. However, as highlighted earlier in this document, if the known source of bleeding is localized to the chest, this test may be helpful. Fluoroscopy Upper GI Series Barium or iodine upper GI series has no role in the diagnosis of acute UGIB. MR Enterography There is no significant literature supporting the use of MR enterography as an imaging test for overt UGIB, which is untreatable by endoscopy. | Nonvariceal Upper Gastrointestinal Bleeding. However, the use of these virtual unenhanced images in place of true unenhanced images is still limited to certain sites and remains a user-specific preference. So, they are not discussed in this document. CTA Abdomen and Pelvis Without and With IV Contrast Although there is no literature to discuss the difference between diagnostic accuracy of CTA abdomen and pelvis without and with IV contrast versus CTA abdomen without and with IV contrast or CTA chest without and with IV contrast, in patients with a known source of bleeding that cannot be controlled endoscopically, the imaging should be tailored to include the site of the bleeding. Overall, CTA has good accuracy for the detection of tumor bleeding, vascular malformations, or diverticular bleeding [7,42,43]. CTA Abdomen With IV Contrast There is no significant literature supporting the use of CTA abdomen with IV contrast. CTA Abdomen Without and With IV Contrast There is no significant literature comparing the use of CTA abdomen and pelvis without and with IV contrast for UGIB. However, as highlighted earlier in this document, if the known source of bleeding if localized to abdomen, this test may be helpful. CTA pelvis may be ordered simultaneously; however, the primary role would be for vascular access mapping. CTA Chest With IV Contrast There is no significant literature supporting the use of CTA chest with IV contrast. CTA Chest Without and With IV Contrast There is no significant literature comparing the use of CTA chest without and with IV contrast for UGIB. However, as highlighted earlier in this document, if the known source of bleeding is localized to the chest, this test may be helpful. Fluoroscopy Upper GI Series Barium or iodine upper GI series has no role in the diagnosis of acute UGIB. MR Enterography There is no significant literature supporting the use of MR enterography as an imaging test for overt UGIB, which is untreatable by endoscopy. | 69413 |
acrac_69413_9 | Nonvariceal Upper Gastrointestinal Bleeding | RBC Scan Abdomen and Pelvis There is no significant literature to support the use of a Tc-99m-labeled RBC scan of the abdomen and pelvis for the diagnosis of GIB when the source of bleeding is obvious but unmanageable on endoscopy. Nonvariceal Upper Gastrointestinal Bleeding Variant 3: Adult. Endoscopy confirms nonvariceal upper gastrointestinal bleeding without a clear source. Initial imaging. This variant is applicable to clinical scenario in which endoscopy shows UGIB but the site or source of the bleeding cannot be determined on endoscopy. Clinically, these patients typically present with overt GIB. Arteriography Visceral VA can be attempted in such cases; however, it also has the limitation that if the bleeding is not active, it may not be seen. VA has the advantage of assessing the entire mesenteric circulation. Provocation techniques including intraprocedure heparin administration, intra-arterial nitroglycerin administration, and low-dose tissue plasminogen activator administration have been shown to increase the sensitivity of VA for the detection of GIB [44]. However, the literature on safety of these maneuvers and the effect on patient outcomes is limited. The other methods described in the literature that can enhance detection of bleeding on VA are glucagon and hyoscine butylbromide administration (to decrease bowel motility and artifacts), the use of carbon dioxide as a contrast medium, and longer injection durations [45]. Limitations include its invasive nature and higher bleeding rate threshold to diagnose obscure bleeding [46]. CT Abdomen and Pelvis With IV Contrast There is no significant literature supporting the use of CT abdomen and pelvis with IV contrast. (Note: CTA is a separate procedure distinct from CT abdomen and pelvis with IV contrast). CT Abdomen and Pelvis Without and With IV Contrast There is no significant literature supporting the use of CT abdomen and pelvis without and with IV contrast. | Nonvariceal Upper Gastrointestinal Bleeding. RBC Scan Abdomen and Pelvis There is no significant literature to support the use of a Tc-99m-labeled RBC scan of the abdomen and pelvis for the diagnosis of GIB when the source of bleeding is obvious but unmanageable on endoscopy. Nonvariceal Upper Gastrointestinal Bleeding Variant 3: Adult. Endoscopy confirms nonvariceal upper gastrointestinal bleeding without a clear source. Initial imaging. This variant is applicable to clinical scenario in which endoscopy shows UGIB but the site or source of the bleeding cannot be determined on endoscopy. Clinically, these patients typically present with overt GIB. Arteriography Visceral VA can be attempted in such cases; however, it also has the limitation that if the bleeding is not active, it may not be seen. VA has the advantage of assessing the entire mesenteric circulation. Provocation techniques including intraprocedure heparin administration, intra-arterial nitroglycerin administration, and low-dose tissue plasminogen activator administration have been shown to increase the sensitivity of VA for the detection of GIB [44]. However, the literature on safety of these maneuvers and the effect on patient outcomes is limited. The other methods described in the literature that can enhance detection of bleeding on VA are glucagon and hyoscine butylbromide administration (to decrease bowel motility and artifacts), the use of carbon dioxide as a contrast medium, and longer injection durations [45]. Limitations include its invasive nature and higher bleeding rate threshold to diagnose obscure bleeding [46]. CT Abdomen and Pelvis With IV Contrast There is no significant literature supporting the use of CT abdomen and pelvis with IV contrast. (Note: CTA is a separate procedure distinct from CT abdomen and pelvis with IV contrast). CT Abdomen and Pelvis Without and With IV Contrast There is no significant literature supporting the use of CT abdomen and pelvis without and with IV contrast. | 69413 |
acrac_69413_10 | Nonvariceal Upper Gastrointestinal Bleeding | (Note: CTA is a separate procedure distinct from CT abdomen and pelvis without and with IV contrast). CT Abdomen and Pelvis Without IV Contrast There is no significant literature supporting the use of CT abdomen and pelvis without IV contrast. (Note: CTA is a separate procedure distinct from CT abdomen and pelvis without IV contrast). CT Abdomen With IV Contrast There is no significant literature supporting the use of CT abdomen with IV contrast. (Note: CTA is a separate procedure distinct from CT abdomen with IV contrast). CT Abdomen Without and With IV Contrast There is no significant literature supporting the use of CT abdomen without and with IV contrast. (Note: CTA is a separate procedure distinct from CT abdomen without and with IV contrast). CT Abdomen Without IV Contrast There is no significant literature supporting the use of CT abdomen without IV contrast. (Note: CTA is a separate procedure distinct from CT abdomen without IV contrast). CT Enterography CT enterography requires the administration of large volumes of neutral oral contrast, which can mask bleeding by dilution. Although multiphasic CT enterography can detect bleeding because of technique parameters that can mirror CTA, the primary use of this protocol is directed more to finding a potential bleeding source when bleeding is of a slow rate, suspected in the small bowel, or occult in nature [41]. CTA Abdomen and Pelvis With IV Contrast There is no significant literature supporting the use of CTA abdomen and pelvis with IV contrast. Dual-energy CT allows for the generation of virtual noncontrast images from a CTA data set. However, the use of these virtual unenhanced images in place of true unenhanced images is still limited to certain sites and remains a user-specific preference. So, they are not discussed in this document. | Nonvariceal Upper Gastrointestinal Bleeding. (Note: CTA is a separate procedure distinct from CT abdomen and pelvis without and with IV contrast). CT Abdomen and Pelvis Without IV Contrast There is no significant literature supporting the use of CT abdomen and pelvis without IV contrast. (Note: CTA is a separate procedure distinct from CT abdomen and pelvis without IV contrast). CT Abdomen With IV Contrast There is no significant literature supporting the use of CT abdomen with IV contrast. (Note: CTA is a separate procedure distinct from CT abdomen with IV contrast). CT Abdomen Without and With IV Contrast There is no significant literature supporting the use of CT abdomen without and with IV contrast. (Note: CTA is a separate procedure distinct from CT abdomen without and with IV contrast). CT Abdomen Without IV Contrast There is no significant literature supporting the use of CT abdomen without IV contrast. (Note: CTA is a separate procedure distinct from CT abdomen without IV contrast). CT Enterography CT enterography requires the administration of large volumes of neutral oral contrast, which can mask bleeding by dilution. Although multiphasic CT enterography can detect bleeding because of technique parameters that can mirror CTA, the primary use of this protocol is directed more to finding a potential bleeding source when bleeding is of a slow rate, suspected in the small bowel, or occult in nature [41]. CTA Abdomen and Pelvis With IV Contrast There is no significant literature supporting the use of CTA abdomen and pelvis with IV contrast. Dual-energy CT allows for the generation of virtual noncontrast images from a CTA data set. However, the use of these virtual unenhanced images in place of true unenhanced images is still limited to certain sites and remains a user-specific preference. So, they are not discussed in this document. | 69413 |
acrac_69413_11 | Nonvariceal Upper Gastrointestinal Bleeding | CTA Abdomen and Pelvis Without and With IV Contrast Besides angiography, CTA abdomen and pelvis without and with IV contrast may become useful when endoscopy shows nonvariceal UGIB without a clear source. As highlighted in previous variants, CTA without and with IV contrast has high accuracy for the detection of GIB. CTA can detect small bowel lesions that may be difficult to see on traditional esophagogastroduodenoscopy but may be better visualized by push enteroscopy [43]. Because CTA can identify a slower bleeding rate than angiography, CTA has been demonstrated to have significantly higher detection of active bleeding as well as localization of the culprit lesion [46]. CTA can also show Dieulafoy lesions that have a very high mortality rate [47,48]. Nonvariceal Upper Gastrointestinal Bleeding CTA Abdomen With IV Contrast There is no significant literature supporting the use of CTA abdomen with IV contrast. CTA Abdomen Without and With IV Contrast There is no significant literature supporting the use of CTA abdomen without and with IV contrast, because lesions not clearly seen on endoscopy may be localized to small bowel, which often requires inclusion of the pelvis. CTA Chest With IV Contrast There is no significant literature supporting the use of CTA chest with IV contrast. CTA Chest Without and With IV Contrast There is no significant literature supporting the use of CTA chest without and with IV contrast. Fluoroscopy Upper GI Series Barium or iodine upper GI series has no role in the diagnosis of acute UGIB. MR Enterography There is no significant literature supporting the use of MR enterography as an imaging test for UGIB, which is confirmed on endoscopy but without a clear source. RBC Scan Abdomen and Pelvis Because CTA evaluation is a well-recognized next step in the management of UGIB if endoscopy confirms bleeding without identification of a source, there is no significant literature outlining the use of nuclear medicine studies in this clinical context. | Nonvariceal Upper Gastrointestinal Bleeding. CTA Abdomen and Pelvis Without and With IV Contrast Besides angiography, CTA abdomen and pelvis without and with IV contrast may become useful when endoscopy shows nonvariceal UGIB without a clear source. As highlighted in previous variants, CTA without and with IV contrast has high accuracy for the detection of GIB. CTA can detect small bowel lesions that may be difficult to see on traditional esophagogastroduodenoscopy but may be better visualized by push enteroscopy [43]. Because CTA can identify a slower bleeding rate than angiography, CTA has been demonstrated to have significantly higher detection of active bleeding as well as localization of the culprit lesion [46]. CTA can also show Dieulafoy lesions that have a very high mortality rate [47,48]. Nonvariceal Upper Gastrointestinal Bleeding CTA Abdomen With IV Contrast There is no significant literature supporting the use of CTA abdomen with IV contrast. CTA Abdomen Without and With IV Contrast There is no significant literature supporting the use of CTA abdomen without and with IV contrast, because lesions not clearly seen on endoscopy may be localized to small bowel, which often requires inclusion of the pelvis. CTA Chest With IV Contrast There is no significant literature supporting the use of CTA chest with IV contrast. CTA Chest Without and With IV Contrast There is no significant literature supporting the use of CTA chest without and with IV contrast. Fluoroscopy Upper GI Series Barium or iodine upper GI series has no role in the diagnosis of acute UGIB. MR Enterography There is no significant literature supporting the use of MR enterography as an imaging test for UGIB, which is confirmed on endoscopy but without a clear source. RBC Scan Abdomen and Pelvis Because CTA evaluation is a well-recognized next step in the management of UGIB if endoscopy confirms bleeding without identification of a source, there is no significant literature outlining the use of nuclear medicine studies in this clinical context. | 69413 |
acrac_69413_12 | Nonvariceal Upper Gastrointestinal Bleeding | However, Tc-99m-labeled RBC scanning can be used to localize a low-rate source of bleeding. Variant 4: Adult. Nonvariceal upper gastrointestinal bleeding; negative endoscopy. Initial imaging. This variant is applicable to patients with no clear source of bleeding despite complete endoscopic evaluation. Clinically, these patients can have obscure bleeding (which may be noted in the form of visible passage of blood or melena or occult bleeding, unexplained iron deficiency anemia, or guaiac-positive stools without visible passage of blood). Small bowel pathology is the frequent source of bleeding in these patients. Arteriography Visceral VA, due to a lower sensitivity for the detection of bleeding and its invasive nature, is considered lower on the diagnostic algorithm compared with more sensitive noninvasive testing [49,50]. Super-selective angiography with intraoperative methylene blue localization may help to diagnose and effectively control bleeding in patients with obscure GIB [51]. In patients with negative endoscopy, VA has false-negative results, because the bleeding is typically detected when the rate is at least 0.5 mL/min. It has been shown that there is no benefit of performing angiography in patients with occult GIB who have a negative CTA study [50]. CT Abdomen and Pelvis With IV Contrast In endoscopy-negative patients, there is no significant literature for CT abdomen and pelvis with IV contrast for GIB evaluation. However, CT may be performed primarily for the evaluation of GI masses rather than the demonstration of bleeding. (Note: CTA is a separate procedure distinct from CT abdomen and pelvis with IV contrast). CT Abdomen and Pelvis Without and With IV Contrast In endoscopy-negative patients, there is no significant literature for CT abdomen and pelvis without and with IV contrast for GIB evaluation. | Nonvariceal Upper Gastrointestinal Bleeding. However, Tc-99m-labeled RBC scanning can be used to localize a low-rate source of bleeding. Variant 4: Adult. Nonvariceal upper gastrointestinal bleeding; negative endoscopy. Initial imaging. This variant is applicable to patients with no clear source of bleeding despite complete endoscopic evaluation. Clinically, these patients can have obscure bleeding (which may be noted in the form of visible passage of blood or melena or occult bleeding, unexplained iron deficiency anemia, or guaiac-positive stools without visible passage of blood). Small bowel pathology is the frequent source of bleeding in these patients. Arteriography Visceral VA, due to a lower sensitivity for the detection of bleeding and its invasive nature, is considered lower on the diagnostic algorithm compared with more sensitive noninvasive testing [49,50]. Super-selective angiography with intraoperative methylene blue localization may help to diagnose and effectively control bleeding in patients with obscure GIB [51]. In patients with negative endoscopy, VA has false-negative results, because the bleeding is typically detected when the rate is at least 0.5 mL/min. It has been shown that there is no benefit of performing angiography in patients with occult GIB who have a negative CTA study [50]. CT Abdomen and Pelvis With IV Contrast In endoscopy-negative patients, there is no significant literature for CT abdomen and pelvis with IV contrast for GIB evaluation. However, CT may be performed primarily for the evaluation of GI masses rather than the demonstration of bleeding. (Note: CTA is a separate procedure distinct from CT abdomen and pelvis with IV contrast). CT Abdomen and Pelvis Without and With IV Contrast In endoscopy-negative patients, there is no significant literature for CT abdomen and pelvis without and with IV contrast for GIB evaluation. | 69413 |
acrac_69413_13 | Nonvariceal Upper Gastrointestinal Bleeding | Although there is not enough literature supporting its use, the test may serve as an initial test for patients with obscure or occult bleeding, especially if an alternative is needed for CTA or enterography. (Note: CTA is a separate procedure distinct from CT abdomen and pelvis without and with IV contrast). CT Abdomen and Pelvis Without IV Contrast In endoscopy-negative patients, there is no significant literature for CT abdomen and pelvis without IV contrast for GIB evaluation. (Note: CTA is a separate procedure distinct from CT abdomen and pelvis without IV contrast). CT Abdomen With IV Contrast In endoscopy-negative patients, there is no significant literature for CT abdomen with IV contrast for GIB evaluation. (Note: CTA is a separate procedure distinct from CT abdomen with IV contrast). Nonvariceal Upper Gastrointestinal Bleeding CT Abdomen Without and With IV Contrast In endoscopy-negative patients, there is no significant literature for CT abdomen without and with IV contrast for GIB evaluation. (Note: CTA is a separate procedure distinct from CT abdomen without and with IV contrast). CT Abdomen Without IV Contrast In endoscopy-negative patients, there is no significant literature for CT abdomen without IV contrast for GIB evaluation. (Note: CTA is a separate procedure distinct from CT abdomen without IV contrast). CT Enterography Although CT enterography is not considered useful for detection of acute bleeding due to the dilution of bleeding from oral contrast, it may be helpful in patients with negative endoscopy to identify a small bowel source of GIB. The sensitivity for detection of the cause of occult GIB is typically low, with measured sensitivity of 25% compared with capsule endoscopy (sensitivity of 87%) and 33% when compared with clinical follow-up in patients with nondiagnostic capsule endoscopy [52,53]. | Nonvariceal Upper Gastrointestinal Bleeding. Although there is not enough literature supporting its use, the test may serve as an initial test for patients with obscure or occult bleeding, especially if an alternative is needed for CTA or enterography. (Note: CTA is a separate procedure distinct from CT abdomen and pelvis without and with IV contrast). CT Abdomen and Pelvis Without IV Contrast In endoscopy-negative patients, there is no significant literature for CT abdomen and pelvis without IV contrast for GIB evaluation. (Note: CTA is a separate procedure distinct from CT abdomen and pelvis without IV contrast). CT Abdomen With IV Contrast In endoscopy-negative patients, there is no significant literature for CT abdomen with IV contrast for GIB evaluation. (Note: CTA is a separate procedure distinct from CT abdomen with IV contrast). Nonvariceal Upper Gastrointestinal Bleeding CT Abdomen Without and With IV Contrast In endoscopy-negative patients, there is no significant literature for CT abdomen without and with IV contrast for GIB evaluation. (Note: CTA is a separate procedure distinct from CT abdomen without and with IV contrast). CT Abdomen Without IV Contrast In endoscopy-negative patients, there is no significant literature for CT abdomen without IV contrast for GIB evaluation. (Note: CTA is a separate procedure distinct from CT abdomen without IV contrast). CT Enterography Although CT enterography is not considered useful for detection of acute bleeding due to the dilution of bleeding from oral contrast, it may be helpful in patients with negative endoscopy to identify a small bowel source of GIB. The sensitivity for detection of the cause of occult GIB is typically low, with measured sensitivity of 25% compared with capsule endoscopy (sensitivity of 87%) and 33% when compared with clinical follow-up in patients with nondiagnostic capsule endoscopy [52,53]. | 69413 |
acrac_69413_14 | Nonvariceal Upper Gastrointestinal Bleeding | However, a study showed a higher sensitivity of CT enterography (88%) for the diagnosis of etiology of occult GIB, significantly higher than capsule endoscopy (sensitivity 38%) in their study population [54]. Advantages of CT enterography over capsule endoscopy are ease of performance and better patient satisfaction [5]. CTA Abdomen and Pelvis With IV Contrast In endoscopy-negative patients, there is no significant literature for CT abdomen and pelvis with IV contrast for GIB evaluation. Dual-energy CT allows for the generation of virtual noncontrast images from a CTA data set. However, the use of these virtual unenhanced images in place of true unenhanced images is still limited to certain sites and remains a user-specific preference. So, they are not discussed in this document. CTA Abdomen and Pelvis Without and With IV Contrast Both CTA and CT enterography are effective imaging tests for the diagnosis of UGIB with negative endoscopy. Studies have demonstrated that there is no significant clinical difference between accuracy of CTA and enterography for endoscopy-negative GIB [55,56]. CTA can detect bleeding rates as slow as 0.3 mL/min [19]. But, as with angiography, intermittent and slow bleeding can be missed, leading to false-negatives. The sensitivity of CTA has been shown to be 81% in high-risk patients (ie, patients requiring 500 mL of transfusion to maintain vital signs), which decreases to 50% in patients with a slow bleed [57]. The portal venous or delayed phase of multiphasic CT may be more useful for detection of GI masses. CTA and CT enterography can serve as triage tools for identifying patients who may benefit from double-balloon endoscopy [58]. CTA Abdomen With IV Contrast In endoscopy-negative patients, there is no significant literature for CTA abdomen with IV contrast for GIB evaluation. | Nonvariceal Upper Gastrointestinal Bleeding. However, a study showed a higher sensitivity of CT enterography (88%) for the diagnosis of etiology of occult GIB, significantly higher than capsule endoscopy (sensitivity 38%) in their study population [54]. Advantages of CT enterography over capsule endoscopy are ease of performance and better patient satisfaction [5]. CTA Abdomen and Pelvis With IV Contrast In endoscopy-negative patients, there is no significant literature for CT abdomen and pelvis with IV contrast for GIB evaluation. Dual-energy CT allows for the generation of virtual noncontrast images from a CTA data set. However, the use of these virtual unenhanced images in place of true unenhanced images is still limited to certain sites and remains a user-specific preference. So, they are not discussed in this document. CTA Abdomen and Pelvis Without and With IV Contrast Both CTA and CT enterography are effective imaging tests for the diagnosis of UGIB with negative endoscopy. Studies have demonstrated that there is no significant clinical difference between accuracy of CTA and enterography for endoscopy-negative GIB [55,56]. CTA can detect bleeding rates as slow as 0.3 mL/min [19]. But, as with angiography, intermittent and slow bleeding can be missed, leading to false-negatives. The sensitivity of CTA has been shown to be 81% in high-risk patients (ie, patients requiring 500 mL of transfusion to maintain vital signs), which decreases to 50% in patients with a slow bleed [57]. The portal venous or delayed phase of multiphasic CT may be more useful for detection of GI masses. CTA and CT enterography can serve as triage tools for identifying patients who may benefit from double-balloon endoscopy [58]. CTA Abdomen With IV Contrast In endoscopy-negative patients, there is no significant literature for CTA abdomen with IV contrast for GIB evaluation. | 69413 |
acrac_69413_15 | Nonvariceal Upper Gastrointestinal Bleeding | CTA Abdomen Without and With IV Contrast In endoscopy-negative patients, there is no significant literature for CTA abdomen without and with IV contrast for GIB evaluation. Because the pathology can be located anywhere in the small bowel, it is more appropriate to image both the abdomen and pelvis. CTA Chest With IV Contrast In endoscopy-negative patients, there is no significant literature for CTA chest with IV contrast for GIB evaluation. CTA Chest Without and With IV Contrast In endoscopy-negative patients, there is no significant literature for CTA chest without and with IV contrast for GIB evaluation. Fluoroscopy Upper GI Series Barium or iodine upper GI series has no role in diagnosis of acute UGIB. MR Enterography Similar to CT enterography, MR enterography requires oral administration of oral contrast that can obscure bleeding. But MR enterography has been used to identify small bowel sources of bleeding in pediatric patients. In a study of 25 pediatric patients with occult GIB, MR enterography was shown to be diagnostic in 79% patients, with a sensitivity and specificity of 86% and 100%, respectively [59]. Few other studies have also compared capsule with MR enterography and demonstrated better diagnostic yields for capsule endoscopy [60,61]. However, there is Nonvariceal Upper Gastrointestinal Bleeding not enough direct evidence to suggest that MR enterography has an advantage over CT enterography or capsule endoscopy in patients with UGIB and negative endoscopy, particularly in adults. RBC Scan Abdomen and Pelvis Tc-99m-labeled RBC scans can detect bleeding rates as low as 0.05 to 0.1 mL/min. Although there is prior literature on the use of RBC abdomen and pelvis scan, no recent studies have compared the accuracy of RBC scan to CTA or angiography for occult UGIB. Prior reports had suggested variable efficacy for the diagnosis of GIB with localization errors, especially when bleeding arises from the stomach and duodenum [62,63]. | Nonvariceal Upper Gastrointestinal Bleeding. CTA Abdomen Without and With IV Contrast In endoscopy-negative patients, there is no significant literature for CTA abdomen without and with IV contrast for GIB evaluation. Because the pathology can be located anywhere in the small bowel, it is more appropriate to image both the abdomen and pelvis. CTA Chest With IV Contrast In endoscopy-negative patients, there is no significant literature for CTA chest with IV contrast for GIB evaluation. CTA Chest Without and With IV Contrast In endoscopy-negative patients, there is no significant literature for CTA chest without and with IV contrast for GIB evaluation. Fluoroscopy Upper GI Series Barium or iodine upper GI series has no role in diagnosis of acute UGIB. MR Enterography Similar to CT enterography, MR enterography requires oral administration of oral contrast that can obscure bleeding. But MR enterography has been used to identify small bowel sources of bleeding in pediatric patients. In a study of 25 pediatric patients with occult GIB, MR enterography was shown to be diagnostic in 79% patients, with a sensitivity and specificity of 86% and 100%, respectively [59]. Few other studies have also compared capsule with MR enterography and demonstrated better diagnostic yields for capsule endoscopy [60,61]. However, there is Nonvariceal Upper Gastrointestinal Bleeding not enough direct evidence to suggest that MR enterography has an advantage over CT enterography or capsule endoscopy in patients with UGIB and negative endoscopy, particularly in adults. RBC Scan Abdomen and Pelvis Tc-99m-labeled RBC scans can detect bleeding rates as low as 0.05 to 0.1 mL/min. Although there is prior literature on the use of RBC abdomen and pelvis scan, no recent studies have compared the accuracy of RBC scan to CTA or angiography for occult UGIB. Prior reports had suggested variable efficacy for the diagnosis of GIB with localization errors, especially when bleeding arises from the stomach and duodenum [62,63]. | 69413 |
acrac_69413_16 | Nonvariceal Upper Gastrointestinal Bleeding | However, the use of single-photon emission CT (SPECT) and SPECT/CT have made the anatomical position of the uptake clear and contributed to the improved localization of the site of GIB [64]. Variant 5: Adult. Postsurgical or traumatic causes of nonvariceal upper gastrointestinal bleeding. Endoscopy is contraindicated. Initial imaging. This variant is applicable to postsurgical or trauma patients with UGIB, contraindicated to upper GI endoscopy. Arteriography Visceral VA is primarily reserved for unstable patients with UGIB and contraindicated for endoscopy, because it can be used for diagnosis and simultaneous treatment [27,65]. Typically, unstable patients with traumatic hemobilia and bleeding after mucosal or submucosal endoscopic mass resection may require VA [4,39,65]. In stable patients, VA may be performed after CTA, which can show the site of bleeding or a pseudoaneurysm [4,39]. CT Abdomen and Pelvis With IV Contrast There is no significant literature supporting the use of CT abdomen and pelvis with IV contrast for postsurgical or post-traumatic UGIB. However, in general, CT of the abdomen and pelvis with IV contrast is routinely used in trauma patients to assess for intra-abdominal injury. In a meta-analysis of 16 studies enrolling 12,514 patients, CT of the abdomen and pelvis had a sensitivity and specificity of 68% and 97%, respectively, for the diagnosis of traumatic hollow viscus injury [66]. (Note: CTA is a separate procedure distinct from CT abdomen and pelvis with IV contrast). CT Abdomen and Pelvis Without and With IV Contrast There is no significant literature supporting the use of CT abdomen and pelvis without and with IV contrast for postsurgical or post-traumatic UGIB. (Note: CTA is a separate procedure distinct from CT abdomen and pelvis without and with IV contrast). CT Abdomen and Pelvis Without IV Contrast There is no significant literature supporting the use of CT abdomen and pelvis without IV contrast for postsurgical or post-traumatic UGIB. | Nonvariceal Upper Gastrointestinal Bleeding. However, the use of single-photon emission CT (SPECT) and SPECT/CT have made the anatomical position of the uptake clear and contributed to the improved localization of the site of GIB [64]. Variant 5: Adult. Postsurgical or traumatic causes of nonvariceal upper gastrointestinal bleeding. Endoscopy is contraindicated. Initial imaging. This variant is applicable to postsurgical or trauma patients with UGIB, contraindicated to upper GI endoscopy. Arteriography Visceral VA is primarily reserved for unstable patients with UGIB and contraindicated for endoscopy, because it can be used for diagnosis and simultaneous treatment [27,65]. Typically, unstable patients with traumatic hemobilia and bleeding after mucosal or submucosal endoscopic mass resection may require VA [4,39,65]. In stable patients, VA may be performed after CTA, which can show the site of bleeding or a pseudoaneurysm [4,39]. CT Abdomen and Pelvis With IV Contrast There is no significant literature supporting the use of CT abdomen and pelvis with IV contrast for postsurgical or post-traumatic UGIB. However, in general, CT of the abdomen and pelvis with IV contrast is routinely used in trauma patients to assess for intra-abdominal injury. In a meta-analysis of 16 studies enrolling 12,514 patients, CT of the abdomen and pelvis had a sensitivity and specificity of 68% and 97%, respectively, for the diagnosis of traumatic hollow viscus injury [66]. (Note: CTA is a separate procedure distinct from CT abdomen and pelvis with IV contrast). CT Abdomen and Pelvis Without and With IV Contrast There is no significant literature supporting the use of CT abdomen and pelvis without and with IV contrast for postsurgical or post-traumatic UGIB. (Note: CTA is a separate procedure distinct from CT abdomen and pelvis without and with IV contrast). CT Abdomen and Pelvis Without IV Contrast There is no significant literature supporting the use of CT abdomen and pelvis without IV contrast for postsurgical or post-traumatic UGIB. | 69413 |
acrac_69413_17 | Nonvariceal Upper Gastrointestinal Bleeding | (Note: CTA is a separate procedure distinct from CT abdomen and pelvis without IV contrast). CT Abdomen With IV Contrast There is no significant literature supporting the use of CT abdomen with IV contrast for postsurgical or post- traumatic UGIB. (Note: CTA is a separate procedure distinct from CT abdomen with IV contrast). CT Abdomen Without and With IV Contrast There is no significant literature supporting the use of CT abdomen without and with IV contrast for postsurgical or post-traumatic UGIB. However, CT of the abdomen without and with IV contrast may be considered in patients who are suspected to have postsurgical UGIB or hemobilia [67]. In a study of postlaparoscopic sleeve gastrectomy patients presenting with UGIB, CT was able to diagnose all patients with surgical site pseudoaneurysm who were successfully treated by surgery or transcatheter arterial embolization [68]. The patients with poststent ulceration or mucosal diseases are, however, often undiagnosed by CT [68]. CT Abdomen Without IV Contrast There is no significant literature supporting the use of CT abdomen without IV contrast for postsurgical or post- traumatic UGIB. CT Enterography There is no significant literature supporting the use of CT enterography for postsurgical or post-traumatic UGIB. CT enterography requires the administration of large volumes of neutral oral contrast, which can mask GIB by dilution. Additionally, a large volume of oral contrast is often not tolerated by patients when acutely ill. However, Nonvariceal Upper Gastrointestinal Bleeding CT enterography can detect GIB because of technique parameters that can mirror CTA and is useful for patients with occult GI or suspected small bowel bleeding [5]. CTA Abdomen and Pelvis With IV Contrast Although CTA can detect arterial post-traumatic bleeding, performance with IV contrast only may not be as helpful. Without contrast images are important to identify sentinel clot and inherently hyperdense intraluminal material. | Nonvariceal Upper Gastrointestinal Bleeding. (Note: CTA is a separate procedure distinct from CT abdomen and pelvis without IV contrast). CT Abdomen With IV Contrast There is no significant literature supporting the use of CT abdomen with IV contrast for postsurgical or post- traumatic UGIB. (Note: CTA is a separate procedure distinct from CT abdomen with IV contrast). CT Abdomen Without and With IV Contrast There is no significant literature supporting the use of CT abdomen without and with IV contrast for postsurgical or post-traumatic UGIB. However, CT of the abdomen without and with IV contrast may be considered in patients who are suspected to have postsurgical UGIB or hemobilia [67]. In a study of postlaparoscopic sleeve gastrectomy patients presenting with UGIB, CT was able to diagnose all patients with surgical site pseudoaneurysm who were successfully treated by surgery or transcatheter arterial embolization [68]. The patients with poststent ulceration or mucosal diseases are, however, often undiagnosed by CT [68]. CT Abdomen Without IV Contrast There is no significant literature supporting the use of CT abdomen without IV contrast for postsurgical or post- traumatic UGIB. CT Enterography There is no significant literature supporting the use of CT enterography for postsurgical or post-traumatic UGIB. CT enterography requires the administration of large volumes of neutral oral contrast, which can mask GIB by dilution. Additionally, a large volume of oral contrast is often not tolerated by patients when acutely ill. However, Nonvariceal Upper Gastrointestinal Bleeding CT enterography can detect GIB because of technique parameters that can mirror CTA and is useful for patients with occult GI or suspected small bowel bleeding [5]. CTA Abdomen and Pelvis With IV Contrast Although CTA can detect arterial post-traumatic bleeding, performance with IV contrast only may not be as helpful. Without contrast images are important to identify sentinel clot and inherently hyperdense intraluminal material. | 69413 |
acrac_69413_18 | Nonvariceal Upper Gastrointestinal Bleeding | Dual-energy CT allows for the generation of virtual noncontrast images from a CTA data set. However, the use of these virtual unenhanced images in place of true unenhanced images is still limited to certain sites and remains a user-specific preference. So, they are not discussed in this document. There is no significant literature supporting the use of CTA abdomen and pelvis with IV contrast only. CTA Abdomen and Pelvis Without and With IV Contrast In traumatic UGIB, it is important to identify the source quickly, safely, and effectively. CT is frequently performed in patients with trauma to assess for visceral injuries. In patients with UGIB, multiphase CTA can effectively and quickly evaluate abdominal vasculature, GIB, and visceral injuries simultaneously. Mesenteric contrast extravasation has a 75% sensitivity for mesenteric injury [69]. CT can help in triage and prognostication, with active bleeding in the arterial or portal venous phase more likely to be life-threatening versus bleeding that first appears in the equilibrium phase [69]. A rare but life-threatening cause of postsurgical or iatrogenic UGIB can be aortoenteric fistula. CTA is the examination of choice [70]. Evidence of a fistula is suggested by gas in a periprosthetic fluid collection, retraction of the contacting intestinal walls, or the presence of a false aneurysm. Extravasation of contrast into the intestinal lumen is diagnostic of aortoenteric fistula but rarely occurs. Even in patients that can undergo endoscopy, CTA is superior and more sensitive compared with endoscopy for the diagnosis of aortoenteric fistula [71]. CTA Abdomen With IV Contrast There is no significant literature supporting the use of CTA abdomen with IV contrast for postsurgical or post- traumatic UGIB. CTA Abdomen Without and With IV Contrast There is no significant literature supporting the use of CTA abdomen without and with IV contrast for postsurgical or post-traumatic UGIB. | Nonvariceal Upper Gastrointestinal Bleeding. Dual-energy CT allows for the generation of virtual noncontrast images from a CTA data set. However, the use of these virtual unenhanced images in place of true unenhanced images is still limited to certain sites and remains a user-specific preference. So, they are not discussed in this document. There is no significant literature supporting the use of CTA abdomen and pelvis with IV contrast only. CTA Abdomen and Pelvis Without and With IV Contrast In traumatic UGIB, it is important to identify the source quickly, safely, and effectively. CT is frequently performed in patients with trauma to assess for visceral injuries. In patients with UGIB, multiphase CTA can effectively and quickly evaluate abdominal vasculature, GIB, and visceral injuries simultaneously. Mesenteric contrast extravasation has a 75% sensitivity for mesenteric injury [69]. CT can help in triage and prognostication, with active bleeding in the arterial or portal venous phase more likely to be life-threatening versus bleeding that first appears in the equilibrium phase [69]. A rare but life-threatening cause of postsurgical or iatrogenic UGIB can be aortoenteric fistula. CTA is the examination of choice [70]. Evidence of a fistula is suggested by gas in a periprosthetic fluid collection, retraction of the contacting intestinal walls, or the presence of a false aneurysm. Extravasation of contrast into the intestinal lumen is diagnostic of aortoenteric fistula but rarely occurs. Even in patients that can undergo endoscopy, CTA is superior and more sensitive compared with endoscopy for the diagnosis of aortoenteric fistula [71]. CTA Abdomen With IV Contrast There is no significant literature supporting the use of CTA abdomen with IV contrast for postsurgical or post- traumatic UGIB. CTA Abdomen Without and With IV Contrast There is no significant literature supporting the use of CTA abdomen without and with IV contrast for postsurgical or post-traumatic UGIB. | 69413 |
acrac_69485_0 | Neuroendocrine Imaging | Introduction/Background Neuroendocrine abnormalities encompass a range of centrally mediated hormonal imbalances and organ-specific pituitary abnormalities. Imaging is generally focused on the pituitary gland and parasellar region and typically follows endocrine evaluation [1-3]. Abnormalities of the pituitary are often an incidental finding on imaging performed for other indications, though these may be associated with occult endocrine dysfunction [4], and focused imaging is commonly requested in follow-up of these lesions. Extrinsic mass effect can result in dysregulation of pituitary hormone release as well as extra-pituitary dysfunction. The hypothalamic pituitary axis consists of two separate neuroendocrine organs: the anterior pituitary system and the posterior pituitary system. The hormones of the anterior pituitary are thyroid-stimulating hormone, adrenal corticotrophic hormone, prolactin, growth hormone, and the gonadotropins. These are secreted under the influence of hypothalamic trophic factors. The posterior pituitary gland consists of axonal terminations of neurons whose cell bodies are located in the hypothalamus. The principal hormones secreted by these cells are oxytocin and vasopressin or antidiuretic hormone. The hypothalamus also participates in complex mediation of food intake, temperature regulation, sleep and arousal, memory, thirst, and other autonomic functions. Special Imaging Considerations This guideline emphasizes the initial imaging for diagnosis. Additional studies may be ordered specifically for image-guided surgical navigation systems or intraoperative imaging platforms. It is recommended that ordering physicians consult with their radiologists as to protocols that can accomplish surgical guidance needs, to minimize patient recall or reimaging [7-10]. | Neuroendocrine Imaging. Introduction/Background Neuroendocrine abnormalities encompass a range of centrally mediated hormonal imbalances and organ-specific pituitary abnormalities. Imaging is generally focused on the pituitary gland and parasellar region and typically follows endocrine evaluation [1-3]. Abnormalities of the pituitary are often an incidental finding on imaging performed for other indications, though these may be associated with occult endocrine dysfunction [4], and focused imaging is commonly requested in follow-up of these lesions. Extrinsic mass effect can result in dysregulation of pituitary hormone release as well as extra-pituitary dysfunction. The hypothalamic pituitary axis consists of two separate neuroendocrine organs: the anterior pituitary system and the posterior pituitary system. The hormones of the anterior pituitary are thyroid-stimulating hormone, adrenal corticotrophic hormone, prolactin, growth hormone, and the gonadotropins. These are secreted under the influence of hypothalamic trophic factors. The posterior pituitary gland consists of axonal terminations of neurons whose cell bodies are located in the hypothalamus. The principal hormones secreted by these cells are oxytocin and vasopressin or antidiuretic hormone. The hypothalamus also participates in complex mediation of food intake, temperature regulation, sleep and arousal, memory, thirst, and other autonomic functions. Special Imaging Considerations This guideline emphasizes the initial imaging for diagnosis. Additional studies may be ordered specifically for image-guided surgical navigation systems or intraoperative imaging platforms. It is recommended that ordering physicians consult with their radiologists as to protocols that can accomplish surgical guidance needs, to minimize patient recall or reimaging [7-10]. | 69485 |
acrac_69485_1 | Neuroendocrine Imaging | Endoscopic surgical landmarks requiring presurgical assessment include evaluation for presence of sinus inflammatory disease, sphenoid sinus pneumatization, bony spurs, variant anatomy, and bony dehiscence overlying the internal carotid arteries within the sphenoid sinuses [11-13]. Discussion of Procedures by Variant Variant 1: Adult. Suspected or known hypofunctioning pituitary gland (hypopituitarism, growth hormone deficiency, growth deceleration, panhypopituitarism, hypogonadotropic hypogonadism). Initial imaging. Pituitary hypofunction can be caused by mass effect from extrinsic solid or cystic lesions, or from intrinsic pituitary abnormalities. Small prolactin-secreting adenomas in males may result in hypogonadotropic Department of Defense, or United States Government. Reprint requests to: [email protected] Neuroendocrine Imaging hypogonadism, with loss of libido and impotence [14]. Other common mass lesions that may affect the neuroendocrine system are germ-line tumors, meningioma, craniopharyngioma, and Rathke cleft cyst, among others [2,15-18]. Metastatic lesions, sarcoid, and other inflammatory processes may involve the sella and parasellar regions as well. Additionally, an empty sella may be seen with herniation of the subarachnoid space into the sella turcica; while this is usually an incidental finding, close to 30% of patients may demonstrate some hypopituitarism upon testing [19]. CT Sella CT can be used to detect bone-destructive lesions of the skull base, which may affect the sella turcica and can also detect larger macroadenomas. Even with an optimized technique, CT for pituitary pathology is insensitive when compared to MRI [12]. Intravenous (IV) contrast may be useful to characterize lesions or assess for soft-tissue invasion, but it should not be considered a first-line imaging test. CT using dual-energy techniques may discriminate larger pituitary masses from other lesions, such as meningioma [18]. | Neuroendocrine Imaging. Endoscopic surgical landmarks requiring presurgical assessment include evaluation for presence of sinus inflammatory disease, sphenoid sinus pneumatization, bony spurs, variant anatomy, and bony dehiscence overlying the internal carotid arteries within the sphenoid sinuses [11-13]. Discussion of Procedures by Variant Variant 1: Adult. Suspected or known hypofunctioning pituitary gland (hypopituitarism, growth hormone deficiency, growth deceleration, panhypopituitarism, hypogonadotropic hypogonadism). Initial imaging. Pituitary hypofunction can be caused by mass effect from extrinsic solid or cystic lesions, or from intrinsic pituitary abnormalities. Small prolactin-secreting adenomas in males may result in hypogonadotropic Department of Defense, or United States Government. Reprint requests to: [email protected] Neuroendocrine Imaging hypogonadism, with loss of libido and impotence [14]. Other common mass lesions that may affect the neuroendocrine system are germ-line tumors, meningioma, craniopharyngioma, and Rathke cleft cyst, among others [2,15-18]. Metastatic lesions, sarcoid, and other inflammatory processes may involve the sella and parasellar regions as well. Additionally, an empty sella may be seen with herniation of the subarachnoid space into the sella turcica; while this is usually an incidental finding, close to 30% of patients may demonstrate some hypopituitarism upon testing [19]. CT Sella CT can be used to detect bone-destructive lesions of the skull base, which may affect the sella turcica and can also detect larger macroadenomas. Even with an optimized technique, CT for pituitary pathology is insensitive when compared to MRI [12]. Intravenous (IV) contrast may be useful to characterize lesions or assess for soft-tissue invasion, but it should not be considered a first-line imaging test. CT using dual-energy techniques may discriminate larger pituitary masses from other lesions, such as meningioma [18]. | 69485 |
acrac_69485_2 | Neuroendocrine Imaging | Solid masses are more easily defined compared to cystic lesions, which may be confused with cerebral spinal fluid in the suprasellar cistern. CT features of selected suprasellar masses may aid in their characterization. Tumor invasion of the cavernous sinuses can be difficult to detect and may be aided by IV contrast as an adjunctive test. Dual-phase imaging with and without IV contrast is not indicated as an initial imaging study. Thin-section acquisition with multiplanar reformatting is essential and vastly improved over direct coronal imaging. This can also aid intraoperative navigation and provide greater osseous detail for anatomy in the sphenoid sinus prior to trans-sphenoidal surgery. CTA Head CT angiography (CTA) is indicated when vascular lesions, such as aneurysm, are suspected, though these rarely cause clinical symptoms referable to the hypothalamic pituitary axis. Cavernous sinus invasion by pituitary masses can be better detected with CTA compared with noncontrast CT. CTA may be part of operative planning or image guidance; however, CTA is not routinely used for initial evaluation [20]. MRA Head MR angiography (MRA) can be a useful adjunct to pituitary MRI and can reliably depict vascular lesions, such as aneurysms [21,22], though these rarely cause clinical symptoms referable to the hypothalamic pituitary axis. Displacement or encasement of vessels in the suprasellar region can be better defined with MRA for surgical planning; however, MRA is not routinely used for initial evaluation. MRI Sella MRI using high-resolution pituitary protocols is the preferred diagnostic imaging modality for evaluation of the pituitary and sellar regions [1,2,23-28]. Anatomy and pathologies involving the pituitary gland, infundibulum, optic chiasm, and vascular structures are reliably depicted and can be characterized on both precontrast and postcontrast imaging, particularly with the aid of high-resolution, focused field-of-view sequences targeted for sellar and parasellar assessment [29-63]. | Neuroendocrine Imaging. Solid masses are more easily defined compared to cystic lesions, which may be confused with cerebral spinal fluid in the suprasellar cistern. CT features of selected suprasellar masses may aid in their characterization. Tumor invasion of the cavernous sinuses can be difficult to detect and may be aided by IV contrast as an adjunctive test. Dual-phase imaging with and without IV contrast is not indicated as an initial imaging study. Thin-section acquisition with multiplanar reformatting is essential and vastly improved over direct coronal imaging. This can also aid intraoperative navigation and provide greater osseous detail for anatomy in the sphenoid sinus prior to trans-sphenoidal surgery. CTA Head CT angiography (CTA) is indicated when vascular lesions, such as aneurysm, are suspected, though these rarely cause clinical symptoms referable to the hypothalamic pituitary axis. Cavernous sinus invasion by pituitary masses can be better detected with CTA compared with noncontrast CT. CTA may be part of operative planning or image guidance; however, CTA is not routinely used for initial evaluation [20]. MRA Head MR angiography (MRA) can be a useful adjunct to pituitary MRI and can reliably depict vascular lesions, such as aneurysms [21,22], though these rarely cause clinical symptoms referable to the hypothalamic pituitary axis. Displacement or encasement of vessels in the suprasellar region can be better defined with MRA for surgical planning; however, MRA is not routinely used for initial evaluation. MRI Sella MRI using high-resolution pituitary protocols is the preferred diagnostic imaging modality for evaluation of the pituitary and sellar regions [1,2,23-28]. Anatomy and pathologies involving the pituitary gland, infundibulum, optic chiasm, and vascular structures are reliably depicted and can be characterized on both precontrast and postcontrast imaging, particularly with the aid of high-resolution, focused field-of-view sequences targeted for sellar and parasellar assessment [29-63]. | 69485 |
acrac_69485_3 | Neuroendocrine Imaging | MRI with and without IV contrast plays an important role in characterizing lesions of the sella, suprasellar cistern, and any cavernous sinus invasion [64,65]. MRI with IV contrast may only be performed for use in operative guidance, and should not be considered a first-line imaging test. MRI can confirm absence or ectopia of the posterior pituitary gland. Pituitary underdevelopment may be suggested on the basis of imaging; however, objective criteria for pituitary hypoplasia do not exist [40,42,48]. An empty sella is well characterized on MRI, even without IV contrast [19]. Venous Sampling Petrosal Sinus Venous sampling of the cavernous sinuses is not useful in the setting of pituitary hypofunction. Venous sampling is reserved for cases in which a definite excess of pituitary hormone is present, medical management has failed, cross-sectional imaging is negative or equivocal, and surgery is planned [66-68]. Radiography Sella Radiography is insensitive and nonspecific for evaluating sellar pathology. Pituitary adenomas are frequently associated with a normal sella size. The sella turcica can be enlarged by the presence of a pituitary adenoma or other pathophysiological conditions such as pulsation of cerebral spinal fluid through a developmental or acquired dehiscence of the diaphragm sella in the empty sella syndrome [69]. Neuroendocrine Imaging CT Sella CT has the ability to identify large pituitary tumors, and contrast-enhanced CT may define some microadenomas [28,70-73], although MRI is more sensitive. Dual-phase imaging with and without IV contrast is not indicated as an initial imaging study. Larger tumors may cause sellar remodeling, including sellar enlargement, bony erosion, supra sellar extension, invasion into the clivus, or sphenoid sinus. CTA Head CTA may be part of operative planning or image guidance; however, CTA is not routinely used in the initial evaluation of known or suspected hyperfunctioning pituitary adenoma [20]. | Neuroendocrine Imaging. MRI with and without IV contrast plays an important role in characterizing lesions of the sella, suprasellar cistern, and any cavernous sinus invasion [64,65]. MRI with IV contrast may only be performed for use in operative guidance, and should not be considered a first-line imaging test. MRI can confirm absence or ectopia of the posterior pituitary gland. Pituitary underdevelopment may be suggested on the basis of imaging; however, objective criteria for pituitary hypoplasia do not exist [40,42,48]. An empty sella is well characterized on MRI, even without IV contrast [19]. Venous Sampling Petrosal Sinus Venous sampling of the cavernous sinuses is not useful in the setting of pituitary hypofunction. Venous sampling is reserved for cases in which a definite excess of pituitary hormone is present, medical management has failed, cross-sectional imaging is negative or equivocal, and surgery is planned [66-68]. Radiography Sella Radiography is insensitive and nonspecific for evaluating sellar pathology. Pituitary adenomas are frequently associated with a normal sella size. The sella turcica can be enlarged by the presence of a pituitary adenoma or other pathophysiological conditions such as pulsation of cerebral spinal fluid through a developmental or acquired dehiscence of the diaphragm sella in the empty sella syndrome [69]. Neuroendocrine Imaging CT Sella CT has the ability to identify large pituitary tumors, and contrast-enhanced CT may define some microadenomas [28,70-73], although MRI is more sensitive. Dual-phase imaging with and without IV contrast is not indicated as an initial imaging study. Larger tumors may cause sellar remodeling, including sellar enlargement, bony erosion, supra sellar extension, invasion into the clivus, or sphenoid sinus. CTA Head CTA may be part of operative planning or image guidance; however, CTA is not routinely used in the initial evaluation of known or suspected hyperfunctioning pituitary adenoma [20]. | 69485 |
acrac_69485_4 | Neuroendocrine Imaging | MRA Head There is no relevant literature regarding the use of MRA in the initial imaging evaluation of suspected or known hyperfunctioning pituitary adenoma. MRI Sella MRI using high-resolution pituitary protocols is generally considered the gold standard for imaging the pituitary gland in cases of suspected hormone-secreting adenoma [15]. MRI can directly visualize the pituitary gland on noncontrast sequences. The addition of IV contrast increases the conspicuity of small adenomas, which are typically seen as hypoenhancing lesions [13,74,75]. MRI with IV contrast may only be performed for use in operative guidance, and should not be considered a first-line imaging test. MRI has been used to characterize tissue consistency [76] and has been used to predict response to therapy in cases of acromegaly [77]. Dynamic contrast-enhanced imaging of the pituitary is advocated by some authors for detection of microadenoma [78-80]. Additionally, recent studies have noted an increased sensitivity for spoiled gradient-echo 3-D T1 sequence in the detection of hormone-secreting adenomas [81]. Hormone-secreting pituitary tumors are more commonly microadenomas (<10 mm), highlighting the need for high-resolution, focused field-of-view, and thin-section imaging. MRI may further characterize hemorrhage within adenomas as areas of decreased signal intensity on T2- weighted images [41,60]. Venous Sampling Petrosal Sinus Petrosal sinus venous sampling is an invasive study reserved for cases in which a definite excess of pituitary hormone is present, medical management has failed, cross-sectional imaging is negative or equivocal, and surgery is planned [67]. When a significant discrepancy in hormone level, usually adrenal corticotrophic hormone, exists between the petrosal sinus and peripheral blood samples, tumor localization is very accurate [66,68]. | Neuroendocrine Imaging. MRA Head There is no relevant literature regarding the use of MRA in the initial imaging evaluation of suspected or known hyperfunctioning pituitary adenoma. MRI Sella MRI using high-resolution pituitary protocols is generally considered the gold standard for imaging the pituitary gland in cases of suspected hormone-secreting adenoma [15]. MRI can directly visualize the pituitary gland on noncontrast sequences. The addition of IV contrast increases the conspicuity of small adenomas, which are typically seen as hypoenhancing lesions [13,74,75]. MRI with IV contrast may only be performed for use in operative guidance, and should not be considered a first-line imaging test. MRI has been used to characterize tissue consistency [76] and has been used to predict response to therapy in cases of acromegaly [77]. Dynamic contrast-enhanced imaging of the pituitary is advocated by some authors for detection of microadenoma [78-80]. Additionally, recent studies have noted an increased sensitivity for spoiled gradient-echo 3-D T1 sequence in the detection of hormone-secreting adenomas [81]. Hormone-secreting pituitary tumors are more commonly microadenomas (<10 mm), highlighting the need for high-resolution, focused field-of-view, and thin-section imaging. MRI may further characterize hemorrhage within adenomas as areas of decreased signal intensity on T2- weighted images [41,60]. Venous Sampling Petrosal Sinus Petrosal sinus venous sampling is an invasive study reserved for cases in which a definite excess of pituitary hormone is present, medical management has failed, cross-sectional imaging is negative or equivocal, and surgery is planned [67]. When a significant discrepancy in hormone level, usually adrenal corticotrophic hormone, exists between the petrosal sinus and peripheral blood samples, tumor localization is very accurate [66,68]. | 69485 |
acrac_69485_5 | Neuroendocrine Imaging | Radiography Sella There is no relevant literature regarding the use of radiography in the initial evaluation of known or suspected hyperfunctioning pituitary adenoma. Variant 3: Adult. Diabetes insipidus. Initial imaging. Central causes of diabetes insipidus may be associated with abnormalities affecting the pituitary stalk and the hypothalamic-pituitary axis. Mass effect or neoplastic invasion may be present because of germ-line tumors, lymphoma, leukemia, Langerhans cell histiocytosis, metastasis, craniopharyngioma, meningioma, and Rathke cleft cyst, among others [2,15-17]. Inflammatory processes (ie, sarcoid, lymphocytic hypophysitis, granulomatous infiltration) occur in the sellar and suprasellar regions as well. Additionally, an empty sella may be seen with Neuroendocrine Imaging herniation of the subarachnoid space into the sella turcica; this is usually an incidental finding, close to 30% of patients may demonstrate some hypopituitarism upon testing [19]. CT Sella CT can detect solid lesions of the suprasellar cistern and may detect infiltrative lesions of the pituitary stalk but also may miss cystic tumors. Multiplanar reconstructions with thin sections and soft-tissue window settings are crucial to visualize the suprasellar cistern and pituitary stalk. CT with IV contrast helps to visualize the enhancing pituitary stalk and infiltrative lesions of the stalk, though MRI is considered the best first-line study. Dual-phase imaging with and without IV contrast is not indicated as an initial imaging study. CTA Head CTA may be part of operative planning or image guidance; however, CTA is not routinely used for initial evaluation of diabetes insipidus [20]. MRA Head There is no relevant literature regarding the use of MRA in the evaluation of diabetes insipidus. | Neuroendocrine Imaging. Radiography Sella There is no relevant literature regarding the use of radiography in the initial evaluation of known or suspected hyperfunctioning pituitary adenoma. Variant 3: Adult. Diabetes insipidus. Initial imaging. Central causes of diabetes insipidus may be associated with abnormalities affecting the pituitary stalk and the hypothalamic-pituitary axis. Mass effect or neoplastic invasion may be present because of germ-line tumors, lymphoma, leukemia, Langerhans cell histiocytosis, metastasis, craniopharyngioma, meningioma, and Rathke cleft cyst, among others [2,15-17]. Inflammatory processes (ie, sarcoid, lymphocytic hypophysitis, granulomatous infiltration) occur in the sellar and suprasellar regions as well. Additionally, an empty sella may be seen with Neuroendocrine Imaging herniation of the subarachnoid space into the sella turcica; this is usually an incidental finding, close to 30% of patients may demonstrate some hypopituitarism upon testing [19]. CT Sella CT can detect solid lesions of the suprasellar cistern and may detect infiltrative lesions of the pituitary stalk but also may miss cystic tumors. Multiplanar reconstructions with thin sections and soft-tissue window settings are crucial to visualize the suprasellar cistern and pituitary stalk. CT with IV contrast helps to visualize the enhancing pituitary stalk and infiltrative lesions of the stalk, though MRI is considered the best first-line study. Dual-phase imaging with and without IV contrast is not indicated as an initial imaging study. CTA Head CTA may be part of operative planning or image guidance; however, CTA is not routinely used for initial evaluation of diabetes insipidus [20]. MRA Head There is no relevant literature regarding the use of MRA in the evaluation of diabetes insipidus. | 69485 |
acrac_69485_6 | Neuroendocrine Imaging | MRI Sella MRI with and without IV contrast using high-resolution pituitary or skull base protocols is preferred in the workup of suspected central diabetes insipidus and in the detection of abnormalities of the hypothalamic- neurohypophyseal axis, which may lead to failure of normal antidiuretic hormone release and transport. Thin- section T1-weighted images are used to directly identify typical T1 signal hyperintensity of normal neurosecretory granules; such a signal may be absent from the sella when an ectopic posterior pituitary gland is present or in long-standing diabetes insipidus. Traumatic etiologies, such as stalk transection or postoperative sella, can be characterized using thin-section T2-weighted images. MRI with and without IV contrast is especially useful for the detection and characterization of inflammatory lesions of the stalk and neoplastic invasion [23,82]. MRI with IV contrast may only be performed for use in operative guidance and should not be considered a first- line imaging test. Venous Sampling Petrosal Sinus There is no relevant literature regarding the use of venous sampling in the evaluation of diabetes insipidus. Radiography Sella There is no relevant literature regarding the use of radiography in the evaluation of diabetes insipidus. CT Sella CT may provide useful information in the clinical setting of sudden onset of pituitary dysfunction by identifying a pituitary or suprasellar mass but is less sensitive than MRI for the detection of acute pituitary hemorrhage [16]. MRI is considered the optimal first-line test. Calcification seen on CT may help to clarify complicated cases of craniopharyngioma, which can occasionally be confused with hemorrhagic pituitary adenoma on MRI. While the use of IV contrast may help distinguish hemorrhage from enhancement, dual-phase imaging with and without IV contrast is not indicated as an initial imaging study. | Neuroendocrine Imaging. MRI Sella MRI with and without IV contrast using high-resolution pituitary or skull base protocols is preferred in the workup of suspected central diabetes insipidus and in the detection of abnormalities of the hypothalamic- neurohypophyseal axis, which may lead to failure of normal antidiuretic hormone release and transport. Thin- section T1-weighted images are used to directly identify typical T1 signal hyperintensity of normal neurosecretory granules; such a signal may be absent from the sella when an ectopic posterior pituitary gland is present or in long-standing diabetes insipidus. Traumatic etiologies, such as stalk transection or postoperative sella, can be characterized using thin-section T2-weighted images. MRI with and without IV contrast is especially useful for the detection and characterization of inflammatory lesions of the stalk and neoplastic invasion [23,82]. MRI with IV contrast may only be performed for use in operative guidance and should not be considered a first- line imaging test. Venous Sampling Petrosal Sinus There is no relevant literature regarding the use of venous sampling in the evaluation of diabetes insipidus. Radiography Sella There is no relevant literature regarding the use of radiography in the evaluation of diabetes insipidus. CT Sella CT may provide useful information in the clinical setting of sudden onset of pituitary dysfunction by identifying a pituitary or suprasellar mass but is less sensitive than MRI for the detection of acute pituitary hemorrhage [16]. MRI is considered the optimal first-line test. Calcification seen on CT may help to clarify complicated cases of craniopharyngioma, which can occasionally be confused with hemorrhagic pituitary adenoma on MRI. While the use of IV contrast may help distinguish hemorrhage from enhancement, dual-phase imaging with and without IV contrast is not indicated as an initial imaging study. | 69485 |
acrac_69485_7 | Neuroendocrine Imaging | As pituitary apoplexy may clinically present with sudden headache and oculomotor palsies; CT may be considered as part of the initial diagnostic evaluation to exclude intracranial hemorrhage or mass lesion, particularly when rapid diagnosis is needed, such as in an emergency department setting. CTA Head CTA may be part of operative planning or image guidance; however, CTA is not routinely used for initial evaluation of pituitary apoplexy [20]. MRA Head There is no relevant literature regarding the use of MRA in the evaluation of pituitary apoplexy. Neuroendocrine Imaging MRI Sella MRI using high-resolution pituitary protocols is the primary modality for the evaluation of suspected pituitary apoplexy [83]. Tumor enlargement, sellar expansion, and intratumoral hemorrhage are well demonstrated on MRI. Noncontrast imaging is sensitive to the detection of hemorrhage and may show T1 signal hyperintensity, low T2 signal, or a hemorrhage fluid level within the pituitary gland [84]. Not all cases of intrapituitary hemorrhage (ie, subacute/necrotic adenoma) are associated with symptomatic pituitary apoplexy; however, caution is needed in interpreting pituitary hemorrhage in the context of clinical symptoms [85]. While pituitary apoplexy is most commonly caused by hemorrhage into an existing pituitary macroadenoma, other soft-tissue masses may have overlapping imaging features. For example, craniopharyngioma or Rathke cleft cysts can have intrinsic T1-signal hyperintensity that is due to proteinaceous content, and a dermoid or teratoma may show high T1 signal because of fat. Inclusion of T1 fat saturation sequences may aid in differentiation of fat from blood products. Ischemic pituitary apoplexy can be detected on contrast-enhanced MRI in the appropriate clinical context by central nonenhancement of the enlarged pituitary, indicating central tumoral necrosis or ischemia. MRI with IV contrast may only be performed for use in operative guidance and should not be considered a first-line imaging test. | Neuroendocrine Imaging. As pituitary apoplexy may clinically present with sudden headache and oculomotor palsies; CT may be considered as part of the initial diagnostic evaluation to exclude intracranial hemorrhage or mass lesion, particularly when rapid diagnosis is needed, such as in an emergency department setting. CTA Head CTA may be part of operative planning or image guidance; however, CTA is not routinely used for initial evaluation of pituitary apoplexy [20]. MRA Head There is no relevant literature regarding the use of MRA in the evaluation of pituitary apoplexy. Neuroendocrine Imaging MRI Sella MRI using high-resolution pituitary protocols is the primary modality for the evaluation of suspected pituitary apoplexy [83]. Tumor enlargement, sellar expansion, and intratumoral hemorrhage are well demonstrated on MRI. Noncontrast imaging is sensitive to the detection of hemorrhage and may show T1 signal hyperintensity, low T2 signal, or a hemorrhage fluid level within the pituitary gland [84]. Not all cases of intrapituitary hemorrhage (ie, subacute/necrotic adenoma) are associated with symptomatic pituitary apoplexy; however, caution is needed in interpreting pituitary hemorrhage in the context of clinical symptoms [85]. While pituitary apoplexy is most commonly caused by hemorrhage into an existing pituitary macroadenoma, other soft-tissue masses may have overlapping imaging features. For example, craniopharyngioma or Rathke cleft cysts can have intrinsic T1-signal hyperintensity that is due to proteinaceous content, and a dermoid or teratoma may show high T1 signal because of fat. Inclusion of T1 fat saturation sequences may aid in differentiation of fat from blood products. Ischemic pituitary apoplexy can be detected on contrast-enhanced MRI in the appropriate clinical context by central nonenhancement of the enlarged pituitary, indicating central tumoral necrosis or ischemia. MRI with IV contrast may only be performed for use in operative guidance and should not be considered a first-line imaging test. | 69485 |
acrac_69485_8 | Neuroendocrine Imaging | Venous Sampling Petrosal Sinus There is no relevant literature regarding the use of venous sampling in the evaluation of pituitary apoplexy. Radiography Sella There is no relevant literature regarding the use of radiography in the evaluation of pituitary apoplexy. Variant 5: Adult. Surveillance postpituitary or sellar mass resection. Delayed surveillance is performed in patients with known subtotal pituitary adenoma resection [86], nonfunctioning pituitary adenomas [87], and postresection of sellar/suprasellar nonadenoma masses guided by tumor pathology and patient symptoms. In the immediate postoperative setting, local complications may be difficult to discern from normal postoperative changes [88,89]. There is variable literature regarding the frequency and duration of routine follow-up for nonfunctioning tumors [90,91]; however, postoperative imaging is often performed >3 months following trans-sphenoidal surgery [88]. CT Sella Assessment for tumor growth or suprasellar extension can be performed using CT, although MRI is generally considered superior. Invasion of the skull base by pituitary tumor is an uncommon but important complication that is best characterized preoperatively with CT [95]. CT with IV contrast is not typically needed in the postoperative setting, unless specifically focused on adjacent structures such as the cavernous sinuses. Dual-phase imaging with and without IV contrast is not indicated as an initial imaging study. CTA Head Routine vascular imaging following pituitary surgery is generally not indicated unless there is known or clinical suspicion for a vascular complication. MRA Head Routine vascular imaging following pituitary surgery is generally not indicated unless there is known or clinical suspicion for a vascular complication. MRI Sella MRI using high-resolution pituitary protocols is the most useful tool for diagnostic assessment following pituitary surgery. | Neuroendocrine Imaging. Venous Sampling Petrosal Sinus There is no relevant literature regarding the use of venous sampling in the evaluation of pituitary apoplexy. Radiography Sella There is no relevant literature regarding the use of radiography in the evaluation of pituitary apoplexy. Variant 5: Adult. Surveillance postpituitary or sellar mass resection. Delayed surveillance is performed in patients with known subtotal pituitary adenoma resection [86], nonfunctioning pituitary adenomas [87], and postresection of sellar/suprasellar nonadenoma masses guided by tumor pathology and patient symptoms. In the immediate postoperative setting, local complications may be difficult to discern from normal postoperative changes [88,89]. There is variable literature regarding the frequency and duration of routine follow-up for nonfunctioning tumors [90,91]; however, postoperative imaging is often performed >3 months following trans-sphenoidal surgery [88]. CT Sella Assessment for tumor growth or suprasellar extension can be performed using CT, although MRI is generally considered superior. Invasion of the skull base by pituitary tumor is an uncommon but important complication that is best characterized preoperatively with CT [95]. CT with IV contrast is not typically needed in the postoperative setting, unless specifically focused on adjacent structures such as the cavernous sinuses. Dual-phase imaging with and without IV contrast is not indicated as an initial imaging study. CTA Head Routine vascular imaging following pituitary surgery is generally not indicated unless there is known or clinical suspicion for a vascular complication. MRA Head Routine vascular imaging following pituitary surgery is generally not indicated unless there is known or clinical suspicion for a vascular complication. MRI Sella MRI using high-resolution pituitary protocols is the most useful tool for diagnostic assessment following pituitary surgery. | 69485 |
acrac_69485_9 | Neuroendocrine Imaging | MRI with and without IV contrast adds additional information when new tumor is suspected in the setting of gross total tumor resection [86], surveillance of smaller known masses [64], and hormone-secreting microadenomas when there is an unexpected hormonal response to medical therapy. In patients with known tumors and normal hormone levels on dopamine agonist therapy, MRI follow-up may be unnecessary [96]. Inclusion of diffusion-weighted imaging may be considered as a tool to differentiate granulation tissue postoperative from residual or recurrent adenoma [97]. Surveillance after resection of nonadenoma sellar/suprasellar masses should be based on the tumor pathology and patient symptoms, generally requiring MRI without and with IV contrast. High-resolution pituitary protocols generally suffice for the assessment of Neuroendocrine Imaging cavernous sinus involvement or complication. MRI with IV contrast may only be performed for use in operative guidance and should not be considered for routine surveillance. Venous Sampling Petrosal Sinus There is no relevant literature regarding the use of venous sampling in the evaluation of the postoperative sella. Radiography Sella There is no relevant literature regarding the use of radiography in the evaluation of the postoperative sella. Variant 6: Child, males younger than 9 years of age; females younger than 8 years of age. Precocious puberty. Initial imaging. Central causes of precocious puberty are often idiopathic; however, this may be related to intracranial neoplasms, trauma, infection, hydrocephalus, and some syndromes with premature gonadotropin-releasing hormone production that is due to infiltration or extrinsic mass effect [98-103]. The appropriateness of imaging in the setting of precocious puberty is debated. Imaging should always follow hormonal studies that suggest a central origin of precocious puberty [104,105]. | Neuroendocrine Imaging. MRI with and without IV contrast adds additional information when new tumor is suspected in the setting of gross total tumor resection [86], surveillance of smaller known masses [64], and hormone-secreting microadenomas when there is an unexpected hormonal response to medical therapy. In patients with known tumors and normal hormone levels on dopamine agonist therapy, MRI follow-up may be unnecessary [96]. Inclusion of diffusion-weighted imaging may be considered as a tool to differentiate granulation tissue postoperative from residual or recurrent adenoma [97]. Surveillance after resection of nonadenoma sellar/suprasellar masses should be based on the tumor pathology and patient symptoms, generally requiring MRI without and with IV contrast. High-resolution pituitary protocols generally suffice for the assessment of Neuroendocrine Imaging cavernous sinus involvement or complication. MRI with IV contrast may only be performed for use in operative guidance and should not be considered for routine surveillance. Venous Sampling Petrosal Sinus There is no relevant literature regarding the use of venous sampling in the evaluation of the postoperative sella. Radiography Sella There is no relevant literature regarding the use of radiography in the evaluation of the postoperative sella. Variant 6: Child, males younger than 9 years of age; females younger than 8 years of age. Precocious puberty. Initial imaging. Central causes of precocious puberty are often idiopathic; however, this may be related to intracranial neoplasms, trauma, infection, hydrocephalus, and some syndromes with premature gonadotropin-releasing hormone production that is due to infiltration or extrinsic mass effect [98-103]. The appropriateness of imaging in the setting of precocious puberty is debated. Imaging should always follow hormonal studies that suggest a central origin of precocious puberty [104,105]. | 69485 |
acrac_69485_10 | Neuroendocrine Imaging | The age of the child at symptom onset is important, where girls <6 and boys younger <9 are most likely to show a central nervous system abnormality and therefore should be screened with MRI [104,106-108]. For girls 6 to 8 years of age, the likelihood of identifying a central nervous system lesion is lower, estimated between 2% to 7%, and neoplastic in 1% [104,109]. The need for routine central nervous system imaging is controversial and requires careful clinical consideration [107,110]. CT Sella CT provides the ability to evaluate the overall size and structure of the sella turcica but offers little in the way of intrasellar and parasellar soft-tissue detail. Larger lesions in the suprasellar cistern can be identified (ie, germinoma, astrocytoma, arachnoid cyst), as well as gross morphological alterations in the configuration and appearance of the ventricles, such as in the setting of hydrocephalus [111]. CT with IV contrast may highlight a solid mass but is rarely indicated as an initial screening examination. CTA Head CTA may be part of operative planning or image guidance; however, CTA is not routinely used for initial evaluation of precocious puberty [20]. MRA Head There is no relevant literature regarding the use of MRA in the evaluation of precocious puberty. MRI Sella MRI is the preferred imaging modality to evaluate the hypothalamic-pituitary axis and parasellar regions with its superior depiction of parenchymal tissue [105,112]. Gadolinium-based contrast adds additional benefit in characterizing lesions. Small pituitary lesions, such as adenomas and Rathke cleft cysts, may be occult without postcontrast sequences. Contrast-enhanced MRI also discriminates between nonenhancing hypothalamic hamartoma and an enhancing astrocytoma. MRI with IV contrast may only be performed for use in operative guidance and should not be considered as a first-line imaging test. | Neuroendocrine Imaging. The age of the child at symptom onset is important, where girls <6 and boys younger <9 are most likely to show a central nervous system abnormality and therefore should be screened with MRI [104,106-108]. For girls 6 to 8 years of age, the likelihood of identifying a central nervous system lesion is lower, estimated between 2% to 7%, and neoplastic in 1% [104,109]. The need for routine central nervous system imaging is controversial and requires careful clinical consideration [107,110]. CT Sella CT provides the ability to evaluate the overall size and structure of the sella turcica but offers little in the way of intrasellar and parasellar soft-tissue detail. Larger lesions in the suprasellar cistern can be identified (ie, germinoma, astrocytoma, arachnoid cyst), as well as gross morphological alterations in the configuration and appearance of the ventricles, such as in the setting of hydrocephalus [111]. CT with IV contrast may highlight a solid mass but is rarely indicated as an initial screening examination. CTA Head CTA may be part of operative planning or image guidance; however, CTA is not routinely used for initial evaluation of precocious puberty [20]. MRA Head There is no relevant literature regarding the use of MRA in the evaluation of precocious puberty. MRI Sella MRI is the preferred imaging modality to evaluate the hypothalamic-pituitary axis and parasellar regions with its superior depiction of parenchymal tissue [105,112]. Gadolinium-based contrast adds additional benefit in characterizing lesions. Small pituitary lesions, such as adenomas and Rathke cleft cysts, may be occult without postcontrast sequences. Contrast-enhanced MRI also discriminates between nonenhancing hypothalamic hamartoma and an enhancing astrocytoma. MRI with IV contrast may only be performed for use in operative guidance and should not be considered as a first-line imaging test. | 69485 |
acrac_3158171_0 | Parathyroid Adenoma | Persistent PHPT is defined as failure to achieve normocalcemia within 6 months of initial parathyroidectomy, whereas recurrent PHPT is defined as hypercalcemia occurring after a normocalcemic interval of 6 months or more after parathyroidectomy [2,10]. Parathyroid reoperations are surgically challenging, with lower cure rates than first- Reprint requests to: [email protected] Parathyroid Adenoma time surgery and higher complication rates. As such, recent international guidelines state that preoperative imaging is essential in the reoperative setting to localize a target parathyroid lesion (or lesions) and to identify postoperative changes from previous parathyroid explorations that can impact a subsequent surgery [10]. Secondary hyperparathyroidism (SHPT) is a failure of calcium homeostasis whereby increased PTH production in response to hypocalcemia (and/or hyperphosphatemia) is unable to correct plasma calcium because of organ failure or reduced calcium availability. This is most commonly due to chronic kidney disease but can also be a result of malabsorption or vitamin D deficiency, among other causes [11]. Rarely, tertiary hyperparathyroidism (THPT) can occur in patients with longstanding SHPT and is characterized by a lack of PTH suppression despite rising serum calcium levels, manifesting as hypercalcemic hyperparathyroidism. This is most commonly encountered following kidney transplantation in patients with longstanding chronic kidney disease [11]. Surgical excision is recommended for medically refractory cases of SHPT and THPT. As these are typically disorders of MGD (ie, parathyroid hyperplasia), the goal of imaging is to identify all eutopic and potential ectopic or supernumerary glands in an attempt to guide the surgical approach [10,12-14]. Discussion of Procedures by Variant Variant 1: Adult or child. Primary hyperparathyroidism. Initial imaging. | Parathyroid Adenoma. Persistent PHPT is defined as failure to achieve normocalcemia within 6 months of initial parathyroidectomy, whereas recurrent PHPT is defined as hypercalcemia occurring after a normocalcemic interval of 6 months or more after parathyroidectomy [2,10]. Parathyroid reoperations are surgically challenging, with lower cure rates than first- Reprint requests to: [email protected] Parathyroid Adenoma time surgery and higher complication rates. As such, recent international guidelines state that preoperative imaging is essential in the reoperative setting to localize a target parathyroid lesion (or lesions) and to identify postoperative changes from previous parathyroid explorations that can impact a subsequent surgery [10]. Secondary hyperparathyroidism (SHPT) is a failure of calcium homeostasis whereby increased PTH production in response to hypocalcemia (and/or hyperphosphatemia) is unable to correct plasma calcium because of organ failure or reduced calcium availability. This is most commonly due to chronic kidney disease but can also be a result of malabsorption or vitamin D deficiency, among other causes [11]. Rarely, tertiary hyperparathyroidism (THPT) can occur in patients with longstanding SHPT and is characterized by a lack of PTH suppression despite rising serum calcium levels, manifesting as hypercalcemic hyperparathyroidism. This is most commonly encountered following kidney transplantation in patients with longstanding chronic kidney disease [11]. Surgical excision is recommended for medically refractory cases of SHPT and THPT. As these are typically disorders of MGD (ie, parathyroid hyperplasia), the goal of imaging is to identify all eutopic and potential ectopic or supernumerary glands in an attempt to guide the surgical approach [10,12-14]. Discussion of Procedures by Variant Variant 1: Adult or child. Primary hyperparathyroidism. Initial imaging. | 3158171 |
acrac_3158171_1 | Parathyroid Adenoma | All patients undergoing imaging in the setting of PHPT should have biochemically proven disease, as imaging has no role in confirming or excluding the diagnosis [2]. As such, assessing the imaging test specificity and negative predictive value is not clinically relevant. Rather, the role of initial imaging in PHPT is to aid in selecting appropriate patients for MIP by accurately identifying and localizing a single parathyroid adenoma; accordingly, these criteria focus on imaging test sensitivity and PPV. Patients with negative or inconclusive imaging results typically remain surgical candidates but will likely require BNE; inconclusive findings on imaging should not preclude surgical referral [2]. Parathyroid Adenoma Multiple imaging modalities may be utilized in combination during the initial imaging evaluation of PHPT in an attempt to maximize the accuracy and confidence of parathyroid localization via concordant imaging results [2,8,15]. This is supported by multiple studies in the literature showing improved sensitivity and PPV in parathyroid lesion localization with a combination of examinations over each examination in isolation [16-23,25]. CT Neck Neck CT for preoperative localization of PHPT is most commonly performed without and with IV contrast (4-D parathyroid CT) [34]. As such, most published data relate to the performance of CT neck without and with IV contrast; however, there are also some data on the performance of CT neck with IV contrast. There is no relevant literature regarding the use of CT neck without IV contrast in the evaluation of PHPT. As the initial imaging study, retrospective studies report the overall sensitivity of CT neck without and with IV contrast to range between 62% and 88% [35,36] and the overall PPV to range between 84% and 90% [32,37]. The largest retrospective study (400 patients) reports an overall sensitivity of 79% and PPV of 90% [32]. | Parathyroid Adenoma. All patients undergoing imaging in the setting of PHPT should have biochemically proven disease, as imaging has no role in confirming or excluding the diagnosis [2]. As such, assessing the imaging test specificity and negative predictive value is not clinically relevant. Rather, the role of initial imaging in PHPT is to aid in selecting appropriate patients for MIP by accurately identifying and localizing a single parathyroid adenoma; accordingly, these criteria focus on imaging test sensitivity and PPV. Patients with negative or inconclusive imaging results typically remain surgical candidates but will likely require BNE; inconclusive findings on imaging should not preclude surgical referral [2]. Parathyroid Adenoma Multiple imaging modalities may be utilized in combination during the initial imaging evaluation of PHPT in an attempt to maximize the accuracy and confidence of parathyroid localization via concordant imaging results [2,8,15]. This is supported by multiple studies in the literature showing improved sensitivity and PPV in parathyroid lesion localization with a combination of examinations over each examination in isolation [16-23,25]. CT Neck Neck CT for preoperative localization of PHPT is most commonly performed without and with IV contrast (4-D parathyroid CT) [34]. As such, most published data relate to the performance of CT neck without and with IV contrast; however, there are also some data on the performance of CT neck with IV contrast. There is no relevant literature regarding the use of CT neck without IV contrast in the evaluation of PHPT. As the initial imaging study, retrospective studies report the overall sensitivity of CT neck without and with IV contrast to range between 62% and 88% [35,36] and the overall PPV to range between 84% and 90% [32,37]. The largest retrospective study (400 patients) reports an overall sensitivity of 79% and PPV of 90% [32]. | 3158171 |
acrac_3158171_2 | Parathyroid Adenoma | For single- gland disease, reported sensitivities and PPVs range from 92% to 94% [30,38] and 88% to 92% [38,39], respectively. For MGD, reported sensitivities and PPVs range from 43% to 67% [39,40] and 89% to 100%, respectively [32,39]. A data-driven rationale for the noncontrast CT phase has been provided (eg, differentiation of parathyroid from thyroid tissue): 22% of parathyroid lesions have an enhancement pattern similar to thyroid on arterial and venous phases and therefore may be missed if the noncontrast phase were excluded [30]. However, a few retrospective reports suggest that eliminating the noncontrast CT phase may not adversely affect test performance [41-43]. The largest retrospective study (278 patients) of CT neck with IV contrast (without the noncontrast phase) reports sensitivity of 55% and PPV of 57% among patients with single-gland disease, and sensitivity of 23% and PPV of 19% among patients with MGD [44]. Several studies have directly compared the performance of neck CT without and with IV contrast to ultrasound (US) and to nuclear medicine scans within the same patient population. Most of these studies found the performance of neck CT without and with IV contrast to be superior [29,31,38,40]; however, a prospective study of 91 patients found neck CT without and with IV contrast (58% sensitivity, 88% PPV) to be inferior to dual isotope (sestamibi and I-123) pinhole subtraction scintigraphy (93% sensitivity, 98% PPV) [45]. MRI Neck MRI neck is an emerging technique for PHPT preoperative localization and may be performed without IV contrast, with IV contrast, or without and with IV contrast. Most studies evaluating the performance of MRI neck for preoperative localization in the setting of PHPT are retrospective and include fewer than 50 patients. As the initial imaging study, the sensitivity of MRI neck without and with IV contrast performed at 1.5T has been reported to be between 64% and 79% [46-48]. | Parathyroid Adenoma. For single- gland disease, reported sensitivities and PPVs range from 92% to 94% [30,38] and 88% to 92% [38,39], respectively. For MGD, reported sensitivities and PPVs range from 43% to 67% [39,40] and 89% to 100%, respectively [32,39]. A data-driven rationale for the noncontrast CT phase has been provided (eg, differentiation of parathyroid from thyroid tissue): 22% of parathyroid lesions have an enhancement pattern similar to thyroid on arterial and venous phases and therefore may be missed if the noncontrast phase were excluded [30]. However, a few retrospective reports suggest that eliminating the noncontrast CT phase may not adversely affect test performance [41-43]. The largest retrospective study (278 patients) of CT neck with IV contrast (without the noncontrast phase) reports sensitivity of 55% and PPV of 57% among patients with single-gland disease, and sensitivity of 23% and PPV of 19% among patients with MGD [44]. Several studies have directly compared the performance of neck CT without and with IV contrast to ultrasound (US) and to nuclear medicine scans within the same patient population. Most of these studies found the performance of neck CT without and with IV contrast to be superior [29,31,38,40]; however, a prospective study of 91 patients found neck CT without and with IV contrast (58% sensitivity, 88% PPV) to be inferior to dual isotope (sestamibi and I-123) pinhole subtraction scintigraphy (93% sensitivity, 98% PPV) [45]. MRI Neck MRI neck is an emerging technique for PHPT preoperative localization and may be performed without IV contrast, with IV contrast, or without and with IV contrast. Most studies evaluating the performance of MRI neck for preoperative localization in the setting of PHPT are retrospective and include fewer than 50 patients. As the initial imaging study, the sensitivity of MRI neck without and with IV contrast performed at 1.5T has been reported to be between 64% and 79% [46-48]. | 3158171 |
acrac_3158171_3 | Parathyroid Adenoma | In 2 prospective studies performed at 3.0T, the sensitivity of MRI neck without and with IV contrast ranged between 64% and 98%, and the PPV ranged between 67% and 95% [49,50]. A third prospective study of MRI without and with IV contrast performed at 3.0T reported accurate localization of parathyroid adenomas in 92% (34/37) of patients with single-gland disease and 74% (35/47) of patients with MGD [51]. For MRI without IV contrast, the reported sensitivity was 67% at 1.5T [52]. There is no relevant literature regarding the use of 3.0T MRI without IV contrast in the initial imaging evaluation of PHPT. Sestamibi Dual-Phase Scan Neck Dual-phase Tc-99m sestamibi planar parathyroid scintigraphy is an accepted and traditionally widely utilized method for parathyroid adenoma localization in PHPT. The sensitivity of this method (typically reported as localization to the correct quadrant or to an ectopic location) varies widely in the literature, ranging from 41% to 96% [45,52-63]. A larger retrospective study of 180 patients reported a sensitivity of 79% [58]. A systematic review of 11 studies reported a pooled sensitivity of 76% [17]. Some variability in these results may arise from varied technical parameters of these scans (eg, timing of early and delayed scans, use of pinhole versus parallel collimators, addition of oblique acquisitions, etc). There is no clear consensus regarding the superiority between sestamibi dual-phase and dual-tracer subtraction planar imaging. One recent study of 63 patients showed greater sensitivity with dual-phase sestamibi (79%) over both dual-tracer (sestamibi and pertechnetate) (69%) and a combined dual-tracer/dual-phase technique (65%) [59]. However, other studies report similar or improved sensitivity of dual-tracer methods (either pertechnetate or I-123, plus sestamibi) over dual-phase sestamibi when utilizing a planar technique [45,54,64,65]. There is a growing Parathyroid Adenoma | Parathyroid Adenoma. In 2 prospective studies performed at 3.0T, the sensitivity of MRI neck without and with IV contrast ranged between 64% and 98%, and the PPV ranged between 67% and 95% [49,50]. A third prospective study of MRI without and with IV contrast performed at 3.0T reported accurate localization of parathyroid adenomas in 92% (34/37) of patients with single-gland disease and 74% (35/47) of patients with MGD [51]. For MRI without IV contrast, the reported sensitivity was 67% at 1.5T [52]. There is no relevant literature regarding the use of 3.0T MRI without IV contrast in the initial imaging evaluation of PHPT. Sestamibi Dual-Phase Scan Neck Dual-phase Tc-99m sestamibi planar parathyroid scintigraphy is an accepted and traditionally widely utilized method for parathyroid adenoma localization in PHPT. The sensitivity of this method (typically reported as localization to the correct quadrant or to an ectopic location) varies widely in the literature, ranging from 41% to 96% [45,52-63]. A larger retrospective study of 180 patients reported a sensitivity of 79% [58]. A systematic review of 11 studies reported a pooled sensitivity of 76% [17]. Some variability in these results may arise from varied technical parameters of these scans (eg, timing of early and delayed scans, use of pinhole versus parallel collimators, addition of oblique acquisitions, etc). There is no clear consensus regarding the superiority between sestamibi dual-phase and dual-tracer subtraction planar imaging. One recent study of 63 patients showed greater sensitivity with dual-phase sestamibi (79%) over both dual-tracer (sestamibi and pertechnetate) (69%) and a combined dual-tracer/dual-phase technique (65%) [59]. However, other studies report similar or improved sensitivity of dual-tracer methods (either pertechnetate or I-123, plus sestamibi) over dual-phase sestamibi when utilizing a planar technique [45,54,64,65]. There is a growing Parathyroid Adenoma | 3158171 |
acrac_3158171_4 | Parathyroid Adenoma | consensus that the addition of single-photon emission CT (SPECT) without or with a coregistered CT (SPECT/CT) improves localization of parathyroid adenomas over a planar dual-phase sestamibi method [59,61,62]. The sensitivity of sestamibi imaging is decreased in the setting of MGD [55,58], concomitant nodular thyroid disease [56,66], small adenomas [53], and mild hypercalcemia [58]. Literature regarding the pediatric population is limited. A retrospective review of 29 patients reported that sestamibi was useful only in older children with a single adenoma in the setting of sporadic PHPT, and added minimal additional information compared with US, with otherwise no utility for sestamibi in neonates or familial PHPT [67]. Sestamibi Dual-Phase Scan with SPECT or SPECT/CT Neck The addition of SPECT or SPECT/CT to a dual-phase sestamibi scan is more commonly utilized than planar imaging alone. Although there are a number of variations in the timing of the SPECT or SPECT/CT acquisition (ie, early, delayed, or both), it is generally accepted that the improved contrast resolution of SPECT or SPECT/CT over planar imaging provides more precise anatomic localization of parathyroid adenomas [68,69]. The sensitivity of dual-phase sestamibi with SPECT or SPECT/CT is decreased by MGD, uptake masked by retained radionuclide in adjacent thyroid or submandibular gland tissue, and smaller size of the adenoma (within these subsets, sensitivity ranges from 24% to 66%) [71-73]. Sestamibi Scan and I-123 Thyroid Scan Subtraction scintigraphy of parathyroid glands using a combination of sestamibi and I-123 sodium iodide is a commonly utilized alternative or adjunct to the dual-phase sestamibi technique. The 2009 European Association of Nuclear Medicine parathyroid guidelines expressed a preference for the dual-tracer sestamibi and I-123 technique, citing improved sensitivity for MGD and improved likelihood of distinguishing sestamibi-avid thyroid nodules from parathyroid lesions [15]. | Parathyroid Adenoma. consensus that the addition of single-photon emission CT (SPECT) without or with a coregistered CT (SPECT/CT) improves localization of parathyroid adenomas over a planar dual-phase sestamibi method [59,61,62]. The sensitivity of sestamibi imaging is decreased in the setting of MGD [55,58], concomitant nodular thyroid disease [56,66], small adenomas [53], and mild hypercalcemia [58]. Literature regarding the pediatric population is limited. A retrospective review of 29 patients reported that sestamibi was useful only in older children with a single adenoma in the setting of sporadic PHPT, and added minimal additional information compared with US, with otherwise no utility for sestamibi in neonates or familial PHPT [67]. Sestamibi Dual-Phase Scan with SPECT or SPECT/CT Neck The addition of SPECT or SPECT/CT to a dual-phase sestamibi scan is more commonly utilized than planar imaging alone. Although there are a number of variations in the timing of the SPECT or SPECT/CT acquisition (ie, early, delayed, or both), it is generally accepted that the improved contrast resolution of SPECT or SPECT/CT over planar imaging provides more precise anatomic localization of parathyroid adenomas [68,69]. The sensitivity of dual-phase sestamibi with SPECT or SPECT/CT is decreased by MGD, uptake masked by retained radionuclide in adjacent thyroid or submandibular gland tissue, and smaller size of the adenoma (within these subsets, sensitivity ranges from 24% to 66%) [71-73]. Sestamibi Scan and I-123 Thyroid Scan Subtraction scintigraphy of parathyroid glands using a combination of sestamibi and I-123 sodium iodide is a commonly utilized alternative or adjunct to the dual-phase sestamibi technique. The 2009 European Association of Nuclear Medicine parathyroid guidelines expressed a preference for the dual-tracer sestamibi and I-123 technique, citing improved sensitivity for MGD and improved likelihood of distinguishing sestamibi-avid thyroid nodules from parathyroid lesions [15]. | 3158171 |
acrac_3158171_5 | Parathyroid Adenoma | A major technical advantage of using I-123 for a subtraction technique is that thyroid and parathyroid images can be acquired simultaneously in a dual-energy window. Reported sensitivity for dual-tracer sestamibi and I-123 subtraction scintigraphy ranges from 75% to 94% [54,74- 76], including a retrospective review of 2,681 patients revealing a sensitivity of 87% and PPV of 92% [77]. The latter study showed that negative or inconclusive results had a higher association with MGD and lower serum calcium levels. There is no clear consensus regarding the superiority between dual-tracer subtraction and sestamibi dual-phase planar imaging. One recent study of 63 patients showed greater sensitivity with dual-phase sestamibi (79%) over both dual-tracer (sestamibi and pertechnetate) (69%) and a combined dual-tracer/dual-phase (65%) technique [59]. However, other studies report similar or improved sensitivity of dual-tracer methods (either pertechnetate or I-123, plus sestamibi) over dual-phase sestamibi when utilizing a planar technique [45,54,64,65]. Sestamibi Scan and I-123 Thyroid Scan with SPECT or SPECT/CT Neck Although there are a number of variations in the timing of the SPECT or SPECT/CT acquisition (ie, early, delayed, or both), it is generally accepted that the improved contrast resolution of SPECT or SPECT/CT over planar imaging provides more precise anatomic localization of parathyroid adenomas [68,69]. However, there are conflicting data in the literature as to what extent the addition of SPECT or SPECT/CT changes the sensitivity of the examination. For example, Bhatt et al [74] evaluated a pinhole protocol, SPECT/CT alone, and a combination of Parathyroid Adenoma pinhole/SPECT/CT, revealing sensitivities of 88%, 69%, and 81%, respectively. Other studies have shown superior performance of the combined sestamibi/I-123 plus SPECT/CT protocol with sensitivities ranging from 86% to 95% and PPV from 86% to 100% [45,75,78,79]. | Parathyroid Adenoma. A major technical advantage of using I-123 for a subtraction technique is that thyroid and parathyroid images can be acquired simultaneously in a dual-energy window. Reported sensitivity for dual-tracer sestamibi and I-123 subtraction scintigraphy ranges from 75% to 94% [54,74- 76], including a retrospective review of 2,681 patients revealing a sensitivity of 87% and PPV of 92% [77]. The latter study showed that negative or inconclusive results had a higher association with MGD and lower serum calcium levels. There is no clear consensus regarding the superiority between dual-tracer subtraction and sestamibi dual-phase planar imaging. One recent study of 63 patients showed greater sensitivity with dual-phase sestamibi (79%) over both dual-tracer (sestamibi and pertechnetate) (69%) and a combined dual-tracer/dual-phase (65%) technique [59]. However, other studies report similar or improved sensitivity of dual-tracer methods (either pertechnetate or I-123, plus sestamibi) over dual-phase sestamibi when utilizing a planar technique [45,54,64,65]. Sestamibi Scan and I-123 Thyroid Scan with SPECT or SPECT/CT Neck Although there are a number of variations in the timing of the SPECT or SPECT/CT acquisition (ie, early, delayed, or both), it is generally accepted that the improved contrast resolution of SPECT or SPECT/CT over planar imaging provides more precise anatomic localization of parathyroid adenomas [68,69]. However, there are conflicting data in the literature as to what extent the addition of SPECT or SPECT/CT changes the sensitivity of the examination. For example, Bhatt et al [74] evaluated a pinhole protocol, SPECT/CT alone, and a combination of Parathyroid Adenoma pinhole/SPECT/CT, revealing sensitivities of 88%, 69%, and 81%, respectively. Other studies have shown superior performance of the combined sestamibi/I-123 plus SPECT/CT protocol with sensitivities ranging from 86% to 95% and PPV from 86% to 100% [45,75,78,79]. | 3158171 |
acrac_3158171_6 | Parathyroid Adenoma | The sensitivity and PPV of a sestamibi/I-123 subtraction technique are least negatively impacted by concomitant thyroid disease when SPECT/CT is included in the protocol [75]. Sestamibi Scan and Pertechnetate Thyroid Scan Subtraction scintigraphy of parathyroid glands using a combination of sestamibi and Tc-99m pertechnetate is a commonly utilized alternative or adjunct to the dual-phase sestamibi technique. Unlike the dual-tracer method with I-123, the pertechnetate method requires 2 separate acquisitions, which prolongs the overall examination time. Although techniques based solely on planar imaging have largely been supplanted by methods with SPECT or SPECT/CT, there are some noteworthy applications of a planar sestamibi and pertechnetate subtraction in recent literature with reported sensitivities ranging from 69% to 79% [59,80]. A meta-analysis including 5 studies utilizing a dual-tracer technique reported sensitivities ranging from 47% to 87%; however, it is unclear which studies relied purely on a subtracted dual-tracer (sestamibi and pertechnetate) planar method and which were performed in combination with a dual-phase sestamibi scan [17]. In a study of 116 patients with PHPT, the addition of pertechnetate subtraction to a dual-phase sestamibi scan increased reader confidence in adenoma localization and changed the final interpretation in 15% of patients from the dual-phase study [81]. MGD decreases the sensitivity of the subtraction sestamibi and pertechnetate scan from 71% to 64% [82]. There is no clear consensus regarding the superiority between dual-tracer subtraction and sestamibi dual-phase planar imaging. One recent study of 63 patients showed greater sensitivity with dual-phase sestamibi (79%) over both dual-tracer (sestamibi and pertechnetate) (69%) and a combined dual-tracer/dual-phase technique (65%) [59]. | Parathyroid Adenoma. The sensitivity and PPV of a sestamibi/I-123 subtraction technique are least negatively impacted by concomitant thyroid disease when SPECT/CT is included in the protocol [75]. Sestamibi Scan and Pertechnetate Thyroid Scan Subtraction scintigraphy of parathyroid glands using a combination of sestamibi and Tc-99m pertechnetate is a commonly utilized alternative or adjunct to the dual-phase sestamibi technique. Unlike the dual-tracer method with I-123, the pertechnetate method requires 2 separate acquisitions, which prolongs the overall examination time. Although techniques based solely on planar imaging have largely been supplanted by methods with SPECT or SPECT/CT, there are some noteworthy applications of a planar sestamibi and pertechnetate subtraction in recent literature with reported sensitivities ranging from 69% to 79% [59,80]. A meta-analysis including 5 studies utilizing a dual-tracer technique reported sensitivities ranging from 47% to 87%; however, it is unclear which studies relied purely on a subtracted dual-tracer (sestamibi and pertechnetate) planar method and which were performed in combination with a dual-phase sestamibi scan [17]. In a study of 116 patients with PHPT, the addition of pertechnetate subtraction to a dual-phase sestamibi scan increased reader confidence in adenoma localization and changed the final interpretation in 15% of patients from the dual-phase study [81]. MGD decreases the sensitivity of the subtraction sestamibi and pertechnetate scan from 71% to 64% [82]. There is no clear consensus regarding the superiority between dual-tracer subtraction and sestamibi dual-phase planar imaging. One recent study of 63 patients showed greater sensitivity with dual-phase sestamibi (79%) over both dual-tracer (sestamibi and pertechnetate) (69%) and a combined dual-tracer/dual-phase technique (65%) [59]. | 3158171 |
acrac_3158171_7 | Parathyroid Adenoma | However, other studies report similar or improved sensitivity of dual-tracer methods (either pertechnetate or I-123, plus sestamibi) over dual-phase sestamibi when utilizing a planar technique [45,54,64,65]. Sestamibi Scan and Pertechnetate Thyroid Scan with SPECT or SPECT/CT Neck Although there are a number of variations in the timing of the SPECT or SPECT/CT acquisition (ie, early, delayed, or both), it is generally accepted that the greater contrast resolution of SPECT or SPECT/CT over planar imaging provides more precise anatomic localization of parathyroid adenomas [68,69]. The addition of a pertechnetate thyroid scan may further improve localization: in a study of 268 patients, planar pertechnetate subtraction in combination with a dual-phase sestamibi scan with SPECT/CT outperformed the dual-phase sestamibi SPECT/CT alone, increasing sensitivity to 93% and PPV to 96% compared with 88% and 92%, respectively [83]. In patients with concomitant thyroid disease, the addition of CT to a dual-tracer sestamibi and pertechnetate SPECT scan increases sensitivity from 80% to 94% [84]. US Parathyroid US of the parathyroid glands is widely utilized as the initial imaging study in PHPT. Both the American Head and Neck Society Endocrine Surgery Section and the American Association of Endocrine Surgeons deem US to be the preferred initial localization study in patients with PHPT, noting the advantage of concomitant thyroid evaluation [2,8]. A 2012 meta-analysis of 19 studies evaluating US in patients with PHPT revealed a pooled sensitivity of 76% and PPV of 93% [70]. A subsequent 2017 meta-analysis of 12 studies (with few studies also included from the 2012 meta-analysis) revealed a pooled sensitivity of 80% [85]. However, the range of sensitivities reported in the literature varies widely. For example, a study of 604 patients comparing US, scintigraphy, and 4-D CT reported a sensitivity of 59% for US [31], with other smaller studies reporting US sensitivities ranging from 44% to 97% [18,20,58,86-90]. | Parathyroid Adenoma. However, other studies report similar or improved sensitivity of dual-tracer methods (either pertechnetate or I-123, plus sestamibi) over dual-phase sestamibi when utilizing a planar technique [45,54,64,65]. Sestamibi Scan and Pertechnetate Thyroid Scan with SPECT or SPECT/CT Neck Although there are a number of variations in the timing of the SPECT or SPECT/CT acquisition (ie, early, delayed, or both), it is generally accepted that the greater contrast resolution of SPECT or SPECT/CT over planar imaging provides more precise anatomic localization of parathyroid adenomas [68,69]. The addition of a pertechnetate thyroid scan may further improve localization: in a study of 268 patients, planar pertechnetate subtraction in combination with a dual-phase sestamibi scan with SPECT/CT outperformed the dual-phase sestamibi SPECT/CT alone, increasing sensitivity to 93% and PPV to 96% compared with 88% and 92%, respectively [83]. In patients with concomitant thyroid disease, the addition of CT to a dual-tracer sestamibi and pertechnetate SPECT scan increases sensitivity from 80% to 94% [84]. US Parathyroid US of the parathyroid glands is widely utilized as the initial imaging study in PHPT. Both the American Head and Neck Society Endocrine Surgery Section and the American Association of Endocrine Surgeons deem US to be the preferred initial localization study in patients with PHPT, noting the advantage of concomitant thyroid evaluation [2,8]. A 2012 meta-analysis of 19 studies evaluating US in patients with PHPT revealed a pooled sensitivity of 76% and PPV of 93% [70]. A subsequent 2017 meta-analysis of 12 studies (with few studies also included from the 2012 meta-analysis) revealed a pooled sensitivity of 80% [85]. However, the range of sensitivities reported in the literature varies widely. For example, a study of 604 patients comparing US, scintigraphy, and 4-D CT reported a sensitivity of 59% for US [31], with other smaller studies reporting US sensitivities ranging from 44% to 97% [18,20,58,86-90]. | 3158171 |
acrac_3158171_8 | Parathyroid Adenoma | Nonlocalizable adenomas by US are most often due to ectopic or far posterior location, MGD, small adenoma size, and concomitant thyroid disease [58,86,91]. The literature in the pediatric population is limited. In a retrospective review of 29 patients, Alagaratnam et al [67] reported that US in 2 neonates resulted in false-negatives; sensitivity improved in older children to 93%. In a separate study of 16 children, sensitivity of US was 60% [7]. Venous Sampling Parathyroid Parathyroid glands tend to drain ipsilaterally and inferiorly relative to their anatomic location. As such, sampling of PTH levels during selective transvenous catheterization of multiple neck and mediastinal veins can be used to Parathyroid Adenoma Variant 2: Adult or child. Primary hyperparathyroidism, recurrent or persistent after parathyroid surgery. Initial imaging. All patients undergoing imaging in the setting of recurrent or persistent PHPT should have biochemically proven disease, as imaging has no role in confirming or excluding the diagnosis [2,10]. As such, assessing the imaging test specificity and negative predictive value is not clinically relevant. Rather, the role of initial imaging in recurrent or persistent PHPT is to accurately identify and localize a parathyroid lesion (or lesions) and identify postoperative changes from previous parathyroid explorations which can impact a subsequent surgery. Accordingly, these criteria focus on imaging test sensitivity and PPV. Multiple imaging modalities may be utilized in combination during the initial imaging evaluation of recurrent or persistent PHPT after prior surgery in an attempt to maximize the accuracy and confidence of parathyroid localization via concordant imaging results [10,15]. | Parathyroid Adenoma. Nonlocalizable adenomas by US are most often due to ectopic or far posterior location, MGD, small adenoma size, and concomitant thyroid disease [58,86,91]. The literature in the pediatric population is limited. In a retrospective review of 29 patients, Alagaratnam et al [67] reported that US in 2 neonates resulted in false-negatives; sensitivity improved in older children to 93%. In a separate study of 16 children, sensitivity of US was 60% [7]. Venous Sampling Parathyroid Parathyroid glands tend to drain ipsilaterally and inferiorly relative to their anatomic location. As such, sampling of PTH levels during selective transvenous catheterization of multiple neck and mediastinal veins can be used to Parathyroid Adenoma Variant 2: Adult or child. Primary hyperparathyroidism, recurrent or persistent after parathyroid surgery. Initial imaging. All patients undergoing imaging in the setting of recurrent or persistent PHPT should have biochemically proven disease, as imaging has no role in confirming or excluding the diagnosis [2,10]. As such, assessing the imaging test specificity and negative predictive value is not clinically relevant. Rather, the role of initial imaging in recurrent or persistent PHPT is to accurately identify and localize a parathyroid lesion (or lesions) and identify postoperative changes from previous parathyroid explorations which can impact a subsequent surgery. Accordingly, these criteria focus on imaging test sensitivity and PPV. Multiple imaging modalities may be utilized in combination during the initial imaging evaluation of recurrent or persistent PHPT after prior surgery in an attempt to maximize the accuracy and confidence of parathyroid localization via concordant imaging results [10,15]. | 3158171 |
acrac_3158171_9 | Parathyroid Adenoma | CT Neck Neck CT in the evaluation of recurrent or persistent PHPT after prior surgery is most commonly performed without and with IV contrast (4-D parathyroid CT) [34], which is reflected in the published data regarding the performance of neck CT for parathyroid localization in the setting of prior failed surgery. There is no literature regarding the use of CT neck with IV contrast specifically limited to patients with recurrent or persistent PHPT. Additionally, there is no relevant literature regarding the use of CT neck without IV contrast in the evaluation of PHPT. In the setting of recurrent or persistent PHPT, retrospective studies report the sensitivity of CT neck without and with IV contrast to range between 50% and 91% and the PPV to range between 69% and 100% [29,36,40,97]. There is no relevant literature regarding the use of MRI without IV contrast in the evaluation of recurrent or persistent PHPT. Sestamibi Dual-Phase Scan Neck There is no relevant literature regarding the use of planar sestamibi dual-phase scans in the evaluation of recurrent or persistent PHPT after prior surgery. Sestamibi Dual-Phase Scan with SPECT or SPECT/CT Neck There are limited data regarding the use of sestamibi dual-phase scans with SPECT or SPECT/CT in the setting of prior surgery: Sestamibi Scan and I-123 Thyroid Scan There is no relevant literature regarding the use of planar sestamibi and I-123 scans in the evaluation of recurrent or persistent PHPT after prior surgery. Sestamibi Scan and Pertechnetate Thyroid Scan There is no relevant literature regarding the use of planar sestamibi and pertechnetate scans in the evaluation of recurrent or persistent PHPT after prior surgery. Sestamibi Scan and Pertechnetate Thyroid Scan with SPECT or SPECT/CT Neck There is no relevant literature regarding the use of sestamibi and pertechnetate scans with SPECT or SPECT/CT in the evaluation of recurrent or persistent PHPT after prior surgery. | Parathyroid Adenoma. CT Neck Neck CT in the evaluation of recurrent or persistent PHPT after prior surgery is most commonly performed without and with IV contrast (4-D parathyroid CT) [34], which is reflected in the published data regarding the performance of neck CT for parathyroid localization in the setting of prior failed surgery. There is no literature regarding the use of CT neck with IV contrast specifically limited to patients with recurrent or persistent PHPT. Additionally, there is no relevant literature regarding the use of CT neck without IV contrast in the evaluation of PHPT. In the setting of recurrent or persistent PHPT, retrospective studies report the sensitivity of CT neck without and with IV contrast to range between 50% and 91% and the PPV to range between 69% and 100% [29,36,40,97]. There is no relevant literature regarding the use of MRI without IV contrast in the evaluation of recurrent or persistent PHPT. Sestamibi Dual-Phase Scan Neck There is no relevant literature regarding the use of planar sestamibi dual-phase scans in the evaluation of recurrent or persistent PHPT after prior surgery. Sestamibi Dual-Phase Scan with SPECT or SPECT/CT Neck There are limited data regarding the use of sestamibi dual-phase scans with SPECT or SPECT/CT in the setting of prior surgery: Sestamibi Scan and I-123 Thyroid Scan There is no relevant literature regarding the use of planar sestamibi and I-123 scans in the evaluation of recurrent or persistent PHPT after prior surgery. Sestamibi Scan and Pertechnetate Thyroid Scan There is no relevant literature regarding the use of planar sestamibi and pertechnetate scans in the evaluation of recurrent or persistent PHPT after prior surgery. Sestamibi Scan and Pertechnetate Thyroid Scan with SPECT or SPECT/CT Neck There is no relevant literature regarding the use of sestamibi and pertechnetate scans with SPECT or SPECT/CT in the evaluation of recurrent or persistent PHPT after prior surgery. | 3158171 |
acrac_3158171_10 | Parathyroid Adenoma | Venous Sampling Parathyroid Parathyroid glands tend to drain ipsilaterally and inferiorly relative to their anatomic location. As such, sampling of PTH levels during selective transvenous catheterization of multiple neck and mediastinal veins can be used to infer the laterality and regional location of parathyroid lesions [8,92,93]. Because of its invasive technique, selective parathyroid venous sampling is typically reserved for reoperative surgical candidates with recurrent or persistent PHPT after noninvasive examinations (eg, US, sestamibi, CT, etc) yield nonlocalizing, equivocal, or discordant results [96,102]. Although parathyroid venous sampling has been in use since the 1990s [102], there are few pertinent studies within the last 10 years, and nearly all evaluate patients who have had prior nonlocalizing examinations, as this is the Parathyroid Adenoma subset of patients for whom this examination is typically reserved. These studies are retrospective and each evaluated fewer than 40 patients, reporting sensitivities ranging from 40% to 93% [92,93,97,103-105]. There is a potential for false regionalization and/or lateralization of parathyroid lesions due to congenitally variant venous anatomy or alteration of regional venous drainage secondary to previous surgical interventions [93]. As with many invasive vascular procedures, venous sampling is potentially associated with serious but uncommon complications [10]. Variant 3: Adult or child. Secondary hyperparathyroidism. Initial imaging. All patients undergoing imaging in the setting of SHPT should have biochemically proven disease, as imaging has no role in confirming or excluding the diagnosis [11]. As such, assessing the imaging test specificity and negative predictive value is not clinically relevant. | Parathyroid Adenoma. Venous Sampling Parathyroid Parathyroid glands tend to drain ipsilaterally and inferiorly relative to their anatomic location. As such, sampling of PTH levels during selective transvenous catheterization of multiple neck and mediastinal veins can be used to infer the laterality and regional location of parathyroid lesions [8,92,93]. Because of its invasive technique, selective parathyroid venous sampling is typically reserved for reoperative surgical candidates with recurrent or persistent PHPT after noninvasive examinations (eg, US, sestamibi, CT, etc) yield nonlocalizing, equivocal, or discordant results [96,102]. Although parathyroid venous sampling has been in use since the 1990s [102], there are few pertinent studies within the last 10 years, and nearly all evaluate patients who have had prior nonlocalizing examinations, as this is the Parathyroid Adenoma subset of patients for whom this examination is typically reserved. These studies are retrospective and each evaluated fewer than 40 patients, reporting sensitivities ranging from 40% to 93% [92,93,97,103-105]. There is a potential for false regionalization and/or lateralization of parathyroid lesions due to congenitally variant venous anatomy or alteration of regional venous drainage secondary to previous surgical interventions [93]. As with many invasive vascular procedures, venous sampling is potentially associated with serious but uncommon complications [10]. Variant 3: Adult or child. Secondary hyperparathyroidism. Initial imaging. All patients undergoing imaging in the setting of SHPT should have biochemically proven disease, as imaging has no role in confirming or excluding the diagnosis [11]. As such, assessing the imaging test specificity and negative predictive value is not clinically relevant. | 3158171 |
acrac_3158171_11 | Parathyroid Adenoma | Rather, as SHPT is typically a disorder of MGD and is therefore treated with BNE, the role of initial imaging is to accurately identify all eutopic and potential ectopic or supernumerary glands in an attempt to decrease surgical failure rates [12-14]; accordingly, these criteria focus on imaging test sensitivity and PPV. It is possible that multiple imaging modalities may be utilized in combination during the initial imaging evaluation of SHPT in an attempt to maximize the accuracy and confidence of parathyroid localization via concordant imaging results. This is supported by limited literature specific to SHPT [12] and otherwise inferred from data regarding PHPT [16-23] showing improved sensitivity and PPV in parathyroid lesion localization with a combination of examinations over each examination in isolation. MRI Neck There is no relevant literature regarding the use of MRI neck performed without IV contrast, with IV contrast, or without and with IV contrast in the evaluation of SHPT. Parathyroid Adenoma neoplasia; patients with SHPT comprised 55% of the cohort. On a per-lesion basis, the reported sensitivity of sestamibi dual-phase scans with SPECT in detecting ectopic glands was 29% [110]. Sestamibi Scan and I-123 Thyroid Scan There is no relevant literature regarding the use of planar sestamibi scan and I-123 thyroid scan in the evaluation of SHPT. Sestamibi Scan and Pertechnetate Thyroid Scan There is no relevant literature regarding the use of planar sestamibi scan and pertechnetate thyroid scan in the evaluation of SHPT. Sestamibi Scan and Pertechnetate Thyroid Scan with SPECT or SPECT/CT Neck There is no relevant literature regarding the use of sestamibi scan and pertechnetate thyroid scan with SPECT or SPECT/CT in the evaluation of SHPT. Venous Sampling Parathyroid There is no relevant literature regarding the use of venous sampling in the evaluation of SHPT. Variant 4: Adult or child. Tertiary hyperparathyroidism. Initial imaging. | Parathyroid Adenoma. Rather, as SHPT is typically a disorder of MGD and is therefore treated with BNE, the role of initial imaging is to accurately identify all eutopic and potential ectopic or supernumerary glands in an attempt to decrease surgical failure rates [12-14]; accordingly, these criteria focus on imaging test sensitivity and PPV. It is possible that multiple imaging modalities may be utilized in combination during the initial imaging evaluation of SHPT in an attempt to maximize the accuracy and confidence of parathyroid localization via concordant imaging results. This is supported by limited literature specific to SHPT [12] and otherwise inferred from data regarding PHPT [16-23] showing improved sensitivity and PPV in parathyroid lesion localization with a combination of examinations over each examination in isolation. MRI Neck There is no relevant literature regarding the use of MRI neck performed without IV contrast, with IV contrast, or without and with IV contrast in the evaluation of SHPT. Parathyroid Adenoma neoplasia; patients with SHPT comprised 55% of the cohort. On a per-lesion basis, the reported sensitivity of sestamibi dual-phase scans with SPECT in detecting ectopic glands was 29% [110]. Sestamibi Scan and I-123 Thyroid Scan There is no relevant literature regarding the use of planar sestamibi scan and I-123 thyroid scan in the evaluation of SHPT. Sestamibi Scan and Pertechnetate Thyroid Scan There is no relevant literature regarding the use of planar sestamibi scan and pertechnetate thyroid scan in the evaluation of SHPT. Sestamibi Scan and Pertechnetate Thyroid Scan with SPECT or SPECT/CT Neck There is no relevant literature regarding the use of sestamibi scan and pertechnetate thyroid scan with SPECT or SPECT/CT in the evaluation of SHPT. Venous Sampling Parathyroid There is no relevant literature regarding the use of venous sampling in the evaluation of SHPT. Variant 4: Adult or child. Tertiary hyperparathyroidism. Initial imaging. | 3158171 |
acrac_3158171_12 | Parathyroid Adenoma | All patients undergoing imaging in the setting of THPT should have biochemically proven disease, as imaging has no role in confirming or excluding the diagnosis [11]. As such, assessing the imaging test specificity and negative predictive value is not clinically relevant. THPT is typically a disorder of MGD in which the role of initial imaging is to accurately identify all eutopic and potential ectopic or supernumerary glands in an attempt to guide the surgical approach [12-14]; accordingly, these criteria focus on imaging test sensitivity and PPV. It is possible that multiple imaging modalities may be utilized in combination during the initial imaging evaluation of THPT in an attempt to maximize the accuracy and confidence of parathyroid via concordant imaging results. This is inferred from data regarding PHPT [16-23] and SHPT [12] showing improved sensitivity and PPV in parathyroid lesion localization with a combination of examinations over each examination in isolation. Parathyroid Adenoma CT Neck There is no relevant literature regarding the use of CT neck in the evaluation of THPT. MRI Neck There is no relevant literature regarding the use of MRI neck performed without IV contrast, with IV contrast, or without and with IV contrast in the evaluation of THPT. Sestamibi Dual-Phase Scan Neck There is no relevant literature regarding the use of planar sestamibi dual-phase scan neck in the evaluation of THPT. Sestamibi Scan and I-123 Thyroid Scan There is no relevant literature regarding the use of planar sestamibi scan and I-123 thyroid scan in the evaluation of THPT. Sestamibi Scan and I-123 Thyroid Scan with SPECT or SPECT/CT Neck There is no relevant literature regarding the use of sestamibi scan and I-123 thyroid scan with SPECT or SPECT/CT in the evaluation of THPT. Sestamibi Scan and Pertechnetate Thyroid Scan There is no relevant literature regarding the use of planar sestamibi scan and pertechnetate thyroid scan in the evaluation of THPT. | Parathyroid Adenoma. All patients undergoing imaging in the setting of THPT should have biochemically proven disease, as imaging has no role in confirming or excluding the diagnosis [11]. As such, assessing the imaging test specificity and negative predictive value is not clinically relevant. THPT is typically a disorder of MGD in which the role of initial imaging is to accurately identify all eutopic and potential ectopic or supernumerary glands in an attempt to guide the surgical approach [12-14]; accordingly, these criteria focus on imaging test sensitivity and PPV. It is possible that multiple imaging modalities may be utilized in combination during the initial imaging evaluation of THPT in an attempt to maximize the accuracy and confidence of parathyroid via concordant imaging results. This is inferred from data regarding PHPT [16-23] and SHPT [12] showing improved sensitivity and PPV in parathyroid lesion localization with a combination of examinations over each examination in isolation. Parathyroid Adenoma CT Neck There is no relevant literature regarding the use of CT neck in the evaluation of THPT. MRI Neck There is no relevant literature regarding the use of MRI neck performed without IV contrast, with IV contrast, or without and with IV contrast in the evaluation of THPT. Sestamibi Dual-Phase Scan Neck There is no relevant literature regarding the use of planar sestamibi dual-phase scan neck in the evaluation of THPT. Sestamibi Scan and I-123 Thyroid Scan There is no relevant literature regarding the use of planar sestamibi scan and I-123 thyroid scan in the evaluation of THPT. Sestamibi Scan and I-123 Thyroid Scan with SPECT or SPECT/CT Neck There is no relevant literature regarding the use of sestamibi scan and I-123 thyroid scan with SPECT or SPECT/CT in the evaluation of THPT. Sestamibi Scan and Pertechnetate Thyroid Scan There is no relevant literature regarding the use of planar sestamibi scan and pertechnetate thyroid scan in the evaluation of THPT. | 3158171 |
acrac_69447_0 | Acute Respiratory Illness in Immunocompromised Patients | ARI constitutes a group of signs and symptoms that develop over a brief interval (hours to weeks), some of which are constitutional (eg, fever, chills, weight loss), and some of which are organ specific (eg, cough, shortness of in complications encountered in breath, chest pain). The respiratory system is frequently involved immunocompromised states. in immunosuppressed patients, and many progress along a rapid and potentially fatal course [1,2]. Noninfectious causes should also be considered when an immunocompromised patient presents with ARI, including such entities as pulmonary edema, drug-induced lung disease, atelectasis, malignancy, radiation-induced lung disease, pulmonary hemorrhage, diffuse alveolar damage, organizing pneumonia, lung transplant rejection, and pulmonary thromboembolic disease [2]. Discussion of Procedures by Variant Variant 1: Acute respiratory illness in immunocompromised patients. Cough, dyspnea, chest pain, or fever. Initial imaging. Radiography Chest Chest radiography is the initial imaging modality of choice for the diagnostic assessment of immunocompromised patients presenting with ARI [3]. The chest radiograph typically shows the presence and extent of pulmonary infection, although radiographs can be normal in up to 10% of patients with proven disease [4]. The pattern and distribution of abnormalities on the chest radiograph, along with changes on serial radiographic examinations, can aid in formulating a differential diagnosis. Chest radiographs may also show the presence of complications of infectious pneumonia, such as empyema or abscess [3,5]. The well-known limitations of chest radiography; however, are the lack of specificity with regard to specific pathogens and overall low sensitivity for detecting subtle abnormalities in immunocompromised patients with symptomatic disease [1]. aResearch Author, University of Southern California Keck School of Medicine, Los Angeles, California. | Acute Respiratory Illness in Immunocompromised Patients. ARI constitutes a group of signs and symptoms that develop over a brief interval (hours to weeks), some of which are constitutional (eg, fever, chills, weight loss), and some of which are organ specific (eg, cough, shortness of in complications encountered in breath, chest pain). The respiratory system is frequently involved immunocompromised states. in immunosuppressed patients, and many progress along a rapid and potentially fatal course [1,2]. Noninfectious causes should also be considered when an immunocompromised patient presents with ARI, including such entities as pulmonary edema, drug-induced lung disease, atelectasis, malignancy, radiation-induced lung disease, pulmonary hemorrhage, diffuse alveolar damage, organizing pneumonia, lung transplant rejection, and pulmonary thromboembolic disease [2]. Discussion of Procedures by Variant Variant 1: Acute respiratory illness in immunocompromised patients. Cough, dyspnea, chest pain, or fever. Initial imaging. Radiography Chest Chest radiography is the initial imaging modality of choice for the diagnostic assessment of immunocompromised patients presenting with ARI [3]. The chest radiograph typically shows the presence and extent of pulmonary infection, although radiographs can be normal in up to 10% of patients with proven disease [4]. The pattern and distribution of abnormalities on the chest radiograph, along with changes on serial radiographic examinations, can aid in formulating a differential diagnosis. Chest radiographs may also show the presence of complications of infectious pneumonia, such as empyema or abscess [3,5]. The well-known limitations of chest radiography; however, are the lack of specificity with regard to specific pathogens and overall low sensitivity for detecting subtle abnormalities in immunocompromised patients with symptomatic disease [1]. aResearch Author, University of Southern California Keck School of Medicine, Los Angeles, California. | 69447 |
acrac_69447_1 | Acute Respiratory Illness in Immunocompromised Patients | bUniversity of Southern California, Los Angeles, California. cPanel Chair, University of Chicago, Chicago, Illinois. dMassachusetts General Hospital and Harvard Medical School, Boston, Massachusetts. eStanford University Medical Center, Stanford, California; The Society of Thoracic Surgeons. fThe University of Texas MD Anderson Cancer Center, Houston, Texas. gThe University of Texas MD Anderson Cancer Center, Houston, Texas. hUniversity of Kentucky, Lexington, Kentucky. iMayo Clinic, Rochester, Minnesota. jVanderbilt University Medical Center, Nashville, Tennessee, American College of Chest Physicians. kMayo Clinic Florida, Jacksonville, Florida. lDuke University School of Medicine, Durham, North Carolina; The Society of Thoracic Surgeons. mUniversity of Kansas Medical Center, Kansas City, Kansas. nSpecialty Chair, University of Wisconsin School of Medicine and Public Health, Madison, Wisconsin. The American College of Radiology seeks and encourages collaboration with other organizations on the development of the ACR Appropriateness Criteria through society representation on expert panels. Participation by representatives from collaborating societies on the expert panel does not necessarily imply individual or society endorsement of the final document. Reprint requests to: [email protected] Variant 2: Acute respiratory illness in immunocompromised patients. Normal, equivocal, or nonspecific chest radiograph. Next imaging study. CT Chest Chest radiographs in immunocompromised patients with ARI may be equivocal or even normal despite a high suspicion for pulmonary disease [1]. In this setting, chest CT has been shown to confer a distinct improvement in sensitivity for detecting subtle parenchymal abnormalities [4]. In one study, CT performed in febrile neutropenic patients with normal chest radiographs showed pneumonia in 60% of cases at least 5 days before the abnormalities became visible on chest radiographs [6]. | Acute Respiratory Illness in Immunocompromised Patients. bUniversity of Southern California, Los Angeles, California. cPanel Chair, University of Chicago, Chicago, Illinois. dMassachusetts General Hospital and Harvard Medical School, Boston, Massachusetts. eStanford University Medical Center, Stanford, California; The Society of Thoracic Surgeons. fThe University of Texas MD Anderson Cancer Center, Houston, Texas. gThe University of Texas MD Anderson Cancer Center, Houston, Texas. hUniversity of Kentucky, Lexington, Kentucky. iMayo Clinic, Rochester, Minnesota. jVanderbilt University Medical Center, Nashville, Tennessee, American College of Chest Physicians. kMayo Clinic Florida, Jacksonville, Florida. lDuke University School of Medicine, Durham, North Carolina; The Society of Thoracic Surgeons. mUniversity of Kansas Medical Center, Kansas City, Kansas. nSpecialty Chair, University of Wisconsin School of Medicine and Public Health, Madison, Wisconsin. The American College of Radiology seeks and encourages collaboration with other organizations on the development of the ACR Appropriateness Criteria through society representation on expert panels. Participation by representatives from collaborating societies on the expert panel does not necessarily imply individual or society endorsement of the final document. Reprint requests to: [email protected] Variant 2: Acute respiratory illness in immunocompromised patients. Normal, equivocal, or nonspecific chest radiograph. Next imaging study. CT Chest Chest radiographs in immunocompromised patients with ARI may be equivocal or even normal despite a high suspicion for pulmonary disease [1]. In this setting, chest CT has been shown to confer a distinct improvement in sensitivity for detecting subtle parenchymal abnormalities [4]. In one study, CT performed in febrile neutropenic patients with normal chest radiographs showed pneumonia in 60% of cases at least 5 days before the abnormalities became visible on chest radiographs [6]. | 69447 |
acrac_69447_2 | Acute Respiratory Illness in Immunocompromised Patients | In another study in patients with pulmonary tuberculosis, 11% of HIV-positive patients had normal chest radiographs [7]. Similarly, up to one-third of patients infected with Pneumocystis jirovecii may have normal chest radiographs [8]. In addition, because of its superior spatial resolution and cross-sectional display of findings, CT provides enhanced characterization of pulmonary parenchymal abnormalities, which is often helpful in formulating a differential diagnosis. MRI Chest CT is preferable to MRI in immunocompromised patients with ARI who have normal, equivocal, or nonspecific chest radiographs. CT remains superior to MRI in the detection and characterization of pulmonary parenchymal infection [9,10]. However, MRI can be considered as an alternative imaging modality, albeit less sensitive. Moreover, once a diagnosis of infection has been established, MRI can be used as a reasonable alternative to CT for follow-up of parenchymal disease [10-12]. FDG-PET/CT Skull Base to Mid-Thigh There is no relevant literature regarding the use of FDG-PET/CT in the evaluation of immunocompromised patients with ARI who have normal, equivocal, or nonspecific chest radiographs. Therefore, FDG-PET/CT should be rarely utilized in this clinical scenario. Furthermore, many pulmonary infections typically have high FDG uptake and, thus, may be mistaken for malignancy [13]. Variant 3: Acute respiratory illness in immunocompromised patients. Abnormal chest radiograph, multiple, diffuse, or confluent opacities. Next imaging study. CT Chest Further investigation with CT is warranted in immunocompromised patients with ARI who have chest radiographs showing multiple, diffuse, or confluent opacities. In this patient population, multiple or diffuse opacities or nodules have a higher probability of representing an atypical opportunistic infection rather than community-acquired bacterial pneumonia [4,14]. | Acute Respiratory Illness in Immunocompromised Patients. In another study in patients with pulmonary tuberculosis, 11% of HIV-positive patients had normal chest radiographs [7]. Similarly, up to one-third of patients infected with Pneumocystis jirovecii may have normal chest radiographs [8]. In addition, because of its superior spatial resolution and cross-sectional display of findings, CT provides enhanced characterization of pulmonary parenchymal abnormalities, which is often helpful in formulating a differential diagnosis. MRI Chest CT is preferable to MRI in immunocompromised patients with ARI who have normal, equivocal, or nonspecific chest radiographs. CT remains superior to MRI in the detection and characterization of pulmonary parenchymal infection [9,10]. However, MRI can be considered as an alternative imaging modality, albeit less sensitive. Moreover, once a diagnosis of infection has been established, MRI can be used as a reasonable alternative to CT for follow-up of parenchymal disease [10-12]. FDG-PET/CT Skull Base to Mid-Thigh There is no relevant literature regarding the use of FDG-PET/CT in the evaluation of immunocompromised patients with ARI who have normal, equivocal, or nonspecific chest radiographs. Therefore, FDG-PET/CT should be rarely utilized in this clinical scenario. Furthermore, many pulmonary infections typically have high FDG uptake and, thus, may be mistaken for malignancy [13]. Variant 3: Acute respiratory illness in immunocompromised patients. Abnormal chest radiograph, multiple, diffuse, or confluent opacities. Next imaging study. CT Chest Further investigation with CT is warranted in immunocompromised patients with ARI who have chest radiographs showing multiple, diffuse, or confluent opacities. In this patient population, multiple or diffuse opacities or nodules have a higher probability of representing an atypical opportunistic infection rather than community-acquired bacterial pneumonia [4,14]. | 69447 |
acrac_69447_3 | Acute Respiratory Illness in Immunocompromised Patients | CT provides enhanced characterization of pulmonary parenchymal abnormalities that are due to its superior spatial resolution and cross-sectional display of findings. Consequently, certain infections can be identified on CT with a higher degree of confidence than on radiographs ARI in Immunocompromised Patients [15-17]. Common CT patterns of disease in patients with ARI include pulmonary nodules, tree-in-bud pattern, lung consolidation, and ground-glass opacities [4]. These patterns have been described in the literature to represent the appearances of a number of pulmonary infections in the immunocompromised host, including P. jirovecii [8,15,16], invasive pulmonary aspergillosis [15,17-20], mucormycosis [21,22], candidiasis [22,23], cytomegalovirus pneumonia [15,24,25], human metapneumovirus [26-28], and mycobacterial pneumonias [29- 32]. CT shows the detailed morphology of parenchymal opacities, as well as the pattern and distribution of disease. For example, in febrile patients having undergone recent stem cell transplantation, the ability of CT to detect halos of ground-glass opacity around pulmonary nodules is helpful in making the early presumptive diagnosis of invasive aspergillosis, thus prompting initiation of empiric antifungal therapy with improved prognosis [17,19]. Furthermore, CT may show the presence of infection complications, such as empyema or abscess that may not be visible on radiographs. MRI Chest CT is preferable to MRI in immunocompromised patients with ARI who have chest radiographs showing multiple, diffuse, or confluent opacities. CT remains superior to MRI in the detection and characterization of pulmonary parenchymal infection [9,10]. However, when the aerated parenchyma is replaced by consolidation, nodules, or masses, MRI can be considered as an alternative imaging modality, albeit less sensitive. Recent studies have shown that MRI can be used for the diagnosis of pulmonary infection [9,12,33]. | Acute Respiratory Illness in Immunocompromised Patients. CT provides enhanced characterization of pulmonary parenchymal abnormalities that are due to its superior spatial resolution and cross-sectional display of findings. Consequently, certain infections can be identified on CT with a higher degree of confidence than on radiographs ARI in Immunocompromised Patients [15-17]. Common CT patterns of disease in patients with ARI include pulmonary nodules, tree-in-bud pattern, lung consolidation, and ground-glass opacities [4]. These patterns have been described in the literature to represent the appearances of a number of pulmonary infections in the immunocompromised host, including P. jirovecii [8,15,16], invasive pulmonary aspergillosis [15,17-20], mucormycosis [21,22], candidiasis [22,23], cytomegalovirus pneumonia [15,24,25], human metapneumovirus [26-28], and mycobacterial pneumonias [29- 32]. CT shows the detailed morphology of parenchymal opacities, as well as the pattern and distribution of disease. For example, in febrile patients having undergone recent stem cell transplantation, the ability of CT to detect halos of ground-glass opacity around pulmonary nodules is helpful in making the early presumptive diagnosis of invasive aspergillosis, thus prompting initiation of empiric antifungal therapy with improved prognosis [17,19]. Furthermore, CT may show the presence of infection complications, such as empyema or abscess that may not be visible on radiographs. MRI Chest CT is preferable to MRI in immunocompromised patients with ARI who have chest radiographs showing multiple, diffuse, or confluent opacities. CT remains superior to MRI in the detection and characterization of pulmonary parenchymal infection [9,10]. However, when the aerated parenchyma is replaced by consolidation, nodules, or masses, MRI can be considered as an alternative imaging modality, albeit less sensitive. Recent studies have shown that MRI can be used for the diagnosis of pulmonary infection [9,12,33]. | 69447 |
acrac_69447_4 | Acute Respiratory Illness in Immunocompromised Patients | Newer pulse sequences on 3T scanners may further enhance the diagnostic performance of MRI [11,34,35]. Therefore, as technology continues to improve, MRI represents a promising imaging modality for the workup of immunocompromised patients with abnormal chest radiographs. Moreover, once a diagnosis of infection has been established, MRI can be used as a reasonable alternative to CT for follow-up of parenchymal disease [10-12]. FDG-PET/CT Skull Base to Mid-Thigh There is no relevant literature regarding the use of FDG-PET/CT in the evaluation of immunocompromised patients with ARI who have chest radiographs showing multiple, diffuse, or confluent opacities. Therefore, FDG- PET/CT should be rarely utilized in this clinical scenario. Furthermore, many pulmonary infections typically cause high FDG uptake and, thus, may be mistaken for malignancy [13]. Image-Guided Transthoracic Needle Biopsy Image-guided transthoracic needle biopsy may play a role in the identification of the specific organism(s) producing pulmonary parenchymal abnormalities in certain clinical scenarios of immunocompromised patients with ARI. Though the diagnostic yield of transthoracic biopsy in detecting infections is significantly less than in detecting malignancies, the underlying pathogen is identified a significant minority of the time, which allows for prompt and appropriate treatment of the infection in these vulnerable patients [36,37]. Transthoracic biopsy demonstrates the greatest value in the diagnosis of fungal infections, even in the presence of normal sputum and blood cultures [36,38,39]. Even if chest radiography reveals the parenchymal opacities, CT should typically still be performed prior to contemplating transthoracic needle biopsy, as certain lesions may be more amenable to bronchoscopic biopsy if they are in close proximity to the airways. | Acute Respiratory Illness in Immunocompromised Patients. Newer pulse sequences on 3T scanners may further enhance the diagnostic performance of MRI [11,34,35]. Therefore, as technology continues to improve, MRI represents a promising imaging modality for the workup of immunocompromised patients with abnormal chest radiographs. Moreover, once a diagnosis of infection has been established, MRI can be used as a reasonable alternative to CT for follow-up of parenchymal disease [10-12]. FDG-PET/CT Skull Base to Mid-Thigh There is no relevant literature regarding the use of FDG-PET/CT in the evaluation of immunocompromised patients with ARI who have chest radiographs showing multiple, diffuse, or confluent opacities. Therefore, FDG- PET/CT should be rarely utilized in this clinical scenario. Furthermore, many pulmonary infections typically cause high FDG uptake and, thus, may be mistaken for malignancy [13]. Image-Guided Transthoracic Needle Biopsy Image-guided transthoracic needle biopsy may play a role in the identification of the specific organism(s) producing pulmonary parenchymal abnormalities in certain clinical scenarios of immunocompromised patients with ARI. Though the diagnostic yield of transthoracic biopsy in detecting infections is significantly less than in detecting malignancies, the underlying pathogen is identified a significant minority of the time, which allows for prompt and appropriate treatment of the infection in these vulnerable patients [36,37]. Transthoracic biopsy demonstrates the greatest value in the diagnosis of fungal infections, even in the presence of normal sputum and blood cultures [36,38,39]. Even if chest radiography reveals the parenchymal opacities, CT should typically still be performed prior to contemplating transthoracic needle biopsy, as certain lesions may be more amenable to bronchoscopic biopsy if they are in close proximity to the airways. | 69447 |
acrac_69447_5 | Acute Respiratory Illness in Immunocompromised Patients | If a CT-guided transthoracic biopsy is planned, the interventionalist can also use the preprocedural CT to determine the optimal patient position and biopsy route. Variant 4: Acute respiratory illness in immunocompromised patients. Abnormal chest radiograph, noninfectious disease suspected. Next imaging study. CT Chest Further investigation with CT is warranted in immunocompromised patients with ARI who have chest radiographs suspected. Immunosuppressed patients are susceptible to a variety of noninfectious pulmonary diseases. Because of its superior spatial resolution and cross-sectional display of findings, CT provides enhanced characterization of pulmonary parenchymal abnormalities, which may assist in formulating a differential diagnosis. Noninfectious causes of ARI in immunocompromised hosts include pulmonary edema, drug-induced lung disease, atelectasis, malignancy (including post-transplant lymphoproliferative disorder), radiation-induced lung disease, pulmonary hemorrhage, diffuse alveolar damage, organizing pneumonia, lung transplant rejection, graft-versus-host disease, and pulmonary thromboembolic disease [2]. For example, many chemotherapeutic and immunomodulatory agents may result in pulmonary toxicity [40]. CT is more sensitive than chest radiography in detecting drug-induced lung injury from agents such as bleomycin, busulfan, carmustine, and cyclophosphamide [40,41]. In patients with a history of pulmonary malignancy, recurrence of the primary tumor or development of lung metastases should always be considered when chest radiography shows nodular opacities; similarly, metastases from an ARI in Immunocompromised Patients extrathoracic primary often manifest as nodules. An organizing pneumonia pattern of lung injury is a well-known complication in immunocompromised patients, particularly as a manifestation of lung transplant rejection and chronic graft-versus-host disease following allogeneic hematopoietic stem cell transplantation [42-44]. | Acute Respiratory Illness in Immunocompromised Patients. If a CT-guided transthoracic biopsy is planned, the interventionalist can also use the preprocedural CT to determine the optimal patient position and biopsy route. Variant 4: Acute respiratory illness in immunocompromised patients. Abnormal chest radiograph, noninfectious disease suspected. Next imaging study. CT Chest Further investigation with CT is warranted in immunocompromised patients with ARI who have chest radiographs suspected. Immunosuppressed patients are susceptible to a variety of noninfectious pulmonary diseases. Because of its superior spatial resolution and cross-sectional display of findings, CT provides enhanced characterization of pulmonary parenchymal abnormalities, which may assist in formulating a differential diagnosis. Noninfectious causes of ARI in immunocompromised hosts include pulmonary edema, drug-induced lung disease, atelectasis, malignancy (including post-transplant lymphoproliferative disorder), radiation-induced lung disease, pulmonary hemorrhage, diffuse alveolar damage, organizing pneumonia, lung transplant rejection, graft-versus-host disease, and pulmonary thromboembolic disease [2]. For example, many chemotherapeutic and immunomodulatory agents may result in pulmonary toxicity [40]. CT is more sensitive than chest radiography in detecting drug-induced lung injury from agents such as bleomycin, busulfan, carmustine, and cyclophosphamide [40,41]. In patients with a history of pulmonary malignancy, recurrence of the primary tumor or development of lung metastases should always be considered when chest radiography shows nodular opacities; similarly, metastases from an ARI in Immunocompromised Patients extrathoracic primary often manifest as nodules. An organizing pneumonia pattern of lung injury is a well-known complication in immunocompromised patients, particularly as a manifestation of lung transplant rejection and chronic graft-versus-host disease following allogeneic hematopoietic stem cell transplantation [42-44]. | 69447 |
acrac_69447_6 | Acute Respiratory Illness in Immunocompromised Patients | Finally, both malignancy and its treatment are risk factors for deep venous thrombosis, which may lead to pulmonary thromboemboli and pulmonary infarcts [45]. MRI Chest There is no relevant literature regarding the use of MRI in the evaluation of immunocompromised patients with ARI, abnormal chest radiographs, and suspicion of noninfectious disease. However, there may be certain clinical scenarios in which MRI can be used as a reasonable alternative to CT. Given that MRI can characterize infectious consolidations, nodules, and masses, it may depict similar findings in noninfectious conditions. FDG-PET/CT Skull Base to Mid-Thigh FDG-PET/CT may occasionally be beneficial in the evaluation of immunocompromised patients with ARI, chest radiographs showing pulmonary parenchymal opacities, and clinical suspicion of noninfectious disease. In patients with a history of pulmonary malignancy, recurrence of the primary tumor or development of lung metastases should always be considered when chest radiography shows nodular opacities; similarly, metastases from an extrathoracic primary often manifest as nodules. However, FDG-PET/CT should be interpreted cautiously, as many pulmonary infections typically cause high FDG uptake and, thus, may be mistaken for malignancy [13]. Certain immunodeficient conditions, such as HIV/AIDS, render patients extremely susceptible to both malignancies and opportunistic infections. AIDS-defining malignancies that may manifest in the lungs include Kaposi sarcoma and non-Hodgkin lymphoma. With regard to these neoplasms, FDG-PET/CT may assist in the specific diagnosis, staging or restaging of disease, and monitoring of therapeutic response [46]. FDG- PET/CT is also an accurate diagnostic modality for the staging and follow-up of patients with post-transplant lymphoproliferative disorder [47]. | Acute Respiratory Illness in Immunocompromised Patients. Finally, both malignancy and its treatment are risk factors for deep venous thrombosis, which may lead to pulmonary thromboemboli and pulmonary infarcts [45]. MRI Chest There is no relevant literature regarding the use of MRI in the evaluation of immunocompromised patients with ARI, abnormal chest radiographs, and suspicion of noninfectious disease. However, there may be certain clinical scenarios in which MRI can be used as a reasonable alternative to CT. Given that MRI can characterize infectious consolidations, nodules, and masses, it may depict similar findings in noninfectious conditions. FDG-PET/CT Skull Base to Mid-Thigh FDG-PET/CT may occasionally be beneficial in the evaluation of immunocompromised patients with ARI, chest radiographs showing pulmonary parenchymal opacities, and clinical suspicion of noninfectious disease. In patients with a history of pulmonary malignancy, recurrence of the primary tumor or development of lung metastases should always be considered when chest radiography shows nodular opacities; similarly, metastases from an extrathoracic primary often manifest as nodules. However, FDG-PET/CT should be interpreted cautiously, as many pulmonary infections typically cause high FDG uptake and, thus, may be mistaken for malignancy [13]. Certain immunodeficient conditions, such as HIV/AIDS, render patients extremely susceptible to both malignancies and opportunistic infections. AIDS-defining malignancies that may manifest in the lungs include Kaposi sarcoma and non-Hodgkin lymphoma. With regard to these neoplasms, FDG-PET/CT may assist in the specific diagnosis, staging or restaging of disease, and monitoring of therapeutic response [46]. FDG- PET/CT is also an accurate diagnostic modality for the staging and follow-up of patients with post-transplant lymphoproliferative disorder [47]. | 69447 |
acrac_69487_0 | Plexopathy | aThe Ohio State University Wexner Medical Center, Columbus, Ohio. bResearch Author, The Ohio State University Wexner Medical Center, Columbus, Ohio. cPanel Chair, University of Utah, Salt Lake City, Utah. dPanel Chair, Mayo Clinic, Jacksonville, Florida. ePanel Vice-Chair, Wake Forest University School of Medicine, Winston Salem, North Carolina. fPanel Vice-Chair, Mallinckrodt Institute of Radiology, Saint Louis, Missouri. gUniversity of Pittsburgh Medical Center, Pittsburgh, Pennsylvania. hJames J. Peters VA Medical Center, Bronx, New York; American Academy of Orthopaedic Surgeons. iMontefiore Medical Center, Bronx, New York. jUniversity of Utah Health, Salt Lake City, Utah. kMayo Clinic, Rochester, Minnesota. lThomas Jefferson University Hospital, Philadelphia, Pennsylvania. mUniversity of California Los Angeles, Los Angeles, California; American Academy of Neurology. nUniversity of Michigan, Ann Arbor, Michigan. oJacobi Medical Center, Bronx, New York. pUniversity of California San Francisco, San Francisco, California. qIndiana University School of Medicine, Indianapolis, Indiana; American College of Physicians. rDuke University, Durham, North Carolina; Neurosurgery expert. sUniversity of Colorado School of Medicine, Anschutz Medical Campus, Aurora, Colorado. tSpecialty Chair, University of Kentucky, Lexington, Kentucky. uSpecialty Chair, Atlanta VA Health Care System and Emory University, Atlanta, Georgia. The American College of Radiology seeks and encourages collaboration with other organizations on the development of the ACR Appropriateness Criteria through representation of such organizations on expert panels. Participation on the expert panel does not necessarily imply endorsement of the final document by individual contributors or their respective organization. Reprint requests to: [email protected] Plexopathy plexus imaging [15-20]. | Plexopathy. aThe Ohio State University Wexner Medical Center, Columbus, Ohio. bResearch Author, The Ohio State University Wexner Medical Center, Columbus, Ohio. cPanel Chair, University of Utah, Salt Lake City, Utah. dPanel Chair, Mayo Clinic, Jacksonville, Florida. ePanel Vice-Chair, Wake Forest University School of Medicine, Winston Salem, North Carolina. fPanel Vice-Chair, Mallinckrodt Institute of Radiology, Saint Louis, Missouri. gUniversity of Pittsburgh Medical Center, Pittsburgh, Pennsylvania. hJames J. Peters VA Medical Center, Bronx, New York; American Academy of Orthopaedic Surgeons. iMontefiore Medical Center, Bronx, New York. jUniversity of Utah Health, Salt Lake City, Utah. kMayo Clinic, Rochester, Minnesota. lThomas Jefferson University Hospital, Philadelphia, Pennsylvania. mUniversity of California Los Angeles, Los Angeles, California; American Academy of Neurology. nUniversity of Michigan, Ann Arbor, Michigan. oJacobi Medical Center, Bronx, New York. pUniversity of California San Francisco, San Francisco, California. qIndiana University School of Medicine, Indianapolis, Indiana; American College of Physicians. rDuke University, Durham, North Carolina; Neurosurgery expert. sUniversity of Colorado School of Medicine, Anschutz Medical Campus, Aurora, Colorado. tSpecialty Chair, University of Kentucky, Lexington, Kentucky. uSpecialty Chair, Atlanta VA Health Care System and Emory University, Atlanta, Georgia. The American College of Radiology seeks and encourages collaboration with other organizations on the development of the ACR Appropriateness Criteria through representation of such organizations on expert panels. Participation on the expert panel does not necessarily imply endorsement of the final document by individual contributors or their respective organization. Reprint requests to: [email protected] Plexopathy plexus imaging [15-20]. | 69487 |
acrac_69487_1 | Plexopathy | Research also continues regarding the use and possible advantages of higher field strength [21] in regards to spatial resolution and contrast [4], volumetric sequences [22], and diffusion tensor imaging [6,23- 26]. Imaging at 1.5T may be beneficial to reduce artifact if metal is present in the area of clinical concern. Primary tumors of the brachial plexus are most commonly benign peripheral nerve sheath schwannomas and neurofibromas, which can be sporadic or can be associated with neurofibromatosis type 2 and type 1, respectively. Malignant peripheral nerve sheath tumors of the brachial plexus are rare and occur more frequently in patients with neurofibromatosis type 1. The most common non-neurogenic primary tumors of the brachial plexus are desmoid tumors and lipomas [27]. Lymphoma can involve the plexus either because of local encasement or nerve infiltration. Extrinsic tumors can directly invade or metastasize to the brachial plexus [29], most commonly due to lung and breast cancer, respectively. Superior sulcus tumors of the lung (Pancoast tumors) often directly invade the lower trunk of the brachial plexus and can be associated with Horner syndrome and pain along the ulnar nerve distribution. Variant 5 describes brachial plexopathy in the setting of a known malignancy or post-treatment syndrome; however, plexopathy can be the first clinical presentation of neoplastic disease. Systemic, inflammatory, and/or immune-mediated processes that involve the brachial plexus include Parsonage- Turner syndrome (ie, neuralgic amyotrophy or brachial plexitis) [30-32], chronic inflammatory neuropathies (eg, chronic inflammatory demyelinating polyradiculoneuropathy, multifocal motor neuropathy, Lewis-Sumner syndrome) [15,33-38], hereditary neuropathies (eg, Charcot-Marie-Tooth syndrome) [39], sarcoidosis [27], and infection [40-42]. The diagnosis of these disorders is typically based on clinical and electrodiagnostic evaluation, as the imaging features can overlap considerably. | Plexopathy. Research also continues regarding the use and possible advantages of higher field strength [21] in regards to spatial resolution and contrast [4], volumetric sequences [22], and diffusion tensor imaging [6,23- 26]. Imaging at 1.5T may be beneficial to reduce artifact if metal is present in the area of clinical concern. Primary tumors of the brachial plexus are most commonly benign peripheral nerve sheath schwannomas and neurofibromas, which can be sporadic or can be associated with neurofibromatosis type 2 and type 1, respectively. Malignant peripheral nerve sheath tumors of the brachial plexus are rare and occur more frequently in patients with neurofibromatosis type 1. The most common non-neurogenic primary tumors of the brachial plexus are desmoid tumors and lipomas [27]. Lymphoma can involve the plexus either because of local encasement or nerve infiltration. Extrinsic tumors can directly invade or metastasize to the brachial plexus [29], most commonly due to lung and breast cancer, respectively. Superior sulcus tumors of the lung (Pancoast tumors) often directly invade the lower trunk of the brachial plexus and can be associated with Horner syndrome and pain along the ulnar nerve distribution. Variant 5 describes brachial plexopathy in the setting of a known malignancy or post-treatment syndrome; however, plexopathy can be the first clinical presentation of neoplastic disease. Systemic, inflammatory, and/or immune-mediated processes that involve the brachial plexus include Parsonage- Turner syndrome (ie, neuralgic amyotrophy or brachial plexitis) [30-32], chronic inflammatory neuropathies (eg, chronic inflammatory demyelinating polyradiculoneuropathy, multifocal motor neuropathy, Lewis-Sumner syndrome) [15,33-38], hereditary neuropathies (eg, Charcot-Marie-Tooth syndrome) [39], sarcoidosis [27], and infection [40-42]. The diagnosis of these disorders is typically based on clinical and electrodiagnostic evaluation, as the imaging features can overlap considerably. | 69487 |
acrac_69487_2 | Plexopathy | CT Myelography Cervical Spine There is no relevant literature regarding the use of CT myelography of the cervical spine in the evaluation of nontraumatic brachial plexopathy. Myelography is not routinely performed for the evaluation of nontraumatic plexopathy because it does not directly evaluate the plexus lateral to the neural foramina. CT Neck There is no relevant literature regarding the use of neck CT in the evaluation of nontraumatic brachial plexopathy. CT offers the next highest level of anatomic visualization of the brachial plexus after MRI and can evaluate for adjacent soft-tissue lesions or tumors that may involve the plexus [43]. CT with IV contrast can be useful for detecting and characterizing soft-tissue masses and tumors, which are in the differential diagnosis of nontraumatic brachial plexopathy and therefore may provide additional information over CT without IV contrast in this setting [43]. Plexopathy CT Cervical Spine There is no relevant literature regarding the use of CT cervical spine in the evaluation of nontraumatic brachial plexopathy. CT cervical spine cannot visualize the preganglionic nerve roots and does not fully evaluate the postganglionic brachial plexus because of the narrow field of view and limited soft-tissue contrast resolution relative to MRI. FDG-PET/CT Whole Body There is no relevant literature to support the use of fluorine-18-2-fluoro-2-deoxy-D-glucose (FDG)-PET/CT in the evaluation of nontraumatic brachial plexopathy in the absence of a known malignancy. MRI Brachial Plexus Brachial plexus MRI has been shown to be useful in evaluating nontraumatic brachial plexopathy because of its superior soft-tissue contrast and good spatial resolution, providing detailed definition of intraneural anatomy as well as localizing pathologic lesions in conditions in which electrodiagnostic and physical findings are nonspecific [6,9,10,27]. | Plexopathy. CT Myelography Cervical Spine There is no relevant literature regarding the use of CT myelography of the cervical spine in the evaluation of nontraumatic brachial plexopathy. Myelography is not routinely performed for the evaluation of nontraumatic plexopathy because it does not directly evaluate the plexus lateral to the neural foramina. CT Neck There is no relevant literature regarding the use of neck CT in the evaluation of nontraumatic brachial plexopathy. CT offers the next highest level of anatomic visualization of the brachial plexus after MRI and can evaluate for adjacent soft-tissue lesions or tumors that may involve the plexus [43]. CT with IV contrast can be useful for detecting and characterizing soft-tissue masses and tumors, which are in the differential diagnosis of nontraumatic brachial plexopathy and therefore may provide additional information over CT without IV contrast in this setting [43]. Plexopathy CT Cervical Spine There is no relevant literature regarding the use of CT cervical spine in the evaluation of nontraumatic brachial plexopathy. CT cervical spine cannot visualize the preganglionic nerve roots and does not fully evaluate the postganglionic brachial plexus because of the narrow field of view and limited soft-tissue contrast resolution relative to MRI. FDG-PET/CT Whole Body There is no relevant literature to support the use of fluorine-18-2-fluoro-2-deoxy-D-glucose (FDG)-PET/CT in the evaluation of nontraumatic brachial plexopathy in the absence of a known malignancy. MRI Brachial Plexus Brachial plexus MRI has been shown to be useful in evaluating nontraumatic brachial plexopathy because of its superior soft-tissue contrast and good spatial resolution, providing detailed definition of intraneural anatomy as well as localizing pathologic lesions in conditions in which electrodiagnostic and physical findings are nonspecific [6,9,10,27]. | 69487 |
acrac_69487_3 | Plexopathy | Tagliafico et al [44] in a blinded, retrospective review studied 157 patients who underwent brachial plexus MRI found an overall sensitivity of 81%, specificity of 91%, positive predictive value of 82%, negative predictive value of 91%, and accuracy of 88% when compared with the reference standard of surgical findings and clinical follow-up. Du et al [45] in a retrospective review studied 191 patients and found that the brachial plexus MRI provided additional information beyond that of clinical evaluation and electrodiagnostic studies in 45% of patients. Hilgenfeld et al [46] in a blinded, retrospective review studied 36 patients and found that brachial plexus MRI could reliably differentiate compressive from noncompressive plexopathy in all patients. MRI with and without IV contrast can be useful for detecting and characterizing several of the etiologies in the differential diagnosis of nontraumatic brachial plexopathy and may provide additional information over MRI without IV contrast in this setting [47]. US Neck There is no relevant literature to support the use of ultrasound (US) as the primary imaging modality for patients with nontraumatic brachial plexopathy in whom clinical and electrodiagnostic evaluation has been inconclusive. US may be a useful supplemental test in selected centers [50]. US has been described as an adjunctive tool for assessment of nerve enlargement in patients with a clinically diagnosed neuropathy [30,35,39,51-55]. US can be very useful for image-guided therapy, including regional anesthesia, which is beyond the scope of this topic. Variant 2: Lumbosacral plexopathy, acute or chronic, nontraumatic. No known malignancy. Initial imaging. This variant encompasses nontraumatic lumbosacral plexopathy occurring in patients without a history of systemic malignancy or post-treatment syndrome. | Plexopathy. Tagliafico et al [44] in a blinded, retrospective review studied 157 patients who underwent brachial plexus MRI found an overall sensitivity of 81%, specificity of 91%, positive predictive value of 82%, negative predictive value of 91%, and accuracy of 88% when compared with the reference standard of surgical findings and clinical follow-up. Du et al [45] in a retrospective review studied 191 patients and found that the brachial plexus MRI provided additional information beyond that of clinical evaluation and electrodiagnostic studies in 45% of patients. Hilgenfeld et al [46] in a blinded, retrospective review studied 36 patients and found that brachial plexus MRI could reliably differentiate compressive from noncompressive plexopathy in all patients. MRI with and without IV contrast can be useful for detecting and characterizing several of the etiologies in the differential diagnosis of nontraumatic brachial plexopathy and may provide additional information over MRI without IV contrast in this setting [47]. US Neck There is no relevant literature to support the use of ultrasound (US) as the primary imaging modality for patients with nontraumatic brachial plexopathy in whom clinical and electrodiagnostic evaluation has been inconclusive. US may be a useful supplemental test in selected centers [50]. US has been described as an adjunctive tool for assessment of nerve enlargement in patients with a clinically diagnosed neuropathy [30,35,39,51-55]. US can be very useful for image-guided therapy, including regional anesthesia, which is beyond the scope of this topic. Variant 2: Lumbosacral plexopathy, acute or chronic, nontraumatic. No known malignancy. Initial imaging. This variant encompasses nontraumatic lumbosacral plexopathy occurring in patients without a history of systemic malignancy or post-treatment syndrome. | 69487 |
acrac_69487_4 | Plexopathy | The differential diagnosis for nontraumatic lumbosacral plexopathy includes entrapment, inflammatory, autoimmune, hereditary, ischemic, and idiopathic etiologies that tend to affect the plexus diffusely, as well as neoplasms or extrinsic compressive lesions that focally involve the plexus [2-4,7,8]. Entrapment neuropathies are a common cause of lumbosacral plexopathy and can result from spinal or extraspinal compression [3]. The clinical and electrodiagnostic features of lumbosacral plexopathy and radiculopathy often overlap, and imaging can help localize the site of nerve compression [1]. In some cases, lumbosacral plexus MRI can detect spinal causes of nerve root compression that may not be detected on a lumbar spine MRI, such as a lateral disc herniation that compresses the distal nerve root lateral to the neural foramen [3]. Lumbosacral plexus MRI may also detect signal abnormalities in the nerve root and plexus distal to the site of spinal neural compression, which may provide additional evidence of the symptomatic nerve root compression level [56]. A commonly described cause of extraspinal nerve entrapment is the piriformis syndrome, in which the sciatic nerve can be compressed by Plexopathy the piriformis muscle due to either the anatomic variation or an associated fibrous band [7,8,57]. Imaging can be useful for detecting nerve abnormalities and/or neuromuscular variants associated with extraspinal nerve compression and can help guide treatment with surgery, interventional, or noninvasive therapy. Primary tumors of the lumbosacral plexus are most commonly benign peripheral nerve sheath schwannomas and neurofibromas, which can be sporadic or can be associated with neurofibromatosis. Malignant peripheral nerve sheath tumors of the lumbosacral plexus are rare and occur more frequently in patients with neurofibromatosis [7]. Other primary malignant or metastatic tumors can also involve the lumbosacral plexus [4]. | Plexopathy. The differential diagnosis for nontraumatic lumbosacral plexopathy includes entrapment, inflammatory, autoimmune, hereditary, ischemic, and idiopathic etiologies that tend to affect the plexus diffusely, as well as neoplasms or extrinsic compressive lesions that focally involve the plexus [2-4,7,8]. Entrapment neuropathies are a common cause of lumbosacral plexopathy and can result from spinal or extraspinal compression [3]. The clinical and electrodiagnostic features of lumbosacral plexopathy and radiculopathy often overlap, and imaging can help localize the site of nerve compression [1]. In some cases, lumbosacral plexus MRI can detect spinal causes of nerve root compression that may not be detected on a lumbar spine MRI, such as a lateral disc herniation that compresses the distal nerve root lateral to the neural foramen [3]. Lumbosacral plexus MRI may also detect signal abnormalities in the nerve root and plexus distal to the site of spinal neural compression, which may provide additional evidence of the symptomatic nerve root compression level [56]. A commonly described cause of extraspinal nerve entrapment is the piriformis syndrome, in which the sciatic nerve can be compressed by Plexopathy the piriformis muscle due to either the anatomic variation or an associated fibrous band [7,8,57]. Imaging can be useful for detecting nerve abnormalities and/or neuromuscular variants associated with extraspinal nerve compression and can help guide treatment with surgery, interventional, or noninvasive therapy. Primary tumors of the lumbosacral plexus are most commonly benign peripheral nerve sheath schwannomas and neurofibromas, which can be sporadic or can be associated with neurofibromatosis. Malignant peripheral nerve sheath tumors of the lumbosacral plexus are rare and occur more frequently in patients with neurofibromatosis [7]. Other primary malignant or metastatic tumors can also involve the lumbosacral plexus [4]. | 69487 |
acrac_69487_5 | Plexopathy | Variant 6 describes lumbosacral plexopathy in the setting of a known malignancy or post-treatment syndrome; however, plexopathy can be the first clinical presentation of neoplastic disease. Non-neoplastic masses involving the lumbosacral plexus can include hematoma, abscess, aneurysm, amyloidosis [4], and endometriosis [58]. CT Myelography Lumbar Spine There is no relevant literature to support the use of CT myelography of the lumbar spine in the evaluation of nontraumatic lumbosacral plexopathy. Myelography is not routinely performed for the evaluation of nontraumatic plexopathy as it does not evaluate the plexus lateral to the neural foramina. CT Abdomen and Pelvis There is no relevant literature to support the use of abdominal and pelvic CT in the evaluation of nontraumatic lumbosacral plexopathy. CT offers the next highest level of anatomic visualization of the lumbosacral plexus after MRI, and can evaluate for adjacent soft-tissue lesions or tumors that may involve the plexus. CT with IV contrast can be useful for detecting and characterizing soft-tissue masses and tumors, which are in the differential diagnosis of nontraumatic lumbosacral plexopathy and therefore may provide additional information over CT without IV contrast in this setting. CT Lumbar Spine There is no relevant literature regarding the use of CT lumbar spine in the evaluation of nontraumatic lumbosacral plexopathy. CT lumbar spine cannot visualize the preganglionic nerve roots and does not fully evaluate the postganglionic lumbosacral plexus because of its narrow field of view and limited soft-tissue contrast resolution relative to MRI. FDG-PET/CT Whole Body There is no relevant literature to support the use of FDG-PET/CT in the evaluation of nontraumatic lumbosacral plexopathy in the absence of a known malignancy. | Plexopathy. Variant 6 describes lumbosacral plexopathy in the setting of a known malignancy or post-treatment syndrome; however, plexopathy can be the first clinical presentation of neoplastic disease. Non-neoplastic masses involving the lumbosacral plexus can include hematoma, abscess, aneurysm, amyloidosis [4], and endometriosis [58]. CT Myelography Lumbar Spine There is no relevant literature to support the use of CT myelography of the lumbar spine in the evaluation of nontraumatic lumbosacral plexopathy. Myelography is not routinely performed for the evaluation of nontraumatic plexopathy as it does not evaluate the plexus lateral to the neural foramina. CT Abdomen and Pelvis There is no relevant literature to support the use of abdominal and pelvic CT in the evaluation of nontraumatic lumbosacral plexopathy. CT offers the next highest level of anatomic visualization of the lumbosacral plexus after MRI, and can evaluate for adjacent soft-tissue lesions or tumors that may involve the plexus. CT with IV contrast can be useful for detecting and characterizing soft-tissue masses and tumors, which are in the differential diagnosis of nontraumatic lumbosacral plexopathy and therefore may provide additional information over CT without IV contrast in this setting. CT Lumbar Spine There is no relevant literature regarding the use of CT lumbar spine in the evaluation of nontraumatic lumbosacral plexopathy. CT lumbar spine cannot visualize the preganglionic nerve roots and does not fully evaluate the postganglionic lumbosacral plexus because of its narrow field of view and limited soft-tissue contrast resolution relative to MRI. FDG-PET/CT Whole Body There is no relevant literature to support the use of FDG-PET/CT in the evaluation of nontraumatic lumbosacral plexopathy in the absence of a known malignancy. | 69487 |
acrac_69487_6 | Plexopathy | MRI Lumbosacral Plexus Lumbosacral plexus MRI is useful in evaluating nontraumatic lumbosacral plexopathy because of its superior soft- tissue contrast and good spatial resolution, providing good definition of intraneural anatomy as well as localizing pathologic lesions in conditions where electrodiagnostic and physical findings are nonspecific [1-4,7,8,61]. The clinical diagnosis of plexopathy can be challenging, and it may be unclear whether neurologic signs and symptoms Plexopathy localize to a single nerve root (radiculopathy) or to the lumbosacral plexus (plexopathy) because of the considerable overlap in these clinical presentations [1]. For this reason, literature evaluating the diagnostic performance of lumbosacral plexus MRI often includes patients who present with radiculopathy as well as plexopathy. Dessouky et al [62], in a retrospective review, analyzed 202 patients who received MRI of lumbosacral plexus for the evaluation of radiculopathy (57%), pelvic pain (28%), or groin pain (15%) and found that 71% of patients had a change in management resulting from MRI findings. Zhang et al [63] in a retrospective review of 137 patients who received MRI lumbosacral plexus with diffusion-weighted neurography for a clinical diagnosis of sciatica found either nerve root compression or abnormal intraneural signal in the nerves in all patients. Chazen et al [56] in a retrospective review of 64 patients with radiculopathy symptoms who underwent lumbosacral plexus MRI and electromyography found abnormal intraneural signal in 45% of lumbosacral plexus MRI examinations and a statistically significant correlation between nerve signal abnormality on MRI and findings of active radiculopathy on electromyography. | Plexopathy. MRI Lumbosacral Plexus Lumbosacral plexus MRI is useful in evaluating nontraumatic lumbosacral plexopathy because of its superior soft- tissue contrast and good spatial resolution, providing good definition of intraneural anatomy as well as localizing pathologic lesions in conditions where electrodiagnostic and physical findings are nonspecific [1-4,7,8,61]. The clinical diagnosis of plexopathy can be challenging, and it may be unclear whether neurologic signs and symptoms Plexopathy localize to a single nerve root (radiculopathy) or to the lumbosacral plexus (plexopathy) because of the considerable overlap in these clinical presentations [1]. For this reason, literature evaluating the diagnostic performance of lumbosacral plexus MRI often includes patients who present with radiculopathy as well as plexopathy. Dessouky et al [62], in a retrospective review, analyzed 202 patients who received MRI of lumbosacral plexus for the evaluation of radiculopathy (57%), pelvic pain (28%), or groin pain (15%) and found that 71% of patients had a change in management resulting from MRI findings. Zhang et al [63] in a retrospective review of 137 patients who received MRI lumbosacral plexus with diffusion-weighted neurography for a clinical diagnosis of sciatica found either nerve root compression or abnormal intraneural signal in the nerves in all patients. Chazen et al [56] in a retrospective review of 64 patients with radiculopathy symptoms who underwent lumbosacral plexus MRI and electromyography found abnormal intraneural signal in 45% of lumbosacral plexus MRI examinations and a statistically significant correlation between nerve signal abnormality on MRI and findings of active radiculopathy on electromyography. | 69487 |
acrac_69487_7 | Plexopathy | Petrasic et al [64] in retrospective review of 23 patients presenting with chronic pelvic pain and/or dysfunction and clinically suspected chronic cauda equina syndrome who underwent MRI lumbosacral plexus found that 78% of patients had a change in diagnosis and 81% had a change in management from the MRI findings. MRI with and without IV contrast can be useful for detecting and characterizing several of the etiologies in the differential diagnosis of nontraumatic lumbosacral plexopathy and may provide additional information over MRI without IV contrast in this setting. MRI Pelvis There is no relevant literature to support the use of MRI of the pelvis (without dedicated plexus imaging) in the evaluation of nontraumatic lumbosacral plexopathy. CT Myelography Cervical Spine CT myelography provides high-resolution imaging capable of detecting traumatic cervical nerve root avulsions and pseudomeningocele formation and can evaluate for other spinal traumatic injuries, such as fracture, hematoma, or cerebrospinal fluid leak [70]. However, CT myelography can only evaluate for preganglionic nerve root injury and does not directly visualize the postganglionic brachial plexus. Therefore, MRI brachial plexus is preferred over CT myelography cervical spine as the first-line imaging test to evaluate for postganglionic brachial plexus injury. CT myelography performed to assess for cervical nerve root avulsion injury should be ideally delayed until approximately 1 month after the initial trauma to allow time for resolution of hemorrhage and formation of a pseudomeningocele [28,68]. Plexopathy FDG-PET/CT Whole Body There is no relevant literature to support the use of FDG-PET/CT in the evaluation of traumatic brachial plexopathy. MRI Brachial Plexus Brachial plexus MRI is considered superior to CT in the evaluation of traumatic brachial plexopathy because of its inherently better soft-tissue contrast and good spatial resolution [6,9,10,27]. | Plexopathy. Petrasic et al [64] in retrospective review of 23 patients presenting with chronic pelvic pain and/or dysfunction and clinically suspected chronic cauda equina syndrome who underwent MRI lumbosacral plexus found that 78% of patients had a change in diagnosis and 81% had a change in management from the MRI findings. MRI with and without IV contrast can be useful for detecting and characterizing several of the etiologies in the differential diagnosis of nontraumatic lumbosacral plexopathy and may provide additional information over MRI without IV contrast in this setting. MRI Pelvis There is no relevant literature to support the use of MRI of the pelvis (without dedicated plexus imaging) in the evaluation of nontraumatic lumbosacral plexopathy. CT Myelography Cervical Spine CT myelography provides high-resolution imaging capable of detecting traumatic cervical nerve root avulsions and pseudomeningocele formation and can evaluate for other spinal traumatic injuries, such as fracture, hematoma, or cerebrospinal fluid leak [70]. However, CT myelography can only evaluate for preganglionic nerve root injury and does not directly visualize the postganglionic brachial plexus. Therefore, MRI brachial plexus is preferred over CT myelography cervical spine as the first-line imaging test to evaluate for postganglionic brachial plexus injury. CT myelography performed to assess for cervical nerve root avulsion injury should be ideally delayed until approximately 1 month after the initial trauma to allow time for resolution of hemorrhage and formation of a pseudomeningocele [28,68]. Plexopathy FDG-PET/CT Whole Body There is no relevant literature to support the use of FDG-PET/CT in the evaluation of traumatic brachial plexopathy. MRI Brachial Plexus Brachial plexus MRI is considered superior to CT in the evaluation of traumatic brachial plexopathy because of its inherently better soft-tissue contrast and good spatial resolution [6,9,10,27]. | 69487 |
acrac_69487_8 | Plexopathy | MRI can identify traumatic nerve root avulsions, which are crucial to detect in order to plan surgical reconstruction and determine prognosis [66]. Wade et al [71] studied 29 consecutive patients requiring brachial plexus exploration following trauma and found that brachial plexus MRI had a diagnostic accuracy of 79% for detecting C5 to T1 nerve root avulsion and that pseudomeningocele as a surrogate marker for root avulsion had a diagnostic accuracy of 68%. MRI can also directly assess the postganglionic brachial plexus and can confirm whether nerve integrity is maintained, differentiating minor stretching injuries from complete nerve disruptions [66]. Tagliafico et al [44], in a blinded retrospective review, studied 38 patients who received brachial plexus MRI for traumatic plexopathy and found a sensitivity of 84%, specificity 91%, positive predictive value of 91%, negative predictive value of 83%, and accuracy of 87% when compared to the reference standard of surgical findings and clinical follow-up. Fuzari et al [69] performed a systematic review of 3 articles reporting diagnostic accuracy of MRI for traumatic brachial plexus injury and found that the studies lacked methodological rigor, thus concluding that more rigorous research should be conducted in this area. Research is ongoing into new MRI sequences that might improve evaluation of traumatic brachial plexopathy, but these are not routinely performed outside of a research setting. For example, diffusion tensor imaging and tractography have been under investigation and show promise in the evaluation of nerve injury and disruption of nerve microstructure [69]. Brachial plexus MRI can also delineate other post-traumatic complications that may contribute to symptoms of plexopathy, such as regional soft-tissue hematoma, traumatic neuromas, and scarring. | Plexopathy. MRI can identify traumatic nerve root avulsions, which are crucial to detect in order to plan surgical reconstruction and determine prognosis [66]. Wade et al [71] studied 29 consecutive patients requiring brachial plexus exploration following trauma and found that brachial plexus MRI had a diagnostic accuracy of 79% for detecting C5 to T1 nerve root avulsion and that pseudomeningocele as a surrogate marker for root avulsion had a diagnostic accuracy of 68%. MRI can also directly assess the postganglionic brachial plexus and can confirm whether nerve integrity is maintained, differentiating minor stretching injuries from complete nerve disruptions [66]. Tagliafico et al [44], in a blinded retrospective review, studied 38 patients who received brachial plexus MRI for traumatic plexopathy and found a sensitivity of 84%, specificity 91%, positive predictive value of 91%, negative predictive value of 83%, and accuracy of 87% when compared to the reference standard of surgical findings and clinical follow-up. Fuzari et al [69] performed a systematic review of 3 articles reporting diagnostic accuracy of MRI for traumatic brachial plexus injury and found that the studies lacked methodological rigor, thus concluding that more rigorous research should be conducted in this area. Research is ongoing into new MRI sequences that might improve evaluation of traumatic brachial plexopathy, but these are not routinely performed outside of a research setting. For example, diffusion tensor imaging and tractography have been under investigation and show promise in the evaluation of nerve injury and disruption of nerve microstructure [69]. Brachial plexus MRI can also delineate other post-traumatic complications that may contribute to symptoms of plexopathy, such as regional soft-tissue hematoma, traumatic neuromas, and scarring. | 69487 |
acrac_69487_9 | Plexopathy | In the post-treatment setting following surgical nerve repair, brachial plexus MRI can be used to study the repaired nerve, assess for complications, and assess for secondary signs of neuropathy, such as degenerative muscular atrophy [72]. MRI with IV contrast usually does not provide significant additional information over MRI without IV contrast for the initial imaging of traumatic brachial plexopathy, though the addition of contrast can help differentiate between vascular structures and nerves. Plexopathy IV contrast usually does not provide significant additional information over MRI without IV contrast for the initial imaging of traumatic brachial plexopathy. Another cause of indirect traumatic injury to the lumbosacral plexus is avulsion fractures at muscular attachment sites, which can cause traumatic edema, hematoma, or inflammation that compresses the adjacent nerve [8]. This can be seen with avulsions of the hamstrings at the ischial tuberosity (injuring sciatic or pudendal nerve), adductor muscles at the inferior pubic symphysis (injuring obturator nerve), or gluteal muscles at the greater trochanter (injuring superior or inferior gluteal nerves) [8]. Similar to avulsion fractures, tendinopathy of the major muscular attachments can also result in local soft-tissue swelling and inflammation that can involve adjacent nerves [3,8]. Iatrogenic injury to the lumbosacral plexus or terminal branches can also occur after childbirth or surgery, such as total hip arthroplasty, gynecologic, or genitourinary surgery [2]. CT Myelography Lumbar Spine CT myelography provides high-resolution imaging of the thecal sac capable of detecting traumatic nerve root avulsion or pseudomeningocele. However, CT myelography can only evaluate for preganglionic nerve root injury and does not directly visualize the postganglionic lumbosacral plexus. Therefore, MRI lumbosacral plexus is superior to CT myelography lumbar spine in the evaluation of postganglionic lumbosacral plexus injury. | Plexopathy. In the post-treatment setting following surgical nerve repair, brachial plexus MRI can be used to study the repaired nerve, assess for complications, and assess for secondary signs of neuropathy, such as degenerative muscular atrophy [72]. MRI with IV contrast usually does not provide significant additional information over MRI without IV contrast for the initial imaging of traumatic brachial plexopathy, though the addition of contrast can help differentiate between vascular structures and nerves. Plexopathy IV contrast usually does not provide significant additional information over MRI without IV contrast for the initial imaging of traumatic brachial plexopathy. Another cause of indirect traumatic injury to the lumbosacral plexus is avulsion fractures at muscular attachment sites, which can cause traumatic edema, hematoma, or inflammation that compresses the adjacent nerve [8]. This can be seen with avulsions of the hamstrings at the ischial tuberosity (injuring sciatic or pudendal nerve), adductor muscles at the inferior pubic symphysis (injuring obturator nerve), or gluteal muscles at the greater trochanter (injuring superior or inferior gluteal nerves) [8]. Similar to avulsion fractures, tendinopathy of the major muscular attachments can also result in local soft-tissue swelling and inflammation that can involve adjacent nerves [3,8]. Iatrogenic injury to the lumbosacral plexus or terminal branches can also occur after childbirth or surgery, such as total hip arthroplasty, gynecologic, or genitourinary surgery [2]. CT Myelography Lumbar Spine CT myelography provides high-resolution imaging of the thecal sac capable of detecting traumatic nerve root avulsion or pseudomeningocele. However, CT myelography can only evaluate for preganglionic nerve root injury and does not directly visualize the postganglionic lumbosacral plexus. Therefore, MRI lumbosacral plexus is superior to CT myelography lumbar spine in the evaluation of postganglionic lumbosacral plexus injury. | 69487 |
acrac_69487_10 | Plexopathy | CT myelography performed to assess for preganglionic lumbosacral nerve root injury should be ideally delayed until approximately 1 month after the initial trauma to allow time for resolution of hemorrhage and formation of a pseudomeningocele. CT Abdomen and Pelvis There is no relevant literature to support the use of CT abdomen and pelvis in the evaluation of traumatic lumbosacral plexopathy. However, this test is often used in the setting of major blunt trauma (in which lumbosacral plexus injuries are most common), and can detect many of the associated findings such as pelvic fractures or hematomas. The role of imaging in the setting of major blunt trauma is addressed in the ACR Appropriateness Plexopathy FDG-PET/CT Whole Body There is no relevant literature to support the use of FDG-PET/CT in the evaluation of traumatic lumbosacral plexopathy. MRI Lumbosacral Plexus Lumbosacral plexus MRI has been shown to be superior to CT in the evaluation of traumatic lumbosacral plexopathy because of its inherently better soft-tissue contrast and good spatial resolution. MRI can directly assess the postganglionic lumbosacral plexus and can confirm whether nerve integrity is maintained, differentiating minor stretching injuries from complete nerve disruptions [4]. Lumbosacral plexus MRI can also delineate other post- traumatic complications that may contribute to symptoms of plexopathy, such as regional soft-tissue hematoma, edema, inflammation, avulsion fracture, tendinopathy, traumatic neuromas, and scarring [8]. MRI to assess the extent of injury should be ideally delayed until approximately 1 month after the initial trauma to allow time for resolution of hemorrhage and edema that can obscure the lumbosacral plexus acutely. MRI with and without IV contrast usually does not provide significant additional information over MRI without IV contrast for the initial imaging of acute traumatic lumbosacral plexopathy, though the addition of contrast can help differentiate between vascular structures and nerves. | Plexopathy. CT myelography performed to assess for preganglionic lumbosacral nerve root injury should be ideally delayed until approximately 1 month after the initial trauma to allow time for resolution of hemorrhage and formation of a pseudomeningocele. CT Abdomen and Pelvis There is no relevant literature to support the use of CT abdomen and pelvis in the evaluation of traumatic lumbosacral plexopathy. However, this test is often used in the setting of major blunt trauma (in which lumbosacral plexus injuries are most common), and can detect many of the associated findings such as pelvic fractures or hematomas. The role of imaging in the setting of major blunt trauma is addressed in the ACR Appropriateness Plexopathy FDG-PET/CT Whole Body There is no relevant literature to support the use of FDG-PET/CT in the evaluation of traumatic lumbosacral plexopathy. MRI Lumbosacral Plexus Lumbosacral plexus MRI has been shown to be superior to CT in the evaluation of traumatic lumbosacral plexopathy because of its inherently better soft-tissue contrast and good spatial resolution. MRI can directly assess the postganglionic lumbosacral plexus and can confirm whether nerve integrity is maintained, differentiating minor stretching injuries from complete nerve disruptions [4]. Lumbosacral plexus MRI can also delineate other post- traumatic complications that may contribute to symptoms of plexopathy, such as regional soft-tissue hematoma, edema, inflammation, avulsion fracture, tendinopathy, traumatic neuromas, and scarring [8]. MRI to assess the extent of injury should be ideally delayed until approximately 1 month after the initial trauma to allow time for resolution of hemorrhage and edema that can obscure the lumbosacral plexus acutely. MRI with and without IV contrast usually does not provide significant additional information over MRI without IV contrast for the initial imaging of acute traumatic lumbosacral plexopathy, though the addition of contrast can help differentiate between vascular structures and nerves. | 69487 |
acrac_69487_11 | Plexopathy | Plexopathy Variant 5: Brachial plexopathy, known malignancy or post-treatment syndrome. Initial imaging. This variant encompasses brachial plexopathy occurring in the setting of a known malignancy or a post-treatment syndrome occurring months to years after radiation treatment for a regional malignancy. In addition, patients can develop brachial plexopathy in the months to years after radiation treatment for regional malignancy, which could be due to recurrent tumor or postradiation injury. Imaging can be helpful in differentiating radiation plexopathy from recurrent malignancy [20,79-81], for which management differs significantly. CT Myelography Cervical Spine There is no relevant literature to support the use of CT myelography for evaluation of brachial plexopathy in the setting of known malignancy or post-treatment syndrome. CT myelography provides high-resolution imaging of the thecal sac capable of detecting thecal sac compression or intradural masses; however, it does not directly visualize the postganglionic brachial plexus. CT Cervical Spine There is no relevant literature regarding the use of CT cervical spine in the evaluation of brachial plexopathy in the setting of known malignancy or post-treatment syndrome. CT cervical spine cannot visualize the preganglionic nerve roots and does not fully evaluate the postganglionic brachial plexus because of its narrow field of view and limited soft-tissue contrast resolution relative to MRI. Plexopathy MRI Brachial Plexus Brachial plexus is considered an optimal imaging modality to evaluate brachial plexopathy in the setting of known malignancy or post-treatment syndrome because of its superior soft-tissue contrast and good spatial resolution [6,9,10,27]. Extrinsic compression, direct tumor invasion, or metastasis to the brachial plexus can be demonstrated on MRI brachial plexus. Perineural tumor spread or lymphoma infiltrating the plexus [76] can also be visualized on MRI. | Plexopathy. Plexopathy Variant 5: Brachial plexopathy, known malignancy or post-treatment syndrome. Initial imaging. This variant encompasses brachial plexopathy occurring in the setting of a known malignancy or a post-treatment syndrome occurring months to years after radiation treatment for a regional malignancy. In addition, patients can develop brachial plexopathy in the months to years after radiation treatment for regional malignancy, which could be due to recurrent tumor or postradiation injury. Imaging can be helpful in differentiating radiation plexopathy from recurrent malignancy [20,79-81], for which management differs significantly. CT Myelography Cervical Spine There is no relevant literature to support the use of CT myelography for evaluation of brachial plexopathy in the setting of known malignancy or post-treatment syndrome. CT myelography provides high-resolution imaging of the thecal sac capable of detecting thecal sac compression or intradural masses; however, it does not directly visualize the postganglionic brachial plexus. CT Cervical Spine There is no relevant literature regarding the use of CT cervical spine in the evaluation of brachial plexopathy in the setting of known malignancy or post-treatment syndrome. CT cervical spine cannot visualize the preganglionic nerve roots and does not fully evaluate the postganglionic brachial plexus because of its narrow field of view and limited soft-tissue contrast resolution relative to MRI. Plexopathy MRI Brachial Plexus Brachial plexus is considered an optimal imaging modality to evaluate brachial plexopathy in the setting of known malignancy or post-treatment syndrome because of its superior soft-tissue contrast and good spatial resolution [6,9,10,27]. Extrinsic compression, direct tumor invasion, or metastasis to the brachial plexus can be demonstrated on MRI brachial plexus. Perineural tumor spread or lymphoma infiltrating the plexus [76] can also be visualized on MRI. | 69487 |
acrac_69487_12 | Plexopathy | In the post-treatment setting, MRI can help differentiate radiation plexopathy from neoplastic plexopathy [20,79,80]. Research is ongoing into new MRI sequences that might improve evaluation of neoplastic brachial plexopathy, but these are not routinely performed outside of a research setting. For example, Yuh et al [20] in a retrospective review of 23 patients who underwent brachial plexus MRI with diffusion-weighted imaging for evaluation of a mass-like or infiltrative lesion found that apparent diffusion coefficient values were significantly different between malignant tumors and postradiation changes or benign tumors. MRI with and without IV contrast can provide additional information over MRI without IV contrast in the setting of malignancy or post-treatment syndromes, as it can improve delineation of tumor margins and/or fibrosis [5,6,9,10,27]. MRI Cervical Spine There is no relevant literature to support the use of MRI of the cervical spine (without dedicated plexus imaging) for evaluation of brachial plexopathy in the setting of known malignancy or post-treatment syndrome. However, MRI cervical spine may be complementary in this clinical scenario because it can better assess for cervical spinal metastases with extraosseous extension of tumor into the neural foramina and epidural space that can compress the brachial plexus nerve roots or spinal cord. In the post-treatment setting, radiation injury or tumor recurrence involving the intradural nerve roots would also be better evaluated with cervical spine MRI. MRI with and without IV contrast can provide additional information over MRI without IV contrast in the setting of malignancy or post- treatment syndromes, as it can improve delineation of tumor margins and/or fibrosis. US Neck US neck is typically not the first-line imaging test for evaluation of brachial plexopathy in the setting of a known malignancy or post-treatment syndrome, and is generally not useful as the primary imaging modality in this clinical scenario. | Plexopathy. In the post-treatment setting, MRI can help differentiate radiation plexopathy from neoplastic plexopathy [20,79,80]. Research is ongoing into new MRI sequences that might improve evaluation of neoplastic brachial plexopathy, but these are not routinely performed outside of a research setting. For example, Yuh et al [20] in a retrospective review of 23 patients who underwent brachial plexus MRI with diffusion-weighted imaging for evaluation of a mass-like or infiltrative lesion found that apparent diffusion coefficient values were significantly different between malignant tumors and postradiation changes or benign tumors. MRI with and without IV contrast can provide additional information over MRI without IV contrast in the setting of malignancy or post-treatment syndromes, as it can improve delineation of tumor margins and/or fibrosis [5,6,9,10,27]. MRI Cervical Spine There is no relevant literature to support the use of MRI of the cervical spine (without dedicated plexus imaging) for evaluation of brachial plexopathy in the setting of known malignancy or post-treatment syndrome. However, MRI cervical spine may be complementary in this clinical scenario because it can better assess for cervical spinal metastases with extraosseous extension of tumor into the neural foramina and epidural space that can compress the brachial plexus nerve roots or spinal cord. In the post-treatment setting, radiation injury or tumor recurrence involving the intradural nerve roots would also be better evaluated with cervical spine MRI. MRI with and without IV contrast can provide additional information over MRI without IV contrast in the setting of malignancy or post- treatment syndromes, as it can improve delineation of tumor margins and/or fibrosis. US Neck US neck is typically not the first-line imaging test for evaluation of brachial plexopathy in the setting of a known malignancy or post-treatment syndrome, and is generally not useful as the primary imaging modality in this clinical scenario. | 69487 |
acrac_69487_13 | Plexopathy | Researchers have investigated whether US might be useful as a supplemental test to evaluate malignant involvement or radiation-induced plexopathy of the brachial plexus [50,83]. Kultur et al [81] in a prospective analysis of 23 patients receiving radiation therapy for breast cancer found statistically significant differences between the ipsilateral and contralateral brachial plexus on shear wave elastography; however, this technique is not routinely used outside of research settings. Variant 6: Lumbosacral plexopathy, known malignancy or post-treatment syndrome. Initial imaging. This variant encompasses lumbosacral plexopathy occurring in the setting of a known malignancy or a post- treatment syndrome occurring months to years after radiation treatment for a regional malignancy. Plexopathy In addition, patients can develop lumbosacral plexopathy in the months to years after radiation treatment for regional malignancy, which could be due to recurrent tumor or postradiation injury. Imaging can be helpful in differentiating radiation plexopathy from recurrent malignancy [20], for which management differs significantly. CT Myelography Lumbar Spine There is no relevant literature to support the use of CT myelography for evaluation of lumbosacral plexopathy in the setting of known malignancy or post-treatment syndrome. CT myelography provides high-resolution imaging of the thecal sac capable of detecting thecal sac compression or intradural masses; however, it does not directly visualize the postganglionic lumbosacral plexus. CT Lumbar Spine There is no relevant literature regarding the use of CT lumbar spine in the evaluation of lumbosacral plexopathy in the setting of known malignancy or post-treatment syndrome. CT lumbar spine cannot visualize the preganglionic nerve roots and does not fully evaluate the postganglionic lumbosacral plexus because of its narrow field of view and limited soft-tissue contrast resolution relative to MRI. | Plexopathy. Researchers have investigated whether US might be useful as a supplemental test to evaluate malignant involvement or radiation-induced plexopathy of the brachial plexus [50,83]. Kultur et al [81] in a prospective analysis of 23 patients receiving radiation therapy for breast cancer found statistically significant differences between the ipsilateral and contralateral brachial plexus on shear wave elastography; however, this technique is not routinely used outside of research settings. Variant 6: Lumbosacral plexopathy, known malignancy or post-treatment syndrome. Initial imaging. This variant encompasses lumbosacral plexopathy occurring in the setting of a known malignancy or a post- treatment syndrome occurring months to years after radiation treatment for a regional malignancy. Plexopathy In addition, patients can develop lumbosacral plexopathy in the months to years after radiation treatment for regional malignancy, which could be due to recurrent tumor or postradiation injury. Imaging can be helpful in differentiating radiation plexopathy from recurrent malignancy [20], for which management differs significantly. CT Myelography Lumbar Spine There is no relevant literature to support the use of CT myelography for evaluation of lumbosacral plexopathy in the setting of known malignancy or post-treatment syndrome. CT myelography provides high-resolution imaging of the thecal sac capable of detecting thecal sac compression or intradural masses; however, it does not directly visualize the postganglionic lumbosacral plexus. CT Lumbar Spine There is no relevant literature regarding the use of CT lumbar spine in the evaluation of lumbosacral plexopathy in the setting of known malignancy or post-treatment syndrome. CT lumbar spine cannot visualize the preganglionic nerve roots and does not fully evaluate the postganglionic lumbosacral plexus because of its narrow field of view and limited soft-tissue contrast resolution relative to MRI. | 69487 |
acrac_69487_14 | Plexopathy | MRI Lumbar Spine There is no relevant literature to support the use of MRI of the lumbar spine (without dedicated plexus imaging) for evaluation of lumbosacral plexopathy in the setting of known malignancy or post-treatment syndrome. However, lumbar spine MRI may be complementary in this clinical scenario because it can better assess for lumbosacral spinal metastases with extraosseous extension of tumor into the neural foramina and epidural space that can compress the lumbosacral plexus nerve roots [3,7]. In the post-treatment setting, radiation injury or tumor Plexopathy recurrence involving the intradural nerve roots would also be better evaluated with lumbar spine MRI. MRI with and without IV contrast can provide additional information over MRI without IV contrast in the setting of malignancy or post-treatment syndromes, as it can improve delineation of tumor margins and/or fibrosis. MRI Lumbosacral Plexus Lumbosacral plexus MRI is considered an optimal imaging modality to evaluate lumbosacral plexopathy in the setting of known malignancy or post-treatment syndrome because of its superior soft-tissue contrast and good spatial resolution [2-4,7,8]. Extrinsic compression or tumor infiltration of the lumbosacral plexus can be well demonstrated on lumbosacral plexus MRI. Perineural tumor spread along the lumbosacral plexus [84,85] or lymphoma infiltrating the plexus [76] can also be visualized on MRI. MRI with and without IV contrast can provide additional information over MRI without IV contrast in the setting of malignancy or post-treatment syndromes, as it can improve delineation of tumor margins and/or fibrosis. Plexopathy Supporting Documents The evidence table, literature search, and appendix for this topic are available at https://acsearch. acr.org/list. The appendix includes the strength of evidence assessment and the final rating round tabulations for each recommendation. For additional information on the Appropriateness Criteria methodology and other supporting documents go to www. | Plexopathy. MRI Lumbar Spine There is no relevant literature to support the use of MRI of the lumbar spine (without dedicated plexus imaging) for evaluation of lumbosacral plexopathy in the setting of known malignancy or post-treatment syndrome. However, lumbar spine MRI may be complementary in this clinical scenario because it can better assess for lumbosacral spinal metastases with extraosseous extension of tumor into the neural foramina and epidural space that can compress the lumbosacral plexus nerve roots [3,7]. In the post-treatment setting, radiation injury or tumor Plexopathy recurrence involving the intradural nerve roots would also be better evaluated with lumbar spine MRI. MRI with and without IV contrast can provide additional information over MRI without IV contrast in the setting of malignancy or post-treatment syndromes, as it can improve delineation of tumor margins and/or fibrosis. MRI Lumbosacral Plexus Lumbosacral plexus MRI is considered an optimal imaging modality to evaluate lumbosacral plexopathy in the setting of known malignancy or post-treatment syndrome because of its superior soft-tissue contrast and good spatial resolution [2-4,7,8]. Extrinsic compression or tumor infiltration of the lumbosacral plexus can be well demonstrated on lumbosacral plexus MRI. Perineural tumor spread along the lumbosacral plexus [84,85] or lymphoma infiltrating the plexus [76] can also be visualized on MRI. MRI with and without IV contrast can provide additional information over MRI without IV contrast in the setting of malignancy or post-treatment syndromes, as it can improve delineation of tumor margins and/or fibrosis. Plexopathy Supporting Documents The evidence table, literature search, and appendix for this topic are available at https://acsearch. acr.org/list. The appendix includes the strength of evidence assessment and the final rating round tabulations for each recommendation. For additional information on the Appropriateness Criteria methodology and other supporting documents go to www. | 69487 |
acrac_3158180_0 | Noncerebral Vasculitis | Introduction/Background Idiopathic vasculitis is a noninfectious inflammation of the vessels that can lead to serious health consequences. It can be a primary inflammatory process or a secondary process because of an underlying disease. Historically, it has been categorized based on whether inflammation is restricted to blood vessels of particular size as large-vessel vasculitis (LVV), medium-vessel vasculitis (MVV), small-vessel vasculitis, and variable-vessel vasculitis. The large vessels are the aorta and its main branches, the medium vessels are the main visceral arteries and initial branches, and the small vessels are the intraparenchymal vessels and analog veins. There is an overlap between these vasculitis types; for example, LVV predominantly involves large vessels; however, it can also affect medium and small vessels. In variable-vessel vasculitis, there is no predominance of vessel size involvement. Tissue biopsy of the large or medium vessels is often not feasible; therefore, imaging plays a crucial role in diagnosing idiopathic vasculitides. Considering the limitations of spatial resolution across all available modalities for small-vessel vasculitis, this manuscript focused on vasculitis mainly involving the large and medium vessels. LVV includes 2 subtypes: giant-cell arteritis (GCA) and Takayasu arteritis (TAK), of which GCA is more common. GCA is an idiopathic, inflammatory, granulomatous vasculitis involving predominantly the large arteries in older patients (>50 years of age). GCA affects the supra-aortic vessels, especially the extracranial branches of the carotid artery, such as the superficial temporal artery (referred to as cranial-GCA [c-GCA]). Classically, a diagnosis of GCA requires temporal artery ultrasound (US) or temporal artery biopsy. However, a growing body of literature has demonstrated the involvement of the extracranial large arteries, particularly the aorta and its main branches, which is known as large-vessel GCA (LV-GCA). | Noncerebral Vasculitis. Introduction/Background Idiopathic vasculitis is a noninfectious inflammation of the vessels that can lead to serious health consequences. It can be a primary inflammatory process or a secondary process because of an underlying disease. Historically, it has been categorized based on whether inflammation is restricted to blood vessels of particular size as large-vessel vasculitis (LVV), medium-vessel vasculitis (MVV), small-vessel vasculitis, and variable-vessel vasculitis. The large vessels are the aorta and its main branches, the medium vessels are the main visceral arteries and initial branches, and the small vessels are the intraparenchymal vessels and analog veins. There is an overlap between these vasculitis types; for example, LVV predominantly involves large vessels; however, it can also affect medium and small vessels. In variable-vessel vasculitis, there is no predominance of vessel size involvement. Tissue biopsy of the large or medium vessels is often not feasible; therefore, imaging plays a crucial role in diagnosing idiopathic vasculitides. Considering the limitations of spatial resolution across all available modalities for small-vessel vasculitis, this manuscript focused on vasculitis mainly involving the large and medium vessels. LVV includes 2 subtypes: giant-cell arteritis (GCA) and Takayasu arteritis (TAK), of which GCA is more common. GCA is an idiopathic, inflammatory, granulomatous vasculitis involving predominantly the large arteries in older patients (>50 years of age). GCA affects the supra-aortic vessels, especially the extracranial branches of the carotid artery, such as the superficial temporal artery (referred to as cranial-GCA [c-GCA]). Classically, a diagnosis of GCA requires temporal artery ultrasound (US) or temporal artery biopsy. However, a growing body of literature has demonstrated the involvement of the extracranial large arteries, particularly the aorta and its main branches, which is known as large-vessel GCA (LV-GCA). | 3158180 |
acrac_3158180_1 | Noncerebral Vasculitis | The American College of Rheumatology classification criteria underperforms in classifying patients with LV-GCA. A retrospective study by Muratore et al [1] demonstrated that American College of Rheumatology criteria are inadequate to classify patients with LV-GCA. Therefore, patients with suspected GCA require supplemental imaging studies in addition to temporal artery US or biopsy [2]. In this regard, the trial of tocilizumab for GCA, which is a large randomized controlled trial in GCA, included patients with imaging-confirmed LV-GCA who did not meet the American College of Rheumatology criteria [3]. Interestingly, patients with LV-GCA compared to other patients with GCA, present less frequently with jaw claudication or ischemic symptoms and have a higher incidence of relapse, have greater cumulative glucocorticoid exposure, and are more frequently treated with steroid-sparing agents, such as the interleukin-6-receptor blocker tocilizumab [1]. Polymyalgia rheumatica (PMR) is a disorder in the same disease spectrum as GCA and can be found in association with GCA or as an isolated phenomenon [4]. Reprint requests to: [email protected] Noncerebral Vasculitis from asymptomatic presentations to nonspecific constitutional symptoms and major ischemic events. Although the histopathology of TAK shares similarities with that of GCA, biopsy material from the large arteries is rarely obtained in TAK. The diagnosis of TAK typically requires a combination of physical examination, laboratory findings, and imaging findings. Variable-vessel vasculitis subtypes such as Cogan or Behcet disease can involve the large vessels, particularly the aorta. Although the literature is limited on imaging features of Cogan or Behcet vasculitis, case reports and series have demonstrated similarities with LVV. Because of the limitation of the peer-reviewed literature, the focus of this manuscript will be on LVV. MVV predominantly involves medium-sized arteries, although arteries of any size can be involved. | Noncerebral Vasculitis. The American College of Rheumatology classification criteria underperforms in classifying patients with LV-GCA. A retrospective study by Muratore et al [1] demonstrated that American College of Rheumatology criteria are inadequate to classify patients with LV-GCA. Therefore, patients with suspected GCA require supplemental imaging studies in addition to temporal artery US or biopsy [2]. In this regard, the trial of tocilizumab for GCA, which is a large randomized controlled trial in GCA, included patients with imaging-confirmed LV-GCA who did not meet the American College of Rheumatology criteria [3]. Interestingly, patients with LV-GCA compared to other patients with GCA, present less frequently with jaw claudication or ischemic symptoms and have a higher incidence of relapse, have greater cumulative glucocorticoid exposure, and are more frequently treated with steroid-sparing agents, such as the interleukin-6-receptor blocker tocilizumab [1]. Polymyalgia rheumatica (PMR) is a disorder in the same disease spectrum as GCA and can be found in association with GCA or as an isolated phenomenon [4]. Reprint requests to: [email protected] Noncerebral Vasculitis from asymptomatic presentations to nonspecific constitutional symptoms and major ischemic events. Although the histopathology of TAK shares similarities with that of GCA, biopsy material from the large arteries is rarely obtained in TAK. The diagnosis of TAK typically requires a combination of physical examination, laboratory findings, and imaging findings. Variable-vessel vasculitis subtypes such as Cogan or Behcet disease can involve the large vessels, particularly the aorta. Although the literature is limited on imaging features of Cogan or Behcet vasculitis, case reports and series have demonstrated similarities with LVV. Because of the limitation of the peer-reviewed literature, the focus of this manuscript will be on LVV. MVV predominantly involves medium-sized arteries, although arteries of any size can be involved. | 3158180 |
acrac_3158180_2 | Noncerebral Vasculitis | Polyarteritis nodosa (PAN) and Kawasaki disease are the 2 types of MVV [6]. PAN affects medium and small visceral vessels (particularly renal arteries), and there is an association with hepatitis B virus. Kawasaki disease is a self-limiting acute necrotizing vasculitis that affects medium and small vessels and is most prevalent in Asian populations. Kawasaki disease commonly affects the coronary arteries in 15% to 20% of patients [7]. Special Imaging Considerations CT and CTA CT or CT angiography (CTA) are cross-sectional imaging modalities with an excellent spatial resolution and faster scan time. Although assessment of the vessel wall is possible with contrast-enhanced CT, the proper modality is CTA. CT without intravenous (IV) contrast material is limited for vascular assessment. However, CTA can be acquired as a biphasic study, including noncontrast and arterial phase, or as a triphasic study with the addition of a delayed or venous phase for proper vessel lumen and wall assessment. In addition, electrocardiogram (ECG)-gated CTA exhibits motionless aortic root and ascending aorta [8]. In most cases, CTA of the chest, abdomen, and pelvis ensures coverage of the entire vasculature in the abdomen. In cases of coronary artery involvement, dedicated coronary CTA displays potential vessel involvement. FDG-PET/CT Whole Body Inflammatory cells in the inflamed vessel wall can accumulate fluorine-18-2-fluoro-2-deoxy-D-glucose (FDG); thus, FDG-PET/CT can be used in the diagnosis of vasculitis [9]. Similar to other indications, patient preparation includes 6 hours of fasting before administering the FDG tracer injection and limiting strenuous exercise before the study. Although FDG-PET/CT can be acquired as a standard 60- to 90-minute delay after the injection for vasculitis, a few studies have suggested that a 120- to 180-minute delay could lead to higher diagnostic accuracy [10,11]. | Noncerebral Vasculitis. Polyarteritis nodosa (PAN) and Kawasaki disease are the 2 types of MVV [6]. PAN affects medium and small visceral vessels (particularly renal arteries), and there is an association with hepatitis B virus. Kawasaki disease is a self-limiting acute necrotizing vasculitis that affects medium and small vessels and is most prevalent in Asian populations. Kawasaki disease commonly affects the coronary arteries in 15% to 20% of patients [7]. Special Imaging Considerations CT and CTA CT or CT angiography (CTA) are cross-sectional imaging modalities with an excellent spatial resolution and faster scan time. Although assessment of the vessel wall is possible with contrast-enhanced CT, the proper modality is CTA. CT without intravenous (IV) contrast material is limited for vascular assessment. However, CTA can be acquired as a biphasic study, including noncontrast and arterial phase, or as a triphasic study with the addition of a delayed or venous phase for proper vessel lumen and wall assessment. In addition, electrocardiogram (ECG)-gated CTA exhibits motionless aortic root and ascending aorta [8]. In most cases, CTA of the chest, abdomen, and pelvis ensures coverage of the entire vasculature in the abdomen. In cases of coronary artery involvement, dedicated coronary CTA displays potential vessel involvement. FDG-PET/CT Whole Body Inflammatory cells in the inflamed vessel wall can accumulate fluorine-18-2-fluoro-2-deoxy-D-glucose (FDG); thus, FDG-PET/CT can be used in the diagnosis of vasculitis [9]. Similar to other indications, patient preparation includes 6 hours of fasting before administering the FDG tracer injection and limiting strenuous exercise before the study. Although FDG-PET/CT can be acquired as a standard 60- to 90-minute delay after the injection for vasculitis, a few studies have suggested that a 120- to 180-minute delay could lead to higher diagnostic accuracy [10,11]. | 3158180 |
Subsets and Splits