Imaging

SAN FRANCISCO — Chest pain represents one of the most common presenting symptoms in emergency departments, and it also represents a diagnostic challenge: Is it a pulmonary embolism? Is it an aortic dissection? Is it coronary artery disease? Or is it nothing?

Now, new CT technology promises to revolutionize this diagnosis, giving the ability to rule out all three conditions with a single 15-second scan.

In theory, this scan can replace stress testing for coronary artery disease, echocardiography or CT for aortic dissection, and CT pulmonary angiography or a ventilation-perfusion scan for pulmonary embolism, Matthew J. Budoff, M.D., said at a cardiovascular imaging conference sponsored by the American College of Cardiology.

Although no diagnostic or prognostic studies on the triple rule out have yet been published, there's some indication that the single scan will have 90% accuracy or better for each of the three conditions, said Dr. Budoff of Harbor-UCLA Medical Center in Torrance, Calif.

The technology involves a 64-slice CT scan from the apex to the base of the lungs. Patients will have to hold their breath for 20–30 seconds as contrast is injected and the images are acquired. Acquisition of the slices will be gated to the heart's rhythm, allowing for stable, high-resolution images of the heart and lungs. The slice thickness will be 0.625 mm.

Software and a sophisticated workstation will allow the clinician to construct three-dimensional images of the heart, lungs, or aorta, and to manipulate three-dimensional and two-dimensional images in a variety of ways.

In addition to aortic dissection, pulmonary embolism, and coronary artery disease, the technique will allow clear views of the pericardium, permitting the diagnosis of calcified or thickened pericardium and sometimes pericarditis.

In addition, “you might pick up pneumonia, and you might pick up pulmonary adhesions or even pericardial adhesions,” Dr. Budoff said. “There are a lot of things you could possibly see. And it could be done during the chest pain episode, which is a great advantage over some of the other modalities where you'd want to wait until their chest pain is quiescent.”

Dr. Budoff described the case of an elderly woman who complained of chest pain and shortness of breath. Because of her age, he was reluctant to order a stress test. The CT angiography showed that her coronary arteries were normal and that her ejection fraction was acceptably high. When he examined the lung images closely, however, he discovered several pulmonary emboli.

“We admitted her to the hospital, put her on heparin, and it all cleared up,” he said.

Despite its promise, the triple rule out does have some limitations. For one thing, it subjects patients to a relatively high dose of radiation—in the neighborhood of 24–30 millisieverts, equivalent to 240–300 chest x-rays.

Because it's a gated study, more contrast must be used and the injection time is longer than for a standard CT. Some patients may have trouble holding their breath for 20–30 seconds.

Then there's the issue of who is going to read these images when a patient presents at 3 a.m. The radiologist staffing the emergency department may not be facile with cardiac CT angiography. Although the images could be transferred over data lines, the cardiologist is not likely to have a workstation with the proper software at home. In all likelihood, someone will have to come to the hospital to read the study.

Still, Dr. Budoff expects the triple rule out to become a routine test in the emergency department, a prospect he greets with mixed emotions.

“We really need to see how this is going to pan out, and work out the reading issues before we start applying this to everybody who comes in with a twinge in their chest or shortness of breath,” he said. “I'm a little leery … to say just because we can do it we should.”

Views of aortic dissection: Left, it is shown as a long, thin dissection flap in the descending aorta; right, the true lumen (larger area) and false lumen are shown endoscopically. Photos courtesy Dr. Matthew J. Budoff

SAN FRANCISCO — Chest pain represents one of the most common presenting symptoms in emergency departments, and it also represents a diagnostic challenge: Is it a pulmonary embolism? Is it an aortic dissection? Is it coronary artery disease? Or is it nothing?

Now, new CT technology promises to revolutionize this diagnosis, giving the ability to rule out all three conditions with a single 15-second scan.

In theory, this scan can replace stress testing for coronary artery disease, echocardiography or CT for aortic dissection, and CT pulmonary angiography or a ventilation-perfusion scan for pulmonary embolism, Matthew J. Budoff, M.D., said at a cardiovascular imaging conference sponsored by the American College of Cardiology.

Although no diagnostic or prognostic studies on the triple rule out have yet been published, there's some indication that the single scan will have 90% accuracy or better for each of the three conditions, said Dr. Budoff of Harbor-UCLA Medical Center in Torrance, Calif.

The technology involves a 64-slice CT scan from the apex to the base of the lungs. Patients will have to hold their breath for 20–30 seconds as contrast is injected and the images are acquired. Acquisition of the slices will be gated to the heart's rhythm, allowing for stable, high-resolution images of the heart and lungs. The slice thickness will be 0.625 mm.

Software and a sophisticated workstation will allow the clinician to construct three-dimensional images of the heart, lungs, or aorta, and to manipulate three-dimensional and two-dimensional images in a variety of ways.

In addition to aortic dissection, pulmonary embolism, and coronary artery disease, the technique will allow clear views of the pericardium, permitting the diagnosis of calcified or thickened pericardium and sometimes pericarditis.

In addition, “you might pick up pneumonia, and you might pick up pulmonary adhesions or even pericardial adhesions,” Dr. Budoff said. “There are a lot of things you could possibly see. And it could be done during the chest pain episode, which is a great advantage over some of the other modalities where you'd want to wait until their chest pain is quiescent.”

Dr. Budoff described the case of an elderly woman who complained of chest pain and shortness of breath. Because of her age, he was reluctant to order a stress test. The CT angiography showed that her coronary arteries were normal and that her ejection fraction was acceptably high. When he examined the lung images closely, however, he discovered several pulmonary emboli.

“We admitted her to the hospital, put her on heparin, and it all cleared up,” he said.

Despite its promise, the triple rule out does have some limitations. For one thing, it subjects patients to a relatively high dose of radiation—in the neighborhood of 24–30 millisieverts, equivalent to 240–300 chest x-rays.

Because it's a gated study, more contrast must be used and the injection time is longer than for a standard CT. Some patients may have trouble holding their breath for 20–30 seconds.

Then there's the issue of who is going to read these images when a patient presents at 3 a.m. The radiologist staffing the emergency department may not be facile with cardiac CT angiography. Although the images could be transferred over data lines, the cardiologist is not likely to have a workstation with the proper software at home. In all likelihood, someone will have to come to the hospital to read the study.

Still, Dr. Budoff expects the triple rule out to become a routine test in the emergency department, a prospect he greets with mixed emotions.

“We really need to see how this is going to pan out, and work out the reading issues before we start applying this to everybody who comes in with a twinge in their chest or shortness of breath,” he said. “I'm a little leery … to say just because we can do it we should.”

Views of aortic dissection: Left, it is shown as a long, thin dissection flap in the descending aorta; right, the true lumen (larger area) and false lumen are shown endoscopically. Photos courtesy Dr. Matthew J. Budoff

SAN FRANCISCO — Chest pain represents one of the most common presenting symptoms in emergency departments, and it also represents a diagnostic challenge: Is it a pulmonary embolism? Is it an aortic dissection? Is it coronary artery disease? Or is it nothing?

Now, new CT technology promises to revolutionize this diagnosis, giving the ability to rule out all three conditions with a single 15-second scan.

In theory, this scan can replace stress testing for coronary artery disease, echocardiography or CT for aortic dissection, and CT pulmonary angiography or a ventilation-perfusion scan for pulmonary embolism, Matthew J. Budoff, M.D., said at a cardiovascular imaging conference sponsored by the American College of Cardiology.

Although no diagnostic or prognostic studies on the triple rule out have yet been published, there's some indication that the single scan will have 90% accuracy or better for each of the three conditions, said Dr. Budoff of Harbor-UCLA Medical Center in Torrance, Calif.

The technology involves a 64-slice CT scan from the apex to the base of the lungs. Patients will have to hold their breath for 20–30 seconds as contrast is injected and the images are acquired. Acquisition of the slices will be gated to the heart's rhythm, allowing for stable, high-resolution images of the heart and lungs. The slice thickness will be 0.625 mm.

Software and a sophisticated workstation will allow the clinician to construct three-dimensional images of the heart, lungs, or aorta, and to manipulate three-dimensional and two-dimensional images in a variety of ways.

In addition to aortic dissection, pulmonary embolism, and coronary artery disease, the technique will allow clear views of the pericardium, permitting the diagnosis of calcified or thickened pericardium and sometimes pericarditis.

In addition, “you might pick up pneumonia, and you might pick up pulmonary adhesions or even pericardial adhesions,” Dr. Budoff said. “There are a lot of things you could possibly see. And it could be done during the chest pain episode, which is a great advantage over some of the other modalities where you'd want to wait until their chest pain is quiescent.”

Dr. Budoff described the case of an elderly woman who complained of chest pain and shortness of breath. Because of her age, he was reluctant to order a stress test. The CT angiography showed that her coronary arteries were normal and that her ejection fraction was acceptably high. When he examined the lung images closely, however, he discovered several pulmonary emboli.

“We admitted her to the hospital, put her on heparin, and it all cleared up,” he said.

Despite its promise, the triple rule out does have some limitations. For one thing, it subjects patients to a relatively high dose of radiation—in the neighborhood of 24–30 millisieverts, equivalent to 240–300 chest x-rays.

Because it's a gated study, more contrast must be used and the injection time is longer than for a standard CT. Some patients may have trouble holding their breath for 20–30 seconds.

Then there's the issue of who is going to read these images when a patient presents at 3 a.m. The radiologist staffing the emergency department may not be facile with cardiac CT angiography. Although the images could be transferred over data lines, the cardiologist is not likely to have a workstation with the proper software at home. In all likelihood, someone will have to come to the hospital to read the study.

Still, Dr. Budoff expects the triple rule out to become a routine test in the emergency department, a prospect he greets with mixed emotions.

“We really need to see how this is going to pan out, and work out the reading issues before we start applying this to everybody who comes in with a twinge in their chest or shortness of breath,” he said. “I'm a little leery … to say just because we can do it we should.”

Views of aortic dissection: Left, it is shown as a long, thin dissection flap in the descending aorta; right, the true lumen (larger area) and false lumen are shown endoscopically. Photos courtesy Dr. Matthew J. Budoff

SAN FRANCISCO — Quantitative nuclear cardiology allows for highly sensitive, specific, and reproducible estimates of a patient's risk and assists in the decision of who should be sent for revascularization, Daniel S. Berman, M.D., said at a cardiovascular imaging conference sponsored by the American College of Cardiology.

When quantitative techniques are used with single proton emission computed tomography (SPECT), the results are operator independent, said Dr. Berman of Cedars-Sinai Medical Center, Los Angeles. A quantitative SPECT assessment of myocardial perfusion and function reduces the reliance on expert observers, standardizes results from center to center, facilitates serial assessments, and ultimately improves patient outcomes.

The technology produces reliable assessments of a number of important parameters of cardiac function. (See box.) And numerous studies have shown how these parameters relate to cardiac risk.

For example, pooled data from more than 17,000 patients show that those with a normal stress myocardial perfusion SPECT had only a 0.6% chance of suffering cardiac death or a nonfatal MI over a mean follow-up of 27 months. This low rate of cardiac events is especially impressive because these were patients with known or suspected coronary artery disease.

This study included patients who were under either exercise or pharmacologic stress. According to another study, a normal stress myocardial perfusion SPECT has less prognostic value if the patient fails to reach at least 70% of the predicted maximal heart rate (PMHR) during exercise. Among more than 5,000 patients, the cardiac event rate for patients who failed to reach 70% PMHR was more than three times that of those who reached 70%–100% PMHR.

Patients who are unable to reach 70% PMHR during exercise need to undergo myocardial perfusion SPECT with pharmacologic stress, Dr. Berman said.

The presence of diabetes is another factor that modifies a patient's risk after myocardial perfusion SPECT. For any given summed stress score (SSS)—an estimate of the overall size and severity of a perfusion defect during stress—nondiabetics have the lowest level of risk, insulin-dependent diabetics have the highest level of risk, and non-insulin-dependent diabetics have an intermediate risk.

SSS alone isn't enough, however. The summed difference score, which subtracts the summed rest score from the SSS, is a more reliable measure.

Better still is to normalize these scores based on the maximum possible score. This yields measures of percent myocardium perfused that are independent of the specific SPECT system employed. When applied to the summed difference score, the percent myocardium perfused is a measure of ischemia.

This measure of ischemia is important in deciding whether to refer patients to revascularization or to treat them with medical therapy. On the one hand, studies have shown that patients with extensive ischemia have a much lower risk of cardiac death with revascularization than with medical therapy. On the other hand, patients with less than about 10% ischemia have a lower risk of cardiac death with medical therapy.

Despite its value, cardiac perfusion SPECT does have a number of limitations, Dr. Berman said. Because the test detects only hydrodynamically significant lesions, it won't pick up early atherosclerosis. It also won't pick up some of the patients at the very highest risk, those with a balanced reduction in perfusion. And it may underestimate the extent of ischemia and cardiovascular disease as well as the amount of viable myocardium.

Measurements Possible With Quantitative Cardiac Perfusion SPECT

Percent hypoperfusion

Percent reversibility

Lung-to-heart ratio

Transient ischemic dilatation

Left ventricular mass

Left ventricular ejection fraction

End diastolic volume

End systolic volume

Wall motion

Wall thickening

Peak filling rate

Source: Dr. Berman

SAN FRANCISCO — Quantitative nuclear cardiology allows for highly sensitive, specific, and reproducible estimates of a patient's risk and assists in the decision of who should be sent for revascularization, Daniel S. Berman, M.D., said at a cardiovascular imaging conference sponsored by the American College of Cardiology.

When quantitative techniques are used with single proton emission computed tomography (SPECT), the results are operator independent, said Dr. Berman of Cedars-Sinai Medical Center, Los Angeles. A quantitative SPECT assessment of myocardial perfusion and function reduces the reliance on expert observers, standardizes results from center to center, facilitates serial assessments, and ultimately improves patient outcomes.

The technology produces reliable assessments of a number of important parameters of cardiac function. (See box.) And numerous studies have shown how these parameters relate to cardiac risk.

For example, pooled data from more than 17,000 patients show that those with a normal stress myocardial perfusion SPECT had only a 0.6% chance of suffering cardiac death or a nonfatal MI over a mean follow-up of 27 months. This low rate of cardiac events is especially impressive because these were patients with known or suspected coronary artery disease.

This study included patients who were under either exercise or pharmacologic stress. According to another study, a normal stress myocardial perfusion SPECT has less prognostic value if the patient fails to reach at least 70% of the predicted maximal heart rate (PMHR) during exercise. Among more than 5,000 patients, the cardiac event rate for patients who failed to reach 70% PMHR was more than three times that of those who reached 70%–100% PMHR.

Patients who are unable to reach 70% PMHR during exercise need to undergo myocardial perfusion SPECT with pharmacologic stress, Dr. Berman said.

The presence of diabetes is another factor that modifies a patient's risk after myocardial perfusion SPECT. For any given summed stress score (SSS)—an estimate of the overall size and severity of a perfusion defect during stress—nondiabetics have the lowest level of risk, insulin-dependent diabetics have the highest level of risk, and non-insulin-dependent diabetics have an intermediate risk.

SSS alone isn't enough, however. The summed difference score, which subtracts the summed rest score from the SSS, is a more reliable measure.

Better still is to normalize these scores based on the maximum possible score. This yields measures of percent myocardium perfused that are independent of the specific SPECT system employed. When applied to the summed difference score, the percent myocardium perfused is a measure of ischemia.

This measure of ischemia is important in deciding whether to refer patients to revascularization or to treat them with medical therapy. On the one hand, studies have shown that patients with extensive ischemia have a much lower risk of cardiac death with revascularization than with medical therapy. On the other hand, patients with less than about 10% ischemia have a lower risk of cardiac death with medical therapy.

Despite its value, cardiac perfusion SPECT does have a number of limitations, Dr. Berman said. Because the test detects only hydrodynamically significant lesions, it won't pick up early atherosclerosis. It also won't pick up some of the patients at the very highest risk, those with a balanced reduction in perfusion. And it may underestimate the extent of ischemia and cardiovascular disease as well as the amount of viable myocardium.

Measurements Possible With Quantitative Cardiac Perfusion SPECT

Percent hypoperfusion

Percent reversibility

Lung-to-heart ratio

Transient ischemic dilatation

Left ventricular mass

Left ventricular ejection fraction

End diastolic volume

End systolic volume

Wall motion

Wall thickening

Peak filling rate

Source: Dr. Berman

SAN FRANCISCO — Quantitative nuclear cardiology allows for highly sensitive, specific, and reproducible estimates of a patient's risk and assists in the decision of who should be sent for revascularization, Daniel S. Berman, M.D., said at a cardiovascular imaging conference sponsored by the American College of Cardiology.

When quantitative techniques are used with single proton emission computed tomography (SPECT), the results are operator independent, said Dr. Berman of Cedars-Sinai Medical Center, Los Angeles. A quantitative SPECT assessment of myocardial perfusion and function reduces the reliance on expert observers, standardizes results from center to center, facilitates serial assessments, and ultimately improves patient outcomes.

The technology produces reliable assessments of a number of important parameters of cardiac function. (See box.) And numerous studies have shown how these parameters relate to cardiac risk.

For example, pooled data from more than 17,000 patients show that those with a normal stress myocardial perfusion SPECT had only a 0.6% chance of suffering cardiac death or a nonfatal MI over a mean follow-up of 27 months. This low rate of cardiac events is especially impressive because these were patients with known or suspected coronary artery disease.

This study included patients who were under either exercise or pharmacologic stress. According to another study, a normal stress myocardial perfusion SPECT has less prognostic value if the patient fails to reach at least 70% of the predicted maximal heart rate (PMHR) during exercise. Among more than 5,000 patients, the cardiac event rate for patients who failed to reach 70% PMHR was more than three times that of those who reached 70%–100% PMHR.

Patients who are unable to reach 70% PMHR during exercise need to undergo myocardial perfusion SPECT with pharmacologic stress, Dr. Berman said.

The presence of diabetes is another factor that modifies a patient's risk after myocardial perfusion SPECT. For any given summed stress score (SSS)—an estimate of the overall size and severity of a perfusion defect during stress—nondiabetics have the lowest level of risk, insulin-dependent diabetics have the highest level of risk, and non-insulin-dependent diabetics have an intermediate risk.

SSS alone isn't enough, however. The summed difference score, which subtracts the summed rest score from the SSS, is a more reliable measure.

Better still is to normalize these scores based on the maximum possible score. This yields measures of percent myocardium perfused that are independent of the specific SPECT system employed. When applied to the summed difference score, the percent myocardium perfused is a measure of ischemia.

This measure of ischemia is important in deciding whether to refer patients to revascularization or to treat them with medical therapy. On the one hand, studies have shown that patients with extensive ischemia have a much lower risk of cardiac death with revascularization than with medical therapy. On the other hand, patients with less than about 10% ischemia have a lower risk of cardiac death with medical therapy.

Despite its value, cardiac perfusion SPECT does have a number of limitations, Dr. Berman said. Because the test detects only hydrodynamically significant lesions, it won't pick up early atherosclerosis. It also won't pick up some of the patients at the very highest risk, those with a balanced reduction in perfusion. And it may underestimate the extent of ischemia and cardiovascular disease as well as the amount of viable myocardium.

Measurements Possible With Quantitative Cardiac Perfusion SPECT

Percent hypoperfusion

Percent reversibility

Lung-to-heart ratio

Transient ischemic dilatation

Left ventricular mass

Left ventricular ejection fraction

End diastolic volume

End systolic volume

Wall motion

Wall thickening

Peak filling rate

Source: Dr. Berman

KEVIN FOLEY, RESEARCH

KEVIN FOLEY, RESEARCH

KEVIN FOLEY, RESEARCH

SAN FRANCISCO — With computed tomography angiography, “patients literally go home with a Band-Aid and a bottle of water” after just 20 minutes, Matthew J. Budoff, M.D., remarked at a cardiovascular imaging conference sponsored by the American College of Cardiology.

With high sensitivity and specificity and images that rival the resolution obtainable with traditional coronary angiography from the catheterization lab, CT angiography will allow many more patients to avoid an invasive procedure, said Dr. Budoff of Harbor-UCLA Medical Center, Torrance, Calif.

After an injection of 80–100 mL of nonionic iodinated contrast solution, up to 4,000 two-dimensional images can be obtained within 20–30 seconds as the patient holds his or her breath. The entire procedure takes 20 minutes, and interpretation takes another 10 minutes.

Sophisticated workstations assemble the stack of 2D images into a three-dimensional reconstruction. Interpretations are made on the basis of the 3D reconstruction with reference to the 2D images.

Dr. Budoff started working with CT angiography in the mid-1990s. In those days it took 3 weeks of full-time computation to assemble a single 3D reconstruction. This same function takes just 30 seconds today.

And these workstations allow the cardiologist to rotate the heart image in three dimensions, to zoom in to interesting features, and to easily reference the original 2D data from any point of interest.

The initial studies of four-slice CT angiography revealed the limitations of this early technique. Only 30% of patients had all three major arteries available for analysis, and in detecting stenosis the sensitivity was just 58% with 76% specificity.

Now, as 16-slice and even 64-slice CT angiography become available, the sensitivity and specificity have improved considerably. Studies have calculated sensitivities as high as 97% and specificities as high as 94%.

Most important, the negative predictive value is 98%–100%. “The benefit of CT angio is that when the coronaries look normal, the coronaries are normal,” Dr. Budoff said.

The temporal resolution of the CT images is about 175 milliseconds, so reducing the heart rate to below 60 beats per minute is important for accuracy and interpretability. Most centers use 100 mg metoprolol 1 hour prior to the study and/or a 5-mg intravenous metoprolol push every 5 minutes until the patient achieves a slow heart rate.

A regular rhythm is also important. With multiple detectors obtaining images at specific parts of the heart cycle, the modality reaches an effective frame speed of 15 images per second. This is slower than the cath lab, but fast enough that the images are free of motion artifact.

CT angiography may be the best technique for imaging the results of bypass grafting as the anastomoses are clearly visible. Other clinical indications for CT angiography are in cases of equivocal results following stress testing; to evaluate patency post angioplasty, post stent, and post bypass surgery; in cases of congenital abnormalities and anomalous coronaries; before and after atrial fibrillation ablation; and before placing a biventricular pacer.

CT angiography is not without its disadvantages, however. It's not very good for visualizing vessels with diameters less than 1.5 mm. It is subject to artifacts from extensive calcification, stents, or extensive clips after bypass grafting. And it subjects patients to a relatively high dose of radiation—about 9.3–11.3 mSv, compared with 2.1–2.3 mSv for the cath lab and 0.1 mSv for a chest x-ray.

CT angiography reveales high-grade stenosis (dark area) in the mid-left anterior descending coronary artery. Courtesy Dr. Matthew J. Budoff

SAN FRANCISCO — With computed tomography angiography, “patients literally go home with a Band-Aid and a bottle of water” after just 20 minutes, Matthew J. Budoff, M.D., remarked at a cardiovascular imaging conference sponsored by the American College of Cardiology.

With high sensitivity and specificity and images that rival the resolution obtainable with traditional coronary angiography from the catheterization lab, CT angiography will allow many more patients to avoid an invasive procedure, said Dr. Budoff of Harbor-UCLA Medical Center, Torrance, Calif.

After an injection of 80–100 mL of nonionic iodinated contrast solution, up to 4,000 two-dimensional images can be obtained within 20–30 seconds as the patient holds his or her breath. The entire procedure takes 20 minutes, and interpretation takes another 10 minutes.

Sophisticated workstations assemble the stack of 2D images into a three-dimensional reconstruction. Interpretations are made on the basis of the 3D reconstruction with reference to the 2D images.

Dr. Budoff started working with CT angiography in the mid-1990s. In those days it took 3 weeks of full-time computation to assemble a single 3D reconstruction. This same function takes just 30 seconds today.

And these workstations allow the cardiologist to rotate the heart image in three dimensions, to zoom in to interesting features, and to easily reference the original 2D data from any point of interest.

The initial studies of four-slice CT angiography revealed the limitations of this early technique. Only 30% of patients had all three major arteries available for analysis, and in detecting stenosis the sensitivity was just 58% with 76% specificity.

Now, as 16-slice and even 64-slice CT angiography become available, the sensitivity and specificity have improved considerably. Studies have calculated sensitivities as high as 97% and specificities as high as 94%.

Most important, the negative predictive value is 98%–100%. “The benefit of CT angio is that when the coronaries look normal, the coronaries are normal,” Dr. Budoff said.

The temporal resolution of the CT images is about 175 milliseconds, so reducing the heart rate to below 60 beats per minute is important for accuracy and interpretability. Most centers use 100 mg metoprolol 1 hour prior to the study and/or a 5-mg intravenous metoprolol push every 5 minutes until the patient achieves a slow heart rate.

A regular rhythm is also important. With multiple detectors obtaining images at specific parts of the heart cycle, the modality reaches an effective frame speed of 15 images per second. This is slower than the cath lab, but fast enough that the images are free of motion artifact.

CT angiography may be the best technique for imaging the results of bypass grafting as the anastomoses are clearly visible. Other clinical indications for CT angiography are in cases of equivocal results following stress testing; to evaluate patency post angioplasty, post stent, and post bypass surgery; in cases of congenital abnormalities and anomalous coronaries; before and after atrial fibrillation ablation; and before placing a biventricular pacer.

CT angiography is not without its disadvantages, however. It's not very good for visualizing vessels with diameters less than 1.5 mm. It is subject to artifacts from extensive calcification, stents, or extensive clips after bypass grafting. And it subjects patients to a relatively high dose of radiation—about 9.3–11.3 mSv, compared with 2.1–2.3 mSv for the cath lab and 0.1 mSv for a chest x-ray.

CT angiography reveales high-grade stenosis (dark area) in the mid-left anterior descending coronary artery. Courtesy Dr. Matthew J. Budoff

SAN FRANCISCO — With computed tomography angiography, “patients literally go home with a Band-Aid and a bottle of water” after just 20 minutes, Matthew J. Budoff, M.D., remarked at a cardiovascular imaging conference sponsored by the American College of Cardiology.

With high sensitivity and specificity and images that rival the resolution obtainable with traditional coronary angiography from the catheterization lab, CT angiography will allow many more patients to avoid an invasive procedure, said Dr. Budoff of Harbor-UCLA Medical Center, Torrance, Calif.

After an injection of 80–100 mL of nonionic iodinated contrast solution, up to 4,000 two-dimensional images can be obtained within 20–30 seconds as the patient holds his or her breath. The entire procedure takes 20 minutes, and interpretation takes another 10 minutes.

Sophisticated workstations assemble the stack of 2D images into a three-dimensional reconstruction. Interpretations are made on the basis of the 3D reconstruction with reference to the 2D images.

Dr. Budoff started working with CT angiography in the mid-1990s. In those days it took 3 weeks of full-time computation to assemble a single 3D reconstruction. This same function takes just 30 seconds today.

And these workstations allow the cardiologist to rotate the heart image in three dimensions, to zoom in to interesting features, and to easily reference the original 2D data from any point of interest.

The initial studies of four-slice CT angiography revealed the limitations of this early technique. Only 30% of patients had all three major arteries available for analysis, and in detecting stenosis the sensitivity was just 58% with 76% specificity.

Now, as 16-slice and even 64-slice CT angiography become available, the sensitivity and specificity have improved considerably. Studies have calculated sensitivities as high as 97% and specificities as high as 94%.

Most important, the negative predictive value is 98%–100%. “The benefit of CT angio is that when the coronaries look normal, the coronaries are normal,” Dr. Budoff said.

The temporal resolution of the CT images is about 175 milliseconds, so reducing the heart rate to below 60 beats per minute is important for accuracy and interpretability. Most centers use 100 mg metoprolol 1 hour prior to the study and/or a 5-mg intravenous metoprolol push every 5 minutes until the patient achieves a slow heart rate.

A regular rhythm is also important. With multiple detectors obtaining images at specific parts of the heart cycle, the modality reaches an effective frame speed of 15 images per second. This is slower than the cath lab, but fast enough that the images are free of motion artifact.

CT angiography may be the best technique for imaging the results of bypass grafting as the anastomoses are clearly visible. Other clinical indications for CT angiography are in cases of equivocal results following stress testing; to evaluate patency post angioplasty, post stent, and post bypass surgery; in cases of congenital abnormalities and anomalous coronaries; before and after atrial fibrillation ablation; and before placing a biventricular pacer.

CT angiography is not without its disadvantages, however. It's not very good for visualizing vessels with diameters less than 1.5 mm. It is subject to artifacts from extensive calcification, stents, or extensive clips after bypass grafting. And it subjects patients to a relatively high dose of radiation—about 9.3–11.3 mSv, compared with 2.1–2.3 mSv for the cath lab and 0.1 mSv for a chest x-ray.

CT angiography reveales high-grade stenosis (dark area) in the mid-left anterior descending coronary artery. Courtesy Dr. Matthew J. Budoff