Imaging

HONOLULU – Ischemic stroke patients at high risk of coronary artery disease rarely receive guideline-supported cardiac stress testing, a study found.

Among 2,377 veterans with stroke, 28% were at high risk of coronary artery disease (CAD), and only 6.2% received CAD screening within 6 months of discharge.

Moreover, 1-year all-cause mortality was significantly lower among high-risk patients who received CAD screening than among their counterparts who did not (5% vs. 19%; P = .018), Dr. Jason Sico said at the International Stroke Conference.

Patrice Wendling/IMNG Medical Media

American Heart Association/American Stroke Association guidelines (Circulation 2003;108:1278-90) recommend that acute ischemic stroke patients at high risk of CAD, defined by a Framingham Risk Score of at least 20%, should receive cardiac screening for occult disease.

Studies have shown that 20%-40% of stroke patients have silent cardiac ischemia, and up to 6% of stroke patients die from cardiac causes or are readmitted with a myocardial infarction in the first 3 months following a stroke, said Dr. Sico, of the department of neurology at Yale New Haven (Conn.) Hospital.

Cardiac stress testing may be underutilized because most medical professionals caring for stroke survivors are not aware of the recommendation, he said in an interview. When they are made aware, one study found that providers may not be convinced that screening for cardiac disease within the stroke population will help their patients or is cost-effective (Stroke 2009;40:3407-9).

"In their favor, there has not been a large prospective study that has demonstrated that cardiac screening for stroke patients improves such important outcomes as mortality and hospital readmission," he added.

The investigators reviewed medical records for a sample of 3,965 patients from 131 Veterans Health Administration facilities admitted for a confirmed diagnosis of ischemic stroke in 2007. Framingham Risk Scores were calculated for 2,377 patients after exclusion of 1,588 patients with a prior cardiac stress test or a known history of CAD, or if they died during hospitalization or had unaccountable data.

In all, 676 (28%) patients had a high Framingham Risk Score of 20% or more, and 1,701 (72%) had a low/intermediate Framingham Risk Score.

Cardiac stress testing within 6 months of discharge from the index stroke was not performed more frequently among high-risk than among low/intermediate-risk patients (6.2% vs. 7.5%; odds ratio, 0.81), Dr. Sico said.

Patients who underwent screening had significantly lower baseline National Institutes of Health Stroke Severity scores than those who did not (mean, 3.3 vs. 4.1; P = .003) and were younger by about 2 years (64.5 years vs. 66.4 years; P = .01). Rates of hypertension, hyperlipidemia, diabetes, and white race were similar between groups, he said at the meeting, which was sponsored by the American Heart Association.

Among all patients, 1-year mortality was significantly lower at 5% in cases where screening was performed, compared with 14% when it was not (P = .001).

Dr. Sico said the strength of the study was the relatively large cohort but that the study was limited by its makeup of primarily male veterans, data from fiscal year 2007, and the inability to explain the reasons for the underutilization of guideline-concordant cardiac screening.

Future work includes understanding the barriers to cardiac testing, the reasons behind the mortality differences among patients who did and did not receive CAD screening, and how implementation of CAD screening guidelines affects outcomes.

"To borrow a page from the diabetes literature, it was dogma that if you were diabetic because it is a coronary equivalent, you should get cardiac stress testing, but when it was prospectively looked at [in the DIAD trial] it really didn’t affect outcome; so we don’t have any prospective [study] in the stroke population to answer this question," he said.

Session moderator Dr. Jennifer Juhl Majersik, of the University of Utah, Salt Lake City, said, "This is a great example of stroke neurologists’ need to look beyond the brain."

The Veterans Health Administration provided funding for the study. Dr. Sico and his coauthors reported no disclosures.

[email protected]

HONOLULU – Ischemic stroke patients at high risk of coronary artery disease rarely receive guideline-supported cardiac stress testing, a study found.

Among 2,377 veterans with stroke, 28% were at high risk of coronary artery disease (CAD), and only 6.2% received CAD screening within 6 months of discharge.

Moreover, 1-year all-cause mortality was significantly lower among high-risk patients who received CAD screening than among their counterparts who did not (5% vs. 19%; P = .018), Dr. Jason Sico said at the International Stroke Conference.

Patrice Wendling/IMNG Medical Media

American Heart Association/American Stroke Association guidelines (Circulation 2003;108:1278-90) recommend that acute ischemic stroke patients at high risk of CAD, defined by a Framingham Risk Score of at least 20%, should receive cardiac screening for occult disease.

Studies have shown that 20%-40% of stroke patients have silent cardiac ischemia, and up to 6% of stroke patients die from cardiac causes or are readmitted with a myocardial infarction in the first 3 months following a stroke, said Dr. Sico, of the department of neurology at Yale New Haven (Conn.) Hospital.

Cardiac stress testing may be underutilized because most medical professionals caring for stroke survivors are not aware of the recommendation, he said in an interview. When they are made aware, one study found that providers may not be convinced that screening for cardiac disease within the stroke population will help their patients or is cost-effective (Stroke 2009;40:3407-9).

"In their favor, there has not been a large prospective study that has demonstrated that cardiac screening for stroke patients improves such important outcomes as mortality and hospital readmission," he added.

The investigators reviewed medical records for a sample of 3,965 patients from 131 Veterans Health Administration facilities admitted for a confirmed diagnosis of ischemic stroke in 2007. Framingham Risk Scores were calculated for 2,377 patients after exclusion of 1,588 patients with a prior cardiac stress test or a known history of CAD, or if they died during hospitalization or had unaccountable data.

In all, 676 (28%) patients had a high Framingham Risk Score of 20% or more, and 1,701 (72%) had a low/intermediate Framingham Risk Score.

Cardiac stress testing within 6 months of discharge from the index stroke was not performed more frequently among high-risk than among low/intermediate-risk patients (6.2% vs. 7.5%; odds ratio, 0.81), Dr. Sico said.

Patients who underwent screening had significantly lower baseline National Institutes of Health Stroke Severity scores than those who did not (mean, 3.3 vs. 4.1; P = .003) and were younger by about 2 years (64.5 years vs. 66.4 years; P = .01). Rates of hypertension, hyperlipidemia, diabetes, and white race were similar between groups, he said at the meeting, which was sponsored by the American Heart Association.

Among all patients, 1-year mortality was significantly lower at 5% in cases where screening was performed, compared with 14% when it was not (P = .001).

Dr. Sico said the strength of the study was the relatively large cohort but that the study was limited by its makeup of primarily male veterans, data from fiscal year 2007, and the inability to explain the reasons for the underutilization of guideline-concordant cardiac screening.

Future work includes understanding the barriers to cardiac testing, the reasons behind the mortality differences among patients who did and did not receive CAD screening, and how implementation of CAD screening guidelines affects outcomes.

"To borrow a page from the diabetes literature, it was dogma that if you were diabetic because it is a coronary equivalent, you should get cardiac stress testing, but when it was prospectively looked at [in the DIAD trial] it really didn’t affect outcome; so we don’t have any prospective [study] in the stroke population to answer this question," he said.

Session moderator Dr. Jennifer Juhl Majersik, of the University of Utah, Salt Lake City, said, "This is a great example of stroke neurologists’ need to look beyond the brain."

The Veterans Health Administration provided funding for the study. Dr. Sico and his coauthors reported no disclosures.

[email protected]

HONOLULU – Ischemic stroke patients at high risk of coronary artery disease rarely receive guideline-supported cardiac stress testing, a study found.

Among 2,377 veterans with stroke, 28% were at high risk of coronary artery disease (CAD), and only 6.2% received CAD screening within 6 months of discharge.

Moreover, 1-year all-cause mortality was significantly lower among high-risk patients who received CAD screening than among their counterparts who did not (5% vs. 19%; P = .018), Dr. Jason Sico said at the International Stroke Conference.

Patrice Wendling/IMNG Medical Media

American Heart Association/American Stroke Association guidelines (Circulation 2003;108:1278-90) recommend that acute ischemic stroke patients at high risk of CAD, defined by a Framingham Risk Score of at least 20%, should receive cardiac screening for occult disease.

Studies have shown that 20%-40% of stroke patients have silent cardiac ischemia, and up to 6% of stroke patients die from cardiac causes or are readmitted with a myocardial infarction in the first 3 months following a stroke, said Dr. Sico, of the department of neurology at Yale New Haven (Conn.) Hospital.

Cardiac stress testing may be underutilized because most medical professionals caring for stroke survivors are not aware of the recommendation, he said in an interview. When they are made aware, one study found that providers may not be convinced that screening for cardiac disease within the stroke population will help their patients or is cost-effective (Stroke 2009;40:3407-9).

"In their favor, there has not been a large prospective study that has demonstrated that cardiac screening for stroke patients improves such important outcomes as mortality and hospital readmission," he added.

The investigators reviewed medical records for a sample of 3,965 patients from 131 Veterans Health Administration facilities admitted for a confirmed diagnosis of ischemic stroke in 2007. Framingham Risk Scores were calculated for 2,377 patients after exclusion of 1,588 patients with a prior cardiac stress test or a known history of CAD, or if they died during hospitalization or had unaccountable data.

In all, 676 (28%) patients had a high Framingham Risk Score of 20% or more, and 1,701 (72%) had a low/intermediate Framingham Risk Score.

Cardiac stress testing within 6 months of discharge from the index stroke was not performed more frequently among high-risk than among low/intermediate-risk patients (6.2% vs. 7.5%; odds ratio, 0.81), Dr. Sico said.

Patients who underwent screening had significantly lower baseline National Institutes of Health Stroke Severity scores than those who did not (mean, 3.3 vs. 4.1; P = .003) and were younger by about 2 years (64.5 years vs. 66.4 years; P = .01). Rates of hypertension, hyperlipidemia, diabetes, and white race were similar between groups, he said at the meeting, which was sponsored by the American Heart Association.

Among all patients, 1-year mortality was significantly lower at 5% in cases where screening was performed, compared with 14% when it was not (P = .001).

Dr. Sico said the strength of the study was the relatively large cohort but that the study was limited by its makeup of primarily male veterans, data from fiscal year 2007, and the inability to explain the reasons for the underutilization of guideline-concordant cardiac screening.

Future work includes understanding the barriers to cardiac testing, the reasons behind the mortality differences among patients who did and did not receive CAD screening, and how implementation of CAD screening guidelines affects outcomes.

"To borrow a page from the diabetes literature, it was dogma that if you were diabetic because it is a coronary equivalent, you should get cardiac stress testing, but when it was prospectively looked at [in the DIAD trial] it really didn’t affect outcome; so we don’t have any prospective [study] in the stroke population to answer this question," he said.

Session moderator Dr. Jennifer Juhl Majersik, of the University of Utah, Salt Lake City, said, "This is a great example of stroke neurologists’ need to look beyond the brain."

The Veterans Health Administration provided funding for the study. Dr. Sico and his coauthors reported no disclosures.

[email protected]

SNOWMASS, COLO. – Fractional flow reserve measurement has rapidly emerged as the tool of choice for deciding in the catheterization laboratory whether to revascularize an intermediate 50%-70% stenosis in a patient with stable ischemic heart disease, according to Dr. E. Murat *Tuzcu.

The evidence is now compelling that fractional flow reserve (FFR) calculated invasively from coronary pressure measurement is superior to angiographic assessment, intravascular ultrasound, or optical coherence tomography for this purpose.

"When the question is, ‘What should we do with the borderline lesion?’ the answer is fractional flow reserve, even in the left main coronary artery," Dr. Tuzcu, professor of medicine at the Cleveland Clinic, asserted at the Annual Cardiovascular Conference at Snowmass.

FFR is the method par excellence for determining if an intermediate stenosis is hemodynamically significant; that is, whether the lesion is responsible for reversible ischemia. If it is, then coronary stenting will improve the patient’s functional status and reduce the likelihood of acute MI and all-cause mortality down the road. If FFR indicates that the stenosis is not responsible for reversible ischemia, however, then PCI won’t improve the patient’s prognosis. FFR has the additional virtues of being fast and simple, and it enables immediate decision-making in the cath lab, he explained.

For Dr. Tuzcu, the game changer for FFR was the DEFER study, which he considers to be one of the most important clinical trials in the field of interventional cardiology in the past decade. It showed that, by using FFR, cardiologists could be more selective in their use of PCI in the setting of stable ischemic heart disease.

DEFER was a multicenter Dutch/Belgian study in which 325 patients underwent FFR measurement just prior to planned PCI for an intermediate stenosis. If the FFR value was less than 0.75 – indicative of reversible ischemia – then PCI was performed as planned. If it was 0.75 or greater, patients were randomized to PCI or to deferred PCI.

At 5 years of follow-up, the rate of cardiac death or acute MI was 3.3% in the 91 patients with an FFR of 0.75 or more in the deferred PCI group – less than 1% per year. That wasn’t significantly different from the 7.9% rate among the 90 patients with an FFR of at least 0.75 who underwent prompt PCI. In contrast, the combined endpoint occurred in 15.7% of the 144 PCI-treated patients with FFR evidence of reversible ischemia due to the target lesion (J. Am. Coll. Cardiol. 2007;49:2105-11.)

DEFER was conducted in the bare metal–stent era. The next major clinical trial advancing FFR was the FAME (Fractional Flow Reserve Versus Angiography for Multivessel Evaluation) study, in which 1,005 patients with multivessel CAD were randomized to PCI with drug-eluting stents guided by FFR or by angiography alone. The FFR definition of reversible ischemia used in FAME and subsequent trials was a value of 0.80 or less.

The 2-year rate of death or MI was 8.4% in the FFR-guided group, significantly less than the 12.9% rate in the angiography-guided patients. For patients with stenotic lesions deferred from PCI on the basis of an FFR greater than 0.80, the 2-year rate of MI was just 0.2%, with a revascularization rate of 3.2% (J. Am. Coll. Cardiol. 2010;56:177-84).

Nearly half of all stenoses were categorized angiographically as intermediate, 50%-70% lesions. FFR classified 35% of such lesions as functionally significant, while 65% were not associated with reversible ischemia (J. Am. Coll. Cardiol. 2010;55:2816-21).

In the FAME-2 trial, 888 patients with stable CAD and at least one functionally significant stenosis with an FFR of 0.80 or less were randomized to PCI plus optimal medical therapy or to optimal medical therapy alone. Recruitment was halted prematurely by the data safety monitoring board because the combined endpoint of death, MI, or urgent revascularization had occurred in 4.3% of the PCI group versus 12.7% of those assigned to optimal medical management alone (N. Engl. J. Med. 2012;367:991-1001).

Dr. Tuzcu noted that optical coherence tomography (OCT) has drawn much interest as a tool for identifying hemodynamically severe coronary stenoses. "It provides great pictures with tremendous resolution. For looking at stent strut coverage, OCT is a star tool. It’s almost like histology," he said.

It can’t, however, hold a candle to FFR for hemodynamic assessment of stenoses, he added, pointing to a recent Spanish study comparing FFR, OCT, and intravascular ultrasound (IVUS) for this purpose. OCT and IVUS displayed moderate diagnostic efficiency, with no clinically meaningful difference between them, but Dr. Tuzcu concurred with the investigators that the low specificity of OCT and IVUS precludes their use in lieu of FFR for functional assessment (J. Am. Coll. Cardiol. 2012;59:1080-9).

The most recent significant development on the FFR front was Dr. Gregg W. Stone’s presentation of the pooled results of the VERDICT and FIRST trials at the Transcatheter Cardiovascular Therapeutics conference in Miami last October. VERDICT and FIRST included 516 patients with 544 intermediate coronary stenoses evaluated by both FFR and IVUS at 24 centers in nine countries. The bottom line was that IVUS-determined minimum luminal cross-sectional area was only modestly correlated with FFR, according to Dr. Stone, professor of medicine and director of cardiovascular research and education at Columbia University Medical Center/New York Presbyterian Hospital.

"I think this clearly showed – and probably conclusively showed – that, while IVUS is useful in many, many settings, it’s probably not the best tool when FFR is available for deciding which lesion is significant and which is not," Dr. Tuzcu said.

"IVUS is a pretty good tool, sometimes, for morphologic assessment. I like it when there’s an issue with the left main coronary artery. I can size the artery, I can understand an ostial lesion, and I can certainly understand better a bifurcation or trifurcation of the left main coronary artery," he added.

The current American College of Cardiology/American Heart Association guidelines give IVUS a class IIa rating as "reasonable" to assess angiographically intermediate stenoses of the left main coronary artery. FFR gets the same relatively tepid IIa rating for assessment of intermediate stenoses in any coronary arteries. In contrast, the latest European Society of Cardiology guidelines on coronary revascularization have bumped up FFR to a class Ia rating, making it the standard for this assessment.

Dr. Tuzcu reported having no financial conflicts.

[email protected]

*CORRECTION, 3/1/2013: In an earlier version of this story, the name of Dr. E. Murat Tuzcu was spelled incorrectly. This version has been updated.

SNOWMASS, COLO. – Fractional flow reserve measurement has rapidly emerged as the tool of choice for deciding in the catheterization laboratory whether to revascularize an intermediate 50%-70% stenosis in a patient with stable ischemic heart disease, according to Dr. E. Murat *Tuzcu.

The evidence is now compelling that fractional flow reserve (FFR) calculated invasively from coronary pressure measurement is superior to angiographic assessment, intravascular ultrasound, or optical coherence tomography for this purpose.

"When the question is, ‘What should we do with the borderline lesion?’ the answer is fractional flow reserve, even in the left main coronary artery," Dr. Tuzcu, professor of medicine at the Cleveland Clinic, asserted at the Annual Cardiovascular Conference at Snowmass.

FFR is the method par excellence for determining if an intermediate stenosis is hemodynamically significant; that is, whether the lesion is responsible for reversible ischemia. If it is, then coronary stenting will improve the patient’s functional status and reduce the likelihood of acute MI and all-cause mortality down the road. If FFR indicates that the stenosis is not responsible for reversible ischemia, however, then PCI won’t improve the patient’s prognosis. FFR has the additional virtues of being fast and simple, and it enables immediate decision-making in the cath lab, he explained.

For Dr. Tuzcu, the game changer for FFR was the DEFER study, which he considers to be one of the most important clinical trials in the field of interventional cardiology in the past decade. It showed that, by using FFR, cardiologists could be more selective in their use of PCI in the setting of stable ischemic heart disease.

DEFER was a multicenter Dutch/Belgian study in which 325 patients underwent FFR measurement just prior to planned PCI for an intermediate stenosis. If the FFR value was less than 0.75 – indicative of reversible ischemia – then PCI was performed as planned. If it was 0.75 or greater, patients were randomized to PCI or to deferred PCI.

At 5 years of follow-up, the rate of cardiac death or acute MI was 3.3% in the 91 patients with an FFR of 0.75 or more in the deferred PCI group – less than 1% per year. That wasn’t significantly different from the 7.9% rate among the 90 patients with an FFR of at least 0.75 who underwent prompt PCI. In contrast, the combined endpoint occurred in 15.7% of the 144 PCI-treated patients with FFR evidence of reversible ischemia due to the target lesion (J. Am. Coll. Cardiol. 2007;49:2105-11.)

DEFER was conducted in the bare metal–stent era. The next major clinical trial advancing FFR was the FAME (Fractional Flow Reserve Versus Angiography for Multivessel Evaluation) study, in which 1,005 patients with multivessel CAD were randomized to PCI with drug-eluting stents guided by FFR or by angiography alone. The FFR definition of reversible ischemia used in FAME and subsequent trials was a value of 0.80 or less.

The 2-year rate of death or MI was 8.4% in the FFR-guided group, significantly less than the 12.9% rate in the angiography-guided patients. For patients with stenotic lesions deferred from PCI on the basis of an FFR greater than 0.80, the 2-year rate of MI was just 0.2%, with a revascularization rate of 3.2% (J. Am. Coll. Cardiol. 2010;56:177-84).

Nearly half of all stenoses were categorized angiographically as intermediate, 50%-70% lesions. FFR classified 35% of such lesions as functionally significant, while 65% were not associated with reversible ischemia (J. Am. Coll. Cardiol. 2010;55:2816-21).

In the FAME-2 trial, 888 patients with stable CAD and at least one functionally significant stenosis with an FFR of 0.80 or less were randomized to PCI plus optimal medical therapy or to optimal medical therapy alone. Recruitment was halted prematurely by the data safety monitoring board because the combined endpoint of death, MI, or urgent revascularization had occurred in 4.3% of the PCI group versus 12.7% of those assigned to optimal medical management alone (N. Engl. J. Med. 2012;367:991-1001).

Dr. Tuzcu noted that optical coherence tomography (OCT) has drawn much interest as a tool for identifying hemodynamically severe coronary stenoses. "It provides great pictures with tremendous resolution. For looking at stent strut coverage, OCT is a star tool. It’s almost like histology," he said.

It can’t, however, hold a candle to FFR for hemodynamic assessment of stenoses, he added, pointing to a recent Spanish study comparing FFR, OCT, and intravascular ultrasound (IVUS) for this purpose. OCT and IVUS displayed moderate diagnostic efficiency, with no clinically meaningful difference between them, but Dr. Tuzcu concurred with the investigators that the low specificity of OCT and IVUS precludes their use in lieu of FFR for functional assessment (J. Am. Coll. Cardiol. 2012;59:1080-9).

The most recent significant development on the FFR front was Dr. Gregg W. Stone’s presentation of the pooled results of the VERDICT and FIRST trials at the Transcatheter Cardiovascular Therapeutics conference in Miami last October. VERDICT and FIRST included 516 patients with 544 intermediate coronary stenoses evaluated by both FFR and IVUS at 24 centers in nine countries. The bottom line was that IVUS-determined minimum luminal cross-sectional area was only modestly correlated with FFR, according to Dr. Stone, professor of medicine and director of cardiovascular research and education at Columbia University Medical Center/New York Presbyterian Hospital.

"I think this clearly showed – and probably conclusively showed – that, while IVUS is useful in many, many settings, it’s probably not the best tool when FFR is available for deciding which lesion is significant and which is not," Dr. Tuzcu said.

"IVUS is a pretty good tool, sometimes, for morphologic assessment. I like it when there’s an issue with the left main coronary artery. I can size the artery, I can understand an ostial lesion, and I can certainly understand better a bifurcation or trifurcation of the left main coronary artery," he added.

The current American College of Cardiology/American Heart Association guidelines give IVUS a class IIa rating as "reasonable" to assess angiographically intermediate stenoses of the left main coronary artery. FFR gets the same relatively tepid IIa rating for assessment of intermediate stenoses in any coronary arteries. In contrast, the latest European Society of Cardiology guidelines on coronary revascularization have bumped up FFR to a class Ia rating, making it the standard for this assessment.

Dr. Tuzcu reported having no financial conflicts.

[email protected]

*CORRECTION, 3/1/2013: In an earlier version of this story, the name of Dr. E. Murat Tuzcu was spelled incorrectly. This version has been updated.

SNOWMASS, COLO. – Fractional flow reserve measurement has rapidly emerged as the tool of choice for deciding in the catheterization laboratory whether to revascularize an intermediate 50%-70% stenosis in a patient with stable ischemic heart disease, according to Dr. E. Murat *Tuzcu.

The evidence is now compelling that fractional flow reserve (FFR) calculated invasively from coronary pressure measurement is superior to angiographic assessment, intravascular ultrasound, or optical coherence tomography for this purpose.

"When the question is, ‘What should we do with the borderline lesion?’ the answer is fractional flow reserve, even in the left main coronary artery," Dr. Tuzcu, professor of medicine at the Cleveland Clinic, asserted at the Annual Cardiovascular Conference at Snowmass.

FFR is the method par excellence for determining if an intermediate stenosis is hemodynamically significant; that is, whether the lesion is responsible for reversible ischemia. If it is, then coronary stenting will improve the patient’s functional status and reduce the likelihood of acute MI and all-cause mortality down the road. If FFR indicates that the stenosis is not responsible for reversible ischemia, however, then PCI won’t improve the patient’s prognosis. FFR has the additional virtues of being fast and simple, and it enables immediate decision-making in the cath lab, he explained.

For Dr. Tuzcu, the game changer for FFR was the DEFER study, which he considers to be one of the most important clinical trials in the field of interventional cardiology in the past decade. It showed that, by using FFR, cardiologists could be more selective in their use of PCI in the setting of stable ischemic heart disease.

DEFER was a multicenter Dutch/Belgian study in which 325 patients underwent FFR measurement just prior to planned PCI for an intermediate stenosis. If the FFR value was less than 0.75 – indicative of reversible ischemia – then PCI was performed as planned. If it was 0.75 or greater, patients were randomized to PCI or to deferred PCI.

At 5 years of follow-up, the rate of cardiac death or acute MI was 3.3% in the 91 patients with an FFR of 0.75 or more in the deferred PCI group – less than 1% per year. That wasn’t significantly different from the 7.9% rate among the 90 patients with an FFR of at least 0.75 who underwent prompt PCI. In contrast, the combined endpoint occurred in 15.7% of the 144 PCI-treated patients with FFR evidence of reversible ischemia due to the target lesion (J. Am. Coll. Cardiol. 2007;49:2105-11.)

DEFER was conducted in the bare metal–stent era. The next major clinical trial advancing FFR was the FAME (Fractional Flow Reserve Versus Angiography for Multivessel Evaluation) study, in which 1,005 patients with multivessel CAD were randomized to PCI with drug-eluting stents guided by FFR or by angiography alone. The FFR definition of reversible ischemia used in FAME and subsequent trials was a value of 0.80 or less.

The 2-year rate of death or MI was 8.4% in the FFR-guided group, significantly less than the 12.9% rate in the angiography-guided patients. For patients with stenotic lesions deferred from PCI on the basis of an FFR greater than 0.80, the 2-year rate of MI was just 0.2%, with a revascularization rate of 3.2% (J. Am. Coll. Cardiol. 2010;56:177-84).

Nearly half of all stenoses were categorized angiographically as intermediate, 50%-70% lesions. FFR classified 35% of such lesions as functionally significant, while 65% were not associated with reversible ischemia (J. Am. Coll. Cardiol. 2010;55:2816-21).

In the FAME-2 trial, 888 patients with stable CAD and at least one functionally significant stenosis with an FFR of 0.80 or less were randomized to PCI plus optimal medical therapy or to optimal medical therapy alone. Recruitment was halted prematurely by the data safety monitoring board because the combined endpoint of death, MI, or urgent revascularization had occurred in 4.3% of the PCI group versus 12.7% of those assigned to optimal medical management alone (N. Engl. J. Med. 2012;367:991-1001).

Dr. Tuzcu noted that optical coherence tomography (OCT) has drawn much interest as a tool for identifying hemodynamically severe coronary stenoses. "It provides great pictures with tremendous resolution. For looking at stent strut coverage, OCT is a star tool. It’s almost like histology," he said.

It can’t, however, hold a candle to FFR for hemodynamic assessment of stenoses, he added, pointing to a recent Spanish study comparing FFR, OCT, and intravascular ultrasound (IVUS) for this purpose. OCT and IVUS displayed moderate diagnostic efficiency, with no clinically meaningful difference between them, but Dr. Tuzcu concurred with the investigators that the low specificity of OCT and IVUS precludes their use in lieu of FFR for functional assessment (J. Am. Coll. Cardiol. 2012;59:1080-9).

The most recent significant development on the FFR front was Dr. Gregg W. Stone’s presentation of the pooled results of the VERDICT and FIRST trials at the Transcatheter Cardiovascular Therapeutics conference in Miami last October. VERDICT and FIRST included 516 patients with 544 intermediate coronary stenoses evaluated by both FFR and IVUS at 24 centers in nine countries. The bottom line was that IVUS-determined minimum luminal cross-sectional area was only modestly correlated with FFR, according to Dr. Stone, professor of medicine and director of cardiovascular research and education at Columbia University Medical Center/New York Presbyterian Hospital.

"I think this clearly showed – and probably conclusively showed – that, while IVUS is useful in many, many settings, it’s probably not the best tool when FFR is available for deciding which lesion is significant and which is not," Dr. Tuzcu said.

"IVUS is a pretty good tool, sometimes, for morphologic assessment. I like it when there’s an issue with the left main coronary artery. I can size the artery, I can understand an ostial lesion, and I can certainly understand better a bifurcation or trifurcation of the left main coronary artery," he added.

The current American College of Cardiology/American Heart Association guidelines give IVUS a class IIa rating as "reasonable" to assess angiographically intermediate stenoses of the left main coronary artery. FFR gets the same relatively tepid IIa rating for assessment of intermediate stenoses in any coronary arteries. In contrast, the latest European Society of Cardiology guidelines on coronary revascularization have bumped up FFR to a class Ia rating, making it the standard for this assessment.

Dr. Tuzcu reported having no financial conflicts.

[email protected]

*CORRECTION, 3/1/2013: In an earlier version of this story, the name of Dr. E. Murat Tuzcu was spelled incorrectly. This version has been updated.

Coronary computed tomographic angiography for rule-out of acute coronary syndrome in the emergency department may have moved ahead of SPECT myocardial perfusion imaging – its main noninvasive imaging rival – on the strength of recent evidence of advantages in time-to-discharge and radiation exposure, according to Dr. Christopher M. Kramer.

Myocardial perfusion imaging (MPI) has for roughly a decade been seen as the standard of care for imaging assistance in ED triage of patients presenting with chest pain and a nondiagnostic ECG.

It attained this status based upon the favorable results of the very large randomized ERASE (ER Assessment of Sestamibi for Evaluation of Chest Pain) trial (JAMA 2002;288:2693-700) and other studies in which MPI was compared to usual ED care.

©Elsevier Inc.

But in more recent randomized trials comparing computed tomographic angiography (CTA) to standard ED evaluation protocols, which now often include MPI, CTA has carried the day, Dr. Kramer said at the annual cardiovascular conference at Snowmass sponsored by the American College of Cardiology.

For example, the multicenter ACRIN-PA trial randomized 1,370 low- to intermediate-risk patients presenting with possible ACS to EDs to either early CTA or standard care, which in most cases included MPI. The CTA group had a higher rate of discharge from the ED: 50%, compared with 23%. They also had a median 18.0-hour length of stay, significantly shorter than the 24.8 hours in the control group (N. Engl. J. Med. 2012;366:1393-403).

Moreover, the CTA group had a higher rate of detection of CAD – 9.0% compared to 3.5% because CTA can detect nonobstructive as well as obstructive plaque, added Dr. Kramer, professor of cardiology and of radiology and director of the cardiovascular imaging center at the University of Virginia, Charlottesville.

The ROMICAT II (Rule Out Myocardial Ischemia/Infarction by Computer-Assisted Tomography) trial was a similar story. ROMICAT II was a multicenter trial in which 1,000 patients who presented to an ED with symptoms suggestive of ACS but a nondiagnostic ECG were randomized to CTA or a standard ED evaluation, most often including MPI. A total of 47% of patients in the CTA group were discharged directly from the ED, compared with just 12% in the control arm. Most notably, the median length of stay in the ED was 8.6 hours with CTA vs. 26.7 hours with a standard evaluation, and the mean length of stay in the hospital was reduced from 30.8 hours with standard evaluation to 23.2 hours, a 7.6-hour reduction (N. Engl. J. Med. 2012;367:299-308).

Follow-up demonstrated no cases of undetected ACS occurred in either group.

The average radiation dose to patients in the CTA arm was 14.3 mSv, compared with about 10 mSv for those patients in the standard evaluation group who got MPI. However, ROMICAT II didn’t use the latest CTA equipment.

"With the modern CTA units, the radiation doses have come down quite a lot. With our flash 256 detector in the ED, we can do a study in a patient with a controlled heart rate and a reasonable BMI [body mass index] with about 1 mSv of radiation now," Dr. Kramer recalled.

Costs in the ED were significantly lower in the CTA group than in controls in ROMICAT II because of the faster patient throughput. However, this savings was neutralized by higher in-hospital costs because the CTA group had more cardiac catheterizations and revascularization procedures because more cases of CAD were detected, as in ACRIN-PA.

"The real advantage to CTA is the rapid discharge from the ED. That’s what patients appreciate. They get their CT angiogram and they can go home within an hour or 2. It really improves their experience in the ED, rather than spending all night waiting for their SPECT MPI results. I don’t think you’re saving costs by doing CTA, though, because of this catch-up phenomenon," Dr. Kramer said.

Aside from the fact that MPI takes 3-5 hours, its other main limitations are that it has difficulty in differentiating acute ischemia from acute infarction or an old infarct, according to Dr. Kramer.

The ability of CTA to detect more cases of CAD by identifying nonobstructive plaque may prove to be of clinical import. That’s the underlying hypothesis of the ongoing National Heart, Lung, and Blood Institute–funded PROMISE (Prospective Multicenter Imaging Study for Evaluation of Chest Pain) study, in which 10,000 patients with symptoms suggestive of CAD are being randomized to an initial CTA or to usual care with a functional test, either MPI, stress echocardiography, or exercise ECG.

This study is being conducted in the offices of primary care physicians and cardiologists rather than in EDs. PROMISE has clinical endpoints as its primary outcomes. The study hypothesis is that intervening in patients identified as having nonobstructive CAD will yield improved outcomes. Results remain several years off, Dr. Kramer said.

ED physicians are eager for high-tech help in quickly and reliably ruling out ACS. Acute chest pain accounts for more than 8 million ED visits annually, but in only 1.19 million admissions for ACS.

Besides MPI and CTA, the other two noninvasive imaging technologies available for use in the ED are cardiac magnetic resonance (CMR) and contrast echocardiography. Neither utilizes radiation – a big plus. Yet neither is as widely used as MPI or CTA. That’s because magnetic resonance takes longer than CTA does, and the scanner may not be available at a moment’s notice, as it really needs to be, when a patient presents with chest pain to the ED. Also, expertise in CMR is not widely available. This is also an issue for contrast echo in the ED.

"The problem with contrast echo is that there are really very few centers around the world that can do it well. It really hasn’t caught on in terms of utilization in the ED," according to the cardiologist.

He reported that he serves as a consultant to Synarc and receives research support from Siemens Medical Solutions.

[email protected]

Coronary computed tomographic angiography for rule-out of acute coronary syndrome in the emergency department may have moved ahead of SPECT myocardial perfusion imaging – its main noninvasive imaging rival – on the strength of recent evidence of advantages in time-to-discharge and radiation exposure, according to Dr. Christopher M. Kramer.

Myocardial perfusion imaging (MPI) has for roughly a decade been seen as the standard of care for imaging assistance in ED triage of patients presenting with chest pain and a nondiagnostic ECG.

It attained this status based upon the favorable results of the very large randomized ERASE (ER Assessment of Sestamibi for Evaluation of Chest Pain) trial (JAMA 2002;288:2693-700) and other studies in which MPI was compared to usual ED care.

©Elsevier Inc.

But in more recent randomized trials comparing computed tomographic angiography (CTA) to standard ED evaluation protocols, which now often include MPI, CTA has carried the day, Dr. Kramer said at the annual cardiovascular conference at Snowmass sponsored by the American College of Cardiology.

For example, the multicenter ACRIN-PA trial randomized 1,370 low- to intermediate-risk patients presenting with possible ACS to EDs to either early CTA or standard care, which in most cases included MPI. The CTA group had a higher rate of discharge from the ED: 50%, compared with 23%. They also had a median 18.0-hour length of stay, significantly shorter than the 24.8 hours in the control group (N. Engl. J. Med. 2012;366:1393-403).

Moreover, the CTA group had a higher rate of detection of CAD – 9.0% compared to 3.5% because CTA can detect nonobstructive as well as obstructive plaque, added Dr. Kramer, professor of cardiology and of radiology and director of the cardiovascular imaging center at the University of Virginia, Charlottesville.

The ROMICAT II (Rule Out Myocardial Ischemia/Infarction by Computer-Assisted Tomography) trial was a similar story. ROMICAT II was a multicenter trial in which 1,000 patients who presented to an ED with symptoms suggestive of ACS but a nondiagnostic ECG were randomized to CTA or a standard ED evaluation, most often including MPI. A total of 47% of patients in the CTA group were discharged directly from the ED, compared with just 12% in the control arm. Most notably, the median length of stay in the ED was 8.6 hours with CTA vs. 26.7 hours with a standard evaluation, and the mean length of stay in the hospital was reduced from 30.8 hours with standard evaluation to 23.2 hours, a 7.6-hour reduction (N. Engl. J. Med. 2012;367:299-308).

Follow-up demonstrated no cases of undetected ACS occurred in either group.

The average radiation dose to patients in the CTA arm was 14.3 mSv, compared with about 10 mSv for those patients in the standard evaluation group who got MPI. However, ROMICAT II didn’t use the latest CTA equipment.

"With the modern CTA units, the radiation doses have come down quite a lot. With our flash 256 detector in the ED, we can do a study in a patient with a controlled heart rate and a reasonable BMI [body mass index] with about 1 mSv of radiation now," Dr. Kramer recalled.

Costs in the ED were significantly lower in the CTA group than in controls in ROMICAT II because of the faster patient throughput. However, this savings was neutralized by higher in-hospital costs because the CTA group had more cardiac catheterizations and revascularization procedures because more cases of CAD were detected, as in ACRIN-PA.

"The real advantage to CTA is the rapid discharge from the ED. That’s what patients appreciate. They get their CT angiogram and they can go home within an hour or 2. It really improves their experience in the ED, rather than spending all night waiting for their SPECT MPI results. I don’t think you’re saving costs by doing CTA, though, because of this catch-up phenomenon," Dr. Kramer said.

Aside from the fact that MPI takes 3-5 hours, its other main limitations are that it has difficulty in differentiating acute ischemia from acute infarction or an old infarct, according to Dr. Kramer.

The ability of CTA to detect more cases of CAD by identifying nonobstructive plaque may prove to be of clinical import. That’s the underlying hypothesis of the ongoing National Heart, Lung, and Blood Institute–funded PROMISE (Prospective Multicenter Imaging Study for Evaluation of Chest Pain) study, in which 10,000 patients with symptoms suggestive of CAD are being randomized to an initial CTA or to usual care with a functional test, either MPI, stress echocardiography, or exercise ECG.

This study is being conducted in the offices of primary care physicians and cardiologists rather than in EDs. PROMISE has clinical endpoints as its primary outcomes. The study hypothesis is that intervening in patients identified as having nonobstructive CAD will yield improved outcomes. Results remain several years off, Dr. Kramer said.

ED physicians are eager for high-tech help in quickly and reliably ruling out ACS. Acute chest pain accounts for more than 8 million ED visits annually, but in only 1.19 million admissions for ACS.

Besides MPI and CTA, the other two noninvasive imaging technologies available for use in the ED are cardiac magnetic resonance (CMR) and contrast echocardiography. Neither utilizes radiation – a big plus. Yet neither is as widely used as MPI or CTA. That’s because magnetic resonance takes longer than CTA does, and the scanner may not be available at a moment’s notice, as it really needs to be, when a patient presents with chest pain to the ED. Also, expertise in CMR is not widely available. This is also an issue for contrast echo in the ED.

"The problem with contrast echo is that there are really very few centers around the world that can do it well. It really hasn’t caught on in terms of utilization in the ED," according to the cardiologist.

He reported that he serves as a consultant to Synarc and receives research support from Siemens Medical Solutions.

[email protected]

Coronary computed tomographic angiography for rule-out of acute coronary syndrome in the emergency department may have moved ahead of SPECT myocardial perfusion imaging – its main noninvasive imaging rival – on the strength of recent evidence of advantages in time-to-discharge and radiation exposure, according to Dr. Christopher M. Kramer.

Myocardial perfusion imaging (MPI) has for roughly a decade been seen as the standard of care for imaging assistance in ED triage of patients presenting with chest pain and a nondiagnostic ECG.

It attained this status based upon the favorable results of the very large randomized ERASE (ER Assessment of Sestamibi for Evaluation of Chest Pain) trial (JAMA 2002;288:2693-700) and other studies in which MPI was compared to usual ED care.

©Elsevier Inc.

But in more recent randomized trials comparing computed tomographic angiography (CTA) to standard ED evaluation protocols, which now often include MPI, CTA has carried the day, Dr. Kramer said at the annual cardiovascular conference at Snowmass sponsored by the American College of Cardiology.

For example, the multicenter ACRIN-PA trial randomized 1,370 low- to intermediate-risk patients presenting with possible ACS to EDs to either early CTA or standard care, which in most cases included MPI. The CTA group had a higher rate of discharge from the ED: 50%, compared with 23%. They also had a median 18.0-hour length of stay, significantly shorter than the 24.8 hours in the control group (N. Engl. J. Med. 2012;366:1393-403).

Moreover, the CTA group had a higher rate of detection of CAD – 9.0% compared to 3.5% because CTA can detect nonobstructive as well as obstructive plaque, added Dr. Kramer, professor of cardiology and of radiology and director of the cardiovascular imaging center at the University of Virginia, Charlottesville.

The ROMICAT II (Rule Out Myocardial Ischemia/Infarction by Computer-Assisted Tomography) trial was a similar story. ROMICAT II was a multicenter trial in which 1,000 patients who presented to an ED with symptoms suggestive of ACS but a nondiagnostic ECG were randomized to CTA or a standard ED evaluation, most often including MPI. A total of 47% of patients in the CTA group were discharged directly from the ED, compared with just 12% in the control arm. Most notably, the median length of stay in the ED was 8.6 hours with CTA vs. 26.7 hours with a standard evaluation, and the mean length of stay in the hospital was reduced from 30.8 hours with standard evaluation to 23.2 hours, a 7.6-hour reduction (N. Engl. J. Med. 2012;367:299-308).

Follow-up demonstrated no cases of undetected ACS occurred in either group.

The average radiation dose to patients in the CTA arm was 14.3 mSv, compared with about 10 mSv for those patients in the standard evaluation group who got MPI. However, ROMICAT II didn’t use the latest CTA equipment.

"With the modern CTA units, the radiation doses have come down quite a lot. With our flash 256 detector in the ED, we can do a study in a patient with a controlled heart rate and a reasonable BMI [body mass index] with about 1 mSv of radiation now," Dr. Kramer recalled.

Costs in the ED were significantly lower in the CTA group than in controls in ROMICAT II because of the faster patient throughput. However, this savings was neutralized by higher in-hospital costs because the CTA group had more cardiac catheterizations and revascularization procedures because more cases of CAD were detected, as in ACRIN-PA.

"The real advantage to CTA is the rapid discharge from the ED. That’s what patients appreciate. They get their CT angiogram and they can go home within an hour or 2. It really improves their experience in the ED, rather than spending all night waiting for their SPECT MPI results. I don’t think you’re saving costs by doing CTA, though, because of this catch-up phenomenon," Dr. Kramer said.

Aside from the fact that MPI takes 3-5 hours, its other main limitations are that it has difficulty in differentiating acute ischemia from acute infarction or an old infarct, according to Dr. Kramer.

The ability of CTA to detect more cases of CAD by identifying nonobstructive plaque may prove to be of clinical import. That’s the underlying hypothesis of the ongoing National Heart, Lung, and Blood Institute–funded PROMISE (Prospective Multicenter Imaging Study for Evaluation of Chest Pain) study, in which 10,000 patients with symptoms suggestive of CAD are being randomized to an initial CTA or to usual care with a functional test, either MPI, stress echocardiography, or exercise ECG.

This study is being conducted in the offices of primary care physicians and cardiologists rather than in EDs. PROMISE has clinical endpoints as its primary outcomes. The study hypothesis is that intervening in patients identified as having nonobstructive CAD will yield improved outcomes. Results remain several years off, Dr. Kramer said.

ED physicians are eager for high-tech help in quickly and reliably ruling out ACS. Acute chest pain accounts for more than 8 million ED visits annually, but in only 1.19 million admissions for ACS.

Besides MPI and CTA, the other two noninvasive imaging technologies available for use in the ED are cardiac magnetic resonance (CMR) and contrast echocardiography. Neither utilizes radiation – a big plus. Yet neither is as widely used as MPI or CTA. That’s because magnetic resonance takes longer than CTA does, and the scanner may not be available at a moment’s notice, as it really needs to be, when a patient presents with chest pain to the ED. Also, expertise in CMR is not widely available. This is also an issue for contrast echo in the ED.

"The problem with contrast echo is that there are really very few centers around the world that can do it well. It really hasn’t caught on in terms of utilization in the ED," according to the cardiologist.

He reported that he serves as a consultant to Synarc and receives research support from Siemens Medical Solutions.

[email protected]

CHICAGO – Dual-energy computed tomography may have a role in clearing patients with embedded bullets and other metallic objects for magnetic resonance imaging who would otherwise be excluded because of safety concerns.

A Swiss analysis of bullets and shotgun pellets found that dual-energy CT had a high discriminatory power to distinguish between projectiles with and without ferromagnetic properties. MR safety depends on whether a metallic object has ferromagnetic properties – something that most patients don’t know or may be unable to communicate during an emergency evaluation, study author Dr. Sebastian Winklhofer said at the annual meeting of the Radiological Society of North America.

Patrice Wendling/IMNG Medical Media

"Dual-energy measurements may contribute to MR safety and allow for MR imaging of patients with retained projectiles," he said.

Retained metallic objects such as bullets, shrapnel, and medical devices are frequently contraindicated for MRI, because they might move, dislodge, or accelerate at dangerously high velocities toward the scanner’s magnet. This has led to patient injury, and in some cases death, if the object is located near or in a critical anatomic structure.

Session comoderator Dr. Seppo Koskinen of Helsinki University Central Hospital, Finland, said he recently had to forgo MR imaging in a patient with a bullet lodged in the spinal canal because the patient didn’t know if it was lead or steel.

"We didn’t do an MR, but maybe with this system we could have," he said in an interview. "So it has a real clinical impact."

Unsafe exams and misinformed refusals to refer or scan a patient are also occurring as a result of confusion over the ever increasing number of metallic orthopedic and cardiovascular implants. For example, a device that is known to be safe on a 1.5 tesla scanner may be contraindicated on a more powerful 3T scanner or on other scanners using different settings.

When the American College of Radiology penned its Safe MR Practices guidelines in 2007, no cardiac pacemakers or implantable defibrillators were labeled as safe or conditionally safe for MR imaging, prompting the college to recommend that routine MRI is inadvisable in patients with these devices.

In its scientific statement on the controversial topic, the American Heart Association says data are available to support MR imaging in patients with cardiovascular devices, but recommends a series of precautions, including careful patient screening and an accurate determination of the device and its properties (Circulation 2007:116:2878-91).

Dual-energy CT has been used in research and clinical settings to assess the composition of various objects, said Dr. Winklhofer of University Hospital Zurich, Switzerland. He highlighted one study showing that dual-energy CT had a 98% specificity in differentiating between urinary stones that did or did not contain uric acid (Invest. Radiology 2010;45:1-6).

In the current study, nine bullets and two shotgun pellets were scanned in an ex vivo chest phantom using second-generation dual-source CT (SOMATOM Definition Flash, Siemens Healthcare), with tube voltages set at 80, 100, 120, and 140 kVp. Two readers, blinded to the ferromagnetic properties of the projectiles, independently assessed CT numbers on 44 images reconstructed using the extended CT scale technique, which helps overcome artifacts arising from metallic objects. A CT number is the density assigned to a voxel in a CT scan.

Dual-energy indices (DEI) were calculated from 80/140 kVP and 100/140 kVP pairs, and receiver operating characteristics analyses were fitted to predict ferromagnetic properties by means of DEI and CT numbers.

The bullets/pellets ranged in diameter from 2 mm to 14 mm; five were ferromagnetic and six nonferromagnetic.

Intrareader agreement was significantly correlated with mean CT number measurements (P less than .001), with excellent intraclass correlation coefficient agreement (Reader 1: ICC = 0.998; Reader 2: ICC = 0.963), Dr. Winklhofer said.

Interreader agreement was also significantly correlated (P less than .001), and again, there was excellent ICC agreement (both readers ICC = 0.988).

In contrast, when the same projectiles were scanned with single-energy CT, no significant differences in CT numbers for ferromagnetic vs. nonferromagnetic projectiles were observed, said Dr. Winklhofer, who received the RSNA Trainee Research prize for his work.

"Single-energy does not allow for differentiation between those types of projectiles, whereas the dual-energy results show it is possible to differentiate between ferromagnetic and nonferromagnetic projectiles," he said.

Finally, the dual-energy indices of ferromagnetic projectiles were significantly higher than were those of nonferromagnetic projectiles for both the 80/140 kVP and 100/140 kVP energy pairs (at a P value less than .10).

Before dual-energy CT is ready for prime time, however, additional studies are needed to prove that the technique is valid in phantoms of other body regions and in humans, Dr. Winklhofer said in an interview. The prospective study is currently underway.

Dr. Winklhofer and Dr. Koskinen reported no conflicts of interest.

[email protected]

CHICAGO – Dual-energy computed tomography may have a role in clearing patients with embedded bullets and other metallic objects for magnetic resonance imaging who would otherwise be excluded because of safety concerns.

A Swiss analysis of bullets and shotgun pellets found that dual-energy CT had a high discriminatory power to distinguish between projectiles with and without ferromagnetic properties. MR safety depends on whether a metallic object has ferromagnetic properties – something that most patients don’t know or may be unable to communicate during an emergency evaluation, study author Dr. Sebastian Winklhofer said at the annual meeting of the Radiological Society of North America.

Patrice Wendling/IMNG Medical Media

"Dual-energy measurements may contribute to MR safety and allow for MR imaging of patients with retained projectiles," he said.

Retained metallic objects such as bullets, shrapnel, and medical devices are frequently contraindicated for MRI, because they might move, dislodge, or accelerate at dangerously high velocities toward the scanner’s magnet. This has led to patient injury, and in some cases death, if the object is located near or in a critical anatomic structure.

Session comoderator Dr. Seppo Koskinen of Helsinki University Central Hospital, Finland, said he recently had to forgo MR imaging in a patient with a bullet lodged in the spinal canal because the patient didn’t know if it was lead or steel.

"We didn’t do an MR, but maybe with this system we could have," he said in an interview. "So it has a real clinical impact."

Unsafe exams and misinformed refusals to refer or scan a patient are also occurring as a result of confusion over the ever increasing number of metallic orthopedic and cardiovascular implants. For example, a device that is known to be safe on a 1.5 tesla scanner may be contraindicated on a more powerful 3T scanner or on other scanners using different settings.

When the American College of Radiology penned its Safe MR Practices guidelines in 2007, no cardiac pacemakers or implantable defibrillators were labeled as safe or conditionally safe for MR imaging, prompting the college to recommend that routine MRI is inadvisable in patients with these devices.

In its scientific statement on the controversial topic, the American Heart Association says data are available to support MR imaging in patients with cardiovascular devices, but recommends a series of precautions, including careful patient screening and an accurate determination of the device and its properties (Circulation 2007:116:2878-91).

Dual-energy CT has been used in research and clinical settings to assess the composition of various objects, said Dr. Winklhofer of University Hospital Zurich, Switzerland. He highlighted one study showing that dual-energy CT had a 98% specificity in differentiating between urinary stones that did or did not contain uric acid (Invest. Radiology 2010;45:1-6).

In the current study, nine bullets and two shotgun pellets were scanned in an ex vivo chest phantom using second-generation dual-source CT (SOMATOM Definition Flash, Siemens Healthcare), with tube voltages set at 80, 100, 120, and 140 kVp. Two readers, blinded to the ferromagnetic properties of the projectiles, independently assessed CT numbers on 44 images reconstructed using the extended CT scale technique, which helps overcome artifacts arising from metallic objects. A CT number is the density assigned to a voxel in a CT scan.

Dual-energy indices (DEI) were calculated from 80/140 kVP and 100/140 kVP pairs, and receiver operating characteristics analyses were fitted to predict ferromagnetic properties by means of DEI and CT numbers.

The bullets/pellets ranged in diameter from 2 mm to 14 mm; five were ferromagnetic and six nonferromagnetic.

Intrareader agreement was significantly correlated with mean CT number measurements (P less than .001), with excellent intraclass correlation coefficient agreement (Reader 1: ICC = 0.998; Reader 2: ICC = 0.963), Dr. Winklhofer said.

Interreader agreement was also significantly correlated (P less than .001), and again, there was excellent ICC agreement (both readers ICC = 0.988).

In contrast, when the same projectiles were scanned with single-energy CT, no significant differences in CT numbers for ferromagnetic vs. nonferromagnetic projectiles were observed, said Dr. Winklhofer, who received the RSNA Trainee Research prize for his work.

"Single-energy does not allow for differentiation between those types of projectiles, whereas the dual-energy results show it is possible to differentiate between ferromagnetic and nonferromagnetic projectiles," he said.

Finally, the dual-energy indices of ferromagnetic projectiles were significantly higher than were those of nonferromagnetic projectiles for both the 80/140 kVP and 100/140 kVP energy pairs (at a P value less than .10).

Before dual-energy CT is ready for prime time, however, additional studies are needed to prove that the technique is valid in phantoms of other body regions and in humans, Dr. Winklhofer said in an interview. The prospective study is currently underway.

Dr. Winklhofer and Dr. Koskinen reported no conflicts of interest.

[email protected]

CHICAGO – Dual-energy computed tomography may have a role in clearing patients with embedded bullets and other metallic objects for magnetic resonance imaging who would otherwise be excluded because of safety concerns.

A Swiss analysis of bullets and shotgun pellets found that dual-energy CT had a high discriminatory power to distinguish between projectiles with and without ferromagnetic properties. MR safety depends on whether a metallic object has ferromagnetic properties – something that most patients don’t know or may be unable to communicate during an emergency evaluation, study author Dr. Sebastian Winklhofer said at the annual meeting of the Radiological Society of North America.

Patrice Wendling/IMNG Medical Media

"Dual-energy measurements may contribute to MR safety and allow for MR imaging of patients with retained projectiles," he said.

Retained metallic objects such as bullets, shrapnel, and medical devices are frequently contraindicated for MRI, because they might move, dislodge, or accelerate at dangerously high velocities toward the scanner’s magnet. This has led to patient injury, and in some cases death, if the object is located near or in a critical anatomic structure.

Session comoderator Dr. Seppo Koskinen of Helsinki University Central Hospital, Finland, said he recently had to forgo MR imaging in a patient with a bullet lodged in the spinal canal because the patient didn’t know if it was lead or steel.

"We didn’t do an MR, but maybe with this system we could have," he said in an interview. "So it has a real clinical impact."

Unsafe exams and misinformed refusals to refer or scan a patient are also occurring as a result of confusion over the ever increasing number of metallic orthopedic and cardiovascular implants. For example, a device that is known to be safe on a 1.5 tesla scanner may be contraindicated on a more powerful 3T scanner or on other scanners using different settings.

When the American College of Radiology penned its Safe MR Practices guidelines in 2007, no cardiac pacemakers or implantable defibrillators were labeled as safe or conditionally safe for MR imaging, prompting the college to recommend that routine MRI is inadvisable in patients with these devices.

In its scientific statement on the controversial topic, the American Heart Association says data are available to support MR imaging in patients with cardiovascular devices, but recommends a series of precautions, including careful patient screening and an accurate determination of the device and its properties (Circulation 2007:116:2878-91).

Dual-energy CT has been used in research and clinical settings to assess the composition of various objects, said Dr. Winklhofer of University Hospital Zurich, Switzerland. He highlighted one study showing that dual-energy CT had a 98% specificity in differentiating between urinary stones that did or did not contain uric acid (Invest. Radiology 2010;45:1-6).

In the current study, nine bullets and two shotgun pellets were scanned in an ex vivo chest phantom using second-generation dual-source CT (SOMATOM Definition Flash, Siemens Healthcare), with tube voltages set at 80, 100, 120, and 140 kVp. Two readers, blinded to the ferromagnetic properties of the projectiles, independently assessed CT numbers on 44 images reconstructed using the extended CT scale technique, which helps overcome artifacts arising from metallic objects. A CT number is the density assigned to a voxel in a CT scan.

Dual-energy indices (DEI) were calculated from 80/140 kVP and 100/140 kVP pairs, and receiver operating characteristics analyses were fitted to predict ferromagnetic properties by means of DEI and CT numbers.

The bullets/pellets ranged in diameter from 2 mm to 14 mm; five were ferromagnetic and six nonferromagnetic.

Intrareader agreement was significantly correlated with mean CT number measurements (P less than .001), with excellent intraclass correlation coefficient agreement (Reader 1: ICC = 0.998; Reader 2: ICC = 0.963), Dr. Winklhofer said.

Interreader agreement was also significantly correlated (P less than .001), and again, there was excellent ICC agreement (both readers ICC = 0.988).

In contrast, when the same projectiles were scanned with single-energy CT, no significant differences in CT numbers for ferromagnetic vs. nonferromagnetic projectiles were observed, said Dr. Winklhofer, who received the RSNA Trainee Research prize for his work.

"Single-energy does not allow for differentiation between those types of projectiles, whereas the dual-energy results show it is possible to differentiate between ferromagnetic and nonferromagnetic projectiles," he said.

Finally, the dual-energy indices of ferromagnetic projectiles were significantly higher than were those of nonferromagnetic projectiles for both the 80/140 kVP and 100/140 kVP energy pairs (at a P value less than .10).

Before dual-energy CT is ready for prime time, however, additional studies are needed to prove that the technique is valid in phantoms of other body regions and in humans, Dr. Winklhofer said in an interview. The prospective study is currently underway.

Dr. Winklhofer and Dr. Koskinen reported no conflicts of interest.

[email protected]