Mitchel L. Zoler, PhD

SAN ANTONIO — The largest-ever, head-to-head comparison of stenting versus surgery for treating severe carotid artery stenosis showed a marked effect of age, with patients older than 70 having fewer adverse outcomes after carotid endarterectomy and patients younger than 70 having fewer complications following carotid angioplasty and stenting.

Although the highly anticipated results from the decade-long Carotid Revascularization Endarterectomy vs. Surgery Trial (CREST) seemed, in simplest terms, to show a dead heat between carotid stenting and surgery (see table), the results reported at the International Stroke Conference actually revealed statistically significant and clinically important differences between the two treatments.

The statistically significant interaction between patient age and outcome will likely play a major role when physicians and patients now decide which intervention to favor for a specific patient.

The CREST results also showed another significant difference between carotid surgery and stenting: Surgery led to a 1.2% increased absolute rate in the incidence of periprocedural MIs, whereas stenting produced a 1.8% increase absolute rate of periprocedural strokes, a finding that will force patients and their physicians to ask themselves which complication they would rather risk.

The patients in CREST answered that question, at least in part, via another outcome measure. Assessment of patient physical and mental quality of life with the 36-item Short Form (SF-36) Health Survey a year after treatment showed that patients who developed new strokes, even “minor” strokes, had statistically significant reductions in both their mental and physical well-being compared with baseline, whereas patients who developed new MIs had, on average, no significant changes in their SF-36 mental and physical scores, Dr. Wayne M. Clark reported while presenting the CREST results.

The CREST report prompted some experts to highlight that the stenting results in the trial came from selected, experienced operators and that it would be a leap to expect comparable results from physicians who were not trained as well as the more experienced operators.

CREST randomized 2,502 patients with either symptomatic carotid stenosis or asymptomatic severe carotid stenosis (at least 60% blockage) at 108 sites in the United States and 9 in Canada. The patients' average age was 69 years, a third were women, and 47% were asymptomatic. The analysis showed no significant effect from either gender or symptom status on outcomes.

The age effect produced the sharpest distinction between stenting and surgery, and confirmed evidence that began emerging a few years ago that carotid stenting poses a special problem for elderly patients. In fact, some of the first suggestions of safety problems that can occur when stenting elderly patients came from the lead-in phase of CREST, a stage that involved nearly 1,600 patients who underwent carotid stenting in the early 2000s as operators in the study established their stenting expertise. The problem has been attributed to the increased difficulty and danger of placing stents and embolic protection devices through elderly patients' tortuous and atherosclerotic arteries.

“I think it's likely that putting in the embolic protection device sets off strokes. Until we have more data to show whether or not the age effect is real, I will take it into account in my patients,” commented Dr. J. Donald Easton, a neurologist at the University of California, San Francisco. Dr. Clark reported the age effect as a continuous variable, without specifying any point estimates of the effect. But based on the line graph he showed, patients who underwent stenting at age 65 had a roughly 20% reduced risk for an adverse perioperative or long-term outcome compared with those who underwent surgery, whereas at age 60 the relative benefit from stenting was about 35%, and at age 50, the rate of adverse outcomes after stenting was less than half the rate after endarterectomy.

The primary adverse-event measure used in CREST was the composite rate of any stroke, myocardial infarction, or death during the 30 days following treatment plus the rate of any ipsilateral stroke during long-term follow-up of up to 4 years. This rate was 7.2% for stenting and 6.8% for endarterectomy, with similar rates of ipsilateral strokes occurring from 31 days to 4 years (2.0% vs. 2.4%, respectively).

In contrast to younger patients, at age 75, the rate of adverse outcomes after stenting rose by about 35% compared with surgery; at age 80, the adverse-outcome rate was more than 50% higher with stenting than with surgery; and at age 85, the adverse event rate was roughly doubled by stenting in comparison with endarterectomy. In patients who were 70 years old, the adverse event rates were essentially identical regardless of which procedure was used.

The CREST results reported so far gave no details on how endarterectomy and stenting fared in asymptomatic patients compared with patients who already had symptoms of carotid disease. In the absence of these data, several experts cautioned that the findings should not be taken as an endorsement of aggressive carotid interventions for asymptomatic patients, especially at a time when medical therapy has become very effective.

CREST was funded by the National Institute of Neurological Disorders and Stroke. Dr. Clark and Dr. Easton had no disclosures relevant to the study.

Source Elsevier Global Medical News

SAN ANTONIO — The largest-ever, head-to-head comparison of stenting versus surgery for treating severe carotid artery stenosis showed a marked effect of age, with patients older than 70 having fewer adverse outcomes after carotid endarterectomy and patients younger than 70 having fewer complications following carotid angioplasty and stenting.

Although the highly anticipated results from the decade-long Carotid Revascularization Endarterectomy vs. Surgery Trial (CREST) seemed, in simplest terms, to show a dead heat between carotid stenting and surgery (see table), the results reported at the International Stroke Conference actually revealed statistically significant and clinically important differences between the two treatments.

The statistically significant interaction between patient age and outcome will likely play a major role when physicians and patients now decide which intervention to favor for a specific patient.

The CREST results also showed another significant difference between carotid surgery and stenting: Surgery led to a 1.2% increased absolute rate in the incidence of periprocedural MIs, whereas stenting produced a 1.8% increase absolute rate of periprocedural strokes, a finding that will force patients and their physicians to ask themselves which complication they would rather risk.

The patients in CREST answered that question, at least in part, via another outcome measure. Assessment of patient physical and mental quality of life with the 36-item Short Form (SF-36) Health Survey a year after treatment showed that patients who developed new strokes, even “minor” strokes, had statistically significant reductions in both their mental and physical well-being compared with baseline, whereas patients who developed new MIs had, on average, no significant changes in their SF-36 mental and physical scores, Dr. Wayne M. Clark reported while presenting the CREST results.

The CREST report prompted some experts to highlight that the stenting results in the trial came from selected, experienced operators and that it would be a leap to expect comparable results from physicians who were not trained as well as the more experienced operators.

CREST randomized 2,502 patients with either symptomatic carotid stenosis or asymptomatic severe carotid stenosis (at least 60% blockage) at 108 sites in the United States and 9 in Canada. The patients' average age was 69 years, a third were women, and 47% were asymptomatic. The analysis showed no significant effect from either gender or symptom status on outcomes.

The age effect produced the sharpest distinction between stenting and surgery, and confirmed evidence that began emerging a few years ago that carotid stenting poses a special problem for elderly patients. In fact, some of the first suggestions of safety problems that can occur when stenting elderly patients came from the lead-in phase of CREST, a stage that involved nearly 1,600 patients who underwent carotid stenting in the early 2000s as operators in the study established their stenting expertise. The problem has been attributed to the increased difficulty and danger of placing stents and embolic protection devices through elderly patients' tortuous and atherosclerotic arteries.

“I think it's likely that putting in the embolic protection device sets off strokes. Until we have more data to show whether or not the age effect is real, I will take it into account in my patients,” commented Dr. J. Donald Easton, a neurologist at the University of California, San Francisco. Dr. Clark reported the age effect as a continuous variable, without specifying any point estimates of the effect. But based on the line graph he showed, patients who underwent stenting at age 65 had a roughly 20% reduced risk for an adverse perioperative or long-term outcome compared with those who underwent surgery, whereas at age 60 the relative benefit from stenting was about 35%, and at age 50, the rate of adverse outcomes after stenting was less than half the rate after endarterectomy.

The primary adverse-event measure used in CREST was the composite rate of any stroke, myocardial infarction, or death during the 30 days following treatment plus the rate of any ipsilateral stroke during long-term follow-up of up to 4 years. This rate was 7.2% for stenting and 6.8% for endarterectomy, with similar rates of ipsilateral strokes occurring from 31 days to 4 years (2.0% vs. 2.4%, respectively).

In contrast to younger patients, at age 75, the rate of adverse outcomes after stenting rose by about 35% compared with surgery; at age 80, the adverse-outcome rate was more than 50% higher with stenting than with surgery; and at age 85, the adverse event rate was roughly doubled by stenting in comparison with endarterectomy. In patients who were 70 years old, the adverse event rates were essentially identical regardless of which procedure was used.

The CREST results reported so far gave no details on how endarterectomy and stenting fared in asymptomatic patients compared with patients who already had symptoms of carotid disease. In the absence of these data, several experts cautioned that the findings should not be taken as an endorsement of aggressive carotid interventions for asymptomatic patients, especially at a time when medical therapy has become very effective.

CREST was funded by the National Institute of Neurological Disorders and Stroke. Dr. Clark and Dr. Easton had no disclosures relevant to the study.

Source Elsevier Global Medical News

SAN ANTONIO — The largest-ever, head-to-head comparison of stenting versus surgery for treating severe carotid artery stenosis showed a marked effect of age, with patients older than 70 having fewer adverse outcomes after carotid endarterectomy and patients younger than 70 having fewer complications following carotid angioplasty and stenting.

Although the highly anticipated results from the decade-long Carotid Revascularization Endarterectomy vs. Surgery Trial (CREST) seemed, in simplest terms, to show a dead heat between carotid stenting and surgery (see table), the results reported at the International Stroke Conference actually revealed statistically significant and clinically important differences between the two treatments.

The statistically significant interaction between patient age and outcome will likely play a major role when physicians and patients now decide which intervention to favor for a specific patient.

The CREST results also showed another significant difference between carotid surgery and stenting: Surgery led to a 1.2% increased absolute rate in the incidence of periprocedural MIs, whereas stenting produced a 1.8% increase absolute rate of periprocedural strokes, a finding that will force patients and their physicians to ask themselves which complication they would rather risk.

The patients in CREST answered that question, at least in part, via another outcome measure. Assessment of patient physical and mental quality of life with the 36-item Short Form (SF-36) Health Survey a year after treatment showed that patients who developed new strokes, even “minor” strokes, had statistically significant reductions in both their mental and physical well-being compared with baseline, whereas patients who developed new MIs had, on average, no significant changes in their SF-36 mental and physical scores, Dr. Wayne M. Clark reported while presenting the CREST results.

The CREST report prompted some experts to highlight that the stenting results in the trial came from selected, experienced operators and that it would be a leap to expect comparable results from physicians who were not trained as well as the more experienced operators.

CREST randomized 2,502 patients with either symptomatic carotid stenosis or asymptomatic severe carotid stenosis (at least 60% blockage) at 108 sites in the United States and 9 in Canada. The patients' average age was 69 years, a third were women, and 47% were asymptomatic. The analysis showed no significant effect from either gender or symptom status on outcomes.

The age effect produced the sharpest distinction between stenting and surgery, and confirmed evidence that began emerging a few years ago that carotid stenting poses a special problem for elderly patients. In fact, some of the first suggestions of safety problems that can occur when stenting elderly patients came from the lead-in phase of CREST, a stage that involved nearly 1,600 patients who underwent carotid stenting in the early 2000s as operators in the study established their stenting expertise. The problem has been attributed to the increased difficulty and danger of placing stents and embolic protection devices through elderly patients' tortuous and atherosclerotic arteries.

“I think it's likely that putting in the embolic protection device sets off strokes. Until we have more data to show whether or not the age effect is real, I will take it into account in my patients,” commented Dr. J. Donald Easton, a neurologist at the University of California, San Francisco. Dr. Clark reported the age effect as a continuous variable, without specifying any point estimates of the effect. But based on the line graph he showed, patients who underwent stenting at age 65 had a roughly 20% reduced risk for an adverse perioperative or long-term outcome compared with those who underwent surgery, whereas at age 60 the relative benefit from stenting was about 35%, and at age 50, the rate of adverse outcomes after stenting was less than half the rate after endarterectomy.

The primary adverse-event measure used in CREST was the composite rate of any stroke, myocardial infarction, or death during the 30 days following treatment plus the rate of any ipsilateral stroke during long-term follow-up of up to 4 years. This rate was 7.2% for stenting and 6.8% for endarterectomy, with similar rates of ipsilateral strokes occurring from 31 days to 4 years (2.0% vs. 2.4%, respectively).

In contrast to younger patients, at age 75, the rate of adverse outcomes after stenting rose by about 35% compared with surgery; at age 80, the adverse-outcome rate was more than 50% higher with stenting than with surgery; and at age 85, the adverse event rate was roughly doubled by stenting in comparison with endarterectomy. In patients who were 70 years old, the adverse event rates were essentially identical regardless of which procedure was used.

The CREST results reported so far gave no details on how endarterectomy and stenting fared in asymptomatic patients compared with patients who already had symptoms of carotid disease. In the absence of these data, several experts cautioned that the findings should not be taken as an endorsement of aggressive carotid interventions for asymptomatic patients, especially at a time when medical therapy has become very effective.

CREST was funded by the National Institute of Neurological Disorders and Stroke. Dr. Clark and Dr. Easton had no disclosures relevant to the study.

Source Elsevier Global Medical News

ATLANTA—Results from a study branded by its principal investigator as underpowered to produce a meaningful result still sparked attention at a major cardiology meeting by fanning the controversy swirling around clopidogrel's role following percutaneous coronary interventions with drug-eluting stents.

The Korean study that tried to test the long-term role of clopidogrel for preventing adverse cardiovascular events following placement of drug-eluting stents (DES) in roughly 2,700 patients “had insufficient statistical power to allow a firm conclusion,” Dr. Seung-Jung Park said at the annual meeting of the American College of Cardiology. That fact mitigated what would have otherwise been a highly surprising and troubling finding: More than a year out from coronary stenting, patients treated with aspirin alone fared no worse than and even trended toward better outcomes compared with patients maintained on dual-antiplatelet therapy with aspirin and clopidogrel.

The underpowered study size might, in other circumstances, have caused the report to be dismissed and quickly forgotten. But two extenuating circumstances instead thrust the study into the spotlight: First, despite its problems, the study simultaneously ran in the New England Journal of Medicine (2010 March 15 [doi:10.1056/NEJMoa1001266]). Second, the report came just days after the Food and Drug Administration on March 12 roiled concerns about clopidogrel's efficacy in patients who recently received a coronary stent by adding a boxed warning to the label of clopidogrel (Plavix) alerting prescribers that certain patients do not metabolize clopidogrel effectively, thereby blunting the drug's efficacy in these people (see article below). Such “poor metabolizers,” the FDA said, comprise an estimated 2%-14% of the American public and perhaps as high as 50% of some Asian populations.

“We see tremendous variability of responsiveness to clopidogrel and aspirin” in patients attributable to genetic differences in features such as the metabolic activation of clopidogrel, said Dr. George D. Dangas, a cardiologist at the Center for Interventional Vascular Therapy at Columbia University in New York. “How can we have a question of [clopidogrel treatment] duration in patients who are not responding? I'm not sure that makes much sense. Perhaps patients in Dr. Park's study were hyporesponders [to clopidogrel] and that's behind what he sees.”

The Korean study enrolled 2,701 patients who had received at least one DES and had been event free while on combined antiplatelet therapy with aspirin and clopidogrel for at least 12 months. Their average age was 62, and 70% were men. A median of 13 months after stent placement, the researchers randomized the patients to continue on 75 mg clopidogrel and 100-200 mg aspirin daily or just aspirin alone. Follow-up continued for a median of 19 months, but the total number of end point events remained low, about a quarter of the expected number, probably because the study involved low-risk patients, said Dr. Park, professor of medicine in the Heart Institute at Asan Medical Center in Seoul, South Korea.

The primary end point, the combined rate of MI or cardiac death, occurred in 1.8% of patients treated with clopidogrel and aspirin and in 1.2% of those on aspirin only, a nonsignificant, 65% relative increased risk of events among patients on the dual-antiplatelet regimen compared with aspirin alone.

In two other outcome measures the worse performance by the combined regimen just missed statistical significance. The combined rate of MI, stroke, or death from any caused occurred in 3.2% of the combined-treatment patients and in 1.8% of the aspirin-alone controls, and the rate of MI, stroke, or cardiac death tallied in 2.7% of the aspirin plus clopidogrel patients compared with 1.3% of patients on aspirin only. Rates of all-cause death and stent thrombosis were nearly identical in both treatment groups.

Many experts who heard these potentially troubling findings that seemingly cast doubt on clopidogrel's efficacy and safety as well as on prolonged dual-antiplatelet therapy following coronary stenting uniformly dismissed the findings as unreliable.

“The answers are not definitive. The lack of power is the primary concern,” said Dr. Laura Mauri, chief scientific officer of the Harvard Clinical Research Institute in Boston.

“We won't know [how long to treat these patients with clopidogrel] until we have an adequately powered study,” said Dr. Dean J. Kereiakes, chief executive officer of the Ohio Heart Health Center in Cincinnati.

While Dr. Dangas agreed that the results were inconclusive, he suggested that they may offer some guidance “until definitive studies come out.” The results were “reassuring that perhaps in patients who did well over the first year [following placement of DES] it might be okay to consider taking them off clopidogrel,” he said.

The study received no industry support. Dr. Park said that he and his associates had no disclosures.

Dr. Dangas reported financial relationships with several pharmaceutical and device companies, including Daiichi-Sankyo, Sanofi-Aventis, Boston Scientific, AstraZeneca, and Cordis. Dr. Mauri reported receiving consulting fees or honoraria from Cordis and Medtronic Vascular. Dr. Kereiakes reported financial relationships with Reva Medical, Eli Lilly, Boston Scientific, Cordis, Devax, and Abbott Vascular, Amylin, and Daiichi Sankyo, among other drug and device makers.

My Take

Results Won't Change My Practice

Despite the study's limited power, it generates some interesting hypotheses. Perhaps we need to consider the level of risk that patients face from major adverse events following coronary stenting with DES when evaluating dual antiplatelet therapy. The new results suggest that in low-risk patients this balance tips in favor of stopping dual antiplatelet drug therapy a year after stenting. It's not clear what mechanism might produce the apparent risk from clopidogrel treatment beyond 1 year in this study.

Asian populations have a high prevalence of cytochrome P2C19 genes that produce little or no active enzyme needed to metabolize clopidogrel to its active form. This may mean that many patients in the study were genetically unable to benefit from clopidogrel treatment.

The new results suggesting that low-risk patients exist who may not benefit from continued clopidogrel treatment, are not convincing. I have sufficient uncertainty that I'm not willing to change my practice, even in low-risk patients. My approach has been to have a low threshold for continuing dual-antiplatelet therapy in DES patients. Until now, all of the data supporting this approach came from observational studies. This is no substitute for prospective, controlled studies, so the Korean study is a laudable first step. What's needed are larger studies with longer follow-up, such as study the Dual Antiplatelet Therapy (DAPT) study, with an expected enrollment of more than 20,000 patients.

DR. ELLIOTT M. ANTMAN is a professor of medicine at Harvard Medical School in Boston. He was principal investigator for TRITON-TIMI 38, the pivotal trial of prasugrel, sponsored by Daiichi-Sankyo. He has financial relationships with Sanofi-Aventis, Momenta, and Eli Lilly, and has received research grants from 22 companies.

ATLANTA—Results from a study branded by its principal investigator as underpowered to produce a meaningful result still sparked attention at a major cardiology meeting by fanning the controversy swirling around clopidogrel's role following percutaneous coronary interventions with drug-eluting stents.

The Korean study that tried to test the long-term role of clopidogrel for preventing adverse cardiovascular events following placement of drug-eluting stents (DES) in roughly 2,700 patients “had insufficient statistical power to allow a firm conclusion,” Dr. Seung-Jung Park said at the annual meeting of the American College of Cardiology. That fact mitigated what would have otherwise been a highly surprising and troubling finding: More than a year out from coronary stenting, patients treated with aspirin alone fared no worse than and even trended toward better outcomes compared with patients maintained on dual-antiplatelet therapy with aspirin and clopidogrel.

The underpowered study size might, in other circumstances, have caused the report to be dismissed and quickly forgotten. But two extenuating circumstances instead thrust the study into the spotlight: First, despite its problems, the study simultaneously ran in the New England Journal of Medicine (2010 March 15 [doi:10.1056/NEJMoa1001266]). Second, the report came just days after the Food and Drug Administration on March 12 roiled concerns about clopidogrel's efficacy in patients who recently received a coronary stent by adding a boxed warning to the label of clopidogrel (Plavix) alerting prescribers that certain patients do not metabolize clopidogrel effectively, thereby blunting the drug's efficacy in these people (see article below). Such “poor metabolizers,” the FDA said, comprise an estimated 2%-14% of the American public and perhaps as high as 50% of some Asian populations.

“We see tremendous variability of responsiveness to clopidogrel and aspirin” in patients attributable to genetic differences in features such as the metabolic activation of clopidogrel, said Dr. George D. Dangas, a cardiologist at the Center for Interventional Vascular Therapy at Columbia University in New York. “How can we have a question of [clopidogrel treatment] duration in patients who are not responding? I'm not sure that makes much sense. Perhaps patients in Dr. Park's study were hyporesponders [to clopidogrel] and that's behind what he sees.”

The Korean study enrolled 2,701 patients who had received at least one DES and had been event free while on combined antiplatelet therapy with aspirin and clopidogrel for at least 12 months. Their average age was 62, and 70% were men. A median of 13 months after stent placement, the researchers randomized the patients to continue on 75 mg clopidogrel and 100-200 mg aspirin daily or just aspirin alone. Follow-up continued for a median of 19 months, but the total number of end point events remained low, about a quarter of the expected number, probably because the study involved low-risk patients, said Dr. Park, professor of medicine in the Heart Institute at Asan Medical Center in Seoul, South Korea.

The primary end point, the combined rate of MI or cardiac death, occurred in 1.8% of patients treated with clopidogrel and aspirin and in 1.2% of those on aspirin only, a nonsignificant, 65% relative increased risk of events among patients on the dual-antiplatelet regimen compared with aspirin alone.

In two other outcome measures the worse performance by the combined regimen just missed statistical significance. The combined rate of MI, stroke, or death from any caused occurred in 3.2% of the combined-treatment patients and in 1.8% of the aspirin-alone controls, and the rate of MI, stroke, or cardiac death tallied in 2.7% of the aspirin plus clopidogrel patients compared with 1.3% of patients on aspirin only. Rates of all-cause death and stent thrombosis were nearly identical in both treatment groups.

Many experts who heard these potentially troubling findings that seemingly cast doubt on clopidogrel's efficacy and safety as well as on prolonged dual-antiplatelet therapy following coronary stenting uniformly dismissed the findings as unreliable.

“The answers are not definitive. The lack of power is the primary concern,” said Dr. Laura Mauri, chief scientific officer of the Harvard Clinical Research Institute in Boston.

“We won't know [how long to treat these patients with clopidogrel] until we have an adequately powered study,” said Dr. Dean J. Kereiakes, chief executive officer of the Ohio Heart Health Center in Cincinnati.

While Dr. Dangas agreed that the results were inconclusive, he suggested that they may offer some guidance “until definitive studies come out.” The results were “reassuring that perhaps in patients who did well over the first year [following placement of DES] it might be okay to consider taking them off clopidogrel,” he said.

The study received no industry support. Dr. Park said that he and his associates had no disclosures.

Dr. Dangas reported financial relationships with several pharmaceutical and device companies, including Daiichi-Sankyo, Sanofi-Aventis, Boston Scientific, AstraZeneca, and Cordis. Dr. Mauri reported receiving consulting fees or honoraria from Cordis and Medtronic Vascular. Dr. Kereiakes reported financial relationships with Reva Medical, Eli Lilly, Boston Scientific, Cordis, Devax, and Abbott Vascular, Amylin, and Daiichi Sankyo, among other drug and device makers.

My Take

Results Won't Change My Practice

Despite the study's limited power, it generates some interesting hypotheses. Perhaps we need to consider the level of risk that patients face from major adverse events following coronary stenting with DES when evaluating dual antiplatelet therapy. The new results suggest that in low-risk patients this balance tips in favor of stopping dual antiplatelet drug therapy a year after stenting. It's not clear what mechanism might produce the apparent risk from clopidogrel treatment beyond 1 year in this study.

Asian populations have a high prevalence of cytochrome P2C19 genes that produce little or no active enzyme needed to metabolize clopidogrel to its active form. This may mean that many patients in the study were genetically unable to benefit from clopidogrel treatment.

The new results suggesting that low-risk patients exist who may not benefit from continued clopidogrel treatment, are not convincing. I have sufficient uncertainty that I'm not willing to change my practice, even in low-risk patients. My approach has been to have a low threshold for continuing dual-antiplatelet therapy in DES patients. Until now, all of the data supporting this approach came from observational studies. This is no substitute for prospective, controlled studies, so the Korean study is a laudable first step. What's needed are larger studies with longer follow-up, such as study the Dual Antiplatelet Therapy (DAPT) study, with an expected enrollment of more than 20,000 patients.

DR. ELLIOTT M. ANTMAN is a professor of medicine at Harvard Medical School in Boston. He was principal investigator for TRITON-TIMI 38, the pivotal trial of prasugrel, sponsored by Daiichi-Sankyo. He has financial relationships with Sanofi-Aventis, Momenta, and Eli Lilly, and has received research grants from 22 companies.

ATLANTA—Results from a study branded by its principal investigator as underpowered to produce a meaningful result still sparked attention at a major cardiology meeting by fanning the controversy swirling around clopidogrel's role following percutaneous coronary interventions with drug-eluting stents.

The Korean study that tried to test the long-term role of clopidogrel for preventing adverse cardiovascular events following placement of drug-eluting stents (DES) in roughly 2,700 patients “had insufficient statistical power to allow a firm conclusion,” Dr. Seung-Jung Park said at the annual meeting of the American College of Cardiology. That fact mitigated what would have otherwise been a highly surprising and troubling finding: More than a year out from coronary stenting, patients treated with aspirin alone fared no worse than and even trended toward better outcomes compared with patients maintained on dual-antiplatelet therapy with aspirin and clopidogrel.

The underpowered study size might, in other circumstances, have caused the report to be dismissed and quickly forgotten. But two extenuating circumstances instead thrust the study into the spotlight: First, despite its problems, the study simultaneously ran in the New England Journal of Medicine (2010 March 15 [doi:10.1056/NEJMoa1001266]). Second, the report came just days after the Food and Drug Administration on March 12 roiled concerns about clopidogrel's efficacy in patients who recently received a coronary stent by adding a boxed warning to the label of clopidogrel (Plavix) alerting prescribers that certain patients do not metabolize clopidogrel effectively, thereby blunting the drug's efficacy in these people (see article below). Such “poor metabolizers,” the FDA said, comprise an estimated 2%-14% of the American public and perhaps as high as 50% of some Asian populations.

“We see tremendous variability of responsiveness to clopidogrel and aspirin” in patients attributable to genetic differences in features such as the metabolic activation of clopidogrel, said Dr. George D. Dangas, a cardiologist at the Center for Interventional Vascular Therapy at Columbia University in New York. “How can we have a question of [clopidogrel treatment] duration in patients who are not responding? I'm not sure that makes much sense. Perhaps patients in Dr. Park's study were hyporesponders [to clopidogrel] and that's behind what he sees.”

The Korean study enrolled 2,701 patients who had received at least one DES and had been event free while on combined antiplatelet therapy with aspirin and clopidogrel for at least 12 months. Their average age was 62, and 70% were men. A median of 13 months after stent placement, the researchers randomized the patients to continue on 75 mg clopidogrel and 100-200 mg aspirin daily or just aspirin alone. Follow-up continued for a median of 19 months, but the total number of end point events remained low, about a quarter of the expected number, probably because the study involved low-risk patients, said Dr. Park, professor of medicine in the Heart Institute at Asan Medical Center in Seoul, South Korea.

The primary end point, the combined rate of MI or cardiac death, occurred in 1.8% of patients treated with clopidogrel and aspirin and in 1.2% of those on aspirin only, a nonsignificant, 65% relative increased risk of events among patients on the dual-antiplatelet regimen compared with aspirin alone.

In two other outcome measures the worse performance by the combined regimen just missed statistical significance. The combined rate of MI, stroke, or death from any caused occurred in 3.2% of the combined-treatment patients and in 1.8% of the aspirin-alone controls, and the rate of MI, stroke, or cardiac death tallied in 2.7% of the aspirin plus clopidogrel patients compared with 1.3% of patients on aspirin only. Rates of all-cause death and stent thrombosis were nearly identical in both treatment groups.

Many experts who heard these potentially troubling findings that seemingly cast doubt on clopidogrel's efficacy and safety as well as on prolonged dual-antiplatelet therapy following coronary stenting uniformly dismissed the findings as unreliable.

“The answers are not definitive. The lack of power is the primary concern,” said Dr. Laura Mauri, chief scientific officer of the Harvard Clinical Research Institute in Boston.

“We won't know [how long to treat these patients with clopidogrel] until we have an adequately powered study,” said Dr. Dean J. Kereiakes, chief executive officer of the Ohio Heart Health Center in Cincinnati.

While Dr. Dangas agreed that the results were inconclusive, he suggested that they may offer some guidance “until definitive studies come out.” The results were “reassuring that perhaps in patients who did well over the first year [following placement of DES] it might be okay to consider taking them off clopidogrel,” he said.

The study received no industry support. Dr. Park said that he and his associates had no disclosures.

Dr. Dangas reported financial relationships with several pharmaceutical and device companies, including Daiichi-Sankyo, Sanofi-Aventis, Boston Scientific, AstraZeneca, and Cordis. Dr. Mauri reported receiving consulting fees or honoraria from Cordis and Medtronic Vascular. Dr. Kereiakes reported financial relationships with Reva Medical, Eli Lilly, Boston Scientific, Cordis, Devax, and Abbott Vascular, Amylin, and Daiichi Sankyo, among other drug and device makers.

My Take

Results Won't Change My Practice

Despite the study's limited power, it generates some interesting hypotheses. Perhaps we need to consider the level of risk that patients face from major adverse events following coronary stenting with DES when evaluating dual antiplatelet therapy. The new results suggest that in low-risk patients this balance tips in favor of stopping dual antiplatelet drug therapy a year after stenting. It's not clear what mechanism might produce the apparent risk from clopidogrel treatment beyond 1 year in this study.

Asian populations have a high prevalence of cytochrome P2C19 genes that produce little or no active enzyme needed to metabolize clopidogrel to its active form. This may mean that many patients in the study were genetically unable to benefit from clopidogrel treatment.

The new results suggesting that low-risk patients exist who may not benefit from continued clopidogrel treatment, are not convincing. I have sufficient uncertainty that I'm not willing to change my practice, even in low-risk patients. My approach has been to have a low threshold for continuing dual-antiplatelet therapy in DES patients. Until now, all of the data supporting this approach came from observational studies. This is no substitute for prospective, controlled studies, so the Korean study is a laudable first step. What's needed are larger studies with longer follow-up, such as study the Dual Antiplatelet Therapy (DAPT) study, with an expected enrollment of more than 20,000 patients.

DR. ELLIOTT M. ANTMAN is a professor of medicine at Harvard Medical School in Boston. He was principal investigator for TRITON-TIMI 38, the pivotal trial of prasugrel, sponsored by Daiichi-Sankyo. He has financial relationships with Sanofi-Aventis, Momenta, and Eli Lilly, and has received research grants from 22 companies.

Major Finding: In patients with type 2 diabetes and a high risk for cardiovascular disease, 2,765 treated with fenofibrate in addition to standard medical therapy had a 2.24%-per-year rate of major fatal or nonfatal cardiovascular events during an average 4.7 years of follow-up. The 2,753 patients randomized to placebo in addition to standard medical therapy had a 2.41%-per-year incidence rate of the end point. The difference in rates between the two groups was not statistically significant.

Data Source: ACCORD, a randomized, controlled lipid trial conducted at 77 sites in the United States and Canada during January 2001–July 2009.

Disclosures: Dr. Ginsberg has financial relationships with several pharmaceutical companies, including Merck and Abbott, which donated the simvastatin and fenofibrate/placebo but had no involvement in ACCORD. The trial was funded by the National Heart, Lung, and Blood Institute.

Atlanta — The failure to significantly reduce the cardiovascular event rate with fenofibrate treatment in a large trial of high-risk type 2 diabetes patients probably occurred because the study enrolled too many of the wrong types of patients to clearly show a benefit from this drug, several experts said.

Instead of focusing on patients with diabetes and dyslipidemia, an elevated serum level of triglycerides, and depressed HDL cholesterol, the Action to Control Cardiovascular Risk in Diabetes (ACCORD) lipid trial enrolled a representative sampling of 5,518 patients with diabetes and a range of triglyceride and HDL cholesterol levels, proving that fibrate treatment, on top of the moderate statin dosage, could not further help most of these patients.

The ACCORD investigators decided to enroll a wide spectrum of patients “to see if [fenofibrate] could apply generally. It's important that we found that fenofibrate on top of a statin will not benefit the majority” of patients with diabetes, Dr. Henry N. Ginsberg said at the annual meeting of the American College of Cardiology.

Existing cholesterol-treatment guidelines from the National Heart, Lung, and Blood Institute—the Adult Treatment Panel III—call for adding a fibrate drug to statin treatment when triglyceride levels are high and HDL cholesterol is low. “I think that's the role for this drug, in patients with the most significant dyslipidemia,” said Dr. Ginsberg, professor of medicine and director of the Irving Institute for Clinical and Translational Research at Columbia University, New York.

In ACCORD, 17% of enrolled patients fell into the subgroup with a plasma triglyceride level of at least 204 mg/dL and a plasma HDL cholesterol that was 34 mg/dL or less. Within this subgroup, fenofibrate treatment produced an improvement in the study's primary end point, the combination of major fatal or nonfatal cardiovascular events, that just missed statistical significance. Tr. Ginsberg said. Concurrently with his report at the meeting, the results were posted online (N. Engl. J. Med. 2010 Mar 14 [doi:10.1056/NEJMoa1001282

“They tested the drug on the wrong patients,” said Dr. Prakash C. Deedwania, a cardiologist at the University of California, San Francisco, in Fresno, Calif. The trial results could potentially have been positive if enrollment had been more focused, he said in an interview.

“The message is that the majority of patients with diabetes don't need a fibrate added,” said Dr. Roger S. Blumenthal, professor of medicine and director of preventive cardiology at Johns Hopkins University in Baltimore. “There was good reason to think the results might have been positive, but in ACCORD it was added to a statin, and a statin by itself is hard to beat.”

In ACCORD, all patients received standard medical therapy for type 2 diabetes and cardiovascular disease risk, including statin therapy with simvastatin. Randomization assigned half the patients to also receive fenofibrate, at a target dosage of 160 mg/day. Although fenofibrate effectively cut triglyceride and HDL cholesterol levels, the incidence of all cardiovascular disease end points examined was similar between the treatment groups: 2.24% per year for the finofibrate group, 2.41% per year for the placebo group.

Dr. Deedwania has been a consultant to or received honoraria from AstraZeneca and Pfizer. Dr. Blumenthal had no relevant disclosures.

My Take

Design Worked Against Fenofibrate

I think of fenofibrate as a triglyceride drug, or possibly as an HDL drug. The median triglyceride level in the ACCORD lipid patients was 162 mg/dL, so it's not very surprising that the overall group did not benefit. It is interesting that patients with high triglycerides and low HDL had some suggestion of benefit, with a P value of .06. Another limitation of the study was that fenofibrate was used on top of a statin.

I wonder what would have happened if it had been used alone, in statin-intolerant patients. Another issue is whether the average 4.7 years of follow-up in the study was long enough. Because the drug works via relatively weak risk factors like triglycerides and HDL cholesterol, perhaps the follow-up was too brief.

The study results clearly show no benefit from fenofibrate for all high-risk patients with diabetes. The results particularly indicated no benefit in women. Further studies should be done to address these issues.

PAUL D. THOMPSON, M.D., is director of preventive cardiology at Hartford (Conn.) Hospital. He disclosed relationships with several pharmaceutical companies including Merck and Abbott, and with the American Board of Internal Medicine, the National Lipid Association, and Genomas. He has served on a data and safety monitoring board for Abbott, and has received other financial benefit from General Electric, Stryker, and Zimmer Holdings.

Vitals

Major Finding: In patients with type 2 diabetes and a high risk for cardiovascular disease, 2,765 treated with fenofibrate in addition to standard medical therapy had a 2.24%-per-year rate of major fatal or nonfatal cardiovascular events during an average 4.7 years of follow-up. The 2,753 patients randomized to placebo in addition to standard medical therapy had a 2.41%-per-year incidence rate of the end point. The difference in rates between the two groups was not statistically significant.

Data Source: ACCORD, a randomized, controlled lipid trial conducted at 77 sites in the United States and Canada during January 2001–July 2009.

Disclosures: Dr. Ginsberg has financial relationships with several pharmaceutical companies, including Merck and Abbott, which donated the simvastatin and fenofibrate/placebo but had no involvement in ACCORD. The trial was funded by the National Heart, Lung, and Blood Institute.

Atlanta — The failure to significantly reduce the cardiovascular event rate with fenofibrate treatment in a large trial of high-risk type 2 diabetes patients probably occurred because the study enrolled too many of the wrong types of patients to clearly show a benefit from this drug, several experts said.

Instead of focusing on patients with diabetes and dyslipidemia, an elevated serum level of triglycerides, and depressed HDL cholesterol, the Action to Control Cardiovascular Risk in Diabetes (ACCORD) lipid trial enrolled a representative sampling of 5,518 patients with diabetes and a range of triglyceride and HDL cholesterol levels, proving that fibrate treatment, on top of the moderate statin dosage, could not further help most of these patients.

The ACCORD investigators decided to enroll a wide spectrum of patients “to see if [fenofibrate] could apply generally. It's important that we found that fenofibrate on top of a statin will not benefit the majority” of patients with diabetes, Dr. Henry N. Ginsberg said at the annual meeting of the American College of Cardiology.

Existing cholesterol-treatment guidelines from the National Heart, Lung, and Blood Institute—the Adult Treatment Panel III—call for adding a fibrate drug to statin treatment when triglyceride levels are high and HDL cholesterol is low. “I think that's the role for this drug, in patients with the most significant dyslipidemia,” said Dr. Ginsberg, professor of medicine and director of the Irving Institute for Clinical and Translational Research at Columbia University, New York.

In ACCORD, 17% of enrolled patients fell into the subgroup with a plasma triglyceride level of at least 204 mg/dL and a plasma HDL cholesterol that was 34 mg/dL or less. Within this subgroup, fenofibrate treatment produced an improvement in the study's primary end point, the combination of major fatal or nonfatal cardiovascular events, that just missed statistical significance. Tr. Ginsberg said. Concurrently with his report at the meeting, the results were posted online (N. Engl. J. Med. 2010 Mar 14 [doi:10.1056/NEJMoa1001282

“They tested the drug on the wrong patients,” said Dr. Prakash C. Deedwania, a cardiologist at the University of California, San Francisco, in Fresno, Calif. The trial results could potentially have been positive if enrollment had been more focused, he said in an interview.

“The message is that the majority of patients with diabetes don't need a fibrate added,” said Dr. Roger S. Blumenthal, professor of medicine and director of preventive cardiology at Johns Hopkins University in Baltimore. “There was good reason to think the results might have been positive, but in ACCORD it was added to a statin, and a statin by itself is hard to beat.”

In ACCORD, all patients received standard medical therapy for type 2 diabetes and cardiovascular disease risk, including statin therapy with simvastatin. Randomization assigned half the patients to also receive fenofibrate, at a target dosage of 160 mg/day. Although fenofibrate effectively cut triglyceride and HDL cholesterol levels, the incidence of all cardiovascular disease end points examined was similar between the treatment groups: 2.24% per year for the finofibrate group, 2.41% per year for the placebo group.

Dr. Deedwania has been a consultant to or received honoraria from AstraZeneca and Pfizer. Dr. Blumenthal had no relevant disclosures.

My Take

Design Worked Against Fenofibrate

I think of fenofibrate as a triglyceride drug, or possibly as an HDL drug. The median triglyceride level in the ACCORD lipid patients was 162 mg/dL, so it's not very surprising that the overall group did not benefit. It is interesting that patients with high triglycerides and low HDL had some suggestion of benefit, with a P value of .06. Another limitation of the study was that fenofibrate was used on top of a statin.

I wonder what would have happened if it had been used alone, in statin-intolerant patients. Another issue is whether the average 4.7 years of follow-up in the study was long enough. Because the drug works via relatively weak risk factors like triglycerides and HDL cholesterol, perhaps the follow-up was too brief.

The study results clearly show no benefit from fenofibrate for all high-risk patients with diabetes. The results particularly indicated no benefit in women. Further studies should be done to address these issues.

PAUL D. THOMPSON, M.D., is director of preventive cardiology at Hartford (Conn.) Hospital. He disclosed relationships with several pharmaceutical companies including Merck and Abbott, and with the American Board of Internal Medicine, the National Lipid Association, and Genomas. He has served on a data and safety monitoring board for Abbott, and has received other financial benefit from General Electric, Stryker, and Zimmer Holdings.

Vitals

Major Finding: In patients with type 2 diabetes and a high risk for cardiovascular disease, 2,765 treated with fenofibrate in addition to standard medical therapy had a 2.24%-per-year rate of major fatal or nonfatal cardiovascular events during an average 4.7 years of follow-up. The 2,753 patients randomized to placebo in addition to standard medical therapy had a 2.41%-per-year incidence rate of the end point. The difference in rates between the two groups was not statistically significant.

Data Source: ACCORD, a randomized, controlled lipid trial conducted at 77 sites in the United States and Canada during January 2001–July 2009.

Disclosures: Dr. Ginsberg has financial relationships with several pharmaceutical companies, including Merck and Abbott, which donated the simvastatin and fenofibrate/placebo but had no involvement in ACCORD. The trial was funded by the National Heart, Lung, and Blood Institute.

Atlanta — The failure to significantly reduce the cardiovascular event rate with fenofibrate treatment in a large trial of high-risk type 2 diabetes patients probably occurred because the study enrolled too many of the wrong types of patients to clearly show a benefit from this drug, several experts said.

Instead of focusing on patients with diabetes and dyslipidemia, an elevated serum level of triglycerides, and depressed HDL cholesterol, the Action to Control Cardiovascular Risk in Diabetes (ACCORD) lipid trial enrolled a representative sampling of 5,518 patients with diabetes and a range of triglyceride and HDL cholesterol levels, proving that fibrate treatment, on top of the moderate statin dosage, could not further help most of these patients.

The ACCORD investigators decided to enroll a wide spectrum of patients “to see if [fenofibrate] could apply generally. It's important that we found that fenofibrate on top of a statin will not benefit the majority” of patients with diabetes, Dr. Henry N. Ginsberg said at the annual meeting of the American College of Cardiology.

Existing cholesterol-treatment guidelines from the National Heart, Lung, and Blood Institute—the Adult Treatment Panel III—call for adding a fibrate drug to statin treatment when triglyceride levels are high and HDL cholesterol is low. “I think that's the role for this drug, in patients with the most significant dyslipidemia,” said Dr. Ginsberg, professor of medicine and director of the Irving Institute for Clinical and Translational Research at Columbia University, New York.

In ACCORD, 17% of enrolled patients fell into the subgroup with a plasma triglyceride level of at least 204 mg/dL and a plasma HDL cholesterol that was 34 mg/dL or less. Within this subgroup, fenofibrate treatment produced an improvement in the study's primary end point, the combination of major fatal or nonfatal cardiovascular events, that just missed statistical significance. Tr. Ginsberg said. Concurrently with his report at the meeting, the results were posted online (N. Engl. J. Med. 2010 Mar 14 [doi:10.1056/NEJMoa1001282

“They tested the drug on the wrong patients,” said Dr. Prakash C. Deedwania, a cardiologist at the University of California, San Francisco, in Fresno, Calif. The trial results could potentially have been positive if enrollment had been more focused, he said in an interview.

“The message is that the majority of patients with diabetes don't need a fibrate added,” said Dr. Roger S. Blumenthal, professor of medicine and director of preventive cardiology at Johns Hopkins University in Baltimore. “There was good reason to think the results might have been positive, but in ACCORD it was added to a statin, and a statin by itself is hard to beat.”

In ACCORD, all patients received standard medical therapy for type 2 diabetes and cardiovascular disease risk, including statin therapy with simvastatin. Randomization assigned half the patients to also receive fenofibrate, at a target dosage of 160 mg/day. Although fenofibrate effectively cut triglyceride and HDL cholesterol levels, the incidence of all cardiovascular disease end points examined was similar between the treatment groups: 2.24% per year for the finofibrate group, 2.41% per year for the placebo group.

Dr. Deedwania has been a consultant to or received honoraria from AstraZeneca and Pfizer. Dr. Blumenthal had no relevant disclosures.

My Take

Design Worked Against Fenofibrate

I think of fenofibrate as a triglyceride drug, or possibly as an HDL drug. The median triglyceride level in the ACCORD lipid patients was 162 mg/dL, so it's not very surprising that the overall group did not benefit. It is interesting that patients with high triglycerides and low HDL had some suggestion of benefit, with a P value of .06. Another limitation of the study was that fenofibrate was used on top of a statin.

I wonder what would have happened if it had been used alone, in statin-intolerant patients. Another issue is whether the average 4.7 years of follow-up in the study was long enough. Because the drug works via relatively weak risk factors like triglycerides and HDL cholesterol, perhaps the follow-up was too brief.

The study results clearly show no benefit from fenofibrate for all high-risk patients with diabetes. The results particularly indicated no benefit in women. Further studies should be done to address these issues.

PAUL D. THOMPSON, M.D., is director of preventive cardiology at Hartford (Conn.) Hospital. He disclosed relationships with several pharmaceutical companies including Merck and Abbott, and with the American Board of Internal Medicine, the National Lipid Association, and Genomas. He has served on a data and safety monitoring board for Abbott, and has received other financial benefit from General Electric, Stryker, and Zimmer Holdings.

Vitals

ORLANDO – High blood levels of a brain natriuretic peptide were associated with poor cognitive function in a study of 950 community-dwelling, healthy, elderly adults.

“This is the first time this [association] has been shown,” Dr. Lori B. Daniels said at the annual scientific sessions of the American Heart Association.

Elevated levels of natriuretic peptide mark the presence of a variety of disease states, especially heart failure, Dr. Daniels said. She suggested several mechanisms that might link production of natriuretic peptide to poor cognitive function, including reduced cardiac output that drops oxygen or nutrient supplies to the brain, atrial fibrillation that creates microemboli, microcirculation deficits that harm both the heart and brain, and genetic predisposition.

Another issue is “which comes first, cardiovascular disease or poor cognitive performance,” said Dr. Daniels, a cardiologist at the University of California, San Diego.

Patients analyzed were enrolled in the Rancho Bernardo study in the early 1970s. Of the more than 5,000 community-dwelling adults in the study, 950 underwent a battery of cognitive function tests from 1997–1999 and had blood specimens drawn; they were the focus of the new analysis. The average age of the 950 participants was 77 years; 61% were women. Two-thirds were hypertensive, 4% were current smokers, 49% drank three or more alcoholic drinks per week, 41% were college graduates, 12% had diabetes, 6% had a history of stroke, and 20% a history of cardiovascular disease.

The researchers used three tests to evaluate cognitive function: The Mini-Mental State Exam (MMSE), which assessed cognitive features, including orientation, attention, calculation, and recall (a score of 24 or less indicated poor cognitive function); the Trail-Making Test B, which gauged executive function (a score of 132 seconds or more indicated poor function); and a category fluency test that asked participants to name as many animals as they could in 1 minute (a score of 12 or less indicated poor function).

MMSE results identified poor function in 7%, the trail-making test B identified poor function in 30%, and category fluency identified poor function in 15%.

Natriuretic peptide levels in the blood specimens were measured using a test that detects N-terminal pro-B-type natriuretic peptide (NT-proBNP). Natriuretic peptide measurements were considered low if the level was less than 450 pg/mL, and high if the level was 450 pg/mL or greater. Among the 950 participants, 79% had a low level and 21% had a high level.

People with high levels of NT-proBNP had significantly worse results in all three cognitive function tests, compared with those who had low levels.

The next step in exploring the link between natriuretic peptide and cognitive function should be a prospective study, Dr. Daniels said.

She received research support from Roche Diagnostics, which markets an NT-proBNP assay.

Tests used to evaluate cognitive function included the MMSE and the Trail-Making Test B.

Source DR. DANIELS

ORLANDO – High blood levels of a brain natriuretic peptide were associated with poor cognitive function in a study of 950 community-dwelling, healthy, elderly adults.

“This is the first time this [association] has been shown,” Dr. Lori B. Daniels said at the annual scientific sessions of the American Heart Association.

Elevated levels of natriuretic peptide mark the presence of a variety of disease states, especially heart failure, Dr. Daniels said. She suggested several mechanisms that might link production of natriuretic peptide to poor cognitive function, including reduced cardiac output that drops oxygen or nutrient supplies to the brain, atrial fibrillation that creates microemboli, microcirculation deficits that harm both the heart and brain, and genetic predisposition.

Another issue is “which comes first, cardiovascular disease or poor cognitive performance,” said Dr. Daniels, a cardiologist at the University of California, San Diego.

Patients analyzed were enrolled in the Rancho Bernardo study in the early 1970s. Of the more than 5,000 community-dwelling adults in the study, 950 underwent a battery of cognitive function tests from 1997–1999 and had blood specimens drawn; they were the focus of the new analysis. The average age of the 950 participants was 77 years; 61% were women. Two-thirds were hypertensive, 4% were current smokers, 49% drank three or more alcoholic drinks per week, 41% were college graduates, 12% had diabetes, 6% had a history of stroke, and 20% a history of cardiovascular disease.

The researchers used three tests to evaluate cognitive function: The Mini-Mental State Exam (MMSE), which assessed cognitive features, including orientation, attention, calculation, and recall (a score of 24 or less indicated poor cognitive function); the Trail-Making Test B, which gauged executive function (a score of 132 seconds or more indicated poor function); and a category fluency test that asked participants to name as many animals as they could in 1 minute (a score of 12 or less indicated poor function).

MMSE results identified poor function in 7%, the trail-making test B identified poor function in 30%, and category fluency identified poor function in 15%.

Natriuretic peptide levels in the blood specimens were measured using a test that detects N-terminal pro-B-type natriuretic peptide (NT-proBNP). Natriuretic peptide measurements were considered low if the level was less than 450 pg/mL, and high if the level was 450 pg/mL or greater. Among the 950 participants, 79% had a low level and 21% had a high level.

People with high levels of NT-proBNP had significantly worse results in all three cognitive function tests, compared with those who had low levels.

The next step in exploring the link between natriuretic peptide and cognitive function should be a prospective study, Dr. Daniels said.

She received research support from Roche Diagnostics, which markets an NT-proBNP assay.

Tests used to evaluate cognitive function included the MMSE and the Trail-Making Test B.

Source DR. DANIELS

ORLANDO – High blood levels of a brain natriuretic peptide were associated with poor cognitive function in a study of 950 community-dwelling, healthy, elderly adults.

“This is the first time this [association] has been shown,” Dr. Lori B. Daniels said at the annual scientific sessions of the American Heart Association.

Elevated levels of natriuretic peptide mark the presence of a variety of disease states, especially heart failure, Dr. Daniels said. She suggested several mechanisms that might link production of natriuretic peptide to poor cognitive function, including reduced cardiac output that drops oxygen or nutrient supplies to the brain, atrial fibrillation that creates microemboli, microcirculation deficits that harm both the heart and brain, and genetic predisposition.

Another issue is “which comes first, cardiovascular disease or poor cognitive performance,” said Dr. Daniels, a cardiologist at the University of California, San Diego.

Patients analyzed were enrolled in the Rancho Bernardo study in the early 1970s. Of the more than 5,000 community-dwelling adults in the study, 950 underwent a battery of cognitive function tests from 1997–1999 and had blood specimens drawn; they were the focus of the new analysis. The average age of the 950 participants was 77 years; 61% were women. Two-thirds were hypertensive, 4% were current smokers, 49% drank three or more alcoholic drinks per week, 41% were college graduates, 12% had diabetes, 6% had a history of stroke, and 20% a history of cardiovascular disease.

The researchers used three tests to evaluate cognitive function: The Mini-Mental State Exam (MMSE), which assessed cognitive features, including orientation, attention, calculation, and recall (a score of 24 or less indicated poor cognitive function); the Trail-Making Test B, which gauged executive function (a score of 132 seconds or more indicated poor function); and a category fluency test that asked participants to name as many animals as they could in 1 minute (a score of 12 or less indicated poor function).

MMSE results identified poor function in 7%, the trail-making test B identified poor function in 30%, and category fluency identified poor function in 15%.

Natriuretic peptide levels in the blood specimens were measured using a test that detects N-terminal pro-B-type natriuretic peptide (NT-proBNP). Natriuretic peptide measurements were considered low if the level was less than 450 pg/mL, and high if the level was 450 pg/mL or greater. Among the 950 participants, 79% had a low level and 21% had a high level.

People with high levels of NT-proBNP had significantly worse results in all three cognitive function tests, compared with those who had low levels.

The next step in exploring the link between natriuretic peptide and cognitive function should be a prospective study, Dr. Daniels said.

She received research support from Roche Diagnostics, which markets an NT-proBNP assay.

Tests used to evaluate cognitive function included the MMSE and the Trail-Making Test B.

Source DR. DANIELS

Major Finding: In patients with type 2 diabetes and a high risk for cardiovascular disease, 2,765 treated with fenofibrate in addition to standard medical therapy had a 2.24%/year rate of major fatal or nonfatal cardiovascular events during an average 4.7 years of follow-up. The 2,753 patients randomized to placebo in addition to standard medical therapy had a 2.41%/year incidence rate of the end point. The difference was not statistically significant.

Data Source: ACCORD, a randomized, controlled lipid trial conducted at 77 sites in the United States and Canada during January 2001–July 2009.

Disclosures: Dr. Ginsberg has financial relationships with several pharmaceutical companies, including Merck and Abbott, which donated the simvastatin and fenofibrate/placebo but had no involvement in ACCORD. Dr. Deedwania has had financial relationships with AstraZeneca and Pfizer. The trial was funded by the National Heart, Lung, and Blood Institute.

ATLANTA — The failure to significantly reduce the cardiovascular event rate with fenofibrate treatment in a large trial of high-risk type 2 diabetes patients probably occurred because the study enrolled too many of the wrong types of patients to clearly show a benefit from this drug, several experts said.

Instead of focusing on patients with diabetes and dyslipidemia, an elevated serum level of triglycerides, and depressed HDL cholesterol, the Action to Control Cardiovascular Risk in Diabetes (ACCORD) lipid trial enrolled a representative sampling of 5,518 patients with diabetes and a range of triglyceride and HDL cholesterol levels.

The ACCORD investigators decided to enroll a wide spectrum of patients “to see if [fenofibrate] could apply generally. It's important that we found that fenofibrate on top of a statin will not benefit the majority” of patients with diabetes, Dr. Henry N. Ginsberg said at the annual meeting of the American College of Cardiology.

Cholesterol-treatment guidelines from the National Heart, Lung, and Blood Institute—the Adult Treatment Panel III—call for adding a fibrate drug to a statin when triglyceride levels are high and HDL cholesterol is low. “I currently use fenofibrate in patients with a serum triglyceride level above 200 mg/dL or with HDL cholesterol in the 30s [mg/dL] or below. That's the group we identified as having potential benefit. A reasonable guess is that fewer than 5% of diabetes patients are on combination treatment with a statin and a fibrate. I think our results suggest that group could be expanded somewhat,” said Dr. Ginsberg, professor of medicine and director of the Irving Institute for Clinical and Translational Research at Columbia University, New York.

In the ACCORD lipid study, 17% of enrolled patients fell into the subgroup with a plasma triglyceride level of at least 204 mg/dL and a plasma HDL cholesterol that was 34 mg/dL or less. Within this subgroup, fenofibrate treatment produced an improvement in the study's primary end point, the combination of major fatal or nonfatal cardiovascular events, that just missed statistical significance. The suggestion of benefit in this subgroup was also notable because it concurred with results from three prior, large studies that also examined the efficacy of a fibrate in patients with diabetes, Dr. Ginsberg said. Concurrently with his report at the meeting, the results were published online (N. Engl. J. Med. 2010 March 14 [doi:10.1056/NEJMoa1001282]).

“They tested the drug on the wrong patients,” said Dr. Prakash C. Deedwania, a cardiologist at the University of California, San Francisco, in Fresno. The trial results could potentially have been positive if enrollment had been more focused, he said in an interview.

ACCORD randomized patients at 77 sites in the United States and Canada during January 2001–October 2005 and followed them for a mean of 4.7 years. All patients received standard medical therapy for type 2 diabetes and cardiovascular disease risk, including statin therapy with simvastatin. Half the patients were randomized to also receive fenofibrate, at a target dosage of 160 mg/day; the rest received placebo. Their average age was 62 years, 31% were women, and two-thirds were white. Although fenofibrate effectively cut triglyceride and HDL cholesterol levels, during follow-up the incidence of all cardiovascular disease end points examined was 2.24%/year with finofibrate and 2.41%/year with placebo.

The only significant subgroup interaction involved gender, Dr. Ginsberg said. Men were more likely to do better on fenofibrate, and women were more likely to have better outcomes on placebo.

My Take

Design Worked Against Fenofibrate

I think of fenofibrate as a triglyceride drug, or possibly as an HDL drug. The median triglyceride level in the ACCORD lipid patients was 162 mg/dL, so it's not very surprising that the overall group did not benefit. It is interesting that the subgroup analysis of patients with high triglycerides and low HDL had some suggestion of benefit, with a P value of .06.

Another limitation of the study was that fenofibrate was used on top of a statin. I wonder what would have happened if it had been used alone, in statin-intolerant patients. Another issue is whether the average 4.7 years of follow-up in the study was long enough. Because the drug works via relatively weak risk factors like triglycerides and HDL cholesterol, perhaps the follow-up was too brief.

The study results clearly show no benefit from fenofibrate for all high-risk patients with diabetes. The results particularly indicated no benefit in women.

Further studies should be done to address these issues.

Major Finding: In patients with type 2 diabetes and a high risk for cardiovascular disease, 2,765 treated with fenofibrate in addition to standard medical therapy had a 2.24%/year rate of major fatal or nonfatal cardiovascular events during an average 4.7 years of follow-up. The 2,753 patients randomized to placebo in addition to standard medical therapy had a 2.41%/year incidence rate of the end point. The difference was not statistically significant.

Data Source: ACCORD, a randomized, controlled lipid trial conducted at 77 sites in the United States and Canada during January 2001–July 2009.

Disclosures: Dr. Ginsberg has financial relationships with several pharmaceutical companies, including Merck and Abbott, which donated the simvastatin and fenofibrate/placebo but had no involvement in ACCORD. Dr. Deedwania has had financial relationships with AstraZeneca and Pfizer. The trial was funded by the National Heart, Lung, and Blood Institute.

ATLANTA — The failure to significantly reduce the cardiovascular event rate with fenofibrate treatment in a large trial of high-risk type 2 diabetes patients probably occurred because the study enrolled too many of the wrong types of patients to clearly show a benefit from this drug, several experts said.

Instead of focusing on patients with diabetes and dyslipidemia, an elevated serum level of triglycerides, and depressed HDL cholesterol, the Action to Control Cardiovascular Risk in Diabetes (ACCORD) lipid trial enrolled a representative sampling of 5,518 patients with diabetes and a range of triglyceride and HDL cholesterol levels.

The ACCORD investigators decided to enroll a wide spectrum of patients “to see if [fenofibrate] could apply generally. It's important that we found that fenofibrate on top of a statin will not benefit the majority” of patients with diabetes, Dr. Henry N. Ginsberg said at the annual meeting of the American College of Cardiology.

Cholesterol-treatment guidelines from the National Heart, Lung, and Blood Institute—the Adult Treatment Panel III—call for adding a fibrate drug to a statin when triglyceride levels are high and HDL cholesterol is low. “I currently use fenofibrate in patients with a serum triglyceride level above 200 mg/dL or with HDL cholesterol in the 30s [mg/dL] or below. That's the group we identified as having potential benefit. A reasonable guess is that fewer than 5% of diabetes patients are on combination treatment with a statin and a fibrate. I think our results suggest that group could be expanded somewhat,” said Dr. Ginsberg, professor of medicine and director of the Irving Institute for Clinical and Translational Research at Columbia University, New York.

In the ACCORD lipid study, 17% of enrolled patients fell into the subgroup with a plasma triglyceride level of at least 204 mg/dL and a plasma HDL cholesterol that was 34 mg/dL or less. Within this subgroup, fenofibrate treatment produced an improvement in the study's primary end point, the combination of major fatal or nonfatal cardiovascular events, that just missed statistical significance. The suggestion of benefit in this subgroup was also notable because it concurred with results from three prior, large studies that also examined the efficacy of a fibrate in patients with diabetes, Dr. Ginsberg said. Concurrently with his report at the meeting, the results were published online (N. Engl. J. Med. 2010 March 14 [doi:10.1056/NEJMoa1001282]).

“They tested the drug on the wrong patients,” said Dr. Prakash C. Deedwania, a cardiologist at the University of California, San Francisco, in Fresno. The trial results could potentially have been positive if enrollment had been more focused, he said in an interview.

ACCORD randomized patients at 77 sites in the United States and Canada during January 2001–October 2005 and followed them for a mean of 4.7 years. All patients received standard medical therapy for type 2 diabetes and cardiovascular disease risk, including statin therapy with simvastatin. Half the patients were randomized to also receive fenofibrate, at a target dosage of 160 mg/day; the rest received placebo. Their average age was 62 years, 31% were women, and two-thirds were white. Although fenofibrate effectively cut triglyceride and HDL cholesterol levels, during follow-up the incidence of all cardiovascular disease end points examined was 2.24%/year with finofibrate and 2.41%/year with placebo.

The only significant subgroup interaction involved gender, Dr. Ginsberg said. Men were more likely to do better on fenofibrate, and women were more likely to have better outcomes on placebo.

My Take

Design Worked Against Fenofibrate

I think of fenofibrate as a triglyceride drug, or possibly as an HDL drug. The median triglyceride level in the ACCORD lipid patients was 162 mg/dL, so it's not very surprising that the overall group did not benefit. It is interesting that the subgroup analysis of patients with high triglycerides and low HDL had some suggestion of benefit, with a P value of .06.

Another limitation of the study was that fenofibrate was used on top of a statin. I wonder what would have happened if it had been used alone, in statin-intolerant patients. Another issue is whether the average 4.7 years of follow-up in the study was long enough. Because the drug works via relatively weak risk factors like triglycerides and HDL cholesterol, perhaps the follow-up was too brief.

The study results clearly show no benefit from fenofibrate for all high-risk patients with diabetes. The results particularly indicated no benefit in women.

Further studies should be done to address these issues.

Major Finding: In patients with type 2 diabetes and a high risk for cardiovascular disease, 2,765 treated with fenofibrate in addition to standard medical therapy had a 2.24%/year rate of major fatal or nonfatal cardiovascular events during an average 4.7 years of follow-up. The 2,753 patients randomized to placebo in addition to standard medical therapy had a 2.41%/year incidence rate of the end point. The difference was not statistically significant.

Data Source: ACCORD, a randomized, controlled lipid trial conducted at 77 sites in the United States and Canada during January 2001–July 2009.

Disclosures: Dr. Ginsberg has financial relationships with several pharmaceutical companies, including Merck and Abbott, which donated the simvastatin and fenofibrate/placebo but had no involvement in ACCORD. Dr. Deedwania has had financial relationships with AstraZeneca and Pfizer. The trial was funded by the National Heart, Lung, and Blood Institute.

ATLANTA — The failure to significantly reduce the cardiovascular event rate with fenofibrate treatment in a large trial of high-risk type 2 diabetes patients probably occurred because the study enrolled too many of the wrong types of patients to clearly show a benefit from this drug, several experts said.

Instead of focusing on patients with diabetes and dyslipidemia, an elevated serum level of triglycerides, and depressed HDL cholesterol, the Action to Control Cardiovascular Risk in Diabetes (ACCORD) lipid trial enrolled a representative sampling of 5,518 patients with diabetes and a range of triglyceride and HDL cholesterol levels.

The ACCORD investigators decided to enroll a wide spectrum of patients “to see if [fenofibrate] could apply generally. It's important that we found that fenofibrate on top of a statin will not benefit the majority” of patients with diabetes, Dr. Henry N. Ginsberg said at the annual meeting of the American College of Cardiology.

Cholesterol-treatment guidelines from the National Heart, Lung, and Blood Institute—the Adult Treatment Panel III—call for adding a fibrate drug to a statin when triglyceride levels are high and HDL cholesterol is low. “I currently use fenofibrate in patients with a serum triglyceride level above 200 mg/dL or with HDL cholesterol in the 30s [mg/dL] or below. That's the group we identified as having potential benefit. A reasonable guess is that fewer than 5% of diabetes patients are on combination treatment with a statin and a fibrate. I think our results suggest that group could be expanded somewhat,” said Dr. Ginsberg, professor of medicine and director of the Irving Institute for Clinical and Translational Research at Columbia University, New York.

In the ACCORD lipid study, 17% of enrolled patients fell into the subgroup with a plasma triglyceride level of at least 204 mg/dL and a plasma HDL cholesterol that was 34 mg/dL or less. Within this subgroup, fenofibrate treatment produced an improvement in the study's primary end point, the combination of major fatal or nonfatal cardiovascular events, that just missed statistical significance. The suggestion of benefit in this subgroup was also notable because it concurred with results from three prior, large studies that also examined the efficacy of a fibrate in patients with diabetes, Dr. Ginsberg said. Concurrently with his report at the meeting, the results were published online (N. Engl. J. Med. 2010 March 14 [doi:10.1056/NEJMoa1001282]).

“They tested the drug on the wrong patients,” said Dr. Prakash C. Deedwania, a cardiologist at the University of California, San Francisco, in Fresno. The trial results could potentially have been positive if enrollment had been more focused, he said in an interview.

ACCORD randomized patients at 77 sites in the United States and Canada during January 2001–October 2005 and followed them for a mean of 4.7 years. All patients received standard medical therapy for type 2 diabetes and cardiovascular disease risk, including statin therapy with simvastatin. Half the patients were randomized to also receive fenofibrate, at a target dosage of 160 mg/day; the rest received placebo. Their average age was 62 years, 31% were women, and two-thirds were white. Although fenofibrate effectively cut triglyceride and HDL cholesterol levels, during follow-up the incidence of all cardiovascular disease end points examined was 2.24%/year with finofibrate and 2.41%/year with placebo.

The only significant subgroup interaction involved gender, Dr. Ginsberg said. Men were more likely to do better on fenofibrate, and women were more likely to have better outcomes on placebo.

My Take

Design Worked Against Fenofibrate

I think of fenofibrate as a triglyceride drug, or possibly as an HDL drug. The median triglyceride level in the ACCORD lipid patients was 162 mg/dL, so it's not very surprising that the overall group did not benefit. It is interesting that the subgroup analysis of patients with high triglycerides and low HDL had some suggestion of benefit, with a P value of .06.

Another limitation of the study was that fenofibrate was used on top of a statin. I wonder what would have happened if it had been used alone, in statin-intolerant patients. Another issue is whether the average 4.7 years of follow-up in the study was long enough. Because the drug works via relatively weak risk factors like triglycerides and HDL cholesterol, perhaps the follow-up was too brief.

The study results clearly show no benefit from fenofibrate for all high-risk patients with diabetes. The results particularly indicated no benefit in women.

Further studies should be done to address these issues.

SAN ANTONIO — The largest-ever, head-to-head comparison of stenting versus surgery for treating severe carotid artery stenosis showed a marked effect of age, with patients older than 70 having fewer adverse outcomes after carotid endarterectomy and patients younger than 70 having fewer complications following carotid angioplasty and stenting.

Although the highly anticipated results from the decade-long Carotid Revascularization Endarterectomy vs. Surgery Trial (CREST) seemed, in simplest terms, to show a dead heat between carotid stenting and surgery (see table), the results reported at the International Stroke Conference actually revealed statistically significant and clinically important differences between the two treatments.

The statistically significant interaction between patient age and outcome will likely play a major role when physicians and patients now decide which intervention to favor for a specific patient.

The CREST results also showed another significant difference between carotid surgery and stenting: Surgery led to a 1.2% increased absolute rate in the incidence of periprocedural myo-cardial infarctions, whereas stenting produced a 1.8% increase absolute rate of periprocedural strokes, a finding that will force patients and their physicians to ask themselves which complication they would rather risk.

The CREST findings also renewed concerns about the appropriateness of any invasive intervention, be it stenting or surgery, for patients with asymptomatic carotid stenosis. The findings raised questions about how CREST differed from another large comparison of stenting and surgery, the International Carotid Stenting Study (ICSS), the results from which also appeared online, coincidentally, on the same day as the CREST report (Lancet 2010 Feb. 26 [10.1016/S0140-6736(10)60239-5

CREST randomized 2,502 patients with either symptomatic carotid stenosis or asymptomatic, severe carotid stenosis (at least 60% blockage) at 108 sites in the United States and 9 in Canada. The patients' average age was 69 years, a third were women, and 47% were asymptomatic. The analysis showed no significant effect from either gender or symptom status on outcomes.

The age effect produced the sharpest distinction between stenting and surgery, and confirmed evidence that began emerging a few years ago that carotid stenting poses a special problem for elderly patients.

Some of the first suggestions of safety problems that can occur when stenting elderly patients came from the lead-in phase of CREST, a stage that involved nearly 1,600 patients who underwent carotid stenting in the early 2000s as operators in the study established their stenting expertise. The problem has been attributed to the increased difficulty and danger of placing stents and embolic protection devices through elderly patients' tortuous and atherosclerotic arteries.

Dr. Wayne M. Clark, who reported the CREST results, said that patients who underwent stenting at age 65 had a roughly 20% reduced risk for an adverse perioperative or long-term outcome compared with those who underwent surgery, whereas at age 60 the relative benefit from stenting was about 35% and at age 50, the rate of adverse outcomes after stenting was less than half the rate after endarterectomy.

The primary adverse-event measure used in CREST was the composite rate of any stroke, myocardial infarction, or death during the 30 days following treatment plus the rate of any ipsilateral stroke during long-term follow-up of up to 4 years. This rate was 7.2% for stenting and 6.8% for endarterectomy, with similar rates of ipsilateral strokes occurring from 31 days to 4 years (2.0% vs. 2.4%, respectively).

In contrast to younger patients, at age 75, the rate of adverse outcomes after stenting rose by about 35% compared with surgery; at age 80, the adverse-outcome rate was more than 50% higher with stenting than with surgery; and at age 85, the adverse event rate was roughly doubled by stenting in comparison with endarterectomy. Dr. Clark reported no disclosures relevant to the study.

Elsevier Global Medical News

SAN ANTONIO — The largest-ever, head-to-head comparison of stenting versus surgery for treating severe carotid artery stenosis showed a marked effect of age, with patients older than 70 having fewer adverse outcomes after carotid endarterectomy and patients younger than 70 having fewer complications following carotid angioplasty and stenting.

Although the highly anticipated results from the decade-long Carotid Revascularization Endarterectomy vs. Surgery Trial (CREST) seemed, in simplest terms, to show a dead heat between carotid stenting and surgery (see table), the results reported at the International Stroke Conference actually revealed statistically significant and clinically important differences between the two treatments.

The statistically significant interaction between patient age and outcome will likely play a major role when physicians and patients now decide which intervention to favor for a specific patient.

The CREST results also showed another significant difference between carotid surgery and stenting: Surgery led to a 1.2% increased absolute rate in the incidence of periprocedural myo-cardial infarctions, whereas stenting produced a 1.8% increase absolute rate of periprocedural strokes, a finding that will force patients and their physicians to ask themselves which complication they would rather risk.

The CREST findings also renewed concerns about the appropriateness of any invasive intervention, be it stenting or surgery, for patients with asymptomatic carotid stenosis. The findings raised questions about how CREST differed from another large comparison of stenting and surgery, the International Carotid Stenting Study (ICSS), the results from which also appeared online, coincidentally, on the same day as the CREST report (Lancet 2010 Feb. 26 [10.1016/S0140-6736(10)60239-5

CREST randomized 2,502 patients with either symptomatic carotid stenosis or asymptomatic, severe carotid stenosis (at least 60% blockage) at 108 sites in the United States and 9 in Canada. The patients' average age was 69 years, a third were women, and 47% were asymptomatic. The analysis showed no significant effect from either gender or symptom status on outcomes.

The age effect produced the sharpest distinction between stenting and surgery, and confirmed evidence that began emerging a few years ago that carotid stenting poses a special problem for elderly patients.

Some of the first suggestions of safety problems that can occur when stenting elderly patients came from the lead-in phase of CREST, a stage that involved nearly 1,600 patients who underwent carotid stenting in the early 2000s as operators in the study established their stenting expertise. The problem has been attributed to the increased difficulty and danger of placing stents and embolic protection devices through elderly patients' tortuous and atherosclerotic arteries.

Dr. Wayne M. Clark, who reported the CREST results, said that patients who underwent stenting at age 65 had a roughly 20% reduced risk for an adverse perioperative or long-term outcome compared with those who underwent surgery, whereas at age 60 the relative benefit from stenting was about 35% and at age 50, the rate of adverse outcomes after stenting was less than half the rate after endarterectomy.

The primary adverse-event measure used in CREST was the composite rate of any stroke, myocardial infarction, or death during the 30 days following treatment plus the rate of any ipsilateral stroke during long-term follow-up of up to 4 years. This rate was 7.2% for stenting and 6.8% for endarterectomy, with similar rates of ipsilateral strokes occurring from 31 days to 4 years (2.0% vs. 2.4%, respectively).

In contrast to younger patients, at age 75, the rate of adverse outcomes after stenting rose by about 35% compared with surgery; at age 80, the adverse-outcome rate was more than 50% higher with stenting than with surgery; and at age 85, the adverse event rate was roughly doubled by stenting in comparison with endarterectomy. Dr. Clark reported no disclosures relevant to the study.

Elsevier Global Medical News

SAN ANTONIO — The largest-ever, head-to-head comparison of stenting versus surgery for treating severe carotid artery stenosis showed a marked effect of age, with patients older than 70 having fewer adverse outcomes after carotid endarterectomy and patients younger than 70 having fewer complications following carotid angioplasty and stenting.

Although the highly anticipated results from the decade-long Carotid Revascularization Endarterectomy vs. Surgery Trial (CREST) seemed, in simplest terms, to show a dead heat between carotid stenting and surgery (see table), the results reported at the International Stroke Conference actually revealed statistically significant and clinically important differences between the two treatments.

The statistically significant interaction between patient age and outcome will likely play a major role when physicians and patients now decide which intervention to favor for a specific patient.

The CREST results also showed another significant difference between carotid surgery and stenting: Surgery led to a 1.2% increased absolute rate in the incidence of periprocedural myo-cardial infarctions, whereas stenting produced a 1.8% increase absolute rate of periprocedural strokes, a finding that will force patients and their physicians to ask themselves which complication they would rather risk.

The CREST findings also renewed concerns about the appropriateness of any invasive intervention, be it stenting or surgery, for patients with asymptomatic carotid stenosis. The findings raised questions about how CREST differed from another large comparison of stenting and surgery, the International Carotid Stenting Study (ICSS), the results from which also appeared online, coincidentally, on the same day as the CREST report (Lancet 2010 Feb. 26 [10.1016/S0140-6736(10)60239-5

CREST randomized 2,502 patients with either symptomatic carotid stenosis or asymptomatic, severe carotid stenosis (at least 60% blockage) at 108 sites in the United States and 9 in Canada. The patients' average age was 69 years, a third were women, and 47% were asymptomatic. The analysis showed no significant effect from either gender or symptom status on outcomes.

The age effect produced the sharpest distinction between stenting and surgery, and confirmed evidence that began emerging a few years ago that carotid stenting poses a special problem for elderly patients.

Some of the first suggestions of safety problems that can occur when stenting elderly patients came from the lead-in phase of CREST, a stage that involved nearly 1,600 patients who underwent carotid stenting in the early 2000s as operators in the study established their stenting expertise. The problem has been attributed to the increased difficulty and danger of placing stents and embolic protection devices through elderly patients' tortuous and atherosclerotic arteries.

Dr. Wayne M. Clark, who reported the CREST results, said that patients who underwent stenting at age 65 had a roughly 20% reduced risk for an adverse perioperative or long-term outcome compared with those who underwent surgery, whereas at age 60 the relative benefit from stenting was about 35% and at age 50, the rate of adverse outcomes after stenting was less than half the rate after endarterectomy.

The primary adverse-event measure used in CREST was the composite rate of any stroke, myocardial infarction, or death during the 30 days following treatment plus the rate of any ipsilateral stroke during long-term follow-up of up to 4 years. This rate was 7.2% for stenting and 6.8% for endarterectomy, with similar rates of ipsilateral strokes occurring from 31 days to 4 years (2.0% vs. 2.4%, respectively).

In contrast to younger patients, at age 75, the rate of adverse outcomes after stenting rose by about 35% compared with surgery; at age 80, the adverse-outcome rate was more than 50% higher with stenting than with surgery; and at age 85, the adverse event rate was roughly doubled by stenting in comparison with endarterectomy. Dr. Clark reported no disclosures relevant to the study.

Elsevier Global Medical News

PHILADELPHIA — The increased atherosclerotic disease that generally accompanies rheumatoid arthritis may not consistently involve carotid artery stenosis, according to two reports at the annual meeting of the American College of Rheumatology.

In one study with 195 rheumatoid arthritis patients and a nearly equal number of controls, carotid atherosclerosis was not clearly linked with coronary atherosclerosis in patients with RA, although the link existed in control patients, said Dr. Jon T. Giles, a rheumatologist at Johns Hopkins Medical Center, Baltimore.

Results from a second study, a meta-analysis of 22 prior reports in a total of 1,384 RA patients, showed that the mean extent of carotid intima-media thickness was “far less than expected.” Patients' average carotid stenosis corresponded to about a 10%–15% increase in cardiovascular risk, compared with similar people without RA, said Dr. Michael T. Nurmohamed, a rheumatologist at the Free University Medical Center in Amsterdam.

But the relationship between RA and carotid disease is more complex, according to other results reported by Dr. Nurmohamed. Preliminary results from measurement of carotid intima-media thickness in 100 patients with RA showed a mean thickness of 0.83 mm—“comparable” to the thickness in patients with type 2 diabetes—and enough stenosis to produce “a significantly increased cardiovascular risk,” Dr. Nurmohamed said.

“What is the best way to assess atherosclerosis in RA patients? For now, there is no recommendation on how to measure” subclinical cardiovascular disease, Dr. Giles said in an interview. No one can say whether measuring coronary disease is better or worse than measuring carotid atherosclerosis. If an RA patient “does not have carotid atherosclerosis, you can't be comfortable that nothing is going on,” he said.

He reported on 195 RA patients seen at the arthritis center at Johns Hopkins during October 2004–May 2008 and enrolled in the ESCAPE-RA (Evaluation of Subclinical Cardiovascular Disease and Predictors of Events in Rheumatoid Arthritis) study. Patients were 45-84 years old at enrollment and met the 1987 ACR classification criteria for RA. Enrollment excluded patients with clinically apparent cardiovascular disease. RA patients were matched by age, sex, and ethnicity with 198 controls who did not have RA and who had been enrolled in the Baltimore cohort of MESA (Multi-Ethnic Study of Atherosclerosis).

The results showed that carotid stenosis was linked to a high level of coronary calcium in both the RA patients and controls. But many RA patients without carotid atherosclerosis nonetheless had an increased prevalence of coronary calcium, an incongruous combination that was not seen in the controls.

“The absence of carotid atherosclerosis cannot rule out coronary atherosclerosis in RA patients in the same way that it does in the general population,” Dr. Giles said. The implication is that “using subclinical carotid atherosclerosis as a surrogate for coronary atherosclerosis in studies of RA patients may be inaccurate.”

The meta-analysis of 22 studies by Dr. Nurmohamed and his associates involved a total of 1,147 controls as well as more than 1,300 RA patients. In 17 of the studies, the carotid intima-media thickness was greater in the RA patients than in the controls. But the average intima-media thickness in the RA patients was 0.71 mm, an average of 0.09 mm larger than in the controls, a difference that corresponds to a modest 10%–15% higher rate of cardiovascular risk.

The low risk level may have occurred because the studies excluded people with cardiovascular disease or risk factors at baseline, a step that may have led to an underestimate of the difference in carotid intima-media thickness between the RA patients and controls.

The carotid data collected directly by Dr. Nurmohamed and his associates came from the CARRÉ (Cardiovascular Research and Rheumatoid Arthritis) study, a prospective study that tracked the incidence of cardiovascular events in patients with RA and in controls. A report from CARRÉ published in September showed the substantially higher level of cardiovascular disease events in 294 patients with RA (13%), compared with 258 controls (5%) (Ann. Rheum. Dis. 2009;68:1395-400).

Additional prospective, controlled studies are needed to further define the cardiovascular disease risk in RA patients, Dr. Nurmohamed said.

Disclosures: Neither Dr. Giles nor Dr. Nurmohamed had any disclosures relevant to their research to report.

'The absence of carotid atherosclerosis cannot rule out coronary atherosclerosis in RA patients.'

Source DR. NURMOHAMED

PHILADELPHIA — The increased atherosclerotic disease that generally accompanies rheumatoid arthritis may not consistently involve carotid artery stenosis, according to two reports at the annual meeting of the American College of Rheumatology.

In one study with 195 rheumatoid arthritis patients and a nearly equal number of controls, carotid atherosclerosis was not clearly linked with coronary atherosclerosis in patients with RA, although the link existed in control patients, said Dr. Jon T. Giles, a rheumatologist at Johns Hopkins Medical Center, Baltimore.

Results from a second study, a meta-analysis of 22 prior reports in a total of 1,384 RA patients, showed that the mean extent of carotid intima-media thickness was “far less than expected.” Patients' average carotid stenosis corresponded to about a 10%–15% increase in cardiovascular risk, compared with similar people without RA, said Dr. Michael T. Nurmohamed, a rheumatologist at the Free University Medical Center in Amsterdam.

But the relationship between RA and carotid disease is more complex, according to other results reported by Dr. Nurmohamed. Preliminary results from measurement of carotid intima-media thickness in 100 patients with RA showed a mean thickness of 0.83 mm—“comparable” to the thickness in patients with type 2 diabetes—and enough stenosis to produce “a significantly increased cardiovascular risk,” Dr. Nurmohamed said.

“What is the best way to assess atherosclerosis in RA patients? For now, there is no recommendation on how to measure” subclinical cardiovascular disease, Dr. Giles said in an interview. No one can say whether measuring coronary disease is better or worse than measuring carotid atherosclerosis. If an RA patient “does not have carotid atherosclerosis, you can't be comfortable that nothing is going on,” he said.

He reported on 195 RA patients seen at the arthritis center at Johns Hopkins during October 2004–May 2008 and enrolled in the ESCAPE-RA (Evaluation of Subclinical Cardiovascular Disease and Predictors of Events in Rheumatoid Arthritis) study. Patients were 45-84 years old at enrollment and met the 1987 ACR classification criteria for RA. Enrollment excluded patients with clinically apparent cardiovascular disease. RA patients were matched by age, sex, and ethnicity with 198 controls who did not have RA and who had been enrolled in the Baltimore cohort of MESA (Multi-Ethnic Study of Atherosclerosis).

The results showed that carotid stenosis was linked to a high level of coronary calcium in both the RA patients and controls. But many RA patients without carotid atherosclerosis nonetheless had an increased prevalence of coronary calcium, an incongruous combination that was not seen in the controls.

“The absence of carotid atherosclerosis cannot rule out coronary atherosclerosis in RA patients in the same way that it does in the general population,” Dr. Giles said. The implication is that “using subclinical carotid atherosclerosis as a surrogate for coronary atherosclerosis in studies of RA patients may be inaccurate.”

The meta-analysis of 22 studies by Dr. Nurmohamed and his associates involved a total of 1,147 controls as well as more than 1,300 RA patients. In 17 of the studies, the carotid intima-media thickness was greater in the RA patients than in the controls. But the average intima-media thickness in the RA patients was 0.71 mm, an average of 0.09 mm larger than in the controls, a difference that corresponds to a modest 10%–15% higher rate of cardiovascular risk.

The low risk level may have occurred because the studies excluded people with cardiovascular disease or risk factors at baseline, a step that may have led to an underestimate of the difference in carotid intima-media thickness between the RA patients and controls.

The carotid data collected directly by Dr. Nurmohamed and his associates came from the CARRÉ (Cardiovascular Research and Rheumatoid Arthritis) study, a prospective study that tracked the incidence of cardiovascular events in patients with RA and in controls. A report from CARRÉ published in September showed the substantially higher level of cardiovascular disease events in 294 patients with RA (13%), compared with 258 controls (5%) (Ann. Rheum. Dis. 2009;68:1395-400).

Additional prospective, controlled studies are needed to further define the cardiovascular disease risk in RA patients, Dr. Nurmohamed said.

Disclosures: Neither Dr. Giles nor Dr. Nurmohamed had any disclosures relevant to their research to report.

'The absence of carotid atherosclerosis cannot rule out coronary atherosclerosis in RA patients.'

Source DR. NURMOHAMED

PHILADELPHIA — The increased atherosclerotic disease that generally accompanies rheumatoid arthritis may not consistently involve carotid artery stenosis, according to two reports at the annual meeting of the American College of Rheumatology.

In one study with 195 rheumatoid arthritis patients and a nearly equal number of controls, carotid atherosclerosis was not clearly linked with coronary atherosclerosis in patients with RA, although the link existed in control patients, said Dr. Jon T. Giles, a rheumatologist at Johns Hopkins Medical Center, Baltimore.

Results from a second study, a meta-analysis of 22 prior reports in a total of 1,384 RA patients, showed that the mean extent of carotid intima-media thickness was “far less than expected.” Patients' average carotid stenosis corresponded to about a 10%–15% increase in cardiovascular risk, compared with similar people without RA, said Dr. Michael T. Nurmohamed, a rheumatologist at the Free University Medical Center in Amsterdam.

But the relationship between RA and carotid disease is more complex, according to other results reported by Dr. Nurmohamed. Preliminary results from measurement of carotid intima-media thickness in 100 patients with RA showed a mean thickness of 0.83 mm—“comparable” to the thickness in patients with type 2 diabetes—and enough stenosis to produce “a significantly increased cardiovascular risk,” Dr. Nurmohamed said.

“What is the best way to assess atherosclerosis in RA patients? For now, there is no recommendation on how to measure” subclinical cardiovascular disease, Dr. Giles said in an interview. No one can say whether measuring coronary disease is better or worse than measuring carotid atherosclerosis. If an RA patient “does not have carotid atherosclerosis, you can't be comfortable that nothing is going on,” he said.

He reported on 195 RA patients seen at the arthritis center at Johns Hopkins during October 2004–May 2008 and enrolled in the ESCAPE-RA (Evaluation of Subclinical Cardiovascular Disease and Predictors of Events in Rheumatoid Arthritis) study. Patients were 45-84 years old at enrollment and met the 1987 ACR classification criteria for RA. Enrollment excluded patients with clinically apparent cardiovascular disease. RA patients were matched by age, sex, and ethnicity with 198 controls who did not have RA and who had been enrolled in the Baltimore cohort of MESA (Multi-Ethnic Study of Atherosclerosis).

The results showed that carotid stenosis was linked to a high level of coronary calcium in both the RA patients and controls. But many RA patients without carotid atherosclerosis nonetheless had an increased prevalence of coronary calcium, an incongruous combination that was not seen in the controls.

“The absence of carotid atherosclerosis cannot rule out coronary atherosclerosis in RA patients in the same way that it does in the general population,” Dr. Giles said. The implication is that “using subclinical carotid atherosclerosis as a surrogate for coronary atherosclerosis in studies of RA patients may be inaccurate.”

The meta-analysis of 22 studies by Dr. Nurmohamed and his associates involved a total of 1,147 controls as well as more than 1,300 RA patients. In 17 of the studies, the carotid intima-media thickness was greater in the RA patients than in the controls. But the average intima-media thickness in the RA patients was 0.71 mm, an average of 0.09 mm larger than in the controls, a difference that corresponds to a modest 10%–15% higher rate of cardiovascular risk.

The low risk level may have occurred because the studies excluded people with cardiovascular disease or risk factors at baseline, a step that may have led to an underestimate of the difference in carotid intima-media thickness between the RA patients and controls.

The carotid data collected directly by Dr. Nurmohamed and his associates came from the CARRÉ (Cardiovascular Research and Rheumatoid Arthritis) study, a prospective study that tracked the incidence of cardiovascular events in patients with RA and in controls. A report from CARRÉ published in September showed the substantially higher level of cardiovascular disease events in 294 patients with RA (13%), compared with 258 controls (5%) (Ann. Rheum. Dis. 2009;68:1395-400).

Additional prospective, controlled studies are needed to further define the cardiovascular disease risk in RA patients, Dr. Nurmohamed said.

Disclosures: Neither Dr. Giles nor Dr. Nurmohamed had any disclosures relevant to their research to report.

'The absence of carotid atherosclerosis cannot rule out coronary atherosclerosis in RA patients.'

Source DR. NURMOHAMED

PHILADELPHIA — Women's risk factors for developing gout are similar to those in men, and baseline serum levels of uric acid may be the most powerful predictor, findings from the Framingham Heart Study show.

Women with no clinical indication of gout but a serum uric acid level of 8.0 mg/dL or greater at baseline had a subsequent 2.7% rate of gout during an average 28 years of follow-up—a 46-fold higher rate than women with a serum uric acid level of less than 5.0 mg/dL at baseline, Dr. Vidula Bhole said at the annual meeting of the American College of Rheumatology.

Serum uric acid likewise posed a powerful risk in men. Those with a level of 8.0 mg/dL or more at baseline had a 3.3% incidence rate during follow-up, 61-fold higher than men who entered the study with a serum level below 5.0 mg/dL, said Dr. Bhole, an epidemiologist at the University of British Columbia in Vancouver.

Even a baseline uric acid level of 5.0-5.9 mg/dL conferred a greater than threefold higher risk for developing gout in women and a greater than fourfold higher risk in men, compared with those whose level was under 5.0 mg/dL. (See box.)

Dr. Bhole and her associates used prospectively collected data from 4,427 individuals who had no history of gout at entry into the Framingham Heart Study in 1948.

The group included 2,476 women, with an average age of 47 years and an average serum uric acid level of 4.0 mg/dL. The group also included 1,967 men who entered at an average age of 46 years and a mean serum uric acid level of 5.1 mg/dL.

The subjects developed 304 cases of gout during an average 28 years of follow-up, with an incidence rate of 1.4 cases/1,000 person-years of follow-up in the women and 4.0 cases/1,000 person-years of follow-up in the men.

An analysis of gout incidence rates relative to baseline serum uric acid showed that, for any baseline level, women developed less gout than men. For example, among people who entered the study with a serum level of 7.0-7.9 mg/dL, the subsequent incidence was 1.3% in women and 1.8% in men.

A multivariate analysis identified several baseline factors linked to a significantly higher rate of incident gout in both genders: age, obesity, heavy alcohol use, hypertension, and diuretic use.

Disclosures: Dr. Bhole said she had no relevant financial relationships. Two of her study colleagues received grant support from and served as consultants to Takeda. One of Dr. Bhole's associates also serves on the advisory board for Savient, a company developing a uric acid–lowering drug.

A baseline uric acid level of 5.0-5.9 mg/dL increased gout risk threefold in women and fourfold in men.

Source DR. BHOLE

Source Elsevier Global Medical News

PHILADELPHIA — Women's risk factors for developing gout are similar to those in men, and baseline serum levels of uric acid may be the most powerful predictor, findings from the Framingham Heart Study show.

Women with no clinical indication of gout but a serum uric acid level of 8.0 mg/dL or greater at baseline had a subsequent 2.7% rate of gout during an average 28 years of follow-up—a 46-fold higher rate than women with a serum uric acid level of less than 5.0 mg/dL at baseline, Dr. Vidula Bhole said at the annual meeting of the American College of Rheumatology.

Serum uric acid likewise posed a powerful risk in men. Those with a level of 8.0 mg/dL or more at baseline had a 3.3% incidence rate during follow-up, 61-fold higher than men who entered the study with a serum level below 5.0 mg/dL, said Dr. Bhole, an epidemiologist at the University of British Columbia in Vancouver.

Even a baseline uric acid level of 5.0-5.9 mg/dL conferred a greater than threefold higher risk for developing gout in women and a greater than fourfold higher risk in men, compared with those whose level was under 5.0 mg/dL. (See box.)

Dr. Bhole and her associates used prospectively collected data from 4,427 individuals who had no history of gout at entry into the Framingham Heart Study in 1948.

The group included 2,476 women, with an average age of 47 years and an average serum uric acid level of 4.0 mg/dL. The group also included 1,967 men who entered at an average age of 46 years and a mean serum uric acid level of 5.1 mg/dL.

The subjects developed 304 cases of gout during an average 28 years of follow-up, with an incidence rate of 1.4 cases/1,000 person-years of follow-up in the women and 4.0 cases/1,000 person-years of follow-up in the men.

An analysis of gout incidence rates relative to baseline serum uric acid showed that, for any baseline level, women developed less gout than men. For example, among people who entered the study with a serum level of 7.0-7.9 mg/dL, the subsequent incidence was 1.3% in women and 1.8% in men.

A multivariate analysis identified several baseline factors linked to a significantly higher rate of incident gout in both genders: age, obesity, heavy alcohol use, hypertension, and diuretic use.

Disclosures: Dr. Bhole said she had no relevant financial relationships. Two of her study colleagues received grant support from and served as consultants to Takeda. One of Dr. Bhole's associates also serves on the advisory board for Savient, a company developing a uric acid–lowering drug.

A baseline uric acid level of 5.0-5.9 mg/dL increased gout risk threefold in women and fourfold in men.

Source DR. BHOLE

Source Elsevier Global Medical News

PHILADELPHIA — Women's risk factors for developing gout are similar to those in men, and baseline serum levels of uric acid may be the most powerful predictor, findings from the Framingham Heart Study show.

Women with no clinical indication of gout but a serum uric acid level of 8.0 mg/dL or greater at baseline had a subsequent 2.7% rate of gout during an average 28 years of follow-up—a 46-fold higher rate than women with a serum uric acid level of less than 5.0 mg/dL at baseline, Dr. Vidula Bhole said at the annual meeting of the American College of Rheumatology.

Serum uric acid likewise posed a powerful risk in men. Those with a level of 8.0 mg/dL or more at baseline had a 3.3% incidence rate during follow-up, 61-fold higher than men who entered the study with a serum level below 5.0 mg/dL, said Dr. Bhole, an epidemiologist at the University of British Columbia in Vancouver.

Even a baseline uric acid level of 5.0-5.9 mg/dL conferred a greater than threefold higher risk for developing gout in women and a greater than fourfold higher risk in men, compared with those whose level was under 5.0 mg/dL. (See box.)

Dr. Bhole and her associates used prospectively collected data from 4,427 individuals who had no history of gout at entry into the Framingham Heart Study in 1948.

The group included 2,476 women, with an average age of 47 years and an average serum uric acid level of 4.0 mg/dL. The group also included 1,967 men who entered at an average age of 46 years and a mean serum uric acid level of 5.1 mg/dL.

The subjects developed 304 cases of gout during an average 28 years of follow-up, with an incidence rate of 1.4 cases/1,000 person-years of follow-up in the women and 4.0 cases/1,000 person-years of follow-up in the men.

An analysis of gout incidence rates relative to baseline serum uric acid showed that, for any baseline level, women developed less gout than men. For example, among people who entered the study with a serum level of 7.0-7.9 mg/dL, the subsequent incidence was 1.3% in women and 1.8% in men.

A multivariate analysis identified several baseline factors linked to a significantly higher rate of incident gout in both genders: age, obesity, heavy alcohol use, hypertension, and diuretic use.

Disclosures: Dr. Bhole said she had no relevant financial relationships. Two of her study colleagues received grant support from and served as consultants to Takeda. One of Dr. Bhole's associates also serves on the advisory board for Savient, a company developing a uric acid–lowering drug.

A baseline uric acid level of 5.0-5.9 mg/dL increased gout risk threefold in women and fourfold in men.

Source DR. BHOLE

Source Elsevier Global Medical News

New criteria for diagnosing rheumatoid arthritis—introduced last October and expected in print later this year—should lead to earlier diagnoses, easier insurance coverage for treatment, and improved patient outcomes, agreed many rheumatologists. The new criteria are also likely to be adopted fairly quickly by most U.S. rheumatologists, experts added.

“It's a paradigm shift: Prevent disease or significantly abrogate it if rheumatoid arthritis is caught early. If you wait for the 1987 criteria to be fulfilled, patients will have established disease. Our goal is to identify and treat patients as early as possible,” said Dr. Clifton O. Bingham III, associate director of the Johns Hopkins Arthritis Center in Baltimore and a member of the panel that came up with the new RA criteria.

A panel of 22 rheumatologists formed by the American College of Rheumatology (ACR) and the European League Against Rheumatism (EULAR) worked for 3 years to devise new RA diagnostic criteria to replace the existing RA classification criteria that were released by the ACR in 1987 (Arthritis Rheum. 1988;31:315-24). An initial public report on the new diagnostic criteria occurred last October at the annual meeting of the ACR in Philadelphia. (See box.) A peer-reviewed, written version of the criteria, as well as reports on the multistage process that led to their creation, should be published later this year, said Dr. Gillian A. Hawker, chief of medicine at Women's College Hospital in Toronto and a project leader.

Although the U.S. rheumatologists who were interviewed for this article weren't completely uniform in their expectations of how the new criteria will likely affect practice, most agreed on several broad consequences, starting with the way by which the new criteria will change the timing of RA diagnosis and treatment.

“This will lead to earlier, definitive assessment and treatment of patients with RA. A major weakness of the previous classification criteria is that they included a bad outcome [and] erosions, and required more extensive disease. We want to treat patients before erosions occur,” said Dr. Eric L. Matteson, professor and chairman of rheumatology at the Mayo Clinic in Rochester, Minn.

“These criteria should help clinicians diagnose patients at an earlier stage, and possibly lead to earlier treatment as well, thereby improving outcomes. They may help qualify patients for therapy at an earlier stage of their disease,” said Dr. Arthur F. Kavanaugh, professor of medicine at the University of California, San Diego.

A major way in which the new criteria enable earlier diagnosis is by setting a lower threshold for the number of involved joints, noted Dr. Michael E. Weinblatt, professor of medicine at Harvard Medical School and associate director of the center for arthritis and joint diseases at Brigham and Women's Hospital, both in Boston. “A lot of times patients don't seek care or don't get referred because only a couple of joints are involved.” The new criteria make it clear that “a couple of joints could be RA.”

These criteria “will allow for earlier diagnosis, but there is a great deal of clinical judgment [involved], and I've certainly diagnosed RA in many patients who did not fulfill the [1987] ACR criteria,” said Dr. Daniel Furst, professor of medicine at the University of California, Los Angeles. “What this does is codify and validate the fact that we make diagnoses earlier” than the old classification criteria allowed. Now rheumatologists and even primary care physicians “will feel comfortable making an earlier diagnosis,” he said.

Earlier diagnosis was a major goal of the panel that came up with the new diagnostic criteria, based on an “increasing concern that we were missing patients with aggressive, erosive disease,” said Dr. Philip J. Mease, a member of the panel and a rheumatologist at Swedish Hospital Medical Center in Seattle. “I'm not sure that the criteria will identify more patients, but they will more precisely identify patients who will have an aggressive course. The hope is that if you start treatment [of these patients] sooner, you may prevent disease progression.”

Experts were split on how confident they were that earlier diagnosis and treatment would lead to better outcomes, although that's what they generally expect.

“We hope it leads to better outcomes. That's the underlying assumption,” Dr. Furst said.

“Earlier treatment of RA means better outcomes, including less irreversible damage,” Dr. Matteson said.

Dr. Mease said that some evidence already exists for improved outcomes from early treatment. He cited results from studies such as TICORA (Tight Control for Rheumatoid Arthritis) (Lancet 2004;364:263-9), BeSt (Behandel-Strategieën) (Arthritis Rheum. 2005:52:3381-90), and CAMERA (Computer-Assisted Management of Early Rheumatoid Arthritis) (Ann. Rheum. Dis. 2007;66:1443-9). Results from all three studies showed that aggressive treatment early in RA led to better outcomes, with lower joint counts, better function, and more significant inhibition of radiologic damage, Dr. Mease said in an interview.

The new criteria should also ease insurance-coverage problems, some experts said. Currently, some insurers ask whether the patient fulfills the 1987 criteria for RA, “and if you answer truthfully, some patients [with early RA] may not fulfill the criteria.” The 2009 criteria “may allow earlier access to medications. This will make it easier to document RA,” Dr. Bingham said.

The number of involved joints has been a diagnostic feature that has held up insurance coverage for some patients, with insurers insisting that patients meet the 1987 standard of at least six involved joints, Dr. Weinblatt said. Dr. Furst and Dr. Matteson also cited experiences with denied insurance coverage, something they anticipate will become a thing of the past, more or less, with the new criteria.

“The first thing insurers ask in prior authorization forms is whether a patient meets the criteria for RA,” Dr. Weinblatt noted in commenting on the new criteria last October. This slowed the use of disease-modifying antirheumatic drugs in some patients. The new criteria will eliminate this barrier in many cases, he said.

Although all the experts who were interviewed agreed that the new criteria accurately reflected current thinking on what constitutes RA, a few envisioned certain situations that could cause problems. One concern involved mixing apples and oranges: Could results from RA patients in prior treatment studies always be appropriately applied to patients whose disease is defined by the new criteria? Dr. Furst asked. Similarly, he wondered whether drug toxicity profiles that were worked out in prior cohorts of RA patients would match the toxicities faced by newly defined RA patients.

Dr. Mease said he worried about a group of patients who are sick but fall short of the diagnostic criteria. These are the patients who present with fewer than 10 involved medium or large joints, low titers of rheumatoid factor and anti–citrullinated protein antibody, and a very high level of C-reactive protein, a constellation showing that the patient “clearly has an inflammatory process,” yet one that would tally a diagnostic score of 4-5 points (depending on symptom duration), which is less than the 6 points needed for a definitive RA diagnosis. Despite such concerns, Dr. Furst noted that the 2009 criteria have higher specificity and sensitivity than did the 1987 criteria. Also, new serologic and genomic tests that will likely emerge in the next several years will further refine diagnoses and will be incorporated into the scoring formula, Dr. Mease said.

“Most rheumatologists accept the concept of a need for early intervention, which these criteria speak to,” Dr. Matteson said. They will be accepted and used by most U.S. rheumatologists, he predicted.

“I think it will improve the outcome of our patients, and so it's a very good thing,” Dr. Furst said.

Disclosures: Dr. Bingham and Dr. Mease both participated on the panel that developed the new diagnostic criteria; this work was sponsored by the ACR and EULAR. The other rheumatologists cited had no role in developing the criteria. The rheumatologists who were interviewed said that because the new criteria do not deal directly with treatment, they did not have financial disclosures relevant to the topic.

Revised criteria will change the timing of RA diagnosis by setting a lower threshold for the number of involved joints.

Source ©J. Cavallini/Custom Medical Stock Photo

The New Diagnostic Criteria in Brief

Patients are definitively diagnosed with RA if they score 6 or more points according to the following criteria:

Joint involvement

▸ 1 medium-large joint (0 points)

▸ 2-10 medium-large joints (1 point)

▸ 1-3 small joints (2 points)

▸ 4-10 small joints (3 points)

▸ More than 10 small joints (5 points)

Serology

▸ Not positive for either rheumatoid factor or anti–citrullinated protein antibody (0 points)

▸ At least one of these two tests are positive at low titer, defined as more than the upper limit of normal but not higher than three times the upper limit of normal (2 points)

▸ At least one test is positive at high titer, defined as more than three times the upper limit of normal (3 points)

Duration of synovitis

▸ Lasting fewer than 6 weeks (0 points)

▸ Lasting 6 weeks or longer (1 point)

Acute phase reactants

▸ Neither C-reactive protein nor erythrocyte sedimentation rate is abnormal (0 points)

▸ Abnormal CRP or abnormal ESR (1 point)

Note: Patients receive the highest point level they fulfill within each domain. For example, a patient with five small joints involved and four large joints involved scores 3 points.

Note: Based on a presentation by Dr. Hawker at the annual meeting of the American College of Rheumatology, October 2009.

New criteria for diagnosing rheumatoid arthritis—introduced last October and expected in print later this year—should lead to earlier diagnoses, easier insurance coverage for treatment, and improved patient outcomes, agreed many rheumatologists. The new criteria are also likely to be adopted fairly quickly by most U.S. rheumatologists, experts added.

“It's a paradigm shift: Prevent disease or significantly abrogate it if rheumatoid arthritis is caught early. If you wait for the 1987 criteria to be fulfilled, patients will have established disease. Our goal is to identify and treat patients as early as possible,” said Dr. Clifton O. Bingham III, associate director of the Johns Hopkins Arthritis Center in Baltimore and a member of the panel that came up with the new RA criteria.

A panel of 22 rheumatologists formed by the American College of Rheumatology (ACR) and the European League Against Rheumatism (EULAR) worked for 3 years to devise new RA diagnostic criteria to replace the existing RA classification criteria that were released by the ACR in 1987 (Arthritis Rheum. 1988;31:315-24). An initial public report on the new diagnostic criteria occurred last October at the annual meeting of the ACR in Philadelphia. (See box.) A peer-reviewed, written version of the criteria, as well as reports on the multistage process that led to their creation, should be published later this year, said Dr. Gillian A. Hawker, chief of medicine at Women's College Hospital in Toronto and a project leader.

Although the U.S. rheumatologists who were interviewed for this article weren't completely uniform in their expectations of how the new criteria will likely affect practice, most agreed on several broad consequences, starting with the way by which the new criteria will change the timing of RA diagnosis and treatment.

“This will lead to earlier, definitive assessment and treatment of patients with RA. A major weakness of the previous classification criteria is that they included a bad outcome [and] erosions, and required more extensive disease. We want to treat patients before erosions occur,” said Dr. Eric L. Matteson, professor and chairman of rheumatology at the Mayo Clinic in Rochester, Minn.

“These criteria should help clinicians diagnose patients at an earlier stage, and possibly lead to earlier treatment as well, thereby improving outcomes. They may help qualify patients for therapy at an earlier stage of their disease,” said Dr. Arthur F. Kavanaugh, professor of medicine at the University of California, San Diego.

A major way in which the new criteria enable earlier diagnosis is by setting a lower threshold for the number of involved joints, noted Dr. Michael E. Weinblatt, professor of medicine at Harvard Medical School and associate director of the center for arthritis and joint diseases at Brigham and Women's Hospital, both in Boston. “A lot of times patients don't seek care or don't get referred because only a couple of joints are involved.” The new criteria make it clear that “a couple of joints could be RA.”

These criteria “will allow for earlier diagnosis, but there is a great deal of clinical judgment [involved], and I've certainly diagnosed RA in many patients who did not fulfill the [1987] ACR criteria,” said Dr. Daniel Furst, professor of medicine at the University of California, Los Angeles. “What this does is codify and validate the fact that we make diagnoses earlier” than the old classification criteria allowed. Now rheumatologists and even primary care physicians “will feel comfortable making an earlier diagnosis,” he said.

Earlier diagnosis was a major goal of the panel that came up with the new diagnostic criteria, based on an “increasing concern that we were missing patients with aggressive, erosive disease,” said Dr. Philip J. Mease, a member of the panel and a rheumatologist at Swedish Hospital Medical Center in Seattle. “I'm not sure that the criteria will identify more patients, but they will more precisely identify patients who will have an aggressive course. The hope is that if you start treatment [of these patients] sooner, you may prevent disease progression.”

Experts were split on how confident they were that earlier diagnosis and treatment would lead to better outcomes, although that's what they generally expect.

“We hope it leads to better outcomes. That's the underlying assumption,” Dr. Furst said.

“Earlier treatment of RA means better outcomes, including less irreversible damage,” Dr. Matteson said.

Dr. Mease said that some evidence already exists for improved outcomes from early treatment. He cited results from studies such as TICORA (Tight Control for Rheumatoid Arthritis) (Lancet 2004;364:263-9), BeSt (Behandel-Strategieën) (Arthritis Rheum. 2005:52:3381-90), and CAMERA (Computer-Assisted Management of Early Rheumatoid Arthritis) (Ann. Rheum. Dis. 2007;66:1443-9). Results from all three studies showed that aggressive treatment early in RA led to better outcomes, with lower joint counts, better function, and more significant inhibition of radiologic damage, Dr. Mease said in an interview.

The new criteria should also ease insurance-coverage problems, some experts said. Currently, some insurers ask whether the patient fulfills the 1987 criteria for RA, “and if you answer truthfully, some patients [with early RA] may not fulfill the criteria.” The 2009 criteria “may allow earlier access to medications. This will make it easier to document RA,” Dr. Bingham said.

The number of involved joints has been a diagnostic feature that has held up insurance coverage for some patients, with insurers insisting that patients meet the 1987 standard of at least six involved joints, Dr. Weinblatt said. Dr. Furst and Dr. Matteson also cited experiences with denied insurance coverage, something they anticipate will become a thing of the past, more or less, with the new criteria.

“The first thing insurers ask in prior authorization forms is whether a patient meets the criteria for RA,” Dr. Weinblatt noted in commenting on the new criteria last October. This slowed the use of disease-modifying antirheumatic drugs in some patients. The new criteria will eliminate this barrier in many cases, he said.

Although all the experts who were interviewed agreed that the new criteria accurately reflected current thinking on what constitutes RA, a few envisioned certain situations that could cause problems. One concern involved mixing apples and oranges: Could results from RA patients in prior treatment studies always be appropriately applied to patients whose disease is defined by the new criteria? Dr. Furst asked. Similarly, he wondered whether drug toxicity profiles that were worked out in prior cohorts of RA patients would match the toxicities faced by newly defined RA patients.

Dr. Mease said he worried about a group of patients who are sick but fall short of the diagnostic criteria. These are the patients who present with fewer than 10 involved medium or large joints, low titers of rheumatoid factor and anti–citrullinated protein antibody, and a very high level of C-reactive protein, a constellation showing that the patient “clearly has an inflammatory process,” yet one that would tally a diagnostic score of 4-5 points (depending on symptom duration), which is less than the 6 points needed for a definitive RA diagnosis. Despite such concerns, Dr. Furst noted that the 2009 criteria have higher specificity and sensitivity than did the 1987 criteria. Also, new serologic and genomic tests that will likely emerge in the next several years will further refine diagnoses and will be incorporated into the scoring formula, Dr. Mease said.

“Most rheumatologists accept the concept of a need for early intervention, which these criteria speak to,” Dr. Matteson said. They will be accepted and used by most U.S. rheumatologists, he predicted.

“I think it will improve the outcome of our patients, and so it's a very good thing,” Dr. Furst said.

Disclosures: Dr. Bingham and Dr. Mease both participated on the panel that developed the new diagnostic criteria; this work was sponsored by the ACR and EULAR. The other rheumatologists cited had no role in developing the criteria. The rheumatologists who were interviewed said that because the new criteria do not deal directly with treatment, they did not have financial disclosures relevant to the topic.

Revised criteria will change the timing of RA diagnosis by setting a lower threshold for the number of involved joints.

Source ©J. Cavallini/Custom Medical Stock Photo

The New Diagnostic Criteria in Brief

Patients are definitively diagnosed with RA if they score 6 or more points according to the following criteria:

Joint involvement

▸ 1 medium-large joint (0 points)

▸ 2-10 medium-large joints (1 point)

▸ 1-3 small joints (2 points)

▸ 4-10 small joints (3 points)

▸ More than 10 small joints (5 points)

Serology

▸ Not positive for either rheumatoid factor or anti–citrullinated protein antibody (0 points)

▸ At least one of these two tests are positive at low titer, defined as more than the upper limit of normal but not higher than three times the upper limit of normal (2 points)

▸ At least one test is positive at high titer, defined as more than three times the upper limit of normal (3 points)

Duration of synovitis

▸ Lasting fewer than 6 weeks (0 points)

▸ Lasting 6 weeks or longer (1 point)

Acute phase reactants

▸ Neither C-reactive protein nor erythrocyte sedimentation rate is abnormal (0 points)

▸ Abnormal CRP or abnormal ESR (1 point)

Note: Patients receive the highest point level they fulfill within each domain. For example, a patient with five small joints involved and four large joints involved scores 3 points.

Note: Based on a presentation by Dr. Hawker at the annual meeting of the American College of Rheumatology, October 2009.

New criteria for diagnosing rheumatoid arthritis—introduced last October and expected in print later this year—should lead to earlier diagnoses, easier insurance coverage for treatment, and improved patient outcomes, agreed many rheumatologists. The new criteria are also likely to be adopted fairly quickly by most U.S. rheumatologists, experts added.

“It's a paradigm shift: Prevent disease or significantly abrogate it if rheumatoid arthritis is caught early. If you wait for the 1987 criteria to be fulfilled, patients will have established disease. Our goal is to identify and treat patients as early as possible,” said Dr. Clifton O. Bingham III, associate director of the Johns Hopkins Arthritis Center in Baltimore and a member of the panel that came up with the new RA criteria.

A panel of 22 rheumatologists formed by the American College of Rheumatology (ACR) and the European League Against Rheumatism (EULAR) worked for 3 years to devise new RA diagnostic criteria to replace the existing RA classification criteria that were released by the ACR in 1987 (Arthritis Rheum. 1988;31:315-24). An initial public report on the new diagnostic criteria occurred last October at the annual meeting of the ACR in Philadelphia. (See box.) A peer-reviewed, written version of the criteria, as well as reports on the multistage process that led to their creation, should be published later this year, said Dr. Gillian A. Hawker, chief of medicine at Women's College Hospital in Toronto and a project leader.

Although the U.S. rheumatologists who were interviewed for this article weren't completely uniform in their expectations of how the new criteria will likely affect practice, most agreed on several broad consequences, starting with the way by which the new criteria will change the timing of RA diagnosis and treatment.

“This will lead to earlier, definitive assessment and treatment of patients with RA. A major weakness of the previous classification criteria is that they included a bad outcome [and] erosions, and required more extensive disease. We want to treat patients before erosions occur,” said Dr. Eric L. Matteson, professor and chairman of rheumatology at the Mayo Clinic in Rochester, Minn.

“These criteria should help clinicians diagnose patients at an earlier stage, and possibly lead to earlier treatment as well, thereby improving outcomes. They may help qualify patients for therapy at an earlier stage of their disease,” said Dr. Arthur F. Kavanaugh, professor of medicine at the University of California, San Diego.

A major way in which the new criteria enable earlier diagnosis is by setting a lower threshold for the number of involved joints, noted Dr. Michael E. Weinblatt, professor of medicine at Harvard Medical School and associate director of the center for arthritis and joint diseases at Brigham and Women's Hospital, both in Boston. “A lot of times patients don't seek care or don't get referred because only a couple of joints are involved.” The new criteria make it clear that “a couple of joints could be RA.”

These criteria “will allow for earlier diagnosis, but there is a great deal of clinical judgment [involved], and I've certainly diagnosed RA in many patients who did not fulfill the [1987] ACR criteria,” said Dr. Daniel Furst, professor of medicine at the University of California, Los Angeles. “What this does is codify and validate the fact that we make diagnoses earlier” than the old classification criteria allowed. Now rheumatologists and even primary care physicians “will feel comfortable making an earlier diagnosis,” he said.

Earlier diagnosis was a major goal of the panel that came up with the new diagnostic criteria, based on an “increasing concern that we were missing patients with aggressive, erosive disease,” said Dr. Philip J. Mease, a member of the panel and a rheumatologist at Swedish Hospital Medical Center in Seattle. “I'm not sure that the criteria will identify more patients, but they will more precisely identify patients who will have an aggressive course. The hope is that if you start treatment [of these patients] sooner, you may prevent disease progression.”

Experts were split on how confident they were that earlier diagnosis and treatment would lead to better outcomes, although that's what they generally expect.

“We hope it leads to better outcomes. That's the underlying assumption,” Dr. Furst said.

“Earlier treatment of RA means better outcomes, including less irreversible damage,” Dr. Matteson said.

Dr. Mease said that some evidence already exists for improved outcomes from early treatment. He cited results from studies such as TICORA (Tight Control for Rheumatoid Arthritis) (Lancet 2004;364:263-9), BeSt (Behandel-Strategieën) (Arthritis Rheum. 2005:52:3381-90), and CAMERA (Computer-Assisted Management of Early Rheumatoid Arthritis) (Ann. Rheum. Dis. 2007;66:1443-9). Results from all three studies showed that aggressive treatment early in RA led to better outcomes, with lower joint counts, better function, and more significant inhibition of radiologic damage, Dr. Mease said in an interview.

The new criteria should also ease insurance-coverage problems, some experts said. Currently, some insurers ask whether the patient fulfills the 1987 criteria for RA, “and if you answer truthfully, some patients [with early RA] may not fulfill the criteria.” The 2009 criteria “may allow earlier access to medications. This will make it easier to document RA,” Dr. Bingham said.

The number of involved joints has been a diagnostic feature that has held up insurance coverage for some patients, with insurers insisting that patients meet the 1987 standard of at least six involved joints, Dr. Weinblatt said. Dr. Furst and Dr. Matteson also cited experiences with denied insurance coverage, something they anticipate will become a thing of the past, more or less, with the new criteria.

“The first thing insurers ask in prior authorization forms is whether a patient meets the criteria for RA,” Dr. Weinblatt noted in commenting on the new criteria last October. This slowed the use of disease-modifying antirheumatic drugs in some patients. The new criteria will eliminate this barrier in many cases, he said.

Although all the experts who were interviewed agreed that the new criteria accurately reflected current thinking on what constitutes RA, a few envisioned certain situations that could cause problems. One concern involved mixing apples and oranges: Could results from RA patients in prior treatment studies always be appropriately applied to patients whose disease is defined by the new criteria? Dr. Furst asked. Similarly, he wondered whether drug toxicity profiles that were worked out in prior cohorts of RA patients would match the toxicities faced by newly defined RA patients.

Dr. Mease said he worried about a group of patients who are sick but fall short of the diagnostic criteria. These are the patients who present with fewer than 10 involved medium or large joints, low titers of rheumatoid factor and anti–citrullinated protein antibody, and a very high level of C-reactive protein, a constellation showing that the patient “clearly has an inflammatory process,” yet one that would tally a diagnostic score of 4-5 points (depending on symptom duration), which is less than the 6 points needed for a definitive RA diagnosis. Despite such concerns, Dr. Furst noted that the 2009 criteria have higher specificity and sensitivity than did the 1987 criteria. Also, new serologic and genomic tests that will likely emerge in the next several years will further refine diagnoses and will be incorporated into the scoring formula, Dr. Mease said.

“Most rheumatologists accept the concept of a need for early intervention, which these criteria speak to,” Dr. Matteson said. They will be accepted and used by most U.S. rheumatologists, he predicted.

“I think it will improve the outcome of our patients, and so it's a very good thing,” Dr. Furst said.

Disclosures: Dr. Bingham and Dr. Mease both participated on the panel that developed the new diagnostic criteria; this work was sponsored by the ACR and EULAR. The other rheumatologists cited had no role in developing the criteria. The rheumatologists who were interviewed said that because the new criteria do not deal directly with treatment, they did not have financial disclosures relevant to the topic.

Revised criteria will change the timing of RA diagnosis by setting a lower threshold for the number of involved joints.

Source ©J. Cavallini/Custom Medical Stock Photo

The New Diagnostic Criteria in Brief

Patients are definitively diagnosed with RA if they score 6 or more points according to the following criteria:

Joint involvement

▸ 1 medium-large joint (0 points)

▸ 2-10 medium-large joints (1 point)

▸ 1-3 small joints (2 points)

▸ 4-10 small joints (3 points)

▸ More than 10 small joints (5 points)

Serology

▸ Not positive for either rheumatoid factor or anti–citrullinated protein antibody (0 points)

▸ At least one of these two tests are positive at low titer, defined as more than the upper limit of normal but not higher than three times the upper limit of normal (2 points)

▸ At least one test is positive at high titer, defined as more than three times the upper limit of normal (3 points)

Duration of synovitis

▸ Lasting fewer than 6 weeks (0 points)

▸ Lasting 6 weeks or longer (1 point)

Acute phase reactants

▸ Neither C-reactive protein nor erythrocyte sedimentation rate is abnormal (0 points)

▸ Abnormal CRP or abnormal ESR (1 point)

Note: Patients receive the highest point level they fulfill within each domain. For example, a patient with five small joints involved and four large joints involved scores 3 points.

Note: Based on a presentation by Dr. Hawker at the annual meeting of the American College of Rheumatology, October 2009.