Clinical news updates from the 2006 AHA Scientific Sessions

This AHA Special Report focuses on the cardiovascular pharmacotherapy trials and news of greatest interest to formulary decision-makers, including: emerging reports on drug-eluting stent safety, FAME, thiazolidinediones and heart failure, MEDAL, CHICAGO, STAMINA-HeFT, testosterone levels in women and coronary heart disease risk, and the use of fluvastatin XL/ezetimibe in patients at high risk for rhabdomyolysis.

DES versus bare metal stents

A series of analyses offers no consensus on the safety of drug-eluting stents (DES) compared with bare metal stents. Depending on the study, DES either result in an increase in major adverse coronary events (MACE) or a decrease in subsequent need for coronary artery bypass graft (CABG) surgery and repeat percutaneous coronary intervention without an excess of adverse clinical events.

In a registry of 9,043 patients undergoing intracoronary stenting at LDS Hospital/University of Utah, the rates of MACE, defined as death, nonfatal myocardial infarction (MI), target vessel revascularization (TVR), or CABG were assessed out to 3 years for patients receiving either bare metal stents or DES.

"After the first year, there were statistically worse outcomes with drug-eluting stents versus bare metal stents," said Joseph B. Muhlestein, MD, co-author of the study, and professor of medicine at the University of Utah, Salt Lake City.

The risk of MACE was increased by 36% (P<.0001) and all-cause death was 93% more likely (P<.0001) with DES, Dr Muhlestein said. Furthermore, although the DES reduced the rate of target lesion revascularization, non-target lesion-related revascularization was more common with the DES, resulting in a significant increase in all forms of coronary revascularization at 3 years.

The DES may have contributed to an acceleration of atherosclerosis in nonstented coronary segments, Dr Muhlestein said, a contention supported by other studies that have demonstrated evidence of endothelial dysfunction in coronary segments distal to the DES.

David O. Williams, MD, director of the cardiovascular laboratory and interventional cardiology at Rhode Island Hospital in Providence, RI, presented findings from a National Heart, Lung, and Blood Institute (NHLBI) registry of 3,223 patients who received either a bare metal stent (October 2001 to March 2002) or a DES (February to May 2004).

Sidney Smith, MD, professor of medicine at the University of North Carolina, Chapel Hill, NC, and past president of AHA, pointed out that the meta-analyses presented in September at the World Congress of Cardiology (WCC) meeting in Barcelona, Spain, demonstrated that late stent thrombosis seemed to be a problem with the DES. Dr Smith noted that Dr Williams' follow-up was only 1 year, whereas Dr Muhlestein's analysis followed patients for MACE at 3 years.

A randomized study of 217 patients who had either a pac-litaxel- or sirolimus-eluting stent or a bare metal stent implanted for ST-segment elevation MI found a strong trend toward better outcomes with the DES compared with the bare metal stent at 18 months. In this study, conducted at University Hospital Basel in Switzerland, the rate of MACE was higher in the bare metal stent group (16.0%) than in the groups assigned to DES (7.7%) that did not quite achieve statistical significance (P=.06). Both DES performed similarly in this study.

A fourth study of 505 patients undergoing primary angioplasty and treated with either bare metal stents or DES found an early advantage to DES with respect to target vessel revascularization. However, DES (either the sirolimus- or the paclitaxel-eluting stent) were no longer significantly superior to bare metal stents after 3 years of follow-up in reducing MACE, said lead investigator Joost Daemen, MD, clinical research fellow in interventional cardiology, Thoraxcenter, Erasmus Medical Center, Rotterdam, Netherlands. Mortality was comparable between the bare metal stents and DES at 3 years.

"Fifty percent of target vessel revascularizations were performed after the first year of follow-up due to stent thrombosis, and 100% of the MIs after the first year were due to stent thrombosis" Dr Daemen said.

One investigator who called for restricting the use of DES was Sanjay Kaul, MD, MPH, whose rationale is based on the results of a cost-effectiveness analysis he presented here. In the analysis, the substitution of DES for bare metal stents appears to be cost-effective only when the TVR rate with bare metal stents is high, which is typically the case in small vessels and longer stenoses.

The purported cost superiority of DES over bare metal stents was derived on the basis of 2 pivotal clinical trials in the United States, which can be misleading, Dr Kaul said. These reports of cost-effectiveness were based on model assumptions that are not consistent with "real-world" practice. For instance, the trials demonstrating cost-effectiveness of DES assumed that 1.4 stents are used per case, that the TVR rate with bare metal stents is >12%, and that the rate of TVR with DES is reduced by 60%.

As exemplified by a real-world study known as BASKET (Basel Stent Kosten Effektivitäts Trial), 1.9 stents were used per case, the TVR rate with bare metal stents was 7.8%, and the reduction in TVR by DES was 41%.

In Dr Kaul's analysis, he assumed that a DES costs $1,600 more than a bare metal stent, either 1.5 or 2.0 stents were used per case, the attributable cost of restenosis was $20,000, and the definition of cost-effectiveness was $30,000 or less per TVR avoided.

Substituting a DES for a bare metal stent was cost-effective only when: 1) the risk of restenosis with a bare metal stent was high (a TVR rate >15%), which constitutes only about 20% of stent-eligible patients, and is typically observed in vessels smaller than 3 mm and stenoses longer than 15 mm; or 2) as "bail-out" use in patients who present with clinical restenosis following a bare metal stent, who represent about 10% to 15% of patients.

"The potential benefits of drug-eluting stents [have] been overestimated and the potential risks have been underestimated," said Dr Kaul, director of the vascular physiology and thrombosis research laboratory, Cedars-Sinai Medical Center, Los Angeles, Calif.

The analysis did not include the possibility of late clinical events due to stent thrombosis, a concern with DES that has been uncovered during the past year. "That aside, just to focus on efficacy and cost, we're making the point that drug-eluting stents are not as cost-effective as they were made out to be. When you add the antiplatelet therapy to offset the risk of stent thrombosis, the cost-effectiveness will be significantly diminished."

Dr Kaul added: "If the risk associated with stent thrombosis resulting in clinical events is true, then it's moot to measure cost-effectiveness."

In related news, AHA issued a statement in early December regarding the use of DES. Information is available at http://www.americanheart.org/. According to AHA president Raymond J. Gibbons, MD, FAHA, "These recent reports have raised appropriate concerns regarding drug-eluting stents, and more research is needed to determine the long-term efficacy and safety of these devices. Until such information is available, patients should receive appropriate anti-platelet therapy according to existing practice guidelines wherever possible. This will usually require both aspirin and clopidogrel for periods of up to a year after stenting and aspirin indefinitely. Patients should not discontinue either aspirin or clopidogrel within the first year without consulting their treating cardiologist."

Dr Gibbons said that AHA is collaborating with 5 other major medical organizations on an advisory, which they plan to issue soon as a source of further guidance.

FAME

Pharmacist intervention improves medication adherence, leads to risk factor improvement

Pharmacist-directed education combined with prepackaging of medications increases compliance with medication schedules, leading to improvements in blood pressure and cholesterol levels among elderly patients, according to a study headed by Allen J. Taylor, MD.

The Federal Study of Adherence to Medications in the Elderly (FAME) tested the effect of a comprehensive pharmacy program on medication adherence in 200 patients aged ≥65 (mean age, 78 years) who were taking 4 or more medications daily. Baseline data were collected during a 2-month run-in phase, which was followed by an intervention phase during Months 2 to 8. Following the intervention phase, 159 patients entered a randomized phase of continued pharmacy care versus return to usual care during Months 8 to 14.

The majority of enrolled patients had drug-treated hypertension (91.5%) and/or hyperlipidemia (80.6%). The patients took an average of 9 chronic daily medications.

The pharmacy care program consisted of individualized medication education using standardized scripts, regular follow-up by pharmacists (every 2 months), and all medications dispensed in time-specified blister packs. During each individualized education session, patients learned about drug names, indications, strengths, adverse effects, and use instructions. The initial visit was scheduled for 1 hour; subsequent visits were scheduled for 30 minutes.

After the intervention phase, medication adherence increased from 61.2% to 96.9%, systolic blood pressure declined from 133.2 mmHg to 129.9 mmHg (P=.019), and low-density lipoprotein (LDL) cholesterol decreased from 91.7 mg/dL to 86.8 mg/dL (P=.001).

After the randomized phase, there was sustained medication adherence in the continued pharmacy group (95.5%), whereas medication adherence fell to 69.1% in the group that reverted to usual care. "There was no durable effect of the program after having it withdrawn," said Dr Taylor, chief of cardiology at Walter Reed Army Medical Center in Washington, DC. The improved persistence in the intervention group resulted in significant reductions in blood pressure compared with the usual care group (10.7 mmHg vs 3.1 mmHg; P=.04), although the LDL cholesterol reduction during the intervention phase was sustained in patients later randomized to usual care.

The patients in this study received their medications through the military health system and thus did not have the financial concerns that many elderly patients do, said Dr Taylor.

The findings suggest that continued provision of blister-packed medications is a key component of a medication adherence program, given that the initial increase in medication adherence during 6 months of the pharmacy care adherence program did not persist in the patients who were randomized to usual care, he noted.

The medication prepackaging cost was 14 cents per month, he said, with the major expense of the program being the pharmacist's time, which may limit the widespread use of the system. Nevertheless, Dr Taylor said a program to improve medication compliance may be warranted, considering the amount of money third-party payors and the Medicare system spend on the drugs themselves.

"Because medication nonadherence is pervasive and morbid, the results of FAME call for greater emphasis within healthcare delivery systems and policy organizations on the development and promotion of clinical programs to enhance medication adherence, particularly among the at-risk elderly," he said.

Thiazolidinediones study

Restrospective analysis examines TZD use in ambulatory patients with diabetes and heart failure

Thiazolidinediones (TZDs) do not appear to increase the risk of heart failure (HF) hospitalization and death in ambulatory patients with diabetes and HF, according to a retrospective analysis of a veterans population.

"I would use TZDs cautiously in patients with diabetes who have class I or II compensated HF," said the lead author of the study, David Aguilar, MD, assistant professor of medicine at Baylor College of Medicine in Houston, Texas. He noted that many clinicians are already using these agents in patients with diabetes and HF without evidence to support that this practice is safe.

TZD use is not recommended for patients with New York Heart Association class III or IV symptoms because they may exacerbate volume status and may worsen HF signs and symptoms.

Among 7,147 veterans with HF and diabetes who were treated in ambulatory clinics at Veterans Affairs medical centers, 818 were being treated with TZDs and 4,700 were not. These patients were the basis of the retrospective study, which had as its primary outcome the time to hospitalization for HF. The secondary outcome was time to death.

During the 2-year follow-up, the rate of HF hospitalizations was not significantly different between the TZD users and non-users (16.4% vs 15.8%, respectively).

The 2-year incidence of death was actually lower in the TZD users (20.5%) than the non-users (25.4%). TZD users tended to be younger with higher ejection fractions, and less likely to be hospitalized for HF in the 2 years preceding the study. When adjusting for these differences, "the benefit with TZD use went away but we still couldn't find a hazard," Dr Aguilar said.

Although these data are reassuring, Dr Aguilar said, he believes that a prospective, randomized trial of TZD use is warranted in patients with diabetes and class I or II symptoms.

MEDAL

Investigational COX-2 inhibitor etoricoxib comparable to diclofenac in CV safety

A comparison of the investigational COX-2 inhibitor etoricoxib and diclofenac found no increase in the risk of cardiovascular events over 3 years with etoricoxib.

The finding suggests that "COX-2 selectivity alone does not increase cardiovascular risk," said Christopher Cannon, MD, lead investigator of the Multinational Etoricoxib and Diclofenac Arthritis Long-Term (MEDAL) study.

At 3 years, there were 320 thrombotic cardiovascular events in the group randomized to etoricoxib and 323 in those randomized to diclofenac, a nonsignificant difference (Table 2). The rate of fatal thrombotic cardiovascular events was 0.17 per 100 patient-years in each group (Table 3).

Etoricoxib did significantly reduce the rate of upper gastrointestinal events by 31%, but the incidence of perforations, obstructions, and major bleeding was not different between etoricoxib and diclofenac (0.30 vs 0.32, respectively, per 100 patient-years).

Robert Califf, MD, vice chancellor for clinical research at Duke University, Durham, NC, voiced concern over the single comparison design and the choice of the comparator. "The single comparison in this trial leaves open the issue of whether naproxen or a different NSAID would have shown a better result," he said.

Dr Cannon noted that diclofenac was a suitable comparator because it is the most widely prescribed NSAID in the world and does not interfere with aspirin's anti-platelet effect.

Although congestive heart failure was rare, it occurred more often in the group receiving 90 mg/d of etoricoxib compared with diclofenac (absolute increase, 0.1% to 0.4%), although this difference was not significant. Discontinuation for edema also was greater with 90 mg/d of etoricoxib than with diclofenac, but the rates were similar for 60 mg/d of etoricoxib and diclofenac. Both doses of etoricoxib were associated with higher rates of discontinuation due to increases in blood pressure compared with diclofenac.

Diclofenac was associated with a higher rate of discontinuation due to hepatic adverse events.

"We're moving into an era where we do have choices and probably should be thinking about different agents for different patients," Dr Cannon said. Direct comparisons of COX-2 inhibitors with ibuprofen and naproxen are needed, he said.

In response to the MEDAL study results, AHA released a statement in which it said that "current evidence indicates that selective COX-2 inhibitors have important adverse cardiovascular effects including increased risk for heart attack, stroke, heart failure, and hypertension." The AHA statement added: "Data suggest that even those NSAIDs with little COX-2 selectivity may not be completely safe for cardiovascular patients, and therefore all NSAIDs should only be prescribed after thorough consideration of the risk-benefit balance."

CHICAGO

Pioglitazone demonstrates anti-atherosclerotic effect in patients with type 2 diabetes

The thiazolidinedione pioglitazone has a beneficial effect on carotid intima media thickness (CIMT) compared with glimepiride in patients with type 2 diabetes, said Theodore Mazzone, MD, lead investigator of the CHICAGO (Carotid Intima Media Thickness in Atherosclerosis Using Pioglitazone) study.

CIMT, a surrogate marker for atherosclerosis and cardiovascular risk, was measured at baseline and again at 24, 48, and 72 weeks in 462 patients with type 2 diabetes who were randomized to either 15 to 45 mg/d of pioglitazone or 1 to 4 mg/d of glimepiride. To be eligible, patients could not have symptomatic coronary artery disease, cerebrovascular disease, peripheral arterial disease, or cardiac failure at baseline.

Hemoglobin A_1c (HbA_1c) levels were similar between the 2 groups until Week 48, when levels in the pioglitazone group became significantly lower than those in the glimepiride group (0.32% difference between the 2 arms; P=.002).

At Week 72, posterior wall CIMT increased by .012 mm in the patients randomized to the glimepiride group and decreased by .001 mm in those in the pioglitazone group (P=.017). The beneficial effect was observed across all subgroups based on age, gender, blood pressure, body mass index, glucose control, and duration of diabetes. Notably, approximately 60% of patients were on statin therapy, and the effect of pioglitazone on CIMT was apparent in those patients as well, said Dr Mazzone, chief of endocrinology, diabetes, and metabolism at the University of Illinois at Chicago.

One patient taking pioglitazone was hospitalized for heart failure (HF), which was reversed upon discontinuation of the drug. Although CHICAGO was not powered to detect differences in clinical events, 10 cardiovascular events occurred in the glimepiride group versus 4 in the pioglitazone group.

The data from CHICAGO add to a growing body of evidence that pioglitazone may reduce cardiovascular risk, Dr Mazzone said. In PROactive (Prospective Pioglitazone Clinical Trial in Macrovascular Events), pioglitazone was associated with a significant reduction in the secondary end point-a composite of death, myocardial infarction, and stroke-compared with placebo, but the reduction in the primary end point, which also included peripheral artery revascularization, in the pioglitazone group was not significant. Viewed in the context of PROactive, the positive effect on CIMT in CHICAGO supports a cardiovascular benefit to pioglitazone, he said.

The finding in CHICAGO also is consistent with preclinical work with pioglitazone, which demonstrated this agent to have anti-inflammatory and anti-atherosclerotic effects, Dr Mazzone said. "We have a solid foundation of experimental evidence showing that pioglitazone is beneficial for preventing atherosclerosis," he said.

Although changes in CIMT cannot be directly extrapolated to a change in clinical event rate, he noted that statins have demonstrated a similar benefit on CIMT that has translated into a reduction in clinical events.

"Pioglitazone could be part of a novel strategy to manage residual cardiovascular risk in type 2 diabetes," Dr Mazzone said.

Peter W.F. Wilson, MD, professor of medicine at Emory University, Atlanta, Ga, said that the low rate of HF with pioglitazone "opens the window a little wider" for the use of thiazolidinediones. The group of patients selected for inclusion may represent a group that was not likely to develop HF with the use of these agents, he said.

Pioglitazone is currently being studied in a trial called PERISCOPE (Pioglitazone Effect on Regression of Intravascular Sonographic Coronary Obstruction Prospective Evaluation) using intravascular ultrasound to measure atherosclerosis in coronary arteries, which is being conducted in patients with a history of cardiovascular disease.

STAMINA–HeFT

Anemia and heart failure: Improvement in outcomes tied to increase in hemoglobin level

A rise in hemoglobin of at least 1 g/dL appears necessary to affect the course of heart failure (HF) in patients with symptomatic HF and anemia, according to a post hoc analysis of STAMINA-HeFT (Studies of Anemia in Heart Failure-Heart Failure Trial).

STAMINA-HeFT was a randomized, double-blind, placebo-controlled study to assess the safety and efficacy of treating anemia with twice-weekly darbepoetin alfa in anemic HF patients. In the study, 319 patients with New York Heart Association class II to IV HF who had hemoglobin (Hb) levels of 9.0 to 12.5 g/dL were randomized to placebo or darbepoetin alfa for 1 year to achieve a target Hb of 14.0 g/dL.

The mean improvement in Hb level in the patients assigned to darbepoetin alfa was 1.8 g/dL. The main results demonstrated a trend toward improvement in exercise duration (the primary end point) with darbepoetin alfa that did not reach statistical significance. The new analysis included only the darbepoetin alfa-treated patients in STAMINA-HeFT, and compared outcomes among those who achieved at least a 1 g/dL increase in Hb after Week 17 and those who failed to achieve at least a 1 g/dL increase in Hb.

The adjusted mean change in exercise duration from baseline to 6 months was 66.2 seconds in the patients with ≥1g/dL increase in Hb, compared with an increase of 14.7 seconds in those with a <1.0 g/dL increase in Hb (P=.037), reported lead investigator Jalal Ghali, MD, Wayne State University Health Center in Detroit, Mich.

Clinical outcomes (all-cause mortality, HF-related hospitalization, and a composite of the 2) also were significantly improved in the patients who achieved at least a 1 g/dL increase in Hb (Table 4).

There were fewer serious adverse events in the patients who achieved ≥1 g/dL increase in Hb.

This retrospective analysis strengthens the hypothesis that raising Hb in anemic HF patients may lead to improved outcomes, said James Young, MD. "By no means are these conclusive data, but they show that the hypothesis is rational," he said. "Pilot studies have given us assurance that in a large number of heart failure patients it's safe to raise hemoglobin with erythropoietin-stimulating therapies and there is a suggestion that patients improve."

Although recent studies indicate that a target Hb in the normal range may be too high for patients with anemia and renal insufficiency or end-stage renal failure, "in the heart failure population, the target Hb is very much up in the air," said Dr Young, professor and chairman, division of medicine, Cleveland Clinic, Cleveland, Ohio.

Patients enrolled in STAMINA-HeFT and other pilot studies of anemia correction in patients with HF are different from those with renal insufficiency, he noted. "It's easier to correct Hb in an HF patient than in one with renal insufficiency or on hemodialysis," he said.

Two large-scale clinical trials that are studying anemia correction with darbepoetin alfa in patients with HF have given no indication of harm with therapy, Dr Young said.

Testosterone study in women

Low testosterone levels associated with increased risk of mortality from CHD in postmenopausal women

Low levels of testosterone are associated with an increased risk of mortality from coronary heart disease (CHD) in postmenopausal women, said Gail A. Laughlin, PhD.

Women's favorable cardiovascular risk compared with men is usually explained by endogenous estrogen levels. However, few studies have evaluated the role of naturally circulating testosterone in older women. Hyperandrogenemia in young women with polycystic ovary syndrome is associated with adverse cardiovascular risk factors, carotid atherosclerosis, and impaired vascular function, said Dr Laughlin, assistant adjunct professor of family and preventive medicine, University of California, San Diego, Calif.

Using enrollees from the Rancho Bernardo Study, an ongoing population-based study of healthy aging in a Southern California community, researchers assessed the incidence of CHD events in 678 postmenopausal women aged 50 to 90 years over a follow-up of 20 years. None of the women were current users of estrogen. Women were stratified into quintiles based on their testosterone levels, with those in the lowest quintile having testosterone levels ≤80 pg/mL.

The study demonstrated that women with testosterone levels in the lowest quintile had a 2-fold increased risk of prevalent CHD, incident CHD (defined as a first-ever myocardial infarction or revascularization procedure), and CHD mortality. The findings were independent of the presence of CHD risk factors, diabetes, and metabolic syndrome.

"These results suggest that androgen deficiency may be a specific risk factor for atherosclerotic heart disease in postmenopausal women," Dr Laughlin said. Whether the results apply only to women who are not current users of estrogen awaits further study.

Rhabdomyolysis study

Fluvastatin XL/ezetimibe combination well tolerated in patients with muscle-related side effects on other statins

The combination of fluvastatin XL and ezetimibe effectively and safely lowers low-density lipoprotein (LDL) cholesterol in patients who previously experienced muscle-related side effects to other statins, said Evan Stein, MD.

In a randomized, double-blind study, 85% of patients who had statin-associated muscle-related side effects had no muscle-related side effects when treated with either fluvastatin XL alone or in combination with ezetimibe, he said.

"Although rhabdomyolysis and myopathy are rare with statins, muscle-related side effects are a significant and growing problem," Dr Stein said. In the PRIMO (Prediction of Muscular Risk in Observational conditions) survey of 7,924 patients in primary care practices who were receiving high-dose statin therapy, the incidence of muscle-related side effects was 10.5%. Extrapolated to the 20 million patients on statin therapy worldwide, about 2 million patients may be expected to discontinue therapy because of myalgia, he said, leaving physicians and patients minimal ability to achieve LDL cholesterol goals.

According to Dr Stein, fluvastatin XL has several unique pharmacologic properties that may minimize the potential for muscle-related side effects. It has low lipophilicity, making it unlikely to diffuse across membranes into muscle cells; an increased first-pass clearance by the liver that results in minimal systemic exposure (virtually undetectable levels in the peripheral circulation); a short half-life that also results in minimal systemic exposure; and limited metabolism by the CYP3A4 enzyme that limits interactions with other drugs.

He said that fluvastatin was the only statin without a death from rhabdomyolysis in an FDA database published in 2001.

"As far as I know, this is the first study that recruited 100% of patients who had muscle-related side effects. Most of the patients had been on more than 1 statin," said Dr Stein, from the Metabolic and Atherosclerosis Research Center, Cincinnati, Ohio.

The 12-week study duration was chosen because most of the PRIMO patients who suffered muscle-related side effects to statin therapy did so within the first 3 months.

Average reductions in LDL cholesterol were 15.6% with ezetimibe alone, 32.8% with fluvastatin XL alone, and 46.1% with the combination.

More patients on the combination achieved LDL cholesterol targets of <100 and <70 mg/dL compared with either monotherapy (Table 5). The incidence of muscle-related side effects and muscle-related side effects that led to discontinuation of the study drug(s) was lower for both fluvastatin XL groups compared with ezetimibe monotherapy.

"The combination is as good as the highest doses of simvastatin, atorvastatin, and moderate doses of rosuvastatin," Dr Stein said, noting the importance of adherence.

Among the fluvastatin recipients, there were no additional patients who experienced muscle-related side effects after the first month of treatment, Dr Stein said.