In the largest such analysis of its kind, the Duke researchers said their findings have important economic and clinical implications for physicians who are deciding whether their heart patients should receive coronary artery bypass surgery, or less-invasive angioplasty, which includes the placement of a stent.
Stents, which were introduced in the U.S. in 1994, are tiny mesh tubes that are inserted at the site of a blockage in a coronary artery that has been opened during balloon angioplasty. The procedure seeks to prevent the artery from becoming blocked again, a process known as restenosis. These blockages, caused by atherosclerotic plaque, can starve the heart of oxygen-rich blood and lead to a heart attack.
Duke cardiologist David Kandzari, M.D., who presented the results of the Duke analysis Nov.7, 2004, at the American Heart Association's annual scientific sessions in New Orleans, said the findings on mortality rates should also be expected to hold true for the latest generation of drug-eluting stents. These stents, which were introduced in 2003, are coated with a drug that keeps blood clots from forming inside them.
"We have found in our long-term analysis that stents do provide a significant early and sustained reduction in the need for subsequent procedures to re-open the treated artery," Kandzari said. "However, we also found that stents do not have any influence on long-term survival.
"Since earlier studies have shown that new drug-eluting stents can lessen the incidence of restenosis, we would expect the need for repeat procedures to decline even more as these stents become more widely used," Kandzari continued. "While earlier trials of drug-eluting stents have demonstrated a significant reduction in repeat procedures, they still have shown no differences in mortality compared with more conventional stents."
Specifically, the researchers found that over the average seven-year follow-up period of their study, 19 percent of patients who received a stent needed another revascularization procedure in the treated artery, compared to 27 percent for those who did not receive a stent. However, the long-term mortality rate for those receiving a stent was 19.9 percent vs. 20.4 percent for those who did not, a disparity which did not statistically differ.
For their analysis, the researchers consulted the Duke Database for Cardiovascular Disease, which keeps detailed clinical data on all heart patients receiving treatment at Duke. The researchers identified 1,288 matched pairs of patients who underwent either balloon angioplasty alone or stenting -- yet all had a similar likelihood of receiving a stent based on their clinical and demographic characteristics.
The patients, 63 percent of whom were male and who had an average age of 59 years, were treated between 1994 and 2002. One in four was diabetic, and one in four had suffered a previous heart attack.
"This study, based as it is on a real-world population of patients, tells us that stents do not save lives, though they do have a profound effect on avoiding repeat procedures," Kandzari said. "We've know that restenosis has never been scientifically associated with increased mortality, but it has been associated with an increased need for revascularization and with a reduction in symptoms such as chest pain."
Given these findings, Kandzari said physicians treating their heart patients should not automatically assume that placing a stent, whether the original bare-metal type or the newer drug-eluting version, will be the end of treatment.
"Many physicians will successfully place a stent and think that's it," Kandzari continued. "The bigger issue is that many of these physicians should also then be prescribing drugs that have a clearly demonstrated beneficial effect on long-term mortality."
Kandzari plans to follow up this study with a similar analysis of the effects of the drug-eluting stents on mortality. Also, the team plans to measure any differences in the quality of life of these patients.
"As we take on more and more difficult and complicated cases in the catheterization lab, we should take a step back to see if there are certain instances when bypass surgery may be the best option," Kandzari said. "There is the temptation out there to just place stents in all patients, no matter what. In some prior trials, the difference in outcome between angioplasty and surgery patients was driven by restenosis, not by differences in mortality.
"However, in the era of conventional stenting, we knew that there we still some instances in which bypass surgery might provide an incremental survival benefit," Kandzari said. "Before routinely placing drug-eluting stents in similar patients, these findings underscore the need for systematic evaluation of drug-eluting stents in these types of patients."
Patients with left main coronary artery disease, for example, appear to fare better with bypass surgery than with angioplasty and stents. Previous study has also suggested this may be true for diabetic heart patients with extensive disease, who appear to benefit the most from bypass surgery, Kandzari said.
"Appropriately, a trial is forthcoming to compare treatment with drug-eluting stents with bypass surgery in diabetic patients," he said.
The study was supported by the Cordis Corp., Miami Lakes, Fla., which develops stents. Kandzari has no financial interest in Cordis.
Other members of the Duke team were Robert Tuttle, M.D., James Zidar, M.D., and James Jollis, M.D.