By AMANDA PEDERSEN
CDU
and DON LONG
CDU Executive Editor

A new meta-analysis published last month in The Lancet is being touted as easing concerns about drug-eluting stents (DES), but it may not do all that much to provide great confidence - or reverse the slump - in this device sector.

The analysis looked at 38 randomized controlled trials involving more than 18,000 patients, and it concluded that the death risks associated with DES and bare-metal stents (BMS) are about the same. The findings compare the two FDA-approved DES products — the Cypher from Cordis (Miami Lakes, Florida), the DES unit of Johnson & Johnson (J&J; New Brunswick, New Jersey) and the Taxus from Boston Scientific (Natick, Massachusetts) — with the conclusions in that comparison tending to favor the Cypher stent.

According to the analysis, patients who received the Taxus were somewhat more likely to suffer heart attacks than those who received the Cypher. The research also indicates a slightly elevated risk of fatal thrombosis associated with Taxus more than 30 days after implantation, compared with BMS. Cypher patients did not run that risk, according to the data, and fewer Cypher patients needed follow-up procedures to deal with a recurrence of coronary blockage.

Christopher Allman, executive director of corporate communications at Cordis, told Cardiovascular Device Update that the 38 trials included in the analysis were funded from different sources. Some of the trials, such as REALITY, SIRIUS, E SIRIUS, RAVEL, were Cordis-funded, he said.

"As the largest and most comprehensive analysis done to date for drug-eluting and bare metal stents, we believe physicians and patients will find these results support the long-term safety of drug-eluting stents, but they also clearly show the differences between the two drug-eluting stents," Allman said.

But Donald Baim, MD, chief medical and scientific officer for Boston Scientific, said in a statement provided to CDU that the meta-analysis failed to reflect that Taxus was compared in its major trials with a better-performing BMS device (the Express) than Cordis chose for the Cypher trials. The Cypher was compared to the BX Velocity stent, Baim said.

The Bx Velocity had higher rates of repeated revascularization at four years and higher rates of stent thrombosis than the Express stent used in the Taxus trials, Baim noted.

Also, he said that two of the major comparisons with BMS in the Boston Scientific trials focused on smaller arteries and longer blockages than those tested in Cordis' trials - conditions where risks of complications would presumably be higher.

"The attempt to force the two different BMS to be equal in the network meta-analysis distorts the estimate of how Cypher and Taxus compare, and leads to the false conclusion that Taxus has 2.7 times as much late-stent thrombosis as Cypher, and 1.4 times as much repeat revascularization as Cypher," Baim said.

The study data put the rate of mortality as similar for all three stents: hazard ratios (HR) were 1.00 (95% credibility interval 0.82 to 1.25) for the Cypher vs. bare-metal stents, 1.03 (CI 0.84 to 1.22) for the Taxus vs. bare-metal stents and 0.96 (CI 0.83 to 1.24) for the Cypher vs. the Taxus, Cordis noted.

While the results can be taken and spun in a variety of ways, they probably won't fuel any renewed uptake of either of the DES devices or of stenting in general. Though J&J/Cordis will banner the positive findings for the Cypher, close observers of the sector will see the study as continuing to rain on the DES parade as having been hyped too early and too often.

Meta-analyses have a variety of built-in flaws, in this case relating to examination of several carefully constructed pre-approval trials of relatively small populations over shorter-term periods rather than in real-world use over longer periods.. And the conclusions, apart from what they say about differences between Cypher and Taxus, also fail to provide any boost in the reputation of DES devices over BMS, which were initially touted, before and shortly after FDA approval, as being far superior.

The initial studies put the need for redo because of in-stent restenosis following BMS implant at 20% to as high as 30%, while DES implant studies reported redo data as low as 3% to 5%.

That math doesn't find its way into the meta-analysis that now says DES restenosis and the need for follow-on procedure is just 50% of that required by BMS use. Though this is a significant difference, it doesn't indicate great cost-effectiveness for DES devices, which run to more than two-and-one-half times the cost of a BMS device.

Additionally, the Lancet report indicates that there is no great difference between DES and BMS in the most significant endpoint for the patient: added months/years of life. This is an endpoint obviously unavailable in the early trials, but the assumption by most clinicians probably was that if DES devices greatly reduced restenosis, they probably also reduced other cardiovascular ills and the rate of death from heart attack or heart failure. But they don't, according to the study.

Finally, the analysis continues to suggest that if DES is an advance, it is an advance that will likely be seen in the 2.0 edition of this technology, coming on-line next year, rather than in the current generation of Cyper, Taxus and BMS devices.

Carbamylation believed to be a culprit in heart disease

Curiosity as to why smokers are at an increased risk of developing heart disease has led researchers, some at the Cleveland Clinic, to a discovery of a marker that has the potential, they say, to be as important as cholesterol is a predictor of heart disease.

The researchers say the study may help explain why one person suffers a heart attack, and another does not. The Cleveland Clinic researchers' findings were published in the Sept. 9 issue of the journal Nature Medicine and titled "Protein carbamylation links inflammation, smoking, uremia and atherogenesis." In addition to smoking, the study also looks at why those with chronic renal disease are more likely to have cardiovascular disease.

"Of all the markers we've tested so far, this one outperforms everything," Stanley Hazen, MD, PhD, section head of preventive cardiology at the Cleveland Clinic and a researcher in the clinic's Lerner Research Institute, told CDU. "And it's independent of cholesterol, HDL, LDL; it's independent and additive to MPO and additive to C-reactive protein and other inflammation markers."

Hazen said the researchers began their effort because they were interested in determining why patients who smoke are at an increased risk to develop heart disease, so they began studying the "chemistry of how smoking affects the inflammation system."

"And it was through this approach that we discovered that carbamylation, which is a post-translational modification of proteins, is increased due to a product that increases in the blood in smoking," he said.

While the researchers are not saying carbamylation is the cause of atherosclerosis, they do "think it is a contributor to that," he said. This mechanism is not only present in high-risk populations, it is also found in those with chronic kidney disease. When proteins in the bloodstream are damaged by carbamylation, it changes the way the cells behave, promoting the accumulation of harmful substances in the arteries and therefore raising the risk of heart disease, the Cleveland Clinic said.

"We think that we have now identified — call it the 'dark matter' that is a heretofore unrecognized underlying pathogenic mechanism that's involved in the development of atherosclerosis in everyone," Hazen said, because, he pointed out, doctors already know that cholesterol is not the "be all and end all of how heart disease occurs." He said that, for example, some people who don't have high cholesterol develop heart disease, and others with high cholesterol don't develop atherosclerosis.

The carbamylation process increases the level of a substance called thiocyanate in plasma, which a white blood cell enzyme called myeloperoxidase uses to covalently modify and alter its function to make proteins like "LDL become more atherogenic, and HDL lose some of its protective function."

The researchers found that a blood test measuring systemic levels of homocitrulline, a molecular marker for the pathway, serves as the "strongest independent predictor of heart disease risk identified thus far," according to the Cleveland Clinic.

"Then what we found is that this was true not just of smokers but in everyone, and the levels of this in the plasma very much depends upon what your diet is ...", Hazen said.

Hazen's research builds on earlier research conducted at the Cleveland Clinic in the early part of this decade centered around myeloperoxidase (MPO), described then as an "abundant leukocyte enzyme" that is elevated in "culprit lesions that have fissured or ruptured in patients with sudden death from cardiac causes," according to a study abstract published in the New England Journal of Medicine in October 2003.

The blood test that the clinic developed was meant to determine whether a person is in danger of experiencing a heart attack within 30 days to six months.

Increased life span driven by improved vascular treatments

Preliminary numbers for 2005 from the Centers for Disease Control and Prevention (Atlanta), reported in mid-September, indicate that life expectancy in the U.S. continues to nudge upward, with CDC's National Center for Health Statistics (NCHS) reporting that life expectancy "at birth for the total population in 2005 reached a record high of 77.9 years." The most conspicuous improvements in mortality by disease category were for heart disease and cerebrovascular disease.

However, neurological diseases and diseases of the lower respiratory system almost completely offset those improvements.

The NCHS report said that the preliminary number-crunch for 2005 "puts life expectancy in the U.S. at nearly 78 years, a figure that has been increasing steadily over the last 50 years." This life expectancy figure was one-tenth of a year of life expectancy greater than in 2004. Life expectancy in 1955 was 69.6 years and in 1995 75.8 years, according to NCHS.

The total number of deaths from "diseases of the heart" in 2005 was just under 650,000, and the age-adjusted death rate for this category fell 3.1%, from 217 to 210.3 per 100,000 people. In comparison, cancer took the lives of slightly less than 560,000, but the drop in the age-adjusted mortality rate for cancer fell only 1.1% to 183.8 per 100,000 in 2005.

Cerebrovascular disease, including stroke, came in as the third leading cause of death in the U.S., taking the lives of about 144,000 annually.

Treatments for this category have apparently had a huge impact if the decline of 6.8% in age-adjusted mortality is any indication - from 50 per 100,000 to 46.6 per 100,000.

The life expectancy number is calculated as the year in which a child is born, to the year he or she is expected to live to, on average. Any comparison to life expectancy of those who had already reached adulthood in 2005 would look different, because overall expectancy at birth takes into account the deaths that will occur in infancy, childhood and adolescence.