Don't Confuse Me With The Facts

Two examples of ignoring facts to reach conclusions you already have.

First, a NEJM study about the risks of Meridia

The study reaffirmed that that patients with heart problems should not be prescribed Meridia," the New York Times (9/2, A25, Harris) reports, a "restriction already included in Meridia's label." But, the journal editors don't think the investigators went "far enough." Gregory D. Curfman, MD, NEJM's executive editor, added, "It wasn't that we disagreed with the interpretation of the authors," but "many patients...have cardiovascular disease and don't know it. How are you supposed to identify those patients who might be put at risk by putting them on drugs like sibutramine?"

In otherwords,  the drug is safe and effective as prescribed for the vast majority of patients but since we never know if everyone who gets it will learn about their risk of heart disease we should take the medicine off the market. 

Is it possible that the NEJM thinks doctors who are prescribing Meridia are too stupid or uninterested in potential medical malpractice suits to determine if someone might be at risk?  I don’t think so.  If every medicine that carries a CV risk was yanked from the market for that reason, you would have to take hundreds of drugs out of commission, including every pain killer.      The editorial's authors also assert that "the heart risk associated with Meridia is not justified by the weight loss seen in the study -- about 9.5 pounds (or 4.5 percent of initial body weight) after one year, on average," CNN /Health.com (9/2, Harding) reports.

And of course Steve Nissen has an informed judgment on the issue..

Steve Nissen, MD, "a cardiologist at the Cleveland Clinic Foundation who has been a vocal critic of recent FDA handling of drug safety concerns, agrees," according to Time (9/2, Park). "We've got a drug here that shows little benefit -- a few pounds of weight loss -- and we trade that for a 28% increased risk of heart attack and 36% increased risk of stroke."

Actually the study concluded: "nonfatal heart attacks occurred in 4.1% of sibutramine users and 3.2% of the placebo group, and nonfatal strokes occurred in 2.6% of sibutramine users and 1.9% of the placebo group,"  

That’s actually a relative risk of 22 percent and 27 percent – or increase in risk of about 1 in 100000 --  but who’s counting? 

 http://www.diahome.org/DIAHome/Home.aspx

The other example of ignoring facts comes from the study of whether gene testing for response to warfarin improves outcomes in the NEJM.   Two other studies suggested genetic testing was important for reducing bleeding associated with warfarin but not for new clot busting drugs.

When you come up with a conglomerate of conclusions that suggest testing is not necessary for different reasons and medicines it would make sense to suggest that you might want to do testing to get it right.   But all studies conclude that genetic testing is not necessary in most cases. 

Let’s take a look at the NEJM article that compared bleeding events in patients treated with  warfarin vs placebo to see why that might not be the best clinical approach:

“Results are presented only for patients of European or Latin American ancestry. Patients with other ancestries were excluded because of small numbers (99 patients in the next largest group) and concern about the potential for population stratification.” 

“Only 18.0% of patients in the CURE population included in our study underwent PCI, and only 14.5% underwent PCI with placement of a stent, as compared with more than 70% in previous studies.”

So the study --  truly a one size fits all approach – excludes African Americans and people undergoing PCI

Further:

"Carriers had a more pronounced reduction in cardiovascular events with clopidogrel treatment as compared with placebo than did noncarriers (hazard ratio with clopidogrel among carriers, 0.55; 95% CI, 0.42 to 0.73; hazard ratio among noncarriers, 0.85; 95% CI, 0.68 to 1.05; P=0.02 for the interaction). A similar interaction was observed with respect to the second composite primary outcome (hazard ratio with clopidogrel among carriers, 0.66; 95% CI, 0.54 to 0.82; hazard ratio among noncarriers, 0.90; 95% CI, 0.76 to 1.06; P=0.03 for heterogeneity)."

Isn’t finding out who  benefits more a good reason for testing?   Eric Topol thinks so:

“Genomics expert Dr Eric Topol (Scripps Research Institute, La Jolla, CA) told heartwire: "Both TRITON and PLATO reinforce the CYP2C19 story . . . that loss-of-function variants lead to diminished clinical impact for clopidogrel. PLATO takes this a step further to now show that the gain-of-function allele *17 is associated with more bleeding."

http://www.theheart.org/article/1114619.do

The rush towards headline grabbing instant analysis undermines a more systematic analysis in the Meridia and genetic testing issues and ignores the fact that RCT’s have limitations.

As another editorial noted with respect to gene testing:

“Besides genetic profiles, the evaluation of the best management should also take into account the clinical determinants of platelet reactivity—from age and sex to body-mass index, diabetes, and inflammation—that modulate platelet function, while also considering the timing from the acute event, she adds. "Prospective studies evaluating different antiplatelet treatments tailored according to the individual characteristics of patients are urgently needed to identify therapeutic strategies that will provide the best benefit for the single patient in this high-risk clinical setting."

Too bad CER funding is not available for such research.