November 16th, 2011

Was AIM HIGH a Hit or a Miss?

Several Cardiology Fellows who are attending AHA.11 this week are blogging together on CardioExchange.  The Fellows include Revathi Balakrishnan, Eiman Jahangir, John Ryan (moderator), and Amit Shah. Read the previous post here. Check back often to learn about the biggest buzz in Orlando.

In the weeks leading up to the AHA, there was much anticipation about the presentation of AIM HIGH amongst my co-fellows and attendings.  Prior to starting fellowship, I did an additional year of training in preventive medicine, so the evidence for (or against) HDL therapy is of particular interest to me. The details of the study were described earlier on CardioExchange , but briefly, patients with cardiovascular disease, cerebrovascular disease or PAD with low HDL and elevated triglycerides were randomized to receive simvastatin combined with either niacin or placebo and were followed for differences in rates of cardiovascular events (CHD death, nonfatal MI, ischemic stroke, ACS, or revascularization). The trial was stopped early back in May, after 3 years of follow up, for “lack of efficacy” and concern for ischemic stroke imbalance (although overall rates were low).  In addition, there was no significant difference for secondary endpoints, for cardiovascular death alone, and no differences between prespecified subgroups (by age, gender, diabetes, metabolic syndrome, history of MI, or statin at entry).

Sitting in the audience during the presentation, one could hear hushed rumblings throughout the crowd; the discussant, Dr. Philip Barter from The Heart Research Institute in Australia delivered probably the most dramatic commentary of the late breaking updates: “This study seriously disturbs me…because it was designed in such a way that it was not possible to get a positive result…” See the CardioExchange summary of the interaction.

He astutely pointed out faults in the inherent design of the trial:

  • The study was powered to see a 25% reduction in cardiovascular events – an ambitious goal for follow up over 2.5 to 7 years
  • Based on prior studies and the lipid goals attained in the study, the predicted CV event rate should actually be 12.5%, i.e., half of what the study was powered to detect
  • The placebo group was placed on low dose niacin and attained HDL elevation; there was only a 4 mg/dl difference in HDL between placebo and treatment group, and only a 5 mg/dL difference in LDL levels.

Dr. Barter pointed out that given these issues it is difficult to draw conclusions from the findings in this trial that are generalizable to clinical practice, as the hypothesis that they sought out to test was not actually successfully tested.  The panel concluded that the question of the benefits of HDL therapy with niacin has not been adequately answered by this trial.

When discussing HDL management of my clinic patients with attendings, I often hear diverse approaches and different opinions. One of the best parts of this conference has been the opportunity to hear experts on the panels dissect and interpret the data.

How have you been approaching patients with low HDL levels? How would you interpret this trial, and will it change your practice?

3 Responses to “Was AIM HIGH a Hit or a Miss?”

  1. This was a great session for fellows to attend because it went into details regarding trial design and statistical power calculations. As trainees, we are exposed to great faculty, but typically we only get to hear the opinion of our own faculty- at this late breaking session, fellows got to hear conflicting opinions from a variety of experts and hopefully form their own opinions- or at the very least appreciate the complexity of clinical care and research.

  2. Douglas Trenkle, DO says:

    CDP required 7 years to prove reduced mortality in 2ndary prevention, and Dr Greg Brown’s work (though small in number) revealed impressive angiographic improvement with niacin in combination therapy(HATS). After the initial critique and discontinuation of AIMHIGH, comprehensive review revealed the flaw(s) in the study design and its premature discontinuation. At our clinic, niacin remains the most difficult lipid lowering therapy to prescribe due to side effects, and counselling and reassuring patients with niacin treatment over the years has made its use challenging…so it is not the outcomes that is the disincentive to prescribing niacin or Niaspan.

  3. Thanks for your input. In my (limited) experience so far, I haven’t had many patients on niaspan. Do you have any tips for counseling patients on adherence to niacin? Do you also recommend additional medications (fish oil) for HDL therapy?