Narrative
Statin medications are aimed at an enzyme in the cholesterol production pathway, and are therefore intended to reduce cholesterol. In some patients elevated cholesterol plays a role in the development of coronary artery disease and heart attacks, so the drugs are intended to reduce the chance of death by reducing heart attacks and strokes.
These data examine the effect of statins for people who have never before had a heart attack or stroke (most of the people who currently take statins). The effectiveness of the statins appears to be reproducible across studies in this group—they do lower cholesterol in most people who took them. But very few people will avoid a heart attack or stroke by virtue of this change. It takes 5 years of daily statin therapy to achieve a 1.6% chance of avoiding a heart attack, and a 0.37% chance of avoiding a stroke. Most disappointing, statins seem unable to prevent death in this group. And most concerning, the drugs may increase diabetes, a serious and life-altering disease.
Caveats
There is controversy about whether mortality is reduced by statins in this group. We do not believe so, but are aware that others interpret these data differently. The analyses suggesting mortality benefits in low risk patients are performed by a single group and they are done in a fashion we do not agree with. Specifically, the CTT group consistently does their comparisons of clinical effects of statin and placebo by using a per-cholesterol-reduction metric. In other words, rather than simply comparing statin and placebo groups head-to-head (as the Therapeutics Initiative and the Ray meta-analysis have done, finding no mortality benefit), the CTT group measures statin effects based on how much cholesterol reduction is achieved. We believe this confounds the analysis and has the potential to advantage statin groups by narrowing the comparison to those whose cholesterol is successfully reduced. It is not clear to us why anyone would choose to promulgate this type of analysis rather than simply presenting comparative results from the two groups, side by side, and we are concerned that it may distort the comparison. In simple comparisons no benefit is found. We feel comfortable that either there is no true mortality benefit in patients without CAD, or that the benefit is too small to be universally agreed upon.
The harms of statins are less publicized than benefits, but are well documented. A recent narrative review of statin myopathy suggests that 10% is a relatively conservative estimate for this side effect, which may be a primary contributor to high rates of drug discontinuation. An additional, more concerning side effect is statin-induced diabetes, as noted in the JUPITER study and a large Women’s Health Initiative cohort, studies that best represent primary prevention cohorts. We use an absolute risk estimate for statin-induced diabetes with the understanding that this may underestimate risk for many. Baseline risk of diabetes is likely the greatest driver of risk for statin-induced diabetes, and the JUPITER trial enrolled patients at lower risk (2.4%) than many, perhaps most, patients who take statins. Our calculation presumes this low baseline risk and that the aggregate chance of diabetes is time-dependent and linear. We have thus extrapolated from the 1.9-year trial data to a 5-year endpoint, the same time endpoint and calculation used for benefits. These data are relevant to new onset diabetes only, and do not address the equally concerning possibility that some diabetic patients may experience statin-induced worsening of their disease, or inability to manage or cure their disease using lifestyle changes. We look forward to data addressing these issues.
Our sense, based on factors noted above, is that the benefits of these drugs are likely exaggerated partly by an unconscious 'hope bias' on the part of readers and authors. This is common in the early literature on therapeutic innovations, and is often attenuated as further literature emerges. In addition, post-marketing surveillance data is in its early stages with this class of drugs and preliminary reports suggest that cognitive decline, tendonopathies, and other side effects may emerge in future literature. Finally, the source of the great majority of these data is industry, which has a spotty history of integrity in trial data reporting, suggesting these data to be a best-case scenario.
In summary, whether statins are an appropriate choice for primary prevention may be best left to individual preferences, and debate continues. For now we have labeled the utility of statins in primary prevention as 'Red' based on certain value judgments. We believe that benefits are best-case and harms may well be underestimated. We also believe that diabetes, a chronic condition with serious long term morbidity, is more important to avoid for most patients than a single event such as a nonfatal heart attack or stroke. Finally, we believe that lifestyle interventions such as Mediterranean diet are substantially more powerful than statin medications in achieving cardiovascular benefits, and come without harms.
For our new onset diabetes calculation we used ASCAPS/TexCAPS and JUPITER data, the same studies in the Cochrane calculation. However, the Cochrane calculation did not adjust for exposure time. In other words, the JUPITER study was stopped at two years whereas the ASCAPS study stopped after 5 years, and the JUPITER data must be carried forward presumptively to derive a risk adjustment for exposure to the drug.
The raw numbers of 270 and 216 new onset diabetes cases from 24 months of exposure to a statin and a placebo (respectively) can be extrapolated, assuming that increased diabetes risk is likely to continue linearly with exposure. This yields 675 and 540 cases at 5 years. Combining these numbers with the AFCAPS numbers yields a relative risk of 1.81 for development of diabetes (not 1.18 as reported by Cochrane). This risk increase translates to an absolute risk of 4%, or 1 in 25.
This may, however, be flawed in its assumption of a linear increase. Perhaps the increase in diabetes risk is frontloaded, and does not rise above that found at two years. In this case the risk increase would be 1.18 as suggested by the Cochrane group, and the absolute risk increase in this case would be 0.43%, or 1 in 233. The answer likely lies between 0.4% and 4%, and we have chosen what we believe to be a conservative estimate of 2% as a midway point in this credible interval.