Author + information
The editorial by Lane et al. (1), which cast doubt on the relationship between depression and myocardial infarction (MI), was inappropriately pessimistic and looked only at part of the evidence. Their argument appears to be that in post-MI patients any apparent relationship between depression and outcome is an epiphenomenon and that the only causal relationship is between disease severity and outcome. Certainly, if all we had were post-MI data on which to base the “depression–heart disease” hypothesis, our case would be weak because the potentially confounding effect of an MI causing or exacerbating depression is hard to dismiss.
The strongest support for the hypothesis comes from a diverse series of prospective studies showing that depression in asymptomatic and apparently healthy subjects is a strong and independent predictor of MI. A recent meta-analysis identified 10 prospective studies (2), where the overall relative risk for depression as an independent risk factor for coronary events was 1.64, a risk between that of passive smoking (n = 1.25) and active smoking (n = 2.5). Nine of the 10 studies obtained positive results. Considering that depression is much harder to quantify than smoking behavior (nine different measures of depression were used), that depression was evaluated only once, and that the average follow-up between the evaluation of depression and the outcome was on average 13.6 years (up to 40 years in one case), this is a truly impressive result.
The argument by Lane et al. that the reason why some of the studies in post-MI patients were positive is that the studies did not adequately control for confounding variables is also open to criticism. The fact that controlling for one variable eliminates the significance of a second one does not necessarily mean that the second factor is unimportant, because it may operate through the first one. Thus, it is conceivable that if we had very good measures of the extent of atherosclerotic plaque, the role of blood cholesterol in causing MI would be “controlled for” by the plaque burden. Would this lead us to conclude that cholesterol is not a risk factor? Clearly, the issue here is which comes first; that is why the prospective studies of disease-free subjects are so important.
What the depression–heart disease hypothesis sorely needs in order to become established or refuted are more observational and interventional studies. In addition to coronary artery disease severity, variations in patient populations and differences in when and how depression was assessed have been other explanations for why some depression–heart disease studies have been negative. We should not forget that many of the early intervention studies attempting to test the lipid–heart disease hypothesis were negative. To date, ENRICHD is the only published study that attempted to reduce recurrence rates by treating depression, and its negative results may well be due to an inadequate treatment effect, as was observed with the lipid-lowering arm of ALLHAT (3).
- American College of Cardiology Foundation
- Lane D.,
- Carroll D.,
- Lip G.Y.H.
- Wulsin L.R.,
- Singal B.M.
- ↵Major outcomes in moderately hypercholesterolemic hypertensive patients randomized to pravastatin vs usual care: the Antihypertensive and Lipid-Lowering Treatment to Prevent Heart Attack Trial (ALLHAT-LLT). JAMA 2002;288:2998–3007