Drs. Tanskanen and Viinamäki Reply

To the Editor: We thank Drs. Terao and Whale for their comments. We admit that the cholesterol/depression/suicide issue is complicated and will remain controversial. At the time our report was published, a discussion with Dr. Partonen and his colleagues had been published elsewhere (1); thus, we do not repeat the entire debate here.

The findings of Dr. Terao and colleagues (Terao et al., 2000) are not convincing. They argued that the group of their cross-sectional study subjects who had the lowest cholesterol levels were more likely to be in a depressive state than those in the other cholesterol groups. However, neither the age- and gender-adjusted nor all adjusted odds ratios were significant in logistic regression analyses. Furthermore, Dr. Terao et al. did not adjust for factors such as smoking, which is an important risk factor in depression and suicide (2).

As Drs. Terao and Whale correctly point out, neurobiologic reports may be difficult to interpret. Serum cholesterol levels are not unanimously associated with serotonergic function. Hibbeln and his co-workers (3) did not find significant correlation between levels of plasma cholesterol and CSF 5-hydroxyindoleacetic acid, a metabolite of serotonin. They further postulated that highly unsaturated essential fatty acids may play some role. In fact, there is recent further evidence that this is true. The infrequent consumption of fish, which is the major source of omega-3 polyunsaturated fatty acids in the human diet, has been shown to be associated with a higher risk of depressive symptoms in the general population (4).

Finally, we would like to remind readers of our major findings. First, the high serum total cholesterol concentration (8.00 mmol/liter or greater, i.e., 309.4 mg/dl) was associated with nearly a two-and-a-half-fold risk of violent suicide. Second, the violent/nonviolent suicide ratio increased linearly from the lowest to the highest cholesterol category, climbing from 1.00 to 4.50. We suggest that the neurobiological background is different for violent and nonviolent suicides. We hope that future research will clarify the picture.