Homocysteine Lowering and Cognitive Performance

Abstract

To the Editor: McMahon et al. (June 29 issue)1 report that lowering plasma homocysteine concentrations by means of treatment with B vitamins did not improve the cognitive performance of healthy elderly volunteers with elevated homocysteine concentrations. Yet in the study population, there was little need to decrease homocysteine concentrations, no cognitive impairment to improve, and no association between elevated homocysteine concentrations and cognitive impairment to dissolve. What does it mean to improve an already quite normal cognitive performance and a mean score on the Mini–Mental State Examination of 29.2, given a possible score of 30? How can one define a clinically meaningful difference between two groups of healthy elderly volunteers? Whereas the usual definition of the upper limit of the normal range is 15 μmol per liter,2-4 McMahon et al. defined as “elevated” homocysteine concentrations as low as 13 μmol per liter. Thus, a good 62% of their healthy volunteers had elevated homocysteine concentrations, as compared with a reported prevalence of hyperhomocysteinemia of 18% among elderly New Zealand women.5 The already small sample was thus diluted by the large presence of elderly persons who were falsely classified as having hyperhomocysteinemia. Consistently, at baseline, the mean plasma folate and vitamin B12 concentrations were already rather high. Moreover, after baseline, the mean homocysteine concentration in the placebo group was below or near the upper limit of the normal range throughout the trial. Ugo Lucca, M.Sc. Mauro Tettamanti, Ph.D.