Background and objective: To identify plasma homocysteine concentrations that could be taken as normal values in our population, and to measure hyperhomocysteinemia prevalence.
Subjects and method: Cross-sectional study performed in all subjects attending or working in a primary health care center for any reason. Information was collected about personal history, cardiovascular risk factors and socio-demografic variables, and plasma homocysteine levels and other biochemical parameters were measured. Distribution of homocysteine concentration was compared in individuals with a history of cardiovascular disease and in a gold-standard population (healthy subjects with normal serum concentrations of B-vitamins). The value of homocysteine concentration chosen as cut-off value was the one that offered an optimal sensitivity/specificity ratio in ROC curves derived from logistic regression models.
Results: 1,636 subjects (51% female and 49% male, mean age 45 [16.3]) were included in the study. Mean plasma homocysteine concentration was 10.7 (4.1) y 8.5 (2.9) micromol/L in men and women, respectively (p < 0.01). Homocysteine levels that best discriminated between cardiovascular disease and gold-standard populations were 10.85 micromol/L in men (sensitivity 58%, specificity 68%), and 9.57 micromol/L in women (sensitivity 50%, specificity 81%). 31.4% of the population (95% CI, 29.1-36.6) presented homocysteine values above these levels.
Conclusions: Hyperhomocysteinemia is not a rare condition in our population. The predictive values obtained imply that measurement of serum homocysteine should be performed only in conjunction with measures of other cardiovascular risk factors. Further research should analyze if homocysteine adds predictive power in cardiovascular risk stratification.