Vitamin B12 and folate levels and lithium administration in patients with affective disorders.

Background: It is unclear whether there is a relationship between lithium administration and vitamin B12 metabolism.

Methods: We compared serum B12, serum folate, and red blood cell folate concentrations in patients receiving and not receiving lithium at two Mood Disorders Clinics. As the two centers differed in vitamin assay methods, data were first analyzed separately and then combined. To rule out an in vitro effect of lithium on the assays, we also added varying amounts of lithium to lithium-free blood samples and measured vitamin concentrations.

Results: Mean serum B12 concentrations were approximately 20% lower in the lithium than in the nonlithium group at each center. This difference was statistically significant for each center and on combination (two-tailed p = .017, .021, and .0009). The parametric effect size for each center and the combined weighted mean effect size were moderate in magnitude (.605, .523, and .565). There was a nonsignificant trend toward an increased prevalence of assay-defined B12 deficiency in the lithium group at one center only, with no cases in either group at the other center and a nonsignificant combined relative risk.

Conclusions: Our data may represent a lithium-associated decrease in serum B12 concentration. The clinical significance of these findings is not yet clear.