Vitamin B₁₂ declines in patients taking metformin

Taking metformin increased the chances of developing vitamin B₁₂ deficiency, according to a new study of type 2 diabetics on insulin.

The Dutch trial randomized 390 patients to receive either 850 mg of metformin or a placebo three times a day for 4.3 years. Their levels of vitamin B₁₂, folate and homocysteine were measured at baseline and 4, 17, 30, 43 and 52 months. The metformin group showed a persistent, progressive decrease in B₁₂ concentrations, with a mean decrease of 19% compared to the placebo group. At the end of the study, patients taking metformin were 7% more likely to be B₁₂-deficient (less than 150 pmol/L) and 11% more likely to have a low level (15 to 220 pmol/L).

The metformin patients also had lower folate concentrations than placebo recipients, although both groups showed an overall increase and the difference between groups was eliminated after adjustments for body mass index and smoking. There was also an increase in homocysteine levels in the metformin group that was not statistically significant across the whole group, but showed that patients who were B₁₂-deficient had higher concentrations of homocysteine (a mean of 23.7 µmol/L compared to 14.9 µmol/L in patients with normal B₁₂ levels). The study was published online by BMJ.

Based on the findings, study authors concluded that regular measurement of B₁₂ concentrations should be considered for patients on long-term metformin treatment. They noted that deficiencies are relatively easy and cheap to treat effectively and safely. However, an editorial that accompanied the study noted several questions that should be answered before such screening becomes routine practice: The study did not determine whether the patients with low levels of B₁₂ suffered any adverse effects of their deficiency, or whether these findings apply to type 2 diabetics not treated with insulin.

It's also possible that simple dietary counseling could solve the problem of B₁₂ deficiency, the editorialists said. If counseling is found to be insufficient, then researchers should conduct randomized trials that screen for deficiency among all type 2 diabetics (not just those on insulin) and look at outcomes and costs of screening and treatment.