Objective: In the last few years a number of studies have found that metformin has a thyrotropin (TSH)-lowering effect in patients with type 2 diabetes (DM2) and that this effect disappears after discontinuation of the drug. Initial studies were carried out in hypothyroid patients on levothyroxine (LT4) treatment. However, this effect was observed only when serum TSH levels were elevated. The aim of this study was to assess the effect of metformin treatment on serum TSH levels in euthyroid DM2 patients.
Design: A retrospective study was conducted in seven primary health care centers in Spain. The study included 278 DM2 patients (110 females). Serum TSH level was tested before and one year after the onset of metformin treatment. We compared both TSH levels by paired t-tests. The sample was divided into four hierarchical clusters (C1 to C4), according to the Euclidean distance of the initial TSH level. No patient received LT4 treatment.
Results: Pre-metformin serum TSH level (2.00 ± 0.76 mU/L) was statistically significantly lower than post-metformin level (2.20 ± 0.87 mU/L; P <0.001). Interestingly, this trend was observed only in clusters 1 and 2 (those with lower baseline TSH levels), whereas the trend was lost in clusters 3 and 4. The difference between pre- and post-metformin TSH levels follows a mathematical model. According to this model, the TSH threshold point level is 3.00 mU/L.
Conclusion: Metformin seems to induce a "buffer effect" on TSH secretion in euthyroid diabetic individuals. Hence, our study found an unreported TSH behavior after metformin treatment.