Table 3:

Crude and adjusted hazard ratios for low thyroid-stimulating hormone levels (< 0.4 mIU/L) associated with the use of metformin monotherapy, compared with sulfonylurea monotherapy, among 59 937 euthyroid patients with type 2 diabetes

VariableNo. events of low TSH levelPerson-years of exposureIncidence rate, per 1000 person-years (95% CI)Crude HRAdjusted HR* (95% CI)
Sulfonylurea, n = 7980528 5766.1 (4.5–8.0)1.001.00 (reference)
Metformin, n = 51 95727063 0474.3 (3.8–4.8)0.710.97 (0.69–1.36)
Duration of use
< 90 d
 Sulfonylurea131 6667.8 (4.2–13.3)1.001.00 (reference)
 Metformin4911 4904.3 (3.2–5.6)0.550.77 (0.41–1.44)
90–180 d
 Sulfonylurea61 1825.1 (1.9–11.0)1.001.00 (reference)
 Metformin448 3105.3 (3.8–7.1)1.041.45 (0.61–3.47)
> 180 d
 Sulfonylurea335 7285.8 (4.0–8.1)1.001.00 (reference)
 Metformin17743 2484.1 (3.5–4.7)0.720.96 (0.64–1.44)
  • Note: CI = confidence interval, HR = hazard ratio, TSH = thyroid-stimulating hormone.

  • * Adjusted for age, sex, year of cohort entry, body mass index, smoking, excessive alcohol use (i.e., alcohol-related disorders such as alcoholism, alcoholic cirrhosis, alcoholic hepatitis and hepatic failure), hemoglobin A1C, duration of diabetes, and use of lithium, amiodarone and glucocorticoids.

  • Hazard ratios were estimated using time-dependent Cox proportional hazards models.