Background/Purpose: Increase in thyroid stimulating hormone (TSH) levels over upper normal limit has been reported in a small percentage of patients treated with febuxostat, but a mechanistic explanation is not yet available.

Methods: In a case-control design study, we tested TSH levels in patients with gout at baseline with that every 6-month during follow-up during treatment with febuxostat. Patients to be started allopurinol and an available measurement of TSH 6 months prior to baseline evaluation were used as controls, a follow-up TSH level being measured at 12-month follow-up. Patients with abnormal TSH levels or previous thyroid disease were not included.

Results: Eighty-eight patients treated with febuxostat and 87 with allopurinol were available for comparison. Patients with febuxostat had more severe hyperuricemia and gout, renal impairment, and previous treatment, but were similar in other characteristics. A similar increase in TSH levels was observed in both groups (0.4 and 0.5 µUI/mL for febuxostat and allopurinol, respectively). At 12-mo, 7/88 (7.9%) of patients on febuxostat and 4/87 (3.4%) of patients on allopurinol showed TSH levels over 5.5 µUI/mL, but the upper quartiles for baseline TSH distribution were higher in patients on febuxostat (Table). Doses prescribed (corrected for estimated glomerular filtration rate in patients on allopurinol) and baseline TSH levels were determinants of TSH levels at 12-month follow-up (Figures). No impact on clinical status or free T4 levels was observed in patients with TSH levels over the normal limit.

Table. TSH levels at baseline evaluation and 12-month follow-up. Doses for allopurinol are shown as prescribed and corrected per dL of estimated glomerular filtration rate.

Conclusion: both allopurinol and febuxostat increase TSH levels in a dose-dependent, with no apparent impact on free T4 levels, suggesting a class (xanthine-oxidase inhibition) effect with no apparent impact on free T4 levels.