Association between valproate treatment for acute phase schizophrenia and risk of new onset hypothyroidism

Abstract Aim To assess the association between acute phase treatment by valproate as an adjunctive drug and risk of new onset hypothyroidism in a large cohort of patients affected by schizophrenia. Methods We conducted a retrospective cohort study in a psychiatric hospital in China between January 2016 and December 2018. We obtained approval from the Ethics Committee of the study hospital prior to the commencement of the study. Patients who were diagnosed with schizophrenia and admitted to the study hospital during the study period with thyroid function tests at admission and during hospitalization were included. Patients with abnormal thyroid function at admission were excluded. Hypothyroidism, defined as TSH>4.2 mU/L or on L-thyroxine treatment, was the primary outcome. The primary exposure was adjunctive valproate plus atypical antipsychotics (AAPD), the secondary exposure was lithium plus AAPD and the comparison group was AAPD only. Adjusted relative risk (RR) and 95% confidence interval (CI) were estimated by log-binomial model to assess the independent association between valproate treatment and risk of hypothyroidism. Results A total of 1622 eligible patients were included the final analysis. Rate of new onset hypothyroidism was 10.7% and 20.9% in AAPD only and valproate plus AAPD groups, respectively. Adjusted RR (95% CI) for valproate plus AAPD was 1.85 (1.44–2.38), with AAPD only group as reference. Similarly, adjusted RR (95% CI) for lithium plus AAPD was 1.93 (1.32–2.69). Conclusion Similar with lithium, valproate as adjunctive drug is associated with increased risk of new onset hypothyroidism during acute phase treatment for schizophrenia.