Background: The long-term risk of hypothyroidism following postpartum thyroid dysfunction (PPTD) is uncertain. Most previous studies have been small, short-term or have lacked a control group.
Objective: To ascertain the long-term risk of hypothyroidism in women following PPTD. Design and participants A 12-year longitudinal study of 409 women (including 71 with PPTD) who previously participated in a PPTD prevalence study.
Measurements: The primary outcome measure was hypothyroidism (defined as TSH greater than 4 mU/l or on thyroxine replacement) at follow-up. Outcomes in women with and without PPTD were compared by logistic regression. Receiver operating characteristic analysis was used to determine the optimal cut-off for baseline TSH as a predictor of hypothyroidism in the cohort.
Results: At follow-up, hypothyroidism was present in 27 of 71 women who had PPTD at baseline (38%) and 14 of 338 women without PPTD (4%). From multivariate analysis, odds ratios (with 95% confidence intervals) for hypothyroidism were - 4.8 (1.6, 14.1) for PPTD; 8.2 (2.8, 24.6) for positive thyroid peroxidase antibodies (TPOAb); 9.7 (2.6, 37.0) for the hypothyroid phase of PPTD and 51.4 (19.2, 137.5) for hypothyroid PPTD with positive TPOAb. A baseline TSH above 2.6 mU/l was the optimal cut-off for predicting hypothyroidism (sensitivity 76%, specificity 86%).
Conclusions: PPTD is a strong predictor of hypothyroidism in the long-term. Women who present with postpartum hypothyroidism or have positive TPOAb are at particularly high risk, suggesting that close long-term follow-up is advisable if thyroxine replacement is not instituted at an early stage.