Comparative Safety of Therapies for Hyperthyroidism Planning Pregnancy: A Network Meta-analysis

Objective: Our intention was to compare the pregnancy safety in women with hyperthyroidism seeking future pregnancy after ATD, RAI or thyroidectomy treatment and to explore the optimum treatment strategy. We hope the results will guide the choice for endocrinologists and patients, and be beneficial for promoting eugenics. Methods: We searched multiple databases though December 2019. The outcome indicators were abortion, preterm birth, IUGR and birth defect. We conducted a frequency-framework network meta-analysis by using Stata and R software. The results of the intervention comparison were expressed as OR with 95%CI, and ranking was assessed using surface under the cumulative ranking (SUCRA) probabilities. Results: The network meta-analysis included 4 retrospective cohort studies with a total enrollment of 480 patients. 1)Thyroidectomy had a lower risk of abortion than RAI [OR=0.77, 95%CI (0.23, 2.56)] and ATD [OR=0.68, 95%CI (0.21, 2.21)]. RAI had a lower risk of abortion than ATD [OR=0.88, 95%CI (0.45, 1.75)]. Based on SUCRA results, thyroidectomy (0.698) was followed by RAI (0.494), ATD (0.308). Thyroidectomy (60.7%) had the highest probability of being first compared to RAI (27.0%) and ATD (12.4%). 2)Thyroidectomy had a lower risk of preterm birth than RAI [OR=0.80, 95%CI (0.26, 2.44)] and ATD [OR=0.79, 95%CI (0.39, 1.59)]. RAI had a lower risk of preterm birth than ATD [OR=0.98, 95%CI (0.42, 2.33)]. Based on SUCRA results, thyroidectomy (0.703) was followed by RAI (0.430), ATD (0.367). Thyroidectomy (57.8%) had the highest probability of being first compared with RAI (30.5%) and ATD (11.8%). 3)Thyroidectomy had a lower risk of IUGR than RAI [OR=0.28, 95%CI (0.03, 3.02)] and ATD [OR=0.83, 95%CI (0.14, 4.86)]. RAI had a higher risk of IUGR than ATD [OR=3.02, 95%CI (0.60, 15.27)]. Based on SUCRA results, thyroidectomy (0.717) was followed by ATD (0.663), RAI (0.120). Thyroidectomy (56.6%) had the highest probability of being first compared with RAI (5.4%) and ATD (37.9%). 4)Thyroidectomy had a lower risk of birth defect than RAI [OR=0.70, 95%CI (0.02, 30.34)] and ATD [OR=0.23, 95%CI (0.01, 4.52)]. RAI had a lower risk of birth defect than ATD [OR=0.32, 95%CI (0.03, 3.12)] (Table 2). Based on SUCRA results, thyroidectomy (0.70) was followed by RAI (0.629), ATD (0.171). Thyroidectomy (55.9%) had the highest probability of being first compared with RAI (41.3%) and ATD (2.8%). Conclusion: Thyroidectomy was the optimum option for women with hyperthyroidism seeking near future pregnancy. The future research direction is to include more samples to conduct head-to-head randomized controlled trials or prospective cohort studies, establish inclusion criteria for various pre-pregnancy conditions or further subgroup analysis, and develop more acceptable, safer, and more manageable treatments that allow for the remission of both thyroid function and autoimmune abnormalities.