The long-term health effects of radioactive iodine (RAI) and antithyroid drug (ATD) treatments compared with surgery for hyperthyroidism remain uncertain.
To compare solid cancer mortality rates associated with RAI and ATD treatments vs surgical management for hyperthyroidism.
This multicenter cohort study assessed patients treated for hyperthyroidism from January 1, 1946, to December 31, 1964, with follow-up through December 31, 2014. Data analysis was performed from August 1, 2019, to April 23, 2020.
Management with RAI, ATDs, surgical intervention, or combinations of these treatments.
Comparisons of solid cancer mortality rates in each treatment group with expected rates from the general population were assessed using standardized mortality ratios (SMRs), and internal comparisons were assessed using hazard ratios (HRs) adjusted for age, sex, and underlying diagnosis (Graves disease or toxic nodular goiter).
Of 31 363 patients (24 894 [79.4%] female; mean [SD] age, 46.9 [14.8] years) included in the study, 28 523 (90.9%) had Graves disease. The median follow-up time was 26.0 years (interquartile range, 12.3-41.9 years). Important differences in patient characteristics existed across treatment groups at study entry. Notably, the drug-only group (3.6% of the cohort) included a higher proportion of patients with prior cancers (7.3% vs 1.9%-4.0%), contributing to an elevated SMR for solid cancer mortality. After excluding prior cancers, solid cancer SMRs were not elevated in any of the treatment groups (SMR for surgery only, 0.82 [95% CI, 0.66-1.00]; SMR for drugs only, 0.90 [95% CI, 0.74-1.09]; SMR for drugs and surgery, 0.88 [95% CI, 0.84-0.94]; SMR for RAI only, 0.90 [95% CI, 0.84-0.96]; SMR for surgery and RAI, 0.66 [95% CI, 0.52-0.85]; SMR for drugs and RAI, 0.94 [95% CI, 0.89-1.00]; and SMR for drugs, surgery, and RAI, 0.85 [95% CI, 0.75-0.96]), and no significant HRs for solid cancer death were observed across treatment groups. Among RAI-treated patients, HRs for solid cancer mortality increased significantly across levels of total administered activity (1.08 per 370 MBq; 95% CI, 1.03-1.13 per 370 MBq); this association was stronger among patients treated with only RAI (HR, 1.19 per 370 MBq; 95% CI, 1.09-1.30 per 370 MBq).
After controlling for known sources of confounding, the study found no significant differences in the risk of solid cancer mortality by treatment group. However, among RAI-treated patients, a modest positive association was observed between total administered activity and solid cancer mortality, providing further evidence in support of a dose-dependent association between RAI and solid cancer mortality.

Author