Context It is not known whether total thyroidectomy is more favorable than medical therapy for patients with amiodarone-induced thyrotoxicosis (AIT). Objective To compare total thyroidectomy with medical therapy on survival and cardiac function in AIT patients. Methods Observational longitudinal cohort study involving 207 AIT patients that had received total thyroidectomy (surgery group, n = 51) or medical therapy (medical therapy group, n = 156) over a 20-year period. AIT types and left ventricular ejection fraction (LVEF) classes were determined at diagnosis of AIT. Cardiac and thyroid function were reevaluated during the study period. Survival was estimated using the Kaplan-Meier method. Results Overall mortality and cardiac-specific mortality at 10 and 5 years, respectively, were lower in the surgery group than in the medical therapy group (P = 0.04 and P = 0.01, respectively). The lower mortality rate of the surgery group was due to patients with moderate to severely compromised LVEF (P = 0.005 vs medical therapy group). In contrast, mortality of patients with normal or mildly reduced LVEF did not differ between the 2 groups (P = 0.281 and P = 0.135, respectively). Death of patients with moderate to severe LV systolic dysfunction in the medical therapy group occurred after 82 days (interquartile range, 56–99), a period longer than that necessary to restore euthyroidism in the surgery group (26 days; interquartile range, 15–95; P = 0.038). Risk factors for mortality were age (hazard ratio [HR] = 1.036) and LVEF (HR = 0.964), whereas total thyroidectomy was shown to be a protective factor (HR = 0.210). LVEF increased in both groups after restoration of euthyroidism, above all in the most compromised patients in the surgery group. Conclusions Total thyroidectomy could be considered the therapeutic choice for AIT patients with severe systolic dysfunction, whereas it is not superior to medical therapy in those with normal or mildly reduced LVEF.

Context Patients with amiodarone-induced thyrotoxicosis (AIT) and severely reduced left ventricular ejection fraction (LVEF) have a high mortality rate that may be reduced by total thyroidectomy. Whether in this subset of patients thyroidectomy should be performed early during thyrotoxicosis or later after restoration of euthyroidism is unsettled. Objectives Mortality rates, including peritreatment mortality and 5-year cardiovascular mortality, and predictors of death, evaluated by Cox regression analysis. Methods Retrospective cohort study of 64 consecutive AIT patients selected for total thyroidectomy from 1997 to 2019. Four groups of patients were identified according to serum thyroid hormone concentrations and LVEF: Group 1 (thyrotoxic, LVEF<40%), Group 2 (thyrotoxic, LVEF≥40%), Group 3 (euthyroid, LVEF<40%), Group 4 (euthyroid, LVEF≥40%). Results Among patients with low LVEF (Groups 1 and 3), mortality was higher in patients undergoing thyroidectomy after restoration of euthyroidism (Group 3) than in those submitted to surgery when still thyrotoxic (Group 1): peritreatment mortality rates were 40% vs 0%, respectively (p=0.048), whereas 5-year cardiovascular mortality rates were 53.3% vs 12.3%, respectively (p=0.081). Exposure to thyrotoxicosis was longer in Group 3 than in Group 1 (112 days, IQR 82.5-140, vs. 76 days, IQR 24.8-88.5, p=0.021). At variance, survival did not differ in patients with LVEF ≥40% submitted to thyroidectomy irrespective of being thyrotoxic (Group 2) or euthyroid (Group 4): in this setting, peritreatment mortality rates were 6.3% vs 4% (p=0.741) and 5-year cardiovascular mortality rates were 12.5% and 20% (p=0.685), respectively. Age (HR 1.104, p=0.029) and duration of exposure to thyrotoxicosis (HR 1.004, p=0.039), but not presurgical serum thyroid hormone concentrations (p=0.577 for free thyroxine, p=0.217 for free triiodothyronine), were independent predictors of death. Conclusions A prolonged exposure to thyrotoxicosis resulted in an increased mortality in patients with reduced LVEF, which may be reduced by early thyroidectomy.

No abstract