OBJECTIVE: We analyzed trends in thyrotoxicosis hospitalizations with and without thyroid storm in the U.S. between 2016 and 2020. METHODS: We analyzed data from the national inpatient database using ICD-10 codes for thyrotoxicosis (E05). We compared demographics using X2 tests. Trends in hospitalization outcomes were assessed using Cuzick's test. Hospital costs were adjusted for inflation using the 2020 consumer price index. Odds of mortality and secondary outcomes were analyzed using multivariable logistic regression. Hospital stay in the 95th percentile was considered prolonged. RESULTS: 33,430 hospitalizations were analyzed. Hospitalization rates declined from 7,444 in 2016 to 5,424 in 2020 (Ptrend=0.002). Mortality rates increased for both hospitalizations without storm (10 [0.17%] in 2016 to 55 [1.30%] in 2020
Ptrend<
0.001) and with thyroid storm (10 [0.62%] in 2016 to 50 [4.15%] in 2020
Ptrend=0.051). There was an uptrend in prolonged hospitalization rates in the total study cohort (11.9% [3,978] to 14.6% [4,881]
Ptrend=0.030). Total hospital costs increased from 6,408 to 49,031: Ptrend<
0.001). A similar uptrend was observed with (5,343 to 9,321
Ptrend <
0.001) and without storm (4,066 to 2,703
Ptrend<
0.001). Thyroid storm was correlated with higher odds of major adverse cardiovascular events (MACEs) (aOR: 1.05
95% CI: 1.02-1.17
P=0.002) including acute heart failure (aOR: 1.15
95% CI: 1.03-1.78
P<
0.001) sudden cardiac death (aOR: 1.23
95% CI: 1.04-2.17
P=0.041), and atrial fibrillation (aOR: 1.17
95% CI: 1.05-2.06
P<
0.001). CONCLUSION: Hospitalization rates for thyrotoxicosis reduced while mortality rates significantly increased. There was an uptrend in healthcare costs, prolonged hospitalization, and the incidence of MACEs.