Determinants of diagnostic delay in thyroid storm: A 12-year single-center cohort study

Abstract Objective Thyroid storm is a life-threatening endocrine emergency. Although predictors of mortality have been described, determinants of diagnostic delay remain unclear. We aimed to identify factors associated with diagnostic delay in thyroid storm and its relationship with severe clinical outcomes. Methods We conducted a retrospective cohort study of adults admitted with thyroid storm at a tertiary-care hospital over 12 years. The primary endpoint was time from symptom onset to diagnosis (days). Factors associated with time to diagnosis were evaluated using multivariable linear regression adjusting for predefined covariates, including age, initial onset versus recurrence, admitting department, presenting symptoms, and cardiac rhythm abnormalities. Secondary analyses examined delayed diagnosis (≥7 days) using multivariable logistic regression. Additional analyses separated diagnostic delay into patient-related and in-hospital components and evaluated admission pathways and referral status. Results Ninety-one patients were included. Median time to diagnosis was 5 days (interquartile range 3.5–7). Older age (1.08 days per 10-year increase; p0.001), initial onset (4.10 days; p0.001), and admission to non-emergency departments (2.76 days; p=0.004) were independently associated with longer diagnostic delay, whereas symptom patterns and thyroid hormone levels were not. Patient-related delay did not differ by admission pathway, but in-hospital delay was longer among non-emergency admissions. Delayed diagnosis was associated with increased need for intensive care and organ support. Conclusion Diagnostic delay in thyroid storm appears driven mainly by clinical pathways rather than symptom patterns or biochemical severity. Earlier consideration of thyroid storm and prompt thyroid testing—particularly at initial onset and in non-emergency settings—may improve outcomes.