Abstract
Background/Objectives: To evaluate whether a TSH-receptor antibody (TRAb)-first, one-sample diagnostic strategy improves etiologic classification of overt hyperthyroidism compared with conventional pathways, and to assess its implications for imaging use, diagnostic accuracy, and cost efficiency. Methods: In this multicentre retrospective study, 274 adults with newly diagnosed overt hyperthyroidism underwent TRAb measurement, thyroid ultrasound, and scintigraphy during a single clinical encounter. Scintigraphy served as the functional reference standard. We compared the diagnostic performance of TRAb and ultrasound, modeled TRAb-first diagnostic algorithms, and estimated the potential impact of reflex TRAb testing on diagnostic workflow and resource use. Results: Graves' disease (GD) accounted for 65% of cases. TRAb showed excellent diagnostic accuracy for GD (sensitivity 92.0%, specificity 96.0%; κ = 0.86) and markedly outperformed ultrasound (sensitivity 66.9%, specificity 62.5%; κ = 0.43). A TRAb-first pathway in which TRAb-positive patients are directly classified as GD and TRAb-negative patients undergo scintigraphy achieved 100% sensitivity, 95.8% specificity, and the lowest overall misclassification rate. Replacing scintigraphy with ultrasound in TRAb-negative patients substantially reduced specificity (~60%) and yielded significant overdiagnosis of GD. Ultrasound identified numerous nodules but detected only one low-risk carcinoma (malignancy rate: 1.2%), suggesting limited oncologic yield. A TRAb-first strategy would have avoided two-thirds of scintigraphies and minimized unnecessary imaging. Conclusions: A TRAb-first diagnostic approach offers the most accurate, efficient, and clinically appropriate pathway for etiologic assessment of overt hyperthyroidism. Scintigraphy should be reserved for TRAb-negative patients, while ultrasound should be used selectively for structural evaluation rather than as part of routine etiologic work-up. Reflex TRAb testing may further streamline care by enabling rapid, one-sample etiologic diagnosis and reducing resource use.