Abstract
PURPOSE: This study was performed to examine the association between preoperative thyroid ultrasound (US) features and the risk of developing hypothyroidism after hemithyroidectomy. METHODS: Patients who underwent hemithyroidectomy between March 2020 and June 2023 were retrospectively included. Postoperative hypothyroidism was defined as a continued need for thyroxine 2 years after surgery. To assess US features suggestive of diffuse thyroid disease, two radiologists evaluated the echogenicity, echotexture, and vascularity of the thyroid parenchyma on preoperative thyroid US images. A US thyroid parenchymal score was defined by summing the scores for these parameters, and a weighted score was also developed using β-coefficients from a multivariable logistic regression model. Model performance was evaluated using the area under the receiver operating characteristic curve (AUC) and integrated discrimination improvement (IDI). RESULTS: Overall, 398 patients were included, of whom 191 (47.9%) developed postoperative hypothyroidism. Preoperative thyroid-stimulating hormone (TSH), body surface area-adjusted remnant thyroid volume, histologic evidence of thyroiditis, and both unweighted and weighted US thyroid parenchymal scores were identified as significant predictors of postoperative hypothyroidism. The unweighted and weighted US scores were statistically significant predictors in the multivariable analysis (unweighted: odds ratio [OR], 1.36 [95% confidence interval (CI), 1.04 to 1.78]; weighted: OR, 3.05 [95% CI, 1.22 to 7.62]). Although adding US scores did not significantly improve the AUC, it led to significant improvements in IDI (P=0.04 and P=0.03, respectively). CONCLUSION: US assessment of the thyroid parenchyma served as an independent predictor of postoperative hypothyroidism and improved discrimination of predicted probabilities based on preoperative TSH.