Efficacy Assessment and Influencing Factors on Superb Microvascular Imaging (SMI) Microflow Patterns in Solid Thyroid...

This study aims to evaluate the diagnostic efficacy of Superficial Microvascular Imaging (SMI) in differentiating benign from malignant thyroid nodules and to identify influencing factors such as patient body mass index (BMI), tumor size, and nodule depth.

A retrospective analysis was conducted involving 560 patients with 783 pathologically confirmed thyroid nodules between January 2020 and July 2023. Patient demographics, including age, sex, and BMI, were recorded. Thyroid nodules were evaluated using conventional ultrasonography and classified by ACR TI-RADS. Nodule size and depth were measured. Subsequently, Superb Microvascular Imaging (SMI) was performed to assess vascular patterns, classified into six types (I-VIb). Diagnostic performance metrics, including sensitivity, specificity, accuracy, positive predictive value (PPV), and negative predictive value (NPV) were calculated based on histopathological results. Logistic regression analyses identified independent predictors of malignancy, while the impact of BMI, tumor size, and nodule depth on SMI performance was analyzed.

Among the 783 thyroid nodules analyzed, 335 were benign and 448 were malignant. Statistically significant differences were observed between benign and malignant nodules in terms of patient age (p < 0.001), sex (p = 0.032), nodule size (p = 0.025), ACR TI-RADS level (p < 0.001), and vascular distribution on SMI (p < 0.001). Multivariate logistic regression analysis identified several independent predictors of malignancy in thyroid nodules. Increasing age (OR: 0.92, 95% CI: 0.90-0.95, p < 0.001) and female sex (OR: 0.36, 95% CI: 0.18-0.70, p = 0.003) were associated with significantly lower odds of malignancy. Among ACR TI-RADS levels, TR5 emerged as a robust independent predictor of malignancy (OR: 49.94, 95% CI: 16.49-151.20, p < 0.001). Regarding SMI vascular distribution, higher vascular types were strongly associated with an increased risk of malignancy. Specifically, Type III (OR: 8.55, 95% CI: 1.82-40.14, p = 0.007), Type IV (OR: 6.77, 95% CI: 2.68-17.12, p < 0.001), Type V (OR: 9.20, 95% CI: 3.69-22.91, p < 0.001), Type VIa (OR: 89.71, 95% CI: 20.36-395.25, p < 0.001), and Type VIb (OR: 220.39, 95% CI: 42.07-1154.47, p < 0.001) demonstrated a stepwise increase in odds ratios, with Type VIb showing the strongest correlation. The diagnostic performance of SMI was high, achieving a sensitivity of 89.5%, specificity of 85.4%, overall accuracy of 87.7%, positive predictive value (PPV) of 89.1%, negative predictive value (NPV) of 85.9%, and an area under the curve (AUC) of 0.896 (95% CI: 0.876-0.920). Notably, nodule depth significantly influenced diagnostic accuracy. Nodules with depths ≤7.5 mm demonstrated superior diagnostic performance compared to those >7.5 mm (DeLong's test; Z = 3.11, p = 0.0019), while tumor size and patient BMI did not correlate with SMI efficacy.

SMI is a promising diagnostic tool for thyroid lesions, demonstrating high diagnostic efficacy. The study highlights the significance of nodule depth as a critical factor influencing SMI's performance. These findings may enhance clinical risk assessment and management strategies for thyroid nodules, ultimately leading to improved patient outcomes.