摘要 目的:探索甲状腺结节患者预后与超声多模态影像特征、血清学指标及中医证候评分的相关性。方法:回顾性分析2023年6月至2024年6月期间200例甲状腺结节患者,采用超声弹性成像、彩色多普勒等技术获取结节参数,检测血清游离甲状腺素(FT4)、游离三碘甲状腺原氨酸(FT3)水平,依据《中医内科学》进行中医证候评分。通过Spearman相关性分析、二元logistic回归及ROC曲线评估各指标与预后的关联。结果:预后不良组的结节形态不规则、边界模糊比例更高,弹性模量值(横断面)、中医证候评分、FT4水平显著升高(P<0.05)。Spearman分析显示上述指标与预后不良呈正相关(r=0.398~0.575, P<0.001)。Logistic回归确定形态不规则、边界模糊、中医证候评分、FT4、横断面平均值为独立危险因素(P<0.05)。ROC曲线显示横断面平均值(AUC=0.926)、中医证候评分(AUC=0.901)及联合预测模型(AUC=0.962)具有较高诊断效能。结论:超声多模态影像特征、血清FT4及中医证候评分对甲状腺结节预后具有较高评估效能,联合模型可更精准预测预后。
ABSTRACT Objective: To explore the correlation between the prognosis of patients with thyroid nodules and the characteristics of multimodal ultrasound imaging, Serological test indicator and TCM syndrome scores. Methods: A retrospective analysis was conducted on 200 patients with thyroid nodules collected between June 2023 and June 2024. Ultrasound elastography and color Doppler for nodule parameters serum free thyroxine (FT4), free triiodothyronine (FT3) level detection, and traditional Chinese medicine syndrome score evaluation based on internal Medicine of Traditional Chinese Medicine were performed. The associations between various indicators and prognostic outcomes were assessed using Spearman correlation analysis, binary logistic regression, and ROC curve analysis. Results: The poor prognosis group exhibited significantly higher proportions of irregular nodule morphology and blurred margins, along with significantly elevated values of elastic modulus (transverse plane average and longitudinal plane average), TCM syndrome scores, and FT4 levels (P<0.05). Spearman analysis revealed positive correlations between these indicators and poor prognosis (r=0.398-0.575, P<0.001). Logistic regression identified irregular morphology, blurred margins, TCM syndrome score, FT4, and transverse plane average as independent risk factors (P<0.05). ROC curve analysis demonstrated high diagnostic performance for transverse plane average (AUC=0.926), TCM syndrome score (AUC=0.901), and the combined predictive model (AUC=0.962). Conclusion: Multimodal ultrasound features, serum FT4 levels, and TCM syndrome scores demonstrated good predictive performance for thyroid nodule prognosis. The combined model significantly improved the predictive accuracy.