基于增强CT定量食管胃底静脉曲张体积预测胃食管静脉曲张Sarin分型的研究

Quantitative Assessment of Esophagogastric Variceal Volume on Contrast-Enhanced CT for Predicting Sarin Classification of Gastroesophageal Varices

摘要: 目的：探讨基于增强CT定量测量的食管胃底静脉曲张体积(esophagogastric variceal volume, EGVV)对胃食管静脉曲张Sarin分型的鉴别价值。方法：回顾性纳入2018年1月至2025年10月于安徽医科大学第二附属医院151例接受增强CT及上消化道内镜检查的肝硬化患者。根据胃镜检查结果及Sarin分型标准，将患者分为GOV1组(n = 107)和GOV2/3组(n = 44)。基于增强CT门静脉期图像，通过三维重建方法测量EGVV，并收集所有患者入院期间的实验室检测结果。采用单因素及多因素二元Logistic回归分析筛选与GOV2/3分型相关的独立因素，并建立预测模型。采用受试者工作特征曲线评价模型的鉴别能力。结果：与GOV1组相比，GOV2/3组患者EGVV明显增高(P < 0.001)。多因素Logistic回归分析显示，EGVV是GOV2/3型胃食管静脉曲张的独立相关因素(OR = 1.353, 95% CI: 1.147~1.595, P < 0.001)，血小板计数亦与GOV2/3分型独立相关(OR = 1.011, 95% CI: 1.003~1.020, P = 0.012)。ROC曲线分析显示，EGVV模型区分GOV1与GOV2/3分型的AUC为0.777 (95% CI: 0.692~0.852)；在EGVV基础上加入PLT后，联合模型AUC为0.778 (95% CI: 0.696~0.853)。DeLong检验显示，两模型间AUC差异无统计学意义(P = 0.995)。结论：增强CT定量测量的食管胃底静脉曲张体积(EGVV)与胃食管静脉曲张Sarin分型密切相关，并可作为区分GOV1与GOV2/3分型的独立影像学指标。

Abstract: Objective: To evaluate the value of quantitative measurement of esophagogastric variceal volume (EGVV) based on contrast-enhanced CT in differentiating Sarin classifications of gastroesophageal varices. Methods: A total of 151 patients with liver cirrhosis who underwent contrast-enhanced CT and upper gastrointestinal endoscopy at the Second Affiliated Hospital of Anhui Medical University between January 2018 and October 2025 were retrospectively included. According to endoscopic findings and the Sarin classification, patients were divided into the GOV1 group (n = 107) and GOV2/3 group (n = 44). EGVV was measured using three-dimensional reconstruction based on portal venous phase CT images, and laboratory data were collected. Univariate and multivariate binary logistic regression analyses were performed to identify factors associated with GOV2/3 classification, and predictive models were established. ROC curve analysis was used to evaluate the performance of the models. Results: EGVV was significantly higher in the GOV2/3 group than in the GOV1 group (P < 0.001). Multivariate Logistic regression analysis demonstrated that EGVV was independently associated with GOV2/3 classification (OR = 1.353, 95% CI: 1.147~1.595, P < 0.001). Platelet count was also independently associated with GOV2/3 classification (OR = 1.011, 95% CI: 1.003~1.020, P = 0.012). ROC curve analysis showed that the EGVV-based model achieved an AUC of 0.777 (95% CI: 0.692~0.852) for distinguishing GOV1 from GOV2/3. After incorporating PLT, the AUC of the combined model was 0.778 (95% CI: 0.696~0.853). DeLong’s test showed no significant difference in AUC between the two models (P = 0.995). Conclusion: Contrast-enhanced CT-derived EGVV is significantly associated with Sarin classification and serves as an independent imaging marker for differentiating GOV1 from GOV2/3 gastroesophageal varices.