肝脏 ›› 2025, Vol. 30 ›› Issue (12): 1682-1686.

• 肝纤维化及肝硬化 • 上一篇    下一篇

肝硬化食管胃底静脉曲张破裂出血的影响因素分析及风险模型构建

许静静, 宗富强   

  1. 215600 张家港 张家港澳洋医院消化内科
  • 收稿日期:2025-02-02 发布日期:2026-02-10
  • 通讯作者: 宗富强,Email:zfq18367299
  • 基金资助:
    苏州市科技发展计划(SYS2020009)

The construction of a risk model for esophageal and gastric variceal bleeding in cirrhotic patients involves analyzing the influencing factors

XU Jing-jing, ZONG Fu-qiang   

  1. Department of Gastroenterology, Zhangjiagang Aoyang Hospital, Zhangjiagang 215600, China
  • Received:2025-02-02 Published:2026-02-10
  • Contact: ZONG Fu-qiang,Email:zfq18367299

摘要: 目的 探讨肝硬化食管胃底静脉曲张破裂出血的影响因素,并构建风险模型。方法 将2021年7月至2023年7月张家港澳洋医院收治的128例肝硬化EGV患者按照8∶2随机分为训练集102例和验证集26例。根据是否发生EGVB分为EGVB组48例和非EGVB组54例。筛查肝硬化EGV患者发生EGVB的危险因素,构建并验证风险模型对肝硬化EGVB的预测效能。结果 EGVB组患者的PLT为(77.31±8.02)×109/L,低于非EGVB组的(89.61±9.75)×109/L(P<0.05),PT、脾静脉内径、脾脏弹性值及EGV严重程度中重度占比分别为(19.82±4.05)s、(11.03±2.75)mm、(16.85±2.14)kPa、91.67%,高于非EGVB组的(15.61±3.27)s、(8.41±2.91)mm、(12.68±2.81)kPa、75.93%,差异均有统计学意义(P<0.05)。PLT、PT、脾静脉内径、脾脏弹性值及EGV严重程度是肝硬化EGV患者发生EGVB的影响因素(P<0.05)。风险模型预测训练集肝硬化EGVB的灵敏度为0.775,特异度为0.865,曲线下面积为0.875。风险模型预测验证集肝硬化EGVB的灵敏度为0.763,特异度为0.841,曲线下面积为0.861。结论 PLT、PT、脾静脉内径、脾脏弹性值及EGV严重程度与肝硬化EGVB的发病有关,构建风险模型有助于早期评估肝硬化EGVB的发生。

关键词: 肝硬化, 食管胃底静脉曲张破裂出血, 影响因素, 风险模型

Abstract: Objective To investigate the influencing factors of esophageal and gastric variceal bleeding (EGVB) in cirrhotic patients and develop a risk model. Methods 128 cirrhotic patients with esophageal and gastric varices (EGV) admitted to Zhangjiagang Aoyang Hospital were randomly divided into a training set of 102 cases and a validation set of 26 cases at an 8∶2 ratio. Based on the occurrence of esophageal and gastric variceal bleeding (EGVB), they were divided into an EGVB group of 48 cases and a non-EGVB group of 54 cases. The risk factors for EGVB in EGV patients with cirrhosis were screened, and a risk model was constructed and validated for predicting EGVB in cirrhosis. Results The PLT level in the EGVB group was (77.31 ± 8.02) × 109/L, which was lower than that in the non-EGVB group [(89.61 ± 9.75) × 109/L, P<0.05]. The PT, splenic vein diameter, spleen elasticity value, and the proportion of moderate-to-severe EGV in the EGVB group were (19.82 ± 4.05) s, (11.03 ± 2.75) mm, (16.85 ± 2.14) kPa, and 91.67%, respectively, all of which were higher than those in the non-EGVB group [(15.61 ± 3.27) s, (8.41 ± 2.91) mm, (12.68 ± 2.81) kPa, and 75.93%, P<0.05]. PLT, PT, splenic vein diameter, spleen elasticity value, and EGV severity are influencing factors for the occurrence of EGVB in cirrhotic patients with EGV (P<0.05). The sensitivity, specificity, and area under the curve of the risk model prediction training set for liver cirrhosis EGVB are 0.775, 0.865, and 0.875, respectively. The sensitivity of the risk model prediction validation set for liver cirrhosis EGVB is 0.763, the specificity is 0.841, and the area under the curve was 0.861. Conclusion PLT, PT, inner diameter of splenic vein, splenic elasticity value, and severity of EGV are related to the occurrence of EGVB in liver cirrhosis. Constructing a risk model is helpful for early assessment of EGVB in liver cirrhosis.

Key words: Cirrhosis, Esophageal and gastric variceal rupture and bleeding, Influencing factors, Risk Model