Center of Liver Disease, Beijing Ditan Hospital Capital Medical University, Beijing 100015, China.
Statistics Room, Beijing Ditan Hospital Capital Medical University, Beijing 100015, China.
Biomed Res Int. 2020 Jun 2;2020:5218930. doi: 10.1155/2020/5218930. eCollection 2020.
To establish a novel nomogram for diagnosing liver fibrosis in patients with chronic hepatitis B virus (HBV) infection and verify the diagnostic performance of the established nomogram.
Patients with chronic HBV infection who met the inclusion and exclusion criteria were enrolled in this retrospective study; 70% and 30% of patients were randomly assigned to training dataset and validation dataset, respectively. The risk factors for liver fibrosis were screened using the univariate and multivariate logistic regression analyses. Based on the results, a nomogram was established and verified.
508 patients with chronic HBV infection were included in this study ( = 355 for training dataset and = 153 for validation dataset). The logistic regression analysis showed that liver stiffness measurement (LSM), platelet (PLT) count, and prothrombin time (PT) were independent risk factors for liver fibrosis ( < 0.01), which were used to establish the nomogram. The consistency index (C-index) of the nomogram established for diagnosing liver fibrosis was 0.875. The calibration line and the ideal line were consistent, which indicated that diagnosis of liver fibrosis by the established model was accurate. The values of area under the receiver operator characteristic (ROC) curve (AUROC) for diagnosing liver fibrosis by the nomogram were 0.857 and 0.862 in the training dataset and validation dataset, respectively, which were noticeably higher than those in the well-known serological models, including the aspartate aminotransferase- (AST-) to-platelet ratio index (APRI) scoring model, fibrosis-4 (FIB-4) scoring model, APAG model (including age, PT, albumin, and -glutamyl transferase), and S-index model (all < 0.05).
LSM, PT, and PLT were found as independent risk factors for liver fibrosis. The established nomogram exhibited an excellent diagnostic performance, and it can more visually and individually evaluate the probability of liver fibrosis in patients with chronic HBV infection.
建立一种诊断慢性乙型肝炎病毒（HBV）感染患者肝纤维化的新诺模图，并验证所建立的诺模图的诊断性能。
本回顾性研究纳入符合纳入和排除标准的慢性 HBV 感染患者；将患者随机分为 70%的训练数据集和 30%的验证数据集。使用单因素和多因素逻辑回归分析筛选肝纤维化的危险因素。基于分析结果，建立并验证诺模图。
本研究共纳入 508 例慢性 HBV 感染患者（训练数据集为 355 例，验证数据集为 153 例）。逻辑回归分析显示，肝硬度测量值（LSM）、血小板计数（PLT）和凝血酶原时间（PT）是肝纤维化的独立危险因素（<0.01），用于建立诺模图。所建立的肝纤维化诊断诺模图的一致性指数（C 指数）为 0.875。校准线与理想线一致，表明该模型对肝纤维化的诊断是准确的。在训练数据集和验证数据集中，该诺模图诊断肝纤维化的受试者工作特征曲线（ROC）下面积（AUROC）值分别为 0.857 和 0.862，明显高于著名的血清学模型，包括天冬氨酸氨基转移酶（AST）-血小板比值指数（APRI）评分模型、纤维化-4（FIB-4）评分模型、APAG 模型（包括年龄、PT、白蛋白和γ-谷氨酰转移酶）和 S 指数模型（均<0.05）。
LSM、PT 和 PLT 被确定为肝纤维化的独立危险因素。所建立的诺模图具有优异的诊断性能，可更直观、个体化地评估慢性 HBV 感染患者肝纤维化的概率。
