1 Department of Radiology, University of Wisconsin School of Medicine & Public Health, E3/311 Clinical Science Center, 600 Highland Ave, Madison, WI 53792-3252.
AJR Am J Roentgenol. 2019 Mar;212(3):547-553. doi: 10.2214/AJR.18.20284. Epub 2019 Jan 15.
The objective was to develop a multiparametric CT algorithm to stage liver fibrosis in patients with chronic hepatitis C virus (HCV) infection.
Abdominal CT and laboratory measures in 469 patients with HCV (340 men and 129 women; mean age, 50.1 years) were compared against the histopathologic Metavir fibrosis reference standard (F0, n = 49 patients; F1, n = 69 patients; F2, n = 102 patients; F3, n = 76 patients; F4, n = 173 patients). From the initial candidate pool, nine CT and two laboratory measures were included in the final assessment (CT-based features: hepatosplenic volumetrics, texture features, liver surface nodularity [LSN] score, and linear CT measurements; laboratory-based measures: Fibrosis-4 [FIB-4] score and aspartate transaminase-to-platelets ratio index [APRI]). Univariate logistic regression and multivariate logistic regression were performed with ROC analysis, proportional odds modeling, and probabilities.
ROC AUC values for the model combining all 11 parameters for discriminating significant fibrosis (≥ F2), advanced fibrosis (≥ F3), and cirrhosis (F4) were 0.928, 0.956, and 0.972, respectively. For all nine CT-based parameters, these values were 0.905, 0.936, and 0.972, respectively. Using more simplified panels of two, three, or four parameters yielded good diagnostic performance; for example, a two-parameter model combining only LSN score with FIB-4 score had ROC AUC values of 0.886, 0.915, and 0.932, for significant fibrosis, advanced fibrosis, and cirrhosis. The LSN score performed best in the univariate analysis.
Multiparametric CT assessment of HCV-related liver fibrosis further improves performance over the performance of individual parameters. An abbreviated panel of LSN score and FIB-4 score approached the diagnostic performance of more exhaustive panels. Results of the abbreviated panel compare favorably with elastography, but this approach has the advantage of retrospective assessment using preexisting data without planning.
本研究旨在开发一种多参数 CT 算法，以对慢性丙型肝炎病毒（HCV）感染患者的肝纤维化进行分期。
对 469 例 HCV 患者（男 340 例，女 129 例；平均年龄 50.1 岁）的腹部 CT 和实验室检查结果与组织学 Metavir 纤维化参考标准（F0 期 49 例，F1 期 69 例，F2 期 102 例，F3 期 76 例，F4 期 173 例）进行比较。在初始候选池中，有 9 项 CT 检查和 2 项实验室检查纳入最终评估（基于 CT 的特征：肝脾容积测量、纹理特征、肝表面结节[LSN]评分和线性 CT 测量；基于实验室的指标：纤维化-4[FIB-4]评分和天冬氨酸转氨酶与血小板比值指数[APRI]）。使用 ROC 分析、比例优势模型和概率进行单变量和多变量逻辑回归。
该模型联合所有 11 个参数鉴别显著纤维化（≥F2）、进展性纤维化（≥F3）和肝硬化（F4）的 ROC AUC 值分别为 0.928、0.956 和 0.972。对于所有 9 项基于 CT 的参数，这些值分别为 0.905、0.936 和 0.972。使用更为简化的两、三或四个参数组合也能获得较好的诊断性能；例如，LSN 评分与 FIB-4 评分联合的二参数模型，其鉴别显著纤维化、进展性纤维化和肝硬化的 ROC AUC 值分别为 0.886、0.915 和 0.932。LSN 评分在单变量分析中表现最佳。
多参数 CT 评估 HCV 相关肝纤维化的表现优于单一参数。由 LSN 评分和 FIB-4 评分组成的简化模型，其诊断性能与更全面的模型相当。该简化模型的结果与弹性成像相比具有优势，但该方法具有利用现有数据进行回顾性评估而无需计划的优势。
