Department of Medicine, Indiana University School of Medicine, Indianapolis, Indiana.
School of Industrial Engineering, Purdue University, West Lafayette, Indiana.
Clin Gastroenterol Hepatol. 2020 Sep;18(10):2305-2314.e12. doi: 10.1016/j.cgh.2020.04.017. Epub 2020 Apr 11.
BACKGROUND & AIMS: Several strategies are available for detecting cirrhosis in patients with nonalcoholic fatty liver disease (NAFLD), but their cost effectiveness is not clear. We developed a decision model to quantify the accuracy and costs of 9 single or combination strategies, including 3 noninvasive tests (fibrosis-4 [FIB-4], vibration-controlled transient elastography [VCTE], and magnetic resonance elastography [MRE]) and liver biopsy, for the detection of cirrhosis in patients with NAFLD.
Data on the diagnostic accuracy, costs, adverse events, and cirrhosis outcomes over a 5-year period were obtained from publications. The diagnostic accuracy, per-patient cost per correct diagnosis of cirrhosis, and incremental cost-effectiveness ratios (ICERs) were calculated for each strategy for base cirrhosis prevalence values of 0.27%, 2%, and 4%.
The combination of the FIB-4 and VCTE identified patients with cirrhosis in NAFLD populations with a 0.27%, 2%, and 4% prevalence of cirrhosis with the lowest cost per person ($401, $690, and $1024, respectively) and highest diagnostic accuracy (89.3%, 88.5%, and 87.5% respectively). The combination of FIB-4 and MRE ranked second in cost per person ($491, $781, and $1114, respectively) and diagnostic accuracy (92.4%, 91.6%, 90.6%, respectively). Compared with the combination of FIB-4 and VCTE (least costly), the ICERs were lower for the combination of FIB-4 and MRE ($2864, $2918, and $2921) than the combination of FIB-4 and liver biopsy ($4454, $5156, and $5956) at the cirrhosis prevalence values tested. When the goal was to avoid liver biopsy, FIB-4 + VCTE and FIB-4 + MRE had similar diagnostic accuracies, ranging from 87.5% to 89.3% and 90.6% to 92.4% for a cirrhosis diagnosis, respectively, although FIB-4 + MRE had a slightly higher cost.
In our cost-effectiveness analysis based on the US health care system, we found that results from FIB-4, followed by either VCTE, MRE, or liver biopsy, detect cirrhosis in patients with NAFLD with a high level of accuracy and low cost. Compared with FIB-4 + VCTE, which was the least costly strategy, FIB-4 + MRE had a lower ICER than FIB-4 + LB.
有多种策略可用于检测非酒精性脂肪性肝病（NAFLD）患者的肝硬化，但它们的成本效益尚不清楚。我们开发了一个决策模型，以量化 9 种单一或联合策略的准确性和成本，包括 3 种非侵入性检测（纤维化 4 指数[FIB-4]、振动控制瞬时弹性成像[VCTE]和磁共振弹性成像[MRE]）和肝活检，用于检测 NAFLD 患者的肝硬化。
从出版物中获得了 5 年内关于诊断准确性、成本、不良事件和肝硬化结局的数据。对于肝硬化患病率分别为 0.27%、2%和 4%的基本情况，计算了每种策略的准确性、每位患者正确诊断肝硬化的成本和增量成本效益比（ICER）。
FIB-4 和 VCTE 的联合使用可在肝硬化患病率为 0.27%、2%和 4%的 NAFLD 人群中识别出肝硬化患者，每人的成本最低（分别为 401 美元、690 美元和 1024 美元），诊断准确性最高（分别为 89.3%、88.5%和 87.5%）。FIB-4 和 MRE 的联合使用在每人的成本方面排名第二（分别为 491 美元、781 美元和 1114 美元），诊断准确性也排名第二（分别为 92.4%、91.6%和 90.6%）。与 FIB-4 和 VCTE 的联合使用（成本最低）相比，FIB-4 和 MRE 的联合使用（ICER 分别为 2864 美元、2918 美元和 2921 美元）在测试的肝硬化患病率下低于 FIB-4 和肝活检的联合使用（ICER 分别为 4454 美元、5156 美元和 5956 美元）。当目标是避免肝活检时，FIB-4+VCTE 和 FIB-4+MRE 的诊断准确性相似，分别为 87.5%至 89.3%和 90.6%至 92.4%，尽管 FIB-4+MRE 的成本略高。
基于美国医疗保健系统，我们的成本效益分析发现，FIB-4 后接 VCTE、MRE 或肝活检可高度准确且低成本地检测 NAFLD 患者的肝硬化。与成本最低的 FIB-4+VCTE 策略相比，FIB-4+MRE 的 ICER 低于 FIB-4+LB。
