Snyder Ned, Nguyen Audrey, Gajula Leka, Soloway Roger, Xiao Shu-Yuan, Lau Daryl T-Y, Petersen John
Department of Internal Medicine, Division of Gastroenterology, The University of Texas Medical Branch, Galveston, Texas 77555-0764, United States.
Clin Chim Acta. 2007 Jun;381(2):119-23. doi: 10.1016/j.cca.2007.02.046. Epub 2007 Mar 19.
Multiple serum markers to estimate hepatic fibrosis in chronic liver disease have been proposed. The AST/Platelet Ratio Index (APRI) is a simple biochemical index that has been shown to be useful and accurate in about 50% of patients with chronic hepatitis C. We determined if the combination of the APRI and the FIBROSpect II, a commercially available hepatic fibrosis marker that measures 3 components of the extracellular hepatic matrix, would further help distinguish mild from significant fibrosis in a group of patients with chronic hepatitis C.
In an outpatient setting, 93 consecutive patients were studied who were undergoing staging liver biopsy for chronic hepatitis C who had a liver biopsy length>or=1.5 cm. All had blood drawn at the time of the biopsy. Liver biopsies were staged for fibrosis by the Batts Ludwig criteria (F0-F4). Patients with previous anti-viral therapy, hepatocellular carcinoma, an organ transplant, or co-infection with HIV or hepatitis B were excluded. The APRI was calculated and FIBROSpect II determined.
The AUC of the ROC curve for the APRI and FIBROSpect II were 0.887 and 0.879 respectively. Using cutoffs of <or=0.42 for mild fibrosis (F0-F1) and >or=1.2 for significant fibrosis, the APRI correctly estimated 19 of 20 patients with mild fibrosis for an NPV of 95.0%, and 31 of 33 patients with significant fibrosis for a PPV of 93.6%. The FIBROSpect II also works best utilizing 2 cutoffs, and using cutoffs of <or=25 and >or=85 it correctly identified 18 of 18 patients with mild fibrosis and all 26 patients with significant fibrosis for an NPV and PPV of 100% for both. Among the 40 patients who could not be classified by the APRI, an additional 16 could be correctly classified using the FIBROSpect II with cutoffs of <or=25 and >or=85. This lowered the indeterminate zone from 43.0 to 25.8%. By combining the APRI and the FIBROSpect II, the AUC for the ROC curve improved significantly to 0.931 (p=0.013).
The APRI and the FIBROSpect II are both accurate tests for separating mild from significant fibrosis. By using the APRI as the initial screen, >50% of patients with mild or significant fibrosis can be correctly identified. If the patient falls in the indeterminate zone, then the more expensive FIBROSpect II could be obtained. This strategy could decrease the number of liver biopsies.
已提出多种血清标志物用于评估慢性肝病中的肝纤维化。天冬氨酸转氨酶/血小板比率指数（APRI）是一种简单的生化指标，已证实在约50%的慢性丙型肝炎患者中有用且准确。我们确定APRI与FIBROSpect II（一种可测量肝细胞外基质3种成分的市售肝纤维化标志物）联合使用是否能进一步帮助区分一组慢性丙型肝炎患者的轻度与重度纤维化。
在门诊环境中，对93例连续接受慢性丙型肝炎肝脏分期活检且活检长度≥1.5 cm的患者进行研究。所有患者在活检时均采血。肝活检根据Batts Ludwig标准（F0 - F4）进行纤维化分期。排除既往接受过抗病毒治疗、患有肝细胞癌、接受过器官移植或合并感染HIV或乙型肝炎的患者。计算APRI并测定FIBROSpect II。
APRI和FIBROSpect II的ROC曲线下面积（AUC）分别为0.887和0.879。对于轻度纤维化（F0 - F1），使用≤0.42的临界值，对于重度纤维化，使用≥1.2的临界值，APRI正确估计了20例轻度纤维化患者中的19例，阴性预测值（NPV）为95.0%，33例重度纤维化患者中的31例，阳性预测值（PPV）为93.6%。FIBROSpect II使用两个临界值时效果也最佳，使用≤25和≥85的临界值，它正确识别了18例轻度纤维化患者中的18例以及所有26例重度纤维化患者，NPV和PPV均为100%。在40例无法通过APRI分类的患者中，使用FIBROSpect II（临界值≤25和≥85）可额外正确分类16例。这将不确定区域从43.0%降至25.8%。通过联合APRI和FIBROSpect II，ROC曲线的AUC显著提高至0.931（p = 0.013）。
APRI和FIBROSpect II都是区分轻度与重度纤维化的准确检测方法。以APRI作为初始筛查，超过50%的轻度或重度纤维化患者可被正确识别。如果患者处于不确定区域，则可采用更昂贵的FIBROSpect II。这种策略可减少肝活检的数量。
