Objectives: Several scoring systems have been developed to noninvasively predict the presence of advanced fibrosis in patients with chronic liver disease. Hepatitis C virus and nonalcoholic fatty liver disease are the 2 most common indications for orthotopic liver transplant and are associated with disease recurrence that can lead to fibrosis progression. Here, we evaluated the performance of commonly used fibrosis scores in assessing the presence of advanced fibrosis in patients after orthotopic liver transplant.
Materials and Methods: Our study consisted of con-secutive patients with hepatitis C virus or nonalcoholic fatty liver disease who underwent a liver biopsy after transplant and had laboratory measurements within 1 week of biopsy. Graft fibrosis was determined by an experienced pathologist (stage F0-F4). Advanced fibrosis was defined as stage F3-F4. The following fibrosis scores were calculated for each patient: aspartate aminotransferase/alanine aminotransferase ratio, aspartate aminotransferase/platelet ratio index, and fibrosis-4 index.
Results: We analyzed 93 patients with median age of 59 years (25th and 75th percentile of 53 and 64 y) and median body mass index of 31.8 kg/m2 (25th and 75th percentile of 27 and 37.6 kg/m2). Of total patients, 41 (44%) were diabetic. Median time to liver biopsy posttransplant was 27.7 months (25th and 75 percentile of 10.8 and 59.9 mo). We found that 54 patients (58%) had no fibrosis, 15 (16.1%) had F1, 8 (8.6%) had F2, 7 (7.5%) had F3, and 9 (9.7%) had F4. Overall, advanced fibrosis (F3-F4) was present in 16 patients. Aspartate aminotransferase/alanine amino-transferase ratio, aspartate aminotransferase/platelet ratio index, and fibrosis-4 index were not significantly different between patients with and without advanced fibrosis (all P > .05). The calculated fibrosis scores had poor diagnostic accuracy for presence of advanced fibrosis posttransplant.
Conclusions: Commonly used liver fibrosis scores are not accurate in predicting the presence of advanced fibrosis in patients after liver transplant.
Key words : Advanced fibrosis, Liver biopsy, Orthotopic liver transplant
Hepatitis C virus (HCV) and nonalcoholic fatty liver disease (NAFLD) are the 2 most common indications for orthotopic liver transplant (OLT) in adult patients. Both conditions have high recurrence rate posttransplant. In patients who undergo OLT with HCV viremia, the recurrence of HCV infection is inevitable.1 When nonalcoholic steatohepatitis is the indication for transplant, recurrence of steatosis ranges between 8.2% and 62.5% within 4 months to 10 years after transplant.2 Disease recurrence carries the risk of progression to fibrosis, cirrhosis, and subsequently graft loss. Therefore, monitoring the progression of fibrosis in the graft is important.
Currently, liver biopsy is the criterion standard to stage hepatic fibrosis. However, it has limitations, including sampling errors and interobserver variabilities.3 It is also an invasive procedure and is associated with complications, including pain (84%),4 bleeding (0.6%),5 and even death (incidence of 1/10 000).6
Several noninvasive markers for liver fibrosis have been studied. Direct fibrosis markers like hyaluronic acid7 and imaging studies to assess liver stiffness8 are usually expensive and not readily available. On the other hand, simple fibrosis scores like aspartate aminotransferase/alanine aminotrans-ferase ratio (AST/ALT ratio), aspartate aminotrans-ferase/platelet ratio index (APRI), and fibrosis-4 index (FIB-4 index) have gained popularity because they are calculated using frequently available clinical variables such as liver enzymes and platelet count. These simple scores have been well studied in chronic liver disease; however, conflicting data exist about their usefulness to assess fibrosis in liver transplant patients.9-12 The NAFLD fibrosis score (NFS) was developed to assess fibrosis stage exclusively in patients with NAFLD and has been validated in multiple cohorts. However, its utility after OLT for NAFLD to predict graft fibrosis has not been well studied.
The aim of our study was to assess the utility of AST/ALT ratio, APRI, and FIB-4 index in predicting advanced fibrosis in adult liver transplant patients.
Materials and Methods
Data were analyzed from a retrospective cohort of adult patients (age ≥ 18) who received liver transplant between June 2003 and October 2013 at the Cleveland Clinic (Cleveland, OH, USA) and who met the following inclusion criteria: patients with HCV or NAFLD who underwent a liver biopsy after OLT for long-term evaluation of graft function and had laboratory values within 1 week of the biopsy. Patients were excluded if the AST or ALT values were more than 5 times the upper limit of normal.
This study was approved by the Institutional Review Board at Cleveland Clinic. Fibrosis in the transplanted graft was determined by an experienced pathologist using the Metavir or Brunt liver fibrosis staging system (F0-F4). A stage F3 corresponds to bridging fibrosis, and a stage F4 corresponds to cirrhosis.13 Advanced fibrosis was defined as F3-F4. The following fibrosis scores were calculated for each patient: AST/ALT ratio, APRI, and FIB-4 index. The NFS was calculated only for patients with NAFLD.
Demographic data, including age, body mass index, diabetes, and history of HCV were recorded for all patients (Table 1). Biochemistry parameters included platelet count, international normalized ratio, albumin, alkaline phosphatase, alanine aminotransferase (ALT), aspartate aminotransferase (AST), and γ-glutamyltransferase.
Calculation of simple liver fibrosis markers
Values for APRI were calculated using the following formula: (AST [U/L]/upper limit of normal)/ (platelet count [109/L] × 100), taking the upper limit of normal to be 40 IU/L.14 The FIB-4 index was calculated using the following formula15: FIB-4 = (age [years] × AST [U/L])/(platelet count [109/L] × ALT [U/L]1/2). For NAFLD patients, we calculated NFS using the following formula16: -1.675 + (0.037 × age [years]) + (0.094 × body mass index [kg/m2]) + (1.13 × IFG/diabetes) + (0.99 × AST/ALT) - (0.013 × platelet count [109/L]) - (0.66 × albumin [g/dL]), where IFG stands for impaired fasting glucose, defined as fasting plasma glucose of more than 100 mg/dL.
Data are presented as median (25th, 75th percentiles) or frequency (percent). Wilcoxon rank sum tests were used to assess differences in scores between patients with and without post-OLT advanced fibrosis. In addition, receiver operating characteristic (ROC) analysis was used to assess the accuracy of the various scores for prediction of advanced fibrosis; areas under ROC curves (AUC) and corresponding 95% confidence intervals are reported. Sensitivity, specificity, and positive and negative predictive values are presented for suggested cutoff values for prediction. P < .05 was considered to be statistically significant. All analyses were performed using SAS software (version 9.4, The SAS Institute, Cary, NC, USA).
Our analysis included 93 patients with median age of 59 years (53-64 y) and median body mass index of 31.8 kg/m2 (27, 37.6 kg/m2), with 41 patients (44%) having diabetes. The median time of the liver biopsy posttransplant was 27.7 months (range, 10.8-59.9 mo) (Table 1).
Characteristics of the nonalcoholic fatty liver disease subgroup
In our patient group, 50 patients had history of NAFLD before transplant. Median age was 60 years, and median body mass index was 33.5 kg/m2, with 56% having diabetes. Median time between transplant and liver biopsy was 14.8 months. Of note, 2 patients had HCV recurrence after liver transplant, which was treated. Both patients achieved sustained virologic response but went on to develop NAFLD after that (Table 1).
Characteristics of the hepatitis C virus subgroup
In our patient group, 43 patients had history of HCV before transplant. Median age was 59 years, and median body mass index was 30.2 kg/m2, with 30.2% having diabetes. Median time between transplant and liver biopsy was 59.5 months (Table 1).
Fibrosis score results
Based on the biopsy results, the fibrosis stage was as follows: F0: 54 (58%), F1: 15 (16.1%), F2: 8 (8.6%),
F3: 7 (7.5%), and F4: 9 (9.7%). Overall, advanced fibrosis (F3-F4) was present in 16 patients (17.2%). The AST/ ALT ratio, APRI, and FIB-4 index were not significantly different between patients with and without advanced fibrosis (P > .05 for all) (Table 2). The calculated fibrosis scores had poor diagnostic accuracy to predict the presence of advanced fibrosis after OLT, as shown in Figure 1 (the area under the ROC curve for AST/ALT ratio, APRI, and FIB-4 to detect advanced fibrosis in the entire cohort was low at 0.52 [95% CI, 0.36-0.68], 0.64 [95% CI, 0.48-0.81], and 0.59 [95% CI, 0.42-0.76], respectively, where CI is confidence interval).
For NAFLD patients, AST/ALT ratio and NFS were significantly higher in patients with advanced fibrosis than in those with nonadvanced fibrosis. For patients with advanced graft fibrosis, AST/ALT ratio was 1.8 (95% CI, 0.93-3.4) versus 0.83 (95% CI, 0.63-1.1) for those without advanced fibrosis. P value was .048, and AUC was 0.84 (95% CI, 0.59-1.00). The NFS was 2.4 in the advanced fibrosis group (95% CI, 2.2-5.5) versus 0.93 (95% CI, -0.38 to 1.9) in the nonadvanced fibrosis group. P value was .026, and AUC was 0.89 (95% CI, 0.75-1.00) (Figure 2). For APRI and FIB-4 index, the difference between the 2 groups did not reach statistical significance (Table 2).
Our study aimed to examine the accuracy of AST/ALT index, APRI, FIB-4, and NFS in predicting advanced fibrosis in liver transplant patients. These scores have been validated in many chronic liver diseases and have shown variable accuracy in predicting fibrosis stage. For patients with HCV, AST/ALT has been correlated with fibrosis stage17; APRI was evaluated in patients with NAFLD and HCV/human immunodeficiency virus and showed correlation with the fibrosis stage with varying reported sensitivity of 30% to 75% and specificity of 86% to 92%.9,18 The FIB-4 index was developed by Sterling and associates, who found that a cutoff value of > 3.25 correlates with area under the ROC of 0.7 and specificity of 97% in predicting advanced fibrosis in patients with HCV/human immunodeficiency virus coinfection.15
In our study, we found these scores to be poor predictors of advanced fibrosis in the graft. Kamphues and associates had similar conclusions regarding FIB-4. The group assessed the stage of liver fibrosis of 135 liver transplant patients (94 with HCV, 41 with alcoholic cirrhosis) using liver biopsy and found that FIB-4 did not correlate with the fibrosis stage.10
Other studies found a significant correlation with fibrosis stage. Kitajima and associates studied AST/ALT ratio, APRI, and FIB-4 in HCV-positive patients after liver transplant and concluded that FIB-4 correlates with advanced fibrosis.11 Kamphues and associates found a significant association between APRI and fibrosis stage (P < .05) in the HCV group with AUC for diagnosis of significant fibrosis (F ≥ 2) and cirrhosis (≥ F4) of 0.68 and 0.63, respectively.10 Pissaia and associates evaluated APRI and FIB-4 in patients who had liver transplant for different indications, including HCV, hepatitis B virus, alcoholic disease, and cholestatic disease. The group found that APRI and FIB-4 significantly correlated with the histologic stage of liver fibrosis regardless of the cause.12 Toniutto and associates evaluated the role of APRI and AST/ALT ratio and other markers in detecting significant fibrosis (defined as Ishak > 2) in liver transplant patients with HCV. They found that APRI and AST/ALT ratio had a good AUC of 0.801 and 0.785, respectively.19 Harada and associates also found a significant correlation between fibrosis stage and APRI in transplant patients with HCV, with a good AUC of 0.7 in detecting significant fibrosis (F > 1).8 Corradi and associates evaluated APRI in HCV liver transplant patients and found significant fibrosis, which correlated with a good AUC of 0.815.20 Most of the available studies were done in patients who had HCV as an indication for liver transplant and focused on identifying significant fibrosis as opposed to advanced fibrosis.
To our knowledge, our study is one of the first studies to assess the use of noninvasive fibrosis scores compared with liver biopsy in NAFLD patients after liver transplant. In the NAFLD subgroup and similar to the HCV subgroup, we found that APRI and FIB-4 had poor accuracy to detect advanced fibrosis. However, AST/ALT ratio was significantly higher in the advanced fibrosis patients. Our results for AST/ALT ratio had AUC of 0.84 and a cutoff value > 1.4, with sensitivity of 66.7% and specificity of 85.1% (Table 3). The NFS was developed by Angulo and associates in an international multicenter study, which included 733 patients with liver biopsy-confirmed NAFLD, showing AUC of 0.88 in the estimation group and 0.82 in the validation group. A cutoff of > 0.676 had high accuracy in identifying advanced fibrosis with a positive predictive value of 90% and 92%, respectively.16 In our study, the NFS was significantly higher in the advanced fibrosis group. We applied the same validated cutoff of > 0.676 and found it had a sensitivity of 100% and specificity of 42.6%. It is important to note that, although the NFS had good sensitivity, it had low positive predictive value of 10% (Table 3).
The study has a number of limitations. First, it is a single center study. Second, biopsies were not done on all of the patients transplanted at our institution as part of a protocol, and the decision to obtain a biopsy for long-term evaluation of graft function was left to the discretion of the treating hepatologist or transplant surgeon.
In conclusion, we found that AST/ALT, APRI, and FIB-4 could not accurately predict advanced fibrosis in the graft after liver transplant. Only AST/ALT and NFS correlated with advanced fibrosis in the graft when the indication of OLT was NAFLD. These findings highlight the need to further investigate new noninvasive markers to decrease reliance on invasive liver biopsy in assessing the progression of fibrosis in liver transplant patients.
Volume : 16
Issue : 5
Pages : 562 - 567
DOI : 10.6002/ect.2016.0340
From the the 1Department of Pediatric Gastroenterology and Hepatology, and the
2Digestive Disease Institute, Cleveland Clinic, Cleveland, Ohio, USA
Acknowledgements: MNK, PCS, JG, JR, SB, ZA, RL, IH, BE, NA were involved in developing the study concept and design; acquisition of data; analysis and interpretation of data; drafting of the manuscript; critical revision of the manuscript. RL performed statistical analysis and provided critical revision of the manuscript for important intellectual content. The authors have no conflicts of interest or sources of funding to declare.
Corresponding author: Naim Alkhouri, Department of Gastroenterology and Hepatology, Cleveland Clinic Foundation, 9500 Euclid Avenue, Cleveland, Ohio 44195, USA
Phone: +1 216 444 9000
Table 1. Patient Characteristics
Table 2. Fibrosis Scores for Prediction of Fibrosis After Orthotopic Liver Transplant
Table 3. Validity of Suggested Cutoff Values
Figure 1. Accuracy of Fibrosis Scores in Predicting Advanced Fibrosis After Liver Transplant in Patients with Nonalcoholic Fatty Liver Disease or Hepatitis C Virus: Receiver Operating Characteristic Analysis
Figure 2. Accuracy of Fibrosis Scores in Predicting Advanced Fibrosis After Liver Transplant in Patients With Nonalcoholic Fatty Liver Disease: Receiver Operating Characteristic Analysis
Figure 3. Accuracy of Fibrosis Scores in Predicting Advanced Fibrosis after Liver Transplantation in Patients with Hepatitis C Virus: Receiver Operating Characteristic Analysis