Journal Logo

Clinical and Translational Research

Assessment of Allograft Fibrosis by Transient Elastography and Noninvasive Biomarker Scoring Systems in Liver Transplant Patients

Beckebaum, Susanne1,2,5; Iacob, Speranta2,3; Klein, Christian G.1,2; Dechêne, Alexander2; Varghese, Joye1; Baba, Hideo A.4; Sotiropoulos, Georgios C.1; Paul, Andreas1; Gerken, Guido2; Cicinnati, Vito R.1,2

Author Information
doi: 10.1097/TP.0b013e3181cc66ca

Liver biopsy is currently the only established method of assessing hepatic inflammation and fibrosis in the nontransplant and transplant settings. However, the usefulness of liver biopsy is limited by its invasiveness, inter- and intraobserver variability, sampling error, and the requirement for repeated examinations to monitor graft function (1). Many transplant centers perform serial protocol biopsies in hepatitis C transplant recipients to monitor disease progression and to determine when to start antiviral treatment. The clinical applications of several methods that are simple, noninvasive, inexpensive, and accessible have been evaluated for diagnosing the extent of hepatic fibrosis. Imaging techniques, especially transient elastography (TE) (FibroScan, Echosens, Paris), have been validated in immunocompetent patients with chronic liver disease (2–5), mostly in hepatitis C virus (HCV)-infected patients (6–8). TE is a novel, rapid, noninvasive, and reproducible method of measuring liver stiffness (LS). There are only few published studies evaluating the predictability of fibrosis using FibroScan in the liver transplantation (LT) setting, all except one (9) focus on HCV transplant patients (10–13). Another study investigated the usefulness of TE in living donor LT recipients in the perioperative period (14).

The diagnostic value of single laboratory tests, combinations of routinely available laboratory values with or without clinical parameters (8, 15–17), direct biochemical markers of hepatic extracellular matrix turnover (18, 19), and more complex assays that are based on a statistical approach such as FibroTest (20), Fibrometer (21), and Hepascore (22), has been assessed in immunocompetent patients. The diagnostic use of many of these noninvasive tests, however, remains limited; standardization of these assays is only partially realized and their usefulness in the LT setting remains to be determined. To date, there is no model available for transplant recipients, which can be used irrespective of the indication for LT.

This is, to our knowledge, the first report of a prospective study that aims to (1) investigate and compare the diagnostic accuracy of TE and multiparameter scores for assessing liver fibrosis in LT patients and (2) develop a model to identify advanced fibrosis in patients transplanted for HCV and non–HCV-related liver diseases.

METHODS

Patients

Table 1 summarizes the clinical features of patients (training set) who were transplanted in the Department of General, Visceral, and Transplantation Surgery at the University Hospital in Essen, Germany and were prospectively enrolled in the study between November 2007 and December 2008.

TABLE 1
TABLE 1:
Characteristics of patients in the training set and validation set

Patients were included if they were more than or equal to 18 years and had complete medical records. Exclusion criteria were abnormal coagulation indices (platelet count<50×109/L and quick prothrombin time<50%), ascites, body mass index (BMI) more than 35 kg/m2, viral coinfection, HCV patients under current antiviral therapy, post–LT-acquired hepatitis B virus (HBV) or HCV infection, disease recurrence other than HCV, presence of acute or chronic rejection, and cholestasis due to nonanastomotic or anastomotic strictures without resolution at the time of enrollment. Equal selection criteria were used for the validation group (Table 1) who was transplanted in the same department as the training group and collected until June 2009 to evaluate the performance of the new scoring system.

The study was conducted in accordance with the Declaration of Helsinki, approved by the Institutional Review Board of the University of Duisburg-Essen (IRB 08-3632), and all patients provided written informed consent before study entry.

Composite Fibrosis Panels

Blood values were measured in our central laboratories and are depicted in an online table (see Supplemental Digital Content 1, http://links.lww.com/TP/A177). The aspartate aminotransferase-to-platelet ratio index (APRI) score (23), Forns index (16), and FibroTest (20) were originally evaluated in chronic HCV patients (7, 24) and more recently in patients with various other liver diseases such as HBV infection, alcohol-related chronic liver disease, and nonalcoholic fatty liver disease (NASH) (21, 25–29). Biopredictive (Paris, France) kindly allowed us to compute the FibroTest and ActiTest results on their website. We also tested the diagnostic value of the Hepascore (22), FibroIndex (30), FIB-4 (31), Lok score (15), and the Benlloch score (32), which is the only validated and published test for HCV LT patients. Commercially available enzyme-linked immunosorbent assays were used to assess serum concentrations of hyaluronic acid (HA), which is included as a parameter in the Hepascore (22). These assays were performed according to the manufacturer's instructions (TECOmedical AG, Sissach, Switzerland).

Liver Histology

Liver biopsies were performed in our LT outpatient unit. Liver biopsy specimens were fixed in formalin, embedded in paraffin, and analyzed by an expert pathologist who had no access to the results of TE or the other noninvasive tests. Liver fibrosis was evaluated according to the staging and grading system of Batts and Ludwig (33). Steatosis was graded and NASH was diagnosed according to the criteria described by Brunt et al. (34). Liver specimens were considered adequate if they were at least 15 mm long or included at least 10 complete portal tracts. Clinical, biochemical data, ultrasonography, and liver stiffness measurement (LSM) were attained on the same day as the liver biopsy.

Liver Stiffness Measurement

LSM by TE was performed using the Fibroscan. Only procedures with 10 validated measurements and a greater than 60% success rate were considered reliable. LSM operators were blinded to the clinical data. The median value of successful measurements was considered to be representative of LS in a given patient only if the interquartile range of all validated measurements was less than 25% of the median value.

Statistical Analysis

Continuous data are expressed as mean±standard deviation unless otherwise indicated. Categorical data are described as frequencies of the subjects with a specific characteristic. Chi-square test or Fisher's exact test was used for comparing categorical data and Student's t test or Mann-Whitney U test was used for comparing continuous variables.

The Kruskal-Wallis test was performed to compare stages of fibrosis with LS for global comparison. An overall alpha of less than 0.05 was chosen to indicate statistical significance. The Mann-Whitney U test was used to compare stages of fibrosis between two groups with LS. To protect the experimentwise error rate at 0.05, each of the four emerging comparisons was tested at alpha=0.01. Area under the receiver operating characteristic (AUROC) curves were constructed to define the best cutoff points to distinguish different stages of fibrosis. The optimal cutoff value was determined at the highest sensitivity with a specificity forced no less than 90%. The z test was used for comparisons between AUROCs of the various fibrosis scores (35).

Two-tailed P values less than 0.05 were considered statistically significant. Statistical analysis was performed using SPSS software (SPSS Inc., Chicago, IL).

RESULTS

A total of 166 patients were enrolled in the training group. Post hoc exclusion criteria included unsuccessful interpretation of liver biopsies in six patients due to inadequate material, chronic ductopenic rejection in two patients, and recurrent autoimmune hepatitis (AIH) in one patient. Thus, the final training group comprised 157 patients.

Indications for LT in both the training and validation sets were HCV-related cirrhosis (n=50 [31.8%] and n=28 [37.8%]), HBV infection (n=24 [15.3%] and n=5 [6.8%]), alcohol-related liver disease (n=22 [14%] and n=11 [14.9%]), autoimmune-related liver diseases (n=14 [8.9%] and n=6 [8.1%]), acute liver failure (n=14 [8.9%] and n=9 [12.2%]), cryptogenic liver disease (n=12 [7.6%] and n=3 [4.1%]), and other causative factors (biliary atresia, Budd-Chiari syndrome, Wilson disease, hemochromatosis, neuroendocrine tumor, hemangioendothelioma, nonalcoholic steatohepatitis, and alpha 1-antitrypsin deficiency; n=21 [13.4%] and n=12 [16.2%]). All patients transplanted for hepatitis C had recurrent viral infection as diagnosed by positive detection of HCV RNA (6.5±26.9×106 IU/mL) and histologic features; none of them had fibrosing cholestatic hepatitis.

All patients in the training and validation groups received tacrolimus (73.2% and 74.4%, respectively) or cyclosporine A-based immunosuppressive therapy. Of those, 28.7% and 21.6% received additional low-dose (2.5–5 mg) steroid therapy, 3.2% and 2.7% azathioprine, 41.4% and 32.4% mycophenolate mofetil, and 5.1% and 5.4% sirolimus therapy. Results of the laboratory values at examination are depicted in an online table (see Supplemental Digital Content 2, http://links.lww.com/TP/A178).

In non-HCV patients diagnosed with moderate fibrosis (F=2) the following risk factors for fibrosis could be identified in the training (n=25) and validation sets (n=11), respectively: metabolic syndrome as defined by the National Cholesterol Education Program-Adult treatment Panel III criteria (36) (n=6 [24.0%] and n=2 [18.2%]), de novo NASH (n=4 [16%] and n=2 [18.2%]), history of anastomotic and nonanastomotic biliary strictures (n=5 [20%] and n=1 [9.1%]), split LT (n=2 [8.0%] and n=2 [18.2%]), prolonged cold ischemia time of more than 11 hr (n=2 [8%] and n=1 [9.1%]), drug-induced hepatitis (n=2 [8%] and n=0]), de novo AIH (n=0 and n=1 [9.1%]), recurrent cytomegalovirus disease-associated hepatitis (n=0 and n=1 [9.1%]), and vascular complications (n=1 [4%] and n=0).

In the non-HCV subgroup with advanced fibrosis risk factors for fibrosis progression in the training (n=30) and validation sets (n=13) included metabolic syndrome (n=8 [26.7%] and n=4 [30.8%]), de novo NASH (n=5 [16.7%] and n=2 [15.4%]), history of anastomotic and nonanastomotic biliary strictures (n=10 [33.3%] and n=5 [38.5%]), split LT (n=2 [6.7%] and n=1 [7.7%]), prolonged cold ischemia time (n=2 [6.7%] and n=1 [7.7%]), de novo AIH (n=3 [10%] and n=0), vascular complications (n=3 [10%] and n=0), and recurrent cytomegalovirus disease-associated hepatitis (n=1 [3.3%] and n=1 [7.7%]). A total of 24 patients (training and validation sets) with moderate or advanced fibrosis had no readily identifiable risk factors for graft damage despite features of chronic hepatitis (37). Most of these patients classified as “idiopathic” chronic hepatitis had mild graft inflammation.

In patients with recurrent HCV infection (training group), the following clinicoserologic parameters were significantly higher when compared with non-HCV patients: fibrosis stage (F≥2, P=0.006), inflammation grade (A≥2, P=0.01), presence of diabetes (38), age, sex, BMI, transaminases, gamma-glutamyl transpeptidase, cholesterol, triglycerides, gamma globulin, and platelet (PLT) count (Table 1; see Table, Supplemental Digital Content 2, http://links.lww.com/TP/A178). In the validation group, the following parameters were significantly different between the HCV (n=28) and non-HCV (n=46) groups: fibrosis stage (F≥2, P=0.003), age, sex, BMI, cholesterol, and PLT count.

Liver Stiffness Measurement by Transient Elastography

The LSM was unreliable in 11 patients (7%; n=2 in the HCV group and n=9 in the non-HCV group) due to overweight or obesity (BMI 29±3.2 kg/m2, body waist circumference 105.4±10.3 cm). The median LSM was 7.05 kPa (range, 2.8–47 kPa) for the entire training group. The median LSM was significantly different between patients with recurrent hepatitis C (8.45 kPa; range, 3–47 kPa) and non-HCV patients (6.1 kPa; 2.8–45 kPa; P=0.005). The median LSM stratified according to the fibrosis stage (F0, F1, F2, F3, and F4) was 3.8, 5.1, 6.1, 11.1, and 30.3 kPa in HCV patients and 4.1, 5.4, 7.9, 9.8, and 15.1 kPa in non-HCV patients. An overall comparison revealed a significant correlation between the fibrosis stage and LS value within both the HCV (Fig. 1A) and non-HCV (Fig. 1B) groups. F0 and F1 categories were combined because of the small number of F0 patients.

FIGURE 1.
FIGURE 1.:
Liver stiffness and FibroTransplant score for each fibrosis stage. Box plots are shown for both (A/C) the hepatitis C virus (HCV) and (B/D) the non-HCV training groups. The top and bottom of the boxes represent the first and third quartiles, respectively. The boxes enclose the interquartile range, with the median value denoted by the horizontal line. Mann-Whitney U test was used to compare stages of fibrosis with liver stiffness measurement and stages of fibrosis with FibroTransplant score between two groups.

Predictive Factors Associated With Advanced Fibrosis (F≥3) After Liver Transplantation

Factors for prediction of F≥3 in the training group were included in the univariate analysis (Table 2). HA was not considered a variable because it is reported to be a marker for graft rejection (39). Independent variables associated with F≥3 were identified by multivariate logistic regression analysis (Table 3). Based on the logistic regression equation, a predictive model was created that allows the calculation of a risk score for advanced fibrosis after LT: FibroTransplant score=1/(1+EXP {− [−20.5+(0.99×presence of HCV infection)+(0.008×time since LT)+(0.096×total protein)+(6.36×international normalized ratio [INR])+(0.277×glucose)+(0.007×alkaline phosphatase [AP])+(0.97×alpha 2-macroglobulin)]}).

TABLE 2
TABLE 2:
Results of univariate analysis for prediction of advanced fibrosis (F≥3)
TABLE 3
TABLE 3:
Results of multivariate analysis for prediction of severe fibrosis (F≥3)

This model includes values between 0 and 1 and comprises HCV infection as a categorical parameter (1=presence of HCV infection; 0=indication for LT other than HCV), and time since LT (months), total protein (g/L), INR, glucose (mmol/L), AP (U/L), and alpha 2-macroglobulin (g/L) as numeric values. The optimal cutoff value for diagnosis of F≥3 for the entire training group was 0.55 with a specificity, sensitivity, positive predictive value (PPV), and negative predictive value (NPV) of 90.2%, 61.8%, 77.2%, and 81.4%, respectively and a diagnostic accuracy of 80.2%. The AUROC curve was 0.87 (95% confidence interval [CI]: 0.81–0.92), indicating a good prediction of advanced graft fibrosis.

The mean FibroTransplant score stratified according to the fibrosis stage (F0–F1, F2, F3, and F4) was 0.21, 0.30, 0.66, and 0.79 in HCV patients and 0.16, 0.21, 0.40, and 0.72 in non-HCV patients. An overall comparison revealed a significant correlation between the fibrosis stage and the FibroTransplant score (Fig. 1C,D).

Validation of the FibroTransplant Score

Characteristic features of the validation set were similar to that of the training set except for age, transaminases, and gamma globulin (Table 1; see Table, Supplemental Digital Content 2, http://links.lww.com/TP/A178). The model provided a high AUROC curve (0.92 [95% CI: 0.83–0.97]) for the whole validation group for prediction of advanced fibrosis. Subdividing the group in HCV and non-HCV patients, we obtained AUROC values of 0.90 (95% CI: 0.73–0.98) and 0.91 (95% CI: 0.79–0.97), respectively. When we applied the cutoff value of 0.55 for prediction of F≥3 to the whole validation group the specificity, sensitivity, PPV, and NPV were 95.8%, 61.5%, 88.9%, and 82.1%, respectively and the diagnostic accuracy was 83.7%.

Comparison of Transient Elastography With Multiparameter Fibrosis Scores

The AUROC curves of LSM for diagnosis of fibrosis F≥1, F≥2, F≥3, and F=4 for the whole training group were 0.90 (95% CI: 0.82–0.95), 0.87 (95% CI: 0.82–0.93), 0.92 (95% CI: 0.87–0.96), and 0.97 (95% CI: 0.96–0.99), respectively. The optimal cutoff value of LSM for diagnosis of F≥3 in the whole training group was 8.6 kPa with a specificity, sensitivity, PPV, and NPV of 90.3%, 83%, 83.9%, and 93.4%, respectively and a diagnostic accuracy of 89.8%. For the HCV and non-HCV groups, the c-statistic for the diagnosis of distinct fibrosis stages is shown in Figure 2A,B. Based on the LSM distribution according to fibrosis stage, etiology (recurrent HCV vs. non-HCV patients), and AUROC curves, we determined the best discriminating cutoff values (specificity of ≥90%; Table 4).

FIGURE 2.
FIGURE 2.:
Diagnostic value of transient elastography to assess liver fibrosis stages F≥1, F≥2, F≥3, F=4. Area under the receiver operating characteristic (AUROC) curves are shown in (A) hepatitis C virus (HCV) patients (n=48) and (B) non-HCV patients (n=98) of the training cohort.
TABLE 4
TABLE 4:
Operative characteristics of transient elastography in assessing the stage of liver fibrosis after liver transplantation (training cohort)

The best performing diagnostic cutoff values for tested fibrosis scores are depicted in Table 5. AUROC values of different scores for fibrosis F≥3 and F=4 stratified by HCV or non-HCV patients are shown in Table 6. For all blood tests, except the Forns index, AUROC values for diagnosis of cirrhotic non-HCV patients were better than those for F≥3 patients. For HCV patients, this was evident for all scores except FibroTransplant and Benlloch.

TABLE 5
TABLE 5:
Performance of fibrosis scores in the detection of F≥3 in the training set
TABLE 6
TABLE 6:
AUROCs (95% CI) for clinicoserologic scores for estimation of F≥3 and F=4 in HCV and non-HCV patients (training set)

In HCV patients, all scores performed better than in non-HCV transplanted patients for estimating advanced fibrosis. In the HCV setting, the Benlloch and FibroTransplant scores had similar AUROC values, which were superior to the other scores for F≥3; the AUROC of the FibroTransplant score for prediction of F≥3 was significantly higher than that of the FibroTest (P=0.002), FibroIndex (P=0.02), Lok score (P=0.005), and Hepascore (P=0.02). There was also a difference between the Benlloch and Lok scores (P=0.01) for F≥3. There was no difference between the AUROCs of the scores for prediction of F=4.

In the non-HCV group, the AUROC of the FibroTransplant score for prediction of F≥3 was significantly higher than that of the Benlloch score (P=0.001), FibroTest (P=0.01), FibroIndex (P=0.03), Lok score (P=0.001), and Forns index (P=0.002). For F=4, there was a significant difference between the AUROCs of the FibroTransplant score and the Benlloch score (P=0.01), Lok score (P=0.03), and Forns index (P=0.002). For predicting F≥3, there was also a significant difference between the Forns index and APRI (P=0.03) and between the Forns index and FIB-4 (P=0.04). For predicting F=4, the AUROC of the Forns index was statistically significant lower than that of the FIB-4 (P=0.02), FibroIndex (P=0.02), and FibroTest (P=0.04).

Other comparisons did not show a statistically significant difference of one over the other score.

In both HCV patients and non-HCV patients, high specificity (100%) and PPV (100%) were obtained by combining LSM and the FibroTransplant score, however, sensitivity was lower (64% and 50%) and NPV was 73.5% and 83.6%, respectively. The corresponding diagnostic accuracies were 82.0% and 85.9%, respectively.

For the ActiTest, cutoff values of 0.78 and 0.45 predicting A more than or equal to 2 provided AUROC curves of 0.60 and 0.69 in HCV and non-HCV patients, respectively. The corresponding specificity, sensitivity, PPV, and NPV were 90.2% and 92.1%, 5.6% and 22.2%, 25% and 36.3%, and 65.2% and 85.4%, respectively.

DISCUSSION

Liver biopsy has shown to be an important diagnostic and prognostic tool in the management of patients after LT. In HCV transplant recipients, accurate assessment of liver fibrosis is important for guiding antiviral treatment decisions. Limited data are available on the long-term histopathology after LT because protocol liver biopsies in nonviral liver disease are a topic of controversy and generally poorly accepted by patients (40). However, an accelerated course of hepatic fibrosis may occur in LT patients despite stable liver blood tests, indicating an urgent need for noninvasive methods to determine ongoing liver fibrosis (41).

In our study, we prospectively investigated the diagnostic accuracy of LSM in patients transplanted for various etiologies of liver diseases and compared this method with several noninvasive scores.

Studies in long-term transplanted patients have shown a high prevalence of histological abnormalities in protocol liver biopsies in the absence of abnormal liver function tests (42, 43). In most LT patients, more than one risk factor is presumably involved in progression of allograft fibrosis (44–46). In HCV patients, viral reinfection is likely to be the major factor contributing to hepatocyte injury. This might explain the more rapid progression of liver fibrosis in hepatitis C patients when compared with those transplanted for other indications. In our study, fibrosis progression in non-HCV patients was attributed to risk factors such as metabolic syndrome, NASH, history of biliary obstruction and to surgical factors. Moreover, profibrogenic effects of calcineurin inhibitors that have been demonstrated in in vitro and in vivo studies may have contributed to fibrosis progression (47, 48). Interestingly, our study also included some cases with diagnosis of “idiopathic” chronic hepatitis with no obvious cause of chronic graft inflammation (37).

It is well known that the extent of hepatic inflammation, sex, BMI, and metabolic syndrome influence LS values (49). The same applies to biliary strictures that may lead to an increase of LS (50). We, therefore, only included patients with successful endoscopic drainage of biliary obstruction or surgical bile duct revision. Our study results indicate that TE performed better in HCV patients than in non-HCV patients but has shown to be a reliable method of assessing severe fibrosis in both groups. Previously reported LS cutoff values for moderate and advanced fibrosis stages in HCV transplanted patients were higher in two studies (10, 12) and similar in a study by Carrion et al. (11). The highest cutoff values were reported in the study by Harada et al. (10) and may be related to the fact that all included HCV patients had undergone living donor LT. Accelerated regeneration of the partial allograft may lead to increased LS values in those cases measured within the first posttransplant year.

Consistent with previous reports from the literature in the nontransplant setting, our results indicate that LSM is less reliable in the intermediate stages of fibrosis (2, 3, 5). We found that in cases not concordant with regard to LSM and histology, TE tended to underestimate the fibrosis grade compared with liver biopsy. LS values in patients with advanced liver fibrosis were higher in HCV than in non-HCV patients. This may be mainly due to the higher degree of graft inflammation indicated by biopsy and elevated transaminases. Interestingly, in both groups, none of the F0 patients were misclassified. There was no misclassification of F4 HCV patients, whereas 3 of 12 cirrhotic non-HCV patients were misclassified as F3. The false-negative results may have been due to histologically proven macronodular cirrhosis separated by sparse septa (n=2) and the absence of allograft inflammation (n=1). Among patients without cirrhosis with a LSM of more than or equal to 17.3 kPa (HCV group, n=1) and more than or equal to 12.6 kPa (non-HCV group, n=4), histology revealed exacerbation of HCV infection reflected by severe graft inflammation (n=1) and extensive fibrosis with numerous septa (n=4).

We developed a model (FibroTransplant score) based on the presence/absence of HCV infection, time since LT, alpha 2-macroglobulin, AP, total protein, INR, and glucose, which accurately distinguished patients with mild to moderate fibrosis from those with advanced fibrosis. As fibrosis progresses, total protein decreases, whereas INR and the concentration of the protease inhibitor alpha 2-macroglobulin increase. Moreover, elevated AP, HCV infection, and diabetes have been described as risk factors for progression to severe fibrosis (51–54).

Patients in the training group had a mean follow-up of 14.2±4.9 months after examination. In the HCV training group, graft failure (defined as allograft dysfunction causing decompensated cirrhosis leading to relisting for LT, retransplantation, or death) occurred more frequently in patients with a FibroTransplant score more than 0.63 than in those with a score less than 0.63 (23.8% vs. 3.4%, respectively; P=0.02). Likewise, in the non-HCV group, graft failure occurred significantly more often in patients with a FibroTransplant score more than 0.38 than in those with a lower score (20% vs. none, respectively; P=0.0005). These results show that the FibroTransplant score might be valuable to prognosticate the outcome of allograft function during follow-up.

Noninvasive clinicoserologic tests for diagnosing fibrosis are being intensively investigated in immunocompetent patients (8, 15–17, 20, 23) and to a low extent in the transplant setting (55–57). The Bologna Liver Transplantation Group (12) assessed the usefulness of noninvasive methods in 36 HCV transplant patients and found that the diagnostic accuracy of LSM was superior to noninvasive scores such as Benlloch, APRI, Forns, FibroTest, and the Doppler resistance index. PLT count is included in several fibrosis scores such as APRI, Forns index, FIB-4, Fibrometer, FibroIndex, and Lok Score. A low number of thrombocytes in transplanted patients, however, can be related to conditions other than fibrosis progression, such as insufficient involution of splenomegaly after LT despite normal portal vein blood flow velocity, normal liver surface, and excellent hepatic function (58).

Among the clinicoserologic scores, the best diagnostic capacity for diagnosis of advanced fibrosis in HCV and non-HCV patients was obtained using the FibroTransplant score. In the HCV group, we obtained AUROCs of 0.89 and 0.90 in the training and validation sets for the FibroTransplant score and AUROCs of 0.86 and 0.87 for the Benlloch score (32). The latter values were similar to those (0.80 and 0.84) published in the original article. The results for the non-HCV group demonstrated markedly higher AUROC values for the FibroTransplant score than for all other tested scores.

Performance of the Lok score (15) for prediction of cirrhosis had no utility in non-HCV patients but was better in our HCV transplant cohort (AUROC 0.93) compared with that originally published in immunocompetent HCV patients (AUROC 0.78). Of note, we yielded the same AUROC value for the Hepascore in F=4 HCV patients as originally reported for nontransplanted HCV patients (22). Moreover, our results confirmed good performance of the Hepascore for the diagnosis of cirrhotic non-HCV transplant patients. However, the Hepascore has some limitations in the LT setting because it includes HA, which is increased in patients with allograft rejection (39).

The ActiTest has been successfully used as a biochemical marker for diagnosing necroinflammatory activity in viral hepatitis (28, 59). In our study with LT patients, however, the diagnostic value of the ActiTest for predicting moderate to severe inflammation was low for both HCV and non-HCV patients. Discrepancies between histologically determined grade of inflammation and ActiTest score resulted from an overestimation of the necroinflammatory activity in 42% and 11% of patients in the A0–1 HCV and non-HCV subgroups, respectively, and an underestimation of graft inflammation in 47% and 78% of patients in the A2–3 subgroups, respectively.

In summary, our results indicate that TE is currently the most accurate noninvasive approach for diagnosing advanced fibrosis and cirrhosis in patients transplanted for HCV and non–HCV-related liver diseases. Among clinicoserologic scores and especially for non-HCV patients, the best performance for predicting advanced fibrosis was obtained by our new model. These noninvasive methods, however, are not expected to replace, but rather to reduce the frequencies of biopsies for monitoring fibrotic changes during follow-up in patients under anticoagulative therapy, with coagulopathy, or those declining a biopsy.

REFERENCES

1. Regev A, Berho M, Jeffers LJ, et al. Sampling error and intraobserver variation in liver biopsy in patients with chronic HCV infection. Am J Gastroenterol 2002; 97: 2614.
2. Ganne-Carrie N, Ziol M, de Ledinghen V, et al. Accuracy of liver stiffness measurement for the diagnosis of cirrhosis in patients with chronic liver diseases. Hepatology 2006; 44: 1511.
3. Corpechot C, El Naggar A, Poujol-Robert A, et al. Assessment of biliary fibrosis by transient elastography in patients with PBC and PSC. Hepatology 2006; 43: 1118.
4. Nahon P, Kettaneh A, Lemoine M, et al. Liver stiffness measurement in patients with cirrhosis and hepatocellular carcinoma: A case-control study. Eur J Gastroenterol Hepatol 2009; 21: 214.
5. Foucher J, Chanteloup E, Vergniol J, et al. Diagnosis of cirrhosis by transient elastography (FibroScan): A prospective study. Gut 2006; 55: 403.
6. Nitta Y, Kawabe N, Hashimoto S, et al. Liver stiffness measured by transient elastography correlates with fibrosis area in liver biopsy in patients with chronic hepatitis C. Hepatol Res 2009; 39: 675.
7. Castera L, Vergniol J, Foucher J, et al. Prospective comparison of transient elastography, Fibrotest, APRI, and liver biopsy for the assessment of fibrosis in chronic hepatitis C. Gastroenterology 2005; 128: 343.
8. Coco B, Oliveri F, Maina AM, et al. Transient elastography: A new surrogate marker of liver fibrosis influenced by major changes of transaminases. J Viral Hepat 2007; 14: 360.
9. Masuzaki R, Yamashiki N, Sugawara Y, et al. Assessment of liver stiffness in patients after living donor liver transplantation by transient elastography. Scand J Gastroenterol 2009; 44: 1115.
10. Harada N, Soejima Y, Taketomi A, et al. Assessment of graft fibrosis by transient elastography in patients with recurrent hepatitis C after living donor liver transplantation. Transplantation 2008; 85: 69.
11. Carrion JA, Navasa M, Bosch J, et al. Transient elastography for diagnosis of advanced fibrosis and portal hypertension in patients with hepatitis C recurrence after liver transplantation. Liver Transpl 2006; 12: 1791.
12. Corradi F, Piscaglia F, Flori S, et al. Assessment of liver fibrosis in transplant recipients with recurrent HCV infection: Usefulness of transient elastography. Dig Liver Dis 2009; 41: 217.
13. Rigamonti C, Donato MF, Fraquelli M, et al. Transient elastography predicts fibrosis progression in patients with recurrent hepatitis C after liver transplantation. Gut 2008; 57: 821.
14. Inoue Y, Sugawara Y, Tamura S, et al. Validity and feasibility of transient elastography for the transplanted liver in the peritransplantation period. Transplantation 2009; 88: 103.
15. Lok AS, Ghany MG, Goodman ZD, et al. Predicting cirrhosis in patients with hepatitis C based on standard laboratory tests: Results of the HALT-C cohort. Hepatology 2005; 42: 282.
16. Forns X, Ampurdanes S, Llovet JM, et al. Identification of chronic hepatitis C patients without hepatic fibrosis by a simple predictive model. Hepatology 2002; 36: 986.
17. Vallet-Pichard A, Mallet V, Nalpas B, et al. FIB-4: An inexpensive and accurate marker of fibrosis in HCV infection. Comparison with liver biopsy and fibrotest. Hepatology 2007; 46: 32.
18. Pungpapong S, Nunes DP, Krishna M, et al. Serum fibrosis markers can predict rapid fibrosis progression after liver transplantation for hepatitis C. Liver Transpl 2008; 14: 1294.
19. Leroy V, Monier F, Bottari S, et al. Circulating matrix metalloproteinases 1, 2, 9 and their inhibitors TIMP-1 and TIMP-2 as serum markers of liver fibrosis in patients with chronic hepatitis C: Comparison with PIIINP and hyaluronic acid. Am J Gastroenterol 2004; 99: 271.
20. Imbert-Bismut F, Ratziu V, Pieroni L, et al. Biochemical markers of liver fibrosis in patients with hepatitis C virus infection: A prospective study. Lancet 2001; 357: 1069.
21. Cales P, Oberti F, Michalak S, et al. A novel panel of blood markers to assess the degree of liver fibrosis. Hepatology 2005; 42: 1373.
22. Adams LA, Bulsara M, Rossi E, et al. Hepascore: An accurate validated predictor of liver fibrosis in chronic hepatitis C infection. Clin Chem 2005; 51: 1867.
23. Pohl A, Behling C, Oliver D, et al. Serum aminotransferase levels and platelet counts as predictors of degree of fibrosis in chronic hepatitis C virus infection. Am J Gastroenterol 2001; 96: 3142.
24. Poynard T, Munteanu M, Ngo Y, et al. FibroTest is effective in patients with normal transaminases, when accuracy is standardized on fibrosis stage prevalence. J Viral Hepat 2008; 15: 472.
25. Zhang YX, Wu WJ, Zhang YZ, et al. Noninvasive assessment of liver fibrosis with combined serum aminotransferase/platelet ratio index and hyaluronic acid in patients with chronic hepatitis B. World J Gastroenterol 2008; 14: 7117.
26. Poynard T, Zoulim F, Ratziu V, et al. Longitudinal assessment of histology surrogate markers (FibroTest-ActiTest) during lamivudine therapy in patients with chronic hepatitis B infection. Am J Gastroenterol 2005; 100: 1970.
27. Nguyen-Khac E, Chatelain D, Tramier B, et al. Assessment of asymptomatic liver fibrosis in alcoholic patients using fibroscan: Prospective comparison with 7 non-invasive laboratory tests. Aliment Pharmacol Ther 2008; 28: 1188.
28. Ngo Y, Benhamou Y, Thibault V, et al. An accurate definition of the status of inactive hepatitis B virus carrier by a combination of biomarkers (FibroTest-ActiTest) and viral load. PLoS One 2008; 3: e2573.
29. Cales P, Laine F, Boursier J, et al. Comparison of blood tests for liver fibrosis specific or not to NAFLD. J Hepatol 2009; 50: 165.
30. Koda M, Matunaga Y, Kawakami M, et al. FibroIndex, a practical index for predicting significant fibrosis in patients with chronic hepatitis C. Hepatology 2007; 45: 297.
31. Sterling RK, Lissen E, Clumeck N, et al. Development of a simple noninvasive index to predict significant fibrosis in patients with HIV/HCV coinfection. Hepatology 2006; 43: 1317.
32. Benlloch S, Berenguer M, Prieto M, et al. Prediction of fibrosis in HCV-infected liver transplant recipients with a simple noninvasive index. Liver Transpl 2005; 11: 456.
33. Batts KP, Ludwig J. Chronic hepatitis. An update on terminology and reporting. Am J Surg Pathol 1995; 19: 1409.
34. Brunt EM, Janney CG, Di Bisceglie AM, et al. Nonalcoholic steatohepatitis: A proposal for grading and staging the histological lesions. Am J Gastroenterol 1999; 94: 2467.
35. Hanley JA, McNeil B. A method comparing the areas under receiver operator characteristic curves derived from the same cases. Radiology 1983; 148: 839.
36. Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults. Executive Summary of The Third Report of the National Cholesterol Education Program (NCEP) Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults (Adult Treatment Panel III). JAMA 2001; 285: 2486.
37. Seyam M, Neuberger JM, Gunson BK, et al. Cirrhosis after orthotopic liver transplantation in the absence of primary disease recurrence. Liver Transpl 2007; 13: 966.
38. American Diabetes Association. Diagnosis and classification of diabetes mellitus. Diabetes Care 2009(31 suppl 1): S62.
39. Fabrega E, Castro B, Crespo J, et al. Different time course of circulating adhesion molecules and hyaluran during hepatic allograft rejection. Transplantation 2000; 69: 569.
40. Mells G, Neuberger J. Protocol liver allograft biopsies. Transplantation 2008; 85: 1686.
41. Abraham SC, Poterucha JJ, Rosen CB, et al. Histologic abnormalities are common in protocol liver allograft biopsies from patients with normal liver function tests. Am J Surg Pathol 2008; 32: 965.
42. Slapak GI, Saxena R, Portmann B, et al. Graft and systemic disease in long-term survivors of liver transplantation. Hepatology 1997; 25: 195.
43. Sebagh M, Rifai K, Feray C, et al. All liver recipients benefit from the protocol 10-year liver biopsies. Hepatology 2003; 37: 1293.
44. Seo S, Maganti K, Khehra M, et al. De novo nonalcoholic fatty liver disease after liver transplantation. Liver Transpl 2007; 13: 844.
45. Scheenstra R, Peeters PM, Verkade HJ, et al. Graft fibrosis after pediatric liver transplantation: Ten years of follow-up. Hepatology 2009; 49: 880.
46. Gor NV, Levy RM, Ahn J, et al. Biliary cast syndrome following liver transplantation: Predictive factors and clinical outcomes. Liver Transpl 2008; 14: 1466.
47. Frizell E, Abraham A, Doolittle M, et al. FK506 enhances fibrogenesis in in vitro and in vivo models of liver fibrosis in rats. Gastroenterology 1994; 107: 492.
48. Khanna A, Kapur S, Sharma V, et al. In vivo hyperexpression of transforming growth factor-beta1 in mice: Stimulation by cyclosporine. Transplantation 1997; 63: 1037.
49. Roulot D, Czernichow S, Le CH, et al. Liver stiffness values in apparently healthy subjects: Influence of gender and metabolic syndrome. J Hepatol 2008; 48: 606.
50. Millonig G, Reimann FM, Friedrich S, et al. Extrahepatic cholestasis increases liver stiffness (FibroScan) irrespective of fibrosis. Hepatology 2008; 48: 1718.
51. Syn WK, Nightingale P, Gunson B, et al. Natural history of unexplained chronic hepatitis after liver transplantation. Liver Transpl 2007; 13: 984.
52. Berenguer M, Ferrell L, Watson J, et al. HCV-related fibrosis progression following liver transplantation: Increase in recent years. J Hepatol 2000; 32: 673.
53. Rowe IA, Webb K, Gunson BK, et al. The impact of disease recurrence on graft survival following liver transplantation: A single centre experience. Transpl Int 2008; 21: 459.
54. Foxton MR, Quaglia A, Muiesan P, et al. The impact of diabetes mellitus on fibrosis progression in patients transplanted for hepatitis C. Am J Transplant 2006; 6: 1922.
55. Toniutto P, Fabris C, Bitetto D, et al. Role of AST to platelet ratio index in the detection of liver fibrosis in patients with recurrent hepatitis C after liver transplantation. J Gastroenterol Hepatol 2007; 22: 1904.
56. Pissaia A Jr, Borderie D, Bernard D, et al. APRI and FIB-4 Scores are useful after liver transplantation independently of etiology. Transplant Proc 2009; 41: 679.
57. Hamelet E, Muntenau M, Thibault V, et al. Interest of fibrotest-actitest (FT-AT) in the follow-up of liver transplanted (LT) patients: A pilot study. J Hepatol 2008; 48: S92.
58. Bolognesi M, Sacerdoti D, Bombonato G, et al. Change in portal flow after liver transplantation: Effect on hepatic arterial resistance indices and role of spleen size. Hepatology 2002; 35: 601.
59. Halfon P, Bourliere M, Deydier R, et al. Independent prospective multicenter validation of biochemical markers (fibrotest-actitest) for the prediction of liver fibrosis and activity in patients with chronic hepatitis C: The fibropaca study. Am J Gastroenterol 2006; 101: 547.
Keywords:

Liver stiffness; Hepatitis C; Liver biopsy; Graft inflammation; Score

Supplemental Digital Content

© 2010 Lippincott Williams & Wilkins, Inc.