1264 BOURSIER ET AL CLINICAL GASTROENTEROLOGY AND HEPATOLOGY Vol. 6, No. 11 Table 1. Study Design: Assignment of Judges According to Patient and Obser

Transcription

1 CLINICAL GASTROENTEROLOGY AND HEPATOLOGY 2008;6: Reproducibility of Liver Stiffness Measurement by Ultrasonographic Elastometry JÉRÔME BOURSIER,*, ANSELME KONATÉ,*, GABRIELLA GOREA,* STÉPHANE REAUD, EMMANUEL QUEMENER,* FRÉDÉRIC OBERTI,*, ISABELLE HUBERT FOUCHARD,*, NINA DIB,*, and PAUL CALÈS*, *University of Angers, IFR 132, HIFIH Laboratory (UPRES 3859), Angers; and University Hospital, Hepatogastroenterology Department, Angers, France Background & Aims: Fibroscan is a noninvasive device that assesses liver fibrosis by liver stiffness evaluation (LSE) with ultrasonographic elastometry. We evaluated LSE reproducibility and its influencing factors. Methods: LSE was performed by 4 experienced physicians (>100 LSEs) in 46 patients with chronic liver disease at 4 different anatomic sites. Additional LSEs were performed for ancillary aims, so that 534 LSEs were available. Results: Overall interobserver agreement for LSE results was considered as excellent, with intraclass coefficient correlation (Ric) of Low LSE level, nonrecommended sites, LSE interquartile range >25%, and body mass index >25 independently decreased agreement. Thus, agreement was fair (Ric 0.53) for LSE <9 kilopascals and excellent (Ric 0.90) beyond. The best measurement site for LSE reproducibility was the median axillary line on the first intercostal space under the liver dullness upper limit, with the patient lying in dorsal decubitus. When LSE results were categorized into fibrosis Metavir stages, interobserver discordance was noticed in about 25% of the cases and was the highest for F2 and F3 stages and the lowest for F4. Intraobserver (Ric 0.94), intersite (Ric ), and interequipment (Ric 0.92) agreements for LSE results were excellent. Preliminary standard ultrasonography or probe pressure changes did not improve interobserver agreement. Conclusions: The best measurement site for LSE is the one generally used for liver biopsy. Reproducibility of LSE is globally excellent but is fair in patient with low liver stiffness. The fibrosis diagnosis by ultrasonographic elastometry in low stages or categorized into fibrosis Metavir stages must be interpreted with caution. Assessing the degree of liver fibrosis in chronic liver disease is very useful for patient management. Noninvasive diagnosis of hepatic fibrosis gained considerable ground during the past few years. 1 The first developed tools were blood tests combining several markers included in algorithms. 2 APRI is the easiest score to use, 3 Fibrotest is the most validated algorithm, 4 and FibroMeter seems the most accurate. 5,6 More recently, a device allowing liver stiffness evaluation (LSE) by ultrasonographic elastometry (FibroScan; EchoSens, Paris, France) was developed. 7 FibroScan is becoming increasingly popular because several recent studies have shown that liver stiffness is well-correlated with fibrosis Metavir stages 8 ;in addition, it is an easy, rapid, noninvasive, and painless examination. However, there are few data about characteristics of LSE reproducibility. 9 The main aim of this prospective study was to evaluate the reproducibility of liver stiffness measurement by transient elastography with Fibroscan. The secondary aims were to evaluate the following influencing factors: observer, device, liver stiffness value, anatomic conditions, body mass index (BMI), knowledge of previous measurement, location by standard ultrasonography, and probe pressure. Patients and Methods Patients All patients included in the study were hospitalized in the Department of Hepato-Gastroenterology of the Angers University Hospital, France, for a clinical assessment of chronic liver disease. The only noninclusion criterion was ascites, which makes LSE impossible. 7 The study protocol conformed to the Helsinki Declaration and was approved by the local Ethics Committee. All patients gave their informed consent before being included. Observers Four physicians specialized in hepatology performed the examinations, and each had already performed at least 100 LSEs before the study: judge A, 750; judge B, 109; judge C, 227; and judge D, 117 LSEs. In a reproducibility study, each patient should classically be examined by all the observers to guarantee the highest homogeneity. However, this design was not possible here because the duration of all LSEs performed by one judge was approximately minutes. Therefore, each patient was only examined by 2 judges. This needed 2 different judge pairs who consequently examined 2 different patient groups (Table 1); judges A and B examined a group of 22 patients (P1), and judges C and D examined a group of 24 patients (P2). Each patient was first examined by the most experienced judge (A for P1, C for P2) and second by the less experienced judge (B for P1, D for P2). Thus, overall interobserver agreement was evaluated by comparing the LSE results between the first (O1: A or C) and second (O2: B or D) observers in the 46 patients included in the study. Naturally, the judge pair (P1 or P2) was statistically controlled as a potential bias. Abbreviations used in this paper: APRI, ASAT-to-Platelets-Ratio-Index; BMI, body mass index; c LSE, crude difference between observer LSE results; IQR, interquartile range; w, weighted kappa index; kpa, kilopascal; LSE, liver stiffness evaluation; Ric, intraclass correlation coefficient; r LSE, relative difference between observer LSE results by the AGA Institute /08/$34.00 doi: /j.cgh

2 1264 BOURSIER ET AL CLINICAL GASTROENTEROLOGY AND HEPATOLOGY Vol. 6, No. 11 Table 1. Study Design: Assignment of Judges According to Patient and Observer Groups Patient group Patients (n) First observer (O1) Liver Stiffness Evaluation Observer group Second observer (O2) P1 22 Judge A Judge B P2 24 Judge C Judge D NOTE. Interobserver agreement was evaluated between O1 and O2. Examination conditions. Examination conditions were those recommended by Echosens. 7 The second observer was blinded to the previous results of the first observer. Definitions. An LSE corresponded to all the measurements recorded during an examination. The LSE success rate (%) was calculated as the ratio of number of valid measurements/total number of measurements. The LSE result was expressed in kilopascals (kpa) and corresponded to the median of all the valid measurements performed within the examination. Metavir fibrosis stages were determined from the LSE result by applying liver stiffness cutoffs previously published by Foucher et al 10 that were calculated in various causes of chronic liver disease as in our study. Interquartile range (IQR) is an index of intrinsic variability of LSE corresponding to the interval around the LSE result containing 50% of the valid measurements between the 25th and 75th percentiles. LSE IQR (kpa) was expressed when necessary as LSE IQR ratio (%): LSE IQR/ LSE result. When 10 valid measurements were obtained, LSE was stopped, and the result was considered as reliable. However, a LSE could not exceed a total of 20 (valid and invalid) measurements. In a previous study, accuracy of LSE for the diagnosis of significant fibrosis or cirrhosis was similar whether 3, 5, or 10 valid measurements were recorded. 11 Thus, as in previous studies, 10,12 we defined an LSE result as acceptable when the LSE success rate was at least 30%, corresponding to 6 valid measurements according to our protocol. Liver Stiffness Evaluation Influencing Factors Measurement site (LSE #1 #4). All judges had to perform an LSE on 4 different sites (Figure 1; Table 2). Site IV corresponded to the usual site for percutaneous liver biopsy. 13 Screen withdrawal (LSE #5). Judges A and D performed a fifth LSE on site IV, with LSE results not shown on the Fibroscan screen. Location by standard ultrasonography (LSE #6). Judges A and D performed a sixth LSE on the best site located by standard ultrasonography, defined by a minimum liver parenchyma thickness of 6 cm without large vessel. 7 LSE repeatability (LSE #7). Judges A and D performed a seventh LSE in 20 patients. This LSE required 20 valid measurements and was performed on the best LSE site according to the observer s opinion. Interequipment reproducibility (LSE #8, #9). Judge A had to perform 2 additional LSEs on a single measurement site of his/her choice, each one with a different Fibroscan device. This evaluation was realized in 17 patients, and results of the first LSE were not communicated to the judge. Probe pressure. In case of difficult LSE (defined as 10 valid measurements after 15 attempts), the judge had to apply a higher probe pressure on the skin and to perform another 15 measurements. High pressure was defined as a grade 9 12 on the 12-grade scale of recommended pressure. This procedure was performed only on the first site for which difficult LSE was observed. Fibrosis Blood Tests To determine which observers provided the reference LSE results, we evaluated the correlation between O1 or O2 LSE results and fibrosis blood tests taken as an independent fibrosis reference: APRI 3 and FibroMeter. 5 Statistics Agreement was estimated with the intraclass correlation coefficient (Ric) for quantitative variables, kappa index for binary variables, and weighted kappa index ( w ) for ordinal variables. Ric combines correlation and similarity but does not take into account chance-expected agreement unlike. 14 Its meaning is Ric 0.87, excellent; 0.87 Ric 0.71, good; 0.71 Ric 0.50, fair; and Ric 0.5, poor agreement. 15 We analysed the independent agreement predictors in the set of 368 LSEs performed by observers O1 and O2 in sites I IV (Table 2). The statistical comparison of agreement required us to use observer discrepancy. Discrepancy was expressed in 2 ways, either crude difference (c LSE expressed in kpa): absolute value of (O2 LSE result O1 LSE result) or relative difference (r LSE expressed in %): absolute value of [(O1 LSE result O2 LSE result)/o2 LSE result]. Statistical software used was SPSS for Windows, version (SPSS Inc, Chicago, IL). Results Patients Forty-six patients were enrolled in the study, 22 in the patient group P1 and 24 in the group P2 (Table 1). Their characteristics at inclusion are summarized in Table 3. Globally, 534 LSEs were performed during the study (Table 2); 184 LSEs were performed on sites I IV by each observer O1 and O2 (46 Figure 1. Location of the liver stiffness measurement sites. LSEs were performed on the first or the second intercostal space under the upper limit of the liver dullness.

3 November 2008 REPRODUCIBILITY OF LIVER STIFFNESS MEASUREMENT 1265 Table 2. Types of LSEs Performed During the Study LSE # Judges LSE (n) a Examination conditions Site Decubitus Axillary line Intercostal space b 1 All 92 I Lateral Median First 2 All 92 II Dorsal Anterior First 3 All 92 III Dorsal Median Second 4 All 92 IV Dorsal Median First 5 A, D 46 IV Screen withdrawal 6 A, D 46 V Best site located by preliminary standard ultrasonography 7 A, D 40 VI 20 valid measurements on the best site (observer opinion) 8, 9 A 34 VII 2 LSEs on the same site, each one with a different Fibroscan device a Total: 534. b Under the upper limit of the liver dullness. patients 4 sites, Table 2): LSE results varied from kpa (median, 11.9), and LSEs were acceptable for both O1 and O2 in 73.4% of the cases. O2 LSE result was considered as the reference LSE result because it was independently associated with the fibrosis blood tests (data not shown). Liver Stiffness Evaluation Agreement The material is the 184 LSEs performed by O1 and O2 observers on sites I IV. Overall interobserver agreement was considered as excellent with Ric of 0.93 (Table 4; Figure 2). After conversion of LSE result into Metavir fibrosis stage, 10 O1/O2 agreement was excellent ( w, 0.80), and fibrosis stages were concordant in around 75% of cases. O1/O2 discordance rate for Metavir fibrosis stages was high in F2 and F3, moderate in F0/1, and low in F4 (Figure 3). Influencing Factors Liver stiffness evaluation result. Interobserver agreement for LSE results increased linearly as a function of liver stiffness (Figure 4). Interobserver agreement was fair in reference LSE results 9 kpa with Ric of 0.53 and excellent when 9 kpa with Ric of Liver stiffness evaluation interquartile range. Interobserver agreement for LSE result was excellent when LSE IQR ratio was 25% and decreased in LSE IQR ratio 25% (Figure 5). Body mass index. Interobserver agreement for LSE result was excellent when the BMI was 25 and decreased in BMI 25 (Figure 6). Liver stiffness evaluation site. Interobserver agreement for LSE results was excellent in each site, but it was the highest in site IV and the lowest in site I (Table 5). LSE agreement was excellent between all sites (Table 5). LSE agreement was excellent between sites IV and II or III (Table 6), suggesting no influence of axillary line and intercostal space tested. LSE agreement between sites I and IV was lower than the previous ones, supporting again a lower reproducibility in lateral than in dorsal decubitus. Screen withdrawal, additional ultrasonography. LSE agreements between LSE #4 and #5 or #6 were excellent (Ric 0.99), suggesting no influence of knowledge of previous measurement on LSE result and no interest of location by standard ultrasonography, respectively. Liver stiffness evaluation repeatability. Agreement between LSE results of the 10 first and the 10 last valid measurements of LSE #7 was excellent (Ric 0.94), suggesting an excellent repeatability and intraobserver agreement of LSE. Table 3. Patient Characteristics at Inclusion as a Function of Patient Group Patient group Both P1 P2 P value a No. of patients Age (y) 34, ,83 34, ,74 34, ,83.91 Male sex (%) BMI (kg/m 2 ) 16.9, , , , , , Cause (%).13 Alcohol Virus Others AST (IU/L) 12, ,271 12, ,271 22, , Bilirubin ( mol/l) 3, ,329 3, ,329 4, , FibroMeter 0.0, , , , , , APRI score 0.15, , , , , , NOTE. Quantitative variables are expressed as median (in bold), interquartiles, and extremes (in italic). a Mann-Whitney test or Fisher test between the patient groups.

4 1266 BOURSIER ET AL CLINICAL GASTROENTEROLOGY AND HEPATOLOGY Vol. 6, No. 11 Table 4. LSE Results and Interobserver Agreement as a Function of Patient Group O1 LSE (kpa) a O2 P value b Ric Patient group c Both P P P value d a O1: judges A (for P1) and C (P2); O2: judges B (P1) and D (P2). b Paired comparison between O1 and O2 by Wilcoxon test (P1 and P2) or t test (all). c Patient groups: P1: judges A (O1) and B (O2); P2: judges C (O1) and D (O2). d Unpaired comparison between P1 and P2 by t test. Interequipment reproducibility. LSE agreement between the 2 Fibroscan devices was excellent (Ric 0.92). Probe pressure. Difficult LSEs were noted in 43 of the 368 LSE performed on sites I IV (11.7%), mainly on site I (86%). A high probe pressure did not significantly increase LSE success rate (P.43) and did not alter LSE result (Ric 0.96). Judge pair. Overall interobserver agreement was excellent in patient group P1 and good in P2 (Table 4). However, LSE results were significantly lower in P2 than in P1 patients (Table 4), precluding a reliable comparison of reproducibility between both judge pairs. Therefore, this confusing factor, judge pair, was evaluated. On one hand, O1 LSE result (P.001), as expected, and judge pair (P.002) were independent predictors of O2 LSE result. On the other hand, O2 LSE result was the only independent predictor of O1 LSE result (P.001), with no independent role for judge pair (P.51). Figure 3. Discordance rate for Metavir fibrosis stage between observers O1 and O2 as a function of anatomic site and reference (O2) Metavir fibrosis stage. Discordance was defined by a difference of at least 1 fibrosis stage between O1 and O2 results. Acceptable liver stiffness evaluation. The rate of acceptable LSE was significantly different between observers O1 and O2, 87.5% and 79.9%, respectively (P.03). However, this difference did not influence interobserver reproducibility because all the previous results were similar whether the statistical analysis was restricted to acceptable LSE in both O1 and O2 observers (detailed data not shown). Observer Discrepancy Crude discrepancy. c LSE was independently associated with O2 LSE IQR at the first step (P.001), O2 LSE result at the second step (P.002), BMI at the third step (P.006), and LSE site at the fourth step (P.01). c LSE grew almost linearly as a function of O2 LSE result (Figure 7). Finally, c LSE was significantly different between the 4 LSE sites and the smallest in site IV (Table 5). Relative discrepancy. r LSE was independently associated only with O2 LSE result (P.02). r LSE, plotted against O2 LSE result, initially decreased and then reached a plateau around 20% 30% by 5 kpa (Figure 7). In addition, r LSE was significantly different between the 4 LSE sites (the smallest for site IV; Table 5). Figure 2. Correlation of LSE results between observers O1 and O2, providing an agreement of Ric of Figure 4. Interobserver agreement for LSE results as a function of different intervals of LSE result (reference: O2) and anatomic sites.

5 November 2008 REPRODUCIBILITY OF LIVER STIFFNESS MEASUREMENT 1267 Figure 5. Interobserver agreement for LSE results as a function of LSE IQR ratio (reference: O2). Agreement on Best Site As already noted, interobserver agreement was the highest in site IV (Table 5). Interobserver agreement (Ric) for LSE results also increased as a function of liver stiffness in site IV, but it was always higher than that of other sites (Figure 4). LSE IQR ratio, BMI, and judge pair had no influence on agreement in site IV (Figures 5, 6, and Table 5, respectively). All these results were similar when only acceptable LSEs in both O1 and O2 observers were included in the statistical analysis. Discussion A recent study has already shown that interobserver LSE agreement was excellent and was influenced by the BMI and especially by the liver fibrosis level. 16 However, these 2 factors were considered only as binary variables in this work (Metavir F 2 and BMI 25), thus providing little information for clinical practice. By evaluating the influence of several factors on LSE reproducibility, our study determined the precise conditions required to obtain a reproducible LSE result. Interobserver Agreement Our study confirmed the excellent overall interobserver LSE agreement reported recently by Fraquelli et al 16 (Ric 0.98), with, however, a slightly lower result, Ric In fact, Fraquelli et al limited their results to a per protocol analysis by using a restrictive and arbitrary definition of acceptable LSE with a success rate 60% and an IQR 30%. This was not an intention to diagnose analysis according to the STARD recommendations. 17 In addition, whereas Fraquelli et al used only 1 fixed site for each observer, in our study agreement was evaluated in 4 sites with slightly variable LSE reproducibility. Influence of Liver Stiffness Level Fraquelli et al 16 showed that interobserver agreement was higher in Metavir F 2 fibrosis stages compared with F 1 stages. Our study confirmed this result; agreement increased with the level of liver stiffness and was fair for liver stiffness under 9 kpa. In addition, LSE result was the only factor independently associated with the relative difference between the 2 observers LSE results and was identified as the second independent predictor of the crude difference. Thus, overall interobserver agreement for LSE depends on the relative prevalence of fibrosis stages in the population studied, as recently demonstrated for fibrosis blood tests. 18 The increase in crude difference c LSE as a function of LSE result could suggest a lower interobserver agreement for high LSE results (Figure 7). However, interobserver agreement was better in high LSE results (Figure 4). Interobserver agreement is thus better reflected by the relative difference r LSE, which was high ( 60%) in low LSE results and thereafter rapidly decreased to stabilize around 20% 30% in LSE 5 kpa (Figure 7). This 30% variation of LSE could have a high implication in clinical practice, especially for medium LSE. For example, a 30% interval around a stiffness measured at 10 kpa (ie, equivalent Metavir F2 3 stages) would result in LSE result from 7 13 kpa, ie, from F0 1 to F4 in equivalent Metavir stages. 9 With that respect, after conversion of liver stiffness in Metavir fibrosis stages according to Foucher cutoffs, 10 our results showed a higher discordance rate between the 2 observers for F2 and F3 stages (Figure 3). Therefore, diagnosis of noncirrhotic fibrosis stages by LSE must be carefully interpreted. Influence of Anatomic Site Site IV was the most suitable site according to LSE reproducibility: (1) interobserver agreement (Ric) was the highest in site IV; (2) crude and relative differences between the 2 observers LSE results were significantly the smallest at site IV; (3) the influence of liver stiffness level, LSE IQR, BMI, and judge pair at site IV was much weaker on interobserver agreement than that observed in all sites (Figures 4, 5, and 6; Table 5). Thus, our study demonstrated that the best measurement site, considering LSE reproducibility, is the median axillary line on the first intercostal space under the upper limit of the liver dullness with the patient lying in the dorsal decubitus position, which corresponds to the site recommended by the FibroScan manufacturer and to the one usually used for liver biopsy. Figure 4 shows that agreement for LSE result was the best for high LSE values recorded on the best measurement site. Thus, on an individual basis, our results suggest that LSE is more devoted to the diagnosis of cirrhosis. In this setting, a recent meta-analysis has shown that LSE was more accurate to diagnose cirrhosis than significant fibrosis. 8 Figure 6. Interobserver agreement for LSE results as a function of BMI.

6 1268 BOURSIER ET AL CLINICAL GASTROENTEROLOGY AND HEPATOLOGY Vol. 6, No. 11 Table 5. Influence of Measurement Site on Interobserver Agreement as a Function of Patient Group (Ric) and of Interobserver Difference Measurement site I II III IV All Patient group P P Both Interobserver difference c LSE a (kpa) b r LSE c (%) b a Crude difference between O1 and O2 LSE results. b P.03 (paired Friedman test between the 4 sites). c Relative difference between O1 and O2 LSE results. Influence of Judge Expertise Interobserver LSE agreement was higher in patient group P1 than in P2 (Table 4). This could be partially explained by the higher mean LSE result in P1 than in P2. Besides, judge pair has an independent influence only on LSE result in O2 observers but not in O1 observers. Taken together, those results indicate a source of variability in P2 and O2, ie, in judge D. Thus, the low expertise effect was more sensitive in this judge as a result of the lower mean level of LSE results. Conversion of Liver Stiffness Evaluation Result Into Metavir Fibrosis Stage On the best measurement site IV, interobserver LSE agreement was excellent. After converting LSE results into Metavir fibrosis stages, interobserver agreement was still excellent. However, calculated fibrosis stages were still discordant in around 15% of the cases (data not shown). This discordance rate raises the difficulty of the interpretation of the LSE result. Actually, we do not know how to use the LSE result in clinical practice; therefore, LSE results are sometimes converted into Metavir fibrosis stage according to published cutoffs. 9 Nevertheless, these cutoffs are heterogeneous among the studies, and besides cause of chronic liver disease 19 and fibrosis stage prevalence, observer variability might participate in such heterogeneity. In addition, the use of 3 cutoffs for 3 diagnostic targets ( F2, F3, F4) cumulates 3 misclassification rates. Thus it seems necessary to privilege the use of unconverted LSE result, but this requires determining its direct clinical value. In this respect, recent studies have evaluated the relationship between LSE result and portal hypertension: portal pressure and large esophageal varices. 10,24 Other Influencing Factors Crude difference of LSE between observers provided more independent factors than relative difference, and they were clinically plausible. Thus, the former is a finer descriptor of influencing factors than the latter. Reference LSE IQR and BMI were independent predictors. The role of BMI has already been suggested. 16 Our results showed that an LSE IQR ratio 25% is required to consider a LSE result as reliable whatever the site. They even suggested that an LSE IQR ratio 10% produces a very reliable result at the best site as a result of an optimal interobserver agreement (Ric of 0.999). Conclusion Overall interobserver agreement for liver stiffness result assessed by ultrasonographic elastometry is globally excellent, but agreement depends on several factors, especially stiffness level, because agreement is fair for LSE results 9 kpa, whatever the anatomic site. Other factors decreasing agreement are a high LSE range (IQR), high BMI, and anatomic site. The best site is that generally used for liver biospy. LSE IQR ratio 25%, especially 10%, produces reproducible LSE in the best site. Converting LSE result into fibrosis Metavir stage decreases interobserver agreement, especially in F2 and F3 stages, and thus should only be used cautiously as an ancillary descriptor of liver fibrosis by ultrasonographic elastometry. Table 6. Intersite Agreement (Ric) for LSE Result Between Anatomic Site IV and Other Sites, With the Ensuing Anatomic Condition Tested Site I II III Site IV vs other site (Ric) Anatomic factor tested Lateral decubitus Axillary line Intercostal space Figure 7. Relationship of the crude or relative difference between the 2 observers LSE results with the LSE result (in reference O2).

7 November 2008 REPRODUCIBILITY OF LIVER STIFFNESS MEASUREMENT 1269 References 1. Rockey DC, Bissell DM. Noninvasive measures of liver fibrosis. Hepatology 2006;43:S113 S Afdhal NH, Nunes D. Evaluation of liver fibrosis: a concise review. Am J Gastroenterol 2004;99: Wai CT, Greenson JK, Fontana RJ, et al. A simple noninvasive index can predict both significant fibrosis and cirrhosis in patients with chronic hepatitis C. Hepatology 2003;38: Poynard T, Morra R, Halfon P, et al. Meta-analyses of FibroTest diagnostic value in chronic liver disease. BMC Gastroenterol 2007;7: Cales P, Oberti F, Michalak S, et al. A novel panel of blood markers to assess the degree of liver fibrosis. Hepatology 2005; 42: Cales P, De Ledinghen V, Halfon P, et al. Evaluating accuracy and increasing the reliable diagnosis rate of blood tests for liver fibrosis in chronic hepatitis C. Liver Int (Epub ahead of press). 7. Sandrin L, Fourquet B, Hasquenoph JM, et al. Transient elastography: a new noninvasive method for assessment of hepatic fibrosis. Ultrasound Med Biol 2003;29: Friedrich-Rust M, Ong MF, Martens S, et al. Performance of transient elastography for the staging of liver fibrosis: a metaanalysis. Gastroenterology 2008;134: Castera L, Forns X, Alberti A. Non-invasive evaluation of liver fibrosis using transient elastography. J Hepatol 2008;48: Foucher J, Chanteloup E, Vergniol J, et al. Diagnosis of cirrhosis by transient elastography (FibroScan): a prospective study. Gut 2006;55: Kettaneh A, Marcellin P, Douvin C, et al. Features associated with success rate and performance of fibroscan measurements for the diagnosis of cirrhosis in HCV patients: a prospective study of 935 patients. J Hepatol 2007;46: de Ledinghen V, Douvin C, Kettaneh A, et al. Diagnosis of hepatic fibrosis and cirrhosis by transient elastography in HIV/hepatitis C virus-coinfected patients. J Acquir Immune Defic Syndr 2006;41: Nousbaum JB. The role of liver biopsy in the management of chronic hepatitis C. Gastroenterol Clin Biol 2002;26(Suppl 2): B Winkfield B, Aube C, Burtin P, et al. Inter-observer and intraobserver variability in hepatology. Eur J Gastroenterol Hepatol 2003;15: Rousselet MC, Michalak S, Dupre F, et al. Sources of variability in histological scoring of chronic viral hepatitis. Hepatology 2005;41: Fraquelli M, Rigamonti C, Casazza G, et al. Reproducibility of transient elastography in the evaluation of liver fibrosis in patients with chronic liver disease. Gut 2007;56: Bossuyt PM, Reitsma JB, Bruns DE, et al. The STARD statement for reporting studies of diagnostic accuracy: explanation and elaboration. Clin Chem 2003;49: Poynard T, Halfon P, Castera L, et al. Standardization of ROC curve areas for diagnostic evaluation of liver fibrosis markers based on prevalences of fibrosis stages. Clin Chem 2007;53: Beaugrand M. Fibroscan: instructions for use. Gastroenterol Clin Biol 2006;30: Bureau C, Metivier S, Peron JM, et al. Transient elastography accurately predicts presence of significant portal hypertension in patients with chronic liver disease. Aliment Pharmacol Ther 2008;27: 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: Lemoine M, Katsahian S, Nahon P, et al. Liver stiffness measurement is correlated with hepatic venous pressure gradient in patients with uncomplicated alcoholic and/or HCV related cirrhosis. Hepatology 2006;44:204A. 23. Vizzutti F, Arena U, Romanelli RG, et al. Liver stiffness measurement predicts severe portal hypertension in patients with HCVrelated cirrhosis. Hepatology 2007;45: Kazemi F, Kettaneh A, N Kontchou G, et al. Liver stiffness measurement selects patients with cirrhosis at risk of bearing large oesophageal varices. J Hepatol 2006;45: Address requests for reprints to: Paul Calès, MD, Service d Hépato- Gastroentérologie, Centre Hospitalier Universitaire, Angers Cedex 09, France. paul.cales@univ-angers.fr; fax: We thank Dermot O Toole for manuscript review. Supported by the following Grant: PHRC 2006 from the French Department of Health.