Diagnosis of liver fibrosis using FibroScan and other noninvasive methods in patients with hemochromatosis: A prospective study

Gastroentérologie Clinique et Biologique (2008) 32, 180—187 Disponible en ligne sur www.sciencedirect.com

ORIGINAL ARTICLE

Diagnosis of liver fibrosis using FibroScan and other noninvasive methods in patients with hemochromatosis: A prospective study Diagnostic de la fibrose hépatique par FibroScan et marqueurs sériques au cours de l’hémochromatose : étude prospective X. Adhoute a, J. Foucher a, D. Laharie a, E. Terrebonne a, J. Vergniol a, L. Castéra a, B. Lovato a, E. Chanteloup a, W. Merrouche a, P. Couzigou a, V. de Lédinghen a,∗,b a

Service d’hépatogastroentérologie, centre d’investigation de la fibrose hépatique, hôpital Haut-Lévêque, 33604 Pessac cedex, France b Inserm, U889, université Victor-Segalen Bordeaux-2, 33076 Bordeaux, France Available online 5 March 2008 Summary Background. — The role of hepatic iron overload in the development of hepatic fibrosis in patients with hemochromatosis is well-established. Transient elastography (FibroScan) is a new noninvasive, rapid, reproducible bedside method, allowing assessment of liver fibrosis by measuring liver rigidity. Objectives. — The aim of this prospective study was to evaluate liver fibrosis with FibroScan and other noninvasive biochemical methods in patients with hemochromatosis (C282Y homozygosity) compared with control patients. Patients and methods. — From January 2004 through October 2006, all consecutive patients with hemochromatosis were evaluated for liver fibrosis using noninvasive methods (FibroScan and biochemical markers). These patients were compared with patients who had chronic cytolysis and no fibrosis on liver biopsy. Results. — One hundred and three consecutive patients (57 cases and 46 controls) were fully investigated. Median FibroScan values were similar in both groups, 5.20 kPa versus 4.9 kPa, respectively. No differences were observed between cases and controls for all biochemical markers. A strong correlation was observed between FibroScan and many biochemical markers,

∗

Corresponding author. E-mail address: [email protected] (V. de Lédinghen).

0399-8320/$ — see front matter © 2008 Elsevier Masson SAS. All rights reserved. doi:10.1016/j.gcb.2007.12.021

Liver fibrosis and hemochromatosis

181

although ferritin levels did not correlate with FibroScan values. The prevalence of patients with FibroScan values greater than 7.1 kPa (cut-off level for significant fibrosis) was 22.8% in patients with hemochromatosis and 0% in the controls (P < 0.0001). Conclusion. — FibroScan and biochemical markers could be reliable noninvasive methods for detecting liver fibrosis in patients with hemochromatosis. Such patients have high FibroScan values more often than do control patients. Further longitudinal and prospective studies are necessary to confirm these preliminary data. © 2008 Elsevier Masson SAS. All rights reserved. Résumé Introduction. — Le rôle de la surcharge en fer dans l’apparition de la fibrose hépatique au cours de l’hémochromatose est bien connu. L’élastométrie impulsionnelle par FibroScan est une méthode non invasive, rapide, reproductible pour évaluer la fibrose hépatique en mesurant l’élasticité hépatique. But. — Le but de cette étude prospective était d’évaluer la fibrose hépatique par FibroScan et marqueurs sériques au cours de l’hémochromatose génétique (mutation C282Y homozygote) par rapport à des témoins. Malades et méthodes. — De janvier 2004 à octobre 2006, tous les malades consécutifs suivis pour hémochromatose ont été évalués par FibroScan et marqueurs sériques de fibrose. Ces malades ont été comparés à des malades ayant une cytolyse chronique sans fibrose à la ponction-biopsie hépatique. Résultats. — Cent trois malades (57 cas et 46 témoins) ont été étudiés. La valeur médiane de FibroScan était identique dans les deux groupes, respectivement 5,20 kPa versus 4,9 kPa. Aucune différence significative n’a été mise en évidence entre les deux groupes pour tous les marqueurs sériques de fibrose. Une forte corrélation a été mise en évidence entre les valeurs de Fibroscan et des marqueurs sériques de fibrose mais le taux de ferritine n’était pas corrélé à la valeur de FibroScan. La prévalence des malades avec FibroScan était supérieure à 7,1 kPa (valeur seuil pour le diagnostic de fibrose significative) était de 22,8 % chez les patients avec hémochromatose et de 0 % chez les témoins (p < 0,0001). Conclusion. — Le FibroScan et les marqueurs sériques de fibrose sont de bonnes méthodes pour évaluer la fibrose hépatique au cours de l’hémochromatose. Les malades suivis pour hémochromatose ont des valeurs de FibroScan plus élevées que les témoins. D’autres études longitudinales et prospectives sont encore nécessaires pour confirmer ces résultats préliminaires. © 2008 Elsevier Masson SAS. All rights reserved.

Introduction Hemochromatosis (HFE-linked iron overload) is the most common inherited autosomal—recessive disorder, affecting about one in 300 individuals of Northern European ancestry [1—3]. Hemochromatosis is associated with mutation of the HFE gene [4]. The mutation is a single-base change that results in the substitution of tyrosine by cysteine at position 282 of the HFE protein (C282Y). Other mutations such as the H63D and S65C mutations have been described [4]. These mutations may have clinical significance only in conjunction with other genetic or environmental factors. Liver biopsy is no longer considered essential for diagnosis of hemochromatosis. Nevertheless, it still provides an assessment of liver fibrosis. Liver biopsy is generally restricted to patients with hemochromatosis and to those suspected of having significant fibrosis or cirrhosis, although ferritin, platelet and aspartate aminotransferase (AST) levels are sufficient to make a diagnosis of cirrhosis in more than 80% of patients [5]. Liver biopsy can also be used by the clinician to assess iron overload, but there is wide variability in measures of hepatic iron load from hepatic biopsy [6]. Moreover, a rapid, noninvasive and costeffective technique based on magnetic resonance imaging

(MRI) is now available for assessing iron load [7]. Thus, liver biopsy is not required for the evaluation of liver iron load. The accuracy of liver biopsy for assessing fibrosis has also come under question because of sampling errors, and intra- and interobserver variability, that can lead to overor underestimation of the stage of fibrosis [8—10]. For this reason, more accurate noninvasive methods, including routine biochemical (APRI, FIB-4, GUCI) and surrogate serum fibrosis markers (FibroTest, Hepascore, Forns score, Lok score), have been developed [11—17]. Transient elastography (FibroScan) is a rapid, noninvasive and reproducible method recently developed for measuring liver rigidity. Recent studies have reported the good performance of FibroScan for the diagnosis of fibrosis and cirrhosis in patients with hepatitis C virus (HCV) infection, with or without human immunodeficiency virus (HIV) infection, in patients with chronic cholestatic disease or under treatment with methotrexate [18—24]. In France, either FibroScan or FibroTest is now recommended for the initial evaluation of liver fibrosis in HCV-naïve patients with no other disease [25]. To our knowledge, however, the usefulness of FibroScan and biochemical markers has never been evaluated in patients with hemochromatosis.

182 The aim of this prospective study was to evaluate liver fibrosis using FibroScan and other noninvasive biochemical methods in patients with hemochromatosis (C282Y homozygosity) and to compare them with control patients.

Patients and methods Inclusion criteria From January 2004 to October 2006, all consecutive patients with hemochromatosis (C282Y homozygosity) were prospectively included in the present study. Some patients were ‘new’ patients (recently diagnosed with hemochromatosis) who had undergone noninvasive evaluation of fibrosis before the first phlebotomy. Some patients were ‘old’ patients (iron-depleted patients), treated with phlebotomy for several months or even years, who had been noninvasively evaluated for fibrosis during the follow-up of their disease. All consecutive patients with persistent, unexplained, elevated alanine aminotransferase (ALT) levels who had had a liver biopsy, but showed no fibrosis (METAVIR score) and a normal or subnormal liver parenchyma were included as controls (and included in the cohort CYTOL) [26]. These patients were all HBsAg-, HCV-Ab- and HIVAb-negative. Also, none of these patients was either an alcohol (< 40 g/day for men and > 20 g/day for women) or drug user, and none had autoimmune liver disease, Wilson’s disease or ␣1 -antitrypsin deficiency. Furthermore, no patient had abnormal thyroid-stimulating hormone levels, cardiac insufficiency or neoplasia. In the control group, liver biopsy sample size was greater or equal to 10 mm. Eligible patients were at least 18 years old. All enrolled patients gave their written informed consent. Clinical data were recorded at inclusion, including age, weight, body mass index (BMI), and presence of diabetes and hypertension.

Noninvasive assessment of liver fibrosis Liver rigidity measurement using FibroScan (EchoSensTM , Paris, France) was performed as previously described [27]. A FibroScan value greater than 7.1 kPa was considered the cut-off level for significant fibrosis [19]. The following parameters were determined by the same laboratory from blood samples taken on the same day that the FibroScan examination was performed: AST; ALT; gamma-glutamyl transpeptidase level (GGT), total bilirubin, ␣2 -macroglobulin, apolipoprotein A1, haptoglobin, ferritin and the platelet count. The AST/platelet ratio index (APRI) was calculated as follows: AST/upper limit of normal (ULN) × 100/platelet count (109 /L) [11]. The FIB-4 index was calculated as follows: [age × AST (IU/L)]/[platelet count (109 /L) × ALT (IU/L)]1/2 [12]. The Forns score was calculated as follows: 7.811—3.131 × ln [platelet count (109 /L)] + 0.781 × ln [GGT (IU/L)] + 3.467 × ln (age) — 0.014 × (cholesterol) [15]. Serum parameters comprising the FibroTest score (Biopredictive, Paris, France) were assessed according to the laboratory’s preanalytical and analytical recommendations [13]. The FibroTest score was computed on the Biopredictive website (www.biopredictive.com). The Hepascore was calcula-

X. Adhoute et al. ted as follows: y = exp [4.185818 − (0.0249 × age) + (0.7464 × sex) + (1.0039 × ␣2 — macroglobulin) + (0.0302 × hyaluronic acid) + (0.0691 × bilirubin) — (0.0012 × GGT)], with age in years, male gender = 1, female gender = 0, ␣2 -macroglobulin in g/L, hyaluronate in ␮g/L, bilirubin in ␮mol/L and GGT in U/L. The Hepascore was calculated using the following equation: y/(1 + y) [14]. The Göteborg University Cirrhosis Index (GUCI) was calculated as follows: AST (IU/L) × Prothrombin time (%) × 100/platelet count (109 /L) [17].

Statistical analysis Descriptive values are expressed as percentages, and mean (± SD) or median (range) values. Results are expressed as means ± SEM. Comparisons of quantitative data were made using Student’s t test, or the nonparametric Mann—Whitney rank-sum test when the data did not exhibit a normal distribution. Qualitative data were analyzed using the chi-square test. Kendall’s tau-b coefficient of correlation and their probabilities (P) were used to evaluate the relationships between FibroScan and FibroTest, Hepascore, FIB-4, GUCI, Lok score, Forns score or APRI. For the whole study population, the odds ratio, together with the 95% confidence interval (CI) and the corresponding P value, were calculated for assessment of relative risk using logistic regression. All reported P values are two-tailed. The data were recorded and analyzed using the SPSS v14.0 software package (SPSS Inc; Chicago, IL, USA).

Results Characteristics of patients Sixty-one patients with hemochromatosis and 46 control patients were included. Four patients with hemochromatosis (7%) and none of the control patients were excluded because of unsuccessful liver rigidity measurement due to obesity. Demographic and clinical characteristics of the 103 analyzed patients are presented in Table 1. The majority of patients with hemochromatosis were men (57.9%), mean age was 54.3 ± 13.7 years and BMI was 25.0 ± 3.5 kg/m2 . The majority of the control patients were women (58.7%), mean age was 46.7 ± 15.3 years and BMI was 23.7 ± 4.2 kg/m2 . Ten patients with hemochromatosis (17.5%) had recent diagnoses of hemochromatosis, and 47 patients (82.5%) were iron-depleted. The recently diagnosed patients had been evaluated for liver fibrosis before the first phlebotomy. Characteristics of patients in these two groups are presented in Table 2. In the recently diagnosed hemochromatosis patients, ferritin levels were higher than in patients under treatment: 1528 ± 1181 ng/ml versus 303 ± 594 ng/ml, P < 0.0001, respectively.

Noninvasive assessment of liver fibrosis In the hemochromatosis patients and control patients, the median values by FibroScan were 5.20 kPa (range: 2.3—75 kPa) and 4.9 kPa (range: 2.6—7.0 kPa), respecti-

Liver fibrosis and hemochromatosis

183

Table 1 Characteristics of the 103 study patients. Caractéristiques des 103 malades.

Men (%) Mean age (years) Mean body mass index (kg/m2 ) Diabetes (%) Hypertension n, (%) Alcohol use (drinks/week) Platelets (g/L) Prothrombin time (%) GGT (IU/L) ALT (IU/L) AST (IU/L) Hyaluronate (␮g/L) Ferritin (ng/mL)

Patients with hemochromatosis (N = 57)

Control patients (N = 46)

P

33 (57.9) 54.3 ± 13.7 25.0 ± 3.5 9 (15.8) 13 (25)b 4.5 ± 10.0 241 ± 77 95.2 ± 7.4 111 ± 337 35.3 ± 37.7 36.8 ± 48.8 79.8 ± 167.1 538.6 ± 876.5

19 (41.3) 46.7 ± 15.3 23.7 ± 4.2 2 (6.7)a 5 (25)c 1.1 ± 3.3 249 ± 96 95.0 ± 8.3 168 ± 165 74.3 ± 63.7 45.5 ± 34.2 37.5 ± 36.4 375.9 ± 595.5

NS 0.009 NS NS NS NS NS NS NS < 0.0001 NS NS NS

NS: not significant. ALT: alanine aminotransferase; AST: aspartate aminotransferase; GGT: gamma-glutamyl transpeptidase. a Missing data in 14 cases. b Missing data in two cases. c Missing data in 21 cases.

vely (P = NS). Results with the use of noninvasive methods for fibrosis evaluation are shown in Table 3. No statistical difference was observed between the two groups in any of the noninvasive tests (Table 3). Also, in both groups, a significant correlation was found between FibroScan and FibroTest values (r = 0.257, P < 0.0001), Forns score (r = 0.160, P = 0.03), Hepascore (r = 0.200, P = 0.01) and GUCI (r = 0.169, P = 0.02). However, no correlation was found between FibroScan values and the Lok score (r = 0.095, P = NS), FIB-4 (r = 0.119, P = NS) or APRI (r = 0.132, P = NS). Noninvasive evaluations of liver fibrosis in patients with hemochromatosis are given in Table 4. Whatever the method (except for APRI and GUCI), there was no significant difference between the two groups, although a slight correlation was found between FibroScan and ferritin levels (r = 0.183, P = 0.02).

Characteristics of patients with FibroScan values greater than 7.1 kPa. The prevalence of patients with FibroScan values greater than 7.1 kPa was 22.8% in patients with hemochromatosis (N = 13) and 0% in control patients (P < 0.0001), and their characteristics are presented in Table 5. In those diagnosed with cirrhosis, a discrepancy between FibroScan and biochemical markers was observed in only one case.

Factors associated with FibroScan values greater than 7.1 kPa. As indicated in Table 6 and using univariate analysis, the only factor associated with FibroScan values greater than 7.1 kPa

Table 2 Characteristics of the 57 patients with hemochromatosis. Caractéristiques des 57 malades avec hémochromatose.

Men (%) Mean age (years) Mean body mass index (kg/m2 ) Diabetes (%) Hypertension n, (%) Alcohol use (drinks/week) Platelets (g/L) Prothrombin time (%) GGT (IU/L) ALT (IU/L) AST (IU/L) Ferritin (ng/mL)

Patients with a recent diagnosis (N = 10)

Iron-depleted patients (N = 47)

P

6 (60.0) 51.7 ± 15.2 24.1 ± 3.4 1 (10) 2 (25)a 1.2 ± 1.8 207 ± 387 94.6 ± 8.8 260.2 ± 714.1 48.9 ± 37.7 59.3 ± 86.6 1528.9 ± 1181.3

27 (57.4) 54.9 ± 13.5 25.2 ± 3.6 8 (17.0) 11 (25)a 5.2 ± 11.1 249 ± 81 95.4 ± 7.2 81.1 ± 180.9 32.4 ± 37.4 32.0 ± 36.1 302.8 ± 594.2

NS NS NS NS NS NS NS NS NS NS NS < 0.001

NS: not significant. ALT: alanine aminotransferase; AST: aspartate aminotransferase; GGT: gamma-glutamyl transpeptidase. a Missing data in some cases.

184

X. Adhoute et al.

Table 3 Noninvasive evaluation of liver fibrosis. Évaluation non invasive de la fibrose.

Median FibroScan value (kPa) Mean FibroTest score Mean Hepascore Mean APRI score Mean FIB-4 score Mean Forns score Mean Lok score Mean GUCI score

Patients with hemochromatosis (N = 57)

Control patients (N = 46)

P

5.2 0.30 ± 0.29 0.39 ± 0.37 0.42 ± 0.60 1.46 ± 1.17 4.40 ± 1.86 0.33 ± 0.22 0.46 ± 0.74

4.9 0.27 ± 0.21 0.24 ± 0.28 0.61 ± 0.71 1.43 ± 1.86 4.47 ± 1.36 0.26 ± 0.19 0.59 ± 0.60

NS NS NS NS NS NS NS NS

NS: not significant.

Table 4 Noninvasive evaluation of liver fibrosis in patients with hemochromatosis. Évaluation non invasive de la fibrose chez les malades avec hémochromatose. Patients with a recent diagnosis (N = 10) FibroScan (kPa) FibroTest Hepascore APRI FIB-4 Forns score Lok score GUCI

10.2 0.32 0.45 0.76 1.83 4.12 0.37 0.93

± ± ± ± ± ± ± ±

11.3 0.36 0.35 1.19 1.64 1.69 0.23 1.59

Iron-depleted patients (N = 47) 8.3 0.30 0.38 0.34 1.38 4.46 0.33 0.36

± ± ± ± ± ± ± ±

13.9 0.27 0.38 0.35 1.05 1.91 0.22 0.39

P NS NS NS 0.05 NS NS NS 0.03

NS: not significant.

was diabetes (odds ratio 4.25, 95% CI 1.04—17.46, P = 0.04). Using multivariate analysis and adding ferritin levels greater than 150 ng/mL, FibroScan values greater than 7.1 kPa were associated with diabetes (odds ratio 8.39, 95% CI 1.35—52.1, P = 0.02) and ferritin levels greater than 150 ng/mL (odds ratio 6.31, 95% CI 1.12—35.39, P = 0.04).

Follow-up of hemochromatosis patients Forty-four patients were followed-up for one to two years. These patients (seven patients with a recent diagnosis of hemochromatosis and 37 iron-depleted patients) had FibroS-

can evaluations every year. The evolution of their FibroScan values is shown in Figs. 1 and 2. In patients with a recent diagnosis of hemochromatosis, five had initial FibroScan values less than 6 kPa that did not change after one or two years, and one patient had an initial value of 8.8 kPa that was 4.9 kPa a year later. The final patient had an initial value of 21.1 kPa that was 21.3 kPa one year later and 8.8 kPa after two years. In iron-depleted patients, FibroScan values did not change significantly: for example, two patients had initial elevated FibroScan values of 53.2 kPa and 21.8 kPa that were 70.6 kPa and 5.7 kPa, respectively, a year later.

Table 5 Characteristics of biochemical markers of fibrosis in patients with hemochromatosis and FibroScan values > 7.1 kPa. Valeurs des marqueurs sériques de fibrose chez les malades avec hémochromatose et valeur de FibroScan > 7,1 kPa. #

FibroScan value (kPa)

Patient characteristics

1 2 3 4 5 6 7 8 9 10 11 12 13

7.2 7.4 7.9 8.1 8.6 8.8 8.8 10.2 14.4 21.1 38.6 70.6 75.0

No fibrosis according to biochemical tests (except Hepascore) No fibrosis according to biochemical tests (except Hepascore) No fibrosis according to biochemical tests No fibrosis according to biochemical tests BMI = 30 kg/m2 ; no fibrosis according to biochemical tests Significant fibrosis according to biochemical tests (except Lok, APRI) Significant fibrosis according to biochemical tests Severe fibrosis according to biochemical tests (except Lok, APRI) Cirrhosis at liver biopsy in 1998 Significant fibrosis according to biochemical tests (except Lok, APRI) Cirrhosis according to biochemical tests Cirrhosis according to biochemical tests Past history of ascites (hepatocellular carcinoma)

Note: biochemical tests included FibroTest, Hepascore, APRI, FIB-4, Forns score, Lok score, GUCI.

Liver fibrosis and hemochromatosis

185

Table 6 Factors associated with FibroScan values > 7.1 kPa. Facteurs associés à des valeurs de FibroSan > 7,1 kPa. Odds ratio

95% confidence interval

P

Univariate analysis Male gender Age > 50 years Body mass index > 2 5 kg/m2 Diabetes Platelets < 200 g/L Ferritin > 150 ng/mL

1.67 1.97 1.50 4.25 1.29 3.84

0.51—5.51 0.56—6.86 0.46—4.85 1.04—17.46 0.36—4.61 0.93—15.91

0.40 0.29 0.50 0.04 0.69 0.06

Multivariate analysis Diabetes Ferritin > 150 ng/mL

8.39 6.31

1.35—52.10 1.12—35.39

0.02 0.04

Fig. 1 Evolution of FibroScan values in patients recently diagnosed with hemochromatosis (N = 7). Évolution des valeurs de FibroScan chez les malades avec diagnostic récent d’hémochromatose (n=7).

Fig. 2 Evolution of FibroScan values in iron-depleted patients (N = 35). Two patients are not represented because of high FibroScan values. Évolution des valeurs de FibroScan chez les malades en traitement d’entretien de l’hémochromatose (n=35). Deux malades n’ont pas été représentés car avaient des valeurs très élevées de FibroScan.

Discussion To our knowledge, this is the first study evaluating noninvasive methods, especially FibroScan, in patients with hemochromatosis. FibroScan and biochemical markers are

efficient techniques for diagnosing significant fibrosis or cirrhosis. In our study using FibroScan and biochemical markers, no differences were observed between patients with hemochromatosis and control patients. Because of the improvement in screening for hemochromatosis in the 1990s, patients with hemochromatosis treated with phlebotomy in the 2000s do not have severe disease or severe fibrosis. Only five patients (8.8%) had FibroScan values greater than 12.5 kPa (cut-off value for the diagnosis of cirrhosis). In four of these patients, biochemical markers correlated with FibroScan values and three patients had symptoms of cirrhosis. In the present study, for the diagnosis of significant fibrosis, we used a cut-off value of 7.1 kPa, the same as for patients with chronic hepatitis C infection [19]. However, other cut-off values have been published [20]. If we had used an 8.7 kPa cut-off value, the diagnosis of significant fibrosis would have included only eight patients with hemochromatosis (14%) and a strong correlation would have been observed for all noninvasive methods. Indeed, all patients with FibroScan values greater than 8.7 kPa had significant fibrosis, at least according to biochemical scores. Cut-off values could probably be optimized according to the etiology and type of lesion. For example, the cut-off value could be defined mainly for chronic hepatitis, with another one given for alcoholic liver disease and nonalcoholic fatty liver or steatohepatitis, where the amount of fibrosis may be different [28]. Therefore, the cut-off value for the diagnosis of cirrhosis or severe fibrosis in patients with hemochromatosis might be estimated in a study comparing liver biopsy and FibroScan. However, such a study would be difficult to set up as guidelines recommend that a liver biopsy be performed only in cases of serum ferritin levels greater than 1000 ng/mL, abnormal AST level and hepatomegaly [2]. As liver blood tests in hemochromatosis are considered unreliable for detecting fibrosis or cirrhosis, the recommendation is to rely only on liver biopsy to assess the severity of liver fibrosis. The risks of liver biopsy can be reduced by operator experience and by using ultrasound guidance. However, complications such as pain, bleeding, pneumothorax, hemothorax, bile peritonitis, hemobilia, punctured kidney or intestine, infection, anxiety and even death are not entirely avoidable [29,30]. In addition to safety issues, the reproducibility of liver biopsy is poor owing to the hete-

186 rogeneity of liver fibrosis and sample size as well as interand intrapathologist variability [8—10,31]. Few studies have demonstrated a role for noninvasive methods in the evaluation of liver fibrosis in patients with iron overload. Jensen et al. showed that serum procollagenIII-peptide has little diagnostic value for assessing liver fibrosis in hemochromatosis patients [32]. Guyader et al. used an algorithm to predict the absence of severe fibrosis in HFE hemochromatosis [33]. Using multivariate analysis, they demonstrated that the combined finding of serum ferritin less than 1000 ␮g/L, absence of hepatomegaly and normal serum AST predicts a low risk of cirrhosis. They also showed that patients with serum ferritin greater than 1000 ␮g/L, elevated AST and a platelet count less than 200 g/L have an 80% chance of having cirrhosis. Morrison et al. have shown that patients with hemochromatosis and serum ferritin levels less than 1000 ␮g/L are unlikely to have cirrhosis [34]. To our knowledge, the diagnostic value of biochemical markers has never been evaluated in liver fibrosis associated with hemochromatosis. We prospectively assessed the performance of FibroScan in comparison to serum markers of fibrosis in patients with hemochromatosis. As liver biopsy is not recommended for all patients with hemochromatosis, for ethical reasons, we did not perform a liver biopsy as the ‘gold standard’. Although no systematic liver biopsies were performed in our study, we consider FibroScan to be a reliable tool for fibrosis assessment, as seen in previous studies of other clinical settings: HCV infection [19,20], HIV—HCV co-infection [18] and cirrhosis of various origin [22,28,35]. There is no reason to believe that the performance of FibroScan would be different in patients with hemochromatosis. Our results suggest that FibroScan may be used reliably as first-line assessments of fibrosis in patients with hemochromatosis. Only two factors (diabetes and ferritin levels) were associated with FibroScan values greater than 7.1 kPa. The association between diabetes and fibrosis is well-known even in patients with hemochromatosis [36,37]. Likewise, ferritin levels are known to be associated with cirrhosis in such patients [33]. No statistical difference was observed between patients with hemochromatosis and control patients in any of the noninvasive tests (Table 2). The control patients were welldefined, having unexplained chronic cytolysis and no fibrosis on liver biopsy. All control patients had FibroScan values less than 7.1 kPa, indicating that the accuracy of FibroScan for excluding significant fibrosis was excellent in this population. Therefore, as no difference was observed between patients with hemochromatosis and control patients in terms of FibroScan values, we hypothesize that FibroScan could identify significant fibrosis in patients with hemochromatosis. In conclusion, FibroScan and biochemical markers may constitute a reliable noninvasive means of detecting liver fibrosis. Liver biopsy could be performed only for the diagnosis of associated liver diseases [such as alcoholic disease, nonalcoholic fatty liver disease (NAFLD) and primary sclerosing cholangitis] or for patients with discrepancies between FibroScan values and biochemical markers. Further longitudinal and prospective studies are necessary to confirm these preliminary data.

X. Adhoute et al.

Conflicts of interest Authors have no financial disclosure.

Acknowledgements The authors thank the following, who collected the data or performed the measurements: Franc ¸oise Carvalho, Danielle Allaix, Sandrine Villars and Manuela Imounga.

References [1] Olynyk JK, Cullen DJ, Aquilia S, Rossi E, Summerville L, Powell LW. A population-based study of the clinical expression of the hemochromatosis gene. N Engl J Med 1999;341:718—24. [2] Pietrangelo A. Hereditary hemochromatosis — a new look at an old disease. N Engl J Med 2004;350:2383—97. [3] Merryweather-Clarke AT, Pointon JJ, Shearman JD, Robson KJ. Global prevalence of putative haemochromatosis mutations. J Med Genet 1997;34:275—8. [4] Feder JN, Gnirke A, Thomas W, Tsuchihashi Z, Ruddy DA, Basava A, et al. A novel MHC class I-like gene is mutated in patients with hereditary haemochromatosis. Nat Genet 1996;13:399—408. [5] Beaton M, Guyader D, Deugnier Y, Moirand R, Chakrabarti S, Adams P. Noninvasive prediction of cirrhosis in C282Y-linked hemochromatosis. Hepatology 2002;36:673—8. [6] Villeneuve JP, Bilodeau M, Lepage R, Cote J, Lefebvre M. Variability in hepatic iron concentration measurement from needle-biopsy specimens. J Hepatol 1996;25:172—7. [7] Gandon Y, Olivie D, Guyader D, Aube C, Oberti F, Sebille V, et al. Noninvasive assessment of hepatic iron stores by MRI. Lancet 2004;363:357—62. [8] Maharaj B, Maharaj RJ, Leary WP, Cooppan RM, Naran AD, Pirie D, et al. Sampling variability and its influence on the diagnostic yield of percutaneous needle biopsy of the liver. Lancet 1986;1:523—5. [9] Regev A, Berho M, Jeffers LJ, Milikowski C, Molina EG, Pyrsopoulos NT, et al. Sampling error and intraobserver variation in liver biopsy in patients with chronic HCV infection. Am J Gastroenterol 2002;97:2614—8. [10] Bedossa P, Dargere D, Paradis V. Sampling variability of liver fibrosis in chronic hepatitis C. Hepatology 2003;38:1449— 57. [11] Wai CT, Greenson JK, Fontana RJ, Kalbfleisch JD, Marrero JA, Conjeevaram HS, et al. A simple noninvasive index can predict both significant fibrosis and cirrhosis in patients with chronic hepatitis C. Hepatology 2003;38:518—26. [12] Sterling RK, Lissen E, Clumeck N, Sola R, Correa MC, Montaner J, et al. Development of a simple noninvasive index to predict significant fibrosis in patients with HIV/HCV coinfection. Hepatology 2006;43:1317—25. [13] Imbert-Bismut F, Ratziu V, Pieroni L, Charlotte F, Benhamou Y, Poynard T. Biochemical markers of liver fibrosis in patients with hepatitis C virus infection: a prospective study. Lancet 2001;357:1069—75. [14] Adams LA, Bulsara M, Rossi E, DeBoer B, Speers D, George J, et al. Hepascore: an accurate validated predictor of liver fibrosis in chronic hepatitis C infection. Clin Chem 2005;51:1867— 73. [15] Forns X, Ampurdanes S, Llovet JM, Aponte J, Quinto L, Martinez-Bauer E, et al. Identification of chronic hepatitis C patients without hepatic fibrosis by a simple predictive model. Hepatology 2002;36:986—92. [16] Lok AS, Ghany MG, Goodman ZD, Wright EC, Everson GT, Sterling RK, et al. Predicting cirrhosis in patients with hepatitis

Liver fibrosis and hemochromatosis

[17]

[18]

[19]

[20]

[21]

[22]

[23]

[24]

[25]

[26]

C based on standard laboratory tests: results of the HALT-C cohort. Hepatology 2005;42:282—92. Islam S, Antonsson L, Westin J, Lagging M. Cirrhosis in hepatitis C virus-infected patients can be excluded using an index of standard biochemical serum markers. Scand J Gastroenterol 2005;40:867—72. de Ledinghen V, Douvin C, Kettaneh A, Ziol M, Roulot D, Marcellin P, et al. Diagnosis of hepatic fibrosis and cirrhosis by transient elastography in HIV/hepatitis C virus-coinfected patients. J Acquir Immune Defic Syndr 2006;41:175—9. Castera L, Vergniol J, Foucher J, Le Bail B, Chanteloup E, Haaser M, 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—50. Ziol M, Handra-Luca A, Kettaneh A, Christidis C, Mal F, Kazemi F, et al. Noninvasive assessment of liver fibrosis by measurement of stiffness in patients with chronic hepatitis C. Hepatology 2005;41:48—54. Laharie D, Zerbib F, Adhoute X, Boue-Lahorgue X, Foucher J, Castera L, et al. Diagnosis of liver fibrosis by transient elastography (FibroScan) and noninvasive methods in Crohn’s disease patients treated with methotrexate. Aliment Pharmacol Ther 2006;23:1621—8. Foucher J, Chanteloup E, Vergniol J, Castera L, Le Bail B, Adhoute X, et al. Diagnosis of cirrhosis by transient elastography (FibroScan): a prospective study. Gut 2006;55:403—8. Corpechot C, El Naggar A, Poujol-Robert A, Ziol M, Wendum D, Chazouilleres O, et al. Assessment of biliary fibrosis by transient elastography in patients with PBC and PSC. Hepatology 2006;43:1118—24. Carrion JA, Navasa M, Bosch J, Bruguera M, Gilabert R, Forns X. Transient elastography for diagnosis of advanced fibrosis and portal hypertension in patients with hepatitis C recurrence after liver transplantation. Liver Transpl 2006;12:1791—8. Fontaine H, Petitprez K, Roudot-Thoraval F, Trinchet JC. Guidelines for the diagnosis of uncomplicated cirrhosis. Gastroenterol Clin Biol 2007;31:504—9. de Ledinghen V, Ratziu V, Causse X, Bail BL, Capron D, Renou C, et al. Diagnostic and predictive factors of significant liver fibrosis and minimal lesions in patients with persistent unexplained elevated transaminases. A prospective multicenter study. J Hepatol 2006;45:592—9.

187 [27] Sandrin L, Fourquet B, Hasquenoph JM, Yon S, Fournier C, Mal F, et al. Transient elastography: a new noninvasive method for assessment of hepatic fibrosis. Ultrasound Med Biol 2003;29:1705—13. [28] Ganne-Carrie N, Ziol M, de Ledinghen V, Douvin C, Marcellin P, Castera L, et al. Accuracy of liver stiffness measurement for the diagnosis of cirrhosis in patients with chronic liver diseases. Hepatology 2006;44:1511—7. [29] Cadranel J, Rufat P, Degos F. Practices of liver biopsy in France: results of a prospective nationwide survey. For the Group of Epidemiology of the French Association for the Study of the Liver (AFEF). Hepatology 2000;32:477—81. [30] Bravo AA, Sheth SG, Chopra S. Liver biopsy. N Engl J Med 2001;344:495—500. [31] Colloredo G, Guido M, Sonzogni A, Leandro G. Impact of liver biopsy size on histological evaluation of chronic viral hepatitis: the smaller the sample, the milder the disease. J Hepatol 2003;39:239—44. [32] Jensen PD, Heickandorff L, Helweg-Larsen HM, Jensen FT, Christensen T, Ellegaard J. Serum procollagen III peptide concentration in iron overload. Eur J Haematol 1996;57: 157—64. [33] Guyader D, Jacquelinet C, Moirand R, Turlin B, Mendler MH, Chaperon J, et al. Noninvasive prediction of fibrosis in C282Y homozygous hemochromatosis. Gastroenterology 1998;115: 929—36. [34] Morrison ED, Brandhagen DJ, Phatak PD, Barton JC, Krawitt EL, El-Serag HB, et al. Serum ferritin level predicts advanced hepatic fibrosis among U.S. patients with phenotypic hemochromatosis. Ann Intern Med 2003;138:627—33. [35] Kazemi F, Kettaneh A, N’kontchou G, Pinto E, Ganne-Carrié N, Trinchet J, et al. Liver stiffness measurement selects patients with cirrhosis at risk of bearing large oesophageal varices. J Hepatol 2006;45:230—5. [36] Adams LA, Angulo P, Abraham SC, Torgerson H, Brandhagen D. The effect of the metabolic syndrome, hepatic steatosis and steatohepatitis on liver fibrosis in hereditary hemochromatosis. Liver Int 2006;26:298—304. [37] Adams LA, Sanderson S, Lindor KD, Angulo P. The histological course of nonalcoholic fatty liver disease: a longitudinal study of 103 patients with sequential liver biopsies. J Hepatol 2005;42:132—8.