Serum Levels of CK18 M30 and Leptin Are Useful Predictors of Steatohepatitis and Fibrosis in Paediatric NAFLD

Transcription

1 ORIGINAL ARTICLE: HEPATOLOGY AND NUTRITION Serum Levels of CK18 M30 and Leptin Are Useful Predictors of Steatohepatitis and Fibrosis in Paediatric NAFLD Emer Fitzpatrick, y Ragai R. Mitry, y Alberto Quaglia, y Munther J. Hussain, z Ruth debruyne, and Anil Dhawan ABSTRACT Background: With the alarming growth in prevalence of paediatric nonalcoholic fatty liver disease (NAFLD), there is a need for noninvasive methods of stratifying disease severity. Our aim was to evaluate a combination of serum biomarkers as a measure of disease activity in paediatric NAFLD. Patients and Methods: Forty-five children with biopsy-proven NAFLD were enrolled. Caspase-cleaved CK18 fragments (CK18 M30), hyaluronic acid, leptin, and adiponectin were measured in serum using enzyme-linked immunosorbent assays and high-sensitivity C-reactive protein using a colorimetric assay. Results: Median age was 12.7 years (55% boys). Median body mass index z score was 1.7. CK18 M30 levels were significantly higher in patients with NAFLD versus controls, median 288 IU/L versus 172 IU/L (P < 0.001), and in those with steatohepatitis, median 347 IU/L versus simple steatosis (NAFLD activity score < 3), median 191 IU/L (P ¼ 0.006). Significant fibrosis (F2) could be differentiated from no/minimal fibrosis (<F2), median 393 IU/L versus 243 IU/L (P ¼ 0.03). Leptin could distinguish <F2 from F2; 28.9 ng/ml versus 70.1 ng/ml (P ¼ 0.037). Adiponectin, hyaluronic acid, and high-sensitivity C-reactive protein did not achieve significance in predicting steatohepatitis nor significant fibrosis. Conclusions: The present study combines use of markers for different processes in the development of steatohepatitis. Serum biomarkers, especially CK18 M30, are useful in stratifying disease severity in paediatric NAFLD. Key Words: NAFLD, nonalcoholic steatohepatitis, noninvasive biomarkers (JPGN 2010;51: 00 00) Nonalcoholic fatty liver disease (NAFLD) is reaching epidemic proportions worldwide, and is associated with the rise in obesity during the past 2 decades. Estimated to occur in 30% of adults in the United States, the paediatric population is not far Received November 27, 2009; accepted March 13, From the Paediatric Liver, GI, and Nutrition Centre, the yinstitute of Liver Studies, King s College London School of Medicine at King s College Hospital, London, UK, and the zdepartment of Paediatric Gastroenterology, Hepatology, and Nutrition, University Hospital Ghent, Belgium. Address correspondence and reprint requests to E. Fitzpatrick, Institute of Liver Studies, King s College Hospital, Denmark Hill, London SE5 9PJ, UK ( emer.fitzpatrick@kcl.ac.uk). The authors report no conflicts of interest. Copyright # 2010 by European Society for Pediatric Gastroenterology, Hepatology, and Nutrition and North American Society for Pediatric Gastroenterology, Hepatology, and Nutrition DOI: /MPG.0b013e3181e376be behind, with up to 10% affected (1,2). The spectrum of the disease ranges from simple steatosis to inflammation and fibrosis (nonalcoholic steatohepatitis or NASH). A total of 10% to 22% percent of those with NASH will ultimately develop cirrhosis and hepatocellular carcinoma (3 5). Part of the difficulty in managing NAFLD is in distinguishing those who have NASH or are at a high risk of developing NASH from those with simple steatosis. Monitoring of disease progression and timing of appropriate therapeutic intervention also need consideration. The natural history of the disease is not yet well characterised. The need for liver transplantation for paediatric NAFLD cirrhosis has recently been reported (6). The criterion standard for diagnosis and progression of disease is liver biopsy; however, there is a need for a less invasive method of screening the population, stratifying disease severity, and studying disease progression. The pathogenesis of NASH is known to involve fat accumulation in the liver together with hepatocyte apoptosis, oxidative stress, inflammation, and fibrosis, and individual markers of these processes have been shown to serve as markers of disease severity (7 9). Wieckowska and colleagues (10) first demonstrated that serum CK18 fragments correlate well with the degree of inflammation in NASH; we have previously reported similar findings in a group of paediatric patients (11). For the present study, we hypothesised that the best predictive power would be achieved by targeting serum markers of the individual pathophysiological processes involved. The principal aim of the present study was to examine whether noninvasive methods using a combination of serum biomarkers accurately predict disease severity in children with NAFLD. PATIENTS AND METHODS Population The study population consisted of 45 children with biopsyproven NAFLD (25 boys). Patients were younger than 18 years of age. The decision to biopsy was made by a paediatric hepatologist and the biopsy was usually done in view of abnormal transaminases and/or splenomegaly. The diagnosis of NAFLD was based on the following criteria: liver features of NAFLD as assessed by a liver pathologist, exclusion of liver disease of other aetiologies, and exclusion of alcohol intake on history. Blood was taken on the day or within 3 months of diagnostic liver biopsy; serum was stored at 808C until analysis. Anthropometric, radiological, and biochemical data were recorded in all of the participants. Serum samples from healthy controls were obtained from an archived sample store. Controls were healthy children matched for age and had no clinical, biochemical, or radiological evidence of liver disease; these children did not undergo liver biopsy. Serum samples JPGN Volume 51, Number 4, October

2 Fitzpatrick et al JPGN Volume 51, Number 4, October 2010 TABLE 1. NAFLD Activity Score and scoring system for fibrosis Steatosis, % Ballooning Description Inflammation (lobular) Score 0 <5 None None Rare or few 1 2 foci/20 field Many 2 4 foci/20 field 3 >66 >4 foci/20 field Fibrosis stage 0 No fibrosis 1a Zone 3 perisinusoidal fibrosis (trichrome stain) 1b Zone 3 perisinusoidal fibrosis (on H&E stain) 1c Periportal/portal fibrosis only 2 Zone 3 plus periportal/portal fibrosis 3 As above with bridging 4 Cirrhosis NAFLD Activity Score is the nonweighted sum of steatosis, ballooning, and inflammation scores. Fibrosis is scored separately as above (12). H&E ¼ haematoxylin and eosin; NAFLD ¼ nonalcoholic fatty liver disease. were taken with the consent of the caregiver and the assent of the child. The present study was approved by the local research ethics committee. Histology Histological specimens were scored according to NAFLD activity score (NAS), also scoring for stage of fibrosis (Table 1) (12). This is a nonweighted sum of steatosis (0 3), ballooning (0 2), and inflammation (0 3). NASH is defined as a score 5, not NASH as a score 2, and a score of 3 and 4 is classified as borderline NASH. Fibrosis was scored using a 5-point scale: F0, no fibrosis; F1, mild/moderate perisinusoidal or portal fibrosis; F2, both perisinusoidal and portal fibrosis; F3, bridging fibrosis; and F4, cirrhosis. A single histopathologist scored the specimens blinded to other markers. Type 1 NASH is characterised by perivenular/ perisinusoidal inflammation and fibrosis, whereas periportal fibrosis is classified as type 2 NASH. Biomarker Assays The CK18 M30-Apoptosense enzyme-linked immunosorbent assay kit (PEVIVA, Bromma, Sweden; kits purchased from Bioaxxess, Malvern, UK) was used to quantify apoptosis-associated neoepitope of CK18 in serum samples. Hyaluronic acid levels were determined using an enzyme-linked immunosorbent assay kit (Corgenix, Peterborough, UK), as were adiponectin (AdipoGen, Seoul, Korea) and leptin (BioVendor, Exeter, UK). All of the assays were performed in duplicate, and absorbance was determined using a microplate reader (Dynex MRX, Dynex Technologies Ltd UK). High-sensitivity CRP was measured by an automated assay (Centaur, Siemens, Surrey, UK). Statistical Analysis Medians and interquartile ranges (IQRs) were determined for all of the continuous variables. Frequencies and percentages were calculated, and the x 2 or the Fisher exact test was used for categorical factors. Comparison of continuous variables between groups was undertaken using the Mann-Whitney U test and the Kruskal-Wallis test. The diagnostic performance of an assay was assessed by the analysis of receiver-operating characteristic (ROC) curves. The probability of a true-positive and true-negative assessment was determined for selected cutoff values. P values <0.05 were considered significant. SPSS version 17.0 software (SPSS Inc, Chicago, IL) was used to perform analysis. RESULTS Characteristics of the Cohort Forty-five children with NAFLD and 13 age-matched controls were studied. Median age at diagnosis of cases was 12.7 years (IQR ), median body mass index 25.3 (IQR ), and median body mass index z score 1.7 (IQR ). The median age of controls was 14.4 years (IQR ). Abdominal pain was the presenting feature in almost 50% (22/45); 17% (8/45) presented with viral illness, at which time liver function tests were performed as part of the workup and an elevated alanine transaminase was noticed. The remainder of the cohort was detected in miscellaneous fashion while undergoing workup for obesity, for example. A family history of insulin resistance or hypertension was found in 65% (29/45) of children; 39% of the children themselves had a blood pressure >95th centile for age and sex. All but 5 children underwent the oral glucose tolerance test; this showed hyperinsulinism in 62%, and 1 child met criteria for type 2 diabetes mellitus. Thirteen percent of children demonstrated acanthosis nigricans. Forty-nine percent (22/45) of children had splenomegaly on ultrasound with an additional 22% (10/45) having a spleen length at the upper limit of normal (Tables 2 and 3). No child had a history of gastrointestinal bleeding, and although endoscopy was performed (mainly for abdominal pain or diarrhoea) in only 4, there was no clinical suspicion of varices in any of the patients. Histology All of the cases underwent percutaneous needle liver biopsy. Overall, 8 (17.8%) scored NAS <2 and thus were classified as simple steatosis; an additional 17 (37.8%) scored 3 or 4 and thus were classified as borderline steatohepatitis. Twenty children (44.4%) scored 5 or more and thus were deemed to have NASH. In terms of fibrosis, 22 (48.9%) scored <F2 or no significant 2

3 JPGN Volume 51, Number 4, October 2010 Serum Levels of CK18 M30 and Leptin in Paediatric NAFLD TABLE 2. Clinical and biochemical parameters of the patients (SS vs borderline NASH vs NASH) Variable SS n ¼ 8 Borderline n ¼ 17 NASH n ¼ 20 P Age, y median (q25, q75) 12.7 (10.2, 14.1) 13.1 (11.9,14.3) 12.4 (10.2, 14.4) NS Male % NS BMI median (q25, q75) 25.9 (19.2, 29) 25.6 (24.7, 30) 24 (22.3, 29) NS BMI z score median (q25, q75) 1.5 (0.8, 1.8) 2 (1.3, 2.1) 1.7 (1.2, 2.1) NS Abnormal BP, % NS Enlarged spleen, % NS Bilirubin, mmol/l, median (q25, q75) 10 (7, 12) 11 (7, 13) 8 (7, 10) NS AST, Iu/L, median (q25, q75) 30 (28, 46) y,z 80 (40, 104) 76 (54, 107) ALT, Iu/L, median (q25, q75) 27 (22, 106) 134 (48, 138) 100 (91, 218) NS GGT, Iu/L, median (q25, q75) 26 (12, 44) 44 (18, 93) 50 (31, 74) NS Albumin, g/l, median (q25, q75) 47 (42 51) 48 (47, 49) 49 (48, 50) NS Platelets, 10 9 /L, median (q25, q75) 324 (251, 367) 300 (261, 352) 339 (286, 313) NS Cholesterol, mmol/l, median (q25, q75) 4.4 (4.3, 5.4) 4.7 (4.2, 5.2) 4.9 (4, 5.5) NS Triglycerides, mmol/l, median (q25, q75) 1.3 (0.8, 1.9) 2 (1.1, 2.8) 1.9 (1.3, 3.1) NS INR median (q25, q75) 0.96 (0.92, 1.01) 0.98 (0.95, 1.01) 0.98 (0.93, 1.03) NS HOMA-IR median (q25, q75) 0.75 (0.51, 0.99) 0.83 (0.52, 1.69) 0.67 (0.61, 1.9) NS ALT ¼ alanine aminotransferase; AST ¼ aspartate aminotransferase; BMI ¼ body mass index; BP ¼ blood pressure; GGT ¼ gamma-glutamyltransferase; HOMA-IR ¼ homeostasis model assessment-insulin resistance; INR ¼ international normalised ratio; SS ¼ simple steatosis; q25 ¼ 25th percentile; q75 ¼ 75th percentile. P values correspond to the comparison of the 3 subject groups, Fisher exact test was used for categorical variables, and Kruskal-Wallis test for continuous variables. Significance was defined as P < For 2-way comparisons Mann-Whitney U test (continuous) and Fisher exact test (categorical) were used. y Indicates P < 0.05 for 2-way comparison between SS and borderline. z Indicates P < 0.05 for 2-way comparison between SS and NASH. fibrosis and 23 (51.1%) F2 and so significant or severe fibrosis. Two children had fibrosis stage 4. The majority of children had type 2 NASH (60%) and an additional 20% had mixed features; no child was found to have classical type 1 NASH. Markers of Apoptosis CK18 M30 levels were significantly higher in those with NAFLD than in controls: 288 IU/L (IQR ) versus 172 IU/L TABLE 3. Clinical and biochemical parameters of patients (no or minimal fibrosis versus significant fibrosis) Variable No/minimal fibrosis (<F2) n ¼ 22 Significant fibrosis (F2) n ¼ 23 P Age, y median (q25, q75) 12.8 (11, 14.3) 12.7 (10.2, 14.4) NS Male% NS BMI median (q25, q75) 24.9 (21, 30) 25.4 (23.3, 29) NS BMI z score median (q25, q75) 1.45 (0.88, 2.06) 1.83 (1.48, 2.14) NS Abnormal BP, % NS Enlarged spleen, % NS Bilirubin, mmol/l, median (q25, q75) 8 (7 13) 9 (7, 12) NS AST, Iu/L, median (q25, q75) 51 (33 86) 77 (50, 109) NS ALT, Iu/L, median (q25, q75) 61 (23 178) 109 (92, 136) NS GGT, Iu/L, median (q25, q75) 37 (20, 72) 43 (29, 65) NS Albumin, g/l, median (q25, q75) 48 (46 50) 48 (47,49) NS Platelets, 10 9 L, median (q25, q75) 319 ( ) 322 ( ) NS Cholesterol, mmol median (q25, q75) 4.9 ( ) 4.7 (4.2, 5.2) NS Triglycerides, mmol median (q25, q75) 1.6 (1.2, 2.9) 1.2 (1.9, 2.8) NS INR median (q25, q75) 0.98 (0.92, 1) 0.98 (0.94, 1.03) NS HOMA-IR median (q25, q75) 0.68 (0.55, 0.94) 0.94 (0.56, 1.96) NS ALT ¼ alanine aminotransferase; AST ¼ aspartate aminotransferase; BMI ¼ body mass index; BP ¼ blood pressure; GGT ¼ gamma-glutamyltransferase; HOMA-IR ¼ homeostasis model assessment-insulin resistance; INR ¼ international normalised ratio; q25 ¼ 25th percentile; q75 ¼ 75th percentile. No significant differences between the groups were found. P values correspond to the comparison of the 2 subject groups; Mann-Whitney U test was used for categorical variables, and x 2 test or Fisher exact test was used for categorical variables. 3

4 Fitzpatrick et al JPGN Volume 51, Number 4, October 2010 FIGURE 1. Serum CK18 M30 levels and disease activity. Serum CK18-M30 levels as predictor of control versus NAFLD reach statistical significance with median levels of 172 IU/L in controls versus 288 IU/L in those with NAFLD (A) and as discriminators of simple steatosis (score 2), borderline NASH (score 3,4), and NASH (score 5) with median levels of 191 IU/L versus 275 IU/L versus 347 IU/L, respectively (B). Fibrosis stage was also distinguishable with median levels of 243 IU/L in those with no or minimal fibrosis (<F2) versus 393 IU/L in those with significant fibrosis ( F2) (C). F2 ¼ fibrosis stage 2; NAFLD ¼ nonalcoholic fatty liver disease; NASH ¼ nonalcoholic steatohepatitis; SS ¼ simple steatosis. (IQR ) (P < 0.001). Median value of CK18 M30 fragments in those with simple steatosis was 191 IU/L (IQR ) versus 275 IU/L (IQR ) for those with borderline NASH and 347 IU/L (IQR ) for those with NASH (P ¼ 0.006). In detecting degree of fibrosis, CK18 M30 fragments were significantly higher in those with significant or severe fibrosis (F2), median 393 IU/L (IQR ) versus no or minimal fibrosis (<F2), median 243 IU/L (IQR ) (P ¼ 0.03) (Fig. 1). The area under the ROC curve (AUROC) for CK18 M30 as a predictor of NASH was 0.85 (CI ), a cutoff value of 207 IU/L giving a sensitivity of 84% and a specificity of 88% with a positive predictive value of 90% and a negative predictive value of 80% (Fig. 2). The AUROC for CK18 M30 in predicting significant or severe fibrosis (score >F2) was 0.66 (CI ), with a cutoff value of 200 IU/L giving a sensitivity of 83% and a specificity of 40%. Adipokines Leptin was a good predictor of fibrosis with median levels of 29 ng/ml in those with no or minimal fibrosis versus in those with significant or severe fibrosis median 70 ng/ml (P ¼ 0.037). Adiponectin did not meet statistical significance as a predictor of either NAS score or fibrosis stage (Fig. 3A D). Hyaluronic Acid We tested hyaluronic acid as a potential biomarker of fibrosis but did not find statistically significant results with a median of 19.5 ng/ml versus 25 ng/ml for fibrosis <2 and 2, respectively (P ¼ 0.17) (Fig. 3E F). High-sensitivity C-reactive Protein High-sensitivity C-reactive protein (hscrp) had a trend towards significance for those with a lower NAS score (<4) (0.25 mg/dl) versus a higher score (1 mg/dl), P ¼ 0.085, but was not predictive of grade of fibrosis (Fig. 3G H). We compared these markers with a simple tool for stratifying disease severity: the aspartate aminotransferase-to-platelet ratio (APRI). APRI did not meet statistical significance in distinguishing no/minimal fibrosis from significant fibrosis (P ¼ 0.64) in contrast to both leptin and CK18 M30 levels, both of which met statistical significance. DISCUSSION There is a real need to develop and validate noninvasive biomarkers of fibrosis for paediatric NAFLD. The condition is not uncommon, and prevalence continues to increase with the evergrowing obesity epidemic; it is estimated that 9% of the paediatric population has hepatic steatosis (2), with one third of these developing steatohepatitis and an unknown number progressing to cirrhosis. The presence and degree of fibrosis is an important indicator of prognosis. In a study of 129 patients with NAFLD with a mean follow-up of 13.7 years, liver-related and all-cause mortality was significantly higher in the NASH group than in those with simple steatosis at baseline. Indeed, the absence of periportal fibrosis at diagnosis gave a negative predictive power of 100% for 4

5 JPGN Volume 51, Number 4, October 2010 Serum Levels of CK18 M30 and Leptin in Paediatric NAFLD FIGURE 2. ROC curve for serum CK18 M30 levels. CK18 M30 levels accurately predict NASH (or borderline NASH) versus simple steatosis in patients with NAFLD. Area under the curve is 0.85 (CI ). A cutoff value of 207 IU/L gives a sensitivity of 84% and a specificity of 88% with a positive predictive value of 90% and a negative predictive value of 80%. ROC curve ¼ receiver-operating characteristic curve; NAFLD ¼ nonalcoholic fatty liver disease. liver-related complications. In that study 41% of patients had evidence of progression of fibrosis on repeat biopsy (4). In a study of 132 individuals with NAFLD during a 10-year follow-up period 22% of those with any fibrosis at diagnosis progressed to cirrhosis versus 4% of those with simple steatosis (3). Thus, it is important to identify children with disease who are at risk for progression as well as those with simple steatosis in whom discharge from followup may be considered. Performing liver biopsy as a screening tool would clearly not be tenable; thus, a noninvasive means of determining the presence of the disease and differentiation of simple steatosis from steatohepatitis is required. The accurate identification of stage of fibrosis in addition to grade of inflammation (NAS) is also important for monitoring progression and in decisions regarding initiation of any treatment. Performing serial biopsy to monitor disease progression is not practical or prudent in view of the risks involved and the invasive nature of the procedure. Both steatohepatitis and fibrosis are taken into account in designing an approach to noninvasive staging and grading of disease. There are thought to be 4 processes in the development of steatohepatitis and the progression of fibrosis in NAFLD following the initial first hit of hepatic steatosis. Markers of each of these processes may be targeted to give an accurate reflection of disease activity at any point in time. Markers of hepatocellular apoptosis may be particularly useful; as free fatty acid accumulates in the cell, lysosomes and mitochondria undergo permeabilisation and activation of the caspase cascade (13). Activation of caspase 3 results in cleavage of CK18 (a major intermediary filament in liver) giving rise to the morphological changes of apoptosis. Once the cell undergoes secondary necrosis, CK18 fragments are released and can be detected in serum. A commercial antibody is available that is capable of recognising the neoepitope exposed on the CK18 fragment following caspase cleavage. These CK18 M30 fragments have recently been shown by a number of studies (10,14,15) to correlate well with severity of NASH. Wieckowska et al (10) reported an AUROC of 0.93 to distinguish NASH from simple steatosis. We found a lower AUROC for diagnosis, however, still with acceptable sensitivity and specificity to distinguish simple steatosis/nonspecific inflammation from NASH. We also demonstrated that CK18 M30 levels were predictive of fibrosis. Second, adipokines are correlated with disease activity in NAFLD (16). These are important inflammatory molecules in the pathophysiology of NASH. In particular, high serum levels of tumour necrosis factor-a and leptin and low levels of adiponectin are associated with greater degree of liver damage (17). Other adipocytokines that may have important roles in NASH are visfatin and resistin (18,19). We found that leptin allowed accurate prediction of grade of fibrosis but did not reflect level of inflammation in the biopsies. Adiponectin did not reach statistical significance but trended towards lower levels in those with significant disease. hscrp is probably more accurate in large population studies of NAFLD but was not helpful for predicting inflammation or fibrosis in our study. Third, many different predictive panels exist for predicting degree of fibrosis including extracellular matrix components, matrix metalloproteinases, and tissue inhibitors of matrix metalloproteinases. Conflicting results exist for these markers as predictors of liver damage (20). Biomarker panels including European fibrosis score have proven useful in NAFLD in both adults (21) and children (22). Hyaluronic acid is perhaps the best-studied marker in paediatric liver disease (23,24). We did not find hyaluronic to be a reliable marker of fibrosis in this particular study. We suggest this may be due to the generally preserved liver function in NAFLD where hyaluronic acid is cleared successfully by the liver by sinusoidal endothelial cells. Finally, markers of oxidative stress including lipid peroxidation products may also be useful markers of disease. These substances are volatile and not always easily measured in serum, however. The relative importance of mitochondrial, peroxisomal, CYP450, nitric oxide synthetase, and myeloperoxidase pathways is not yet known (25). The children studied were those who presented to a tertiarylevel pediatric hepatology unit usually referred in view of abnormal transaminases or abnormal appearance of liver on ultrasound. Thus, there is an inherent bias in those who underwent investigation. However, we would argue that these results reflect real-life practice, and despite referral bias, there were still a significant number of those with mild disease in terms of both steatohepatitis and fibrosis. The high incidence of splenomegaly in this population is perhaps not surprising when it is considered that splenomegaly is 1 of the indications for biopsy. However, the presence of splenomegaly in even mild disease is interesting and perhaps additional evaluation of sinusoidal capillarisation in these children may be valuable as reported by Vajro et al (26). Serum biomarker levels were correlated to severity of histopathological damage in this study. Kleiner et al (12) defined NASH based on histological criteria scoring system devised and validated in populations including children. However, pathological appearance in children is known to be different from that in adults; for example, children rarely have Mallory bodies and demonstrate little hepatocyte ballooning. The pattern of fibrosis is often different, children demonstrating a type 2 pattern or an overlap of type 1 and type 2 in 80% of cases (27). We found a particularly high 5

6 Fitzpatrick et al JPGN Volume 51, Number 4, October 2010 FIGURE 3. Adipokines, hyaluronic acid, and hscrp as markers of disease activity. Leptin is a good predictor of fibrosis with levels of 29 ng/ml in those with minimal fibrosis versus 70 ng/ml in those with significant fibrosis (P ¼ 0.037), but was not useful in distinguishing NASH from simple steatosis (A, B). Adiponectin did not meet statistical significance for either NAS score or fibrosis stage (C, D). Neither hyaluronic acid (E, F) nor hscrp (G, H) met significance. F2 ¼ fibrosis stage 2; HA ¼ hyaluronic acid; hscrp ¼ high-sensitivity C-reactive protein; NAS ¼ NAFLD activity score; NASH ¼ nonalcoholic steatohepatitis; SS ¼ simple steatosis. Simple steatosis is defined by a NAS score <2; Borderline NASH is NAS score 3 4; NASH is NAS

7 JPGN Volume 51, Number 4, October 2010 Serum Levels of CK18 M30 and Leptin in Paediatric NAFLD incidence of type 2 NASH in the present study. Although fibrosis per se is not included in the definition score, we consider it an essential parameter of monitoring disease. In the present study, we examined a combination of markers of 3 processes: fibrosis, inflammation, and apoptosis. We demonstrated that these markers can serve as individual predictors of disease: leptin for fibrosis, CK18 M30 for both inflammation and fibrosis, whereas hscrp and adiponectin trend towards significance. Ideally, a multicentre validation study should be undertaken to evaluate the usefulness of these markers in clinical practice. REFERENCES 1. Patton HM, Sirlin C, Behling C, et al. Pediatric nonalcoholic fatty liver disease: a critical appraisal of current data and implications for future research. J Pediatr Gastroenterol Nutr 2006;43: Schwimmer JB, Deutsch R, Kahen T, et al. Prevalence of fatty liver in children and adolescents. Pediatrics 2006;118: Matteoni CA, Younossi ZM, Gramlich T, et al. Nonalcoholic fatty liver disease: a spectrum of clinical and pathological severity. Gastroenterology 1999;116: Ekstedt M, Franzen LE, Mathiesen UL, et al. Long-term follow-up of patients with NAFLD and elevated liver enzymes. Hepatology 2006; 44: Farrell GC, Larter CZ. Nonalcoholic fatty liver disease: from steatosis to citrhosis. Hepatology 2006;43:S Feldstein AE, Charatcharoenwitthaya P, Treeprasertsuk S, et al. The natural history of nonalcoholic fatty liver disease in children: a followup study for up to 20-years. Gut 2009;58: Wieckowska A, McCullough AJ, Feldstein AE. Noninvasive diagnosis and monitoring of nonalcoholic steatohepatitis: present and future. Hepatology 2007;46: Yoneda M, Mawatari H, Fujita K, et al. Noninvasive assessment of liver fibrosis by measurement of stiffness in patients with nonalcoholic fatty liver disease (NAFLD). Dig Liver Dis 2008;40: Mandato C, Lucariello S, Licenziati MR, et al. Metabolic, hormonal, oxidative, and inflammatory factors in pediatric obesity-related liver disease. J Pediatr 2005;147: Wieckowska A, Zein NN, Yerian LM, et al. In vivo assessment of liver cell apoptosis as a novel biomarker of disease severity in nonalcoholic fatty liver disease. Hepatology 2006;44: Mitry RR, De Bruyne R, Quaglia A, et al. Noninvasive diagnosis of nonalcoholic fatty liver disease using serum biomarkers. Hepatology 2007;46: Kleiner DE, Brunt EM, Van Natta M, et al. Design and validation of a histological scoring system for nonalcoholic fatty liver disease. Hepatology 2005;41: Ribeiro PS, Cortez-Pinto H, Sola S, et al. Hepatocyte apoptosis, expression of death receptors, and activation of NF-kappaB in the liver of nonalcoholic and alcoholic steatohepatitis patients. Am J Gastroenterol 2004;99: Feldstein AE, Wieckowska A, Lopez AR, et al. Cytokeratin-18 fragment levels as noninvasive biomarkers for nonalcoholic steatohepatitis: a multicenter validation study. Hepatology 2009;50: Diab DL, Yerian L, Schauer P, et al. Cytokeratin 18 fragment levels as a noninvasive biomarker for nonalcoholic steatohepatitis in bariatric surgery patients. Clin Gastroenterol Hepatol 2008;6: Jarrar MH, Baranova A, Collantes R, et al. Adipokines and cytokines in non-alcoholic fatty liver disease. Aliment Pharmacol Ther 2008; 27: Manco M, Marcellini M, Giannone G, et al. Correlation of serum TNFalpha levels and histologic liver injury scores in pediatric nonalcoholic fatty liver disease. Am J Clin Pathol 2007;127: Diehl AM, Li ZP, Lin HZ, et al. Cytokines and the pathogenesis of nonalcoholic steatohepatitis. Gut 2005;54: Pagano C, Soardo G, Pilon C, et al. Increased serum resistin in nonalcoholic fatty liver disease is related to liver disease severity and not to insulin resistance. J Clin Endocrinol Metab 2006;91: Guha IN, Parkes J, Roderick PR, et al. Non-invasive markers associated with liver fibrosis in non-alcoholic fatty liver disease. Gut 2006; 55: Guha IN, Parkes J, Roderick P, et al. Noninvasive markers of fibrosis in nonalcoholic fatty liver disease: validating the European Liver Fibrosis Panel and exploring simple markers. Hepatology 2008;47: Nobili V, Parkes J, Bottazzo G, et al. Performance of ELF serum markers in predicting fibrosis stage in pediatric non-alcoholic fatty liver disease. Gastroenterology 2009;136: Dhawan A, Trivedi P, Cheeseman P, et al. Serum hyaluronic acid as an early prognostic marker in biliary atresia. J Pediatr Surg 2001;36: Hartley JL, Brown RM, Tybulewicz A, et al. Hyaluronic acid predicts hepatic fibrosis in children with hepatic disease. J Pediatr Gastroenterol Nutr 2006;43: Sanyal AJ. Mechanisms of Disease: pathogenesis of nonalcoholic fatty liver disease. Nat Clin Pract Gastroenterol Hepatol 2005;2: Vajro P, DeChiara C, D Armiento M, et al. Serum hyaluronic acid and laminin in children with chronic hepatitis B. Eur J Gastroenterol Hepatol 1994;6: Schwimmer JB, Behling C, Newbury R, et al. Histopathology of pediatric nonalcoholic fatty liver disease. Hepatology 2005;42: