Skip Navigation LinksHome > July 2013 - Volume 57 - Issue 1 > Serum Bilirubin Level Is Inversely Associated With Nonalcoho...
Journal of Pediatric Gastroenterology & Nutrition:
doi: 10.1097/MPG.0b013e318291fefe
Original Articles: Hepatology and Nutrition

Serum Bilirubin Level Is Inversely Associated With Nonalcoholic Steatohepatitis in Children

Puri, Kanika*; Nobili, Valerio§; Melville, Katherine; Corte, Claudia D.§; Sartorelli, Maria R.§; Lopez, Rocio; Feldstein, Ariel E.||; Alkhouri, Naim

Free Access
Article Outline
Collapse Box

Author Information

*Department of Pediatrics, SUNY Downstate, Brooklyn, NY

Department of Pediatric Gastroenterology

Quantitative Health Sciences, Cleveland Clinic, Cleveland, OH

§Liver Unit, “Bambino Gesù” Children's Hospital and Research Institute, Rome, Italy

||Department of Pediatric Gastroenterology, Rady Children's Hospital, University of California, San Diego.

Address correspondence and reprint requests to Naim Alkhouri, MD, Department of Pediatric Gastroenterology, A-31, Digestive Disease Institute, Cleveland Clinic, 9500 Euclid Avenue, Cleveland, OH 44195 (e-mail:

Received 12 December, 2012

Accepted 7 March, 2013

The authors report no conflicts of interest.

Collapse Box


Objectives: Oxidative stress has been implicated in the development of nonalcoholic fatty liver disease (NAFLD) and progression to the more severe form, nonalcoholic steatohepatitis (NASH), in children. We aimed to study the clinical correlation between bilirubin, a potent endogenous antioxidant with cytoprotective properties, and histopathological findings in pediatric patients with NAFLD.

Methods: We included consecutive children with biopsy-proven NAFLD and obtained demographic, clinical, and histopathological data. We performed logistic regression analysis to assess the clinical factors associated with the histological features of NASH or fibrosis.

Results: From a total of 302 biopsies, 67% (203) had evidence of NASH, whereas 64.2% had some degree of fibrosis (stage 1 in 51%, stage 2 in 6.3%, and stage 3 in 6.6%). Mean total bilirubin was significantly lower in the NASH group compared with the non-NASH group (0.65 ± 0.24 vs 0.73 ± 0.22 mg/dL, P = 0.007). Higher total bilirubin levels were negatively correlated with the presence of steatosis and the NAFLD activity score (P < 0.05), whereas a trend in that direction was observed for presence of fibrosis and inflammation (P = 0.051). On multivariable analysis, higher bilirubin levels were significantly associated with a decreased likelihood of a histological diagnosis of NASH on biopsy (odds ratio 0.29, 95% CI 0.10–0.85, P = 0.024).

Conclusions: In children with NAFLD, there is an inverse relation between serum bilirubin levels and the presence of NASH on biopsy. This may be secondary to the antioxidant effect of bilirubin.

Nonalcoholic fatty liver disease (NAFLD) is the leading cause of chronic liver disease in children and adolescents in industrialized countries (1). Its strong association with obesity, insulin resistance, and metabolic syndrome (MetS) is well established (2). The rising incidence of pediatric NAFLD has paralleled the global epidemic of childhood obesity (3). Despite the increased recognition of this condition, the exact pathogenesis for the development of NAFLD or the progression from the relatively benign form of hepatic steatosis to the more severe forms of nonalcoholic steatohepatitis (NASH) and advanced fibrosis still remains elusive. The traditional “two-hit” hypothesis (4) proposed in 1998 has been replaced by a “multiple-hit” hypothesis, in which many diverse processes act in parallel, leading to liver injury (5). The mismatch between increased oxidative stress and decreased antioxidant capacity is considered to be an important mechanism in the pathogenesis of NAFLD (6). Overproduction of reactive oxygen species promotes hepatocyte fat accumulation, inflammation, and fibrosis through induction of lipid peroxidation as well as increased production of inflammatory cytokines and adipokines (7).

Bilirubin is a potent endogenous antioxidant with cytoprotective properties (8–10). The antioxidant effect of bilirubin in inhibiting lipid peroxidation may even exceed that of vitamin E, a strong antioxidant (8,11). Bilirubin is suggested to have a protective effect in diseases involving increased oxidative stress such as atherogenesis, coronary artery disease, and peripheral arterial disease (12–14). Higher levels of bilirubin have been shown to be inversely associated with insulin resistance and MetS in children and adolescents (15). In a cross-sectional study involving obese children, serum bilirubin levels were found to be lower in patients with NAFLD as compared with those without NAFLD (16). In this study, variant UGT1A1*6 genotype, which contributes to the increased serum bilirubin levels, was associated with a lower risk of NAFLD. In another study, unconjugated hyperbilirubinemia was associated with less severe disease on histopathology and/ or transient elastography in adult patients with NAFLD (17). To our knowledge, there are no data looking at the association of serum bilirubin and histopathology in children with NAFLD. Thus, the aim of the study was to assess the relation between severity of liver injury and serum bilirubin in a well-characterized group of children with biopsy-proven NAFLD. We hypothesized that serum bilirubin, a potent antioxidant, has a favorable response on NAFLD disease activity and is associated with less severe histopathological changes.

Back to Top | Article Outline



Consecutive pediatric patients diagnosed as having NAFLD at the “Bambino Gesù” Children's Hospital and Research Institute, Rome, Italy, and enrolled in a prospectively maintained database were included in the study. This large database has been used in part for previous research studies (18). Inclusion criteria were persistently increased serum aminotransferase levels, diffusely echogenic liver on imaging studies suggestive of fatty liver, and a biopsy consistent with the diagnosis of NAFLD (18,19). Exclusion criteria were hepatic virus infections (hepatitis A, B, C, D, E, and G; cytomegalovirus; and Epstein-Barr virus), alcohol consumption, history of parenteral nutrition, and use of drugs known to induce steatosis (eg, valproate, amiodarone, prednisone) or to affect body weight and carbohydrate metabolism. Autoimmune liver disease, metabolic liver disease, Wilson disease, and α1-antitrypsin–associated liver disease were ruled out using standard clinical, laboratory, and histological criteria. The study was approved by the ethics committee of the “Bambino Gesù” Children's Hospital and Research Institute.

Back to Top | Article Outline
Demographic and Clinical Data

We obtained demographic data (including age at first visit and sex) and recorded the clinical variables, including waist circumference (WC), height, and weight. We calculated the body mass index (BMI) and its standard deviation score (z score) (20). We defined obesity by a BMI ≥95th percentile adjusted for age and sex. MetS was defined as the presence of ≥3 of the following 5 criteria (21): abdominal obesity (defined by WC ≥90th percentile for age) (22); hypertriglyceridemia as triglyceride (TG) >95th percentile for age, sex, and race (23); low high-density lipoprotein cholesterol as concentrations <5th percentile for age and sex (23); hypertension as systolic or diastolic blood pressure >95th percentile for age and sex (24); and impaired fasting glucose or known type 2 diabetes mellitus (25). Laboratory data including aspartate aminotransferase and alanine aminotransferase, total bilirubin, and serum γ-glutamyltransferase were obtained. The homeostasis model assessment index of insulin resistance (HOMA-IR) (26), insulin sensitivity index (27), and the quantitative insulin sensitivity check index (28) were calculated as surrogate markers of insulin sensitivity. All of the data were uniformly collected at the time of patient evaluation to avoid recall bias.

Back to Top | Article Outline
Liver Histology

The clinical indications for liver biopsy included the need to assess the presence of NASH and degree of fibrosis or to identify other likely independent or competing liver diseases. Liver biopsy was performed in all of the children, after an overnight fast, using an automatic core biopsy 18-gauge needle (Biopince, Amedic, Sweden) under general anesthesia and ultrasound guidance. A Sonoline Omnia ultrasound machine (Siemens, Munich, Germany) equipped with a 5-MHz probe (5.0 C 50, Siemens) and a biopsy adaptor was used. We performed 2 biopsy passes within different liver segments for each subject and recorded the length of the liver specimen (in millimeters). Only samples with a length ≥15 mm and including at least 5 to 6 complete portal tracts (29) were considered adequate for the purpose of the study. Biopsies were evaluated by a single hepatopathologist who was blinded to clinical and laboratory data. Biopsies were routinely processed (ie, formalin fixed and paraffin embedded) and sections of liver tissue, 5-μm thick, were stained with hematoxylin-eosin, Van Gieson, periodic acid-Schiff diastase, and Prussian Blue stain.

Patients were divided into 2 groups according to the pathologic diagnosis: NASH and non-NASH (simple steatosis). Liver biopsy features were graded according to the NAFLD activity scoring (NAS) system proposed by Kleiner et al (30). Briefly, grade of steatosis was scored as 0 ≤5%, 1 = 5% to 33%, 2 ≥33% to 66%, and 3 ≥66%; grade of lobular inflammation was scored as 0 = no foci, 1 = <2 foci/200× field, 2 = 2 to 4 foci/200× field, and 3 = >4 foci/200× field; and grade of ballooning was scored as 0 = none, 1 = few ballooning cells, and 2 = many cells/prominent ballooning. The grades of steatosis (0–3), lobular inflammation (0–3), and ballooning (0–2) were then combined to determine the NAS (0–8). Fibrosis was scored as 0 = none, 1 = periportal or perisinusoidal fibrosis, 2 = perisinusoidal and portal/periportal fibrosis, 3 = bridging fibrosis, and 4 = cirrhosis.

Back to Top | Article Outline
Statistical Analysis

Descriptive statistics were computed for all factors. Continuous variables are presented as mean ± standard deviation, or median (25th–75th percentiles) and categorical variables as frequencies and percentages. Student t test was used to compare bilirubin levels between subjects with and without NASH and a Wilcoxon rank sum test was used to compare bilirubin groups (0–0.4, 0.5–0.8, and >0.8). The same was done for any fibrosis (fibrosis stage >0) and moderate fibrosis (fibrosis state >1). To assess correlations between bilirubin levels and histological features, Spearman correlation coefficients were used. In addition, multivariable logistic regression analysis was used to further assess whether bilirubin levels are associated with NASH or fibrosis after adjusting for possible confounders. An automated stepwise variable selection method performed on 1000 bootstrap samples was used to choose the final models; the bilirubin level and HOMA-IR were forced into the models and variables with inclusion rates of at least 30% were included in the models. A P < 0.05 was considered statistically significant. All of the analyses were performed using SAS version (9.2 software, SAS Institute, Cary, NC) and R (version 2.13.1, R Foundation for Statistical Computing, Vienna, Austria).

Back to Top | Article Outline


Patient Characteristics

A total of 302 children (36.4% boys) with biopsy-proven NAFLD were included. The baseline anthropometric, clinical, and laboratory data are summarized in Table 1. The mean age at the initial visit was 12.3 ± 3.1 years. More than half of the patients had MetS, and 268 patients (88.7%) were obese. Patients with NASH had significantly higher BMI percentile, WC percentile, TG level, aspartate aminotransferase level, prevalence of MetS, and lower high-density lipoprotein level than patients in the non-NASH group. The main histological features are included in Table 2. Sixty-seven percent of the cases (203 patients) were classified as NASH according to pathologist diagnosis. Three-fourths of the patients had mild severity of lobular inflammation, whereas two-thirds revealed at least mild portal inflammation. Ballooning was present in only 47% of patients. Some degree of fibrosis was seen in the majority of patients (64%): mild fibrosis (stage 1) in 155 patients (51%), moderate fibrosis (stage 2) in 19 patients (6%), and advanced fibrosis (stage 3) in 20 patients (7%). None of the patients included in the study had liver cirrhosis. The mean NAS was 3.8 ± 1.6 (4.5 ± 1.4 for NASH and 2.2 ± 0.655 for non-NASH, P < 0.001).

Table 1
Table 1
Image Tools
Table 2
Table 2
Image Tools
Back to Top | Article Outline
Serum Bilirubin Levels in Relation to Histopathological Features

Patients with NASH had significantly lower bilirubin levels as compared with those without NASH (0.65 ± 0.24 vs 0.73 ± 0.22, P = 0.007) (Fig. 1). A significant negative correlation was noted between levels of serum bilirubin and the histological severity of NAFLD as assessed by the NAS score (r = −0.14, P = 0.017), and between bilirubin levels and presence of steatosis (r = −0.12, P = 0.038) (Table 3). A negative correlation trend was also observed between bilirubin levels and other histological features of NAFLD, including the presence of lobular inflammation and fibrosis, but this did not reach statistical significance (r = −0.11, P = 0.051 for both). More important, on multivariate regression analysis, lower bilirubin levels were independently associated with the presence of NASH on liver biopsy (odds ratio 0.29, 95% CI 0.10–0.85, P = 0.024) after adjusting for multiple confounders including WC percentile, international normalized ratio, TG level, HOMA-IR, and presence of MetS (Table 4). For each 0.1-mg/dL increase in bilirubin level, the likelihood of having NASH decreased by 7.1%.

Figure 1
Figure 1
Image Tools
Table 3
Table 3
Image Tools
Table 4
Table 4
Image Tools
Back to Top | Article Outline


The principal finding of this study is the negative association between serum bilirubin levels and the severity of liver injury on histopathology in children with NAFLD. Serum bilirubin levels negatively correlate with the presence of NASH on liver biopsy. After adjusting for WC percentile, international normalized ratio, TG level, HOMA-IR, and MetS, for every 0.1-mg/dL increase in bilirubin, the likelihood of having NASH decreases by 7.1%. There was also an inverse association between bilirubin levels and NAS. To our knowledge, this is the first study to show an inverse relation between serum bilirubin levels and presence of NASH on liver biopsy in children with NAFLD.

The association between bilirubin levels and NAFLD was first suggested in a cross-sectional study involving 28 obese children with NAFLD, diagnosed on liver ultrasound (16). The mean bilirubin in this group was lower when compared with the obese children without NAFLD. Another prospective cohort study involving Korean adults, 30 to 59 years of age, found that higher serum bilirubin levels were significantly associated with a lower risk of developing NAFLD, diagnosed on ultrasound (31). The histopathological data were not obtained in either of these studies, so that patients with severe disease activity such as NASH with advanced fibrosis could not be differentiated from those with benign hepatic steatosis. Thus, the association between bilirubin levels and NAFLD activity could not be assessed. Our results are consistent with 2 other studies showing a significant relation between bilirubin levels and NAFLD severity in adult patients with NAFLD (17,32). In a study involving 204 adult patients with NAFLD diagnosed by ultrasound, 42 patients had an adequate liver biopsy and 25 were diagnosed as having NASH (17). The proportion of patients with NASH was significantly lower among patients with unconjugated hyperbilirubinemia than those without it (40% vs 70%, P = 0.05). In addition, unconjugated hyperbilirubinemia was also associated with the absence or decreased severity of liver fibrosis. In another large retrospective study involving 508 adult patients with evidence of NAFLD on liver biopsy, it was found that unconjugated hyperbilirubinemia (defined as unconjugated bilirubin >1 mg/dL) was inversely associated with NASH (odds ratio 16.1, 95% CI 3.7–70.8, P < 0.001) on liver biopsy (32).

Bilirubin is formed by enzymatic reduction of biliverdin by biliverdin reductase. It has been shown to protect against lipid peroxidation and to prevent the formation of reactive oxygen species (9). Bilirubin has been shown to protect the cells from a 10,000-fold excess of peroxide, confirming its antioxidant cytoprotective properties (10). This study also showed that the depletion of biliverdin reductase led to tripling of reactive oxygen species levels and a marked augmentation of cell death. In addition to antioxidant effects, serum bilirubin is inversely associated with insulin resistance and MetS in children and adolescents (15). It is believed that low biliverdin reductase activity causes a dysregulation of insulin signaling, leading to insulin resistance. Bilirubin may also be protective for NAFLD by affecting fatty acid metabolism. Bilirubin has been shown to prevent lipolysis in animal models (33). These studies provide biologically plausible explanations for the protective effect of bilirubin in patients with NAFLD.

The main strengths of our study are the inclusion of a large group of children with biopsy-proven NAFLD with the full spectrum of disease, along with an extensive characterization of their clinical, metabolic, and histological profile. This gives us an opportunity to identify a subgroup of patients who may be at a higher risk for developing NASH and may require even closer follow-up. Traditionally, lower serum bilirubin levels were thought to be associated with preserved liver function, and were reassuring for the physicians; however, our study highlights that this may not be true. On the contrary, children with NAFLD who have low serum bilirubin concentrations may be the ones who are at a higher risk for having NASH, and in turn, at higher risk for complications; however, it is important to emphasize that the bilirubin levels encountered in our study were within the normal range, and further evaluation is warranted in children with higher values to rule out overlap with other cholestatic liver diseases.

Our study has some limitations. First, the fact that patients were seen at a large referral tertiary care medical center was evidenced by the fact that two-thirds of our patients with NAFLD had NASH on biopsy. The results may not be generalized to different populations and settings. Second, most of our children were white, making it difficult to determine whether the association between NAFLD severity and bilirubin levels is different among other ethnic groups. Bilirubin levels are determined, in part, by genetic variants of the UGT1A1 gene, the prevalence of which varies among races (16). Third, our study was not designed to identify the causal relation between bilirubin and severity of liver injury. We do not know if increased serum bilirubin prevents NASH, or if NASH somehow decreases bilirubin levels.

In conclusion, our study shows for the first time in children an inverse correlation between serum bilirubin levels and the presence of NASH on liver biopsy. These findings suggest a possible protective role of bilirubin on NAFLD disease activity, likely secondary to its antioxidant properties. Further studies are needed to confirm this association and investigate mechanistic pathways linking bilirubin levels and advanced disease.

Back to Top | Article Outline

The authors thank M. Jeffrey Maisels, MD, for critical review of the manuscript.

Back to Top | Article Outline


1. Pacifico L, Nobili V, Anania C, et al. Pediatric nonalcoholic fatty liver disease, metabolic syndrome and cardiovascular risk. World J Gastroenterol 2011; 17:3082–3091.

2. Schwimmer JB, Deutsch R, Rauch JB, et al. Obesity, insulin resistance, and other clinicopathological correlates of pediatric nonalcoholic fatty liver disease. J Pediatr 2003; 143:500–505.

3. Ogden CL, Carroll MD, Kit BK, et al. Prevalence of obesity and trends in body mass index among US children and adolescents, 1999–2010. JAMA 2012; 307:483–490.

4. Day CP, James OFW. Steatohepatitis: a tale of two. Gastroenterology 1998; 114:842–845.

5. Tilg H, Moschen AR. Evolution of inflammation in nonalcoholic fatty liver disease: the multiple parallel hits hypothesis. Hepatology 2010; 52:1836–1846.

6. Malaguarnera M, Di Rosa M, Nicoletti F, et al. Molecular mechanisms involved in NAFLD progression. J Mol Med 2009; 87:679–695.

7. Alisi A, Cianfarani S, Manco M, et al. Non-alcoholic fatty liver disease and metabolic syndrome in adolescents: pathogenetic role of genetic background and intrauterine environment. Ann Med 2012; 44:29–40.

8. Stocker R, Yamamoto Y, McDonagh AF, et al. Bilirubin is an antioxidant of possible physiological importance. Science 1987; 235:1043–1046.

9. Stocker R. Antioxidant activities of bile pigments. Antioxid Redox Signal 2004; 6:841–849.

10. Baranano DE, Rao M, Ferris CD, et al. Biliverdin reductase: a major physiologic cytoprotectant. Proc Natl Acad Sci U S A 2002; 99:16093–16098.

11. Neuzil J, Stocker R. Free and albumin-bound bilirubin are efficient co-antioxidants for alpha-tocopherol, inhibiting plasma and low density lipoprotein lipid peroxidation. J Biol Chem 1994; 269:16712–16719.

12. Sedlak TW, Snyder SH. Bilirubin benefits: cellular protection by a biliverdin reductase antioxidant cycle. Pediatrics 2004; 113:1776–1782.

13. Mayer M. Association of serum bilirubin concentration with risk of coronary artery disease. Clin Chem 2000; 46:1723–1727.

14. Perlstein TS, Pande RL, Beckman JA, et al. Serum total bilirubin level and prevalent lower-extremity peripheral arterial disease: National Health and Nutrition Examination Survey (NHANES) 1999 to 2004. Arterioscler Thromb Vasc Biol 2008; 28:166–172.

15. Lin LY, Kuo HK, Hwang JJ, et al. Serum bilirubin is inversely associated with insulin resistance and metabolic syndrome among children and adolescents. Atherosclerosis 2009; 203:563–568.

16. Lin YC, Chang PF, Hu FC, et al. Variants in the UGT1A1 gene and the risk of pediatric nonalcoholic fatty liver disease. Pediatrics 2009; 124:e1221–e1227.

17. Kumar R, Rastogi A, Maras J, et al. Unconjugated hyperbilirubinemia in patients with non-alcoholic fatty liver disease: a favorable endogenous response. Clin Biochem 2011; 45:272–274.

18. Nobili V, Marcellini M, Devito R, et al. NAFLD in children: a prospective clinical-pathological study and effect of lifestyle advice. Hepatology 2006; 44:458–465.

19. Manco M, Bedogni G, Marcellini M, et al. Waist circumference correlates with liver fibrosis in children with non-alcoholic steatohepatitis. Gut 2008; 57:1283–1287.

20. Cole TJ, Bellizzi MC, Flegal KM, et al. Establishing a standard definition for child overweight and obesity worldwide: international survey. BMJ 2000; 320:1240–1243.

21. Boney CM, Verma A, Tucker R, et al. Metabolic syndrome in childhood: association with birth weight, maternal obesity, and gestational diabetes mellitus. Pediatrics 2005; 115:e290–e296.

22. Fernández JR, Redden DT, Pietrobelli A, et al. Waist circumference percentiles in nationally representative samples of African-American, European-American, and Mexican-American children and adolescents. J Pediatr 2004; 145:439–444.

23. American Academy of PediatricsNational Cholesterol Education Program: report of the Expert Panel on Blood Cholesterol Levels in Children and Adolescents. Pediatrics 1992; 89 (3 Pt 2):525–584.

24. Report of the Second Task Force on Blood Pressure Control in Children—1987Task Force on Blood Pressure Control in Children. National Heart, Lung, and Blood Institute, Bethesda, Maryland. Pediatrics 1987; 79:1–25.

25. Genuth S, Alberti K, Bennett P, et al. Follow-up report on the diagnosis of diabetes mellitus. Diabetes care 2003; 26:3160.

26. Matthews D, Hosker J, Rudenski A, et al. Homeostasis model assessment: insulin resistance and β-cell function from fasting plasma glucose and insulin concentrations in man. Diabetologia 1985; 28:412–419.

27. Matsuda M, DeFronzo RA. Insulin sensitivity indices obtained from oral glucose tolerance testing: comparison with the euglycemic insulin clamp. Diabetes Care 1999; 22:1462–1470.

28. Katz A, Nambi SS, Mather K, et al. Quantitative insulin sensitivity check index: a simple, accurate method for assessing insulin sensitivity in humans. J Clin Endocrinol Metab 2000; 85:2402–2410.

29. Poynard T, Halfon P, Castera L, et al. Variability of the area under the receiver operating characteristic curves in the diagnostic evaluation of liver fibrosis markers: impact of biopsy length and fragmentation. Aliment Pharmacol Ther 2007; 25:733–739.

30. Kleiner DE, Brunt EM, Van Natta M, et al. Design and validation of a histological scoring system for nonalcoholic fatty liver disease. Hepatology 2005; 41:1313–1321.

31. Chang Y, Ryu S, Zhang Y, et al. A cohort study of serum bilirubin levels and incident non-alcoholic fatty liver disease in middle aged Korean workers. PloS ONE 2012; 7:e37241.

32. Hjelkrem M, Morales A, Williams CD, et al. Unconjugated hyperbilirubinemia is inversely associated with non-alcoholic steatohepatitis (NASH). Aliment Pharmacol Ther 2012; 35:1416–1423.

33. Shepherd R, Moreno F, Cashore W, et al. Effects of bilirubin on fat cell metabolism and lipolysis. Am J Physiol 1979; 237:E504–E508.


antioxidant; bilirubin; children; histology; nonalcoholic fatty liver disease; nonalcoholic steatohepatitis

© 2013 by European Society for Pediatric Gastroenterology, Hepatology, and Nutrition and North American Society for Pediatric Gastroenterology,


Article Tools



Search for Similar Articles
You may search for similar articles that contain these same keywords or you may modify the keyword list to augment your search.

Connect With Us





Visit on your smartphone. Scan this code (QR reader app required) with your phone and be taken directly to the site.