Digestive Diseases and Sciences

, Volume 54, Issue 10, pp 2225–2230

Clinical and Histological Features of Nonalcoholic Fatty Liver Disease in Children

Authors

  • Jae Sung Ko
    • Department of PediatricsSeoul National University College of Medicine
  • Jung Min Yoon
    • Department of PediatricsSeoul National University College of Medicine
  • Hye Ran Yang
    • Department of PediatricsSeoul National University College of Medicine
  • Jae Kyung Myung
    • Department of PathologySeoul National University College of Medicine
  • Hye Ryeung Kim
    • Department of PathologySeoul National University College of Medicine
  • Gyeong Hoon Kang
    • Department of PathologySeoul National University College of Medicine
  • Jung-Eun Cheon
    • Department of RadiologySeoul National University College of Medicine
    • Department of PediatricsSeoul National University College of Medicine
Original Article

DOI: 10.1007/s10620-009-0949-3

Cite this article as:
Ko, J.S., Yoon, J.M., Yang, H.R. et al. Dig Dis Sci (2009) 54: 2225. doi:10.1007/s10620-009-0949-3

Abstract

Background

Nonalcoholic fatty liver disease (NAFLD) is becoming more frequently diagnosed as the prevalence of obesity in children increases rapidly.

Aim

The aim of this study was to investigate the correlation of clinical findings with histopathologic features in children with NAFLD.

Methods

We reviewed the clinical data and liver histology results of children with biopsy-proven NAFLD at Seoul National University Hospital. NAFLD was classified as simple steatosis, type 1 nonalcoholic steatohepatitis (NASH), characterized by ballooning degeneration and perisinusoidal fibrosis, or type 2 NASH, characterized by portal inflammation and portal fibrosis.

Results

Among 80 total patients, 84% were male. All patients were obese or overweight. Insulin resistance was present in 96% of children. Perisinusoidal fibrosis was noted in 45% of children and portal fibrosis was noted in 77%. Simple steatosis was present in 22% of children, type 1 NASH in 34%, and type 2 NASH in 44%. No differences were found among NAFLD subtypes or NAFLD activity score with regard to sex, blood pressure, or levels of aminotransferase, fasting lipid, or insulin. Children with NASH were older and had higher body mass index than those with simple steatosis. Patients with type 2 NASH had higher body mass index and advanced fibrosis compared with patients with type 1 NASH.

Conclusions

Obesity and older age are associated with development of NASH. Type 2 NASH is the most common form and associated with a greater severity of obesity and advanced fibrosis.

Keywords

Fatty liver diseaseObesityHistologyChild

Introduction

For the last several decades, nonalcoholic fatty liver disease (NAFLD) and nonalcoholic steatohepatitis (NASH) have been increasingly recognized with the increasing prevalence of pediatric obesity. NAFLD is a clinicopathologic diagnosis characterized histologically by accumulation of macrovesicular fat in hepatocytes, without consumption of excessive amounts of alcohol. The spectrum of NAFLD extends from simple steatosis to NASH and cirrhosis. The definition of steatosis is accumulation of fat in the liver, and NASH is defined as fatty liver disease causing liver inflammation and varying degrees of hepatic fibrosis. Simple steatosis has a relatively benign course, whereas NASH can progress to cirrhosis and eventually to hepatocellular carcinoma [1]. The true prevalence of NAFLD remains unknown, but recent data show that the prevalence of NAFLD in children and adolescents ranges from 2.6 to 9.6% in the USA, Japan, and Korea [24]. The prevalence of NAFLD in overweight and obese children ranges from 12 to 80% [5]. Because of the diversity of definitions of fatty liver disease and diagnostic modalities, the distribution of prevalence estimates for overweight and obese children varies widely.

The pathologic identification of NASH was first described by Ludwig et al. [6]. Brunt et al. [7] have proposed grades or stages of NASH. The pathology committee of the NASH Clinical Research Network designed and validated a histological feature scoring system that addresses the full spectrum of lesions caused by NAFLD and proposed a NAFLD activity score (NAS) for use in clinical trials [8]. A unique histologic pattern of NASH in children has been described, in which the inflammation and fibrosis are accentuated in the portal area, in contrast with the zone 3 injury pattern typically seen in adult NASH patients. This distinct pattern (termed type 2 NASH) was observed in 51% of children with NAFLD, whereas only 17% showed the typical features of NASH (steatosis, ballooning degeneration, and perisinusoidal fibrosis) seen in the adult population [9]. However, clinical and pathologic evaluations of NAFLD in Asian children are very limited. The aim of this study was to investigate the correlation of clinical characteristics and histopathologic features in children with NAFLD.

Methods

Patients

We reviewed the clinical findings and histopathologic features of 80 children with biopsy-proven NAFLD who visited the Department of Pediatrics at Seoul National University Hospital between January 1995 and June 2008. All subjects were diagnosed with NAFLD following the exclusion of other causes of chronic hepatitis, including hepatitis A, B, and C viruses, autoimmune hepatitis, Wilson’s disease, drug toxicity, total parenteral nutrition, and chronic alcohol intake. Approval for this study was obtained from the Ethical Committee of the Seoul National University Hospital.

Demographics and Laboratory Assessment

Each subject’s chart was reviewed for age, sex, weight, height, body mass index (BMI), and blood pressure. BMI was calculated as the weight (kg) divided by the height (m) squared. Weight, height, and BMI status (percentile) were determined according to age and sex, based on 2007 Korean National growth charts [10]. Obesity was defined as a BMI > 95th percentile, adjusted for age and sex; overweight was defined as a BMI between the 85th and 95th percentiles. We measured the serum concentrations of aspartate aminotransferase (AST), alanine aminotransferase (ALT), fasting glucose, total cholesterol (TC), triglycerides (TG), HDL cholesterol, and insulin after overnight fasting. Results of the blood chemistry obtained at the time of biopsy were recorded. Insulin resistance was determined by the homeostatic model assessment (HOMA-IR = (insulin × glucose)/22.5) [11].

Liver Histopathologic Evaluation

Liver biopsy specimens were evaluated for the following, according to the validated histologic scoring system of Kleiner et al. [8]. Quantitative steatosis was scored as grade 0 (<5% macrovesicular fat), grade 1 (5–33%), grade 2 (34–66%), or grade 3 (>66%). Portal inflammation was graded as 0 (none to minimal) or 1 (greater than minimal). Lobular inflammation was graded as 0 (no foci), 1 (<2 foci per 200× field), 2 (2–4 foci per 200× field), or 3 (>4 foci per 200× field). Ballooning degeneration of hepatocytes was graded as 0 (none), 1 (few balloon cells), or 2 (many cells and prominent ballooning). Fibrosis was graded as stage 0 (none), stage 1a (mild perisinusoidal), stage 1b (moderate perisinusoidal), stage 1c (portal/periportal fibrosis only), stage 2 (perisinusoidal and portal/periportal), stage 3 (bridging fibrosis), or stage 4 (cirrhosis). For analysis, fibrosis stage was grouped into four categories: none (stage 0), perisinusoidal or periportal (stage 1), perisinusoidal and portal/periportal (stage 2), bridging fibrosis (stage 3), or cirrhosis (stage 4). NAFLD activity score (NAS) was calculated as the sum of the scores of steatosis (0–3), lobular inflammation (0–3), and hepatocyte ballooning (0–2).

The pattern of histopathologic features was classified into three different forms of NAFLD, as described by Schwimmer et al.: simple steatosis, type 1 NASH (characterized by steatosis, ballooning degeneration, and perisinusoidal fibrosis), and type 2 NASH (characterized by steatosis, portal inflammation, and portal fibrosis) [9].

Statistical Analysis

Comparisons of the clinical and biochemical features between groups were performed with respect to the type of NAFLD and NAS. The results are expressed as means (means ± standard deviations [SDs]) or number (percentage) of patients with a given condition. Frequency data were compared using the Chi-square test or Fisher’s exact test. Wilcoxon rank sum was performed to assess the significance of differences for continuous variables and the ANOVA test was used for comparison of the three groups. P-values < 0.05 were considered statistically significant. The data were analyzed using the SPSS 12.0 software program (SPSS, Chicago, IL, USA).

Results

Clinical Findings and Liver Histology

Among 80 total patients, there were 67 (84%) males and 13 (16%) females. The mean age was 12.0 ± 2.8 years. Obesity was present in 52 (65%) children, and 28 (35%) were overweight. The mean BMI was 27.6 ± 5.4 kg/m2. Mean levels of AST and ALT were 81.8 ± 54.1 IU/l and 169.2 ± 138.1 IU/l, respectively. Insulin resistance, as indicated by HOMA-IR > 2.0, was present in 96% of children. Type 2 diabetes mellitus was present in one child (1%).

The histological findings are summarized in Table 1. Hepatocyte ballooning was noted in 48 children (60%). Mallory’s hyaline was found in 16 cases (20%). Perisinusoidal fibrosis and lobular inflammation were found in 45 and 86% of patients, respectively. Portal inflammation was present in 43% of subjects, and portal fibrosis was present in 77%. Most subjects showed mild to moderate fibrosis (stage 1 or 2). Bridging fibrosis and cirrhosis were observed in six patients (8%) and one patient (1%), respectively.
Table 1

Histological findings of patients with NAFLD

Grade/stage

Steatosis n (%)

Lobular inflammation n (%)

Hepatocyte ballooning n (%)

Fibrosis n (%)

0

11 (13.8)

32 (40.0)

18 (22.5)

1

14 (17.5)

45 (56.3)

34 (42.5)

38 (47.5)

2

39 (48.8)

24 (30.0)

14 (17.5)

17 (21.3)

3

27 (33.8)

6 (7.5)

4

1 (1.3)

NAFLD Types

Simple steatosis was present in 18 patients (22%), type 1 NASH in 27 patients (34%), and type 2 NASH in 35 patients (44%). In comparison to children with simple steatosis, those with NASH were significantly older and had significantly higher BMI. Additionally, patients with type 2 NASH had higher BMI than those with type 1 NASH. The frequency of type 2 NASH was 25% in patients with BMI < 95th percentile and 57% in those with BMI > 97th percentile (P = 0.03). No differences were found among the three histopathologic types based on age, sex, or diastolic and systolic blood pressure. Laboratory data, including AST, ALT, fasting lipid, glucose, insulin levels, and HOMA-IR, did not differ according to the type of NAFLD (Table 2).
Table 2

Clinical findings of children with NAFLD

 

Simple steatosis (n = 18)

Type 1 NASH (n = 27)

Type 2 NASH (n = 35)

P-value

Age (years)

10.4 ± 3.9

12.6 ± 2.4

12.3 ± 2.3

0.03a

Sex (male:female)

16:2

22:5

29:6

NS

BMI (kg/m2)

24.3 ± 4.3

27.7 ± 3.9

29.2 ± 6.3

0.007b

Systolic BP

123.4 ± 14.4

127.0 ± 11.0

128.9 ± 16.0

NS

Diastolic BP

68.4 ± 8.2

67.3 ± 8.8

68.3 ± 10.8

NS

AST (IU/l)

70.2 ± 61.1

76.7 ± 52.1

91.7 ± 51.6

NS

ALT (IU/l)

160.4 ± 213.1

150.8 ± 104.6

187.9 ± 112.0

NS

GGT (IU/l)

60.9 ± 81.1

57.6 ± 36.0

57.8 ± 27.5

NS

TC (mg/dl)

183.9 ± 37.1

178.2 ± 36.3

182.0 ± 32.3

NS

TG (mg/dl)

132.6 ± 71.6

149.2 ± 61.5

129.7 ± 52.7

NS

HDL-C (mg/dl)

54.6 ± 15.8

44.9 ± 11.3

48.1 ± 11.4

NS

Fasting glucose (mg/dl)

98.6 ± 14.8

101.8 ± 54.3

91.8 ± 8.8

NS

Insulin (μIU/ml)

19.6 ± 12.0

25.8 ± 18.0

24.2 ± 13.0

NS

HOMA-IR

4.7 ± 3.3

6.9 ± 6.5

5.5 ± 3.2

NS

Values are expressed as means ± SD

BMI body mass index; BP blood pressure; AST aspartate aminotransferase; ALT alanine aminotransferase; GGT γ-glutamyltransferase; TC total cholesterol; TG triglyceride; HDL-C high-density lipoprotein cholesterol; NS not significant

aSimple steatosis versus NASH

bSimple steatosis versus type 1 NASH versus type 2 NASH

Results of histologic grading and staging according to the type of NASH are described in Table 3. No differences were found in the severity of steatosis, lobular inflammation, or hepatocyte ballooning between the types of NASH. Lobular inflammation was observed in most of the patients with type 1 or type 2 NASH. Advanced fibrosis, including stage 3 or 4, was more common in type 2 NASH patients than in type 1 NASH patients (P = 0.005). Six of seven patients with advanced fibrosis had type 2 NASH.
Table 3

Histologic grading and staging, according to NASH type

 

Type 1 NASH (n = 27)

Type 2 NASH (n = 35)

P-value

Steatosis grade

    0

0

0

NS

    1

2 (7.4%)

5 (14.3%)

    2

16 (59.3%)

17 (48.0%)

    3

9 (33.3%)

13 (37.1%)

Lobular inflammation

    0

3 (11.1%)

0

NS

    1

17 (63.0%)

22 (62.9%)

    2

7 (25.9%)

13 (37.1%)

    3

0

0

Hepatocyte ballooning

    0

5 (18.5%)

13 (37.1%)

NS

    1

11 (40.7%)

19 (54.3%)

    2

11 (40.7%)

3 (8.6%)

Fibrosis stage

    0

0

1 (2.9%)

0.005

    1

20 (74.1%)

17 (48.6%)

    2

6 (22.2%)

11 (31.4%)

    3

1 (3.7%)

5 (14.3%)

    4

0

1 (2.9%)

NS not significant

NAFLD Activity Score

The mean NAS value was 4.0 ± 1.5. No differences were found among the three NAS groups with regard to age, sex, and BMI. Laboratory test values, including AST, ALT, fasting lipid, fasting glucose, insulin levels, and HOMA-IR, did not differ according to NAS (Table 4).
Table 4

Clinical characteristics of the patients according to their NAFLD activity scores

 

NAS 1–3 (n = 28)

NAS 4–5 (n = 38)

NAS 6–7 (n = 14)

P-value

Age (years)

11.7 ± 3.7

12.4 ± 2.4

11.3 ± 1.6

NS

Sex (male:female)

26:2

30:8

11:3

NS

BMI (kg/m2)

27.0 ± 5.7

27.9 ± 5.8

27.8 ± 4.1

NS

AST (IU/l)

86.1 ± 70.3

80.6 ± 44.1

76.6 ± 43.7

NS

ALT (IU/l)

182.1 ± 202.1

157.6 ± 82.5

174.9 ± 104.4

NS

GGT (IU/l)

65.0 ± 70.1

56.0 ± 27.9

51.8 ± 24.7

NS

TC (mg/dl)

195.9 ± 35.0

171.7 ± 32.3

177.5 ± 30.5

NS

TG (mg/dl)

137.0 ± 53.8

136.4 ± 60.2

136.7 ± 68.3

NS

Fasting glucose (mg/dl)

96.2 ± 13.5

99.4 ± 46.0

90.5 ± 7.8

NS

Insulin (μIU/ml)

23.5 ± 14.6

25.5 ± 16.3

22.0 ± 11.7

NS

HOMA-IR

5.5 ± 3.8

6.6 ± 5.8

4.8 ± 2.5

NS

The values are expressed as means ± SD

NAS NAFLD activity score; BMI body mass index; AST aspartate aminotransferase; ALT alanine aminotransferase; GGT γ-glutamyltransferase; TC total cholesterol; TG triglyceride; NS not significant

Discussion

The incidence of pediatric NAFLD is rising as childhood obesity becomes increasingly prevalent. Histology-based autopsy data suggest that the prevalence of fatty liver is 9.6% in children ages 2–19 [2]. Recent studies have shown that NAFLD is the most common histological diagnosis in patients with unexplained, persistently abnormal liver function tests [12, 13]. An epidemiological study has suggested that NAFLD is the most important cause of cryptogenic cirrhosis [14].

The conventional radiologic studies used in the diagnosis of fatty liver include ultrasound, computed tomography, and magnetic resonance imaging. These conventional methods are useful in the identification of steatosis but cannot differentiate NASH. Thus, percutaneous liver biopsy remains the standard for distinguishing simple steatosis from NASH and for assessing disease severity, using histological grading and staging [1]. The differentiation of NASH from simple steatosis is clinically important because simple steatosis is associated with a favorable outcome. NASH is a progressive fibrotic disease in which cirrhosis and liver-related death occur in up to 20 and 12% of patients, respectively, over a 10-year period, whereas only 3% of patients with simple steatosis progress to steatohepatitis or cirrhosis [15].

Schwimmer et al. reported that the prevalence of simple steatosis, type 1 NASH, and type 2 NASH was 16, 17, and 51%, respectively. A study from Italy showed that type 1 NASH was observed in only 2.4% of subjects and that type 2 NASH was present in 28.6% [16]. In our study, type 2 NASH was the most common form, present in 44% of children with NAFLD, whereas 34% had type 1 NASH. This observation confirms that type 2 NASH is the major form in children of Asian descent [9]. However, unlike previous studies [9, 16], type 1 NASH was not a minor occurrence in our pediatric patients. In addition, histological features of type 1 NASH, including hepatocyte ballooning (60%), lobular inflammation (86%), and perisinusoidal fibrosis (45%), were found frequently. This higher prevalence of adult-type NASH in our study may be related to ethnic differences. A lesser degree of interobserver agreement in scoring the histologic features of pediatric NAFLD should also be considered [8]. Liver biopsy was usually performed in obese children with persistently abnormal serum aminotransferase levels. For these reasons, the prevalence of simple steatosis appears lower than that of NASH. In an autopsy-based series, simple steatosis was observed in 77% of subjects with fatty liver [2].

The patients with NAFLD in this study had moderate increases in serum aminotransferase levels. The type of NAFLD and NAS value were not associated with biochemical parameters, including HOMA-IR and levels of AST, ALT, fasting lipid, glucose, and insulin. Our findings are consistent with previous reports demonstrating no difference in laboratory data between subjects with type 1 and type 2 NASH [9]. A recent study has shown that AST level is a significant predictor of NAS and fibrosis severity but cannot replace liver biopsy in evaluating pediatric NAFLD [17]. The NAS included only features that were potentially reversible in the short term. Liver fibrosis is less reversible and is thought to be a long-term result of disease activity [8]. Older age and presence of obesity were independent indicators of the presence of liver fibrosis in children with NAFLD [16]. We showed that children with NASH were older and had higher BMI compared with those with simple steatosis. Children with type 2 NASH had higher BMI than those with type 1 NASH. This is consistent with the prior observation that children with type 2 NASH had a greater severity of obesity [9]. Our study shows that lobular inflammation is observed in most NASH patients and does not differentiate between subtypes of NASH. As shown in Schwimmer et al.’s study [9], children with advanced fibrosis are more likely to have type 2 NASH than type 1 NASH. Liver fibrosis suggests more severe and progressive liver damage. The clinical outcome of type 2 NASH, compared with type 1 NASH, remains to be elucidated.

Insulin resistance has been implicated in the initiation of NAFLD. In this study, 96% of patients with NAFLD demonstrated insulin resistance, which was defined as HOMA-IR > 2 [18, 19]. In a consecutive series of 43 children who had biopsy-proven NAFLD, 95% met the criteria for insulin resistance when results from HOMA-IR analysis and a related arithmetic estimate of insulin resistance (QUICK-I) were combined [20].

In summary, most children with NAFLD demonstrated insulin resistance. Obesity and older age were associated with development of NASH. Type 2 NASH was the most common form and type 1 NASH was not a minor occurrence. Children with type 2 NASH had a greater severity of obesity and advanced fibrosis. The type of NAFLD or NAS value was not associated with laboratory values, including AST, ALT, fasting lipid, and insulin resistance. Further studies are necessary to investigate the natural course of type 1 and type 2 NASH.

Acknowledgments

This study was supported by grant no. 0420080830 from SNUH Research Fund.

Copyright information

© Springer Science+Business Media, LLC 2009