Recent advances in understanding and managing pediatric nonalcoholic fatty liver disease

Introduction

Nonalcoholic fatty liver disease (NAFLD) represents a clinical spectrum of liver abnormalities resulting from excessive fat accumulation within hepatocytes. These abnormalities may range from simple steatosis to nonalcoholic steatohepatitis (NASH) with evident inflammation and cell injury. In some, NASH progresses to advanced fibrosis, cirrhosis, or increased risk for hepatocellular carcinoma. The prevalence of NAFLD has increased over the last several decades in tandem with the rise in obesity globally. Currently, NAFLD represents the most common etiology of chronic liver disease in children and adolescents, affecting nearly 10% of the pediatric population in some parts of the developed world^1,2. Over the last decade, it has become a leading indication for liver transplantation in adults³, mandating the need for early recognition and treatment.

Earlier-onset NASH may result from a distinct underlying pathogenesis, which is demonstrable by differences in certain features relating to histopathology, evident in a subset of prepubertal children⁴. Emerging data support the idea that individuals whose NAFLD is diagnosed in childhood have an increased risk of morbidity and mortality in adulthood⁵. These findings show that affected children may be afflicted with a more aggressive disease phenotype. It follows that, when NAFLD presents earlier, these children are at risk for a more advanced liver disease entering adulthood. Also, they may be at heightened risk for other comorbidities related to metabolic syndrome, including cardiovascular disease.

Some unique pediatric histologic features include a more pronounced fat accumulation, increased portal inflammation and fibrosis, and a less pronounced lobular inflammation and ballooning degeneration. Earlier presentation of disease, along with discrepant features of histology, suggests that children and adults may differ in regard to pathophysiology, and children may display enhanced genetic and environmental vulnerability. Some in utero and perinatal stressors such as premature birth may predispose pediatric patients to NAFLD⁶. Recently, our understanding of NAFLD has improved significantly with advances in diagnosis, etiopathogenesis, genetic predispositions, and intervention.

Epidemiology

The prevalence of pediatric NAFLD is difficult to ascertain since most studies use indirect indicators of hepatic steatosis. Furthermore, the definitive diagnosis of NASH requires a histologic diagnosis via liver biopsy, which is not feasible in population-based studies. With indirect measures, including ultrasound-based imaging or elevated serum aminotransaminase values, the prevalence of NAFLD has been reported as 7% among children and adolescents in the general population and upwards of 34% among obese children¹. When autopsy studies from Poland, Turkey, and the United States are considered, the worldwide prevalence of NAFLD ranges from 4.2 to 9.6% and increases to 38% in those with obesity^2,7–9. Hispanic children, particularly of Mexican descent, have the highest incidence of NAFLD when body mass index (BMI) z-score, gender, and age are controlled for, whereas black children, who have the lowest reported incidence, seem to be relatively protected from the accumulation of hepatic fat (HF)^2,10. There is a male predominance in pediatric studies and the problem is most often detected in the early pubertal period¹¹.

It is important to recognize that current prevalence studies likely do not reflect the true prevalence of disease given the lack of uniform screening methodologies and consistency in diagnostic criteria. Data on the prevalence of NASH, as opposed to NAFLD without NASH, require autopsy-based studies of liver histology in children and adolescents dying from non-medical causes.

Pathophysiology

The clinical spectrum of NAFLD results from specific lipotoxicity (or lipotoxicities) with or without augmented oxidative stress that is found in connection with chronic excess of hepatic lipids. Hepatocellular steatosis in some provides the substrate for invoking hepatic inflammation that predisposes patients to cell injury, apoptosis, stellate cell activation, and production of excess matrix. Why certain individuals develop progressive disease whereas others are less affected remains less clear. Although the majority of children with NAFLD are obese, not all children with obesity develop NAFLD. NAFLD can occur, albeit much less often, in lean individuals. In these cases, further evaluation is especially crucial to assess for alternative etiologies. Several conditions, such as Wilson’s disease, appear similar to NAFLD and may have available therapies^12,13.

Developmental programming: perinatal influence

Many aspects of the mother–child dyad have been examined to identify potentially modifiable risk factors for infant and maternal NAFLD. Maternal obesity, antepartum hyperglycemia, gestational diabetes, excessive weight gain during pregnancy, cesarean section, early-life exposure to antibiotics, and the absence of breastfeeding have all been associated with increased risk for obesity in offspring and the development of NAFLD^31,32.

Altered intestinal microbiomes in the infants of obese mothers result in altered bile acid metabolism, increased short-chain fatty acids, and increased intestinal permeability with bacterial translocation. Dysbiosis also has implications for the bone marrow with recruitment of poorly functioning macrophages. Hepatic inflammation secondary to the translocated bacteria, or their endotoxins which stimulate hepatocyte TLRs triggering a pro-inflammatory cascade, is inadequately controlled because of these hyporesponsive macrophages. A secondary hit, such as that found in Western diets, results in hepatic steatosis and obesity^33,34. See Figure 1 for an illustrative summary of the pathophysiology of NAFLD.

Figure 1. Interplay between genetics, epigenetics, the microbiome, and environmental factors that contribute to NAFLD.

GCKR, glucokinase regulatory protein; GDM, gestational diabetes mellitus; HBW, high birth weight; LBW, low birth weight; PNPLA3, patatin-like phospholipase domain–containing protein 3; TM6SF2, transmembrane 6 superfamily member 2.

Screening/diagnosis

There is a current discrepancy in recommendations between societal guidelines in regard to the utility of screening for pediatric NAFLD. In 2017, both the American Association for the Study of Liver Diseases (AASLD)¹¹ and the North American Society for Pediatric Gastroenterology, Hepatology, and Nutrition (NASPGHAN)¹⁴ published independent guidelines for the screening, diagnosis, and management of pediatric NAFLD.

NASPGHAN recommends screening for obese children beginning at ages 9 to 11 years as well as overweight children with additional risk factors, including insulin resistance, dyslipidemia, or a family history of NAFLD or cryptogenic cirrhosis. Screening consists of serum alanine aminotransferase (ALT) measurement, using the upper limit of normal (ULN) of 22 U/L for girls and 26 U/L for boys. Patients with an ALT value persistently greater than twice the ULN for at least three months should undergo further evaluation and work-up¹⁴. These recommendations are based on expert opinion, whereas AASLD Guidelines do not recommend such screening in adults or children; this is due to a lack of established evidence related to current benefits associated with screening, particularly in the current environment where pharmacologic interventions remain unapproved¹¹. The guidelines purport that since lifestyle modification aimed at weight loss, improved diet, and exercise remains the standard recommendation for all overweight children and adults, a diagnosis of NAFLD will not alter management. This essentially calls into question the utility of screening in the absence of alternative treatment. However, clinicians should consider that confirmation of an existing comorbidity, such as NAFLD, in pediatric obesity may have a role in improving adherence to lifestyle modifications.

In 2015, the European Society for Pediatric Gastroenterology, Hepatology and Nutrition (ESPGHAN) Hepatology Committee published a position paper regarding the role of liver biopsy³⁵. They stated that liver biopsy is required for a definitive diagnosis of NAFLD or NASH but is not proposed in screening. In agreement with AASLD pediatric guidelines, liver biopsy should be performed to exclude other diseases, if advanced disease is suspected, before pharmacologic or surgical treatment, and in clinical research trials³⁵.

Monitoring

Although percutaneous liver biopsy remains the clinical and research standard for the diagnosis of NASH and for assessing histologic improvement during treatment, several imaging modalities have demonstrated promise as surrogate markers of certain features relating to histologic improvement.

Prevention

Maintaining a healthy lifestyle is the cornerstone of NAFLD prevention. Pediatric health-care providers must regularly assess BMI z-scores. This assessment will allow for an allotment of time during routine health-care visits to discuss appropriate diet, exercise, and family involvement in maintaining weight within an appropriate range. This often requires that obstacles to implementing advice be identified with the help of other ancillary providers, including dietitians, social workers, exercise physiologists, psychologists, and educators.

As previously mentioned, growing evidence suggests that perinatal factors and aspects of the maternal–infant dyad play a contributory role in the development of obesity and the metabolic syndrome, including its hepatic component, NAFLD³². These potentially modifiable risk factors may present an opportunity for early disease prevention. Controlling antepartum hyperglycemia, preventing gestational diabetes, and avoiding excessive weight gain during pregnancy may lead to a decrease in childhood obesity and NAFLD. The promotion of prolonged breastfeeding may also provide long-term benefit for both mother and child³². Further studies are needed to evaluate the long-term benefit of such interventions.

Management

End-points for treatment studies vary substantially. Commonly pursued and easily measured goals are a decrease in HF fraction and a decrease in or resolution of inflammation and cell injury (that is, NASH) or fibrosis, which are difficult to measure without liver biopsy. This is more difficult in pediatric populations in particular. A sustained decrease in serum ALT from baseline (generally greater than or equal to 50%) is commonly used as a surrogate marker for earlier stages studies relating to histologic improvement and response to treatment^42,43. Fibrosis measures currently accepted by the US Food and Drug Administration (FDA) require histologic assessment obtained via liver biopsy, which is not always feasible in pediatric studies, especially in those controlled with a placebo arm. In children, NAFLD comorbidities (insulin resistance, dyslipidemia, and hypertension) are also important comorbidities to treat in order to improve future outcomes.

Limitations

One of the primary challenges in pediatric NAFLD has been the variability in clinical trial designs with relatively small sample sizes, heterogeneity of populations, and length of treatments. The absence of validated or accepted non-invasive markers to assess for disease diagnosis or response remains a significant barrier to clinical trial designs and monitoring. Treatment end-points vary across studies. ALT and ultrasound are frequently used as primary outcome surrogates in lieu of liver biopsy given the mandates for trial participation benefits related to clinical research in minors⁵¹. Though widely available, these surrogates have not demonstrated sufficient relation with histology. Continued development of age-validated non-invasive biomarkers and imaging will facilitate future study conduct and monitoring of therapeutic response.

Conclusions

Pediatric NAFLD has emerged as the leading cause of chronic liver disease in children and adolescents worldwide. The pathogenesis of early-onset NAFLD is secondary to a complex interplay involving distinct genetic, metabolic, environmental, and microbiological factors which may begin in utero. Lifestyle modifications involving physical activity and dietary changes remain the only practical therapeutic intervention now available, albeit long-term efficacy is limited by poor adherence. Advancements in non-invasive biomarkers and imaging techniques offer promise in assessing the presence and changes in steatosis and fibrosis. Further validation will facilitate longitudinal, natural history, and therapeutic efficacy studies.