Portal inflammation as index of steatohepatitis in children with nonalcoholic fatty liver disease


  • Potential conflict of interest: Nothing to report.

Portal Inflammation as Index of Steatohepatitis in Children with Nonalcoholic Fatty Liver Disease

To the Editor:

We read with great interest the article by Brunt et al.1 In this study, the authors investigated the relationship between portal chronic inflammation and clinical features in the subjects with nonalcoholic fatty liver disease (NAFLD).

The study was conducted on biopsies and clinical parameters of 728 adults and 205 children who were enrolled in the Nonalcoholic Steatohepatitis (NASH) Clinical Research Network. The major findings indicate that 83% of adult biopsies had mild or more than mild portal chronic inflammation, whereas in children this value reached up to 90%. Interestingly, in both groups, portal chronic inflammation (more than mild) correlates with advanced fibrosis.

In our cohort of pediatric patients, the predominant pattern of steatosis distribution is panacinar or azonal, and only in a few cases the steatosis was located in zone 1 or in zone 3. No Mallory bodies were seen, and the prevalent pattern of fibrosis was portal/periportal. Bridging fibrosis was present in only four cases and cirrhosis was absent.2 Moreover, the analysis of histological features of about 200 of our pediatric patients, already described in a previous study,3 demonstrated that a definite steatohepatitis (NASH) was more often associated with moderate and severe steatosis, mild/severe ballooning injury, and mild portal and lobular inflammation.

Brunt and coworkers demonstrated that in pediatric biopsies, “more than mild” compared with “none” was associated with steatosis location (zone 1 accentuation) and the pattern of portal/periportal fibrosis (or more advanced fibrosis). Interestingly, during a recent histopathology re-evaluation of our pediatric patients with NAFLD, we found no association between portal chronic inflammation and histologic severity of the disease. More than one possible explanation could explain this apparent discrepancy. First, our series of children differ on obesity degree, homeostasis model assessment of insulin resistance, and median values of alanine aminotransferase from the NASH Clinical Research Network ones. Second, because just two of our (consecutive) pediatric patients had “none” portal chronic inflammation, we could not trichotomize the outcome in those patients with “none”, “mild”, and “more than mild” as assessed by the NASH Clinical Research Network study. Third, the U.S. pediatric population is characterized by a quite different ethnic mix compared to our population, which included only Caucasian children. The same is true concerning genetic predisposition, lifestyle, and eating habits, which may be relevant in determining the framework of the complex clinical-histologic associations that characterize NAFLD in children.

In conclusion, although we believe that greater emphasis and importance should be given to the specific definition of grade/location and nature of inflammatory infiltrate, other studies are needed to confirm the importance of correlation between portal chronic inflammation and advanced NASH in specific pure pediatric populations.

Anna Alisi Ph.D.*, Rita Devito M.D.†, Valerio Nobili M.D.*, * Liver Unit, Bambino Gesù Children's Hospital and Research Institute, Rome, Italy, † Laboratory of Pathological Anatomy, Bambino Gesù Children's Hospital and Research Institute, Rome, Italy.