Autoimmune hepatitis (AIH) remains a management conundrum. The established approach relies upon two phases: (1) early immune suppression via rapid acting corticosteroids; (2) maintenance with azathioprine or other antimetabolite agents. While these latter agents have their own adverse effects, they play a key role in eliminating long term adverse effects secondary to chronic steroid use. This retrospective cohort study by Flatley et al. 1 offers important quantitative data on the metabolic consequences in AIH patients receiving corticosteroid therapy by following 494 patients over a median of 9 years. Conceptually, this work adds to the body of literature regarding the complications of early phase of treatment. The authors report a 15% incidence of new-onset diabetes over the study period, with higher incidence ratios in the first year. Once developed, diabetes rarely resolved even after stopping prednisolone, suggesting permanent metabolic disruption in most cases. The authors attribute this rate of diabetes development to differences in reporting frequencies, older age and extended prednisolone duration in the current study, the latter of which is a notable feature of this cohort: patients were maintained on an average prednisolone dose of 10 mg daily for 2–3 years before considering discontinuation. This may in part be due to the fact that the study cohort was initiated in 1987 and practices regarding steroid utilisation have changed considerably in the last 40 years. In fact, the American Association for the Study of Liver Disease (AASLD) Practice Guidelines 2 recommend tapering steroids by 6 months of treatment initiation. As a result, up to 50% of patients discontinuing prednisolone within twelve months have new-onset diabetes rates of < 5% 3-5. Furthermore, new-onset diabetes and anytime-diabetes are associated with the development of cirrhosis, and new-onset diabetes is associated with liver-related death and transplantation. Prior studies have established worse outcomes in patients with diabetes and liver disease due to hepatitis C, alcohol or MASLD. However, the identification that this also exists in AIH strengthens the importance of prevention of diabetes in this population. Whether diabetes directly accelerates liver disease progression or represents a marker of disease severity and metabolic dysfunction remains to be determined. The study is limited by a retrospective single-center design. However, the comprehensive capture of patients over 36 years, access to primary data for diabetes ascertainment, and detailed outcome tracking provide advantages over some multicenter registries where complete case capture may be challenging. Perhaps the biggest limiting factor is the long study span, which, while allowing for robust patient recruitment, also introduces practice variability that may be directly contributing to higher diabetes incidence numbers. The data highlights the importance of reducing the incidence of new-onset diabetes by limiting corticosteroids regimens. Appropriately focusing on controlling hepatic inflammation, we may have underestimated the long-term metabolic cost. Earlier tapering and discontinuation following sustained complete biochemical remission is warranted. Patients with multiple diabetes risk factors might benefit from closer monitoring and more aggressive steroid-sparing strategies. The findings contribute useful quantitative data to inform evidence-based approaches to AIH therapy, regarding the balance between achieving disease remission and limiting metabolic complications. Pranab M. Barman: conceptualization, writing – original draft, writing – review and editing. The author has nothing to report. This article is linked to Flatley et al. paper. To view this article, visit https://doi.org/10.1111/apt.70188 and https://doi.org/10.1111/apt.70575. The data that support the findings of this study are available from the corresponding author upon reasonable request.
Pranab Barman (Tue,) studied this question.