Correlation between long-term outcome and steroid therapy in type 1 autoimmune pancreatitis: relapse, malignancy and side effect of steroid

Abstract

Objectives. Autoimmune pancreatitis (AIP) responds well to corticosteroid therapy (CST), and CST is essential to induce remission. However, the correlation between long-term outcome and CST has not been evaluated. We aimed to clarify the correlation between long-term outcome of AIP and CST. Material and methods. We retrospectively evaluated relapse, risk of malignancy and side effects of CST by focusing on the correlation with CST in 84 patients with type 1 AIP. Results. The incidence of relapse was 23.8%. The frequency of relapse after CST administration was significantly lower in patients taking CST for >6 months than in those who did not (22% versus 67%; p = 0.036). The incidence of malignancy was 10.7%. The standardized incidence ratio of malignancy was 2.14 [95% confidence interval 0.74–3.54]. There were no significant correlations between development of malignancy and CST. The incidences of total and serious side effects due to CST were 75% and 19.1%, respectively. Relapse was the only significant independent predictive risk factor for serious side effects in a multivariate analysis (odds ratio 4.065; 95% confidence interval 1.125–14.706; p = 0.032). The cumulative dose of corticosteroid was significantly higher in patients with serious side effects than in those without (12,645 mg versus 7322 mg; p = 0.041). Conclusions. CST reduces relapse of AIP. However, CST causes serious side effects, particularly in relapsing patients. Alternative maintenance therapy to prevent relapse is needed.