Introduction

Although an increased risk of acute pancreatitis (AP) in inflammatory bowel disease (IBD) has been previously described, there are only small series, and data on clinical management are lacking. We aimed to investigate the prevalence, etiology, and outcome of AP in IBD patients, including management, in a multicenter international study.

Aims & Methods

A two-step retrospective study was conducted involving 34 centers across Europe. The first step consisted of a survey questionnaire examining the number of AP (according to the Atlanta criteria) in adult IBD patients observed during a 10-year study period (2011-2020), and the second one in collecting data on IBD characteristics, onset of AP, etiology, clinical management, and outcome. The study's primary aim was to estimate the prevalence of AP in IBD patients based on the total number of IBD patients observed in the study period indicated in the survey form. Secondary aims included investigating the etiology of AP in IBD patients, the relationship between IBD medications and AP, the role of different imaging modalities in improving the diagnosis of pancreatic abnormalities, and the clinical outcome of AP. The study protocol has been disseminated through the ECCO (European Crohn’s and Colitis Organization) platform after review and support by Clinical Research Committee.

Results

The expected prevalence of AP in IBD patients, according to the survey results, was 1.13% (780 out of 68989). However, the number of entered patients in the database was 501 (0.73%), but only 398 (0.58%) fulfilled the inclusion criteria. The mean age was 33.6± 14.3, 52% were female, 225 (56.5%) were CD, and 173 (46.4%) were UC. AP was clinically mild in most (86.9%) and, when imaging was performed (n=347, 86.7%), a normal pancreatic gland was described in 81 (23.3%), whereas edematous AP in 218 (62.8%). The most frequent etiology was the drug-induced (220, 55.3%), due to azathioprine in 172 (77.8%), with AP occurring at 2 mg/kg dosage, after a median duration of 1 month. Other causes included gallstones (59, 14.8%) and autoimmune (8.3%). In 54 (13.5%) AP was considered idiopathic. During a median follow-up of 67 months [IQR 34-96] from the index episode, 52 (13%) patients had AP recurrence, and 6 (1.5%) patients progressed to chronic pancreatitis. According to a subgroup analysis by IBD type, CD patients were more often smokers, had a family history of IBD and/or pancreatic disease, extraintestinal manifestations (arthropathy), and a history of surgery, usually ileocecal resection. Ileal location with or without colonic involvement was present in 90.3%, with non-stricturing non-penetrating behavior in most (56%). Medications (66.7%) and gallstones (17.8%) were CD's most common AP etiologies. In contrast, no risk factors for AP were found in UC patients, and extensive disease was described in 102 (59%). The etiology differed significantly from the CD group (p < 0.0001), since in addition to medication (40.5%), autoimmune pancreatitis was found in 13.9% and idiopathic AP in 22%.

Conclusion

PANDORA study established a 0.58% prevalence of AP in IBD patients, lower than expected. The AP is usually mild both clinically and radiologically. In 14% of cases no imaging is performed, in 23% pancreas is normal, suggesting that the real prevalence might be even lower. An ileal location in CD and extensive colitis in UC are usually reported and azathioprine seems the most common cause of AP in this setting, especially a few weeks after its introduction.