Pediatric Drug-Associated Pancreatitis Reveals Concomitant Risk Factors and Poor Reliability of Causality Scoring: Report From INSPPIRE.

Abstract

Drug-associated acute pancreatitis (DAP) studies typically focus on single acute pancreatitis (AP) cases. We aimed to analyze the (1) characteristics, (2) co-risk factors, and (3) reliability of the Naranjo scoring system for DAP using INSPPIRE-2 (the INternational Study group of Pediatric Pancreatitis: In search for a cuRE-2) cohort study of acute recurrent pancreatitis (ARP) and chronic pancreatitis (CP) in children. Data were obtained from ARP group with ≥1 episode of DAP and CP group with medication exposure ± DAP. Physicians could report multiple risk factors. Pancreatitis associated with Medication (Med) (ARP+CP) was compared to Non-Medication cases, and ARP-Med vs CP-Med groups. Naranjo score was calculated for each DAP episode. Of 726 children, 392 had ARP and 334 had CP; 51 children (39 ARP and 12 CP) had ≥1 AP associated with a medication; 61% had ≥1 AP without concurrent medication exposure. The Med group had other risk factors present (where tested): 10 of 35 (28.6%) genetic, 1 of 48 (2.1%) autoimmune pancreatitis, 13 of 51 (25.5%) immune-mediated conditions, 11 of 50 (22.0%) obstructive/anatomic, and 28 of 51 (54.9%) systemic risk factors. In Med group, 24 of 51 (47%) had involvement of >1 medication, simultaneously or over different AP episodes. There were 20 ARP and 4 CP cases in "probable" category and 19 ARP and 7 CP in "possible" category by Naranjo scores. Medications were involved in 51 of 726 (7%) of ARP or CP patients in INSPPIRE-2 cohort; other pancreatitis risk factors were present in most, suggesting a potential additive role of different risks. The Naranjo scoring system failed to identify any cases as "definitive," raising questions about its reliability for DAP.