Abstract
BACKGROUND: Cannabis use is increasing globally, with a parallel rise in Cannabis Use Disorder (CUD). Chronic pancreatitis (CP), a progressive inflammatory condition, is associated with acute pancreatitis (AP) flares and an elevated risk of pancreatic cancer (PC). Although cannabis is often used for pain management in CP, its impact on PC risk and AP flare frequency is unclear. METHODS: We conducted a retrospective cohort study using TriNetX to identify adults with CP, stratified by CUD status. Patients with pre-existing PC were excluded. Propensity score matching (1:1) was applied for demographics, behavioral factors, and comorbidities. The primary outcome was PC incidence; the secondary was AP flare frequency. Hazard ratios (HR) were calculated using Cox proportional hazards regression. Sensitivity analysis adjusted for opioid use disorder. RESULTS: Before matching, the CUD cohort (n = 10,864) had higher rates of alcohol and nicotine use than controls (n = 42,160). After matching, 6,858 patients per group remained with balanced covariates (SMD < 0.1). Mean follow-up was shorter in the CUD cohort (736 ± 422 vs. 896 ± 368 days). CUD was associated with a significantly reduced risk of PC (67 vs. 274 cases; HR: 0.263, 95% CI: 0.202-0.344; p < 0.001) but a modest increase in AP flare risk (HR: 1.102, 95% CI: 1.043-1.166; p = 0.001). Results were consistent in the sensitivity analysis. CONCLUSIONS: Among patients with CP, CUD was associated with lower rates of PC detection during available follow-up, but a slightly increased risk of AP flares. These findings warrant further prospective and mechanistic studies to clarify cannabis's role in pancreatic disease.