Outcomes of Acute Pancreatitis Treatment in the Cannabis Users: A Nationwide Propensity Matched Analysis

Abstract

Introduction: Cannabis is most commonly used illicit drug in the world and is also one of the most commonly abused drugs among alcohol drinkers. Experimental studies have shown conflicting results on the effects of cannabis on the severity and outcomes of acute pancreatitis (AP). The purpose of this study is to ascertain the clinical effects and outcomes in AP in cannabis users by using the largest nationwide propensity-matched cluster. Methods: We assessed the Nationwide Inpatient Sample (NIS) from Jan. 2010-Dec. 2014 and identified adult AP patients and relevant comorbidities with and without cannabis use using ICD-9 CM codes (Table 1). Propensity-score matching was performed using multivariate analysis while adjusting for potential confounders, and categorical & continuous variables were compared in matched cohorts using Chi-square and Student’s T-test, respectively. Primary outcomes were all-cause mortality and other complications, whereas secondary outcomes were the length of stay (LOS) (days) and total hospital charges. Results: The propensity-matched clusters comprised of 5576 patients in both the groups(Table 2). Results were non-significant for all-cause mortality (11%vs. 15% p=0.432) but they were significant for lesser need for total parenteral nutrition (TPN) (1.6% vs. 2.5% p=0.001) in the cannabis-AP group. There were no significant differences in complication rates of SIRS (1.7% vs. 2.0%, p=0.265), acute respiratory distress syndrome (ARDS) (1.4%vs.1.7% p=0.164), disseminated intravascular coagulation (DIC) (0.00%vs.0.1% p=0.095), peritonitis (0.3%vs. 0.3% p=0.857), pancreatic pseudocysts/necrosis (8.3%vs.9.3% p=0.057) between both the groups. The hospital stay was shorter (mean 4.0 vs 4.4 days) and total hospital charges were lower (mean $26,107.40 vs $29,756.74) in cannabis-AP group (p<0.001) Conclusion: This study is our attempt to compare outcomes in AP patients in relation to cannabis use using the largest US inpatient database where we found a lesser need for TPN, shorter LOS and lower hospital charges, which appear to be inter-related. Complication rates tended to be lower but non-significant for the presence of SIRS, lower all-cause in-hospital mortality, development of pancreatic pseudocyst/necrosis, ARDS, and DIC. The results should be interpreted with caution because of inability to adjust multiple other confounding factors and lack of information about type, timing, route, amount, and duration of the cannabis consumption.93_A Figure 1. No Caption available.93_B Figure 2. No Caption available.