Cannabis use disorder and severe sepsis outcomes in cancer patients: Insights from a national inpatient sample.

Abstract

Background: The burden of cannabis use disorder (CUD) in the context of its prevalence and subsequent cardiopulmonary outcomes among cancer patients with severe sepsis is unclear.

Aim: To address this knowledge gap, especially due to rising patterns of cannabis use and its emerging pharmacological role in cancer.

Methods: By applying relevant International Classification of Diseases, Ninth and Tenth Revision, Clinical Modification codes to the National Inpatient Sample database between 2016-2020, we identified CUD(+) and CUD(-) arms among adult cancer admissions with severe sepsis. Comparing the two cohorts, we examined baseline demographic characteristics, epidemiological trends, major adverse cardiac and cerebrovascular events, respiratory failure, hospital cost, and length of stay. We used the Pearson χ ² d test for categorical variables and the Mann-Whitney U test for continuous, non-normally distributed variables. Multivariable regression analysis was used to control for potential confounders. A P value ≤ 0.05 was considered for statistical significance.

Results: We identified a total of 743520 cancer patients admitted with severe sepsis, of which 4945 had CUD. Demographically, the CUD(+) cohort was more likely to be younger (median age = 58 vs 69, P < 0.001), male (67.9% vs 57.2%, P < 0.001), black (23.7% vs 14.4%, P < 0.001), Medicaid enrollees (35.2% vs 10.7%, P < 0.001), in whom higher rates of substance use and depression were observed. CUD(+) patients also exhibited a higher prevalence of chronic pulmonary disease but lower rates of cardiovascular comorbidities. There was no significant difference in major adverse cardiac and cerebrovascular events between CUD(+) and CUD(-) cohorts on multivariable regression analysis. However, the CUD(+) cohort had lower all-cause mortality (adjusted odds ratio = 0.83, 95% confidence interval: 0.7-0.97, P < 0.001) and respiratory failure (adjusted odds ratio = 0.8, 95% confidence interval: 0.69-0.92, P = 0.002). Both groups had similar median length of stay, though CUD(+) patients were more likely to have higher hospital cost compared to CUD(-) patients (median = 94574 dollars vs 86615 dollars, P < 0.001).

Conclusion: CUD(+) cancer patients with severe sepsis, who tended to be younger, black, males with higher rates of substance use and depression had paradoxically significantly lower odds of all-cause in-hospital mortality and respiratory failure. Future research should aim to better elucidate the underlying mechanisms for these observations.