The Association of Chronic Opioid Use with Resource Utilization and Outcomes after Emergency General Surgery

Abstract

Introduction

Chronic opioid use is prevalent among patients undergoing emergent surgery. We sought to understand it on the outcomes of the most common emergency surgery procedures, Appendectomy and Cholecystectomy.

Methods

We used the National Inpatient Sample to identify chronic opioid use in emergency appendectomies (n = 953) and cholecystectomies (n = 2826) from 2005 to 2014. Primary outcome was length of stay (LOS), and secondary outcomes included total charges and mortality. LOS was analyzed with multivariate Poisson regression, total charges with multivariate linear regression.

Results

For Appendectomy, the opioid abuse group was younger, had similar gender and racial demographics, had more Medicaid and private insurance and less self-pay, and had no clinically significant differences in comorbidities. Those with chronic opioid use had a 24% increased LOS (20–29%, p < .001) and $5532(±$881, p < .001) higher hospital charges. Mortality was very rare and not different (0.2% vs 0.6%, aOR 0.54 [0.11–2.58], p = .44). For Cholecystectomy, the opioid abuse group was similar in age and gender, had slightly more white individuals, had a slightly different payor mix including higher rate of private insurance, and had no clinically significant differences in comorbidities. Patients with preoperative chronic opioid abuse showed a 14% increased LOS (12–16%, p < .001) and $5352 (± $1065, p < .001) higher hospital charges, but no significant increase in mortality (0.7% vs 0.6%, aOR 1.58 [0.77–3.25], p = .22).

Conclusion

Patients with chronic opioid abuse did not have increased mortality following EGS but had increased resource utilization and LOS. These findings may help explore the impact of opioid abuse on hospital and societal cost.

Keywords:

• Opioid abuse
• emergency general surgery
• cholecystectomy
• appendectomy
• hospital charges
• length of stay
• in-hospital mortality

Disclosure statement

No potential conflict of interest was reported by the author(s).