Chronic steroid use as an independent risk factor for perioperative complications.

Background: Corticosteroid use continues to rise nationally. Studies have evaluated the impact of chronic steroid use on surgical outcomes in smaller populations. This study investigated the impact of chronic steroid use on perioperative surgical outcomes in a surgical cohort of more than 5 million surgical patients, using a statistically rigorous methodology.
Methods: The National Surgical Quality Improvement Program Database was queried 2008-2016 to evaluate chronic steroid use. Patient demographics, comorbidities, and outcomes were compared, using χ ² and t test analysis, and then repeated after propensity score matching. Finally, a double-adjustment logistic regression was utilized, yielding odds ratios to assess the effect of chronic steroids on perioperative outcomes within the matched population.
Results: Between 2008 and 2016, a total of 5,244,588 patients met inclusion criteria, of whom 181,901 (3.5%) were taking steroids for a minimum of 30 days before surgery. Patients on chronic steroids had significantly more comorbidities compared with the remaining population. After propensity score matching and double-adjusted logistic regression, chronic steroid use was found to be associated with increased surgical complications and poorer surgical outcomes. Chronic steroid use significantly increased a patient's risk of having a hospital stay longer than 30 days by 19%, risk of readmission within 30 days by 58%, risk of reoperation by 21%, and risk of death by 32%.
Conclusion: After controlling for differences in comorbidities and demographics, patients on chronic steroids have significantly poorer perioperative outcomes. Chronic steroid use should be evaluated and, if possible, addressed before surgery, given their significant impact on surgical outcomes.
(Copyright © 2018 Elsevier Inc. All rights reserved.)