Background: Chronic obstructive pulmonary disease (COPD) is associated with significant morbidity, mortality, and cost with the greatest proportion incurred treating acute exacerbations of COPD (AECOPD). While guidelines recommend oral steroids for the treatment of AECOPD, parental corticosteroids are still used in the inpatient setting; such use is associated with high costs and adverse effects. The effect of oral versus parenteral use on length of stay (LOS) of patients with AECOPD is not well documented, therefore, we performed a retrospective analysis of patients hospitalized for AECOPD examining the relationship between the mode of steroid administration and LOS.
Methods: We identified hospitalized patients with a primary diagnosis of an AECOPD at an academic medical center between 2011 and 2015 using billing records. We included patients with age 40 years or older, smoking history of at least 20 pack-years, and who received at least 3 days of steroid treatment for AECOPD. Patients with underlying malignancy, asthma, chronic steroid use, ventilator-dependence, tracheostomy, or index admission LOS greater than 14 days were excluded. We used the following criteria for determination of the severity of illness: severity of illness by University HealthSystem Consortium (UHC®) utilizing billing criteria and ICU care during hospitalization. A generalized linear model was used to assess the relationship between LOS and the mode of steroids administration with and without adjustment for severity of illness at admission and ICU utilization.
Results: Of the 655 patients admitted with an AECOPD, 50.5% were female, 39.9% Caucasian, 58.2% African-American, 50% had Medicare, 20% had Medicaid, and 3% were self-pay. Mean (SD) age was 61.7 (9.7) years and LOS was 3.8 (2.5) days. At least one parental dose of corticosteroid was given to 277 (42.3%) patients and oral steroids exclusively were given to 378 (57.7%) patients. As compared to patients who were given oral steroids, patients given parental steroids had higher severity of illness (Chi-Square = 16.4, p=0.0009) and were more likely to require ICU level of care during hospitalization (20.9% vs 1.6%, p<0.0001). Adjusting for severity of illness using generalized linear regression, LOS was longer for patients receiving parental steroids compared to patients who received only oral steroids (Estimated difference = 0.56 days, p<0.004).
Conclusions: Hospitalized patients with AECOPD who received parental steroids had longer LOS than those receiving solely oral steroids independent of the severity of illness. Several factors may have contributed to increased LOS among patients given parenteral corticosteroids including greater risk of side effects and greater time spent in transitioning from parental to oral steroids. While this area needs further investigation, our results suggest that hospitalists should generally not use parental steroids for patients with AECOPD, especially those with lower severity.