Long-term effect of hyperoxemia during chronic obstructive pulmonary disease exacerbation managed by emergency medical service and emergency department: a prospective, exploratory study.

Abstract

Long-term effects of hyperoxemia during acute exacerbation of chronic obstructive pulmonary disease (AECOPD) remained unknown. We aimed to explore these effects of hyperoxemia during AECOPD. This was an exploratory follow-up study of a cohort with AECOPD managed by Emergency Medical Service and two emergency departments (EDs). Patients were classified as hyperoxemic (PaO2 > 65 mmHg) or nonhyperoxemic (PaO2 ≤ 65 mmHg). Patients discharged from ED/inpatient care were followed up prospectively for 1 year. The primary outcome was 1-year all-cause mortality in hyperoxemic vs. nonhyperoxemic groups. Secondary outcomes were 3-month all-cause mortality and median number of repeat AECOPD hospitalizations within 1 year. We generated Kaplan-Meier curves and compared them using log-rank test. The primary outcome was also analyzed using Cox proportional-hazards model. We reported crude and adjusted hazard ratios, their 95% confidence intervals (CIs) and P values. We adjusted for two a priori predictors of delayed mortality; age ≥ 70 years and repeat AECOPD hospitalizations. A total of 231 patients were analyzed. One-year mortality rates in hyperoxemic vs. nonhyperoxemic groups were 26/137 (19.0%) and 12/94 (12.8%), respectively (P = 0.693). Although Kaplan-Meier curves showed divergent courses favoring nonhyperoxemic group, log-rank test was not statistically significant (P = 0.203). The crude and adjusted hazard ratios (reference: nonhyperoxemic group) were 1.55 (95% CIs, 0.78-3.08; P = 0.207) and 1.57 (95% CIs, 0.79-3.13; P = 0.196), respectively. Secondary outcomes did not differ. Our study reported no effect on 1-year all-cause mortality associated with hyperoxemia during AECOPD. Further studies are needed to prove/disprove our findings.