227 Impact of Active versus Passive Preoxygenation on Emergency Department Mortality in Kigali, Rwanda

Abstract

Preoxygenation for endotracheal intubation (EI) is well studied in high-income countries (HIC). However, its effect on emergency department (ED) airway management in low- and middle-income countries (LMIC) is not well characterized. This study compared the impacts of active versus passive preoxygenation methods on ED mortality among patients presenting for emergency care and undergoing EI at the University Teaching Hospital-Kigali (UTH-K). A prospective cohort of patients requiring ED EI, accrued continuously over twelve months (January 1st, 2017 – December 31st, 2017) with documented preoxygenation methods, were evaluated. The exposure of interest was active preoxygenation (defined as bag-valve mask or positive pressure ventilation) versus passive preoxygenation (defined as non-rebreather mask or oxygen facemask). The primary outcome was ED mortality. Collected data included: duration of preoxygenation, EI indication, clinical characteristics, and pre-intubation vital signs. Magnitudes of effects were quantified using multivariable regression models to yield adjusted odds ratios (aOR) with 95% confidence intervals (CI). Of 194 patients undergoing EI, 163 met inclusion and were analyzed. Median age was 38.7 years (IQR 6-84), 72% were male, with 52% trauma patients. Within the cohort, 73.6% received passive preoxygenation while 26.4% were actively preoxygenated. The shock index (SI) was higher than 0.9 in 45% of those with passive preoxygenation and 58.6% of those who were actively preoxygenated. The majority of both passively (68.3%) and actively preoxygenated (53.9%) patients were preoxygenated for 3-5 minutes. Actively preoxygenated patients had higher ED mortality (81.4%) as compared to passively preoxygenated patients (45.8%) (p<0.001). This translated to significantly lower adjusted odds of ED mortality for those with passive preoxygenation in multivariable models controlling for EI indication, pre-intubation oxygen saturation, pre-intubation SI, and intubation method (aOR 0.30, 95% CI: 0.11, 0.82, p=0.02). In adjusted analyses, passively preoxygenated patients had much lower odds of ED mortality. This association could be due to the impacts of active preoxygenation methods or potentially unmeasured confounding factors. Further research is needed to better understand this clinical approach in LMICs where there exists limited data on preoxygenation methods in emergency care and where there are often barriers to oxygen availability.