Background Formalised emergency departments (ED) are in early development in sub-Saharan Africa and there are limited data on emergency airway management in those settings. This study evaluates characteristics and outcomes of ED endotracheal intubation, as well as risk factors for mortality, at a teaching hospital in Rwanda.

Methods This was a prospective observational study of consecutive patients requiring endotracheal intubation at the University Teaching Hospital of Kigali ED conducted between 1 January and 31 December 2017. A standardised data collection tool was used to record patient demographics, preintubation clinical presentation, indication for intubation, vital signs. medications and equipment used, and periintubation complications. The primary outcome was in-hospital mortality. Univariate associations were determined for risks of mortality.

Results Of 198 intubations were analysed, 72.7% were male and the median age was 35 years (IQR 23–51). Airway protection was the most common indication for intubation (73.7%). Rapid sequence intubation was performed in 74.2% of cases; sedative-only facilitated intubation in 20.6% and non-drug assisted in 5.2%. The most common agents used were Ketamine for sedation (85.4%) and vecuronium for paralysis (65.7%). All patients were successfully intubated within three attempts, 85.4% on the first attempt. During intubation, 23.1% of patients experienced hypoxia, 6.7% aspiration and 3.6% cardiac arrest. Median ED length of stay was 2 days. Outcome data were available for 164 patients of whom 67.7% died. Bonferroni-corrected univariate analysis demonstrated that mortality was associated with higher postintubation shock index (p=0.0007) and lower postintubation systolic blood pressure (SBP) (p=0.0006).

Conclusion The first-attempt and overall success rates for intubation in this ED in Rwanda were comparable to those in high-income countries (HIC). Mortality postintubation is associated with lower postintubation SBP and higher postintubation shock index. The high complication and mortality rates suggest the need for better resources and training to address differences in compared with HIC.