Epidemology and outcomes of critically ill patients in emergency department at Kigali University Teaching Hospital (CHUK)

Abstract

Background and Objectives: The introduction of Emergency Medicine in Rwanda in 2013 has been associated with a mortality reduction in critically ill patients presenting to CHUK. However, while awaiting available Intensive Care Unit (ICU) beds, very sick patients often board in the Emergency department. The aim of this study is to: 1) identify the rate of critically ill patients in ED at CHUK, 2) assess the outcomes of those patients, 3) assess the types of critical management provided to them in ED at CHUK, and 4) determine factors associated to outcomes in critically ill patients. Methods: A descriptive prospective cohort study analyzed all patients 15 years who presented to CHUK between April and June 2022 with modified South African Triage Scores of Red with alarm, Red without alarm, and Orange. Results: Of 320 patients, 66.9% were male and median age was 40 years. Patients were triaged as Orange (65.3%), Red without alarm (22.8%), and Red with alarm (11.9%). Presentations were categorized as: medical emergencies (47.1%), traumatic injury (44.1%), and surgical emergencies (7.5%). Mortality was 12.2% and highest among medical emergencies (16.5%, p=0.046) and increased significantly with triage color (p<0.001): Red with alarm (46.1%), Red without alarm (30.8%) and Orange (23.1%). CPR (10.3%) and endotracheal intubation (8.8%) were the most frequent critical interventions provided. Median length of stay was 33.5 hours and boarding time was 22.2 hours. Length of stay was statistically associated with mortality (p=0.024). In addition, mortality was associated with the following interventions: CPR, intubation, and use of vasopressors (p<0.05). Conclusions: Our study shows high mortality rate despite providing critical management, which can be attributed to prolonged length of stay. Due to limitations, this pilot study is recommended to be extended to other hospital departments for longer periods and to understand factors predicting mortality such that they can be addressed at systems level.