Apnea Test for Brain-Death Determination: A Seven-Years Retrospective Study in a French University Hospital

BackgroundThe apnea test is essential for the diagnosis of death by neurologic criteria but clinical data are scarce. Research QuestionHow often and why apnea testing was not done or was aborted, how often the apnea test was completed, and the results of aborted and completed tests (positive/negative/inconclusive). Study Design and MethodsThis retrospective observational cohort study enrolled patients screened for death by neurologic criteria in four intensive care units at a French university hospital in 2015-2022. The main outcomes were the proportion of patients in whom the apnea test was not done due to safety concerns, the proportion of tested patients in whom the test was aborted because of adverse events, completed tests, and the results of tests. Adverse events were described. We measured arterial pH, PaO2, and PaCO2 at the start and end of each test. In patients with multiple tests, only the first was considered for the study. ResultsOf 371 patients screened for death by neurologic criteria, 52 (14%) did not undergo apnea testing, mainly because of respiratory or hemodynamic instability. Among all apnea tests, 94% were positive, 2% negative, and 4% inconclusive. The test was aborted in 34 (11%) patients due to adverse events, of which the most common was hypoxemia; five other severe adverse events were recorded. Test duration varied substantially even when considering only completed tests. The proportion of patients who became donors was not significantly different across the groups with completed tests, aborted tests, or no test. InterpretationOur data confirm that in most cases, the apnea test is conducted, completed and generally safe for diagnosing death by neurologic criteria. However, further research is needed to identify risk factors for adverse events during testing, contraindications to testing, and the optimal protocol for conducting the apnea test according to the characteristics of each patient.