Predicting time to death after withdrawal of life-sustaining therapy

Abstract

Purpose: Predicting time to death following the withdrawal of life-sustaining therapy is difficult. Accurate predictions may better prepare families and improve the process of donation after circulatory death. Methods: We systematically reviewed any predictive factors for time to death after withdrawal of life support therapy. Results: Fifteen observational studies met our inclusion criteria. The primary outcome was time to death, which was evaluated to be within 60 min in the majority of studies (13/15). Additional time endpoints evaluated included time to death within 30, 120 min, and 10 h, respectively. While most studies evaluated risk factors associated with time to death, a few derived or validated prediction tools. Consistent predictors of time to death that were identified in five or more studies included the following risk factors: controlled ventilation, oxygenation, vasopressor use, Glasgow Coma Scale/Score, and brain stem reflexes. Seven unique prediction tools were derived, validated, or both across some of the studies. These tools, at best, had only moderate sensitivity to predicting the time to death. Simultaneous withdrawal of all support and physician opinion were only evaluated in more recent studies and demonstrated promising predictor capabilities. Conclusions: While the risk factors controlled ventilation, oxygenation, vasopressors, level of consciousness, and brainstem reflexes have been most consistently found to be associated with time to death, the addition of novel predictors, such as physician opinion and simultaneous withdrawal of all support, warrant further investigation. The currently existing prediction tools are not highly sensitive. A more accurate and generalizable tool is needed to inform end-of-life care and enhance the predictions of donation after circulatory death eligibility.