About the tool: This multivariate logistic regression model was developed to predict progression to circulatory death after withdrawal of care among potential DCD heart donors. Inputs should be from the time point closest to withdrawal of care.
In the model development cohort, probability >50% demonstrated sensitivity of 0.85 and specificity of 0.52 with respect to the outcome of circulatory death. Assuming the outcome prevalence observed in the development cohort, this yielded positive predictive value of 76% and negative predictive value of 66%. Probability <20% yielded a negative predictive value of 100%, meaning a donor exceedingly unlikely to progress to circulatory death.
This tool is experimental and should not be used alone to guide patient care. Physicians and other healthcare professionals who use this tool should exercise their own clinical judgment before making any medical decisions.