Nomogram for Predicting Time to Death After Withdrawal of Life-Sustaining Treatment in Patients With Devastating Neurological Injury.

Reliable prediction of time of death after withdrawal of life-sustaining treatment in patients with devastating neurological injury is crucial to successful donation after cardiac death. Herein, we conducted a study of 419 neurocritical patients who underwent life support withdrawal at four neurosurgical centers in China. Based on a retrospective cohort, we used multivariate Cox regression analysis to identify prognostic factors for patient death, which were then integrated into a nomogram. The model was calibrated and validated using data from an external retrospective cohort and a prospective cohort. We identified 10 variables that were incorporated into a nomogram. The C-indexes for predicting the 60-min death probability in the training, external validation and prospective validation cohorts were 0.96 (0.93-0.98), 0.94 (0.91-0.97), and 0.99 (0.97-1.00), respectively. The calibration plots after WLST showed an optimal agreement between the prediction of time to death by the nomogram and the actual observation for all cohorts. Then we identified 22, 26 and 37 as cut-points for risk stratification into four groups. Kaplan-Meier curves indicated distinct prognoses between patients in the different risk groups (p < 0.001). In conclusion, we have developed and validated a nomogram to accurately identify potential cardiac death donors in neurocritical patients in a Chinese population.