Factors influencing time to death after withdrawal of life support in neurocritical patients

Abstract

Objective: Improving our ability to predict the time of death after withdrawal of life-sustaining measures (WLSM) could have a significant impact on rates of organ donation after cardiac death and allocation of appropriate medical resources. We sought to determine which pre-WLSM clinical factors were associated with earlier time to death in patients with catastrophic neurologic disease.

Methods: We retrospectively analyzed all patients who underwent WLSM from 2002 to 2008 in a neurologic intensive care unit. Individuals who died within 60 minutes were compared to those who died beyond this time from the point of WLSM. Patients declared brain dead or not intubated and cases with insufficient data were excluded. Demographic, clinical, laboratory, and radiographic data were reviewed. Statistical analysis was based on multivariate logistic regression.

Results: A total of 149 comatose patients satisfied our inclusion criteria. A total of 75 patients had cardiac arrest in <60 minutes; 57% were male and 52% were older than 66 years. Ischemic stroke (30%) and intraparenchymal hemorrhage (52%) were the most frequent diagnoses. Absent corneal (odds ratio [OR] = 4.24, 95% confidence interval [CI] 1.57-11.5, p = 0.005) and cough reflexes (OR = 4.46, 95% CI 1.93-10.3, p = 0.0005), extensor or absent motor response (OR = 2.83, 95% CI 1.01-7.91, p = 0.048), and an oxygenation index greater than 4.2 (OR = 3.36, 95% CI 1.33-8.5, p = 0.011) were associated with earlier death.

Conclusions: Specific neurologic signs and respiratory measurements are associated with earlier death after withdrawal of life-sustaining measures in the neurologic intensive care unit. This subset of comatose patients with irreversible neurologic injury may be suitable for organ donation after cardiac death protocols. These attributes need validation in a prospective data set.