Factors associated with palliative withdrawal of mechanical ventilation and time to death after withdrawal.

Abstract

In imminently dying patients, mechanical ventilation withdrawal is often a comfort measure and avoids prolonging the dying process. The aim of the study was to identify factors associated with palliative withdrawal of mechanical ventilation and time to death after extubation. Logistic regression models were used to identify factors associated with palliative withdrawal of mechanical ventilation. Cox proportional hazards models were used to determine factors associated with time to death after extubation. We retrospectively evaluated 322 patients who died on mechanical ventilation or after palliative ventilator withdrawal at a single tertiary care center. Of the 322 ventilated deaths, 159 patients had palliative withdrawal of mechanical ventilation and 163 patients died on the ventilator. Clinical service was associated with palliative withdrawal of mechanical ventilation: Patients withdrawn from the ventilator were less likely to be on the surgery service and more likely to be on the neurology/neurosurgical service. The median time to death was 0.9 hours (range 0-165 hours). Fraction of inspired oxygen (FIO2) greater than 70% (hazard ratio [HR] 1.92, 95% confidence interval [CI ]1.24-2.99) and a requirement for vasopressors (HR 2.06, 95% CI 1.38-3.09) were associated with shorter time to death. Being on the neurology/neurosurgical service at the time of ventilator withdrawal was associated with a longer time to death (HR 0.60, 95% CI 0.39-0.92). Palliative withdrawal of mechanical ventilation was performed in only half of dying mechanically ventilated patients. Because clinical service rather than physiologic parameters are associated with withdrawal, targeted interventions may improve withdrawal decisions. Considering FIO2 and vasopressor requirements may facilitate counseling families about anticipated time to death.