Prospective Study of Nonbeneficial Care in Neurocritical Care Unit

Objective: We sought to determine whether a set of simple criteria can identify patients in the neuroscience intensive care unit (NICU) at high risk of poor outcome and delivery of nonbeneficial care early in the course of their illness. Secondarily, factors affecting limitation of care protocols were assessed. Methods: We prospectively identified patients who were admitted to the NICU with partial loss of brainstem reflexes persisting for >24 hours due to an intrinsic lesion of the brain (trauma, stroke, hemorrhage, etc.). Results: The study included 102 patients. Seventy-two of them died after a mean of 16 days (median: 8 days), and 23 remained comatose, locked-in, or in a vegetative state. Four were conscious and following commands, while 3 were minimally conscious, episodically obeying simple commands. Three out of 4 conscious patients were young males with traumatic brain injuries. Patients who remained full code spent a mean of 22.2 days in the NICU, compared with 10.4 for those who had withdrawal of care (P = 0.022) and 11.9 for patients who received a do-not-resuscitate order (P = 0.045). Time to death did not differ significantly between the groups. Overall, institution of various limitations of care protocols correlated positively with older age (odds ratio [OR] = 1.07, P = 0.0008), being treated on the neurology service (OR = 4.4, P = 0.043), and having health insurance (OR = 5.4, P = 0.03). Conclusions: We identified simple criteria that can be used to identify patients in the NICU setting for whom continued aggressive care is likely nonbeneficial. Our analysis revealed demographic, social, and economic factors correlating with proxies' willingness to consider limitation of care.