On the difficulty of neurosurgical end of life decisions.

OBJECTIVE: To analyse the process of end of life decisions in a neurosurgical environment.
METHODS: All 113 neurosurgical patients, who were subject to so called end of life decisions within a one year period were prospectively enrolled in a computerised data bank. Decision pathways according to patient and physician related parameters were assessed.
RESULTS: Leading primary diagnoses of the patients were traumatic brain injury and intracranial haemorrhage. Forty-five patients had undergone an emergency neurosurgical operation prior to end of life decision, N = 69 were conservatively treated, which included intracranial pressure recording, or they were not offered neurosurgical care because of futile prognosis. N = 111 died after a median of two (zero to nine) days. Two, in whom the end of life decisions were revised, survived. Clear decisions to terminate further treatment were made by a senior staff member on call being informed by the senior resident on call (27.4%), difficult decisions on the basis of extensive round discussions (71.7%), and very difficult decision by an interdisciplinary ethical consult (0.9%). Decisions were further substantiated by electrophysiological examinations in N = 59.
CONCLUSION: End of life decisions are to be considered standard situations for neurosurgeons. These decisions may reach a high rate of "positive" prediction, if substantiated by electrophysiological examinations as well as on the grounds of clinical experience and respect for the assumed will of the patient. The fact that patients may survive following revision of an end of life decision underlines the necessity for repeated reassessment of these decisions. Ethical training for neurosurgeons is to be encouraged.