Surgical intervention for severe head injury: ethical considerations when performing life-saving but non-restorative surgery

The aim of this study was to compare the predicted outcome with observed outcome in those patients who have had a unilateral decompressive craniectomy following evacuation of an intracranial mass lesion and to consider some of the ethical issues that need to be addressed when performing life-saving but non-restorative surgery. By using the web-based outcome prediction model developed by the CRASH trial collaborators predicted and observed outcomes were compared for those patients who had had a unilateral decompression after evacuation of a mass lesion in the two major neurotrauma hospitals in Western Australia between 2004 and 2008. Three cases were selected with differing outcome predictions. For the three selected cases the predicted risk of an unfavourable outcome at 6 months was 65.8%, 78.9% and 91.3%, respectively. For the 11 patients in this cohort with an outcome prediction between 61% and 70%, the observed outcome at 18 months (GOS) was: 5 had a good outcome, 4 were moderately disabled, and 3 were severely disabled. For the ten patients with an outcome prediction between 90-100%, observed outcome confirmed: one patient was moderately disabled, four patients were severely disabled, one patient was in a permanent vegetative state, and four patients had died. As the index of injury severity (as adjudged by the CRASH outcome prediction model) increases, clinical decision making and discussion with surrogates must reflect the evidence provided by observed outcome, prior to life-saving but potentially non-restorative decompressive surgery.