Decompressive Craniectomy in Patients with Traumatic Brain Injury: Are the Usual Indications Congruent with Those Evaluated in Clinical Trials?

Abstract

In patients with traumatic brain injury (TBI), multicenter randomized controlled trials have assessed decompressive craniectomy (DC) exclusively as treatment for refractory elevation of intracranial pressure (ICP). DC reliably lowers ICP but does not necessarily improve outcomes. However, some patients undergo DC as treatment for impending or established transtentorial herniation, irrespective of ICP. We performed a population-based cohort study assessing consecutive patients with moderate-severe TBI. Indications for DC were compared with enrollment criteria for the DECRA and RESCUE-ICP trials. Of 644 consecutive patients, 51 (8%) were treated with DC. All patients undergoing DC had compressed basal cisterns, 82% had at least temporary preoperative loss of ≥1 pupillary light reflex (PLR), and 80% had >5mm of midline shift. Most DC procedures (67%) were "primary," having been performed concomitantly with evacuation of a space-occupying lesion. ICP measurements influenced the decision to perform DC in 18% of patients. Only 10 and 16% of patients, respectively, would have been eligible for the DECRA and RESCUE-ICP trials. DC improved basal cistern compression in 76%, and midline shift in 94% of patients. Among patients with ≥1 absent PLR at admission, DC was associated with lower mortality (46 vs. 68%, p=0.03), especially when the admission Marshall CT score was 3-4 (p=0.0005). No patients treated with DC progressed to brain death. Variables predictive of poor outcome following DC included loss of PLR(s), poor motor score, midline shift ≥11mm, and development of perioperative cerebral infarcts. DC is most often performed for clinical and radiographic evidence of herniation, rather than for refractory ICP elevation. Results of previously completed randomized trials do not directly apply to a large proportion of patients undergoing DC in practice.