Craniectomy With Hemorrhage Evacuation May Have Role in TBI Preemptive Craniectomy With Craniotomy: What Role in the Management of Severe Traumatic Brain Injury?

Background: Craniectomy in severe traumatic brain injury (TBI) may be performed to manage intractable elevated intracranial pressure (ICP) or to avoid postoperative development of elevated ICP following craniotomy for a hemorrhagic mass lesion. Objective: To clarify the role of craniectomy performed during removal of hemorrhagic mass lesions in TBI. Design: Single-institution retrospective medical record review. Participants/Methods: Patients with TBI from January 1, 2000, to June 30, 2006, treated with craniotomy alone were compared to those treated with craniectomy to prevent postoperative ICP elevation using data from Scripps Mercy Hospital Trauma Registry. Reasons for craniectomy were ascertained from operative reports. Patients with decompressive craniectomy for refractory ICP were excluded. Propensity score analysis was adjusted for selection biases in evaluating effects of craniectomy on survival. Results: 135 patients (68.5%) had craniotomy and 62 (31.5%) had craniectomy to manage TBI. Craniectomy patients were significantly younger than craniotomy patients (41.5 vs 51.1 years) and had lower admission Glasgow Coma Scale (GCS) scores (mean, 7.6 vs 11.8). Craniotomy injury mechanisms of falls were more prevalent than were craniectomy mechanisms (55.6% vs 32.3%), but pedestrian versus auto mechanisms were less prevalent (5.2% vs 19.4%, respectively). Epidural hematomas were more common in craniotomy patients (19.3% vs 3.2%). Progressive injury on preoperative CT occurred more commonly in craniectomy (29% vs 11.1%). Preoperative ICP monitoring was more common in craniectomy (17.7% vs 5.2%) and more frequently >20 mm Hg or labile (100.0% vs 57.1%). Postoperative ICP monitoring was more common in craniectomy (77.4% vs 32.6%) and more frequently (not significantly) >20 mm Hg or labile (45.8% vs 29.5%). Craniectomy was performed sooner after admission (7.8 vs 27.1 hours). Craniectomy was performed for excessive brain swelling (67.7%) and young patient age (14.5%). Mortality was higher after craniectomy (41.9% vs 23.0%); propensity score analysis controlling for GCS motor score, age, and Abbreviated Injury Score showed equivalent mortality (craniectomy 41%, craniotomy 43%). Craniectomy utilization by individual neurosurgeons ranged from 8.6% to 75.0%. Conclusions: When controlled for other variables, craniectomy is not associated with increased mortality compared to craniotomy alone. Reviewer's Comments: The authors attempt to show that craniectomy performed with surgery for hemorrhage intracranial mass lesions in TBI can preempt postoperative elevated ICP. Unfortunately, the study has too many faults to be convincing. Craniectomy patients are different than craniotomy patients: different mechanisms of injury, lower presenting GCS, fewer epidural hematomas, younger age, and earlier operative intervention, among others. ICP monitoring was inconsistent, and its indications were not specified. Craniectomy was used for cerebral edema in only 67% of cases, and use appeared to be surgeon-specific. Use of craniectomy in cases other than those in which malignant cerebral edema precludes replacement of the bone flap is not more clearly defined by this study. (Reviewer-N. Scott Litofsky, MD).