Cerebral perfusion pressure for prediction of recurrent intracranial hypertension after primary decompressive craniectomy

Background: Decompressive craniectomy (DC) with dural grafting may be performed in patients with moderate (Glasgow-Coma-Scale [GCS] score 9-12 points) or severe traumatic brain injury (TBI; GCS score ≤ 8 points) and threatening herniation. However, its effectiveness especially after primary craniectomy is still discussed due to missing evidence of improved outcome. The objectives of this study were to show the incidence of recurrent intracranial hypertension after primary DC, to identify predictive parameters for secondary DC, and to evaluate the long-term neurological performance 12 months after TBI. Methods: Between 01/1997 and 06/2001 all consecutive patients admitted with moderate or severe isolated TBI were enrolled in this study. They were treated according to the guidelines of the European Brain Injury Consortium, and the American Association of Neurosurgical Surgeons (AANS) for the management of severe TBI. Process and clinical data as well as every intervention were registered prospectively. The long-term neurological status was reassessed using the Glasgow Outcome Score (GOS) 12 months after TBI. Statistical comparison was performed using Mann-Whitney-U test, and multivariate testing by means of logistic regression analysis. Results: Fifty-one (43 males, 8 females; median age 51.4 years) of 119 isolated TBI patients were included. Ten patients (8 males, 2 females; median age 38.4 years) underwent secondary extended or contralateral DC in their clinical course. Three of them (30%) died at a median of 1 day after revision respectively 6 days after TBI. According to univariate analysis, secondary DC significantly correlated with arterial hypotension (p = 0.020) and otorrhagia at admission (p = 0.041), skull base fracture (p = 0.011) and decreased maximum cerebral perfusion pressure (CPP; p = 0.006) after primary surgery. Multivariate analysis identified decreased maximum CPP as the only independent predictive parameter (p = 0.036) for secondary DC and unfavourable GOS after 12-months follow-up. Conclusion: Arterial hypotension, otorrhagia at admission and skull base fractures are negatively influencing the mortality and morbidity of patients with isolated moderate or severe TBI. However, only decreased maximum CPP may independently indicate secondary DC after primary craniectomy in case of recurrent intracranial hypertension.