Article

Intracranial pressure monitoring in brain-injured patients is associated with worsening of survival.

Division of Burns, Department of Surgery, Trauma Critical Care, University of Texas Southwestern Medical School, Dallas, Texas, USA.
The Journal of trauma (Impact Factor: 2.35). 03/2008; 64(2):335-40. DOI: 10.1097/TA.0b013e31815dd017
Source: PubMed

ABSTRACT The Brain Trauma Foundation (BTF) recommends intracranial pressure (ICP) monitoring in traumatic brain injury (TBI) patients with Glasgow Coma Scale (GCS) of 8 or less, and an abnormal brain computed tomography. However, benefits of ICP monitoring have not been documented. We hypothesized that BTF criteria for ICP monitoring in blunt TBI do not identify patients who are likely to benefit from it.
The National Trauma Data Bank (1994-2001) was analyzed. Inclusion criteria were blunt TBI, head-abbreviated injury score (AIS) 3 to 6, age 20 to 50 years, GCS </=8, abnormal brain computed tomographic scan, and intensive care unit admission for 3 days or more. Early deaths (<48 hours) and delayed admissions (>24 hours after injury) were excluded. Patients who underwent ICP monitoring (n = 708) were compared with those did not (n = 938). Multivariate logistic regression was used to determine the relationship between ICP monitoring and survival, while controlling for overall injury severity, TBI severity, craniotomy, associated injuries, comorbidities, and complications.
ICP monitoring was performed in only 43% of patients who met BTF criteria. There were no group differences in age, gender, or GCS. After adjusting for multiple potential confounding factors including, admission GCS, age, blood pressure, head AIS, and injury severity score (ISS), ICP monitoring was associated with a 45% reduction in survival (OR = 0.55; 95% CI, 0.39-0.76; p < 0.001).
ICP monitoring in accordance with current BTF criteria is associated with worsening of survival in TBI patients. A prospective randomized controlled trial of ICP-guided therapy is needed. Until then, the use of ICP monitoring should not be used as a quality benchmark.

0 Bookmarks
 · 
111 Views