New Radiologic Parameters Predict Clinical Outcome after Decompressive Craniectomy

Abstract

OBJECTIVE:
Decompressive craniectomy (DC) is an established part of the management of patients with increased intracranial pressure due to malignant middle cerebral artery (MCA) infarction or traumatic brain injury (TBI). The aim of this study was to determine prognostic radiologic parameters regarding the functional outcome of patients with increased intracranial pressure (ICP) undergoing DC. Special focus was put on the potential differences between malignant MCA infarction and TBI.
METHODS:
A total of 113 patients were analyzed for their clinical course where preoperative and postoperative radiologic features in computed tomography (CT) scans were correlated to the clinical outcome assessed by the Glasgow Outcome Scale. The difference between presurgical and postsurgical midline shift (ΔMLS) and the ratio between the diameter of the affected and contralateral hemisphere (HDratio) in presurgical and postsurgical CT scans were calculated.
RESULTS:
ΔMLS (MCA infarction group) and postsurgical HDratio (TBI group) were found to be highly correlating with the clinical outcome on Spearman-correlation testing and underwent further analysis using a binary logistic regression model to evaluate their prognostic value on the outcome, which showed the predictive power of ΔMLS in malignant MCA infarction patients (odds ratio [OR] 0.715; confidence interval [CI] 0.551-0.865). Postsurgical HDratio correlated significantly (OR 0.620; CI 0.384-0.901) with the outcome in the TBI group.
CONCLUSION:
ΔMLS is an objectifiable parameter, predicting outcome in malignant MCA infarction. In contrast, ΔMLS was of no predictive value in TBI patients. Here postsurgical HDratio serves as a strong predictor of clinical outcome. We recommend applying postsurgical HDratio to TBI patients in order to estimate their clinical outcome and adjust treatment.

Bibliographical data

Original languageEnglish
ISSN1878-8750
DOIs
Publication statusPublished - 04.2016
PubMed 26548831