Magnetic resonance spectroscopy abnormalities in traumatic brain injury: A meta-analysis - 14/11/17

Doi : 10.1016/j.neurad.2017.09.004

Matthew Brown ^a,^⁎ , Hediyeh Baradaran ^a, Paul J. Christos ^b, Drew Wright ^b, Ajay Gupta ^a,^c, Apostolos J. Tsiouris ^a
^a Departments of Radiology, Healthcare Policy, Research, New York Presbyterian Hospital, Weill Cornell Medicine, 525 East 68th Street, 10065 New York, NY, USA
^b Healthcare Policy and Research, New York Presbyterian Hospital, Weill Cornell Medicine, 525 East 68th Street, 10065 New York, NY, USA
^c Feil Family Brain and Mind Research Institute, Weill Cornell Medicine, 407 E 61st St, 10065 New York, NY, USA

^⁎Corresponding author. 420 E. 70th St, New York, NY 10021, USA. Fax: +212 746 8597.

Sous presse. Épreuves corrigées par l'auteur. Disponible en ligne depuis le Tuesday 14 November 2017
Cet article a été publié dans un numéro de la revue, cliquez ici pour y accéder

Highlights

•	Significant changes occur on magnetic resonance spectroscopy following TBI.
•	There is strong evidence of MRS abnormalities in moderate-to-severe TBI.
•	There is evidence of MRS abnormalities in the subacute and chronic time phases.
•	The literature does not conclusively show changes following mild TBI.
•	There is no conclusive evidence of MRS abnormalities within the first seven days of injury.

Le texte complet de cet article est disponible en PDF.

Abstract

Background and purpose

Despite traumatic brain injury (TBI) being common, evaluation with imaging remains challenging. Magnetic resonance spectroscopy (MRS) shows promise in detecting changes of brain metabolite concentrations following TBI; however, currently there are only small studies available without conclusive evidence of the technique's efficacy. The purpose of this systematic review and meta-analysis was to evaluate the association between TBI and MRS metabolite changes.

Materials & methods

A comprehensive literature search was performed looking for studies reporting brain metabolite concentrations in both TBI and control subjects. Included studies reported values for both adult TBI and control subjects. Cumulative and subgroup meta-analyses were performed using a random effects model.

Results

The literature search returned an initial 898 manuscripts, of which 36 (which included 748 unique subjects) met study criteria. Cumulatively, NAA/Cr ratios in TBI patients showed a significant decrease as compared to controls (standardized mean deviation [SMD]=−0.88, P<0.0001). When broken into subgroups by severity, the severe and mixed TBI subgroups showed decreases, but the mild TBI (mTBI) subgroup did not. When stratified by time, a significant decrease was seen in the subacute and chronic phases but not the acute phase. Cumulative post-TBI Cho/Cr levels were increased compared to controls (SMD=0.69, P=0.0002). Significant changes were seen in all subgroups except the mild and mixed mTBI subgroups and the acute phase subgroup.