Citation

Gomes-Osman J, Cortes M, Guest J, Pascual-Leone A. J. Neurotrauma 2015; 33(5): 425-438.

Affiliation

Beth Israel Deaconess Medical Center, Harvard Medical School, Berenson-Allen Center for Noninvasive Brain Stimulation, Department of Neurology, Boston, Massachusetts, United States ; apleone@bidmc.harvard.edu.

Copyright

(Copyright © 2015, Mary Ann Liebert Publishers)

DOI

10.1089/neu.2014.3812

PMID

26415105

Abstract

While various approaches have been proposed in clinical trials aimed at improving motor function after spinal cord injury in humans, there is still limited information regarding the scope, methodological quality and evidence associated with single-intervention and multi-intervention approaches. A systematic review carried out using the PubMed search engine and the keywords "spinal cord injury motor recovery" identified 1,973 records, of which 39 were selected (18 from the search records and 21 from reference list inspection). Study phase (clinicaltrials.org criteria) and methodological quality (Cochrane criteria) were assessed. Studies included proposed a broad range of single-intervention (encompassing cell therapies, pharmacology, electrical stimulation, rehabilitation) (encompassing cell therapies, pharmacology, electrical stimulation, rehabilitation), and multi-intervention approaches (that combined more than one strategy). The highest evidence level was for phase III studies supporting the role of multi-intervention approaches that contained a rehabilitation component. Quality appraisal revealed that the percentage of selected studies classified with high risk of bias by Cochrane criteria was as follows: random sequence generation=64%; allocation concealment=77%; blinding of participants and personnel=69%; blinding of outcome assessment=64%; attrition=44%; selective reporting=44%. The current literature contains a high proportion of studies with a limited ability to measure efficacy in a valid manner due to low methodological strength in all items of the Cochrane risk of bias assessment. Recommendations to decrease bias are discussed, and include increased methodological rigor in the study design and recruitment of study participants, and the use of electrophysiologic and imaging measures that can assess functional integrity of the spinal cord (and may be sufficiently sensitive to detect changes that occur in response to therapeutic interventions).

Language: en