A critical appraisal of clinical practice guidelines on pharmacological treatments for spinal cord injury.

Background Context: Spinal cord injury brings devastating consequences and huge economic burden. Different authoritative organizations have developed different guidelines for pharmacological treatments of spinal cord injury, but there is a lack of a critical appraisal of them.

Purpose: To systematically review and appraise guidelines regarding their recommendations for pharmacological treatments for spinal cord injury.

Study Design: Systematic review.

Methods: We searched Medline, Embase, Cochrane, and Web of Science from January 2000 to January 2022 as well as guideline-specific databases (eg, Congress of Neurological Surgeons) and Google Scholar. We included the most updated guideline containing evidence-based recommendations or consensus-based recommendations developed by specific authoritative organizations if multiple versions were available. We appraised guidelines through the Appraisal of Guidelines for Research and Evaluation, 2nd edition instrument consisting of six domains (eg, applicability). With supporting evidence, recommendations were classified as: for, against, neither for nor against. We utilized an evidence assessment system to categorize the quality of supporting evidence as poor, fair, or good.

Results: Eight guidelines developed from 2008 to 2020 were included, but all of them scored lowest in the domain of applicability among all six domains. Twelve pharmacological agents (eg, methylprednisolone) were studied. For methylprednisolone, three guidelines (3/8=37.5%) recommended for (one evidence-based and two consensus-based), three (3/8=37.5%) recommended against (all evidence-based), and two (2/8=25%) recommended neither for nor against. For monosialotetrahexosylganglioside (GM-1), one guideline (1/4=25%) recommended for (consensus-based), one (1/4=25%) recommended against (evidence-based), and two (2/4=50%) recommended neither for nor against. For other agents (eg, minocycline), most guidelines (3/5=60%) recommended neither for nor against, one (1/5=20%) recommended against naloxone (evidence-based) and nimodipine (evidence-based), and one (1/5=20%) recommended for neural growth factor (consensus-based). The quality of most of the supporting evidence was poor, and the rest was fair.

Conclusions: There were inconsistencies among recommendations for methylprednisolone and GM-1. Evidence-based recommendations tended to recommend against, whereas consensus-based recommendations tended to recommend for.