Review of treatment trials in human spinal cord injury: Issues, difficulties, and recommendations

OBJECTIVE: To provide a comprehensive review of the treatment trials in the field of spinal cord injury, emphasizing what has been learned about the effectiveness of the agents and strategies tested and the quality of the methodology. The review aims to provide Useful information for the improvement of future trials. The review audience includes practitioners, researchers, and consumers. METHODS: All publications describing organized trials since the 1960s were analyzed in detail, emphasizing randomized, prospective controlled trials and published Phase I and II trials. Trials were categorized into neuroprotection, surgery, regeneration, and rehabilitation trials. Special attention was paid-to design, outcome measures, and case selection. RESULTS: There are 10 randomized prospective control trials in the acute phase that have provided much useful information. Current neurological grading systems are greatly improved, but still have significant shortcomings, and independent, trained, and blinded examiners are mandatory. Other trial designs should be considered, especially those using adaptive randomization. Only methylprednisolone and thyrotropin-releasing hormone have been shown to,be effective, but the results of the former are controversial, and studies involving the latter involved too few patients. None of the surgical trials has proven effectiveness. Currently, a multitude of cell-based Phase I trials in several countries are attracting large numbers of patients, but such treatments are unproven in effectiveness and may cause harm. Only a small number are being conducted in a randomized or blinded format. Several consortia have committed to a promise to improve the conduct of trials. CONCLUSION: A large number of trials in the field of spinal cord injury have been conducted, but with few proven gains for patients. This review reveals several shortcomings in trial design and makes several recommendations for improvement.