ABSTRACT
Background: Low back pain is a global public health concern and a major cause of life-time disability. The use of non-steroidal anti-inflammatory drugs (NSAIDS) for low back pain is a popular practice unlike exercise which seems less attractive. However, to foster evidence-based decision and care, we evaluated the comparative efficacy of NSAIDS and exercise for low back pain.
Methods: This is a network meta-analysis of randomized control trial articles published in English language, centered on use of NSAIDs and/or exercise to improve pain in patient with low back pain.
Articles were searched for on MEDLINE, PubMed, CINAHL, Academic Search Complete, and PsycINFO. Search results were exported into EndNote20, where duplicate articles were removed. Full text screening was done using COVIDENCE (Extraction 2). Data was selected and extracted according to predesigned eligibility criteria using a standardized data extraction table. We used a random-effect model of meta-analysis to estimate the efficacy of NSAIDS and exercise, and network meta-analysis to compare their separate efficacies. We appraised the quality of the included studies with the PeDro scale.
Results: Relative to control/placebo, there were significant reductions in pain intensity with NSAIDs (SMD = -0.769 95% CI = -1.290 to -0.248; I2 = 96.634) and exercise (SMD = -1.563 (95% CI = -2.784 to -0.342). Direct comparison of NSAIDs and exercise showed marginal significant reduction in pain intensity amongst individuals who received exercise compared to NSAIDs (SMD= -0.138 (95% CI = -0.384 – 0.660), P =0.605). Indirectly, result shows no significant difference between pharmacotherapy and acupuncture ((SMD = 0.023 ((95% CI = -0.688 to 0.733), p = 0.950). Overall, in favor of exercise, we obtained a combined estimate of SMD = -0.483 (95% CI =-2.059 to 1.093).
