Abstract
BACKGROUND: To provide high-quality evidence for extracorporeal shock wave therapy (ESWT) for low back pain (LBP). METHODS: A computerized search screened trials of ESWT for LBP, and the time range was from the establishment of databases to August 2023, including CNKI, CBMdisc, Wanfang Data, VIP, PubMed, Web of Science, Embase, and Cochrane databases. The outcomes were extracted, including patients' pain, physical and psychological conditions, and adverse reactions. Meta-analysis was completed using Revman 5.4 and Stata 15 software, and GRADEpro software quantitated rate the evidence and assigned a recommendation strength. RESULTS: This meta-analysis included 22 studies involving 1749 patients. After treatment, patients in the ESWT group had lower scores in the visual analog scale (VAS) (mean difference [MD] = -1.14, 95% confidence interval [CI] (-1.47, -0.80), P < .00001), Oswestry disability index (ODI) [MD = -6.01, 95%CI (-7.97, -4.05), P < .00001], and Beck depression inventory [MD = -3.89, 95%CI (-6.59, -1.20), P = .005]; finger-floor distance [MD = -3.71, 95%CI (-5.26, -2.17), P < .00001] was shorter; Japanese orthopedic association scores [MD = 3.20, 95%CI (1.56, 4.85), P = .0001] were higher; at the 3-month follow-up, the VAS [MD = -1.29, 95%CI (-2.39, -0.19), P = .02] and ODI [MD = -5.95, 95%CI (-10.06, -1.84), P = .005] remained lower; adverse reactions [odds ratio = 0.19, 95%CI (0.07, 0.52), P = .001] were less, and the valid cases [odds ratio = 3.84, 95%CI (2.47, 5.96), P < .00001] were more. Sensitivity analyses revealed that the results were stable; Egger test found no publication bias in VAS (P = .270) and valid cases (P = .747); the trim and fill analysis tested the robustness of ODI scores due to the publication bias (P = .029), and the effect sizes before and after the trim and fill were consistent, which means the result was robust. CONCLUSION: ESWT for LBP has efficacy and safety in improving pain, dysfunction, and psychological conditions compared to other therapies. ESWT has shown advantages in terms of long-term efficacy. Randomized controlled trials with larger sample sizes and more objective outcomes are required.