Objective: We sought to evaluate the efficacy and safety of lumbar fusion versus nonoperative care for the treatment of chronic low back pain associated with degenerative disk disease.

Methods: A comprehensive duplicate electronic database search including PubMed, Embase, Cochrane Library, and China National Knowledge Infrastructure was conducted to identify relevant prospective studies published up to June 30, 2020. The main outcomes including clinical results, complications, and all-cause additional surgeries were presented in the form of short- (<2 years) and long-term (≥2 years) follow-up results. Analyses were performed in Review Manager software 5.3.

Results: Six prospective studies involving 159 patients for short-term follow-up and 675 for long-term follow-up were included. The 2 interventions exhibited little difference in regard to short- and long-term Oswestry Disability Index and visual analog scale scores for back and leg pain, except that lumbar fusion might bring about lower additional surgery rate (risk ratio, 0.40; 95% confidence interval, 0.17 to 0.98; P = 0.04), and higher complication rate (risk ratio, 21.46; 95% confidence interval, 4.34 to 106.04; P = 0.0002) in the long term.

Conclusions: The present meta-analysis determined that fusion surgery was no better than nonoperative treatment in terms of the pain and disability outcomes at either short- or long-term follow-up. It is necessary for clinicians to weigh the risk of complications associated with fusion surgery against additional surgeries after nonoperative treatment. Considering lax patient inclusion criteria in the existing randomized clinical trials, the result needs to be further confirmed by high-quality research with stricter selection criteria in the future.