Long-term reoperation after decompression with versus without fusion among patients with degenerative lumbar spinal stenosis: a systematic review and meta-analysis

The debate over adding fusion after decompression in lumbar spinal stenosis patients without spondylolisthesis is due to the "absence of evidence" in its benefits, particularly in reoperation. However, this “absence of evidence” does not indicate “evidence of absence.” To investigate the reoperation rates following the addition of fusion after decompression in patients with lumbar spinal stenosis without spondylolisthesis. Systematic review and meta-analysis. We searched Medline, Embase, Web of Science, and Google Scholar databases on December 12, 2021, with an updated search conducted on April 06, 2023. Inclusion criteria were adult patients with lumbar spinal stenosis. Exclusions comprised cases of spondylolisthesis and instabilities. The occurrence of reoperation was summarized using odds ratios (OR), while other outcomes were presented as mean differences. We employed a Cox-based shared-frailty model with random effects for the time-to-event analysis of reoperation. Additionally, we used a 2-stage method to validate our estimates. Heterogeneity variance within the random-effects model was estimated using the Hartung-Knapp-Sidik-Jonkman method. A total of 1973 studies were identified and screened, of which 48 met selection criteria, and 17 were included in the meta-analysis. Comparison between fusion and non-fusion groups in patients with lumbar stenosis and neurological claudication revealed no significant difference in reoperation rates (odds ratio: 1.13 [95% CI: 0.88 to 1.46]; 8016 participants; 14 studies; I2 = 0%). Bayesian analysis indicated an 8.9-fold likelihood of similar reoperation rates. Time-to-reoperation analysis revealed a 16.46 months delay in the fusion group, though not statistically significant (mean difference: 16.46 [95% CI: -3.13–36.04]; 83 participants; 3 studies; I2 = 46%). Consistently, ODI, back pain, and leg pain VAS showed no significant differences. The certainty of the evidence was low for odds of reoperation and leg pain VAS, and very low for the remaining outcomes. In lumbar spinal stenosis patients without spondylolisthesis, the addition of fusion after decompression showed limited benefits in terms of reoperation rates, ODI, and leg pain.