Long-Term Prognosis of Surgical and Non-Surgical Treatment for Lumbar Spinal Stenosis: A Retrospective Cohort Study.

OBJECTIVE

To evaluate the long-term outcomes and prognostic factors of surgical versus non-surgical treatment for lumbar spinal stenosis (LSS).

METHODS

This retrospective cohort study included 210 patients with LSS (122 surgical, 88 non-surgical) from three tertiary spine centers, with a minimum five-year follow-up. Primary outcomes included Oswestry Disability Index (ODI), pain scores, and maintaining minimal clinically important difference (MCID). Secondary outcomes encompassed quality of life measures, walking capacity, and patient satisfaction. Prognostic factors were analyzed using Cox proportional hazards models.

RESULTS

At baseline, the surgical group exhibited more severe symptoms (ODI: 47.5±12.9 vs 41.3±14.7, P<0.001) and higher prevalence of multi-level stenosis (63.1% vs 49.2%, P=0.007). At six months, surgical patients demonstrated greater improvement in ODI (23.5 vs 11.8 points, P<0.001) and leg pain (VAS reduction: 4.6 vs 2.0, P<0.001). This advantage persisted at one year, with 74.6% of surgical patients achieving MCID compared to 42.2% in the non-surgical group. Long-term follow-up (mean 7.1±1.7 years) revealed sustained but attenuated treatment effects, with higher rates of maintaining MCID in the surgical group (63.2% vs 46.4%, P=0.004). Younger age, predominant leg pain, shorter symptom duration, and absence of depression predicted favorable surgical outcomes, while spondylolisthesis negatively impacted non-surgical outcomes.

CONCLUSION

Surgical decompression provides superior early improvement in functional status and pain compared to non-surgical management, with benefits partially maintained beyond five years. However, outcome convergence over time suggests that both approaches can be effective for appropriately selected patients. Prognostic factors identified in this study may guide personalized treatment decisions for patients with LSS. This study suggests an association between surgical decompression and superior initial improvement. However, due to the non-randomized design, these findings must be interpreted with caution.