Incremental Benefit of Revision Surgery After Adult Spinal Deformity Surgery: At What Point Does the Value of Multiple Revisions Diminish in ASD Corrective Surgery and Does It Depend on the Etiology

Introduction: Incidence of revision surgery after surgical correction of ASD remains high. However, the benefits of multiple revision surgery on patient outcomes are not well-defined. This study aims to assess impact of revision surgery on surgical outcomes, alignment, and HRQLs.

Methods: ASD patients were stratified by revisions (R1, R2, or R3 greater than 2). Baseline demographic and surgical details were from timepoint visit prior to revision. Follow-up based on timepoint 2Y after final revision. 3 revision groups were PSMws using many-to-one matching adjusting for revision baseline. Radiographics SVA, PI-LL, PT, and GAP score assessed alignment post-revision via ANOVA. MVA assessed differences in outcomes after final revision surgery.

Results: After PSM, 252 patients remained (R1: 108; R2: 108; R3: 36). Revision baseline deformity (SVA, PI-LL, PT, GAP) showed no significant differences between groups. 2 years after final revision surgery, ANOVA showed significant differences between R1 and R3, and R2 and R3, with those undergoing greatest number of revisions R3 demonstrating less improvement in deformity measures. Pre-revision ODI scores showed a decline with increasing revisions. R1 and R2 had greater difference between baseline and 2 year ODI scores, indicating a sustained benefit to revision (-13.0 vs -13.9 vs -9.7). Percentage of those reaching SCB in ODI from pre-revision baseline to years were R1 71.3%, R2 60.2%, and R3 38.9%, with significant difference found between R1 and R3, and R2 and R3 (all p < .05). MVA demonstrated a significantly lower odds of reaching MCID in ODI for R2 and R3 compared to R1. Overall, surgical indications for revisal increased with number of revisions: R3 had significantly greater percentage of surgical and mechanical complications leading to revision.

Conclusion : These results suggest that after two revisions, improvement in radiographic alignment may be limited. However, patient-reported outcomes do not show the same diminishing returns with the appropriate indication.