Introduction: Several systems have been developed to classify thoracolumbar alignment in adult spinal deformity (ASD), however with increasing focus on value-based healthcare, the cost effectiveness of meeting additional alignment criteria has yet to be explored. The objective of this study is to compare cost effectiveness of meeting additional alignment criteria.

Methods: This is a retrospective study of a prospective, single-center ASD-database. Operative patients who met criteria for ASD with baseline (BL) and 1-year (1Y) radiographic and health related quality of life (HRQL) data were included. At BL and 1Y, patients were assessed using four published alignment systems: SRS-Schwab (includes: Pelvic-Tilt(PT), Sagittal-Vertical-Axis(SVA) and Pelvic-Incidence minus-Lumbar-Lordosis(PI-LL)); GAP Score, a sagittal disproportion score out of 13; Age-Adjusted Alignment, which adjusts SRS-Schwab parameters to age-adjusted ideal; Roussouly Type, assessed as ‘Match’ or ‘Mismatch’ their theoretical type. Patients were classified as meeting SRS-Schwab if 1Y PT, PI-LL, or SVA decreased in severity or maintained a score of 0 at 1Y. Patients were classified as meeting GAP, if 1Y GAP score decreased in severity or maintained a score between 0 and 2 at 1Y. Meeting age-adjusted and Roussouly were based on matching at 1Y. Patients were separated into 5 groups by number of criteria (0 – 4) met at 1Y and cost per Quality-Adjusted-Life-Year(QALY) was calculated by group. Within groups meeting 1, 2, or 3 criteria, cost/QALY was calculated overall, regardless of which combination of criteria were met and by specific combinations of criteria met. ANCOVA estimated marginal means of complications and reoperations adjusting for BL age, deformity, approach and invasiveness. ANCOVA also estimated marginal means for ODI, adjusting for BL age, gender, deformity, Charlson Comorbidity Index (CCI) and frailty index (FI). A cost analysis was completed on the PearlDiver database accounting for approach, revisions, complications and comorbidities. For QALY analysis, utility was calculated using ODI converted to SF-6D. A 3% discount rate was applied to account for residual decline to life expectancy (LE), 78.70 years. Trendline analysis noted changes over time.

Results: 364 patients were included (58.90±14.55yrs, 81.54%-female, 26.85±5.42kg/m²). Of these, 15-met 0-of-4, 62-met 1-of-4, 130-met 2-of-4, 125-met 3-of-4, & 32-met all 4 alignment systems. Controlling for BL age, deformity, approach & invasiveness, patients meeting 0-of-4 had higher rates of major complications than all other groups, p=.004. 1Y cost was highest in patients meeting 0-of-4 and lowest in patients meeting any 1-of-4,p=.009 (Figure 1). Patients that met 0-of-4 had lowest QALYs gained at 1Y(0.10) whereas 1/4 had the highest(0.20) (Table 1). Cost/QALY at LE was highest in pts that met 0/4($113,583.41) and lowest in those meeting 1-of-4 ($41,990.58),p<.05. Of those meeting any 1 system, patients meeting age-adjusted had lowest Cost/QALY at LE ($41,559.31).

Discussion: Patients that improved in any one of the alignment systems at 1Y had lower complication rates, lower cost, higher QALYs gained and lowest Cost/QALY at 1Y. Of patients meeting only 1 criteria, age-adjusted proved to be most cost-effective. These data suggest that in terms of maximizing cost effectiveness, it may not be necessary to realign patients all 4 classification systems.