Evaluation of the Utility of Morphological Alignment in Revision Adult Spinal Deformity Corrective Surgery: Do Revisions Obscure Recognition, Implementation, and Outcomes?

Introduction: Determination of optimal spinal alignment and shape becomes increasingly complicated by the presence of existing hardware in ASD revision surgery.

Methods: We included ASD patients undergoing revision surgery with a minimum of 2-year follow-up undergoing fusion from at least L1 to sacrum. Patients were stratified based on whether they had prior fusion from L1 and proximal to sacrum(PF) or not(NPF). Roussouly morphotype based on SS “current” was defined as: Type 1(T1): SS < 35° and 3 or fewer lordotic vertebra(kyphosis-lordosis inflection point was caudal to L2/3); Type 2 (T2): SS < 35° and 4 or more lordotic vertebrae(inflection point was cranial to L2);Type 3-AP(T3A): 35°≤SS < 45°, PI < 50, and PT < 5°;Type 3(T3): 35°≤SS < 45°, PI≥50°,or PT≥5°; and Type 4(T4):SS≥45°. Subanalysis conducted on postoperative match of Roussouly based on PI “theoretical”:Type 1(T1): PI < 45, LL apex below L4; Type 2(T2):PI < 45, LL apex above L4-L5 space;Type 3(T3PI): 45< PI < 60;Type 4(T4PI): PI>60.

Results: 334 patients met inclusion(Age:63±10,77%F,BMI:27.6±5.1kg/m2,frailty:3.5±1.5,CCI:1.9±1.7). Overall, 9.0% had PF. Roussouly categories preoperatively: 40% T1,26% T2, 24% T3, 1% T3AP, 9% T4. Overall, 48% met Roussouly targets postoperatively, with 35% of those in T1 meeting the correct target type, 60% in T2, 49% in T3, 0% in T3AP, and 72% in T4. Evaluation of PF depicted no difference in rates of matching overall and by type. Furthermore, no difference in PF and NPF rates by type was found when assessing only those who were mismatched at baseline. However, in those matched preoperatively(54% of the overall cohort), depicted a 40% lower rate of remaining matched in PF(PF 29% vs NPF 49%,p=.14), which was most prominent in T2(NPF 59% vs 17%,p=.03). Becoming mismatched when matched preoperatively was associated with 11x likelihood of not achieving MCID for ODI in those with prior fusion(OR:11.928,[1.081-131.614],p=0.043).

Conclusion : Prior lumbar fusion affected rates of matching ideal spinal morphological targets for Type 2 who were matched preoperatively. These findings indicate that significant pelvic retroversion, in the setting of low PI, was more challenging to identify and correct in those undergoing revision.