224. Does Roussouly type affect revision surgery rates for proximal junctional kyphosis (PJK) and adjacent segment degeneration (ASD) after fusion for adult scoliosis

BACKGROUND CONTEXT As spine types have been classified into 4 Roussouly types based upon their sacral slope and shape, it is unclear if certain types lead to higher or lower rates of adjacent segment breakdown after spinal fusion for scoliosis. PURPOSE There are certain Roussouly types that may predispose patients to a higher rate of proximal segment breakdown compared to others. STUDY DESIGN/SETTING Retrospective review of a single center North American spine center scoliosis cases from 2004 to 2016. PATIENT SAMPLE A total of 193 patients with minimum 2-year follow up were included. Mean follow up was 45 months. OUTCOME MEASURES PJK and ASD can occur after fusion for adult scoliosis. It is unclear if one Roussouly type predisposes or protects against proximal segment breakdown. We wished to investigate if certain Roussouly types were associated with higher or lower rates of adjacent segment breakdown. METHODS Retrospective review at a single North American spine center scoliosis cases. Radiographs were reviewed, and all cases were classified by Roussouly type (1 to 4) based upon preoperative 3-foot standing lateral radiographs. Records were reviewed for revision surgery proximal to the construct only, for either ASD or PJK, and patients were excluded if surgery was performed for infection, wound problems, implant failure, or surgery at the original index levels. Fusion types were classified as short ( RESULTS Patients were classified as type 1: 48, type 2: 66, type 3: 56, and type 4: 23. When evaluated by Roussouly type, there was no difference in revision surgery rates for ASD or PJK. However, when evaluating only type 2 spines, short fusion had a revision surgery rate of 35.1% (13/37) compared to 13.8% (7/29) for long fusions (p=0.0491). With regards to the entire cohort of 193 patients, short segment fusions had a revision surgery rate at the proximal level of 33.6% (36/107) compared to a long fusion rate of 16.3% (14/86), p=0.0014. CONCLUSIONS Short segment fusions had a statistically higher rate of surgery for proximal pathology for all spine types, and Roussouly type 2 spines had a higher rate of proximal pathology surgery with short fusions compared to long fusions. We could not detect a difference of revision surgery rates for proximal pathology for adult scoliosis based solely upon Roussouly type. FDA DEVICE/DRUG STATUS This abstract does not discuss or include any applicable devices or drugs.