143. Determining the best vertebra for measuring pelvic incidence and spinopelvic parameters in transitional anatomy

Abstract

<h3>BACKGROUND CONTEXT</h3> Sacralization of L5 is a condition that affects 15% of the population. Because of the increased L5 articulations to the sacrum, it is unclear how pelvic incidence and spinopelvic parameters should be measured. <h3>PURPOSE</h3> Determine whether patients with L5 sacralization should be radiographically measured with relation to the L5 or S1 vertebrae. <h3>STUDY DESIGN/SETTING</h3> Retrospective analysis. <h3>PATIENT SAMPLE</h3> Prospective multicenter adult spinal deformity database. <h3>OUTCOME MEASURES</h3> ODI, SF36 PCS, radiographic spinopelvic parameters. <h3>METHODS</h3> Linear regression modeling was used to determine normative TPA and PI-LL based on PI and age in a database of asymptomatic subjects. In an adult spinal deformity database, patients with radiographic evidence of L5 sacralization, had PI, LL and TPA measured from the superior endplate S1 and then also from L5. Patients with instrumentation below L4 at baseline were excluded. The differences in TPA and PI-LL from normative were calculated in the sacralization cohort relative to L5 and S1 and correlated to ODI. Patients were grouped based on PI-LL Schwab modifier (0, +, ++) utilizing their L5 PI-LL and S1 PI-LL. Baseline ODI and SF36 PCS were compared across and within groups and their respective SRS Schwab class. <h3>RESULTS</h3> Among 1,179 ASD patients, 276 (23.4%) had transitional anatomy, 176 with sacralized L5 (14.9%) and 100 (8.48%) with lumbarization of S1. The 176 with sacralized L5 were analyzed. PI (24.5 ± 11.0 vs 55.7 ± 12.0, p=0.001), TPA (11.2 ± 12.0 vs 20.3 ± 12.5, p=0.001) and PI-LL (0.67 ± 21.1 vs 11.4 ± 20.8, p=0.001) measured utilizing the L5 superior endplate were significantly smaller than those who were measured relative to S1. When measured from S1, 76 (43. %) of patients were SRS Schwab 0, 45 (25.6%) were Class +, and 55 (31.3%) were class ++ compared to 124 (70.5%), 22 (12.5%), and 30 (17.0%) measured from L5 respectively. There were significant differences in ODI and PCS as the Schwab grade increased regardless of L5 or S1 measurement. The L5 group had higher PCS functional scores for Schwab 0 and ++ relative to same grades in the S1 group. Offset from normative TPA (0.5 ± 11.1 vs 9.6 ± 10.8, p=0.001) and PI-LL (0.67 ± 21.1 vs 11.4 ± 20.8, p=0.001) were smaller when measuring from L5. Moreover, S1 measurements were more correlated with disability by ODI (TPA offset from normative, S1: R=0.326 vs L5: R=0.285; PI-LL Offset from normative, S1: R=0.318 vs L5: R=0.274). <h3>CONCLUSIONS</h3> Measuring the PI at L5 in sacralized anatomy results in underestimating the PILL Schwab grade in a percentage of patients and less correlation with HRQLs. Patients with sacralized L5s should have spinopelvic parameters measured relative to S1. <h3>FDA DEVICE/DRUG STATUS</h3> This abstract does not discuss or include any applicable devices or drugs.