187. Lumbar lordosis is primarily lost in the upper lumbar spine in ASD patients with sagittal deformities

Abstract

<h3>BACKGROUND CONTEXT</h3> In asymptomatic individuals, two-thirds of the lordosis comes from the L4-S1 level. It is commonly believed that the majority of loss of lordosis in adult spinal deformity (ASD) is attributable to the distal (L4-S1) levels. <h3>PURPOSE</h3> This study aims to investigate the location of loss of lordosis in a cohort of ASD patients prior to surgical intervention. <h3>STUDY DESIGN/SETTING</h3> Retrospective review of prospective databases. <h3>PATIENT SAMPLE</h3> This study included 119 symptomatic adult volunteer and 357 adult spinal deformity patients. <h3>OUTCOME MEASURES</h3> Regional and focal alignment. Lordosis distribution. <h3>METHODS</h3> A registry of asymptomatic volunteers was used to build age and pelvic incidence (PI) adjusted normative models of PI-LL, L1-L4 lordosis, L4-S1 lordosis, TL kyphosis (T10-L2), and thoracic kyphosis (T1-T12). Our study cohort was a registry of surgical primary ASD patients without coronal malalignment (SRS-Schwab Type=N). The formulas developed in the asymptomatic population were applied to the ASD group to calculate normative values for each patient. The ASD population was divided into four groups based on quartiles of PI-LL: no, mild, moderate, and severe PI-LL mismatch. Within each group, paired t-tests were performed to compare actual and calculated normative sagittal alignment; actual alignment was also expressed in percentage of normative values. The level of significance was p<0.05. <h3>RESULTS</h3> The asymptomatic cohort used included 119 asymptomatic volunteers (50.7yo±17, PI: 52°±11.4) with the following regional alignment: L4-S1=34°, L1-L4=23°, TKL=3°, and TK=49°. The study cohort included 357 ASD patients (64.6yo, 58.5%F). The PI-LL of the 4 quartiles were -10°, 10°, 20°, and 40°. There were no significant differences in PI or in any of the coronal Cobb angles. Comparing each ASD group to the calculated normative values; the "no mismatch" group had a greater L4-S1 (+8°), a greater T10-L2 kyphosis (+14°) and T1-T12 kyphosis (+8°), but no difference in L1-L4. The "mild mismatch" group had a smaller L1-L4 (-12°), greater TLK (+5°), smaller TK (-7°) and no difference in L4-S1. The "moderate mismatch" had a smaller L4-S1 (-7°), L1-L4 (-15°), TK (-15°) but no difference in TLK. Finally, the "severe mismatch" group had a smaller L4-S1 (-15°), L1-L4 (-26°), and TK (-27°) but no difference in TKL (all with p<0.001). The analysis by percentage of actual alignment vs the calculated age and PI matched normative values permitted to identify the driver(s) of the sagittal malalignment. The "no mismatch" had an excess of TLK (+510%) compensated by an excess in L4-S1 (+27%). The "mild mismatch" had a loss of L1-L4 (-48%) with a normal L4-S1, while the "moderate mismatch" had mainly a deficit in L1-L4 (-66%) associated to a 22% loss on L4-S1. Finally, "severe-mismatch" presented a kyphotic L1-L4 (-115% vs norm) with a 46% deficit in L4-S1 <h3>CONCLUSIONS</h3> In this cohort of ASD patients with limited coronal deformity, the majority of the loss of lordosis is attributable to the proximal segment. As the deformity worsened, the loss of lordosis also involved the distal lumbar segments. <h3>FDA DEVICE/DRUG STATUS</h3> This abstract does not discuss or include any applicable devices or drugs.