P45. The predictive value of a novel site-specific MRI-based bone quality assessment, endplate bone quality (EBQ), for Severe Cage Subsidence among patients undergoing standalone lateral lumbar interbody fusion

Abstract

<h3>BACKGROUND CONTEXT</h3> Methods to measure bone quality using opportunistic imaging have recently been developed, including quantitative computed tomography (QCT), which allows for site-specific volumetric bone mineral density (vBMD) measurement, such as pedicles or endplates. Studies have shown these site-specific QCT measurements had more predictive value for site-specific complications than standard L1/2 measurements. Recently, MRI-based assessments of the vertebral bone quality (VBQ), which measures trabecular bone quality, were also introduced. However, there have been few studies that investigate the association between site-specific MRI bone assessment and osteoporosis related complications in patients undergoing lumbar interbody fusion. In this work, we created a novel site-specific MRI-based assessment of the endplate bone quality (EBQ) and assessed its predictive value for severe cage subsidence following standalone lateral lumbar interbody (SA-LLIFF) fusion. <h3>PURPOSE</h3> The purpose is to introduce a novel MRI-based site-specific bone quality assessment, EBQ, and investigate the predictive value of EBQ for severe cage subsidence after SA-LLIF compared with standard VBQ. <h3>STUDY DESIGN/SETTING</h3> A retrospective observational study at a single academic institution. <h3>PATIENT SAMPLE</h3> Patients undergoing SA-LLIF between 2008-2019 with available preoperative CT and T1-weighted MRIs of the lumbar spine as well as postoperative lumbar radiographs or CTs. <h3>OUTCOME MEASURES</h3> Severe cage subsidence (Grade 2 or 3) within 5-14 month after SA-LLIF. <h3>METHODS</h3> We retrospectively reviewed the records of SA-LLIF patients. Cage subsidence was assessed using the classification of Marchi et al. and severe cage subsidence was defined as grade 2 or 3. We measured VBQ as shown in the original article by Ehresman et al. EBQ was measured using preoperative non-contrast T1-weighted MRIs of the lumbar spine. The regions of interest were the subchondral bones of upper and lower endplates at the operated level, and the average signal intensity of both endplates was divided by that of the cerebrospinal fluid space at the level of L3. As a confounding factor, vBMD was measured using asynchronous QCT. Bivariate and multivariable analysis with generalized linear mixed models were performed with setting binary severe subsidence as the outcome. Statistical significance was set at p=0.05. The quality of fit was assessed using Akaike's information criterion. <h3>RESULTS</h3> A total of 205 levels in 89 patients were included in the final analysis. There were 50 levels (24.4%) which demonstrated severe subsidence. QCT-vBMD was significantly lower in severe subsidence group (Mean (SD) 97.4 (34.4) vs 110.1(33.4), p=0.021). Both VBQ and EBQ were significantly higher in severe subsidence group (VBQ: Mean (SD) 2.67 (1.08) vs 2.39 (0.44), p=0.010; EBQ Mean (SD) 5.09 (2.20) vs 4.31 (1.09), p=0.001). In multivariate analyses adjusted with QCT-vBMD, EBQ showed significant association with severe cage subsidence (OR (95% confidential interval) = 1.94 (1.36-3.63), p=0.038), whereas VBQ only showed a marginal trend (p=0.071). The EBQ based prediction model for severe subsidence showed better goodness of fit compared to VBQ based model. (AIC EBQ 190 vs VBQ 192). <h3>CONCLUSIONS</h3> We introduced a novel MRI-based site-specific bone quality assessment. High EBQ was an independent factor for severe cage subsidence after SA-LLIF and the EBQ-based model showed better goodness of fit compared to VBQ-based model. Given the availability of preoperative MRIs, EBQ assessment prior to SA-LIFF may provide insight into a patient's risk for severe subsidence. <h3>FDA DEVICE/DRUG STATUS</h3> This abstract does not discuss or include any applicable devices or drugs.