P73. A novel preoperative MRI-based lumbar muscle health calculation to predict patient-reported health-related quality of life scores

Abstract

BACKGROUND CONTEXT Poor lumbar muscle health has been implicated as a source of disability for patients with low back/radicular pain. We wanted to evaluate the relationship between muscle health and patient reported quality of life scores. PURPOSE Determine how preoperative muscle health impacts patient reported health related quality of life scores. STUDY DESIGN/SETTING Retrospective review of imaging and outcome scores. PATIENT SAMPLE A total of 92 adult patients with lumbar spine pathology requiring MIS lumbar decompression/fusion. OUTCOME MEASURES Health-related-quality of life (HRQL) scores, paralumbar muscle cross sectional area (PL-CSA), lumbar indentation value (LIV), Goutallier classification. METHODS We performed a retrospective review of patients that had lumbar decompression and/or fusion surgery after failing nonop management for degenerative pathologies. We quantified muscle health using PL-CSA, LIV and lumbar muscle fat atrophy using the Goutallier classification. T2 MRI axial slices from the disc space at the operative level were analyzed. We graded fat atrophy on a 1-4 scale and the LIV was calculated using a published technique of measuring the distance from the tip of a spinous process and a line across the muscular fascia. We used a standardized protocol of measuring cross sectional area of the paralumbar muscle (PL-CSA). We scaled the PL-CSA by finding the ratio of PL-CSA/BMI. HRQOL scores collected included VAS leg, VAS back, ODI, SF-12 mental and physical health and PROMIS. We performed a linear regression analysis to determine the relationship of LIV, PL-CSA, PL-CSA/BMI and the HRQOLs. We performed an ANOVA analysis to identify the relationship between Goutallier classification and HRQOLs listed. We combined our measurements to create a score to quantify muscle health and determined whether this score correlated with HRQOLs based on an ANOVA analysis. RESULTS A total of 92 patients were included within our analysis. The average age was 57.9+/-14.4 years old (49 men and 43 women). There were 104 levels operated on within this cohort. The most common preoperative diagnosis was lumbar spinal stenosis (58 patients). We found that the average LIV, PL-CSA and PL-CSA/BMI was 16.1 +/-7.5 mm, 4004 +/-1210 mm^2 and 153.5 +/- 45.0 mm^2/BMI. The PL-CSA/BMI ratio significantly correlated with preop SF-12 PHS (p = 0.03), VAS back (0.007) and VAS Leg (p = 0.002). Patients with less than 130 of the PL-CSA/BMI ratio had statistically significant worse PROMIS (35.9 vs 29.7, p=0.007), ODI (39.4 vs 50.2, p=0.01), SF-12 PHS (35.5 vs 28.7, =0.001), VAS leg (7.3 vs 5.5, p=0.007) and VAS back (7.9 vs 4.9, p=0.002) scores. We combined our results and scored each patient from 1-3 based upon whether there LIV 10mm (0 or 1), Goutallier Classification >2 or 130 or CONCLUSIONS Muscle health contributes significantly to preoperative disability. When patients have combined low PL-CSA/BMI, goutallier classification and LIV this corresponds to significantly worse HRQOLs. This lumbar muscle health score is a valid screen for patients with poor muscle health which may be contributing to their disability. FDA DEVICE/DRUG STATUS This abstract does not discuss or include any applicable devices or drugs.