D30-05 CT-Measured Pectoralis Muscle Loss Is an Extrapulmonary Biomarker Associated With Longitudinal COPD Morbidity

Abstract Rationale Abnormal and progressive muscle loss is prevalent in smokers and people with chronic obstructive pulmonary disease (COPD). Computed tomography (CT)-derived pectoralis muscle volume (PMV) is cross-sectionally associated with spirometry, exercise capacity, and respiratory-related symptom burden in people at risk-for and with COPD. However, it is unclear whether pectoralis muscle loss is associated with longitudinal COPD morbidity. Therefore, the objective of this study was to determine whether the baseline or rate of PMV loss is associated with emphysema progression, lung function decline, exercise capacity decline, and symptom burden worsening. Methods Data from ever-smokers without COPD and individuals with COPD (forced expiratory volume in 1-second-to-forced vital capacity ratio (FEV1/FVC)0.70) from the Canadian Cohort Obstructive Lung Disease (CanCOLD) study who obtained inspiratory chest CT scans at the baseline and 3-year follow-up visit was analyzed. Future changes to COPD morbidity were evaluated using annualized differences in volume-adjusted lung density (ΔV-LD15; g/L/year), FEV1 (ΔFEV1; mL/year), peak rate of oxygen consumption during cardiopulmonary exercise testing (ΔV’O2,peak; mL/kg/min/year), and St. George’s respiratory questionnaire scores (ΔSGRQ; N/year) between follow-up and baseline. The complete pectoralis muscle was automatically segmented from all available chest CT scans using a previously developed deep-learning model, and the PMV was defined as the number of segmented voxels (in cm3) in the standard muscle Hounsfield unit range of -29, 150. Longitudinal PMV measurements were extracted by co-registering and simultaneously cropping pectoralis segmentations to a matched field-of-view, before calculating their annualized difference (ΔPMV; cm3/year). Associations between baseline PMV with changes in morbidity indices were evaluated using multivariable linear regression models adjusted for baseline age, sex, height, BMI, obesity, smoking-status, pack-years, and center ID. ΔPMV models were also adjusted for baseline FEV1, baseline PMV, and the baseline outcome. Results Among 473 individuals included in this study, the mean±SD age was 65.7±9.5yrs, 190 (40.2%) were female, and 302 (63.8%) had COPD. Smaller baseline PMV was not associated with ΔV-LD15, ΔFEV1, or ΔV’O2,peak (p 0.05), but was associated with greater symptom burden as evaluated by SGRQ (ΔSGRQ: β (95% CI) = +0.027 (0, 0.053) per -10cm3; p = 0.047). Meanwhile, ΔPMV was not associated with ΔFEV1, ΔV’O2,peak, or ΔSGRQ (p 0.05), but was associated with lung density decrease (ΔV-LD15: β (95% CI) = -0.235g/L/year (-0.403, -0.067) per -10cm3/year; p = 0.006). Conclusions CT-derived pectoralis muscle volume measurements showed prognostic utility as an extrapulmonary imaging biomarker that could potentially identify ever-smokers at risk of COPD and people with COPD susceptible to accelerated longitudinal COPD morbidity. This abstract is funded by: D. Genkin acknowledges salary support from the Natural Sciences and Engineering Research Council (NSERC). D. Jensen holds a Canada Research Chair in Clinical Exercise & Respiratory Physiology (Tier 2) from the Canadian Institutes of Health Research. M. Kirby acknowledges support from NSERC and the Canada Research Chair Program (Tier 2).