Thoracic CT-derived skeletal muscle volume discriminated DXA-defined sarcopenia with an AUC of 0.795 (95% CI 0.69-0.90) in patients with pulmonary arterial hypertension.
Cohort (n=80)
Does chest CT-derived skeletal muscle volume accurately diagnose sarcopenia compared to DXA in patients with PAH?
Thoracic skeletal muscle tissue volume measured on routine chest CT performs similarly to DXA for identifying low muscle mass in patients with pulmonary arterial hypertension.
Effect estimate: AUC 0.795 (95% CI 0.69-0.90)
Abstract Rationale Sarcopenia is defined by low skeletal muscle mass and strength and is associated with increased risk of frailty, hospitalization, and mortality. Guidelines recommend whole body dual-energy-X-ray absorptiometry (DXA) derived muscle mass and physical performance testing for diagnosis, although DXA is not standard in patients with pulmonary arterial hypertension (PAH). Chest computed tomography (CT) imaging is routinely obtained and could offer a novel diagnostic tool. We sought to validate CT-derived skeletal muscle volume indexed to height2 (MVI) against DXA-defined muscle mass criteria for sarcopenia. Methods In a prospective cohort of patients with PAH, we obtained standardized research non-contrast chest CTs at baseline and 1 year as well as whole body-DXA. We compared the performance of CT-derived MVI (estimated from skeletal muscle tissue volume within the thoracic body region in cm3/m2) to DXA-derived total appendicular lean muscle mass indexed to height2 (ALMMI). Low muscle mass was defined using DXA derived EWGSOP2 criteria (ALMMI 7 kg/m2 in males, 5.5 kg/m2 in females). We used a linear regression model to predict ALMMI from thoracic CT-derived MVI and sex. We assessed the performance of our derived measures using receiver operating characteristic curve analysis. Results The study population consisted of 80 patients with PAH, predominantly female (82.5%). The most common ethnicity was non-Hispanic white (63.8%), and the most common etiology of PAH was connective tissue disease (32.5%) followed by idiopathic PAH. Twenty-six percent of the cohort had low muscle mass by DXA. Compared to those with normal muscle mass, low muscle mass patients were older (median age 66.3 vs 55.4 years, p = 0.013), had smaller waist (35.0 vs 39.5 inches, p 0.001) and hip circumferences (37.0 vs 42.0 inches, p 0.001). Additionally, sarcopenic patients had decreased fat mass indexed to height2 (8.10 vs 12.80 , p 0.001) and decreased thoracic muscle volume (774.3 vs 999.9 c 3, p = 0.006) when compared to those with normal muscle mass. CT-MVI and DXA-ALMMI were strongly positively correlated (r = 0.57, 95%CI 0.40-0.71, p 0.001). The AUC for the discrimination of DXA-derived sarcopenia by CT-measured appendicular lean mass measures was 0.795 (95% CI 0.69-0.90, figure). Conclusions Thoracic skeletal muscle tissue volume on clinically obtained CT performs similarly to DXA in identifying low muscle mass in patients with PAH. Chest CT imaging is routinely performed in PAH, potentially allowing for increased diagnosis of sarcopenia at no additional cost, ionizing radiation exposure, or patient inconvenience. Further validation is required in larger cohorts. This abstract is funded by: K23 HL141584, R01 HL173533
Farley et al. (Fri,) conducted a cohort in Pulmonary Arterial Hypertension (PAH) (n=80). Chest CT-derived skeletal muscle volume indexed to height2 (MVI) vs. DXA-derived total appendicular lean muscle mass indexed to height2 (ALMMI) was evaluated on Discrimination of DXA-derived sarcopenia (AUC 0.795, 95% CI 0.69-0.90). Thoracic CT-derived skeletal muscle volume discriminated DXA-defined sarcopenia with an AUC of 0.795 (95% CI 0.69-0.90) in patients with pulmonary arterial hypertension.