Predictive value of coronary artery calcification for osteoporosis in COPD patients: a retrospective AI-based study.
Authors
Affiliations (4)
Affiliations (4)
- Department of Radiology, Affiliated Hospital of Putian University, Putian, Fujian, China.
- College of Biomedical Engineering and Instrument Science, Zhejiang University, Hangzhou, China.
- Department of Medical Ultrasonics, Fujian Maternity and Child Health Hospital, College of Clinical Medicine for Obstetrics and Gynecology and Pediatrics, Fujian Medical University, Fuzhou, China.
- Department of Pediatric Critical Care Medicine, Affiliated Hospital of Putian University, Putian, Fujian, China.
Abstract
Patients with chronic obstructive pulmonary disease (COPD) are at high risk for osteoporosis (OP), yet OP screening remains suboptimal in practice. The "bone-vascular axis" hypothesis suggests shared mechanisms between vascular calcification and bone loss. Coronary artery calcification (CAC), a reliable imaging marker of atherosclerosis, may serve as a surrogate indicator of OP risk in this population. Advances in artificial intelligence (AI) now enable automated CAC quantification from routine chest CT scans, offering a potentially efficient, low-cost strategy for opportunistic OP risk assessment in COPD patients. This study aimed to evaluate the predictive value of artificial intelligence (AI)-based CAC scoring, derived from routine chest computed tomography (CT) scans, for identifying osteoporosis in patients with COPD. The retrospective study included COPD patients who underwent dual-energy X-ray absorptiometry (DXA) and non-gated chest CT between 2020 and 2025. Osteoporosis was defined as a T-score ≤ -2.5. An AI tool automatically quantified Agatston scores for all four coronary arteries. Statistical analyses involved intergroup comparisons and adjusted logistic regression, with ROC curves assessing predictive performance. Amongst the 262 patients, 117 were diagnosed with osteoporosis. The osteoporosis group was characterized by older age, lower BMI, and greater coronary artery calcification scores in all arteries. Following adjustment for confounding factors including age, sex, and BMI, left anterior descending coronary artery calcification (LAD-CAC; OR = 1.01, 95% CI: 1.00-1.02, <i>P</i> = 0.036) and left circumflex coronary artery calcification (LCX-CAC; OR = 1.11, 95% CI: 1.02-1.20, <i>P</i> = 0.021) were identified as independent predictors of osteoporosis. The area under the curve (AUC) values were 0.703 (sensitivity 51.3%, specificity 87.6%) and 0.625 (sensitivity 31.6%, specificity 93.1%), respectively. Coronary artery calcium scores, particularly LAD-CAC and LCX-CAC, automatically obtained via AI software from routine chest CT scans, are independently associated with osteoporosis in COPD patients and demonstrate moderate predictive capability. This finding suggests a predictive utility for early, convenient, and cost-free screening of OP risk by leveraging routine imaging examinations already performed in COPD patients.