Abstract / Summary
Mitral annular calcification (MAC) has been associated with coronary artery disease in population-based studies, but its relationship with preoperative coronary artery stenosis (CAS) severity, particularly in elderly orthopedic fracture patients, remains unclear. This study aimed to evaluate the association between MAC and CAS severity, and to determine whether MAC provides incremental discriminatory value beyond traditional risk factors. We retrospectively analyzed 191 consecutive orthopedic fracture patients who underwent both transthoracic echocardiography for MAC assessment and coronary computed tomography angiography (CCTA) for preoperative cardiac risk stratification between December 1, 2022, and December 31, 2022. When CCTA revealed a stenosis judged to require intervention (typically > 70% diameter reduction, or > 50% with high-risk features), the patient was referred for invasive coronary angiography (ICA). CAS was graded on an ordinal scale (0–IV) and also dichotomized at three cutoffs: any stenosis (grade ≥ I), obstructive stenosis (grade ≥ III), and occlusive disease (grade IV). Ordinal logistic regression was used to assess the association between MAC and CAS grade, with progressive adjustment for covariates across four model specifications: Model A (age, sex, hypertension), Model B (Model A + diabetes, hyperlipidemia, aortic valve calcification), Model C (12 exploratory variables), and a modified Revised Cardiac Risk Index (RCRI) model (age, sex, MAC, exploratory modified RCRI-like score). Model A served as the primary analysis, while Models B, C, and the modified RCRI model were conducted as sensitivity analyses to assess the robustness of the association. Receiver operating characteristic (ROC) analysis was used to quantify the incremental discriminatory value of MAC beyond traditional risk factors. MAC was present in 77 patients (40.3%), with prevalence increasing from 12.0% in Grade 0 to 95.8% in Grade IV. In ordinal logistic regression, MAC was the factor most strongly associated with higher stenosis grade across all models: Model A OR 12.56 (95% CI 6.11–25.84, p < 0.001) and modified RCRI model OR 11.73 (95% CI 5.69–24.17, p < 0.001). Consistent results were obtained in exploratory sensitivity analyses (Models B and C). For the binary outcome of obstructive stenosis (Grade ≥ III), adding MAC to the baseline clinical model (age, sex, hypertension) improved the AUC from 0.73 (95% CI 0.64–0.81) to 0.86 (95% CI 0.80–0.92), Δ=+0.13 (DeLong test with bootstrap, p = 0.001). The modified RCRI-like model achieved an AUC of 0.88 (95% CI 0.82–0.93). The MAC-CAS association persisted in the CCTA-only subgroup ( n = 152, OR 4.35–6.32, all p < 0.001), suggesting that the association is not entirely driven by referral to invasive angiography, although referral selection bias for CCTA remains. Factors independently associated with MAC included aortic valve calcification (OR 7.79), hypertension (OR 3.27), male (OR 2.49) and age (OR 1.04 per year). MAC demonstrated a strong and independent association with CAS severity in this selected cohort of elderly orthopedic fracture patients. A simplified four-variable model combining MAC and the exploratory modified RCRI-like score achieved an AUC of 0.88. However, these findings are hypothesis-generating and do not establish MAC as a clinical prediction tool. Given that echocardiography is routinely performed in this population, MAC assessment requires no additional imaging or cost. Importantly, these findings are derived from a highly selected cohort—only 12% of screened fracture patients were included—and should be considered specific to patients referred for advanced cardiovascular imaging. The conclusions do not apply to the broader population of patients undergoing urgent orthopedic surgery. Prospective multi-center studies in unselected fracture populations with hard clinical endpoints are warranted to validate these findings.