Abstract Body: Background
Atherosclerotic carotid artery disease is associated with a high risk of major adverse cardiac events (MACE) even with optimal risk factor management and surgical interventions such as carotid endarterectomy (CEA). Aortic stenosis (AS) is an age-related valve disease that is associated with a high rate of MACE. An artificial intelligence-based AS estimation algorithm (AI-AS) can identify AS with high accuracy using electrocardiogram. This study evaluates the prognostic value of AI-AS in patients with carotid artery disease undergoing CEA.
Methods
We included patients with carotid artery disease undergoing CEA at Mayo Clinic from 2002-2023 if they had >30 days of follow-up and an electrocardiogram within one year before CEA. The electrocardiogram closest to the CEA date was used to calculate AI-AS (AS probability ranging from 0-1). MACE was defined as the composite of cardiac events, ischemic cerebrovascular accidents, and all-cause mortality during follow-up after CEA. A receiver operating characteristic curve and the Youden index were used to identify the optimal AI-AS cut-off value for predicting MACE. Kaplan Meier plot was used to depict MACE-free survival. Regression models were used to evaluate the predictive value of AI-AS for future MACE adjusting for clinical covariates.
Results
A total of 665 patients with a median age of 72 [IQR:66, 78] years and 438 (65.9%) males were included. Over a median follow-up time of 7.1 [IQR:2.7, 11.5] years, MACE occurred in 419 (63%) of the patients, including 187 (44.6%) mortality, 161 (38.4%) cardiac events, and 71 (16.9%) cerebrovascular accidents. Median AI-AS was 0.32 [IQR:0.09, 0.55] in the total cohort, 0.34 [IQR:0.12, 0.56] in patients with MACE, and 0.30 [IQR:0.07, 0.53] in patients without MACE (P=0.038). AI-AS was a significant predictor of MACE (HR:2.38, 95%CI:1.64-3.46, P<0.001) in univariate Cox regression analysis. The optimal cut-off value of AI-AS for MACE was 0.419 (area under the curve: 0.55). Patients with high AI-AS (≥0.419) had a significantly lower MACE-free survival in the Log-rank test (Figure 1) and in the multivariable Cox regression model adjusted for covariates compared to low AI-AS (<0.419) patients (Table 1).
Conclusion
AI-AS is an independent predictor of MACE in patients with carotid artery disease after CEA and can be used as a risk-stratifying tool in this patient population. Additional studies are needed to elucidate the mechanistic link between AS and MACE in these patients.