Lipoprotein(a) level and ischemic cardiovascular risk assessment using artificial intelligence-based portable fundus camera in thai patients with coronary artery disease (cad) who underwent cardiac catheterization
Abstract:
Background: Lipoprotein(a), or Lp(a), constitutes a noteworthy risk factor contributing to cardiovascular disease, with its levels exhibiting significant variations among different ethnicities. Emerging data indicate that fundus cameras utilizing artificial intelligence (AI) technology have the capacity to evaluate cardiovascular risk factors. Currently, limited data exists regarding Lp(a) levels and the assessment of cardiovascular risk utilizing artificial intelligence (AI)-based fundus cameras in Thai patients diagnosed with coronary artery disease (CAD)Methods: Data were procured from individuals admitted to King Chulalongkorn Memorial Hospital for cardiac catheterization between January and April 2024. Participants were stratified into two groups based on cardiac catheterization results: CAD group and non-CAD group. Blood samples were collected for Lp(a) level assessment, and all participants underwent fundus examinations. The primary objective was to discern the Lp(a) level difference between individuals with and without confirmed coronary artery disease through cardiac catheterization. The secondary objective aimed to determine the cutoff value, utilizing a portable AI-based fundus camera, capable of predicting coronary artery disease.Results: Among 120 patients examined, the non-coronary artery disease (CAD) group (n = 40) exhibited a median (IQR) Lp(a) level of 16.5 (8.5, 33.7) nmol/L, while the CAD group (n = 80) showed a higher level of 30.3 (15.5, 88) nmol/L. A Wilcoxon signed-rank test resulted in a p-value < 0.01, indicating a statistically significant difference between the two groups. Notably, Lp(a) levels exceeding 125 nmol/L, a threshold outlined in the 2022 EAS guidelines to indicate cardiovascular risk, were observed in 13 CAD patients and none in the non-CAD group. For AI-based fundus camera risk assessment, the identified cutoff value associated with the most effective prediction of cardiovascular risk is set at