Key Points:
- Women are underdiagnosed and undertreated for CAD, despite higher prevalence of non-obstructive CAD, leading to worse outcomes
- Using AI-enhanced assessment of coronary artery plaque characteristics, investigators suggest better evaluation of CAD risk compared to traditional risk scores
- The CONFIRM2 registry study analyzed symptomatic patients with no prior CAD to assess for evidence of CAD with plaque characterization
- Despite a higher prevalence of plaque in men, women demonstrated incrementally higher rates of elevated risk plaque characteristics, which the study investigators suggest should be considered in future guidelines
Women suffer from underdiagnosis and undertreatment of coronary artery disease (CAD). They often have a distinct clinical presentation, a higher prevalence of non-obstructive disease, and suffer from adverse outcomes at higher rates. Traditional clinical risk calculators have low accuracy and suffer from poorer performance in female patients. Sex-specific differences in atherosclerotic plaque profile by CT have been reported, but data on association with CVD risk are limited. Here, artificial intelligence-based quantitative CT (AI-QCT) may have a role in CV risk prediction, enabling evaluation of features like total plaque volume (TPV), calcified plaque (CP), non-calcified plaque (NCP), high-risk plaque (HRP), and more.
The study aimed to define sex-specific patterns of atherosclerosis by AI-QCT features for prediction of MACE compared to clinical risk scores in a mulicenter international study registry across 18 sites in 11 countries: the CONFIRM2 registry (Coronary CT Angiography Evaluation for Evaluation of Clinical Outcomes). Included patients were symptomatic with clinical indication for CCTA with no prior history of CAD with access to CCTA of at least 64-detector row. The study evaluated plaque based on TPA, NCP volume < 350 HU, CP, low attenuation plaque (LAP) volume, HRP, and percentage atheroma volume (%PAV), as well as stenosis severity. Patients were followed on average for 4.8 years. 4163 patients were screened, resulting in 3551 patients ultimately after excluding asymptomatic patients, prior CAD patients, or those without clinical information. Five percent of patients experienced major adverse cardiovascular events during follow up time. Patients were well balanced with equal proportion of men and women, similar BMI, and similar proportion of ASCVD risk factors, but of note, more women enrolled endorsed atypical symptoms of angina (39.5% vs 35.7%; p = 0.0195).
Results demonstrated higher proportion of non-obstructive disease among women vs men (79.7% vs 71.6%; p<0.001). In addition, using a multivariate Cox regression analysis adjusted for age and cardiovascular risk factors, despite a higher AI-QCT CAD burden in men, similar increments in AI-QCT-derived plaque features (namely TPV, NCP, CP, and PAV) conferred a higher relative risk for MACEs in women versus men, whereas clinical risk score performance was moderate to poor.
Given these findings, the investigators suggest using AI-QCT CAD risk stratification with CCTA versus traditional risk scores for individualized risk assessment. Furthermore, they advocate for reinforcement of ASCVD therapy and prevention, and they raise a consideration of sex-specific treatment guidelines given differential findings for men versus women in this study.