Screening Tool to Identify Patients with Advanced Aortic Valve Stenosis

Area of Science:

• Cardiology
• Public Health
• Medical Informatics

Background:

• Aortic stenosis (AS) poses a significant clinical challenge in Western countries.
• Current clinical practice lacks effective screening algorithms for AS.
• Developing a predictive model is crucial for early detection and management.

Purpose of the Study:

• To develop and validate a risk prediction model for moderate to severe aortic stenosis (AS).
• To identify key demographic and clinical variables associated with AS risk.
• To facilitate targeted screening strategies for AS.

Main Methods:

• Cross-sectional analysis of echocardiographic reports from 2013-2018.
• Inclusion of unique patients aged 40-95 years.
• Logistic regression modeling using demographics and comorbidities; C-statistic and calibration plot for performance assessment.

Main Results:

• Analysis of 38,788 reports identified 10.8% of patients with ≥moderate AS.
• Key predictors included age, male gender, Caucasian race, obesity (BMI ≥ 30), and cardiovascular factors.
• The model achieved a C-statistic of 0.77 and demonstrated good calibration, with a risk score range from 0.0002 to 0.7711.

Conclusions:

• A validated risk prediction model for moderate to severe AS was developed using a large cohort.
• The model incorporates readily available demographic and clinical data.
• This tool has the potential to guide targeted screening for advanced AS in the general population.