Risk prediction models for heart failure admissions in adults with congenital heart disease

Area of Science:

• Cardiology
• Public Health
• Medical Informatics

Background:

• Heart failure (HF) is a leading cause of mortality in adult patients with congenital heart disease (ACHD).
• Existing risk prediction models do not specifically address HF hospitalization (HFH) in the ACHD population.
• There is a critical need for a clinically relevant tool to identify ACHD patients at elevated risk for HFH.

Purpose of the Study:

• To develop a one-year risk prediction system for HFH in ACHD patients.
• To identify key clinical predictors of incident HFH within this population.
• To create a practical risk score for stratifying HFH risk in ACHD.

Main Methods:

• A retrospective cohort of 29,991 ACHD patients (aged 18-64) was analyzed using data from the Quebec CHD Database (1995-2010).
• A nested case-control sample was used to identify one-year HFH predictors via multivariable logistic regression.
• A risk score was developed based on adjusted odds ratios of significant predictors.

Main Results:

• The cumulative risk of HFH by age 65 was 12.58% in the ACHD cohort.
• Significant one-year HFH predictors included age ≥ 50, male sex, CHD lesion severity, recent HFH history, pulmonary arterial hypertension, chronic kidney disease, coronary artery disease, systemic arterial hypertension, and diabetes mellitus.
• The developed risk score (0-19) showed excellent prediction performance, with rapidly increasing HFH risk beyond a score of 8.

Conclusions:

• Approximately one-eighth of the ACHD population experiences HFH before age 65.
• A risk prediction model incorporating age, sex, CHD lesion severity, recent HFH history, and comorbidities effectively identifies ACHD patients at high risk for HFH.
• The developed risk score offers a valuable tool for clinical risk stratification and management of ACHD patients.