Long-term cardiovascular outcomes in a population-based multicentric cohort of northern Portugal: Validation of the ESC/EAS prognostic risk classification

Autores da FMUP

• Cristina Maria Rodrigues Pinheiro Gavina

  Autor

• Tiago Salgado De Magalhães Taveira Gomes

  Autor

Participantes de fora da FMUP

• Carvalho, Daniel Seabra
• Afonso-Silva, Marta
• Abreu, Daniela Brandao
• Canelas-Pais, Mariana
• Araujo, Francisco

Unidades de investigação

• Medicina da Comunidade, Informação e Decisão em Saúde

Abstract

BACKGROUND: Cardiovascular (CV) risk scores identify individuals at higher long-term risk of CV events that may benefit from more aggressive preventive interventions. OBJECTIVE: To assess the association of CV-risk categories and criteria with long-term CV events. METHODS: Observational cohort study between 2000-2019 on patients aged 40-80 years, followed by 14 primary care centers assisted by 1 hospital in Portugal. Follow-up began when electronic health records data allowed for CV-risk classification and dynamic reassessment per 2019 ESC/EAS Guidelines. Inclusion criteria required at least one appointment with a primary care physician within three years before follow-up initiation. We assessed the 10-year adjusted hazard-ratio of combined CV death and non-fatal atherosclerotic cardiovascular disease (ASCVD) hospitalization, across SCORE risk categories and criteria, using Cox proportional hazards models adjusted for sex, age, competing comorbidities, and medication. RESULTS: The study included 161 681 observations from 87 035 unique patients. During the observation period, 71 787 patients were classified as low/moderate, 51 476 as high and 38 418 as very-high CV-risk categories. In the very-high group, prevalent comorbidities were hypertension (69%), hypercholesterolemia (69%) and type 2 diabetes (61%), and 13% were hospitalized for ASCVD. The adjusted 10-year hazard ratio of the composite of CV death or ASCVD hospitalization was 2.10 (95% CI: 1.91- 2.32) for high-risk and 3.56 (95% CI: 3.21-3.96) for very-high-risk patients (low-risk as reference). CONCLUSION: Our study reinforces the prognostic relevance of CV-risk stratification for long-term prediction of CV death and ASCVD hospitalization in an unselected cohort, independently of sex, age, competing comorbidities and medication. (c) 2024 National Lipid Association. Published by Elsevier Inc. All rights are reserved, including those for text and data mining, AI training, and similar technologies.

Keywords

• Atherosclerosis; Hypolipidemic agents; Cholesterol; LDL; Medical records systems; Computerized; Cardiovascular risk