Abstract 48: Long-Term Patterns in the Development of Multiple Cardiovascular Diseases: The Framingham Heart Study

Abstract

Background: Having a first CVD event increases the risk for subsequent events; however, the long-term patterns in the development and sequence of multiple CVDs, including stroke, myocardial infarction (MI) and chronic heart failure (CHF) are unknown. The aim of this study was to identify distinct long-term patterns in the order and timing of MI, stroke and CHF occurrence. Methods: We used publicly available data from the Framingham Heart Study (FHS). The occurrence of fatal/non-fatal MI, stroke and CHF were examined separately using discrete mixture modeling implemented in SAS (Proc Traj) to identify trajectory groups for the risk of each CVD event starting at age 30. We included both first and subsequent events. Clusters of disease specific trajectory groups were examined. Baseline demographics and risk factors were compared across clusters. Results: Among 5,079 participants (ppts) in FHS, we identified 8 unique patterns in the development of CVDs (see figure). The majority, 72.5%, of Framingham ppts experienced average age- related increases in the yearly risk for all three endpoints of MI, stroke and heart failure (Average Risk group) other groups experienced early or higher risks for specific CVDs including the High CHF Risk group (6% of ppts), Early CHF group (2%), High Stroke Risk group (9%), High CHF and Stroke Risk group (1%), High MI Risk group (4%), High MI and CHF Risk group (1%), Early CHF and MI Risk group (1%). Groups in which stroke and/or HF risk was elevated had higher prevalence of smoking and greater baseline BP levels; those groups at elevated risk of MI had higher total cholesterol levels and higher BMI. Conclusions: We identified distinct patterns in development and ordering of MI, stroke and CHF associated with diverse risk factor profiles. By understanding the life-course and likely sequence of CVD events related to distinct risk factor profiles, clinicians may be able to consider personalized prevention strategies with the highest likelihood of preventing first CVD events and reducing the overall burden of CVD.