Studies on register-based family history of cardiovascular disease : from preclinical to recurrent disease

Background

Family history of coronary heart disease (CHD) is a known risk-factor for incident atherosclerotic cardiovascular disease (ASCVD). The underlying mechanisms are, however, poorly understood and the risk effects are seemingly independent of polygenic risk scores for ASCVD. Furthermore, the evidence of family history of CHD as a risk factor in specific settings, such as in the diagnostics of acute ASCVD events and in secondary prevention is conflicting, which may be partly due to misreporting of family history in studies relying on self-reported data.

Methods and results

In order to find registered parents and siblings of index subjects, the index population of each study was linked to the Swedish Multi-Generation Register (MGR). Information on registered diagnoses and causes of death in relatives, including myocardial infarction (MI), any coronary revascularization procedure, and in some cases ischemic stroke, was retrieved from the Swedish National Patient register and the Cause of Death Register, along with the age of diagnosis. This data was consequently used to construct different definitions of register- based family history of CHD and ASCVD.

Study I. The first consecutive visit of 28,188 adult patients seeking any of four emergency departments in Stockholm due to chest pain between 2013 and 2016 were included. Patients with no or only one parent identifiable in MGR were excluded. In multivariable logistic regression, a family history of early-onset CHD was associated with major adverse cardiac events, including MI and unstable angina, as well as cardiovascular deaths for those discharged from the emergency department, independently of traditional cardiovascular risk factors. The fully adjusted odds ratio for major adverse cardiac events was 1.62 (95% confidence interval [CI] 1.35 - 1.94) and family history had a larger effect on the outcome in subjects <65 years old and with low initial high-sensitivity cardiac troponin T values, with significant statistical interactions.

Study II. In total, 25,615 patients aged 18-76 years that attended the 1-year follow-up revisit after their first MI between 2005 and 2013 in the Swedish Web- system for Enhancement and Development of Evidence-based care in Heart disease Evaluated According to Recommended Therapies (SWEDEHEART) were included. Patients with no or only one parent identifiable in MGR were excluded. The hazard ratio (HR) for recurrent ASCVD (rASCVD) comparing patients with a family history of ASCVD to those without was estimated using Cox proportional- hazards regression, adjusted for cardiovascular risk factors and socioeconomic determinants. A family history of early-onset ASCVD was associated with an increased risk of rASCVD, with a fully adjusted HR of 1.22 (95% CI 1.05 - 1.42) during a median follow-up of 7.2 years. The addition of family history of early- onset ASCVD to a previously validated risk-model for rASCVD improved risk prediction significantly, albeit with small absolute increments.

Study III. A total of 4,251 subjects free of known CHD, from the population-based Swedish Cardiopulmonary bioImage Study (SCAPIS), that had undergone coronary computed tomography angiography and proteomics analysis and with identifiable relatives in MGR were included. In the main analysis, 38 plasma proteins were associated with a family history of early-onset CHD using a false discovery rate of 0.05, whereas 27 remained significant after adjusting for metabolic risk factors. Follistatin (FST) was positively associated with family history and only slightly attenuated when adjusting for metabolic risk factors. Furthermore, 79 proteins were associated with the coronary atherosclerotic burden and differed across subgroups of family history status. In interaction analysis, 18 proteins showed steeper associations with the coronary atherosclerotic burden in subjects with family history as compared to those without such history. In two-sample Mendelian randomization, there was evidence of causal associations between three proteins and CHD; the proprotein convertase subtilisin/kexin type 9 (PCSK9) enzyme, platelet endothelial cell adhesion molecule 1 (PECAM-1) and FST.

Study IV. We included 25,302 subjects from the SCAPIS cohort with both parents identifiable in MGR. The SCAPIS core questionnaire on family history of MI in parents and siblings, with and without information on age of onset, was compared to registered diagnoses and deaths due to MI in relatives from registers. The agreement for self-reported family history of MI in any parent or sibling as compared to register-based history was modest, with a Cohen's K = 0.491, 95% CI 0.480 - 0.503, a sensitivity of 57.6%, and specificity of 89.0%. Agreement measures for maternal and paternal history were similar, however, early-onset disease was generally less accurately reported. In subgroup analyses, female subjects consistently reported family history more accurately than male counterparts and having a university degree or a high income was indicative of slightly more accurate reporting.

Conclusions

A family history of CHD and ASCVD is an important cardiovascular risk factor in both diagnostic and secondary preventive settings. Family history of particularly early-onset CHD is an independent risk factor for major adverse cardiac events in chest pain patients seeking emergency care, especially in those with otherwise low absolute risk, such as younger patients and those with low initial cardiac troponin values. Family history of early-onset ASCVD also increases the risk for disease recurrence, independently of traditional risk factors, and improves risk prediction in this setting. Furthermore, we present evidence of specific biological processes in familial CHD, as a number of circulating plasma proteins were associated with family history, independently of traditional risk factors. Furthermore, there was evidence of modifying effects of family history on proteins associated with the coronary atherosclerotic burden, indicating distinct effects of these proteins in subjects with family history. We found potentially causal associations for three proteins and CHD, of which FST is a novel marker. Lastly, self-reported family history only to a lesser extent captures registered diagnoses of MI in relatives, demonstrating that the use of register- based family history in research is more robust than self-reports. This enhanced understanding of family history as a risk factor for ASCVD could improve targeted therapies, as well as risk stratification for early diagnostics and treatments.

List of scientific papers

I. Family history of coronary artery disease is associated with acute coronary syndrome in 28,188 chest pain patients Agnes Wahrenberg, Patrik K.E. Magnusson, Andrea Discacciati, Lina Ljung, Tomas Jernberg, Mats Frick, Rickard Linder, Per Svensson European Heart Journal: Acute Cardiovascular Care. 2020; 9(7): 741-747. https://doi.org/10.1177/2048872619853521

II. Cardiovascular family history increases the risk of disease recurrence after a first myocardial infarction Agnes Wahrenberg, Ralf Kuja-Halkola, Patrik K.E. Magnusson, Henrike Habel, Anna Warnqvist, Kristina Hambraeus, Tomas Jernberg, Per Svensson Journal of the American Heart Association. 2021; 10(23): e022264. https://doi.org/10.1161/jaha.121.022264

III. Plasma protein profile associated with a family history of early- onset coronary heart disease Agnes Wahrenberg, Lars Lind, Natan Åberg, Henrike Häbel, Marika Ström, Anders Mälarstig, Patrik K.E. Magnusson, Ralf Kuja-Halkola, Göran Bergström, Gunnar Engström, Emil Hagström, Tomas Jernberg, Stefan Söderberg, Carl Johan Östgren, Per Svensson [Submitted]

IV. Validation of self-reported family history of myocardial infarction using nationwide health care data Agnes Wahrenberg, Karin Leander, Henrike Häbel, Patrik K.E. Magnusson, Ralf Kuja-Halkola, Göran Bergström, Lars Lind, Emil Hagström, Gunnar Engström, Tomas Jernberg, Stefan Söderberg, Carl Johan Östgren, Per Svensson [Manuscript]