Background: Circulating biomarkers may improve prediction of cardiovascular (CV) events in the general population, beyond traditional risk factors.
Aim: To investigate if circulating N-terminal pro-brain natriuretic peptide (NT-proBNP), high sensitivity troponin I (hsTNI) and microRNAs (miRNAs) could improve risk prediction among asymptomatic individuals.
Methods: Participants with no history of CV disease were recruited from the general population (n=2940, age 64±11 years, 67% male). Fatal/non-fatal CV events (myocardial infarction (MI), unstable angina, ischemic stroke, transient ischemic attack, heart failure) were recorded over a median 8 years follow-up. Associations between known risk factors (age, gender, ethnicity, cigarette smoking, body mass index, systolic blood pressure, diabetes, family history of MI, blood pressure medications), plasma NT-proBNP, hsTNI and CV events were tested with Cox regression and log-rank tests. In an exploratory substudy, associations between CV events and plasma levels of 3 miRNAs previously associated with CV events <1 year (miR-223, miR-146a, miR-127) were tested in 56 participants who had a CV event and 39 event-free controls (age- and gender-matched, median follow-up 6 years).
Results: A total of 287 (13%) participants experienced a CV event during follow-up. For all participants, higher levels of hsTNI and NT-proBNP were strongly associated with CV events (event rates for tertiles: hsTNI 4%, 13%, 22%; NT-proBNP 8%, 14%, 25%; p<0.001; Fig 1a,b). These associations were independent of known risk factors (hsTNI p<0.001; NT-proBNP p=0.008). In the miRNA substudy, higher levels of miR-223 were associated with CV events (event rate for tertiles: 37%, 67%, 68%, p=0.021, Fig 1c), but this was not independent of NT-proBNP (p=0.866).
Conclusions: Plasma hsTNI, NT-proBNP, and miR-223 are associated with increased CV risk in healthy participants, but only hsTNI and NT-proBNP provide independent risk prediction.