-
Views
-
Cite
Cite
Magnus N Lyngbakken, Helge Røsjø, Oddgeir L Holmen, Håvard Dalen, Kristian Hveem, Torbjørn Omland, Temporal Changes in Cardiac Troponin I Are Associated with Risk of Cardiovascular Events in the General Population: The Nord-Trøndelag Health Study, Clinical Chemistry, Volume 65, Issue 7, 1 July 2019, Pages 871–881, https://doi.org/10.1373/clinchem.2018.301069
Close - Share Icon Share
Abstract
Cardiac troponins are associated with cardiovascular risk in the general population, but whether temporal changes in cardiac troponin I provide independent prognostic information remains uncertain. Using a large community-based cohort with follow-up close to the present day, we aimed to investigate the associations between temporal changes in cardiac troponin and cardiovascular events.
We measured cardiac troponin I with a high-sensitivity assay (hs-cTnI) in 4805 participants attending both the second (HUNT 2, 1995–97) and third wave (HUNT 3, 2006–2008) of the prospective observational Nord-Trøndelag Health (HUNT) Study. We constructed statistical models with both relative and absolute changes of hs-cTnI from HUNT 2 to HUNT 3. A composite end point of cardiovascular death or first admission for myocardial infarction or heart failure was generated.
Participants with relative decrease in hs-cTnI were more frequently younger and female and had lower blood pressure and body mass index. Participants with relative increase in hs-cTnI more frequently were older and male, with higher systolic blood pressure. The adjusted hazard ratio (HR) for relative increase in hs-cTnI was 1.68 (95% CI, 1.16–2.42) and the adjusted HR for relative decrease was 1.19 (95% CI, 0.84–1.68). Absolute increases in hs-cTnI exhibited similar prognostic properties as relative increases in hs-cTnI. The most recent measurement of hs-cTnI outperformed the change variables in discrimination and reclassification models.
Both relative and absolute increases in hs-cTnI are independently associated with cardiovascular risk. For refinement of risk prediction models, the most recent measurement of hs-cTnI should be preferred in clinical practice.
