Trajectories of high-sensitivity cardiac troponin I in the two decades before cardiovascular death in Whitehall II

Abstract

Abstract Background High-sensitivity cardiac troponin may be a promising biomarker that could be used for personalised cardiovascular risk prediction and monitoring in the general population. Temporal changes in high-sensitivity cardiac troponin before cardiovascular death are largely unexplored. Purpose Using the longitudinal Whitehall II cohort, we evaluated whether three serial high-sensitivity cardiac troponin I measurements over 15 years improved prediction of cardiovascular death when compared to a single time point at baseline. Methods Whitehall II is an ongoing longitudinal observation cohort study of 10,308 civil servants, and we included participants who had at least one cardiac troponin measurement and outcome data available. We constructed time trajectories to evaluate the temporal pattern of cardiac troponin I in those who died from cardiovascular disease as compared to those who did not. Cox regression and joint models were used to investigate the association of cardiac troponin I in relation to cardiovascular death using single time point (at baseline) and repeated measurements (at baseline, 10 and 15 years), respectively. The discriminative ability was assessed by the concordance index. Results In total, we included 7,293 individuals (mean age of 58 years [SD 7] at baseline, 29.4% women). Of these, 5,818 (79.8%) and 4,045 (55.5%) individuals had a second and third cardiac troponin I concentration measured, respectively. Cardiovascular death occurred in 281 (3.9%) individuals during a median follow-up of 21.4 [IQR, 15.8 to 21.8] years. In the 21-year trajectories of cardiac troponin I, we observed higher baseline concentrations in those who died due to cardiovascular disease as compared to those who did not (median 5 [IQR, 2 to 9] ng/L versus 3 [IQR, 2 to 5] ng/L respectively, Figure). Cardiac troponin I was an independent predictor of cardiovascular death, and the hazard ratio (HR) derived from the joint model that included serial cardiac troponin measurements was higher than the HR derived from the single time point model (single time point model: adjusted HR 1.53, 95% Confidence Interval [CI] 1.37 to 1.70 per naturally log transformed unit of cardiac troponin I, versus repeated measurements model: adjusted HR 1.79, 95% CI 1.58 to 2.02). The discriminative ability of the cardiac troponin model improved when using repeated measurements (concordance index of unadjusted cardiac troponin models, single time point: 0.668 versus repeated measurements: 0.724). Conclusions Our study shows that cardiac troponin I trajectories were persistently higher among individuals who died from cardiovascular disease. Cardiac troponin I is a strong independent predictor of cardiovascular death, and incorporating repeated measurements of cardiac troponin improves cardiovascular risk prediction in the general population. Funding Acknowledgement Type of funding sources: Foundation. Main funding source(s): Cardiac troponin I measurements and analysis were supported by were supported by Siemens Healthineers. The study was supported by Health Data Research UK which receives its funding from HDR UK Ltd (HDR-5012) funded by the UK Medical Research Council, Engineering and Physical Sciences Research Council, Economic and Social Research Council, Department of Health and Social Care (England), Chief Scientist Office of the Scottish Government Health and Social Care Directorates, Health and Social Care Research and Development Division (Welsh Government), Public Health Agency (Northern Ireland), British Heart Foundation and the Wellcome Trust. NLM is supported by the British Heart Foundation through a Senior Clinical Research Fellowship (FS/16/14/32023), Programme Grant (RG/20/10/34966) and a Research Excellence Award (RE/18/5/34216). The funders had no role in the study and the decision to submit this work to be considered for publication.