High-sensitivity cardiac troponin I and risk of dementia: 25-year longitudinal study in the Whitehall II cohort

ObjectiveWe hypothesise that subclinical myocardial injury during midlife, indexed by increases in cardiac troponin I, is associated with accelerated cognitive decline, smaller structural brain volume, and higher risk of dementia. DesignLongitudinal cohort study SettingCivil service departments in London (Whitehall II study) Participants5985 participants aged 45-69 had cardiac troponin I measured by high-sensitivity assay at baseline (1997-99) for prospective cohort analyses. A nested case-control sample of 3475 participants (695 dementia cases and 2780 matched controls) was used for backward cardiac troponin I trajectory analysis. 641 participants provided magnetic resonance imaging (MRI) scans for brain volume analysis. Main outcome measuresIncident dementia cases were ascertained from national hospital episode statistics, mental health and mortality registers until 2023. Cognitive testing was performed at six waves over 25 years (1997-99, 2002-04, 2007-09, 2012-13, 2015-16, 2019-22). Brain volume metrics were derived from structural MRI scans (2012-16). ResultsFor prospective cohort analyses, 606 (10.1%) incident cases of dementia were recorded over a median follow-up of 24.8 years. Doubling of cardiac troponin was associated with 11% higher risk of dementia (HR=1.11, 95% CI: 1.04 to 1.19). Participants with increased cardiac troponin at baseline had a faster decline of cognitive function with age. Compared to participants with concentrations below the limit of quantitation (<2.5 ng/L), those in the upper third (>5.2 ng/L) had similar global cognitive z score at age 60, but had 0.10 (95% CI: 0.02 to 0.18) standard deviations lower score at age 80, and 0.19 (0.03 to 0.35) standard deviations lower score at age 90. Participants with dementia had increased cardiac troponin concentrations compared with those without dementia between 7 and 25 years before diagnosis. Compared to those with low cardiac troponin level (<2.5 ng/L at baseline) those with concentrations >5.2 ng/L had lower grey matter volume and higher hippocampal atrophy 15 years later, equivalent to ageing effects of 2.7 and 3 years, respectively. ConclusionsSubclinical myocardial injury at midlife was associated with higher dementia risk in later life. What is already known on this topicA systematic review of observational studies suggests higher cardiac troponin concentrations are associated with poorer cognitive function and increased dementia risk. Formal meta-analysis was not performed due to the small number of available studies. Existing studies assessed cardiac troponin once and had relatively short follow-up time. Evidence is lacking on the time course of cardiac troponin level before diagnosis in dementia cases compared with controls. What this study addsPeople with increased cardiac troponin I concentrations in mid-life had faster cognitive decline and were more likely to develop dementia over 25 years of follow-up. Backward trajectory analysis using three measurements using a high-sensitivity assay showed that people with dementia had higher cardiac troponin levels as early as 25 years before dementia diagnosis compared with those without dementia. People in the magnetic resonance imaging subcohort with higher cardiac troponin I concentrations at baseline had lower grey matter volume and hippocampal atrophy 15 years later. Subclinical myocardial injury in midlife, by indicating long-term risk of dementia, is unlikely to be due to preclinical changes before dementia onset and may lie on the causal pathway to dementia.