A high incidental rise in cardiac troponin I carries a higher mortality risk in older patients than in those with a diagnosed acute coronary syndrome.

A high incidental rise in cardiac troponin I carries a higher mortality risk in older patients than in those with a diagnosed acute coronary syndrome SIR—Cardiac troponins are highly sensitive and specific biomarkers for myocardial injury [1]. Furthermore, in the context of an acute coronary syndrome (ACS), baseline elevations of troponin have prognostic significance, with a higher risk of death or recurrent infarction early after presentation in patients with a raised troponin [2]. Although tro-ponin is a well-established prognostic marker among patients presenting with unstable coronary syndromes, incidental troponin rises are often encountered in clinical practice where neither an ACS nor another disease associated with an increase in troponin level [3] had been considered. In older populations, the presentation of an ACS is often non-specific (e.g. falling, confusion) and thus greater reliance is placed upon the troponin result by clinicians for diagnosis. These troponin elevations may not however signify thrombotic ischaemia, and the high sensitivity but low specificity of the test among patients with a low pre-test probability of disease [4] thus confers prognostic uncertainty in this group. This may divert attention from the true underlying clinical problem, leading to unnecessary further cardiac evaluation. We have previously reported that older people with no apparent cause for a troponin rise were found to have higher short-term readmission rates and mortality compared with those with an ACS or disease known to be associated with a troponin rise [5]. As an incidental rise in troponin appears to be more prevalent in older people [6], we sought to examine longer-term case fatality of an incidental rise in troponin I in older patients. Methods Study population Two hundred and thirty-seven patients with a troponin I level ≥ 0.04 µg/l measured at least 8 h after the onset of symptoms were recruited from all medical admitting services within a university hospital in the UK from January to May 2004. Our recruitment strategy and the methodology have been reported previously [5]. In brief, we prospectively recruited all patients aged 65 years and over with admission troponin I level ≥ 0.04 µg/l at least 8 h after onset of ischaemic symptoms according to the local reference laboratory values (Beckman Coulter assay, Access). We took a second recording of troponin I 12 h after admission if in non-specific presentations. Patients were prospectively recruited and placed by us into one of three categories based on history, examination, diagnosis made during …