Association between troponin level and medium-term mortality in 20 000 hospital patients.

Abstract

Cardiac troponin (cTn) concentrations above the manufacturer recommended upper limit of normal (ULN) are frequently seen in hospital patients without a clinical presentation consistent with type 1 myocardial infarction, and the significance of this is uncertain. The aim of this study was to assess the relationship between medium-term mortality and cTn concentration in a large consecutive hospital population, regardless of whether there was a clinical indication for performing the test. This prospective observational study included 20 000 consecutive in-hospital and outpatient patients who had a blood test for any reason at a large teaching hospital, and in whom a hs-cTnI assay was measured, regardless of the original clinical indication. Mortality was obtained via NHS Digital. A total of 20 000 patients were included in the analysis and 18 282 of these (91.4%) did not have a clinical indication for cardiac troponin I (cTnI) testing. Overall, 2825 (14.1%) patients died at a median of 809 days. The mortality was significantly higher if the cTnI concentration was above the ULN (45.3% vs 12.3% p<0.001 log rank). Multivariable Cox analysis demonstrated that the log₁₀ cTnI concentration was independently associated with mortality (HR 1.76 (95% CI 1.65 to 1.88)). Landmark analysis, excluding deaths within 30 days, showed the relationship between cTnI concentration and mortality persisted. In a large, unselected hospital population, in 91.4% of whom there was no clinical indication for testing, cTnI concentration was independently associated with medium-term cardiovascular and non-cardiovascular mortality in the statistical model tested.