Background: Cardiac troponin is a cornerstone of the initial diagnostic workup in patients with suspected cardiovascular disease. Elevated troponin above the 99^th percentile of the upper reference limit indicates an increased risk for major adverse events, and patients usually require urgent treatment, including coronary angiography. Meanwhile, patients with detectable troponin levels below the 99^th percentile represent a more heterogeneous collective at need for further risk stratification, but the implications of detectable troponin below the 99^th percentile remain incompletely understood. This study aims to assess the prognostic relevance of detectable troponin below the upper reference limit compared to troponin lower than the detectable range in patients with cardiovascular disease.

Methods: The Essen Coronary Artery Disease (ECAD) registry was screened for patients with detectable troponin below the 99^th percentile and without detectable troponin upon admission. Patients with ST-segment elevation myocardial infarction, patients with no data on admission troponin, and patients with admission troponin above the 99^th percentile were excluded. A Siemens Dimension Troponin I (2004-2014; detectable limit: 40 ng/L, 99^th percentile: 70 ng/L) and a contemporary Centaur high-sensitive Troponin I Ultra (2014-2019; detectable limit: 6 ng/L, 99^th percentile: 40 ng/L) assay were used. Overall survival was defined as the primary endpoint. Cox regression analysis was used to determine the association of troponin groups with incident mortality, adjusting for age, sex, low-density lipoprotein (LDL) cholesterol, smoking status, systolic blood pressure, and family history of premature cardiovascular disease.

Results: 14,776 patients (mean age 65.35 ± 12.74 years, 71.3% male) were included in the analysis after screening of 40,461 consecutive patients with hospital admissions between 2004 and 2019. 11,965 patients had troponin levels below the detectable limits, while 2,811 patients had detectable troponin below the 99^th percentile. During a mean follow-up of 4.25 ± 3.76 years, 2379 (16.1%) deaths of any cause occurred. The overall mortality was higher in patients with detectable troponin below the 99^th percentile compared to patients without detectable troponin (20.8% vs. 15.0%, p < 0.001). In multivariable regression analysis, detectable troponin below the 99^th percentile was significantly associated with all-cause mortality (HR 1.71; 95% CI 1.46-2.01; p < 0.001). To determine the role of troponin levels in patients with detectable troponin below the 99^th percentile, patients were stratified in tertiles. A significant stepwise relationship with increasing overall mortality between the tertiles of troponin levels as compared to troponin levels below the detectable range as reference was determined (tertile 1, HR 1.62 (1.39-1.90); tertile 2, HR 1.88 (1.63-2.16); tertile 3, HR 2.02 (1.74-2.35)).

Conclusions: Detectable troponin below the 99^th percentile was an independent predictor of overall mortality in patients with suspected cardiovascular disease, and a gradually higher risk for mortality with increasing troponin levels was identified. Therefore, the data indicate that detectable troponin in patients with suspected cardiovascular disease may serve to identify patients at high risk for adverse outcomes.