Defining normality in a European multinational cohort: Critical factors influencing the 99th percentile upper reference limit for high sensitivity cardiac troponin I

- A questionnaire and biomarkers were used to define a cardio-healthy cohort.
- Male gender, BNP, HbA1c and smoking were associated with hs-cTnI concentration.
- Introduction of selection criteria decreased the impact of age on hs-cTnI value.
- Non-parametric analysis results were higher than from robust statistical analysis.
- 99th percentile URL ranged from 23.7 to 7.1 ng/L depending on analysis.

ObjectiveTo establish and critically evaluate the 99th percentile upper reference limit (URL) for high-sensitivity cardiac troponin I (hs-cTnI) in a large healthy European cohort using different selection criteria.Methods1368 presumably healthy individuals from 9 countries were evaluated with surrogate biomarkers for diabetes (glycated hemoglobin [HbA1c] < 48 mmol/mol), myocardial (B-type natriuretic peptide [BNP] < 35 pg/mL) and renal dysfunction (estimated glomerular filtration rate [eGFR] > 60 mL/min/1.73 m2), and dyslipidemia to refine the healthy cohort. The 99th percentile URLs were independently determined by the non-parametric and robust methods.ResultsThe use of biomarker selection criteria resulted in a decrease of the 99th percentile URL for hs-cTnI from 23.7 to 14.1 ng/L and from 11.2 to 7.1 ng/L, when using the non-parametric percentile and robust methods, respectively; a further reduction after exclusion of individuals with dyslipidemia was noted. Male gender, BNP, HbA1c and smoking status were independently associated with hs-cTnI concentration in the presumably healthy population, while the impact of age, present in the univariate analysis, decreased after adjustments for gender and surrogate biomarkers. The BNP-based inclusion criterion had the most pronounced effect on the 99th percentile URL, excluding 21% of the study participants and decreasing its value to 11.0 (7.1) ng/L according to the non-parametric (robust) method. Gender, but not age-specific, differences at 99th percentile URL have been identified.ConclusionThe selection of a reference population has a critical impact on the 99th percentile value for hs-cTnI. A uniform protocol for the selection of the healthy reference population is needed.