Table 1: Studies reporting percentile values for hs-cTnT in different reference populations. †=median [IQR]; =only range provided; CI: confidence interval; cTn: cardiac troponin; eGFR: estimated glomerular filtration rate; UK: United Kingdom; US: United States; ARIC: Atherosclerosis Risk in Communities study; CHS: Cardiovascular Health Study; DHS: Dallas Heart Study.

StudyStudy objectiveYearLocationStudy populationAge,
mean ± SD
Population, according to sex percentile (ng/L) [95% CI]Comments
Males (%)Females (%)MalesFemales

hs-cTnT (Roche Diagnostics)

Collinson et al. [40]To determine the effect of patient selection on the reference percentile2012UK54558 259 (47.5)286 (52.5)22.812.8Reference population selection based on: medical history, biomarkers and cardiac imaging

Apple et al. [61]To systematically assess percentiles of cTn concentrations in a single population for a large number of assays2012US52418-64272
(52)
252 (48)2013Reference population selection based only on health questionnaire interviews

Giannitsis et al. [62]To validate the hs-cTnT assay2010US61644 ± 13.8309 (50.2)307 (49.8)14.510Reference population selection based only on medical records

Saenger et al. [75]To evaluate the analytical performance of the hs-cTnT assay in a multicenter, international trial2011US, Europe53337268 (50.3)265 (49.7)15.59Reference population selection based only on medical records

1600 (whole cohort)61 ± 14872 (54.6)728 (45.4)21.8 16.3 Reference population selection based on medical history, biomarkers and cardiac imaging; population stratified by age;
Various subpopulations included: high heterogeneity and reference population unable to achieve the recommended statistical power to determine the percentile for each subgroup
Franzini et al. [76]To determine the upper reference limit for cTnT in Italian apparently healthy subjects2015Italy553<20270 (48.8)283 (51.2)10.9 6.8
87220-64503 (57.7)369 (42.3)23.2 10.2
175>6599 (56.6)76 (43.4)36.8 28.6

Mingels et al. [77]To study the improvements made by new hs-cTn assays in detecting exercise-induced cTn release2009US47951 264 (55.1)215 (44.9)168Reference population selection based only on medical records

1540 (whole cohort)57 ± 8733 (47.6)807 (52.4)16 12 Reference population selection based on: medical history and biomarkers; population stratified by age
Kimenai et al. [78]To assess sex-specific and age-specific percentile upper reference limits of hs-cTnT and hs-cTnI in a single reference cohort2016Netherlands28340-49120 (42.4)163 (57.6)16 12
94650-64443 (46.8)503 (53.2)14 12
31165-75170 (54.7)141 (45.3)28 27

Koerbin et al. [79]To evaluate the analytical characteristics of the hs-cTnT assay2010Australia11125-7462 (55.9)49 (44.1)12.911Reference population selection based on medical history, biomarkers and cardiac imaging

DHS: 197843.2 ± 9.6873 (44.1)1105 (55.9)17 11 Reference population selection based on: progressive cohorts restriction based on clinical history, imaging and/or laboratory tests
Gore et al. [44]To determine the percentile values in three large community-based subcohorts, restricted by healtiness criteria2014USARIC: 757561 ± 92972 (39.2)4603 (60.8)26 15
CHS: 137472 ± 6489 (35.6)885 (64.4)34 24

Mueller et al. [80]To assess percentile in a blood donors population2016Austria40235
259 (64.4)143 (35.6)13.911.3Reference population selection based on: no overt cardiovascular disease, eGFR>90 ml/min

Ungerer et al. [81]Determine and compare percentile cut-offs of 3 cTn assays in a cohort of blood donors2016Australia2004Male: 43.7 1299 (64.8)705
(35.2)
31.3 [90% CI: 25.0-57.5]20.2 [90% CI: 9.9-51.7]Reference population selection based on: health questionnaire
Female: 33.2

Yang et al. [95]Establish percentile in a healthy Chinese population2016China1725Male:
54 ± 20
818 (47.4)907 (52.6)Several according to ageSeveral according to ageReference population selection based on clinical history, physical examination, lab tests
Female:
54 ± 19

Monneret et al. [107]Establish age and sex specific percentile in patients without CKD2018France2707Male:
62
1548 (57.2)1159
(42.8)
Several according to ageSeveral according to ageReference population selection based on age partitioning and outliers removal. Cut-off obtained with an analytical imprecision-based approach
Female:
63

Welsh et al. [108]Evaluating the influence of several variables, including sex, on the percentile levels of hs-cTnT and hs-cTnI2018Scotland1950135-658126 (41.7)11375 (58.3)Several according to ageSeveral according to ageReference population selection based on general population; health questionnaire; lab tests