Utilizing saliva for non-invasive detection of exercise-induced myocardial injury with point-of-care cardiac troponin-I

Point-of-care (POC) cardiac troponin-I (cTnI) measurement methods often involve immunoassays, which can provide a momentary view of cTnI levels but the current modality highly restricts access to and frequency of testing in a sports and exercise medicine setting due to the requirement of a blood draw. This study aimed to compare cTnI concentrations in saliva and serum in athletes before (T1), early (T2), 4 h (T3), and 24 h (T4) after exercise. 82 male runners (age: 26.82 ± 2.49 years) were recruited and then randomly assigned to two groups using computer-generated permuted blocks with a 2:1 ratio via sequentially numbered opaque envelopes. 54 participants (group 1) completed a 5-km time-trial, while 28 participants (group 2) did not undergo this exercise. POC testing device was used to quantify salivary and serum concentrations of cTnI in both groups at T1, T2, T3, and T4. In group 1, salivary and serum concentrations of cTnI increased at T2 (0.41 ± 0.06 ng/mL and 0.48 ± 0.06 ng/mL) compared to T1 (0.18 ± 0.04 ng/mL and 0.22 ± 0.04 ng/mL), reaching the highest values at T3 (0.62 ± 0.05 ng/mL and 0.76 ± 0.05 ng/mL) with the subsequent return to baseline values at T4 (0.16 ± 0.03 ng/mL and 0.22 ± 0.03 ng/mL). In group 2, there were no time-dependent changes in cTnI levels in both saliva (T1: 0.17 ± 0.04 ng/mL, T2: 0.16 ± 0.03 ng/mL, T3: 0.16 ± 0.04 ng/mL, T4: 0.16 ± 0.04 ng/mL) and serum (T1: 0.22 ± 0.04 ng/mL, T2: 0.22 ± 0.04 ng/mL, T3: 0.21 ± 0.03 ng/mL, T4: 0.21 ± 0.04 ng/mL). Salivary and serum concentrations of cTnI were significantly lower in group 2 compared to group 1 at T2 and T3; there was no difference between groups at T1 and T4. Deming regression and Passing-Bablok regression revealed that there was differential bias (at T3), but proportional agreement (at T1, T2, T3, T4) between salivary and serum levels of cTnI in both groups. The Bland-Altman method indicated that there was a negative differential bias but no proportional bias in the data. Recalibration of the new measurement approach (measurement of cTnI levels in saliva) by using the MethodCompare R package was effective in removing existing bias, as evidenced by its similar precision to the reference method (measurement of cTnI levels in serum), particularly at T2, T3, and T4. In athletic settings, quantification of cTnI levels in saliva utilizing the POC-cTnI-Getein1100 assay may be a useful non-invasive tool in evaluating whether exercise-induced increases in cTnI levels are transient or there are acutely or chronically elevated cTnI concentrations.