Cardiac and renal biomarkers in recreational runners following a 21 km treadmill run

Abstract Background Highly trained athletes running 42 km or more demonstrate elevated cardiac biomarkers, ventricular dysfunction, and decreased glomerular filtration rate (GFR). Whether similar changes occur in the much larger population of recreational runners following half‐marathon distance running is unclear. Hypothesis Recreational runners exhibit changes in myocardial and renal biomarkers, including ventricular strain, after a half‐marathon treadmill run. Methods 10 recreational subjects (mean age 36.5 ± 6.5 years) ran 21 km on a treadmill (mean completion time 121.6 ± 16.1 minutes). Serum high‐sensitivity troponin T (hsTnT), amino‐terminal pro‐brain natriuretic peptide (NT‐proBNP), creatinine, and neutrophil gelatinase‐associated lipocalin (NGAL) were measured prior to, 1 hour post‐, and 24 hours post‐exercise. Pre‐ and post‐exercise echocardiograms were performed. Results All biomarkers increased 1 hour post‐exercise: hsTnT by 8.5 ± 8.5 pg/ml (P < .05), NT‐ProBNP by 26.2 ± 22.8 pg/ml (P < .05) and NGAL by 29.5 ± 37.7 ng/ml (P=NS). By 24 hours post‐run, these biomarkers declined toward baseline levels. Right ventricle (RV) free wall and left ventricle global longitudinal strain decreased by 5.5% and 1.8%, respectively (P < .001). Changes in NGAL correlated well with changes in serum creatinine (R = 0.79, P < .01) and GFR (R = −0.73, P < .05). Faster 21 km completion times, and a larger reduction in post‐exercise RV strain, were associated with higher NGAL levels: (R = −0.75, P = .01) and (R = 0.66, P < .05), respectively. Conclusion A 21 km run in recreational runners is associated with transient ventricular stunning and reversible changes in myocardial and renal biomarkers. Whether repeated bouts of similar activity contributes to chronic cardiac or kidney dysfunction deserves further evaluation.


| Study population and protocol
Study participants were recruited from a local recreational running club. Eligible subjects were recreational runners 21 years or older with a weekly training distance of under 50 km. 12 In contrast, highly trained endurance athletes have average weekly training distances of 120 to 200 km. 13 Potential participants were screened by completing a comprehensive questionnaire. Individuals with any history of cigarette smoking, diabetes, hypertension, hyperlipidemia, consumption of any heart rate lowering medication, and those with a family history of sudden cardiac death were excluded from this study.
Training history was ascertained for all participants including average running distance per week, number of years of running, and number of half-and full-marathons completed. Measurements of height, weight, peripheral blood pressure, and resting heart rate were obtained prior to the run. Holter monitors were attached to all runners for the duration of the run to track their exercise heart rates. Maximum heart rate for each participant was calculated by subtracting their age from 220. Level of intensity was then calculated by dividing average heart rate during activity by maximum heart rate. Based on the American College of Sports Medicine (ACSM) guidelines, a range of 30% to 49% of maximum heart rate corresponds to "light" intensity, 50% to 69% to "moderate" and 70% to 89% to "hard". 14 All subjects completed an indoor 21 km run at their own pace within an air-conditioned gymnasium on a commercial grade treadmill (Precor C946i, Woodinville, WA) with full manual settings. Isotonic beverages and plain water were made available throughout the run and participants were encouraged to hydrate liberally. Fluid intake of the subjects was not monitored. Transthoracic echocardiography (TTE) was performed pre-run and within an hour of completing the 21 km run. Peripheral blood samples were obtained pre-run, within 1 hour post-run and within 24 hours post-run.
The protocol was approved by the local institutional review board ethics committee and all subjects gave written informed consent prior to study participation.

| Biochemical studies
Serum creatinine was measured by the Enzymatic Creatinine 2 method using the Advia 2400 analyzer (Siemens Healthcare Diagnostics, Erlangen, Germany). The reference range for this assay is 9-2652 umol/L, and upper limit of normal is 125 umol/L and 90 umol/L in males and females, respectively. Estimated glomerular filtration rate (GFR) was then calculated using the Modification of Diet in Renal Disease (MDRD) equation. 15 High-sensitivity troponin T (hsTnT) and NT-proBNP were measured using electrochemiluminescence immunoassays on a cobas e411 analyzer (Roche Diagnostics GmbH, Mannheim, Germany). For hsTnT, the 99th percentile for normal subjects is 14 pg/ml. For NT-ProBNP, the upper limit of normal in subjects 75 years of age and below is considered to be 125 pg/ml.

| Echocardiography
Subjects underwent M-mode, two-dimensional (2D) and doppler TTE using a Vivid 7 Dimension ultrasound system (GE Healthcare, Milwaukee, WI) equipped with a 2.5 MHz probe, in accordance with American Society of Echocardiography (ASE) guidelines. 16,17 These were analyzed by a neutral physician blinded to the results of the endurance exercise.
Left ventricular (LV) ejection fraction and stroke volume were measured using the biplane method of disks. Fractional shortening was calculated using the difference in LV end-diastolic and endsystolic diameters via M-mode. Pulsed wave tissue doppler imaging was performed to obtain peak early (e') and late (a') myocardial velocities and the ratio of early diastolic transmitral flow velocity to e' (E/e').
Indices of myocardial deformation obtained by speckle-tracking analysis using the EchoPac software (Version 11.1.8, GE Healthcare, Horten, Norway). LV global longitudinal strain (GLS) was obtained as the average segmental value based on three apical imaging planes.
Right ventricular (RV) free wall strain was assessed along the length of the free wall from apical 4-chamber images. Other measures of RV systolic function that is, 2D fractional area change (FAC), tricuspid annular plane systolic excursion (TAPSE) and myocardial performance index (via pulsed tissue Doppler) were also obtained. 18 Right atrial pressure (RAP) was calculated. 19,20 Inferior vena caval dimension was measured during expiration, just proximal to its junction with the hepatic vein. 18 Mean pulmonary artery pressure (mPAP) was estimated using Mahan's regression equation. 20 (Table 1) The study population consisted of 10 subjects (40% female) with mean age of 36.5 ± 6.5 years and mean body mass index of 21.9 ± 1.7 kg/m 2 .

| Statistical analysis
Participants had 4.25 ± 3.1 years of running experience, and ran 29 ± 9.1 km weekly. Mean pre-exercise values for systolic and diastolic blood pressure and heart rate were 120 ± 11.6 mmHg, 72.8 ± 8.4 mmHg and 59 ± 6.7 beats per minute respectively. All participants were free of modifiable cardiovascular risk factors.
Participants completed the 21 km treadmill run in an average time of 121.6 ± 16.1 minutes (122.7 ± 20 minutes for male subjects, compared with 120 ± 10.1 minutes for females). These timings correspond to the 60 to 70th percentile and 30 to 40th percentile for recreational male and female runners respectively, and 40 to 50th percentile overall. 21 Average heart rate during the run was 149.8 ± 11.5 bpm. This translated to 81.6 ± 5.1% of their age-predicted maximum heart rate.
Mean NT-proBNP at baseline was 13.5 ± 10 pg/ml and increased to 39.7 ± 29 pg/ml (P < .01) within 1 hour post-run. In no subject was the upper limit of normal exceeded. As with hsTnT, levels of NT-proBNP declined at 24 hours post-run to 26.8 ± 17 pg/ml (P = .06).
There was no correlation between NT-proBNP and hsTnT.

| DISCUSSION
In this study, we assessed the effects of exercise in recreational runners completing a 21 km run with continuous access to oral rehydration in a controlled, air-conditioned indoor environment.
Highly sensitive biomarkers were used to determine if an endurance run of 21 km was associated with potential myocardial or renal injury. reported prevalence rates as high as 17%. [23][24][25] Conversely, other studies encompassing individuals with a broad spectrum of physical activity levels have shown minimal late gadolinium enhancement. 26,27 These conflicting data suggest endurance exercise and myocardial scarring remains a highly controversial topic.
Our study showed a statistically significant reduction in LV GLS, in keeping with previous work on athletes participating in prolonged endurance activity. 28  Tricuspid annulus systolic velocity (S') (cm/s) 12.6 ± 2.1 11.7 ± 2.6 .235 Myocardial performance index 0.54 ± 0.14 0.58 ± 0.11 .45 Mean pulmonary arterial pressure (mmHg) 9.7 ± 7.4 16.9 ± 6.6 .033 Tricuspid utilizing direct measurement of plasma volumes as well as both urinary and blood biomarkers will be necessary to elucidate any potential heart-kidney relationship during endurance activity.
Limitations of our study included its small sample size, making generalisability an issue. Additionally, cardiac magnetic resonance, the current imaging gold standard, was not utilized for more accurate assessment of cardiac volumes and function due to operational constraints. However, each subject served as their own control, and any inaccuracies in echocardiographic assessment should be systematic.
We did not perform a 24-hour post-run echocardiographic assessment, which would have provided more information on the reversibility of the observed changes in ventricular strain. Objective baseline cardiopulmonary fitness via maximum oxygen uptake (VO 2 max) of study participants was not available, hence, the ACSM guidelines were used as a surrogate for maximum heart rate and subsequent calculation of exercise intensity.

| CONCLUSION
In this study of recreational runners completing a 21 km treadmill run, we observed a reversible elevation of cardiac biomarkers of myocardial injury and hemodynamic stress, which was associated with ventricular stunning, as evidenced by reduced biventricular strain. We further observed changes in markers of renal function consistent with RIFLE stage I AKI,

ACKNOWLEDGMENTS
We would like to thank Daw Hla Yee, Jie-Ying Lee, and Peini Too who provided valuable assistance and helped to facilitate the study.