Acute impact of first-time marathon running on the heart in middle-aged men Authors: Inarota Laily, Niels van Steijn, Tom G H Wiggers, Martijn Froeling, R Nils Planken, Sjoerd M Verwijs, Ferdinand H de Haan, Ehsan Motazedi, Evert A L M Verhagen, Harald T Jørstad, Adrianus J Bakermans Published: 2025 ABSTRACT AbstractBackgroundMiddle-aged men seem to be particularly susceptible to adverse cardiovascular effects of endurance exercise. The acute impact of training, marathon finishing and recovery on the heart has not been extensively investigated in such at-risk novice marathon runners.MethodsWe prospectively quantified cardiac changes in middle-aged men (41.2±4.5 years; n=17) who participated in their first marathon run at four time points: at baseline, after 16 weeks of training, within 10 hours after completing the 2021 Amsterdam Marathon run and after 4 weeks of recovery. Measurements included comprehensive 3-Tesla MRI examination, echocardiography, ECG and blood sample assays of high-sensitivity troponin-T (hs-TnT) and N-terminal pro-B-type natriuretic peptide (NT-proBNP).ResultsAcute effects associated with finishing the marathon run were a <10% reduction in left ventricular end-diastolic volume in absence of other morphological changes, and only minor changes in systolic contractile mechanics and diastolic filling patterns. Neither quantitative myocardial mapping with MRI nor late gadolinium enhancement revealed any signs of post-marathon myocardial injury. ECGs did not reveal any abnormalities. Circulating levels of hs-TnT and NT-proBNP had increased markedly after the marathon, with hs-TnT exceeding the clinical upper reference limit for all participants. All observed marathon running-induced effects were transient.ConclusionsExtensive phenotyping using state-of-the-art quantitative imaging modalities paired with physical assessments and blood assays in this selected cohort offers no evidence to support the notion that first-time marathon running in healthy middle-aged men has a detrimental impact on the heart.Keywords: Biomarkers, Magnetic Resonance Imaging, Echocardiography WHAT IS ALREADY KNOWN ON THIS TOPIC High-intensity endurance exercise increases circulating levels of biomarkers that could signal myocardial injury. The acute impact of first-time marathon running on the heart of middle-aged men has not been extensively investigated. High-intensity endurance exercise increases circulating levels of biomarkers that could signal myocardial injury. The acute impact of first-time marathon running on the heart of middle-aged men has not been extensively investigated. WHAT THIS STUDY ADDS In middle-aged male novice marathon runners, modestly elevated levels of troponin-T and N-terminal pro-B-type natriuretic peptide were accompanied by transient subclinical alterations in systolic and diastolic function. No evidence of any myocardial injury was found with MRI, echocardiography or ECG. In middle-aged male novice marathon runners, modestly elevated levels of troponin-T and N-terminal pro-B-type natriuretic peptide were accompanied by transient subclinical alterations in systolic and diastolic function. No evidence of any myocardial injury was found with MRI, echocardiography or ECG. HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY Our findings suggest that the transient and modest exercise-induced release of biomarkers does not reflect myocardial injury, but rather a physiological response to stress. Our findings suggest that the transient and modest exercise-induced release of biomarkers does not reflect myocardial injury, but rather a physiological response to stress. Introduction Beneficial effects of physical exercise have been extensively studied, providing evidence for current recommendations on minimum amounts of physical exercise needed to maintain or improve health.1 However, there may be upper limits to the amount and intensity of vigorous exercise that, if exceeded, could be detrimental to cardiovascular health.2 Foundational to this hypothesis are studies that document elevated circulating levels of cardiac troponin after exercise.3 While the exercise-induced release of such biomarkers may be associated with myocardial tissue damage,4 it remains a surrogate measure for cardiac maladaptation or injury. Until now, potential adverse effects of extensive endurance exercise have not been investigated extensively with state-of-the-art imaging in combination with other quantitative assays.5 Recreational marathon running has become a popular endurance sports activity, particularly attracting middle-aged men seeking to improve health through physical exercise. Simultaneously, this population seems to be more susceptible to adverse cardiovascular effects of endurance exercise than experienced athletes.6 7 MRI is particularly suited to non-invasively quantify myocardial tissue properties that can inform on potentially detrimental effects of marathon running, such as oedema, fibrosis and compromised cardiomyocyte integrity.8 Thus far, most imaging studies have only included experienced or elite runners and lack follow-up measurements to investigate the potentially transient nature of any marathon running-induced changes. As such, the acute impact of marathon training, finishing and recovery on the heart has not been extensively investigated specifically in middle-aged male novice marathon runners. Here, we aimed to prospectively quantify any cardiac changes in middle-aged men who trained for and participated in their first marathon run using state-of-the-art quantitative techniques (figure 1). We hypothesised that comprehensive examinations at various time points would reveal acute adverse myocardial effects of marathon running that may explain the exercise-induced release of myocardial injury biomarkers. Study outline. Healthy middle-aged male first-time marathon runners (n=17) were examined at baseline with MRI, echocardiography, blood sampling, treadmill cardiopulmonary exercise testing and ECG. Subjects were then trained for 16 weeks, followed by a battery of post-training (ie, premarathon) evaluations that were identical to baseline assessments. After 2 weeks of tapering, blood samples were drawn on the morning (am) of the marathon. Within <10 hours after finishing the marathon, MRI, echocardiography, ECG and repeated blood sampling were performed to investigate the acute effects of finishing a marathon run. After 4 weeks of recovery, a final set of assessments was conducted to evaluate the transient nature of any marathon running-induced changes. Figure created with Servier Medical Art (https://smart.servier.com) under a Creative Commons BY 4.0 licence (https://creativecommons.org/licenses/by/4.0/). Methods This exploratory prospective cohort study was approved by the local ethics committee (NL70800.029.19). All participants provided written informed consent before inclusion in the study. The study9 was centred around the 2021 Amsterdam Marathon. We included 25 apparently healthy men aged 35–50 years who were motivated to participate in a marathon run for the first time, allowing for a drop-out rate of 25% to achieve a cohort of n=18 finishers that could still be accommodated within the Amsterdam University Medical Centers’ capacity to perform 3-Tesla MRI examinations directly after the marathon. Inclusion criteria included: never trained for or completed a marathon, never ran >21.1 km in a single race or training in the previous year and never trained or competed on a semiprofessional or professional level in endurance sports. Inclusion or exclusion was not based on participant’s prior regular physical activity per se. Metrics of interest We aimed to assess cardiac morphological changes in middle-aged men during training, marathon running and after recovery. Cardiovascular examinations were conducted at four time points (figure 1): (1) before beginning marathon training (baseline); (2) after 16 weeks of guided, unsupervised training (post training; 2 weeks pre-marathon); (3) within 10 hours after completing the marathon (post-marathon) and (4) 4 weeks after the marathon (recovery). Other readouts included: (1) quantitative myocardial parameter mapping with MRI, (2) echocardiography, (3) cardiorespiratory fitness, (4) ECG characteristics and (5) circulating levels of cardiac biomarkers, that is, blood concentrations of high-sensitivity troponin-T and N-terminal pro-B-type natriuretic peptide (NT-proBNP). Participants documented their unsupervised training efforts in a logbook and transferred data from their personal runner’s watches to allow for a training intensity estimate. As a measure of exercise intensity, mean heart rate during the race was collected from runner’s watches and expressed as a percentage of their post-training maximal heart rate during cardiopulmonary exercise testing (CPET). Study procedures Study procedures were conducted according to standard clinical practice, following appropriate guidelines or recommendations. Venous blood samples were collected with every MRI examination, as well as <1 hour before the marathon start, within 1 hour after finishing and again 1–4 hours after finishing the marathon run. Samples were processed for immediate assays or aliquoted for appropriate storage until later analyses. Standardised 12-lead ECG recordings at rest were evaluated according to recommendations for interpretation in athletes.10 Cardiorespiratory fitness was assessed with CPET until exhaustion according to a modified treadmill (Valliant; Lode, Groningen, The Netherlands) ergometry protocol11 at baseline and post-training. Standardised transthoracic echocardiography was used to quantify atrial and ventricular dimensions12 and function and valvular function. Cardiac MRI examinations were performed at 3 Tesla using three similar MR systems (Ingenia; Philips, Best, The Netherlands). Acquisitions included cine MRI and short-axis tagged MRI to quantify13 chamber morphology, function, torsion and strains, respectively, as well as native T1 mapping, T2 mapping, diffusion tensor imaging, late gadolinium enhancement (LGE) MRI and extracellular volume fraction (ECV) mapping (figure 2) to inform on myocardial tissue characteristics such as diffuse or focal fibrosis, scarring, oedema and diffusivity and anisotropy of tissue water.8 14 To avoid any MR system-dependent and interobserver variability, each participant was scanned by the same operator on the same MR system throughout the entire study. Quantitative myocardial parameter mapping with MRI. Mid-ventricular short-axis MRI acquired at 3 Tesla within 3 hours after finishing a marathon run. The examinations included cine MRI (A, B) and tagged MRI (C, D) for chamber morphology, function, torsion and strain analyses, native T1 mapping (E), T2 mapping (F), diffusion tensor imaging for mapping tissue water diffusivity (MD) (G) and extracellular volume fraction (ECV) mapping (H). Statistical analyses Descriptive summaries are presented as mean±SD or median (IQR), as appropriate. Missing values of biomarker concentrations below the detection limit were imputed by their respective detection limit to allow efficient effect size estimation. Outcome changes between time points were evaluated using the Wilcoxon signed-rank test. Training effects were thus evaluated by comparing baseline with post-training measurements, marathon running effects by comparing post-training (ie, pre-marathon) with post-marathon measurements and subsequent recovery effects by comparing premarathon with recovery measurements. In all analyses, the statistical significance level was set to α=0.05. Given our study’s observational nature and relatively small sample size, we did not adjust for multiple testing. Statistical analyses were performed using SPSS Statistics V.28.0 (IBM Corporation, Armonk, New York, USA) and RStudio V.2024.04.1+748 integrated with R V.4.4.0 (posit Team; Posit Software, Boston, Massachusetts, USA). Results We included 25 participants for baseline examination in May 2021. After baseline evaluation, we excluded three subjects due to cardiovascular pathology that mandated a restrictive sports advice.15 Of the remaining 22 participants, two subjects were lost to follow-up, three subjects dropped out for personal reasons and 17 (77%) started and finished the marathon and completed the study protocol. Here, we report the effects of training, participation and recovery in these first-time marathon finishers (n=17). Baseline characteristics The 17 finishers (online supplemental table S1) were healthy, physically active men aged 41.2±4.5 years with a body mass index of 23.4±2.2 kg/m2 and a training history of 31.2±11.2 metabolic equivalent of task hours per week in the 3 months before inclusion. Body surface area-indexed MRI-measured (table 1) left ventricular (LV) and right ventricular (RV) end-diastolic volumes (EDVs) and LV mass were in the upper range of normal values (100±19 mL/m2, 102±16 mL/m2 and 77±15 g/m2, respectively). Cardiac MRI (table 1), echocardiography (table 2), ECG (online supplemental table S3) and blood panel (online supplemental table S1), (online supplemental table S4) results were within normal reference ranges for middle-aged healthy men,11 12 16 17 while baseline maximal oxygen uptake VO2max (54.6±6.3 mL/min/kg; online supplemental table S2) was high, indicating that our participants were relatively fit. Indeed, all had a history of running, road cycling and/or other sports (eg, soccer, field hockey, tennis) for more than 2 hours per week for >1 year up to 29 years prior to this study. Table 1MRI results on cardiac morphology and function of middle-aged male first-time marathon runners (n=17) at baseline, after 16 weeks of training (2 weeks premarathon), directly post marathon and after 4 weeks of recoveryBaselinePost trainingPost marathonRecoveryHeart rate at MRI(beats/min)55 (51 to 58)55 (48 to 60)75 (69 to 81)*54 (51 to 63)†LV mass (g)146 (128 to 168)144 (127 to 173)144 (129 to 171)141 (130 to 175)LV EDV (mL)189 (168 to 221)181 (162 to 225)184 (160 to 220)‡186 (163 to 215)LV ESV (mL)69 (62 to 76)67 (55 to 88)67 (55 to 72)69 (60 to 77)LV SV (mL)121 (110 to 151)117 (107 to 138)119 (102 to 142)113 (109 to 143)LV CO (L/min)6.9 (6.3 to 7.4)6.8 (5.6 to 7.2)9.2 (8.0 to 10.2)*6.7 (6.0 to 7.9)LV EF (%)64 (60 to 67)61 (59 to 67)65 (61 to 69)63 (59 to 68)LV GCS (%)−37 (−40 to −33)−38 (−41 to −34)−38 (−43 to −33)−34 (−40 to −31)LV GLS (%)−25 (−27 to −23)−25 (−28 to −22)−26 (−30 to −25)−25 (−27 to −23)RV EDV (mL)198 (169 to 236)188 (173 to 224)193 (167 to 227)198 (171 to 223)RV ESV (mL)79 (64 to 85)74 (65 to 87)72 (63 to 87)80 (61 to 89)RV SV (mL)119 (109 to 145)112 (108 to 137)118 (106 to 140)112 (109 to 139)RV CO (L/min)7.0 (6.3 to 7.4)6.6 (5.6 to 7.1)9.0 (8.3 to 10.1)*6.7 (6.0 to 7.8)RV EF (%)60 (58 to 65)61 (58 to 65)61 (60 to 68)60 (57 to 64)RV GLS (%)−29 (−31 to −24)−26 (−29 to −24)−31 (−32 to −27)‡−30 (−32 to −25)LA ESV (mL)100 (88 to 127)110 (90 to 124)97 (83 to 117)101 (88 to 127)LA EDV (mL)43 (33 to 57)45 (36 to 55)45 (39 to 58)49 (36 to 59)LA EF (%)57 (55 to 64)57 (56 to 61)54 (51 to 59)‡55 (51 to 61)LA GLS (%)37 (31 to 40)33 (30 to 36)§30 (25 to 35)33 (27 to 36)RA EDV (mL)67 (52 to 74)65 (49 to 77)62 (53 to 69)59 (50 to 74)RA ESV (mL)110 (85 to 131)118 (90 to 134)109 (80 to 133)102 (84 to 138)RA GLS (%)26 (22 to 27)27 (25 to 32)26 (19 to 29)27 (23 to 33)RA EF (%)41 (34 to 48)44 (39 to 48)§43 (34 to 49)44 (36 to 49)LV endocardial circumferential shortening (−)−0.24 (−0.27 to −0.23)−0.24 (−0.25 to −0.22)−0.23 (−0.26 to −0.21)−0.24 (−0.26 to −0.23)LV torsion (rad)−0.12 (0.14 to −0.11)−0.12 (−0.14 to −0.11)−0.15 (−0.16 to −0.14))‡−0.12 (−0.13 to −0.11)Torsion/shortening ratio (rad)0.50 (0.46 to 0.56)0.52 (0.47 to 0.54)0.59 (0.51 to 0.62)0.47 (0.43 to 0.54)Mean diffusivity (mm2/s)1.57 (1.53 to 1.62)1.59 (1.54 to 1.63)1.63 (1.57 to 1.67)1.62 (1.57 to 1.64)ADC (10-3 mm2/s)1.56 (1.50 to 1.61)1.60 (1.54 to 1.64)1.59 (1.55 to 1.68)1.63 (1.53 to 1.67)Fractional anisotropy (−)0.32 (0.30 to 0.35)0.33 (0.31 to 0.36)0.36 (0.32 to 0.37)0.33 (0.31 to 0.41)Native T1 (ms)1195 (1188 to 1205)1194 (1184 to 1213)1193 (1187 to 1212)1202 (1182 to 1206)Myocardial T2 (ms)55 (55 to 56)55 (54 to 56)54 (53 to 55)55 (54 to 56)Haematocrit (L/L)0.45 (0.43 to 0.46)0.44 (0.42 to 0.45)0.44 (0.43 to 0.45)0.44 (0.43 to 0.45)†Blood native T1 (ms)1799 (1743 to 1842)1817 (1764 to 1856)1840 (1797 to 1866)1794 (1717 to 1853)ECV (%)¶26.1 (24.8 to 27.2)26.1 (25.0 to 27.4)25.7 (23.9 to 26.3)25.9 (24.7 to 27.5)ECV (%)**26.1 (24.9 to 27.4)26.1 (24.8 to 26.6)25.1 (23.9 to 25.9)‡26.1 (24.8 to 27.8)Data are presented as median (IQR). Symbols indicate significant results of Wilcoxon signed-rank tests:*p<0.001 compared to post training (marathon effect).† p<0.05 compared to post training (ie, incomplete recovery). ‡p<0.05.§p<0.05 compared to baseline (training effect).¶ECV corrected with individual haematocrit values measured at the time of MRI.**ECV corrected with haematocrit as a uniform constant. ADC, apparent diffusion coefficient; CO, cardiac output; ECV, extracellular volume fraction; EDV, end-diastolic volume; EF, ejection fraction; ESV, end-systolic volume; GCS, global circumferential strain; GLS, global longitudinal strain; LA, left atrium; LV, left ventricle; RA, right atrium; RV, right ventricle; SV, stroke volume. MRI results on cardiac morphology and function of middle-aged male first-time marathon runners (n=17) at baseline, after 16 weeks of training (2 weeks premarathon), directly post marathon and after 4 weeks of recovery Data are presented as median (IQR). Symbols indicate significant results of Wilcoxon signed-rank tests: p<0.001 compared to post training (marathon effect). p<0.05 compared to post training (ie, incomplete recovery). p<0.05. p<0.05 compared to baseline (training effect). ECV corrected with individual haematocrit values measured at the time of MRI. ECV corrected with haematocrit as a uniform constant. ADC, apparent diffusion coefficient; CO, cardiac output; ECV, extracellular volume fraction; EDV, end-diastolic volume; EF, ejection fraction; ESV, end-systolic volume; GCS, global circumferential strain; GLS, global longitudinal strain; LA, left atrium; LV, left ventricle; RA, right atrium; RV, right ventricle; SV, stroke volume. Table 2ECG results on cardiac morphology and function of middle-aged male first-time marathon runners (n=17) at baseline, after 16 weeks of training (2 weeks premarathon), directly post marathon and after 4 weeks of recoveryBaselinePost trainingPost marathonRecoveryLV IDd (cm)5.0 (4.8 to 5.4)5.2 (4.8 to 5.5)5.0 (4.7 to 5.2)*5.1 (4.8 to 5.3)LV IDs (cm)3.1 (2.9 to 3.4)3.3 (2.9 to 3.4)3.1 (2.8 to 3.3)3.2 (2.8 to 3.5)IVS (cm)0.97 (0.88 to 1.06)1.03 (0.94 to 1.09)1.07 (0.93 to 1.08)0.99 (0.85 to 1.06)LV EDV (mL)132 (114 to 149)141 (110 to 174)118 (100 to 132)†120 (106 to 146)LV ESV (mL)55 (44 to 71)63 (44 to 70)48 (40 to 62)*50 (46 to 57)LA ESV (mL)57 (49 to 70)59 (48 to 72)53 (49 to 66)60 (48 to 67)E (m/s)0.77 (0.71 to 0.92)0.79 (0.66 to 0.84)0.66 (0.57 to 0.79)†0.69 (0.62 to 0.78)‡A (m/s)0.52 (0.45 to 0.58)0.49 (0.39 to 0.57)0.54 (0.48 to 0.66)†0.46 (0.41 to 0.55)E/A ratio (−)1.6 (1.3 to 1.9)1.7 (1.4 to 1.9)1.2 (1.0 to 1.4)§1.5 (1.2 to 1.8)E DT (ms)197 (160 to 217)197 (182 to 220)191 (162 to 233)198 (180 to 234)e’ septal (cm/s)13.2 (11.5 to 14.3)12.7 (11.7 to 14.3)11.2 (10.4 to 12.3)†12.0 (11.2 to 12.8)e’ lateral (cm/s)14.7 (12.4 to 17.9)17.4 (15.1 to 18.4)15.4 (12.4 to 16.3)†16.7 (14.9 to 18.6)E/e’ (−)5.9 (4.9 to 6.7)5.0 (4.5 to 6.0)¶5.0 (4.5 to 5.9)4.8 (4.4 to 5.9)LV GLS (%)−18 (−21 to −16)−20 (−21 to −18)−18 (−20 to −18)−19 (−21 to −18)TAPSE (cm)2.6 (2.4 to 3.0)2.5 (2.2 to 2.8)2.5 (2.2 to 2.8)2.4 (2.1 to 2.6)Aortic sinus of Valsalva (cm)3.5 (3.2 to 3.9)3.5 (3.4 to 3.9)3.5 (3.4 to 4.0)3.5 (3.4 to 3.9)TV S’ (cm/s)16.1 (14.2 to 18.5)16.0 (15.0 to 16.6)16.5 (15.1 to 17.8)14.9 (12.6 to 15.9)Data are presented as median (IQR). Symbols indicate significant results of Wilcoxon signed-rank tests: *p<0.05.†p<0.01.‡p<0.05 compared to post training (ie, incomplete recovery). §p<0.001.¶p<0.05 compared to baseline (training effect)A, peak atrial filling flow velocity; DT, deceleration time; e’, early diastolic mitral annular velocity; E, peak early filling flow velocity; EDV, end-diastolic volume; ESV, end-systolic volume; GLS, global longitudinal strain; IVS, interventricular septum; LA, left atrium; LV, left ventricle; LV GLS, left ventricle global longitudinal strain; LV IDd, left ventricle inner diameter at diastole; LV IDs, left ventricle inner diameter at systole; TAPSE, tricuspid annular plane systolic excursion; TV S', tricuspid annular velocity. ECG results on cardiac morphology and function of middle-aged male first-time marathon runners (n=17) at baseline, after 16 weeks of training (2 weeks premarathon), directly post marathon and after 4 weeks of recovery Data are presented as median (IQR). Symbols indicate significant results of Wilcoxon signed-rank tests: p<0.05. p<0.01. p<0.05 compared to post training (ie, incomplete recovery). p<0.001. p<0.05 compared to baseline (training effect) A, peak atrial filling flow velocity; DT, deceleration time; e’, early diastolic mitral annular velocity; E, peak early filling flow velocity; EDV, end-diastolic volume; ESV, end-systolic volume; GLS, global longitudinal strain; IVS, interventricular septum; LA, left atrium; LV, left ventricle; LV GLS, left ventricle global longitudinal strain; LV IDd, left ventricle inner diameter at diastole; LV IDs, left ventricle inner diameter at systole; TAPSE, tricuspid annular plane systolic excursion; TV S', tricuspid annular velocity. Impact of training (baseline vs post-training) Self-reported training effort was 171±71 min/week and 30.8±13.0 km/week of running time and distance in the first 4 weeks, and 213±70 min/week and 37.6±12.4 km/week at 2 weeks premarathon. After 16 weeks, their VO2 at the first ventilatory threshold had increased slightly (32.5 (27.4–34.2) mL/min/kg to 33.3 (28.4–40.5) mL/min/kg, p=0.044; online supplemental table S2), while VO2max and other parameters of cardiorespiratory fitness remained similar (online supplemental table S2). Body weight (−1.3 (−2.8 to –0.3)%, p=0.004) and total cholesterol (−0.4 (−0.8 to –0.2) mmol/L, p=0.006) decreased, while other parameters remained similar (online supplemental table S1). MRI and echocardiography volumetry as well as systolic and diastolic function remained essentially unaltered (tables1 2). We detected a reduction in the LV peak early filling flow velocity (E) to early diastolic mitral annular velocity (e’) ratio (E/e’) with echocardiography (5.9 (4.9–6.7) to 5.0 (4.5–6.0), p=0.046; table 2), and in the LA global longitudinal strain (LA GLS) with MRI (37 (31–40)% to 33 (30–36)%, p=0.049; table 1). Combined, these data show that 16 weeks of guided, unsupervised training led to small improvements in cardiovascular risk factors, without major changes in cardiorespiratory fitness, cardiac morphology or cardiac function. Acute cardiovascular effects of finishing a marathon run (pre-marathon vs post marathon) The finishers (n=17) completed the 2021 Amsterdam Marathon in 3:57 hours:min±34 min at a mean heart rate of 161±8 beats/min, at 87.8±3.5% of their post-training maximal heart rate. Cardiac morphology LV EDV decreased (−22 (−45 to –6) mL on echocardiography, p=0.007; −9 (−15 to 0) mL on MRI, p=0.020). RV EDV, LA end-systolic volume (ESV) and RA ESV remained similar. LV myocardial mass did not change. These data from two different imaging modalities (tables1 2) show that, other than a reduction in LV EDV, there were no changes to overall cardiac morphology immediately after finishing a marathon. Systolic function LV ESV decreased slightly on echocardiography (−14 (−24 to 1) mL, p=0.049), while RV ESV did not change, preserving LV and RV stroke volumes and LV and RV ejection fractions (EFs). Driven by an elevated post-marathon resting heart rate, cardiac output was substantially higher (table 1). LV GLS and global circumferential strain (GCS) remained similar, while RV GLS was somewhat larger post marathon (p=0.017). LA and RA GLS did not change. Consistent with LV GCS estimated from cine MRI, tagged MRI confirmed that LV endocardial circumferential shortening had not changed. Notably, LV torsion was higher (p=0.017), although this was not reflected in an increased (p=0.078) torsion-to-shortening ratio (TSR) post marathon (figure 3). Left ventricular (LV) torsion plotted against LV endocardial circumferential shortening throughout the cardiac cycle. Short-axis tagged MRI was used to quantify myocardial wall motion in middle-aged male first-time marathon runners (n=17). The direction of time is indicated by the arrows, showing a linear trajectory during LV contraction that reflects the torsion-to-shortening ratio. Note the higher absolute torsion at end-systole post-marathon (red), compared with baseline (blue) and post-training (ie, 2 weeks pre-marathon; green) curves. The torsion-to-shortening curve normalised after 4 weeks of recovery (orange). Negative torsion is defined as counterclockwise rotation of the apex relative to the base, when viewed from the apex. Diastolic function LV E decreased (p=0.002), while peak atrial filling flow velocity (A) increased (p=0.007) on marathon running, resulting in a substantially lowered E/A ratio post marathon (1.7 (1.4–1.9) to 1.2 (1.0–1.4), p <0.001). E velocity deceleration time remained similar, and although the e’ at both the septal (p=0.010) and the lateral (p=0.002) side decreased markedly, this did not translate into a marathon running-induced reduction of the E/e’ ratio (table 2). Myocardial tissue characteristics Myocardial native T1 did not change on marathon running, while myocardial T2 showed a trend towards a small reduction post marathon (p=0.056). Myocardial ECV decreased (p=0.013) if haematocrit correction was applied as a uniform constant, but remained similar if individual haematocrit values measured at the time of MRI were used (p=0.179). Haematocrit did not decrease due to marathon running (p=0.254). We did not identify pathological LGE in any participant. Mean diffusivity and fractional anisotropy of myocardial tissue water remained unchanged. Electrocardiography Resting heart rate was high (75±8 beats/min, p=0.002) on the afternoon after the marathon. The QT interval (defined as the time from the onset of the QRS complex to the end of the ST–T segment) had shortened (−16 (−40 to –5) ms, p=0.036), translating into a corrected QT interval increase (+16 (–2 to 36) ms, p=0.020). No abnormalities in P-waves, T-waves, QRS complexes or ST (end of QRS complex to onset of T wave) segments were observed (online supplemental Table S3). Biomarkers Circulating levels of troponin-T (detection limit, 5 ng/L) increased more than four-fold from 8 (<5 to 8) ng/L on the morning of the marathon to 34 (24–46) ng/L within 1 hour post marathon (p<0.001) and were still elevated (p<0.001) at 25 (21–40) ng/L but decreased (p=0.001) after 4 hours (figure 4, (online supplemental Table S4). Premarathon, NT-proBNP was below the detection limit (<50 ng/L) in 16 subjects, increased to 85 (<50 to 108) ng/L within 1 hour post marathon (p=0.003) and remained high at 97 (53–110) ng/L after 4 hours. Creatine kinase myocardial band isoenzyme increased from 2.4 (1.8–3.3) µg/L premarathon to 17.4 (9.1–28.6) µg/L within 1 hour post marathon (p<0.001) and further increased to 23.2 (10.7–48.9) µg/L after 4 hours (p<0.001). Circulating levels of myocardial biomarkers. Blood samples were obtained at baseline, on the morning before the start of the marathon run, within 1 hour after finishing and again 1–4 hours after finishing the marathon run. A recovery sample was taken at 4 weeks after the marathon run. Clinical upper reference limits (URLs) are indicated by the green horizontal line. Grey curves represent data for individual participants (n=17), with the red curve indicating the group mean values for high-sensitivity troponin-T (hs-TnT) (A), N-terminal pro-B-type natriuretic peptide (NT-proBNP) (B) and creatine kinase myocardial band isoenzyme (CK-MB) (C). Note that all participants exceeded the URL for hs-TnT after finishing the marathon. Recovery (premarathon vs recovery) After 4 weeks of recovery, nearly all MRI and echocardiography parameters had reverted to premarathon ranges. On resting ECG (online supplemental Table S3), heart rate was lower (p=0.004), with slight mean prolongations of the QRS complex (+3 (1–10) ms, p=0.035) and QT interval (+41 (24–52) ms, p=0.003) relative to premarathon durations. Circulating biomarker concentrations had returned to baseline values (online supplemental Table S4). Taken together, our data indicate that marathon running-induced changes were all transient. Discussion Strong evidence linking a physical exercise-induced release of myocardial injury biomarkers with detrimental cardiac effects is lacking. Here, with a detailed and non-invasive quantitative assessment in a real-life setting of male runners training for, participating in and recovering from their first marathon run, we found no indications of permanent or transient myocardial injury induced by vigorous physical exercise. Instead, the modestly elevated levels of troponin-T and NT-proBNP were accompanied by transient subclinical alterations in systolic and diastolic function. As such, our findings do not support the hypothesis that first-time marathon running in middle-aged runners has detrimental cardiac effects and do not explain the exercise-induced release of myocardial injury biomarkers. Rather, our imaging findings reinforce the notion that the transient and modest exercise-induced release of such biomarkers reflects physiological stress rather than myocardial injury.3 An early echocardiography study18 reported elevated troponin-T and NT-proBNP levels correlating with decreased RV and LV function, but such correlations have not been corroborated by others3 nor by the present work. An MRI study in recreational athletes19 found a small reduction in LV EDV similar to the present work and, likewise, could not establish any evidence of myocardial oedema, inflammation or fibrosis, nor relate any myocardial dysfunction with troponin-I or NT-proBNP levels. Using tagged MRI, we found a marked ~20% increase in systolic LV torsion directly post marathon. This finding aligns with previous observations in healthy male amateur runners studied under controlled laboratory conditions.20 Systolic function in terms of LV EF, LV GCS or circumferential shortening was not affected, thus preserving pumping performance post marathon with augmented myocardial wringing. Yet, an increase in TSR, a measure of transmural myocardial contractility differences,21 was not detected in our cohort. Thus, any (transient) subendocardial perfusion deficits and/or tissue damage that may arise from the prolonged LV pressure overload during marathon running appear unlikely. Indeed, quantitative myocardial mapping with MRI as well as LGE did not reveal any post-marathon subendocardial injury. The transient changes in diastolic filling patterns are also consistent with previous echocardiography studies18 20 and do not meet the criteria for qualification as diastolic dysfunction. Therefore, the augmented torsion during systole22 as well as the altered diastolic filling23 can likely be attributed to inotropic stimulation by catecholamines released during high-intensity endurance exercise,24 rather than reflecting detrimental functional impairment. Our observation that ECV decreases under the assumption of a constant haematocrit value seemingly contradicts the previous results of ECV increasing post marathon.14 With measurement time points and MR methodology and equipment being essentially identical for both studies, a potential explanation could be that these discrepancies may be mediated by cohort differences. The previous work8 focused on a smaller group of experienced marathon runners (n=11) of higher age (51 (50–56) years), where baseline ECV appeared to be higher (27.6±2.6%) than in our novice marathon runners in the present work (26.1 (24.8–27.2)%). This may suggest that the well-trained hearts already had a larger interstitial volume fraction that would readily expand25 during prolonged endurance exercise (ie, ECV increase). In contrast, cardiomyocyte swelling26 in untrained hearts would reduce the interstitial volume (ie, ECV decreases) on marathon running. These effects are not yet corroborated by concomitant changes in measured water diffusion parameters, and both studies were conducted in relatively small cohorts. As such, this hypothesis on the potentially differential impact of a single marathon run on the novice versus endurance-trained heart as well as any accumulative effects of endurance training and exercise needs further investigation. Limitations: the primary limitation of our study is the limited capacity for conducting all MRI examinations shortly after the marathon finish, resulting in a relatively small sample size of male runners. This practical challenge is common for imaging studies that aim to investigate acute effects of marathon running8 27 and has been mitigated previously by pooling data from multiple marathon events18 20 or by expanding the post-marathon time window to 24 hours28 up to 2 weeks.29 Parallel employment of three similar MR systems enabled us to examine the 17 finishers within a narrow time window of <10 hours after completion of a real-life marathon. This timing coincides with peak levels of circulating biomarkers of myocardial injury, aligning our measurements for detecting any such marathon running-induced injury. Despite our versatile range of sensitive imaging readouts and a fairly homogeneous cohort that delivered a solid endurance exercise effort, we could not establish any direct evidence of myocardial injury. A larger sample size would increase the power to detect small but potentially meaningful effects. Moreover, our cohort’s homogeneous nature and relatively high fitness level hamper generalisability of our results, for example, to less fit runners,30 or to younger or older runners. Notably, we excluded three subjects due to existing cardiovascular pathology that became apparent during this study’s extensive baseline evaluation. We cannot rule out that marathon running in the presence of existing (subclinical) cardiovascular pathology will induce detrimental cardiac effects. Subjects who were lost to follow-up (n=2) or dropped out for personal reasons (n=3) had a below-average VO2max and documented less prior physical activity (data not shown) than the 17 marathon finishers, suggesting a selection bias towards fitter subjects through the course of training for the marathon event. The exclusion of female participants is an important limitation to the present work. The current findings quantify the acute impact of marathon running in a well but narrowly defined cohort of middle-aged male runners, but generalisation to female runners is therefore limited. We note that although middle-aged male runners are at higher risk of endurance exercise-induced adverse cardiovascular events,6 7 similar investigations of female runners are warranted to unravel any sex-specific effects of marathon running. In conclusion, this exploratory study investigated the acute impact of marathon running on the heart in male novice marathon runners. Acute effects of finishing the marathon run were a <10% reduction in LV EDV in the absence of other morphological changes, and only small changes in systolic contractile mechanics and diastolic filling patterns, while circulating levels of biomarkers putatively signalling myocardial injury increased. Importantly, marathon running-induced effects were transient. The comprehensive test battery in our extensively phenotyped cohort offers no evidence to support the notion that first-time marathon running in healthy middle-aged men has a detrimental impact on the heart. Supplementary material Acknowledgements We thank Ted Ronteltap for cardiopulmonary exercise testing, Sandra van den Berg-Faay for study planning, blood sampling and MR acquisitions, Rianne H.A.C.M. de Bruin-Bon for echocardiography examinations and analyses, Arno A. van de Bovenkamp, Chantal J. Nieuwenhuizen and Myrte Wennen for their skilful assistance during the study weekends, and Kiki T.J. Geurkink for MRI volumetry analyses. Parts of this work were presented at the EAPC Annual Congress ESC Preventive Cardiology 2025 in Milan, Italy. Footnotes Funding: This work was supported by the Indonesia Endowment Fund for Education (LPDP, grant 201803220412678). Dr Jørstad is supported by an unrestricted grant from NOC*NSF (Nederlands Olympisch Comité - Nederlandse Sport Federatie) and Stichting Heart to Handle. Provenance and peer review: Not commissioned; externally peer reviewed. Patient consent for publication: Not applicable. Ethics approval: This study involves human participants and was approved by Amsterdam University Medical Center Ethics Committee (NL70800.029.19). Participants gave informed consent to participate in the study before taking part. Data availability free text: The data underlying this article will be shared on reasonable request to the corresponding author. Data availability statement Data are available upon reasonable request. References Associated Data Supplementary Materials Data Availability Statement Data are available upon reasonable request.