Journal Logo

Basic Sciences: Original Investigations

Short-term effects of marathon running: no evidence of cardiac dysfunction


Author Information
Medicine & Science in Sports & Exercise: October 1999 - Volume 31 - Issue 10 - p 1414
  • Free


Some controversy exists regarding the short-term effects of strenuous endurance exercise on the hearts of healthy individuals and trained endurance athletes. Whether endurance exercise could cause myocardial ischemia and/or necrosis is still a subject of debate. In this regard, although serum levels of some enzymes such as creatine kinase MB (CK MB) might increase after prolonged exertion (i.e., marathon running) (2,10,11,13,26,30) the exact muscle source of released intracellular components is often difficult to interpret. Moreover, a few reports have shown controversial results using more specific markers for detecting myocardial injury such as cardiac troponin T (TnT-c) or I (TnI-c) (2,7,10,11,19,21,22,25). While some authors have reported significant increases in TnT-c to pathological levels after ultramarathon running (21,25), others have only reported isolated increases in healthy trained athletes after vigorous exercise (marathon running or endurance cycling) (19), or even no changes during such an extreme endurance event as the Giro d’Italia for professional cyclists (7). Although most studies have evidenced no overall alterations in serum TnI-c in response to endurance exercise such as marathon (2,10) or altitude ultramarathon running (11,22), it must be stressed that elevated levels of TnI-c have been reported in a few samples (10).

In contrast, several investigators have suggested that cardiac dysfunction might occur after prolonged exercise such as ultramarathon running or long distance triathlons (13–15,24,26,27,31,34). Based on echocardiographic findings, previous reports have suggested that left ventricular dysfunction (i.e., transient reductions in systolic and diastolic performance) occurs during extreme exercise (13,26,27,34). Nevertheless, such a finding has not been corroborated in other studies (11,32). In addition, echocardiographic abnormalities might exist with no concomitant changes in serum markers of myocardial injury (11).

In the majority of the cases in which cardiac alterations have been reported, mostly well trained athletes were selected as subjects, i.e., ultraendurance athletes or triathletes able to complete extremely demanding events such as the Hawaii Ironman (11,13,14,26,27). The popularity of marathon running, however, has considerably increased in Western societies in the last two decades. Thousands of individuals of varying ages and fitness levels (finishing times ranging from ∼ 2 h 10 min to more than 5 h) participate in such events. In this regard, the question remains as to whether a demanding but accessible sport such as marathon running might have a deleterious effect in the hearts of healthy, not well trained runners (i.e., finishing time > 4 h). In addition, no research has assessed the effects of marathon running on both specific biochemical cardiac markers and echocardiographic parameters in healthy individuals.

The purpose of this investigation was to analyze the short-term effects of marathon running on both markers of cardiac damage and echocardiographic parameters in a group of runners with a wide range of fitness levels.



To be eligible for this investigation, subjects were required to meet the following criteria: 1) 20–45 yr of age; 2) in good health, as determined by a normal physical examination and routine laboratory tests within the previous year; 3) no history of cardiopulmonary disease; 4) cardiac examination (including 12-lead electrocardiogram and M-mode two-dimensional and Doppler echocardiogram) with no pathological results; and 5) no history of use of cardiovascular drugs.

Twenty-two marathoners (17 male and 5 female) were enrolled in the study. Values of subjects’ mean (± SD) age, height, and weight were the following: 34 ± 5 yr (range 23–44), 169.3 ± 9.4 cm, and 66.0 ± 9.2 kg. All of them completed the Madrid Marathon (1997). This race (42 km) is held annually in the month of April over a hilly circuit at an altitude of ∼ 600 m. The temperature ranged from 11° to 22°C, humidity from 60 to 65%. Informed consent was obtained from each subject in accordance with the regulations of the Complutense University.

Maximal Exercise Tests

Before the initiation of the study protocol, each subject performed an exercise test to determine maximal oxygen uptake (V̇O2max) with the use of an automated breath-by-breath system (CPX, Medical Graphics, St. Paul, MN). V̇O2max was determined by an incremental treadmill protocol to exhaustion (LE-6000, Erich Jaeger, Wuerzburg, Germany) in which running velocity was increased by 1 km·h−1 each min, starting at 8 km·h−1. Treadmill inclination was kept constant at 1.0%. All exercise tests were terminated voluntarily by the subjects or when established criteria of test termination were met (1). Heart rate (HR) was continuously monitored during the tests from modified 12-lead ECG tracings (EK56, Hellige, Freiburg, Germany).

Biochemical Markers

Venous blood was drawn by clean venipuncture (antecubital vein) from each subject five times during the study: 48 h before the race, immediately after the race, and 6, 24, and 48 h postcompetition. The subjects did not perform a hard training session within 48 h before the study protocol nor within 48 h after the marathon.

All blood samples were collected into sterile chilled tubes. The samples were allowed to clot at room temperature and then were centrifuged at 2,000 g for 20 min. Separated serum was aliquoted and stored at −20 or −80°C for the determination of myoglobin, total creatine kinase catalytic activity (total CK), mass concentration of creatine kinase isoenzyme MB (CK-MB mass), TnT-c, and TnI-c. Whole blood for determination of hemoglobin (Hb) and hematocrit (Hct) was also collected in EDTA tubes maintained at room temperature to correct relative changes in plasma volume by using Hb and Hct values from each test, according to the method described by Dill and Costill (12).

Total CK.

The serum was kept at −20°C until measured with a Boehringer Mannheim (Mannheim, Germany) Hitachi System 747–200; reference values of our laboratory for healthy adults are 0–190 U·L−1.


The serum was kept at −20°C until measured with an immunoturbidimetric assay (Tina-quant, Boehringer Mannheim). The reference values for men and women are 16–76 and 7–64 ng·mL−1, respectively, and the sensitivity of the test is 3 ng·mL−1.

CK-MB mass.

The serum was kept at −20°C until measured with an electrochemiluminiscence immunoassay (Elecsys, Boehringer Mannheim). The reference values are 0–5 ng·mL−1 and the sensitivity is 0.15 ng·mL−1.


The serum was stored at −20°C until measured with an electrochemiluminiscence immunoassay (Elecsys, Boehringer Mannheim). This is a rapid, single-step sandwich type assay (total duration of ∼ 9 min). The sensitivity of the test is 0.01 ng·mL−1 with a cut-off value of 0.1 ng·mL−1 for healthy individuals.


The serum was kept at −80°C until measured with a highly specific immunoenzymometric assay (ERIA, Sanofi Diagnostics Pasteur, Marnes-la-Coquette, France). The assay, performed in a total time of ∼ 30 min, is a two-site immunometric procedure using two different monoclonal antibodies both directed against human TnI-c which show no cross-reactivity with skeletal muscle TnI or other cardiac proteins. The sensitivity of the test is 0.03 ng·mL−1 with a cut-off value of 0.1 ng·mL−1 in healthy blood donors. Finally, the relative index of CK-MB mass concentration to total catalytic activity of CK (CK-MB mass/total CK) was calculated by dividing mass concentration in ng·mL−1 by total activity in U·L−1 and multiplying by 100. An upper reference value of 4.0 was used based on previous research (17).


Each subject underwent three echocardiographic studies: 1) at baseline, 2 to 5 d before the race; 2) within 30 min (mean ± SD 16 ± 7 min, range 6–28 min) after finishing the race; and 3) 24 to 36 h later. Transtorathic M-mode, two-dimensional, and Doppler echocardiographic examinations were performed with a Toshiba SSH-140A (Toshiba Medical Systems, Madrid, Spain) using a 2.5 MHz transducer. Athletes were placed in the left lateral position (45°) with the head slightly tilted. Imaging location and gain settings were adjusted to yield optimal definition of endocardial and epicardial borders. An electrocardiographic tracing was simultaneously displayed on the screen. All examinations were recorded on a Panasonic AG-7330 videocassette recorder for subsequent analysis over three to five high quality frozen frames. All measurements were performed by the same experienced investigator, which ensured that the subjects and the transducer positions were similar in all three studies. Two-dimensionally guided M-mode echocardiograms were obtained from the left parasternal long axis view, and left ventricular diastolic (LVDD) and systolic diameters (LVSD), posterior wall (PW), and septal (IVS) thickness were measured both at end diastole and end systole, according to the recommendations of the American Society of Echocardiography (33). Left ventricular end-diastolic (LVDV) and end-systolic (LVSV) volumes were calculated according to the formula of Teichholz et al. (35) and corrected for m2 of body surface area (BSA).

Left ventricular systolic function.

Left ventricular ejection fraction (LVEF) and fractional shortening (LVFS), and mean velocity of circumferential fiber shortening corrected for heart rate (mVCFc) were computed as parameters of LV systolic function (16). LVEF was calculated as 100·(LVDV − LVSV)/LVDV and fractional shortening (LVFS) as 100·(LVDD − LVSD)/LVDD. Mean velocity of circumferential fiber shortening (mVCF) was measured as 10·LVFS/LVET (where LVET is LV ejection time) and rate corrected (mVCFc) by ET/ √ RR. Thus, mVCFc = LVDD − LVSD/LVDD·(ET/ √ RR). Cuff sphygmomanometer blood pressure was measured simultaneously with performance of echocardiographic studies. Systolic blood pressure (Ps) was combined with LVSD and end systolic posterior wall thickness (PWs) to provide an index of meridional end-systolic wall stress (ςm = 1.33·Ps·LVSD/4·PWs·(1 + PWs/LVSD)) (18). In contrast, the relationship existing between ςm and mVCFc in each of the three echocardiographic evaluations was assessed by linear regression analysis. This relationship is thought to be a sensitive index of left ventricular inotropic state and independent from loading conditions (9).

Left ventricular diastolic function.

Two-dimensionally guided pulsed Doppler recordings of left ventricular inflow at the level of the mitral annulus were obtained from the apical four chamber view to assess left ventricular filling dynamics. For each Doppler profile, peak velocities (centimeter per second) of left ventricular inflow in early (E) and late (A) diastole, and the ratio of early to late (E/A) diastolic flow velocity, were calculated. The isovolumic relaxation period (IRP, in milliseconds) was measured by pulsed Doppler from the end of the aortic flow to the start of the mitral flow. IRP corrected for HR (IRP/HR) was also computed.

Three to five beats were analyzed and averaged for M-mode, two-dimensional, and Doppler measurements. Before the study protocol, intraobserver reproducibility of selected M-mode echocardiographic measurements was assessed. A total of 30 subjects that were not selected for this investigation were evaluated on two different occasions. Comparison of these measurements (including wall thickness and cavity dimensions) showed no significant difference between repeated measurements (P < 0.05) and close correlations (r > 0.81).


A one-way repeated-measures analysis of variance (ANOVA) was used to determine if there existed a significant difference in the totality of subjects between mean values of the different biochemical and echocardiographic parameters been evaluated before and after the race, respectively. As previously mentioned, the relationships between ςm and mVCFc were assessed by linear regression analysis. Each regression equation was compared with each other for parallelism and a common intercept with the large sample Z test. Finally, ANCOVA with HR as the covariate and E/A as the dependent variable was applied to assess the effect of exercise-induced variations of HR on E/A data. This test allows to test the significance of adjusted mean differences among a group of subjects (i.e., E/A before the race vs E/A immediately after the race, etc.) after removing variation in the results caused by unwanted factors (covariates).

The level of significance was set at 0.05 for all statistical analyses. All data are expressed as means ± SD.


Maximal Tests and Racing Performance

The mean V̇O2max value of the subjects was 55.7 ± 9.1 mL·kg−1·min−1 (range: 42.6–68.4). The performance time in the marathon was lower than 3 h in seven subjects (32% of total), between 3 and 4 h in nine subjects (41% of total), and higher than 4 h in six subjects (27% of total).

Biochemical Markers

Markers of skeletal muscle damage.

Mean levels of total CK were below reference limits before the marathon and increased to reach values above such limits in all postrace samples (Fig. 1). Pre-exercise values were significantly lower than those measured at 6, 24, or 48 h postexercise (P < 0.01, P < 0.01, and P < 0.05, respectively). The results obtained immediately after the race were also lower than those corresponding to 6 or 24 h of recovery (P < 0.05). Mean values of myoglobin, on the other hand, also ranged within normal limits before the race, and they were above reference levels in the samples collected 0, 6, and 24 h after the marathon (Fig. 1). Significant differences (P < 0.01) existed between prerace samples and those collected at 0 or 6 h postexercise. However, both mean values obtained at race finish or after 6 h significantly decreased (P < 0.01) after 24 or 48 h of recovery.

Figure 1
Figure 1:
Markers of skeletal muscle damage: total creatine kinase catalytic activity (total CK) and myoglobin. Values are means ± SD, and dotted lines represent normal upper limits. * Pre-exercise vs 6, 24, and 48 h postexercise (P < 0.01, P < 0.01, and P < 0.05, respectively). ** 0 h postexercise vs 6 and 24 h postexercise, P < 0.01. § pre-exercise vs 0 and 6 h postexercise, P < 0.01. §§ 0 and 6 h postexercise vs 24 and 48 h postexercise (P < 0.01).

CK-MB mass and CK-MB mass/total CK index.

Average mass concentration of CK-MB was within normal limits before the marathon (Fig. 2), although significant increases were found in all the evaluations performed after the competition in comparison with resting conditions (P < 0.01 for 6 and 24 h postexercise, and P < 0.05 for 48 h postexercise). Significant differences also existed between 0 h postexercise vs both 6 and 24 h postexercise (P < 0.01). In contrast, mean values of the CK-MB mass/total CK index were consistently within normal limits (<4.0) (17). No significant differences existed during the study protocol, except for a lower value in the last samples in comparison with those collected 0 and 6 h postrace (P < 0.05). Finally, individual values were above reference values only: 1) in all five samples of one male subject; and 2) in the sample collected at 6 h postexercise in another male subject.

Figure 2
Figure 2:
Mass concentration of creatine kinase isoenzyme MB (CK-MB mass) and CK-MB mass/total creatine kinase index (CK-MB mass/total CK). Values are means ± SD and dot lines represent normal upper limits. * Pre-exercise vs 6, 24, and 48 h postexercise (P < 0.01, P < 0.01, and P < 0.05, respectively). ** 0 h postexercise vs 6 and 24 h postexercise, P < 0.01. § 48 h postexercise vs 0 and 6 h postexercise, P < 0.05.

TnT-c and TnI-c.

All subjects had no detectable (<0.01 ng·mL−1) serum TnT-c before running the marathon (Fig. 3). Except in one male subject, each individual value measured after the race was below the commonly accepted level of 0.1 ng·mL−1 indicative of myocardial injury (5). Concerning the aforementioned subject, elevated levels were found at 0, 6, and 48 h postrace. However, His CK-MB mass/CK index was consistently within normal limits. Significant differences (P < 0.05) existed between means only when comparing the samples collected at race finish with those collected before the race or after 24 and 48 h of recovery. In contrast, all subjects had no detectable (<0.03 ng/mL) TnI-c before or after the race (Fig. 3) except for the aforementioned male subject. This subject presented elevated levels of TnI-c (slightly above 0.1 ng·mL−1) only in the last sample. Finally, no significant differences existed between means.

Figure 3
Figure 3:
Markers of myocardial damage: cardiac troponin T (TnT-c) and I (TnI-c). Values are means ± SD and dot lines represent normal upper limits. * 0 h postexercise vs preexercise and vs 24 and 48 h postexercise, respectively.P < 0.05. No significant differences existed for TnI-c levels.

Echocardiographic Parameters

Results are shown in Table 1.

Table 1
Table 1:
Echocardiographic results.

Heart rate and blood pressure.

Heart rates were significantly higher (P < 0.01) after the race when compared with pre-exercise or recovery conditions. In contrast, systolic blood pressure was considerably reduced at race finish (P < 0.01). No significant change was observed in diastolic blood pressure.

Left ventricular diameters and volumes.

LVSD and LVDD, LVSV and LVDV significantly decreased (P < 0.01) after the marathon in comparison with the other two conditions.

Left ventricular systolic function.

LVEF and LVFS significantly increased at race finish in comparison with pre-exercise conditions (P < 0.01 and P < 0.05, respectively). On the other hand, mVCFc was not altered during the study, whereas average value of ςm was significantly lower at race finish than in the other two conditions (P < 0.01). Finally, the relationships between ςm and mVCFc in each of the three echocardiographic evaluations are shown in Figure 4. No significant differences were found between the three conditions (P > 0.05).

Figure 4
Figure 4:
Regression lines for meridional end-systolic wall stress (ςm) plotted against mean velocity of circumferential fiber shortening corrected for heart rate (mVCFc) on prerace, finish, and recovery (24–36 h) recordings (top): no significant differences existed among the three conditions in the slope of relationship or in the y-intercept (P > 0.05). The regression lines corresponding to pre-exercise and 0 h postexercise are shown (bottom) together with individual data points and 95% confidence intervals of pre-exercise regression line (dotted lines).

Left ventricular diastolic function.

In comparison with the evaluations performed before the marathon or after 24–36 h of recovery, E and A exhibited lower and higher values, respectively, at race finish (P < 0.01). E/A was significantly reduced after the marathon (P < 0.01) and returned to baseline levels after 24–36 h of recovery. The results of the ANCOVA test showed that such response persisted (P < 0.01) after removing variation because of changes in HR with exercise. Finally, no changes were found in IRP during the study although IRP/HR was significantly reduced (P < 0.01) after competition in comparison with both pre-exercise and recovery conditions.

Echocardiographic parameters were within normal limits in the subject with elevated levels of TnT-c or TnI-c after the race.


To date, no research has assessed the effects of marathon running on both specific biochemical cardiac markers and echocardiographic parameters in healthy individuals. To our knowledge, only one study has used such approach to analyze the short-term effects of extraordinary endurance exercise (163 km altitude ultramarathon over a hilly course) on the human heart (11). Although no physiological data of the subjects (such as V̇O2max) were documented, there is little doubt that only experienced, highly trained subjects are likely to participate in this type of event. In fact, only 14 of a total of 23 subjects selected for the study were able to cover the distance. In addition, the reduced total number of runners (N = 83) participating in the race showed that this type of exercise is limited only to a minority of athletes with a solid background in endurance training and competition. Therefore, it could not be inferred from their study whether more accessible endurance events have a deleterious effect in the heart of average runners. In contrast, thousands of runners with a wide variety of fitness levels (performances ranging from less than 2 h 30 min to more than 5 h) participate each year in city marathons, as it is the case of the one chosen for our investigation (∼ 5000 participants in 1997, of which more than 90% were able to cover the total distance).

Nonspecific Biochemical Markers

Mean levels of total CK and myoglobin were below reference limits before the marathon and significantly increased after exertion to reach values above reference limits. In this regard, there is little question that raised levels of these two proteins result from muscle trauma, especially after exercises largely involving excentric contractions (10). Our results, indeed, are in agreement with previous studies conducted with marathon runners (10,26,30). Mass concentration of CK-MB, on the other hand, was within normal ranges before the marathon, although mean levels significantly increased in the samples collected after the competition. Furthermore, reference levels were surpassed in postrace samples. Our results confirm those of previous studies that have suggested that, after severe physical exertion, elevation of CK-MB mass may occur in healthy individuals not only above reference limits (4) but also into the range suggestive of myocardial damage (10,29). Nevertheless, studies using myocardial imaging techniques suggest that serum CK-MB elevations probably do not reflect cardiac damage but rather derive from skeletal muscle (29). Indeed, the skeletal muscles of trained athletes might contain elevated CK-MB levels related to the degree of training and stress (10). Therefore, other more specific markers (such as cardiac troponin) should be used as actual indicators of myocardial injury in healthy runners.

On the other hand, mean values of the CK-MB mass/total CK index were consistently within normal limits (<4.0). In addition, individual values were above reference values only in some postrace samples. Our data are in line with those reported by Ordóñez-Llanos et al. (30) and confirm the utility of the CK-MB mass/total CK ratio in evaluating the source of high CK-MB levels associated to skeletal muscle damage.

Cardiac Troponin

Some authors have recently suggested that intense endurance exercise may induce subclinical myocardial injury in healthy athletes (21,25). Such an assumption, however, was only based on the observation of elevated TnT-c levels after a marathon (21) or ultramarathon (25) without any additional evidence of heart damage (i.e., echocardiographic parameters). Moreover, the findings reported by Laslett et al. (21) were possibly caused by the cross-reactivity of the “first generation” of the TnT-c assay that was used by the authors. Indeed, Mair et al. (22) found increases in TnT-c after an Alpine marathon (67 km) measured with the first generation assay which were not corroborated when using either the “second generation” TnT-c assay or TnI-c as a biochemical marker for cardiac injury. Furthermore, using the second-generation assay, Bonetti et al. (7) found no single value of serum TnT-c above cut-off limits (0.1 ng·mL−1) in a total of 64 samples collected from 25 professional cyclists during such an extreme endurance event as the Giro d’Italia (22 consecutive competition days during which a total distance of 3.740 km is covered). On the other hand, another confounding factor comes from the fact that TnT-c, a protein which is suppressed in healthy adult skeletal muscle (3), could be re-expressed in response to skeletal injury (2), as might the case in endurance events largely involving excentric contractions such as marathon or ultramarathon running. Therefore, such a confounding factor could have also explained the elevated values of TnT-c reported by some authors either in a large number of subjects (21,25) or in isolated cases (19), as well as those increases found in the postrace samples of one subject in our investigation. In this regard, the use of TnI-c has allowed us to accurately determine whether cardiac tissue damage existed after the marathon. TnI-c, indeed, is an isoform of troponin I not found in skeletal muscle (even during development) that has been shown to possess extremely high specificity for the detection of cardiac injury (2,20). Some authors have also analyzed the effects of strenuous exercise on serum levels of TnI-c, showing no increases in response to such demanding endurance exercise as marathon (2,10) or altitude ultramarathon running (11,22). Nevertheless, some controversy also arises from these studies with comparable results. Indeed, elevated levels of TnI-c have been reported in few postexercise samples in the studies by Cummings et al. (1 subject of a total of 11) and by Dávila-Román et al. (1 subject of a total of 23) (10,11). Our results are in agreement with previous research, since increases above cut-off values were only found in one of a total of 88 postrace samples (<1.0%). Moreover, in the only sample in which elevated levels of TnI-c levels were found, the obtained value was slightly higher than the reference upper limit, and far below those levels (2–3 ng·mL−1) indicative of myocardial injury (22).

Since TnI-c is an ideal heart-specific cell marker that is uniformly distributed throughout atrial and ventricular chambers (10), it can be concluded that marathon running does not induce myocardial injury in healthy subjects with a wide range of fitness levels.

Echocardiographic Parameters

Our results showed no overall decrease in LV contractility after exercise. Although LVEF and LVFS actually increased at race finish, other indicators such as mVCFc or the stress-shortening relationship were not altered. On the other hand, the changes observed in diastolic function at the end of the marathon were reversed after 24–36 h of recovery. Several studies have presented evidence of reduced LV function after prolonged exercise (6,8,13,14,24,26,27,31,32,36). In other reports, however, several echocardiographic indicators of LV function were not altered after endurance exercise (11,32). In most of the studies showing echocardiographic alterations, exercise was extraordinarily demanding. In the majority of the cases, both duration of exercise (i.e., 24 h run or 12 h triathlon) and/or environmental conditions (i.e., 24° and 42°C with a 40–85% humidity in the Hawaii Ironman) were extreme (13,15,26). In this regard, it could be hypothesized that LV function might be transiently altered mostly after ultraendurance events, especially when held under severe climatic conditions. In these cases, electrolyte disturbances and/or severe loss of plasma volume, combined with the extraordinary length of exercise might compromise LV function. In fact, Dávila-Román et al. (11) found no alteration in the left ventricle of runners after completing an ultraendurance alpine marathon under more favorable weather conditions (2°−24°C, 10–60% of humidity).

However, the information concerning the effects of less demanding endurance events such as marathon races on echocardiographic parameters is somewhat controversial. Although research has reported altered LV function after actual marathon running (6,8) or after a 170 min treadmill run (34), Perrault et al. (32) reported no evidence of LV performance impairment after a marathon in a group of moderately to well trained runners (finishing time 2 h 19 min–3 h 46 min). In their study, fractional shortening was unchanged despite decreased wall stress, suggesting that the duration and/or intensity of exercise was not sufficient to induce significant cardiac function impairment. In line with this report, our findings suggest that marathon running per se does not negatively affect LV systolic function in runners of different fitness levels, at least when environmental conditions are not extreme (in our study the temperature was ∼ 17°C and the mean loss of plasma volume of the subjects was below 7%). Moreover, LVEF and LVFS were increased at the completion of the race. In this regard, we also examined the stress-shortening relation. According to Colan et al. (9), this relationship is an indicator of the myocardial inotropic state which incorporates afterload and heart rate, and which is independent on preload over the physiological range. In contrast with previous research in which the aforementioned relationship was reported to be significantly altered at race finish (13,36), we found no significant differences among the three evaluations. Such a finding suggests that, when considering mVCFc (LV contractility) independent from loading conditions, LV inotropic state was not significantly altered in our subjects. The increase in LVEF and LVFS at race finish could be partly attributed to the higher heart rates at the moment of the second echocardiographic evaluation in comparison with the other two conditions. Increased heart rate, indeed, is known to improve overall left ventricular performance, presumably through a rate-dependent increase in left ventricular calcium availability (23,36), although the exact mechanism underlying this effect remains to be clarified (36). In addition, such improvement in systolic performance could have been mediated by an exercise-induced increase in sympathetic-adrenal activity. Although we did not measure circulating catecholamines, it has been documented that the serum levels of epinephrine and norepinephrine significantly increase during a marathon in runners of different performance levels (28).

In contrast, an inverse relationship exists between plasma catecholamines and the mileage run (37). Thus, the latter phenomenon might have played a role in the altered LV systolic function reported by some authors after ultraendurance exercise. However, some studies suggest that prolonged exposure to catecholamines during exercise might trigger a downregulation of β-adrenoreceptors which in turn might be responsible for an acute decrease in left ventricular contractility after endurance exercise (36). In this regard, it could be hypothesized that the duration of the exercise protocol chosen for our investigation (∼ 2 h 20 min to 5 h) was not sufficient to produce such effect. Conversely, and in agreement with previous research (13,14,33), analysis of left ventricular filling pattern showed a decrease in the ratio of early-to-late flow velocities after the race, independent from exercise-induced increase in heart rate. In any case, such alteration in LV diastolic properties was transient, since mean values returned to baseline levels after recovery.


In summary, our investigation represents the first attempt to assess the effects of marathon running on both biochemical and echocardiographic indicators of cardiac function. Both the protocol and the subject population chosen for our study lead us to conclude that marathon running per se (with no extreme environmental conditions) does not adversely affect the hearts of healthy individuals independently from their training status. Therefore, such type of exercise can be safely practiced by most healthy individuals even with a low fitness level after an adequate training program.

This study was partially supported by a grant from DGICYT (PM 95–0190). The authors acknowledge Luis San José from Toshiba Medical Systems (Madrid, Spain) for his technical assistance. We are also grateful to Pedro Cuesta (Departamento de Apoyo a la Investigación, Centro de Proceso de Datos de la Universidad Complutense de Madrid) for his assistance in statistical analysis.


1. American College of Sports Medicine. Guidelines for Exercise Testing and Prescription. Philadelphia, PA: Lea & Febiger, 1986, pp. 95–96.
2. Adams, J. E., G. S. Bodor, V. G. Dávila-Román et al. Cardiac troponin I: a marker with high specificity for cardiac injury. Circulation 88:101–106, 1993.
3. Anderson, P. A. W., N. N. Malouf, A. E. Oakeley, E. D. Pagani, and P. D. Allen. Troponin T isoform expression in humans: a comparison among normal and failing adult heart, fetal heart, and adult and fetal skeletal muscle. Circulation 69:1226–1233, 1991.
4. Apple, F. S., M. A. Rogers, D. C. Casal, L. Lewis, J. L. Ivy, and J. W. Lampe. Skeletal muscle creatine kinase mb mass measurement/total CK activity ratio. Clin. Chem. 32:291–295, 1986.
5. Bakker, A. J., J. P. M. C. Gorgels, B. van Vlies, F. D. M. Haagen, and R. Smits. The mass concentrations of serum troponin T and creatine kinase-MB are elevated before creatine kinase and creatin kinase-MB activities in acute myocardial infarction. Eur. J. Clin. Chem. Clin. Biochem. 31:715–724, 1993.
6. Boynton, M., E. Winslow, R. Ling, and D. Mehlman. Are echocardiographic estimates of left ventricular function altered by running a marathon? (Abstract). Med. Sci. Sports Exerc. 19:S84, 1987.
7. Bonetti, A., F. Tirelli, R. Albertini, C. Monica, M. Monica, and G. Tredici. Serum cardiac troponin T after repeated endurance exercise events. Int. J. Sports Med. 17:259–262, 1996.
8. Chan, K., A. Pipe, and M. Barrie. Acute effect of marathon running on left ventricular function (Abstract). Med. Sci. Sports Exerc. 19:S83, 1987.
9. Colan, S., K. Borow, and A. Neumann. The left ventricular end-systolic wall stress-velocity of fiber shortening relation: a load independent index of myocardial contractility. J. Am. Coll. Cardiol. 4:715–724, 1984.
10. Cummins, B., M. Auckland, and P. Cummins. Cardiac-specific troponin-I radioimmunoassay in the diagnosis of acute myocardial infarction. Am. Heart J. 113:1333–1344, 1987.
11. Dávila-Román, V. G., T. M. Guest, P. G. Tuteur, W. J. Rowe, J. H. Ladenson, and A. S. Jaffe. Transient right but not left ventricular dysfunction after strenuous exercise at high altitude. J. Am. Coll. Cardiol. 30:468–473, 1997.
12. Dill, D. B. and D. L. Costill. Calculation of percentage change in volumes of blood, plasma, and red cells in rehydration. J. Appl. Physiol. 37:247–248, 1979.
13. Douglas, P. S., M. L. O’Toole, D. B. Hiller, K. Hackney, and N. Reichek. Cardiac fatigue after prolonged exercise. Circulation 76:1206–1213, 1987.
14. Douglas, P. S., M. L. O’Toole, and J. Woolard. Regional wall motion abnormalities after prolonged exercise in the normal left ventricle. Circulation 82:2108–2114, 1990.
15. Douglas, P. S., M. L. O’Toole, W. D. B. Hiller, and N. Reichek. Different effects of prolonged exercise on the right and left ventricles. J. Am. Coll. Cardiol. 15:64–69, 1990.
16. Ehsani, A. A., W. H. I. Martin, G. W. Heath, and E. F. Coyle. Cardiac effects of prolonged and intense exercise training in patients with coronary artery disease. Am. J. Cardiol. 50:246–251, 1982.
17. El Allaf, M., J. P. Chapelle, D. El Allaf, et al. Differentiating muscle damage from myocardial injury by means of the serum creatine kinase (CK) isoenzyme MB mass measurement/total CK activity ratio. Clin. Chem. 32:291–295, 1986.
18. Grossman, W., D. Jones, and M. P. McLaurin. Wall stress and pattern of hypertrophy in the human left ventricle. J. Clin. Invest. 56:56–64, 1975.
19. Koller, A., J. Mair, M. Mair, C. Calzolari, C. Larue, and B. Puschendorf. Diagnosis of myocardial injury in marathon runners. Ann. N. Y. Acad. Sci. 752:234–235, 1995.
20. Larue, C., C. Calzolari, J. P. Bertinchant, F. Leclercq, R. Grolleau, and B. Pau. Cardiac-specific immunoenzymometric assay of troponin I in the early phase of acute myocardial infarction. Clin. Chem. 39:972–979, 1993.
21. Laslett, L., E. Eisenbud, and R. Lind. Evidence of myocardial injury during prolonged strenuous exercise. Am. J. Cardiol. 78:488–490, 1996.
22. Mair, J., W. Schobersberger, A. Koller, et al. Risk for exercise-induced myocardial injury for athletes performing prolonged strenuous endurance exercise (Letter). Am. J. Cardiol. 80:543–544, 1997.
23. Maughan, L., K. Sunagawa, D. Surkhoff, W. Graves, W. Hunter, and K. Sagawa. Effect of heart rate on the canine end-systolic pressure-volume relationship. Circulation 72:454–459, 1985.
24. McKechnie, J. K., W. P. Leary, T. D. Noakes, J. C. Kallmeyer, E. T. M. MacSearraigh, and L. R. Olivier. Acute pulmonary oedema in two athletes during a 90-km running race. S. Afr. Med. J. 56:261–265, 1979.
25. Musha, H., J. Nagashima, T. Awaya, K. Oomiya, H. Takada, and M. Murayama. Myocardial injury in a 100-km ultramarathon. Curr. Ther. Res. Clin. E. 58:587–593, 1997.
26. Niemelä, K. O., I. J. Palatsi, M. J. Ikheimo, J. T. Takkunen, and J. Vuori. Evidence of impaired left ventricular performance after an uninterrupted competitive 24 hour run. Circulation 70:350–356, 1984.
27. Niemelä, K. O., I. Palatsi, M. Kaheimo, J. Airaksinen, and J. Takkunen. Impaired left ventricular diastolic function in athletes after utterly strenuous prolonged exercise. Int. J. Sports Med. 8:61–65, 1987.
28. O’Brien, M. J., C. A. Viguie, R. S. Mazzeo, and G. Brooks. Carbohydrate dependence during marathon running. Med. Sci. Sports Exerc. 25:1009–1017, 1993.
29. Ohman, E. M., K. K. Teo, A. H. Johnson, et al. Abnormal cardiac enzyme responses after strenuous exercise: alternative diagnostic aids. Br. Med. J. 285:1523–1526, 1982.
30. Ordóñez-Llanos, J., J. R. Serra-Grima, J. Mercé-Muntañola, and F. González-Sastre. Ratio of creatine kinase 2 mass concentration to total creatine kinase activity not altered by heavy physical exercise. Clin. Chem. 38:224–2227, 1992.
31. Osbakken, M. and R. Locko. Scintigraphic determination of ventricular function and coronary perfusion in long-distance runners. Am. Heart J. 108:296–304, 1984.
32. Perrault, H., F. Péronnet, R. Lebeau, and R. A. Nadeau. Echocardiographic assessment of left ventricular performance before and after marathon running. Am. Heart J. 112:1026–1031, 1986.
33. Sahn, D. J., A. Demaria, J. Kisslo, and A. Weyman. Recommendations regarding quantitation in m-mode echocardiography: results of a survey of echocardiographic measurements. Circulation 58:1072–1083, 1978.
34. Seals, D. R., M. A. Rogers, J. M. Hagberg, C. Yamamoto, P. E. Cryer, and A. A. Ehsani. Left ventricular dysfunction after prolonged strenuous exercise in healthy subjects. Am. J. Cardiol. 61:875–879, 1988.
35. Teicholz, L. E., T. Kreulen, M. V. Herman, and R. Gorlin. Problems in echocardiographic volume determinations: echocardiographic-angiographic correlations in the presence or absence of asynergy. Am. J. Cardiol. 37:7–11, 1976.
36. Vanoverschelde, J-L. J., L. T. Younis, J. A. Melin, et al. Prolonged exercise induces left ventricular dysfunction in healthy subjects. J. Appl. Physiol. 70:1356–1363, 1991.
37. Ward, M. M., I. N. Mefford, G. W. Black, and W. M. Bortz. Exercise and plasma catecholamine release. In: Exercise Endocrinology. K. Fotherby and S. B. Pal (Eds.). Berlin: Walter de Gruyter & Co., 1985, pp. 263–294.


© 1999 Lippincott Williams & Wilkins, Inc.