Author + information
- Received December 24, 1996
- Revision received March 26, 1997
- Accepted April 16, 1997
- Published online August 1, 1997.
- Víctor G Dávila-Román, MD, FACCA,*,
- Thomas M Guest, MDABCD,
- Peter G Tuteur, MDABCD,
- William J Rowe, MDABCD,
- Jack H Ladenson, PhDABCD and
- Allan S Jaffe, MD, FACCE
- ↵*Dr. Vı́ctor G. Dávila-Román, Cardiovascular Division, Box 8086, Washington University, 660 South Euclid Avenue, Saint Louis, Missouri 63110.
Objectives. We sought to evaluate whether prolonged exercise in ultramarathon runners results in left ventricular (LV) damage.
Background. Strenuous exercise has been reported to cause LV damage.
Methods. Fourteen runners who completed an ultramarathon at high altitude underwent echocardiography, finger-tip oximetry and blood measurements of cardiac troponin I (cTnI) and creatine kinase, MB fraction (CK-MB) levels before, immediately after and 1 day after the race.
Results. At baseline, the echocardiograms showed normal LV and right ventricular (RV) size and function in all subjects, as well as mild tricuspid regurgitation in nine subjects, with normal estimated pulmonary artery systolic pressure (mean 28 mm Hg). At baseline, all oximetric readings and CK-MB measurements were normal, and cTnI was undetectable. Immediately after the race, the echocardiograms showed the expected augmentation of global and segmental LV function in all subjects. Although the RV was normal in nine subjects, five developed marked RV dilation and hypokinesia, paradoxic septal motion, pulmonary hypertension and wheezing. CK-MB values were elevated in all subjects. In all but one subject cTnI was undetectable. In that subject there was a small elevation in cTnI accompanied by severe RV dysfunction and pulmonary hypertension. At the 1-day follow-up study, the echocardiographic measurements had normalized in all subjects.
Conclusions. In trained athletes, strenuous exercise at high altitude did not result in LV damage. However, wheezing, reversible pulmonary hypertension and RV dysfunction occurred in a third of those completing the race. The incidence and pathogenesis of these findings remain to be determined.
The short-term effects of strenuous exercise on the hearts of healthy individuals and trained endurance athletes are a subject of debate. Myocardial dysfunction thought to be due to cardiac damage has been reported in several studies after strenuous or prolonged exercise [1–6]. In these studies the levels of the MB isoenzyme of creatine kinase (CK-MB) or the potential nonspecific effects on left ventricular (LV) function of exercise, as assessed by echocardiography, were evaluated. However, elevations in CK-MB can reflect the skeletal muscle damage that may accompany extreme exertion , and LV regional wall motion can be altered in the absence of cardiac damage by alterations in ventricular loading, compliance or interventricular dependence, alone or in combination. Thus, the findings of previous studies that have used one or both of these techniques to assess myocardial damage are difficult to interpret because of a lack of specificity of these tests.
Cardiac troponin I (cTnI) is a regulatory cardiac protein that has been shown to have extremely high specificity for the detection of cardiac injury [7, 8]. Because cTnI is not found in skeletal muscle, even during development , elevated levels in the blood do not occur in response to short- or long-term skeletal muscle injury, unless there is concomitant cardiac muscle injury. Furthermore, the sensitivity of cTnI for the detection of cardiac damage is comparable to that of CK-MB [7–10]. Accordingly, we used measurements of cTnI conjointly with echocardiography to assess cardiac function in endurance athletes after ultraexercise at a high altitude in an attempt to determine the effects of strenuous exercise on the heart.
Twenty-three of the 81 runners participating in the 1994 Hardrock-100 race were enrolled in the study. This race is a 163-km, high altitude mountain run (with elevation changes ranging from 2,350 to 4,300 m) held annually in Silverton, Colorado during the month of July. Written informed consent, in a format approved by the Human Studies Committee at the Washington University School of Medicine, was obtained from all subjects. The mean (±SD) age of the study participants was 43 ± 8 years (range 27 to 56); there were 22 men and one woman. Of the 23 runners enrolled, the 14 who completed the race are the subjects of this report. All subjects were experienced endurance athletes who had participated in at least two ultramarathon races during the preceding 12 months. During the race, temperatures ranged from 2° to 24°C, and humidity from 10% to 60%. All athletes were allowed to drink fluids as needed.
Before the race, baseline two-dimensional and Doppler echocardiograms were recorded, finger-tip oximetric readings were made and blood samples were drawn for measurements of cTnI and CK-MB. In the 14 subjects who finished the race, all tests were repeated immediately after the race and all but pulse oximetry were repeated 18 to 24 h later.
Transthoracic echocardiography (HP Sonos 1500) was performed in the four standard views in all subjects on three occasions: 1) at baseline (the day before the race, after refraining from strenuous exercise for 24 h); 2) within 20 min (mean 8 ± 2, range 5 to 20) after finishing the race; and 3) 18 to 24 h later. The images were stored on ½-in. (1.27-cm) videotape (S-VHS) and digitized in a cine loop, quad screen format (TomTec Imaging Systems) for subsequent side by side comparison. All studies were performed by an experienced cardiac sonographer, which ensured that the subjects and the transducer position were similar in all three studies. In all instances, optimal visualization of the LV and right ventricular (RV) endocardial borders and cardiac chambers was sought . The apical four-chamber views were digitized at end-diastole and end-systole by use of five consecutive beats at sustained end-expiration to obtain the RV and LV cavity areas and volumes. The fractional area change (FAC) was calculated in all beats as follows: FAC = [(End-diastolic area − End-systolic area)/End-diastolic area] × 100. The LV volumes and ejection fraction were calculated by the method of summation of disks (modified Simpson rule) from the apical four-chamber view. The global and segmental wall motion was evaluated independently by two experienced echocardiographers (who had no knowledge of the clinical and biochemical data) using a semiquantitative scoring system with a 16-segment model .
Color flow, pulsed and continuous wave Doppler recordings were obtained from the apical four-chamber view at the level of the tricuspid valve. Color flow studies were performed at the same gain and color-coding settings . The pulmonary artery systolic pressure (PASP, in mm Hg) was calculated from the pressure gradient between the RV and right atrium with continuous wave Doppler echocardiography across the tricuspid valve as follows: PASP = 4 (V)2+ 10, where V is the peak velocity (in m/s) of the tricuspid valve regurgitant jet, and 10 mm Hg is the estimated right atrial pressure for our normal group of subjects .
1.3 Pulse oximetry.
Oxygen saturation was determined after 5 min of continuous recordings by finger-tip oximetric measurements with subjects in the supine position.
1.4 Evaluation of molecular markers.
Blood was drawn into tubes containing no preservatives and centrifuged at 2,000gfor 15 min. The serum samples were stored at −70°C, thawed once and assayed [15–17]. Assays for CK-MB and cTnI were performed, and judgments concerning whether levels were normal or abnormal were made by individuals unaware of the echocardiographic findings. The concentration of CK-MB (upper reference limit ≥6.7 ng/ml, lower limit of detection 2.2 ng/ml) was measured with a commercially available immunoassay (Stratus creatine kinase-MB, Baxter Dade) .
The cTnI concentration was measured by an immunoassay on the Baxter Stratus analyzer in a preliminary research application that uses two cTnI-specific monoclonal antibodies and has no detectable cross reactivity with skeletal muscle troponin I [8, 15]. The lower limit of detection of cTnI is 1.5 mg/ml; the upper reference limit is >3.1 ng/ml (95% cutoff value by nonparametric analysis) [7, 10].
All results are expressed as the mean value ± SD. Comparisons of measurements obtained before the race to those obtained immediately after the race and at 18 to 24 h after the race were performed by repeated measures analysis of variance or by the paired ttest. Comparisons between the groups were made by the unpaired ttest. Statistical significance was determined at p < 0.05.
2.1 Baseline results.
cTnI was not detected in any of the subjects. All CK-MB measurements (mean 2.0 ± 1.2 ng/ml, range 0.6 to 4.8) and all oximetric readings (mean 94 ± 1%, range 92% to 97%) were within reference limits. All echocardiograms showed normal LV and RV size and function (Table 1). Doppler echocardiograms showed mild tricuspid regurgitation in nine subjects and no abnormalities in the rest. The mean tricuspid regurgitant jet was 2.1 ± 0.1 m/s (range 1.8 to 2.6); the estimated mean PASP was 28 ± 2 mm Hg (range 25 to 30).
2.2 Immediate postrace results.
For the group as a whole, heart rate was significantly increased and diastolic blood pressure was significantly decreased from baseline. Systolic blood pressure was unchanged (Table 2). Although CK-MB was elevated in all finishers (mean 65 ± 33 ng/ml, range 16 to 130), cTnI remained undetectable in all but one subject, and this elevation was small (5 ng/ml) (Fig. 1).
All echocardiograms showed the expected augmentation in global LV function (i.e., a significant decrease in the LV end-systolic and end-diastolic areas and volumes, and no change in the ejection fraction). The RV was normal in nine subjects. However, five subjects developed marked RV dilation and global RV hypokinesia, which was corroborated by a significant increase in the RV end-systolic areas and end-diastolic areas, and a significant decrease in the RV FAC (40.4 ± 4.8% vs. 33.2 ± 3.8%, baseline vs. immediate postrace for the five subjects, respectively, p = 0.009). The RV echocardiographic measurements before and after the race are listed in Table 3. All five subjects also developed paradoxic septal motion along with an increase in the tricuspid valve regurgitant velocity corresponding to elevated PASP (mean 28 ± 2 mm Hg at baseline vs. 55 ± 10 mm Hg immediately after the race for the same five subjects, p < 0.005). The same five subjects experienced wheezing during the race and at its completion. Three were treated with beta2-antagonists, which relieved the wheezing. The subject with the solitary elevated value of cTnI had marked RV dysfunction and the highest PASP (65 mm Hg, an increase of 35 mm Hg from baseline).
The age, weight and oxygen saturation were similar in individuals who developed postrace pulmonary hypertension and RV dysfunction and those who did not (Table 4).
2.3 Results at 18- to 24-h follow-up.
At this interval CK-MB values were within reference limits in all subjects, and cTnI was undetectable in all but one. The subject with the elevated cTnI immediately after the race had a detectable but normal (1.8 ng/ml) cTnI level 18 h after the race had ended. All echocardiograms were normal; specifically, the five subjects who developed pulmonary hypertension and RV dysfunction immediately after the race now had normal echocardiograms.
2.4 Pulmonary function evaluation in subjects with wheezing.
The five subjects who developed wheezing, pulmonary hypertension and RV dysfunction immediately after the race underwent subsequent pulmonary function testing at rest and with exercise at the Washington University Pulmonary Function Laboratory. Baseline evaluation of rest pulmonary function, including pulmonary physical examination, spirometry, lung volume measurements and tests for diffusing capacity and arterial blood gas content, showed no evidence of airway obstruction or impairment of gas exchange at rest. Bicycle ergometric exercise (≥225 W) to heart rates ≥140 beats/min was carried out for 45 to 50 min in cold ambient air (3.5° to 4.5°C). During exercise, a repeat pulmonary physical examination, arterial blood gas analysis and echocardiographic and spirometric examinations were performed. One hour after completion of exercise, a methacholine challenge test was administered [18, 19]. Ausculatory wheezes were heard in two of five subjects, but postexercise forced expiratory volume at 1 min did not change. The methacholine challenge test was weakly positive (16 mg/ml) in one subject and negative in the other four. All echocardiograms recorded during and after exercise showed the expected augmentation in both RV and LV function. The Doppler-estimated PASP remained normal and unchanged from baseline.
In this study, 14 highly trained athletes who participated in an ultramarathon at high altitudes did not develop abnormalities in global or segmental LV wall motion that was evident on the echocardiograms. Five subjects developed marked pulmonary hypertension, RV dysfunction and wheezing. All but one subject had cTnI levels that remained within the normal reference range. The subject with elevated cTnI had the most RV dilation and the most severe pulmonary hypertension (estimated PASP of 65 mm Hg). It is therefore likely that the elevated cTnI immediately after the race was due to injury to the RV rather than to the LV. Thus, in trained individuals, strenuous exercise at high altitude does not result in LV damage or dysfunction.
3.1 Biochemical markers of cardiac injury.
Previous studies have relied on measurements of CK-MB to document cardiac injury. However, CK-MB can be elevated in patients with acute and chronic muscle disease, in marathon runners and in patients with chronic renal failure in the absence of clinical, electrocardiographic or echocardiographic evidence of myocardial injury [20–24]. We measured cTnI, a more sensitive and specific marker of cardiac injury, and found that it remained within the normal range in all but one of our subjects, as discussed earlier. Our group has previously shown that cTnI is a highly sensitive and cardiac-specific marker of myocardial injury, even in those with skeletal muscle injury, including marathon runners [7, 8]. Furthermore, cTnI is more specific than CK-MB for the detection of myocardial injury . Therefore, elevations of CK-MB in our study were most likely due to skeletal muscle damage.
3.2 Postexercise LV dysfunction.
Previous investigators have suggested, based on echocardiographic findings, that LV dysfunction occurs during extreme exercise [1, 2]. We found no evidence of global or segmental abnormalities of the LV on serial echocardiograms recorded after ultraexercise. Our data differ slightly from the data of previous investigators, and may be explained by one or more of the following. Given the RV dilation/dysfunction that resulted from acute pressure overload, it is possible that the finding of the involvement of the interventricular septum and the LV apex is due to ventricular interdependence, thus giving the appearance of LV dysfunction or damage, or both. This is consistent with the observations by Douglas et al. of segmental echocardiographic changes in the septum and apex after strenuous exercise. Alternatively, in the setting of high altitude exercise, the RV may be protective of the LV, although this protection may not occur during prolonged exercise at lower altitudes. In addition, rapidly reversible wall motion abnormalities may have been missed on the immediate postrace echocardiograms, although our time to imaging after exercise was similar to that in previous studies [2, 4, 5]. In untrained individuals, ventricular dysfunction and acute myocardial infarction can be triggered by strenuous physical activity .
3.3 Postexercise RV dysfunction.
The RV echocardiographic abnormalities observed were dramatic and unexpected. Five (36%) of 14 subjects developed marked RV dilation, paradoxic septal motion and a decrease in global RV function. It is likely that pulmonary hypertension initiated the chain of events by causing increased RV afterload, RV dilation and a decrease in FAC, an echocardiographic load-dependent measurement of cardiac function. Injury to and/or dysfunction of the RV secondary to acute pulmonary hypertension has been documented previously [26–28]. Adams et al. measured serial CK-MB levels in patients with acute pulmonary emboli and no evidence of ischemic heart disease and found that 8% had the rise and fall in enzyme levels characteristic of acute myocardial infarction, in addition to echocardiographic findings suggestive of RV infarction . In response to ultraexercise, RV dilation has been described by Douglas et al. in a subset of 41 athletes participating in the Hawaii Ironman Triathlon. These investigators showed, using echocardiography, that the RV dilated immediately after the race and returned to baseline within 2 days. However, because they did not report a Doppler-estimated PASP, it is unknown whether these athletes developed exercise-induced pulmonary hypertension. Thus, the phenomena that we observed in this study has been partially described in earlier studies, and it may have been more pronounced in our subjects because of the high altitude or the more severe exercise performed, or both. Although the concept that acute pulmonary hypertension can induce injury to and/or dysfunction of the RV is well established, the causes of elevated pulmonary artery pressures and wheezing after rigorous exertion at high altitude are less clear; they could result from exercise-induced asthma or a mild form of high altitude pulmonary edema, or both.
3.4 Exercise-induced pulmonary hypertension.
The development of pulmonary hypertension secondary to hypoxemia associated with asthma has been described [29, 30]. McFadden and other investigators [30–38]have suggested that asthma, and especially exercise-induced asthma, is primarily a vascular phenomenon, resulting in vascular engorgement of the bronchial circulation, inflammation and edema of the bronchial mucosa and thus airway narrowing manifested by audible wheezing [31–37]. The five runners who developed pulmonary hypertension and RV dilation/dysfunction in this study also experienced wheezing, which was relieved by administration of beta2agonists. Each of them indicated that this had occurred previously at lower altitudes; we observed wheezing in two subjects when we evaluated them at sea level.
Studies in patients with acute mountain sickness and in those with LV dysfunction support the concept that asthma may be induced by vasoconstriction [39–41]. In subjects with high altitude pulmonary edema, pulmonary artery vasoconstriction and pulmonary hypertension develop in response to acute hypoxia. The pulmonary hypertension results in increased capillary and epithelial permeability, which causes wheezing and dyspnea. Asthma and high altitude pulmonary edema have been shown to be prevented by prophylactic treatment with the calcium channel blocker nifedipine, supporting a vascular mechanism in the cause of these two diseases [42–45]. Oxygen has been therapeutically successful, but we found no reports of the use of aerosolized beta2agonists. The incidence of the kind of reversible exercise-induced pulmonary hypertension, wheezing and RV dilation/dysfunction we observed is unknown. Although it is likely that the high altitude accentuated the clinical syndrome, it is possible that untrained individuals may experience severe respiratory symptoms at lower altitudes.
These data suggest that in well trained athletes, strenuous exercise at high altitude does not result in LV damage or dysfunction. However, exercise-induced pulmonary hypertension, wheezing and RV dilation/dysfunction occurred in a third of those completing the ultramarathon at high altitude. The incidence and pathogenesis of these findings remain to be determined.
We thank Chris Bauman, RCDS, Vonnie Landt and Carol Williams for technical assistance; Elizabeth Engeszer for editorial assistance; and the race directors of the Hardrock 100 Marathon for invaluable assistance with the study.
☆ This work was supported in part during the tenure of a Minority Scientist Development Award from the American Heart Association (Dallas, Texas) to Dr. Dávila-Román.
- creatine kinase, MB fraction
- cardiac troponin I
- fractional area change
- left ventricle, left ventricular
- pulmonary artery systolic pressure
- right ventricle, right ventricular
- Received December 24, 1996.
- Revision received March 26, 1997.
- Accepted April 16, 1997.
- The American College of Cardiology
- Niemelä KO,
- Palatsi IJ,
- Ikheimo MJ,
- Takkunen JT,
- Vuori JJ
- Douglas PS,
- O’Toole ML,
- Hiller DB,
- Hackney K,
- Reichek N
- Douglas PS,
- O’Tolle ML,
- Woolard J
- Adams JE III.,
- Bodor GS,
- Dávila-Román VG,
- et al.
- Adams JE III.,
- Abendschein DR,
- Jaffe S
- Adams JE III.,
- Schechtman KB,
- Landt Y,
- Ladenson JH,
- Jaffe AS
- Bommer W,
- Weinert L,
- Neumann A,
- Neef J,
- Mason DT,
- DeMaria A
- Yock PG,
- Popp RL
- Bodor GS,
- Porter S,
- Landt Y,
- Ladenson JH
- Vaidya HC,
- Maynard Y,
- Dietzler DN,
- Ladenson JH
- Juniper EF,
- Frith PA,
- Dunnett C,
- Cockcroft DW,
- Hargreave FF
- Kishgegian AA,
- Feinberg NV
- Apple RS,
- Rogers MA,
- Sherman WM
- Schwartz JG,
- Prihoda TJ,
- Stuckey JH,
- Gage CL,
- Darnell ML
- Coma-Canella I,
- Gamallo C,
- Onsurbe PM,
- Lopez-Sendon J
- Chandler Deal E Jr.,
- McFadden ER Jr.,
- Ingram RH Jr.,
- Strauss RH,
- Jaeger JJ
- Zawadski DK,
- Lenner KA,
- McFadden ER Jr..
- Gilbert IA,
- Fouke JM,
- McFadden ER Jr..
- Gilbert IA,
- Winslow CJ,
- Lenner KA,
- Nelson JA,
- McFadden ER Jr..
- Hultgren HN,
- Grover RF,
- Hartley LH