|
|
||||||||||
|
J Am Coll Cardiol, 2004; 44:144-149, doi:10.1016/j.jacc.2004.02.057 © 2004 by the American College of Cardiology Foundation |



* Service de Cardiologie, Paris, France
Unité d'Épidémiologie et de Recherche Clinique, Paris, France
Centre d'Investigations Cliniques, Hôpital Européen Georges Pompidou, Paris, France
Institut C
ur Effort Santé, Paris, France
Manuscript received November 18, 2003; revised manuscript received January 24, 2004, accepted February 17, 2004.
* Address for correspondence: Dr. Eric Abergel, Service de Cardiologie, Hôpital Européen Georges Pompidou, 20 rue Leblanc, 75908 Paris Cédex 15, France.
eric.abergel{at}egp.ap-hop-paris.fr
| Abstract |
|---|
|
|
|---|
BACKGROUND: Knowledge of cardiac changes in athletes, who are at particularly high risk of sudden cardiac death, is mandatory to detect hypertrophic cardiomyopathy (HCM) or dilated (DCM) cardiomyopathy.
METHODS: We carried out echocardiographic examinations on 286 cyclists (group A) and 52 matched sedentary volunteers (group C); 148 cyclists participated in the 1995 "Tour de France" race (group A1), 138 in the 1998 race (group A2), and 37 in both (group B).
RESULTS: In groups A, A1, A2, and C, respectively, diastolic left ventricular diameter (LVID) was 60.1 ± 3.9 mm, 59.2 ± 3.8 mm, 61.0 ± 3.9 mm, and 49.0 ± 4.3 mm (A vs. C and A1 vs. A2, p < 0.0001), and maximal wall thickness (WT) was 11.1 ± 1.3 mm, 11.6 ± 1.3 mm, 10.6 ± 1.1 mm, and 8.6 ± 1.0 mm (A vs. C and A1 vs. A2, p < 0.0001). Among group A, 147 (51.4%) had LVID >60 mm; 17 of them had also a below normal (<52%) left ventricular ejection fraction (LVEF). Wall thickness exceeded 13 mm in 25 athletes (8.7%) (always <15 mm), 23 with LVID >55 mm. In group B, LVID increased (58.3 ± 4.8 mm to 60.3 ± 4.2 mm, p < 0.001) and WT decreased (11.8 ± 1.2 mm to 10.8 ± 1.2 mm, p < 0.001) with time.
CONCLUSIONS: Over one-half of these athletes exhibited unusual LV dilation, along with a reduced LVEF in 11.6% (17 of 147), compatible with the diagnosis of DCM. Increased WT was less common (always <15 mm) and scarce without LV dilation (<1%), eliminating the diagnosis of HCM. Serial examinations showed evidence of further LV dilation along with wall thinning. These results might have important implications for screening in athletes.
| ||||||||||||||||||||||||||
| Methods |
|---|
|
|
|---|
Measurements.
For each subject, at least three two-dimensional-guided M-mode recordings of the LV were recorded on videotapes. Paper tracings were obtained simultaneously. All recordings were performed as recommended by Devereux et al. (6) and by the American Society of Echocardiography (7). Readings were taken after all recordings had been completed, and LV measurements were averaged over three cardiac cycles. The parameters recorded included left ventricular internal diameters at end-diastole (LVIDd) and left ventricular internal diameters at end-systole (LVIDs), end-diastolic interventricular septal wall thickness (IVSTd), and posterior wall thickness (PWTd). All readings were taken from anonymous tracings by a single investigator (E.A.). The Devereux-modified American Society of Echocardiography-cube formula was used to calculate LV mass (6) as: 0.8 [1.04 (LVIDd + PWTd + IVSTd)3 LVIDd3] + 0.6 g, which was expressed as a function of BSA (indexed left ventricular mass, g/m2). Relative WT was calculated as (IVSTd + PWTd)/LVIDd, with a distinction between eccentric (relative WT <0.44) and concentric (relative WT
0.44) patterns. Left ventricular volumes were calculated using the Teichholz formulas [7/(2.4 + LVID) x LVID3] and were used to calculate left ventricular ejection fraction (LVEF, %); cardiac output and index were derived from LVEF and heart rate. Endocardial fractional shortening (eFS) and midwall fractional shortening (mFS) were calculated as previously described (8). End-systolic meridional stress (ESS) was calculated as [(0.334 x SBP x LVIDs)/(PWTs x [1 + (PWTs/LVIDs]) (9), where SBP is the systolic blood pressure and PWTs is the systolic posterior WT.
Left ventricular hypertrophy (LVH) was defined as IVSTd and/or PWTd >13 mm and abnormal cavity dilation as LVIDd >60 mm (10,11). Thresholds corresponding to mean ± 1.96 SD of data obtained in the control group were also used to define LV dilation (LVIDd >57.4 mm or LVIDd >30.6 mm/m2), low LVEF (<52%), and low mFS (<14.5%). To take into account afterload, we used the relationship between LVEF and ESS in the control group and in the athletes. Using each model, we calculated the lower limit of the 95% confidence interval of a predicted LVEF for each individual ESS (12). The difference between this last value and the measured value was used to classify subjects: athletes with difference
0 were considered as having abnormally low LVEF.
Statistical analysis. Statistical analysis was performed with the StatView software (Version 4.5, SAS, Inc., Cary, North Carolina). Results are presented as means ± 1 SD. Differences between two groups were compared by an unpaired t test, and differences between more than two groups were tested by one-way analysis of variance. The paired t test was used to assess change within a group. For paired comparisons, the Bonferroni correction was used to take into account multiple comparisons. Differences were considered to be statistically significant if p < 0.05.
| Results |
|---|
|
|
|---|
|
|
|
60% in 111 athletes, between 40% and 52% in 20 athletes, and between 52% and 56% in 38 athletes (Fig. 1). Although cardiac output and cardiac index were higher in athletes than in controls, indexes of endocardial and midwall LV function (LVEF, eFS, mFS) were all significantly lower in athletes (Table 1). Moreover, LVEF was lower in the 147 cyclists with LV dilation (>60 mm) than in the remaining 139 cyclists (60.5 ± 6.9% vs. 62.8 ± 5.6%, p = 0.0027). Conversely, among the cyclists with abnormal LV dilation, 67 of 147 (46%) had an LVEF <60%, vs. 39 of 139 (28%) of those with normal LV size (p = 0.0021). The mFS was abnormally low in 26 cyclists (22 with eccentric LVH) who showed lower LVEF (51.5 ± 4.9% vs. 62.6 ± 5.6% for the remaining 260 athletes, p < 0.0001). Finally, 17 athletes (14 in 1998, 3 in 1995) showed both LV dilation (>60 mm) and low LVEF (<52%) (Fig. 1). Ten of these athletes had also a low mFS (<14.5%). Comparisons between measured LVEF and predicted LVEF (lower limit of the 95% confidence interval) are shown in Figure 3. All the controls are above the zero line: measured LVEF values are always higher than predicted LVEF values. In the subgroup of 17 cyclists with both LVIDd >60 mm and LVEF <52%, 7 cyclists are below or at the zero line, which means that they might be classified as abnormal. Moreover, three cyclists with LVEF <52% and an LVIDd between 58 mm and 60 mm also have lower than predicted LVEF values.
|
|
| Discussion |
|---|
|
|
|---|
LV structure evaluation. Endurance cycling has the greatest impact on LV cavity dimension and WT, due to a combination of endurance training (cycling) and isometric exercise (with the arms). Thus, while LV enlargement exceeding 60 mm (13) is unusual in non-elite athletes, this is observed in about 14% of elite athletes (11). We report here the largest proportion ever published of athletes with substantial LV enlargement (51% >60 mm, up to 73 mm). The major determinants of dilation are usually BSA, as is the case here, and participation in "high-impact" sports such as cycling, cross-country skiing, canoeing, rowing, and soccer. However, unlike in a previous study (11), considerable LV dilation persisted here after correction for BSA. Even in the group of 62 athletes with BSA <1.8 m2, 22 (35%) exhibited abnormal LV dilation (Table 3). Increased LV WT was less common (9% of the athletes) and similar to that previously reported (14). It never exceeded 15 mm and rarely occurred in the absence of cavity dilation (<1%), allowing easy elimination of the diagnosis of HCM (Fig. 2) (15).
|
52%) showed an increased LV size (>60 mm). In a previous study (11), athletes with LV dilation showed a normal global systolic function. Moreover, in a small group of 21 less trained elite cyclists (16), as demonstrated by a lower mean LV mass (200 g vs. 266 g in our study), LVH was not associated with significant abnormalities of cardiac function. It has been suggested that reduced fractional shortening in runners (who show predominant LV dilation) could be due to a decreased preload along with a normal afterload and contractile state (17). It has also been shown that 39% of football players have a moderately reduced resting LVEF (50% to 55%), which appropriately increases with exercise (mean, 76%) (18). In the present study, whatever the approach used to define abnormal LV function (association of LV dilation and low LVEF or low afterload-corrected LVEF) (Fig. 3), a depressed myocardial function can be suggested in a small group of cyclists (17). It is unlikely that these subjects had a pre-existing DCM, as this condition is very scarce in the general population (<36 of 100,000) (19,20). The potential role of exercise-induced tachycardia can be excluded, as all cyclists carefully monitored their heart rate during training and the races, with a similar average exercise heart rate (around 80% of their maximal heart rate to avoid muscular exhaustion). It is also possible that illicit drugs might induce LV function deterioration. This factor has not yet been studied in detail, even though there is evidence that some drugs are becoming more popular among athletes (3). Recent works do not mention potential drug abuse (2), or eliminated this possibility on the questionable basis that athletes denied the use of illicit drugs (11,21). Moreover, the combined effects of multiple drugs remain unknown.
Serial cardiac changes.
Successive examinations in 1995 and 1998 in the subgroup of 37 cyclists evaluated twice showed further LV dilation along with wall thinning with time. This was not associated with a decrease in cyclist performances (Indurain won the 1995 Tour de France3,635 kmat a mean speed of 39 km/h, whereas Pantani won the 1998 Tour de France3,711 kmat a mean speed of 40 km/h). In 1998, LV diameters were markedly increased, with the largest dimension ever reported in athletes (73 mm). The 2-mm difference in diameter observed between 1995 and 1998 in this cohort is of the magnitude observed between athletes practicing isotonic and isometric sports (2). It seems unlikely that this difference was due to poor reproducibility of these anonymous readings. In a previous study involving 96 cyclists, the mean interobserver or intraobserver difference for LV diameters was
0.3 mm (22). A second bias could be due to seasonal changes in LV morphology: WT is lower during the resting season than during the competitive season, whereas diameters remain unchanged (23). In our study, all measurements were taken at the same moment in the competitive season in 1995 and 1998. Moreover, the difference observed in the group of cyclists examined twice was similar to that observed in the whole population, eliminating the possibility that the extremely large LV cavities found in certain athletes are due to a specific genetic background (24). Finally, it has been suggested that heart rate affects M-mode echocardiographic LV measurements (25). However, in our study, heart rate was not significantly different between the two evaluations. In the absence of such bias, LV changes may be due to changes in the type and intensity of physical training. However, to induce a 2-mm difference in LV diameter at the same moment of the season, training technology must change dramatically, and no such change has been reported. Moreover, more intensive and/or prolonged physical training would have lead to wall thickening (23,26), which was not observed in our population. The unexpected and significant decrease in WT observed with time might also be due to the increasing problem of drug abuse.
Conclusions. In this large homogeneous cohort of highly trained elite cyclists, marked echocardiographic LV dilation is frequent and often associated with a low LVEF, which raises the problem of the differential diagnosis with a dilated cardiomyopathy. Abnormal increases in WT were less common, always moderate, and usually associated with cavity dilation, eliminating a primary pathologic hypertrophy. The unexpected occurrence of cavity dilation and wall thinning with time raises questions about excessive physical training and/or pharmacologic interventions. These results might have important implications for myocardial disease screening in highly trained elite athletes.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
P. Luthi, M. Zuber, M. Ritter, E. N. Oechslin, R. Jenni, B. Seifert, S. Baldesberger, and C. H. Attenhofer Jost Echocardiographic findings in former professional cyclists after long-term deconditioning of more than 30 years Eur J Echocardiogr, March 1, 2008; 9(2): 261 - 267. [Abstract] [Full Text] [PDF] |
||||
![]() |
R. Q Osborn, W. C Taylor, K. Oken, M. Luzano, M. Heckman, and G. Fletcher Echocardiographic characterisation of left ventricular geometry of professional male tennis players Br. J. Sports Med., November 1, 2007; 41(11): 789 - 792. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. Gizurarson, M. Lorentzon, T. Ramunddal, F. Waagstein, L. Bergfeldt, and E. Omerovic Effects of complete heart block on myocardial function, morphology, and energy metabolism in the rat Europace, June 1, 2007; 9(6): 411 - 416. [Abstract] [Full Text] [PDF] |
||||
![]() |
R. Hetzer, M. Dandel, C. Knosalla, J. G. Burniston, V. G. Florea, D. Rott, D. Leibowitz, G. J. Vanderwilt, M. H. Yacoub, and E. J. Birks Left Ventricular Assist Devices and Drug Therapy in Heart Failure N. Engl. J. Med., February 22, 2007; 356(8): 869 - 872. [Full Text] [PDF] |
||||
![]() |
T Venckunas, A Stasiulis, and R Raugaliene Concentric myocardial hypertrophy after one year of increased training volume in experienced distance runners Br. J. Sports Med., August 1, 2006; 40(8): 706 - 709. [Abstract] [Full Text] [PDF] |
||||
![]() |
E. Abergel and A. A. Hagege Detection of Pathologic or Physiologic Left Ventricular Remodeling in Athletes J. Am. Coll. Cardiol., May 17, 2005; 45(10): 1731 - 1731. [Full Text] [PDF] |
||||
![]() |
G. A. Whalley, R. N. Doughty, and J. C. Baldi Reply J. Am. Coll. Cardiol., May 17, 2005; 45(10): 1731 - 1731. [Full Text] |
||||
![]() |
L. H Naylor, L. F Arnolda, J. A Deague, D. Playford, A. Maurogiovanni, G. O'Driscoll, and D. J Green Reduced ventricular flow propagation velocity in elite athletes is augmented with the resumption of exercise training J. Physiol., March 15, 2005; 563(3): 957 - 963. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | SUBSCRIPTIONS | CURRENT ISSUE | PAST ISSUES | CARDIOSOURCE | SEARCH | HELP | FEEDBACK |