Author + information
- Received January 17, 2003
- Revision received March 28, 2003
- Accepted April 17, 2003
- Published online August 6, 2003.
- Domingo Hernández, MD*,* (, )
- Alejandro de la Rosa, MD†,
- Antonio Barragán, MD†,
- Ysamar Barrios, PhD§,
- Eduardo Salido, MD§,
- Armando Torres, MD*,
- Basilio Martín, MD*,
- Ignacio Laynez, MD†,
- Amelia Duque, MD†,
- Antonia De Vera, MD§,
- Victor Lorenzo, MD* and
- Antonio González, MD‡
- ↵*Reprint requests and correspondence:
Dr. Domingo Hernández, Servicio de Nefrologia, Hospital Universitario de Canarias, Ofra s/n. 38320, La Laguna, Tenerife, Spain.
Objectives We studied the impact of the angiotensin-converting enzyme (ACE)/DD genotype on morphologic and functional cardiac changes in adult endurance athletes.
Background Trained athletes usually develop adaptive left ventricular hypertrophy (LVH), and ACE gene polymorphisms may regulate myocardial growth. However, little is known about the impact of the ACE/DD genotype and D allele dose on the cardiac changes in adult endurance athletes.
Methods Echocardiographic studies (including tissue Doppler) were performed in 61 male endurance athletes ranging in age from 25 to 40 years, with a similar period of training (15.6 ± 4 h/week for 12.6 ± 5.7 years). The ACE genotype (insertion [I] or deletion [D] alleles) was ascertained by polymerase chain reaction (DD = 27, ID = 31, and II = 3). Athletes with the DD genotype were compared with their ID counterparts.
Results The DD genotype was associated with a higher left ventricular mass index (LVMI) than the ID genotype (162.6 ± 36.5 g/m2vs. 141.6 ± 34 g/m2, p= 0.031), regardless of other confounder variables. As a result, 70.4% of DD athletes and only 42% of ID athletes met the criteria for LVH (p = 0.037). Although systolic and early diastolic myocardial velocities were similar in DD and ID subjects, a more prolonged E-wave deceleration time (DT) was observed in DD as compared with ID athletes, after adjusting for other biologic variables (210 ± 48 ms vs. 174 ± 36 ms, respectively; p = 0.008). Finally, a positive association between DT and myocardial systolic peak velocity (medial and lateral peak Sm) was only observed in DD athletes (p = 0.013, r = 0.481).
Conclusions The ACE/DD genotype is associated with the extent of exercise-induced LVH in endurance athletes, regardless of other known biologic factors.
Adaptive left ventricular hypertrophy (LVH) is a well-known feature of the athlete's heart (1). Physical exercise activates the renin-angiotensin system (RAS) (2), and this system may regulate myocardial growth (2,3). Indeed, angiotensin II stimulates cardiac protein synthesis, whereas bradykinins may have an antiproliferative effect. The angiotensin-converting enzyme (ACE) is a key enzyme in the production of angiotensin II, as well as degradation of bradykinins, and it may modulate cardiac growth. Thus, an upregulated RAS after a long period of training may contribute to ventricular mass growth. However, not all endurance athletes with similar training develop left ventricular (LV) mass growth to the same extent, suggesting that genetic factors may modulate heart size (4).
Genes encoding RAS components regulate ACE levels and angiotensin II expression and function (5). In particular, an insertion/deletion (I/D) polymorphism of the ACE gene affects the plasma and tissue levels of ACE activity. The presence of each D allele has an additive effect on ACE levels, so that DD subjects have the highest levels of this enzyme (6). Therefore, a positive association between the D allele dose and exercise-induced hypertrophy appears likely.
Despite this evidence, the role of the ACE/DD genotype in the development of the athlete's heart is unclear. Prospective and cross-sectional studies have found a preponderance of D allele carriers among individuals with exercise-induced cardiac changes, but a D allele dose effect was not demonstrated. Indeed, significant cardiac growth in response to physical training (7–9)and established exercise-induced LVH (10)were similarly achieved by both DD and ID subjects. Other reports, however, have not observed such an association (11,12). Additionally, some studies included athletes younger than 20 years (8,12), and it is known that the most pronounced and homogeneous cardiac changes induced by exercise occur among elite athletes (13)or endurance athletes ranging in age from 20 to 39 years (14). Thus, the impact of the ACE/DD genotype and D allele dose on heart size could be greater in adult endurance athletes who have undergone vigorous training programs over longer periods. Finally, little information has been provided about Doppler-derived diastolic function in these subjects (7–12), which could distinguish pathologic from physiologic LVH (15,16).
We investigated the effect of the ACE/DD genotype on morphologic and functional cardiac changes, assessed by echocardiography, in 61 male endurance athletes (age range 25 to 40 years) with long periods of intensive training.
A group of 61 white male endurance athletes from the Canary Islands were studied. All of them participated primarily in isotonic sports (36 long- and middle-distance runners, 4 swimmers, 11 endurance cyclists, and 10 triathlonists). Each athlete had trained for at least ≥10 h/week for the last 5 years (15.6 ± 4 h/week for 12.6 ± 5.7 years). None of the athletes had a history of cigarette smoking, hypertension, coronary artery disease, diabetes mellitus, renal or hepatic dysfunction, or a positive family history of hypertrophic cardiomyopathy. Likewise, no medication was taken by any of the athletes for any reason, including anabolic steroids. Blood pressure was measured in a sitting position from the right arm after 15 min of rest. Mean systolic blood pressure was 117 ± 12 mm Hg (range 140 to 90 mm Hg), and mean diastolic blood pressure was 65 ± 7 mm Hg (range 50 to 80 mm Hg).
This study was approved by the Ethics Committee of the Hospital Universitario and conducted according to the Declaration of Helsinki. All athletes gave written, informed consent before participating in the study.
Fasting biochemical parameters (serum creatinine, glucose, and complete blood count) were measured by means of a computerized autoanalyzer (Hitachi 717, Boehringer Mannheim, Mannheim, Germany).
Echocardiographic studies were performed with an ultrasonoscope system equipped with a tissue Doppler program (Hewlett-Packard Sonos 5500 Advanced Diagnostics, Palo Alto, California), using a 2- to 4-MHz multifrequency transducer. Echocardiograms were recorded following a standard protocol at or just below the tips of the mitral valve leaflets, with the transducer applied to the third or fourth intercostal space. All measurements were analyzed by two experienced readers (A.R. and A.B.), on an average of three to five cardiac cycles, without previous knowledge of the genetic typing. In our laboratory, interobserver variability for these measurements was below 7%, similar to previous results from our laboratory (17).
Standard echocardiographic studies consisted of M-mode, two-dimensional, and Doppler blood flow measurements. Cavities and cardiac walls were measured following the recommendations of the American Society of Echocardiography (18). Left ventricular mass was determined according to the method of Devereux and Reichek (19)and indexed to body surface area to yield the left ventricular mass index (LVMI). The cut-off level defining LVH was LVMI >143 g/m2(20). Systolic function was assessed by the ejection fraction. Pulsed-wave Doppler of transmitral flow was used to assess overall diastolic function. The Doppler indexes measured were peak early velocity (E) and peak atrial velocity (A) in centimeters per second, deceleration time of the E-wave (DT), and LV isovolumic relaxation time (ms). In addition, the E/A ratio was calculated.
Pulsed-wave tissue Doppler echocardiography was performed from the apical four-chamber view for lateral and medial sites. The sample volume (gated at 5 mm) was placed over the mitral annulus when the patient was breathing normally. Peak velocities of the medial and lateral mitral annulus were acquired. Filters were set to exclude high-frequency signals, and the Nyquist limit was adjusted to a velocity range of −15 to 20 cm/s. Thus, the following measurements were obtained: peak velocity of the myocardial systolic wave (peak Sm[cm/s]), peak velocity of the E-wave (peak Em[cm/s]), and peak velocity of the A-wave (peak Am[cm/s]). The Em/Amratio was calculated.
Determination of genotypes
The ACE I/D polymorphism was determined by polymerase chain reaction amplification of a fragment of intron 16 of the ACE gene (21). Deoxyribonucleic acid (DNA) was purified from 3 ml blood by proteinase K digestion, phenol extraction, and ethanol precipitation, following standard protocols. Approximately 0.1 μg DNA was amplified with specific primers, and the amplification product was analyzed by gel electrophoresis. Samples yielding exclusive amplification of the D allele (and therefore potentially typed as DD) were subjected to a second independent amplification, as described by Lindpaintner et al. (21). We also estimated the genotype frequencies for the I/D polymorphism at the ACE locus among 246 DNA samples from nonathletic healthy individuals, which have been previously reported (22).
Differences in the distribution of ACE genotypes between the athletes and healthy individuals was tested by the chi-square exact test. A comparison between the genotypes observed and the predictions from the Hardy-Weinberg model was also done by chi-square analysis. Likewise, frequencies of DD and ID genotypes with LVH were evaluated by the chi-square test. Given that only three subjects were homozygous for the I allele, we statistically compared DD and ID genotypes. Thus, comparisons of continuous variables between both groups were performed by the ttest. General factorial analysis of covariance was used to compare LVMI, as well as DT, between DD and ID genotypes, after adjusting for other potential confounding variables. Linear regression analysis was used to correlate quantitative echocardiographic parameters with different clinical and biochemical data. All data are expressed as the mean value ± SD. A p value <0.05 was considered significant, and all p values are two-tailed. All computations were made using the statistical package SPSS version 10.0 for Windows (SPSS Inc., Chicago, Illinois). Exact tests were performed by Statxact version 5.0 (Cytel Software Co., Cambridge, Massachusetts).
All samples typed as DD with the first assay (primer hace3s and hace3as) were confirmed to lack the I allele by using the second assay (primers hace5a and hace5c), and no disagreement was found between both assays. This is consistent with previous experience in our laboratory, as long as purified DNAs, rather than cell lysates, are used as a template for the amplification reaction. The distribution of the DD, ID, and II genotypes in athletes was 44% (n = 27), 51% (n = 31), and 5% (n = 3), respectively. For the 246 samples from nonathletic healthy individuals tested in our laboratory, the following frequencies for DD, ID, and II were obtained: 37.8%, 49.2%, and 13%, respectively. These differences in genotype frequencies were not statistically significant (chi-square = 3.354, p = 0.187).
Table 1shows the characteristics and biochemical data of endurance athletes according to ACE gene polymorphism. To assess the impact of the D allele dose on cardiac changes, individuals with the DD genotype were compared with those with the ID genotype. There were no significant differences with regard to body mass index, years of active competitive sports, endurance training (h/week), heart rate, hemoglobin levels, and systolic and diastolic blood pressure between the groups, although ID subjects showed a tendency toward a higher age and serum creatinine than DD athletes (Table 1). Morphologic echocardiographic data are summarized in Table 2. Interestingly, DD athletes showed a significantly greater LVMI than ID athletes, even after adjusting for age, body mass index, hemoglobin levels, years of training, and blood pressure. As a result, 70.4% of DD athletes and only 42% of ID athletes met the criteria for LVH (chi-square = 4.71, p = 0.037). None of the subjects with the II genotype showed echocardiographic data of cardiac hypertrophy.
Table 3shows standard and tissue Doppler echocardiographic measurements in athletes with the DD, ID, and II genotype. Although systolic and early diastolic myocardial velocities were similar in DD or ID individuals, the DT was significantly more prolonged in DD athletes (Table 3). This difference persisted after adjusting for confounding variables such as age, heart rate, LVMI, and time of training (Fig. 1, Table 3). Other diastolic parameters were not different between DD and ID subjects. Finally, a significant correlation between DT and myocardial systolic peak velocity (medial and lateral peak Sm) was only observed in DD athletes (p = 0.013, r = 0.481). No other associations were found between echocardiographic parameters and clinical data.
This study demonstrates that the ACE/DD genotype is associated with the development of exercise-induced cardiac growth in endurance athletes. As a result, a different diastolic filling pattern, as evidenced by a longer DT, was also present in these subjects.
The ACE genotype has been associated with cardiac growth in the presence of several pathologic processes (5,22), but this does not permit one to definitely conclude that this polymorphism mediates the development of LVH. Thus, for a better understanding of the role of this polymorphism on cardiac growth, we only included male endurance athletes (age range 25 to 40 years) after a long and intensive period of training. Additionally, we explored diastolic function in our athletes by using standard and tissue Doppler echocardiography to distinguish pathologic from physiologic LVH. Under these conditions, the effect of genetic factors might be more evident than was previously assumed, and it could clarify the contribution of this polymorphism to the development of LVH.
Synthesis of several RAS components, including ACE, has been shown to take place in the myocardium and to be upregulated during myocardial growth (3). Angiotensin-converting enzyme is a key enzyme in the production of angiotensin II, as well as the degradation of bradykinins, and it may therefore participate in cardiac growth. Physical exercise activates the RAS, resulting in an increase of angiotensin II levels (2). Thus, a physiologic role of RAS in the development of the athlete's heart appears likely. However, the degree to which LV mass changes in response to exercise is highly variable, suggesting that genetic factors may play an important role in this process.
An I/D polymorphism of the ACE gene, located in the noncoding region of the gene (intron 16), affects the plasma and tissue levels of ACE activity. Normal individuals homozygous for the D allele show the highest plasma and tissue ACE levels (6), including the heart (23). Thus, healthy subjects with the DD genotype may have a higher risk of LVH than subjects with other genotypes (ID or II) (24). High-performance athletes seem to be an ideal population to study the possible association between ACE polymorphism and the athlete's heart, as adaptive LVH is very common in endurance athletes (1). Thus, during top-level training, genetically increased ACE activity might influence local tissue angiotensin II generation, thereby stimulating cardiac growth. In other words, the ACE/DD polymorphism may play a permissive role in the development of LVH when the cardiac growth machinery is activated. Accordingly, endurance athletes being homozygous for the D allele may potentially have a higher risk than ID or II subjects of developing significant LVH after a long period of training.
In agreement with these arguments, our endurance athletes with the DD genotype showed a higher LV mass than those with the ID genotype, after a similar intensive training regimen. We statistically compared echocardiographic data from DD and ID subjects, because only three subjects were homozygous for the I allele; nevertheless, none of the II individuals met criteria for LVH, as previously reported (7,8). The effect of the DD genotype was independent of other known biologic factors that affect cardiac growth, as documented in normal individuals (25). This suggests that ACE may be a rate-limiting step in tissue angiotensin II generation or degradation of kinins, when a greater D allele dose and intensive training concur. This mechanism might also explain the reported interaction of a polymorphism at the bradykinin receptor B2BKR (26). In any case, this positive association does not definitely prove a direct biologic effect of this polymorphism on the athlete's heart, and, obviously, larger studies will be required to confirm this issue.
Prospective studies have found a preponderance of D allele carriers who have experienced exercise-induced cardiac changes, but no effect of the D allele dose was demonstrated. In fact, significant cardiac growth in response to physical training was similarly achieved in both DD and ID subjects (7–9). Likewise, in a cross-sectional study, LV mass was not different between DD and ID athletes (10). Other investigators, however, have not observed such an association (11,12). There are several possible reasons for the differences between these reports and our study. First, most studies that previously investigated this association included athletes younger than 20 years (8,12), and it is known that the most pronounced exercise-induced cardiac changes take place in elite, adult athletes (13). Moreover, more uniform cardiac growth has been reported in endurance athletes ranging in age from 20 to 39 years (14). Thus, the impact of the ACE/DD genotype and D allele dose on cardiac growth might have been underestimated in previous studies. Finally, LVH is a multifactorial process and may be influenced by ethnic heterogeneity or population stratification, among others factors.
Previous reports have demonstrated a normal diastolic filling pattern in exercise-induced cardiac growth (15,27), but in most of them, the genetic influence of RAS on diastolic function was not documented. The higher availability of local angiotensin II in DD subjects could mediate changes in diastolic properties. Several criteria have been proposed to distinguish exercise-induced from pathologic hypertrophy, including reversal of the E/A ratio, prolongation of the DT, and/or a lesser systolic annular peak velocity, among others (16,28). Our DD athletes had a longer DT than their ID counterparts, but tissue Doppler systolic and diastolic velocities, which most accurately distinguish physiologic from pathologic LVH (16), were similar in both groups. Angiotensin II leads to prolonged diastolic filling (29), so that an early diastolic change might be expected in DD athletes with LVH. However, angiotensin II–induced myocardial hypercontractility, which enhances diastolic myocardial elastic recoil, may balance this effect, providing a proper LV filling and stroke volume in these subjects. In favor of this view was the finding of a positive correlation between DT and peak Smonly in DD athletes.
Recent studies have shown higher frequencies of the I allele among endurance athletes (30). However, there have been conflicting reports about this association, and the representation of ACE alleles among elite athletes might vary depending on the type of sports activity (31). The trend toward a lower frequency of I alleles found in our sample of athletes was not statistically significant, and it could be merely due to chance.
The DD genotype has been associated with a variety of adverse cardiovascular effects (5), and LVH is an independent predictor for the development of heart failure in the general population (32). Whether exercise-induced cardiac growth also carries an increased risk of sudden death in DD endurance athletes is of great concern. The long-term effect of ACE polymorphism on the athlete's heart requires futures studies to answer this important question.
There are several potential limitations of this study. Genetic association studies may only provide hypotheses rather than proof for a biologic effect. Thus, a molecular marker of exercise-induced cardiac growth appears to be rather limited, unless we understand the mechanism whereby angiotensin II increases LV mass. In addition, we could not rule out other genetic mutations responsible for hypertrophic cardiomyopathy in our study. However, the absence of this disorder in family members of our DD subjects makes this unlikely. Finally, we have no data on our athletes' heart size before they initiated systematic and long-term training. Future longitudinal and prospective studies will be needed to clarify the prognostic significance of ACE I/D polymorphism on the athlete's heart.
The ACE/DD genotype is associated with the extent of exercise-induced LV growth in adult endurance athletes, regardless of other known biologic factors.
The authors are grateful to the endurance athletes for their cooperation in this study.
☆ This work was supported by a grant (PI 50/00) from Consejeria de Sanidad y Consumo (Gobierno de Canarias), Fundación Canaria de Investigación y Salud (FUNCIS), Santa Cruz de Tenerife, Spain.
- angiotensin-converting enzyme
- deoxyribonucleic acid
- deceleration time of the E-wave
- left ventricular
- left ventricular hypertrophy
- left ventricular mass index
- peak Sm
- myocardial systolic peak velocity
- renin-angiotensin system
- Received January 17, 2003.
- Revision received March 28, 2003.
- Accepted April 17, 2003.
- American College of Cardiology Foundation
- Higaki J.,
- Aoki M.,
- Morishita R.,
- et al.
- Karjalainen J.,
- Mäntysaari M.,
- Viitasalo M.,
- Kujala U.
- Montgomery H.E.,
- Clarkson P.,
- Dollery C.M.,
- et al.
- Myerson S.G.,
- Montgomery H.E.,
- Whittingham M.,
- et al.
- Karjalainen J.,
- Kujala U.M.,
- Stolt A.,
- et al.
- Nishimura T.,
- Yamada Y.,
- Kawai C.
- Schannwell C.M.,
- Schneppenheim M.,
- Plehn G.,
- Marx R.,
- Strauer B.E.
- Vinereanu D.,
- Florescu N.,
- Sculthorpe N.,
- Tweddel A.C.,
- Stephens M.R.,
- Fraser A.G.
- Devereux R.B.,
- Reichek N.
- Krumhloz H.M.,
- Larsson M.,
- Levy D.
- Danser A.H.,
- Schalekamp M.A.,
- Bax W.A.,
- et al.
- Harrap S.B.,
- Dominiczak A.F.,
- Freiser R.,
- et al.
- Pluim B.M.,
- Zwinderman A.H.,
- van der Laarse A.,
- van der Wall E.E.
- Lewis J.F.,
- Spirito P.,
- Pelliccia A.,
- Maron B.J.
- Myerson S.,
- Hemingway H.,
- Budget R.,
- Martin J.,
- Humphries S.,
- Montgomery H.