Cardiac Remodeling in Women Athletes and Implications for Cardiovascular Screening

Long-term athletic training is associated with left ventricular (LV) remodeling, including increased dimensions in cavity size, wall thickness, and mass, which are regarded as a physiologic adaptation to the increased hemodynamic load induced by chronic and intensive exercise training (i.e., “athlete’s heart”) (1,6,10,20,21). Although numerous studies have extensively described this cardiac remodeling in athletes, most were limited to males, and information regarding female athletes has remained incomplete. In the past decade, however, the increasing participation of women at the highest levels of competitive sports, while exposing athletes to more substantial alterations in cardiac morphology, has offered us the opportunity to investigate the characteristics of cardiac remodeling in elite female athletes as well as the gender-specific differences. We had the opportunity to explore this issue, by evaluating a large population of highly trained and elite female athletes in comparison with male elite teammates (15). We took advantage of the large database of the National Institute of Sports Medicine, where large populations of competitive athletes, of both sexes, have been consecutively examined with echocardiography as a part of the medical program here implemented for elite Italian athletes (16). Our study included a group of 600 female athletes, free of structural cardiovascular disease, who were engaged in a wide range of 27 different sports, with long participation in competitions, including about one third of elite athletes who had achieved recognition in international and Olympic events. CARDIAC DIMENSIONS IN FEMALE ATHLETES In the overall group of female athletes, LV end-diastolic cavity dimension ranged from 40 to 66 mm (mean 49 ± 4) (Fig. 1).FIGURE 1—Distribution of left ventricular end-diastolic cavity dimensions in a large population of 600 consecutive female athletes, engaged in a spectrum of 27 different disciplines, routinely examined with echocardiography at National Institute of Sports Medicine (: solid columns ). A substantial minority (8%) of these athletes showed cavity dimension that exceeded the upper limits of normal. For comparison, distribution of cavity dimension in a group of 65 sedentary female controls, of the same age and body size, is shown ( shaded columns ).Although the vast majority of athletes showed absolute values within the normal limits (i.e., end-diastolic diameter ≤ 54 mm) (5), a substantial minority (about 8%), had an LV cavity size that was enlarged and occasionally markedly dilated (i.e., ≥60 mm). Maximal LV thickness (usually corresponding to the anterior ventricular septum) ranged from 6 to 12 mm (mean 8.2 ± 0.9). Maximum LV wall thickness was greater than 10 mm in a minority of female athletes (only nine subjects) and did not exceed the upper limits of normal (i.e., 12 mm) (5) in any of the 600 athletes. Relative wall thickness was 0.34 ± 0.03, with a wide range from 0.24 to 0.50. Finally, LV mass (normalized to body surface) area was 80 ± 16 g·−2 and was above the accepted normal limits (i.e., 110 g·−2) (2) in a 6%. COMPARISON OF FEMALE ATHLETES WITH SEDENTARY CONTROLS When female athletes and sedentary female controls (of comparable age, body size, and racial composition) were compared, athletes showed enlarged LV cavity dimension (average, +6%), increased wall thickness (average, +14%), relative wall thickness (average, +9%), and mass normalized to body size (average, +25%). Compared to sedentary controls, athletes also showed a slightly enlarged left atrial dimension (average, +4%) (15). Despite morphologic differences, athletes did not show alterations in the indexes of LV systolic function (ejection fraction was greater than 50% in each); also, diastolic filling pattern as assessed by Doppler echocardiography was normal, including early diastolic peak flow velocity (73 ± 13 cm·s−1 vs 72 ± 11 cm·s−1 in controls; not significant (ns)) and deceleration of early peak flow velocity (521 ± 133 ˙s−1 vs 515 ± 120 ˙s−1 (2) in controls; ns); however, late (atrial) peak flow velocity was lower in athletes than controls (30 ± 8 vs 35 ± 8; P < 0.001) as a consequence of the bradycardia typical of trained subjects. Consequently, athletes also showed an increased ratio of the early to late peak flow velocities (2.6 ± 0.9 vs 2.2 ± 0.6 of controls; P < 0.001) (15). COMPARISON OF FEMALE WITH MALE ATHLETES A group of 738 male athletes (reported as part of a previous analysis (17)) of similar age, ethnic origin, sporting disciplines, and intensity of training, was used for comparison with the group of 600 female athletes. Women showed a smaller LV cavity dimension (−11%), maximum wall thickness (−23%), relative wall thickness (−9%), and mass normalized to body size (−31%) compared to male athletes. Also, aortic root (−9%) and left atrial dimension (−14%) were smaller in female athletes. LV cavity size in male athletes showed a wide range of values (44−66 mm), which was similar to that observed in female athletes (40−66 mm) (Fig. 1). In contrast, LV wall thickness showed a broader range of values in male (7−16 mm) than in female (6−12 mm) athletes (17). Indeed, LV wall thickness exceeded the upper limits of normal in a small subset of elite male athletes (2%), whereas it remained within the accepted normal limits in all female athletes (Fig. 2).FIGURE 2—Distribution of maximum LV wall thickness in the 600 female athletes (: solid columns ). For comparison, distribution of maximum wall thickness in a group of 738 male athletes of the same ethnic origin, age range, and spectrum of sport disciplines participated in is shown ( shaded columns ).Although 2% of these athletes exceed the upper limits of normal (i.e., 12 mm), women rarely have wall thickness >11 mm and none exceeds the normal limit.The gender-related differences in LV wall thickness were not related to the different body size (or composition) of the female athletes because normalization of the LV wall thickness for body surface area or height did not eliminate differences among sexes, and men continued to significantly exceed women (15). DETERMINANTS OF LV REMODELING IN FEMALE ATHLETES In the overall group of 600 athletes, the impact of different determinants on LV dimensions was assessed by stepwise regression analysis, which demonstrated that about 50% of the variability in LV cavity dimension was associated with body size, older chronological age, and lower resting HR (which in our population also reflect the intensity and duration of athletic conditioning). ANCOVA also confirmed the significant impact of the type of sport and showed that endurance disciplines (such as cycling, cross-country skiing, rowing/canoeing) had the greatest effect on LV cavity dimensions. Other disciplines, such as team ball (which include aerobic and anaerobic exercise training) showed a moderate impact on LV cavity dimension; finally, technical disciplines (such as equestrian or yachting) had only a minimal effect, as shown in Figure 3.FIGURE 3—Representation of the impact of different sport disciplines on LV cavity dimension assessed by stepwise regression analysis and ANCOVA in our population of 600 competitive female athletes. Hockey is field hockey.These findings are consistent with previous investigations in male athletes, in which different training profiles have been shown to alter cardiac dimensions in different ways, with endurance disciplines demonstrating the greatest impact on LV cavity dimension and wall thickness (17). Other factors may explain part of the gender-related LV dimensional differences. Gender itself is an independent determinant, and, although a large proportion of the differences between males and females are due to different average body size, other mechanisms possibly implicated are the lower absolute blood pressure response to exercise (4), and the lower availability of anabolic androgenic hormones (which stimulate cardiac protein synthesis (13)) in female athletes. Finally, genetic factors have recently achieved major recognition and have stimulated several investigations that suggest their role in the cardiac remodeling in athletes (7,14). OUTER LIMITS OF LV REMODELING IN FEMALE ATHLETES AND IMPLICATIONS FOR CARDIOVASCULAR SCREENING The upper limits to which absolute LV dimensions are increased with athletic conditioning in women have particular relevance to the differential diagnosis between the athlete’s heart and structural cardiovascular disease. In fact, female athletes not uncommonly show an enlarged LV cavity dimension (i.e., end-diastolic dimension ≥ 54 mm) and, occasionally, markedly dilated (≥60 mm) that overlap into a distinctively pathologic range observed in patients with dilated cardiomyopathy (DCM) (3,9). This morphologic finding raises a differential diagnosis between an extreme cardiac adaptation to intensive exercise training and a pathologic cardiac condition with the potential for adverse clinical consequences. The correct identification of physiologic LV dilatation may avoid an unnecessary withdrawal of the athlete from competitions, and the unjustified loss of the varied benefits (including economic) derived from sport (11). In DCM patients, the LV cavity is disproportionately enlarged and is modified to a more spherical shape (3); in trained athletes LV cavity enlargement is associated with slight enlargement of the right ventricle; indeed, the physiologically dilated LV cavity maintains the ellipsoid shape, with the mitral valve normally positioned and without mitral regurgitation (18,19). The most definitive evidence of DCM is the presence of global systolic dysfunction (i.e., ejection fraction less than 50%), and/or evidence of segmental wall motion abnormalities. Instead, athletes with physiologic LV cavity enlargement do not show global or segmental systolic dysfunction or abnormal diastolic filling (18). Finally, LV cavity enlargement is more common in athletes training in largely aerobic disciplines, such as cycling, cross-country skiing, rowing, and long-distance running and is associated with superior physical performance (18). The upper limits of absolute LV wall thickness in female athletes rarely exceed 11 mm and never overlap into the abnormal range (i.e., ≥13 mm), compatible with hypertrophic cardiomyopathy (HCM) (12). This observation appears to differ importantly from that found in male athletes in whom LV wall thicknesses may exceed the upper limits of normal (i.e., 12 mm) in an important minority (17). Therefore, intense athletic conditioning apparently does not represent a sufficient stimulus to increase LV wall thicknesses in women up to the gray zone of borderline LV hypertrophy, and such athletes do not show morphologic changes that resemble HCM (15). Considering that male and female patients with HCM show a similar magnitude of LV wall thickening (8), the presence of LV wall thickness of at least 13 mm in a female athlete is unlikely to represent a physiologic consequence of athletic conditioning and more likely is an expression of a primary pathologic hypertrophy, such as HCM.