Exercise-induced cardiac remodeling (EICR) is the process of structural and functional myocardial adaptation that develops in response to sustained exercise training. Exercise-induced cardiac remodeling is a complex process with numerous determinants, including the athlete’s age, sex, ethnicity, genetics, and sport type.1 Although EICR was first described more than a century ago,2 it remains an area of active investigation for 2 basic reasons. First, the study of EICR continues to refine our understandings of adaptive human biology and the role of the cardiovascular system in human performance. Second, EICR provides critical data for the effective clinical management of athletes and highly active individuals. Differentiating EICR from the diseases associated with the risk for adverse events during athletic participation is among the most challenging and important tasks for the sports cardiologist. At the most basic level, differentiating adaptive physiologic development from occult pathologic change requires a comprehensive understanding of what constitutes normal for a given athlete subtype.
In this issue of JAMA Cardiology, Engel et al3 from Columbia University Medical Center present echocardiographic data from a sizable cohort of professional basketball players. Data derived from National Basketball Association’s (NBA’s) compulsory annual preparticipation screening protocol, unique in the global sports world in that it includes stress echocardiography, was used to generate a robust normative database. The athletes included in this 2-year study were young (mean [SD] age, 25.4 [4.3] years), largely African American (406 of 526 [77.2%]), and men who were most notable for their height (mean [SD], 200.2 [8.8] cm). Key findings from this effort can be summarized as follows. Measurements of the left heart structure including left ventricular (LV) cavity size, LV wall thickness, and LV mass exceeded normal cut points proposed for use in the general population in a significant number of athletes. However, each of these LV metrics was directly proportional to body size and not significantly different from more sedentary populations after appropriate scaling. Second, LV hypertrophy (LVH), the hallmark of the athlete’s heart, was observed in 144 of 526 NBA athletes (27.4%), with ethnicity identified as a key determinant of the prevalence and geometric pattern of LVH. Third, LV function, as assessed using ejection fraction, was mildly depressed (<50%) in a small subset of players who were noted to have larger ventricles than athletes with higher ejection fractions. All athletes with mildly reduced ejection fractions had appropriate ejection fraction augmentation during exercise provocation. Finally and perhaps of greatest interest, aortic root diameter, despite the immense body size of these athletes, exceeded 40 mm in very few athletes (24 of 526 [4.6%]), and when dilation was observed, it was of a very small magnitude (maximally recorded dimension, 42 mm). Unlike LV variables, which retained a direct relationship with height across the measurement spectrums, aortic root diameter reached a plateau at a body height of approximately 212 cm. This important finding confirms the notion that only mild aortic dilation should be considered physiologic among athletes4 and suggests that athletes at the extreme end of the height spectrum should not be expected to demonstrate proportionally marked aortic dilation.
A consortium of experts recently convened by the American College of Cardiology5 identified the development of normative, sport-specific, athlete databases as a key area of scientific pursuit. The study by Engel et al3 provides a crucial piece of this puzzle and as such, the authors are to be commended for their work. Their data underscore the need for physicians who care for competitive basketball players, and perhaps other tall athletes from alternative sporting disciplines, to consider body size during the interpretation of cardiac imaging data as a matter of routine. In addition, this study reminds us that a mildly reduced LV ejection fraction, particularly among athletes with marked eccentric LVH (concomitant wall thickening and chamber dilation), most often represents physiologic features and not disease. This finding is best explained by the concept that stroke volume, not ejection fraction, is physiologically regulated and that athletes with large ventricles simply need to eject a smaller percentage of end-diastolic volume under resting conditions. As nicely demonstrated in this study, augmentation of LV ejection fraction during exercise may be a useful tool for differentiating the low ejection fraction of health from that of mild myopathy.
Finally, this study enhances our understanding of the association between aortic size and athletic participation in perhaps the most clinically relevant population. Unlike ventricular size, which increases proportionally with body size, aortic size may further increase dramatically after extended isotonic exercise training and thus the upper limits of normal cannot be applied. We should now routinely use a ceiling for physiologic aortic dilation. Specifically, athletes with aortic dimensions that exceed age and body size–scaled normal limits should be considered at risk for aortopathy and prohibited from competitive athletics or relegated to close clinical surveillance as appropriate on an individual case-by-case basis.6
As is often the case, this important study raises as many questions as it answers. At the most fundamental level, the inherent limitations of cross-sectional data are crucial. Cross-sectional studies, such as that of Engel et al,3 provide snap-shot observations that can generate hypotheses but cannot establish causal relationships. As such, we can make no meaningful conclusions about whether the LV dilation and LVH observed in this cohort of NBA athletes represent the process of EICR or sport-independent factors. The somewhat striking direct linear relationships between LV variables and body size coupled with LVH that was driven by ethnicity and seemingly not by attributes of sport (eg, duration or intensity of exercise exposure) suggest that basketball players may in fact undergo far less EICR than athletes from other sporting disciplines. This cardinal uncertainty has direct clinical relevance, particularly to basketball athletes who are subjected to serial preparticipation measurements or longitudinal surveillance. Specifically, whether cardiac variables change during a multiyear professional career or during shorter periods of prescribed detraining, a tool often used during the clinical assessment of gray zone hypertrophy, remains unknown. Repeated-measure studies in both scenarios will be required to clarify the pros and cons of annual athlete assessment and the diagnostic utility of prescribed detraining. In addition, the present study leaves unanswered questions about the determinants of data heterogeneity among basketball players. For example, the hemodynamic physiology and attendant EICR of a relatively small, agile point guard with high endurance may differ significantly from that of a power forward or a center. The effect of field or court position on EICR, and thus sport-specific normative data, has been delineated clearly in other sporting disciplines.7 To what degree this phenomenon and other established determinants of EICR apply to basketball athletes is unknown.
Finally, we are left to consider LVH among basketball athletes. Left ventricular hypertrophy is classically subdivided into eccentric and concentric geometric variants, which is a dichotomization that informs physiology and prognosis. Among athletes, these forms of LVH have been associated with the relative contributions of isometric and isotonic hemodynamic stress, the chronicity of training, and underlying genetics. In the present study, a more recently developed LVH classification scheme that further subdivides concentric and eccentric LVH based on the presence or absence of absolute LV dilation is applied. This approach was initially developed for use in hypertensive cohorts and has been demonstrated to predict prognosis with more accuracy than the conventional approach.8 To what degree this schema should be applied to athletes remains uncertain, and caution is urged in attempting to translate lessons from patients with hypertensive heart disease. Specifically, eccentric dilated LVH, a geometric subtype shown to confer risk among patients with hypertension but also the exact phenotype found among healthy endurance athletes, accounted for 37 of 134 cases of LVH (27.6%) among basketball players. Are these basketball players healthy or not? At present, use of this newer LVH schema in athletic cohorts remains uncertain with respect to how it informs us about physiologic adaptation and untested from a prognostic perspective.
The areas of uncertainty delineated above set the stage for future work and do not diminish the value of the current NBA study. Physicians who care for professional basketball players now have a valuable resource at their fingertips. Ideally, this study will set the stage for future longitudinal efforts in this population and similarly visionary cross-sectional work in other distinct athlete populations. The absence of definitive normative data is at the core of many passionately debated sports cardiology topics, including the role of preparticipation screening. For now, we must remember that we cannot act in the best interest of the individual patient or the society at large until we understand what is normal and what is not.
Corresponding Author: Aaron L. Baggish, MD, Cardiovascular Performance Program, Massachusetts General Hospital, Yawkey Ste 5B, 55 Fruit St, Boston, MA 02114 (email@example.com).
Published Online: February 24, 2016. doi:10.1001/jamacardio.2015.0289.
Conflict of Interest Disclosures: The author has completed and submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest and none were reported.
Baggish AL. Cardiac Variables in Professional Basketball PlayersLooking Closely at the Normal Big Athlete (NBA). JAMA Cardiol. 2016;1(1):87-88. doi:10.1001/jamacardio.2015.0289