Article Text

Download PDFPDF

Screening athletes for cardiovascular disease in Africa: a challenging experience
  1. Christian Schmied1,
  2. Fernando M Di Paolo2,
  3. A Yacine Zerguini3,
  4. Jiri Dvorak4,
  5. Antonio Pelliccia2
  1. 1Cardiovascular Center/Sports Cardiology, University Hospital Zurich, Zurich, Switzerland
  2. 2Institute of Sport Medicine and Science, CONI, Rome, Italy
  3. 3Clinique Chahrazed, FIFA Medical Center of Excellence, Cheraga, Alger, Algeria
  4. 4Fédération Internationale de Football Association (FIFA), and Schulthess Clinic, Zurich, Switzerland
  1. Correspondence to Dr Antonio Pelliccia, Institute of Sport Medicine and Science, Department of Medicine, Largo P. Gabrielli, 1, Rome 00197, Italy; ant.pelliccia{at}libero.it

Abstract

Aims Preparticipation cardiovascular (CV) screening has been advocated as an efficient strategy to reduce sudden cardiac death in Caucasian athletes. At present, uncertainty remains if such strategy is feasible and efficient in native African athletes. To this scope, we performed a CV screening in an African setting.

Methods 210 male Gabonian football players were examined with history, physical examination, ECG and echocardiography.

Results On history, 19 players (9%) referred atypical chest discomfort/oppression. Familial sudden death was referred by 36 (17%). No anomalies were detected at physical examination. ECG showed large proportions of ‘training-related’ abnormalities, that is, ST-segment elevation in precordial leads in 150 (71.4%), and isolated increase in R/S-wave voltage in 116 (55.2%). A substantial subset (12.4%) showed ‘training-unrelated’ abnormalities, that is, inverted T-waves in 10 (4.8%), left atrial enlargement in 8 (4%), deep Q-waves in 3 (1.4%). On echocardiography, one athlete meet criteria for hypertrophic cardiomyopathy (HCM); none showed evidence for arrhythmogenic right ventricular cardiomyopathy (ARVC) or dilated cardiomyopathy (DCM). Other abnormalities included mitral valve prolapse in three, atrial septal defect in two and pulmonary hypertension in one.

Conclusions About 12% of native African athletes showed ECG abnormalities unrelated to training and requiring additional testing and periodical follow-up. Structural abnormalities were found, however, in a minority (5%), including HCM in one, but no ARVC or DCM. In conclusion, this study demonstrates that preparticipation CV screening is efficient to identify (or raise suspicion) for CV abnormalities in native African athletes, but challenging for conclusive identification of cardiac diseases in the difficult scenario of a developing African country.

  • Cardiology

Statistics from Altmetric.com

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Introduction

Strategies to prevent sudden cardiac death (SCD) in young athletes have raised large interest and fuelled intense debate within the scientific community, with particular regard to feasibility and efficacy of preparticipation screening to timely detect cardiac diseases at risk.1–10 Dispute is mostly focused on the utility of 12-lead ECG in trained individuals, who present a large variety of changes, in most instances related to training and deprived of clinical significance.5–10 Implementation of 12-lead ECG in the preparticipation screening is more controversial in black athletes, who not uncommonly show a broad spectrum of abnormalities of uncertain clinical significance,11–16 and are exposed to greater risk of SCD, due to a large prevalence of undetected hypertrophic cardiomyopathy (HCM).17 ,18

Correct identification of cardiovascular (CV) abnormalities at-risk requires a specific knowledge in black athletes, and particularly in native Africans, who currently represent a relevant and increasing proportion of elite and professional athletes over the planet, with a large proportion of them moving to Europe and the USA. Moreover, appropriate assessment of cardiac risk in this selected population has become a medical priority which conveys substantial economic, social and legal implications.

At present, many obstacles are seen related to the implementation of preparticipation CV screening worldwide for competitive athletes, as suggested by the scientific societies (European Society of Cardiology, ESC) and sport governing organisations (International Olympic Committee and Federation International Football Association, FIFA).5 ,19 ,20 At present, no systematic screening programmes have been realised for native athletes in Africa, and the feasibility and efficacy of such programmes in developing countries remains largely unknown. Therefore, in the present study, we sought to assess the feasibility and efficacy of CV screening in a cohort of native athletes performed directly in the African setting.

Methods

Study population

In the present study, we chose to perform a preparticipation CV screening in Libreville, capital city of Gabon, a West-African sub-Saharan country (figure 1).21

Figure 1

Social, demographic and health characterisation of the Gabon, as derived from web information.21. This figure is only reproduced in colour in the online version.

We examined 210 native male Africans, who were regularly engaged in high-level, competitive football. Specifically, all players in the range 18–22 years of age and members of the Gabonian Football Association were invited to participate. All did agree and there were no subsequent withdrawals. All players were of Gabonian origin.

Players participated in football academies, that is, sport organisations where they trained on a regular basis, under the guidance of professional coaches. Their training programme was scheduled on daily basis and included technical work with the ball and general conditioning programmes; matches were scheduled according to the national football championship timetable. A subset of 17 players (8%) was part of the junior national team, and they gather within the national team throughout most periods of the year.

Each player underwent a CV screening, including athlete's (and family) history, physical examination (PE), 12-lead ECG and transthoracic echocardiography. As in every FIFA precompetition screening we used the official FIFA PCMA form.20 The medical history, PE and ECG were performed with the help of local Gabonian physicians. In all athletes blood pressure measurements were performed at rest, after a few minutes of lying down in supine position. Echocardiography was performed by two experienced cardiologists (CS and FDP). Analysis and interpretation of the ECG and echocardiographic findings was performed by the same cardiologists.

Electrocardiography

The 12-lead ECG was performed with a ‘Schiller AT-101’ (©Schiller AG, Baar, Switzerland) with the subject in supine position after a few minutes of rest during quiet respiration, and recorded at 25 mm/s. ECG analysis was performed according to the present ECG-interpretation standards from the ESC and considered as ‘training-related’ (ie, benign) and ‘training-unrelated’ (potentially expression of CV disease).22 Specifically, we assessed heart rate (bpm), PR interval (s), QRS duration (s), QT interval corrected for the heart rate23 (s), presence of Q waves (≥2 mm in depth, in ≥2 leads), R/S-wave amplitude in precordial leads (S1+R5) (mm), with Sokolow-Lyon criterion for left ventricular hypertrophy (LVH) (positive if >35 mm),24 presence and shape (concave or domed) of ST-segment elevation (≥1 mm from baseline, in ≥2 contiguous leads), presence of ST-segment depression (≥0.5 mm, in ≥2 contiguous leads), T-wave inversion (≥2 mm in depth, in ≥2 contiguous leads, with exclusion of III, aVR,V1). Furthermore, marked QRS axis deviation (right ≥+120°, or left −30° to −90°) or right ventricular enlargement (R/S-wave amplitude >10.5 mV in V1+S5 or 6) were assessed.

Echocardiography

Two-dimensional and Doppler echocardiographic studies were performed with ‘GE Vivid q’ (‘©GE Healthcare, Littel Chalfont, UK’). Images of the heart were obtained in multiple cross-sectional planes by standard transducer positions. M-mode echocardiograms were derived from two-dimensional images under direct anatomic visualisation and recorded at 100 mm/s. Measurements of end-diastolic and end-systolic left ventricular (LV) cavity dimensions, anterior ventricular septal and posterior free wall thicknesses were obtained as recommended.25 LV mass was calculated by the formula of Devereux et al26 and was indexed to body surface area. Right ventricular (RV) area (end of diastole), left atrium (LA) and right atrium (RA) area (end of systole) were measured in the four-chamber view.

Ejection fraction was assessed in the apical four-chamber and two-chamber views and quantified according to modified Simpson rule. Parameters of LV filling were obtained with pulsed Doppler echocardiography, as previously described.27 The extent and distribution of LV hypertrophy was assessed from two-dimensional echocardiography and differentiation of physiological LV remodelling from pathologic conditions (ie, HCM) was based on previously described and widely accepted criteria.28 ,29 Diagnosis of arrhythmogenic cardiomyopathy was based on current, revised criteria30 and LV non-compaction on criteria proposed by Jenni et al31 Mitral valve prolapse was diagnosed on evidence of thickened leaflets (>5 mm) and systolic displacement of >2 mm, in the parasternal long-axis view.32 Finally, a systolic RV/RA-gradient of >30 mm Hg was considered evidence of pulmonary hypertension.

Statistical methods

Data were expressed as mean±SD. Differences between means were assessed with unpaired Student t test. Differences of proportions were assessed by χ2 test. A two-tailed p value <0.05 was considered as statistically significant.

Results

Athletes’ mean age was 18.6 years (range 18–22), height 172±6.4 cm (158–191), weight 64±7.4 kg (46–85) and body surface area 1.7±0.13 m2 (1.43–2.09).

History and PE

A subset of 19 athletes (9%) referred atypical chest discomfort/oppression, not specifically related to physical exercise. None reported syncope or palpitations. Large proportion (110, or 52%) reported ‘flu-like’ symptoms, and 40 (19%) diarrhoea within the last 4 weeks. Family history revealed systemic hypertension in 84 (40%), sudden or unexplained death in 36 (17%).

The PE revealed no relevant findings, except innocent systolic murmurs in 52 (24.7%), that showed no structural abnormalities on echocardiography. Blood pressure was 125±4/72±2 mm Hg. Systolic blood pressure was >140 (and <150) mm Hg in 12 athletes, and >90 (and <100) mm Hg in two.

12-lead ECG

‘Training-related’ ECG changes (table 1). The most common ECG alteration was ST-segment elevation (ie, early repolarisation pattern) observed in 150 (71.4%); in the majority (n=116) with a concave-shaped ST-segment and positive T-wave; in the remaining (n=34) with a dome-shaped ST-segment and biphasic/inverted terminal T-wave restricted to anterior precordial leads V1–V4. The second most frequent alteration was an isolated increase in R/S-wave voltage (suggestive for LV hypertrophy) present in 116 (55.2%). Less frequent findings were incomplete right bundle branch block in 31 (14.8%) and PR interval prolongation >0.20 s in 24 (11.4%).

Table 1

Distribution of ‘training-related’ ECG abnormalities (isolated or in combination) in 210 native African athletes

‘Training-unrelated’ ECG changes (table 2). Of the ECG abnormalities considered potential expression of cardiac disease, inverted T-waves were observed in 10 athletes (4.8%): specifically, in precordial leads V1–V4 in 9 (4.3%), and diffuse in inferior-lateral (III, VF) and precordial (V5, V6) leads in 1 (0.5%). Left atrial enlargement was present in eight (4%), deep Q-waves in three (1.4%), Wolff-Parkinson-White (WPW) pattern in three (1.4%) and ST-segment depression in two (1%). Five athletes had borderline QTc interval prolongation (≥0.44, <0.46 s). The training-unrelated abnormalities were present in 26 athletes, with an overall prevalence of 12.4%.

Table 2

Distribution of ‘training-unrelated’ ECG abnormalities (isolated or in combination) in 210 native African athletes

Eventually, all athletes presenting ECG abnormalities considered possible expression of cardiac disease were referred to the local physicians responsible for medical care of the football teams, with the purpose to enter a clinical follow-up and further investigation. Specific alert was raised for the three athletes with WPW pattern, with clarification of the potential clinical complications of this abnormality.

Echocardiography

LV cavity dimensions were within the normal limits in the vast majority of athletes (177 or 84%), mildly enlarged (end-diastolic diameter 55–59 mm) in a subset (29 or 14%), and markedly enlarged (end-diastolic diameter ≥60 mm) in only four athletes (2%; table 3). LV systolic function was normal in each athlete. A substantial proportion (19 or 9%) had LV wall thickness at upper limits (ie, 12 mm), but none showed increased wall thickness of >13 mm. Left atrial and aortic root remained within the normal dimensional limits in all.

Table 3

ECG and Doppler parameters in 210 native African athletes

Individual analysis showed that one of athletes with borderline wall thickness (12–13 mm) had remarkably small LV cavity size (end-diastolic diameter, 39 mm) and, in association, diffuse negative T-wave in inferior and lateral leads, with an overall clinical picture strongly suggestive for HCM. The other 18 athletes with mild wall thickening (12 mm) had normal or increased LV cavity size, with normal LV function, considered compatible with athlete's heart.

None of the athletes had morphological or functional RV changes suggestive of arrhythmogenic right ventricular cardiomyopathy (ARVC), including the nine with inverted T-waves in anterior precordial leads. None of the athletes with markedly enlarged LV cavity (≥60 mm) showed global or regional LV dysfunction or clinical correlates suggestive of dilated cardiomyopathy.

Other abnormalities included mitral valve prolapse (with mild regurgitation) in three, type II atrial septal defect (with normal-size right atrium) in two and increased pulmonary systolic pressure (44 mm Hg) in one. As an incidental finding, a slight thickening of the aortic valves (possible consequence of unrecognised rheumatic disease) was observed in eight athletes, and a meshwork of false tendons and/or hypertrabeculation in the mid-ventricular and apical LV cavity was seen in 69 (33%). This latter pattern was judged not to fulfil the diagnostic criteria for LV non-compaction, based on normal LV cavity dimension and function, and lack of clinical correlates.32

Relation of the ECG abnormalities to echocardiographic findings

Vast majority of the electrocardiographic ‘training-unrelated’ abnormalities remained deprived of clinical correlates or structural cardiac abnormalities detectable by echocardiography, with the exception of the athlete with HCM, who showed diffuse ventricular repolarisation abnormalities.

In addition, left axis deviation was the only ECG abnormality in one athlete with atrial septal defect, and left atrial enlargement pattern was seen in one with mitral valve prolapse (figure 2).

Figure 2

Flow chart showing the echocardiographic findings in African athletes according to ECG abnormalities, that is, ‘training-related’ (benign) and ‘training-unrelated’ (possibly expression of cardiac disorder). Among athletes showing ECG abnormalities unrelated to training, one hypertrophic cardiomyopathy was diagnosed. This figure is only reproduced in colour in the online version.

Discussion

Preparticipation CV screening has been advocated as a feasible and reliable strategy to identify individuals at risk and reduce the occurrence of SCD in young athletes.33 However, such favourable experience was attained in a cohort of Caucasian athletes, living in a developed European country and perplexity has been raised that similar results might not be duplicated in a different scenario, such as in young athletes living in a developing country.

The present investigation was planned to address the question of feasibility and efficacy of a CV screening performed in an African setting, with participants comprised of young, native Africans. Screening competitive athletes in Africa, where top medical priorities are different21 may be seen as an eccentric viewpoint. However, it can also be anticipated that for the young African generation participation and achievement in professional sport represent a fast track to gain public recognition, escape poverty and progress in social classes. As a consequence, issues related to competitive/professional sport are surrounded by far large social attention than we commonly perceive.

From our experience, certain considerations emerged, as outlined here. First, history revealed mostly atypical symptoms (in 9%), larger than previously reported in Caucasian athletes (5%),34 which ultimately remained deprived of clinical significance.

The 12-lead ECG showed large proportion of innocent changes, consequence of athletic training, such as isolated increase of R/S-wave voltages suggestive of LV hypertrophy (in 55.2%) and ST-segment elevation (in 71.4%). This latter pattern presented a concave shape with positive T-wave in the majority of athletes (as described in Caucasians35); however, a substantial subset of Africans (16%), larger than Caucasians (3%),16 showed a convex, domed pattern, with terminal negative T-waves in anterior precordial leads. The large proportion of domed ST-segment elevation in Africans seems to be a race-related finding,15 ,16 although the causative mechanisms remain to be elucidated.

Overall, large majority (about 90% in our cohort) of native African athletes presented a variety of ECG changes considered ‘training-related’, as previously reported in athletes of African-American and Afro-Caribbean descent.12–15 These alterations (ie, isolated voltage criteria for LVH, and ST-segment elevation) were unrelated to the presence of cardiomyopathies or other cardiac diseases, and in the setting of preparticipation screening should not be considered an indication for additional diagnostic testing, as it has been already stated for Caucasian athletes.20

On the other hand, a substantial proportion of ECG abnormalities (about 12%) were considered unrelated to the training according to the present ECG interpretation,22 and raised concern for possible underlying cardiac disease. Most of these alterations comprised T-wave inversion confined to anterior precordial leads (V1–V4), that after thorough clinical and echocardiographic investigation remained deprived of clinical significance or structural correlates (specifically, arrhythmogenic cardiomyopathies). Our experience supports the view that T-waves inversion confined to anterior precordial leads may be a distinctive racial trait in native Africans, as observed in black athletes of second/third generation living in Europe.15 However, we recommended continued clinical surveillance and periodical investigation.

On the other hand, a small minority of athletes (0.5%) showed deep and diffuse T-wave inversion, which was associated with HCM. This observation suggests that such repolarisation abnormalities mandate careful diagnostic investigation and continued clinical surveillance in African athletes, as already recommended in Caucasians.36

Overall, the proportion of abnormal ECG patterns, unrelated to training and raising clinical concern was double in young African compared with Caucasian athletes (12% vs 5–7%),34 ,37 confirming that implementation of ECG screening in multiethnic populations, including large proportion of black individuals, is more problematic than in a homogeneous Caucasian population, and requires better understanding of race-related ECG differences.

LV cavity dimensions were within normal values in the vast majority of athletes, and markedly enlarged in only 2%, that is, a lower proportion than observed in Caucasian athletes (14%).38 This finding can be explained by the relatively low body size of our cohort of young athletes, in consideration of the close interdependence of LV cavity size and anthropometric characteristics.38 ,39 However, this finding may also support the hypothesis that native black athletes develop, in response to exercise training, a more concentric LV remodelling compared with Caucasians.13 ,16 Clinical implication derived from this observation is, therefore, that a distinctly eccentric LV remodelling in an African athlete should be viewed with caution, and considered likely as expression of a non-physiological condition.

It is noteworthy that none of the athletes presented a marked increase in wall thickness above 13 mm. This finding, unexpected in a cohort of African athletes, was likely due to the average small body size of the Bantu football players,16 as well as the type of sport (ie, football), which does not represent physiologic stimulus for substantial LV wall thickening.28 ,29 These observations suggest that the question of differential diagnosis between athlete's heart and HCM in native African athletes may be less common than believed to be, and can be managed as previously suggested.29

In the context of our screening, routine echocardiography identified one athlete with HCM (who also showed ECG abnormalities); in addition, only a small subset (about 5%) of athletes showed structural abnormalities, including mitral valve prolapse, atrial septal defect and pulmonary hypertension. Therefore, in our cohort routine echocardiography added only a mild contribution to the identification of cardiac conditions at risk and, in our opinion, should not be considered as routine testing in African athletes, such as in Caucasians.16 ,35 ,37

Limitations

Our study conveys certain limitations, inherent to the study design. First, the study population was comprised of a selected cohort of elite male football players, which cannot be considered representative of the entire African athletes population, encompassing a variety of subjects with broad range of age, types of sport participated in and both genders. Indeed, our study design did not include CMR, genetic testing and family screening and, therefore, we cannot exclude with certainty the presence of HCM (or other cardiomyopathies) in athletes with abnormal ECG findings. Finally, our study was the first experience of screening in an African country and designed as ‘cross-sectional’ pilot study; therefore, we have no data derived from longitudinal follow-up to support our medical management. However, this study is relevant because represents the first attempt to implement the CV screening in the real-world of a developing country, and limitations here quoted were unavoidable part of the study design.

In conclusion, about 12% of native African athletes showed ECG abnormalities potentially suggestive of cardiac disease, requiring additional testing and periodical follow-up. Structural cardiac abnormalities, however, were found in a small minority (5%), including HCM in one, but no ARVC or dilated cardiomyopathy. Therefore, this study demonstrates that preparticipation CV screening is efficient enough to identify (or raise suspicion) for CV abnormalities in native African athletes, but remains challenging for conclusive identification of cardiac diseases in the difficult scenario of a developing African country.

What are the new findings?

  • The assessment of cardiac risk in athletes has become a medical priority, which also conveys substantial economic, social and legal implications. Indeed, correct identification of abnormalities at-risk and appropriate cardiovascular (CV) management requires a specific knowledge in black athletes, and particularly in native Africans, who currently represent a relevant and increasing proportion of elite and professional athletes over the planet, with larger proportion of them moving to Europe and the USA.

  • So far, no systematic CV screening programmes have been implemented for native athletes in Africa, and the feasibility and efficacy of such programmes remains largely unknown. The novelty of the present study is the study population, comprised of native black athletes examined directly in the setting of a developing country, and the information relative to the ECG and cardiac morphological features of native African athletes.

  • Overall, about 90% of native African athletes presented ECG changes considered ‘training-related’, (ie, isolated voltage criteria for LVH, and ST-segment elevation) that were unrelated to the presence of cardiomyopathies or other cardiac disease, and in the setting of preparticipation screening should be viewed as benign consequence of athletic training and should not represent indication for additional diagnostic testing.

  • About 12% of young native African athletes showed abnormal ECG patterns, unrelated to training and raising indication for additional testing and follow-up. This proportion is about double compared with that of Caucasian athletes.

  • In the context of our screening, echocardiography identified one athlete with HCM (who also showed ECG abnormalities); in addition, only a small subset (about 5%) with structural abnormalities, including mitral valve prolapse, atrial septal defect and pulmonary hypertension. Therefore, in our cohort echocardiography added only a modest contribution to the identification of conditions at risk.

How might it impact clinical practice in the near future?

  • This study demonstrates that preparticipation CV screening is efficient to identify (or raise suspicion) for CV abnormalities in native African athletes, but challenging for conclusive identification of cardiac diseases in the difficult scenario of a developing African country.

  • ECG abnormalities not related to training and potentially expression of cardiac disease were found in about 12% of athletes and remained largely deprived of structural correlates on echocardiography and without apparent clinical significance. Therefore, implementation of CV screening in African countries remains a benevolent project, but requires additional efforts to solve the ambiguities of ECG abnormalities in a substantial subset of athletes.

  • This study shows several difficulties that physicians encounter when performing a CV screening of native African athletes, including the correct understanding and appreciation of the cardiac symptoms, the evaluation of certain ECG abnormalities, such as the inverted T-waves in anterior precordial leads that are found in a large proportion of individuals, the management of uncertain morphologic abnormalities, such as borderline LV hypertrophy in young athletes, in whom information derived from genetic analysis and long-term follow-up are usually missing.

Acknowledgments

The authors gratefully acknowledge FIFA (Fédération Internationale de Football Association) for the funding of the study. The authors appreciate the Fédération Gabonaise de Football, the Football League in Gabon and the technical staffs for their support. They also thank Pr Cyrille Mouyopa, Dr Dissouva and Dr Moukendi and their staff in Clinique Médicale Moukouti in Libreville (Gabon). They owe a considerable debt of gratitude to the players who participated in the examinations. Finally, they are indebted to Filippo M. Quattrini, MD, PhD for the careful revision of the manuscript and pertinent suggestions.

References

View Abstract

Footnotes

  • Contributors CS substantially contributed to the study design, acquisition of the data, analysis and interpretation of the data, drafting and revising the manuscript, approval of the final version. FMDP substantially contributed to the acquisition of the data, analysis and interpretation of the data, drafting and revising the manuscript, approval of the final version. AYZ substantially contributed to the study design, support to the logistics, acquisition of the data and, revising the manuscript, approval of the final version. JD substantially contributed to the conception and design of the study; interpretation of the data, support to the logistics, revising the manuscript, approval of the final version. AP substantially contributed to the study design, analysis and interpretation of data, drafting and revising the manuscript, approval of the final version.

  • Funding This study was funded by the Fédération Internationale de Football Association (FIFA).

  • Competing interests None.

  • Patient consent Obtained.

  • Ethics approval The study design was approved by the institutional review board, and the Conseil National de ĹOrdre des Medecins, Algeria.

  • Provenance and peer review Not commissioned; externally peer reviewed.

  • ▸ References to this paper are available online at http://bjsm.bmj.com