Echocardiographic Screening of Anomalous Origin of Coronary Arteries in Athletes with a Focus on High Take-Off

Anomalous aortic origin of coronary arteries (AAOCA) represents a rare congenital heart disease. However, this disease is the second most common cause of sudden cardiac death in apparently healthy athletes. The aim of this systematic review is to analyze the feasibility and the detection rate of AAOCA by echocardiography in children and adults. A literature search was performed within the National Library of Medicine using the following keywords: coronary artery origin anomalies and echocardiography; then, the search was redefined by adding the keywords: athletes, children, and high take-off. Nine echocardiographic studies investigating AAOCA and a total of 33,592 children and adults (age range: 12–49 years) were included in this review. Of these, 6599 were athletes (12–49 years). All studies demonstrated a high feasibility and accuracy of echocardiography for the evaluation of coronary arteries origin as well as their proximal tracts. However, some limitations exist: the incidence of AAOCA varied from 0.09% to 0.39% (up to 0.76%) and was lower than described in computed tomography series (0.3–1.8%). Furthermore, echocardiographic views for the evaluation of AAOCA and the definition of “minor” defects (e.g., high take-off coronary arteries) have not been standardized. An echocardiographic protocol to diagnose the high take-off of coronary arteries is proposed in this article. In conclusion, the screening of AAOCA by echocardiography is feasible and accurate when appropriate examinations are performed; however, specific acoustic windows and definitions of defects other than AAOCA need to be standardized to improve sensitivity and specificity.

The aim of the present manuscript is to systematically review the feasibility and the detection rate of echocardiography for the screening of AAOCA, with special attention paid to high take-off, and to highlight its strengths and limitations.

Search Strategy
The search strategy and study selection were carried out according to PRISMA guidelines. Potential publications were identified from a systematic search in the National Library of Medicine (PubMed access to MEDLINE citations; http://www.ncbi.nlm.nih. gov/PubMed/)/(accessed on 24 June 2020). The search strategy included a mix of medical subject headings and free text terms for the key concepts, starting from coronary artery origin anomalies and echocardiography. The search was further refined by adding the keywords athletes, children, and high take-off. In addition, we identified other potentially relevant publications using a manual search of references from all eligible studies and review articles, as well as from the Science Citation Index Expanded on the Web of Science. Two reviewers (M.C., M.K.) independently assessed all identified reports, and a consensus was reached for inclusion in the analysis. Titles and abstracts of all articles identified by the above search strategy were evaluated. Studies were excluded if they: (i) evaluated children with congenital heart defects (CHDs) that are often associated with coronary artery anomalies (e.g., transposition of the great arteries, tetralogy of Fallot), (ii) were focused on anomalous origin of the coronary artery from pulmonary artery (ALCAPA), Kawasaki or coronary artery fistula, (iii) used imaging modalities other than echocardiography (e.g., CT scan), and (iv) were written in languages other than English.

Feasibility of Visualizing Coronary Arteries' Origins
A good feasibility of visualizing origin of coronary arteries was described by most of the authors [15,19,21,24,31], including studies published in the 1990s [19,20]; in three studies, the feasibility was not reported [13,22,23]. Overall, the feasibility of coronary artery origin visualization varied from to 90% [20] to 98.5% [21] and was higher in children compared to adults [21]-see Table 2. The correct visualization of the left common artery (LCA) origin was considered feasible for 98% [15] to 100% [31] of the subjects, while the feasibility for the visualization of right coronary artery (RCA) origin varied from 80% [19] to 96% [24]. When the visualization refers to the proximal course of the coronary arteries, the feasibility was lower, ranging from 81-82% [15] to 98.5% [21].

Echocardiographic Views Employed to Visualize Origin of Coronary Artery
The origin of coronary arteries was assessed in the parasternal short-axis view by all the authors [13,[19][20][21][22][23][24]31], which is well in accordance with recent guidelines [26]. Parasternal long-axis [13,23] or para-sagittal planes [23] views, also recommended [26], were employed only by two authors. Only two studies [21,22] used the color Doppler with reduced color gain (e.g., 15-40 cm/s) to detect coronary flow. Thus, the use of specific protocols for the evaluation of coronary artery origin may increase the detection rate of anomalies; indeed, Thankvel and colleagues [22] reviewed their experience before and after the introduction of a new screening method for the evaluation of AAOCA that extends the parasternal short-axis view into the ascending aorta in children and adolescents. They found that the detection rate of AAOCA improved from 0.02% (in 5669 subjects screened) to 0.22% (in 6428 subjects screened).

Detection Rate of Major and Minor AAOCAs by Echocardiography
Major AAOCAs were usually defined as RCA and LCA origins from opposite wrong sinus ( Figure 1), a single ostium coronary artery, or left circumflex artery (CFx) originating from right coronary sinus [14,15,20,23,24,31]. The detection rate of major AAOCAs greatly varied among the different studies, from 0.0% [15,24,31] to 0.09% [20] and up to 0.76% [23]. The positive predictive value of echocardiography (with confirmation at either coronary angiography or CT angiography) in the diagnosis of major AAOCAs was high, varying from 87.5% [21] to 100% [20]. Minor AAOCAs were described for 1.5% [13] to 2.6% [19] of the cases: they included separate ostia for left anterior descending artery and CFx from left sinus, two distinct ostia in the right sinus for RCA and the conus branch, and a short left main coronary artery-<5 mm-or small fistulas. A case of minor AAOCA from our case series, with two distinct ostia originating from the left sinus, is shown in Figure 2. However, definition and clinical significance of minor and major AAOCAs varied among the studies. Indeed, the study by Lytrivi and colleagues [23], reporting the highest number of AAOCAs (111 patients out of a cohort of 14,546 subjects), did not distinguish between major and minor defects. Gerling et al. included into low-risk AAOCAs also high take-off of coronary arteries with acute angle that may be at risk of SCD [33]. Four studies did not evaluate minor AAOCAs [15,22,24,31]. Minor AAOCAs were described for 1.5% [13] to 2.6% [19] of the cases: they included separate ostia for left anterior descending artery and CFx from left sinus, two distinct ostia in the right sinus for RCA and the conus branch, and a short left main coronary artery-<5 mm-or small fistulas. A case of minor AAOCA from our case series, with two distinct ostia originating from the left sinus, is shown in Figure 2. However, definition and clinical significance of minor and major AAOCAs varied among the studies. Indeed, the study by Lytrivi and colleagues [23], reporting the highest number of AAOCAs (111 patients out of a cohort of 14,546 subjects), did not distinguish between major and minor defects. Gerling et al. included into low-risk AAOCAs also high take-off of coronary arteries with acute angle that may be at risk of SCD [33]. Four studies did not evaluate minor AAOCAs [15,22,24,31].

Symptoms, ECG, Stress Testing, and Clinical Management
Information about clinical data, including the indication to echocardiography, is reported in only four studies [13,20,21,23] (Table A1 in Appendix A). Most of the examinations were just screenings [13,20,21,23], while in a limited case there was a clinical indication. Among the 59 AAOCAs evaluated in these studies, only 10 patients presented symptoms (chest pain and/or dyspnea) and one case was resuscitated after cardiac arrest related to ventricular fibrillation; in this case, a major AAOCA was found, with LCA originating from the wrong sinus (e.g., the right sinus of Valsalva). The basal electrocardiogram was completely normal [13,20] or showed no specific defects [21,23], such as left or right ventricular hypertrophy, T wave inversion or St depression in V5-V6, left-axis deviation [21,23] in these symptomatic subjects. A stress test was performed only in 23 out of 59 AAOCAs subjects, being positive only in five cases [13,20,21]. Myocardial scintigraphy was employed as stress testing modality and resulted positive in 4 out of 9 cases [20,21]. Unroofing surgery was performed in 10 cases [21,23] (including two with positive scintigraphy, one urgent case and seven cases where results of stress tests are not available), while two authors reported indication for sport eligibility [13,20]. Athletes with major AAOCAs were disqualified [20], while athletes with high RCA take-off did not undergo sports restriction [13].

Symptoms, ECG, Stress Testing, and Clinical Management
Information about clinical data, including the indication to echocardiography, is reported in only four studies [13,20,21,23] (Table A1 in Appendix A). Most of the examinations were just screenings [13,20,21,23], while in a limited case there was a clinical indication. Among the 59 AAOCAs evaluated in these studies, only 10 patients presented symptoms (chest pain and/or dyspnea) and one case was resuscitated after cardiac arrest related to ventricular fibrillation; in this case, a major AAOCA was found, with LCA originating from the wrong sinus (e.g., the right sinus of Valsalva). The basal electrocardiogram was completely normal [13,20] or showed no specific defects [21,23], such as left or right ventricular hypertrophy, T wave inversion or St depression in V5-V6, left-axis deviation [21,23] in these symptomatic subjects. A stress test was performed only in 23 out of 59 AAOCAs subjects, being positive only in five cases [13,20,21]. Myocardial scintigraphy was employed as stress testing modality and resulted positive in 4 out of 9 cases [20,21]. Unroofing surgery was performed in 10 cases [21,23] (including two with positive scintigraphy, one urgent case and seven cases where results of stress tests are not available), while two authors reported indication for sport eligibility [13,20]. Athletes with major AAOCAs were disqualified [20], while athletes with high RCA take-off did not undergo sports restriction [13].

Coronary Artery High Take-Off
High take-off coronary artery is a rare anomaly [33][34][35][36][37][38][39][40][41][42][43] that may present in isolation or associated with other congenital cardiac malformations [35][36][37][38][39][40][41][42][43], mainly identified for the RCAs (up to 84.46% of cases) [33]. There is still limited literature on the visualization and definition of RCA high take-off by echocardiography [13,14,23]; although there is not a consensus on the definition of high take-off by echocardiography, all the studies included in the present review defined high take-off as an origin above or distal to the sinutubular junction (STJ) [13,14,23]. An example of high take-off of RCA from our case series is reported in Figure 3. A very recent [13] study, of 1045 consecutive elite adolescent football players, identified coronary high take-off origin in 13 subjects (i.e., 1.14%). Eccentric RCA origin with a high take-off and partial intra-arterial course was observed in two cases (with no slit-like ostium and no intramural course); high take-off origin of the RCA with acute angle was observed in one case; high take-off of the RCA origin (with no intramural or slit-like orifice) was observed in 11 cases. Of the latter 11 cases, diagnosis was feasible only from the parasternal long-axis view, where the ostium of the RCA was measured from 2.3 to 6.8 mm above the sinutubular junction [13]. Lytrivi et al. [24] documented RCA high take-off in 53 cases (0.36%), LCA high take-off in four cases, and high take-off of both the coronary arteries in two cases of a valuable cohort of 14,546 pediatric subjects [23].
Healthcare 2021, 9, x 8 of 15 included in the present review defined high take-off as an origin above or distal to the sinutubular junction (STJ) [13,14,23]. An example of high take-off of RCA from our case series is reported in Figure 3. A very recent [13] study, of 1045 consecutive elite adolescent football players, identified coronary high take-off origin in 13 subjects (i.e., 1.14%). Eccentric RCA origin with a high take-off and partial intra-arterial course was observed in two cases (with no slit-like ostium and no intramural course); high take-off origin of the RCA with acute angle was observed in one case; high take-off of the RCA origin (with no intramural or slit-like orifice) was observed in 11 cases. Of the latter 11 cases, diagnosis was feasible only from the parasternal long-axis view, where the ostium of the RCA was measured from 2.3 to 6.8 mm above the sinutubular junction [13]. Lytrivi et al. [24] documented RCA high take-off in 53 cases (0.36%), LCA high take-off in four cases, and high take-off of both the coronary arteries in two cases of a valuable cohort of 14,546 pediatric subjects [23]. There is relatively limited literature on high take-off of LCA [26], and echocardiographic reports are extremely limited [14,23]. Only one study among the nine included in this review reported four cases of high LCA. Of these, a normal intracardiac anatomy was found in one case and associated defects (including one ventricular septal defect, one patent arterial duct, one aortic coarctation) were found in three cases [23]. No There is relatively limited literature on high take-off of LCA [26], and echocardiographic reports are extremely limited [14,23]. Only one study among the nine included in this review reported four cases of high LCA. Of these, a normal intracardiac anatomy was found in one case and associated defects (including one ventricular septal defect, one patent arterial duct, one aortic coarctation) were found in three cases [23]. No clear indications for the appropriate acoustic window to be used for evaluating the LCA take-off were reported in this study.
Discrepancies exist in the definition of major and minor AAOCAs, especially regarding the classification of a high take-off of coronary artery origin. Unfortunately, limited echocardiographic data are currently available in the literature [13,14,23]. Furthermore, most of the CT studies adopted a different definition as compared to echocardiographic articles. Indeed, in most cases, CT studies defined high take-off as a height >1 cm or >20% the depth of the sinus above STJ [26], a height >0.25 cm above the sinutubular junction, and a minority any height above the STJ [26]. Echocardiographic studies all used the latter definition, identifying the high take-off as an origin above or distal to the STJ [13,14,23]. The use of different cut-offs to define a high take-off has the consequence of affecting prevalence [33]; indeed, in CT studies where the definition of a height >1 cm or >20% the depth of the sinus above sinutubular junction was used, the incidence of RCA high take-off was 0.202%. Conversely, in those that employed a height >0.25 cm above the sinutubular junction as a definition, it decreased to 0.199% [33]. When the high take-off of RCA was defined as any height of origin above STJ, the prevalence of this defect increased [33] up to 0.364% [26]. In echocardiographic studies that used the latter definition, the reported incidence of RCA high take-off was even higher than in CT studies, ranging from 0.36% [23] to 1.14% [13]. This definition may overestimate the prevalence of high take-off but, most importantly, it may have the consequence to classify benign variants as malignant anomalies potentially at risk of SCD [33]. As recently reviewed [26], 3 of 12,899 (0.023%) cases of high take-off coronaries that originated more than one centimeter above the sinutubular junction were associated with SCD. Notably, although preferable in adults, the use of fixed criteria (such as 1 cm above the sinutubular junction) may have relevant limitations in children where aortic dimensions are smaller than in adults; therefore, some authors [44] proposed the adoption of relative criteria, such as coronary orifices that arise 120% or more of the depth of the sinus of Valsalva or 20% or more the depth of the sinus above the STJ. A comprehensive evaluation of a high take-off of coronary arteries should also include other important characteristics, such as the presence of slit-like ostium, stenosis, the interarterial course, and intramural course [26,[33][34][35]. Notably, a high take-off associated with acute angulation from the aorta, an intramural or an interarterial course comprising 4% of the defects, are more at risk for the development of SCD [26]. This specific characteristic may be studied by echocardiography [13], but this imaging technique has inherent limitations for an accurate definition of these characteristics that need a comprehensive multimodality approach for their definition, as suggested by the current recommendations [26]. Furthermore, based on previously published studies, no clear indications exist for the appropriate acoustic window to be used for other coronary anomalies, such as the high take-off of coronary arteries, and particularly for LCA. Current recommendations [26] suggest using a coronal subcostal view to visualize LCA origin; however this window can be easily utilized only in neonates and children. We propose in this article an additional acoustic window that may allow for evaluation of high take-off of LCA by using a modified apical five-chamber view (see Figure 4). So far, we have tested this projection in a limited number of cases with suspicion of LCA high take-off or when origin of LCA was not seen by conventional short-axis and parasternal long-axis views, with encouraging results.
published studies, no clear indications exist for the appropriate acoustic window to be used for other coronary anomalies, such as the high take-off of coronary arteries, and particularly for LCA. Current recommendations [26] suggest using a coronal subcostal view to visualize LCA origin; however this window can be easily utilized only in neonates and children. We propose in this article an additional acoustic window that may allow for evaluation of high take-off of LCA by using a modified apical five-chamber view (see Figure 4). So far, we have tested this projection in a limited number of cases with suspicion of LCA high take-off or when origin of LCA was not seen by conventional short-axis and parasternal long-axis views, with encouraging results.
Current American Heart Association/American College of Cardiology (AHA/ACC) guidelines [52] suggest that, in asymptomatic athletes with a coronary artery originating from the wrong sinus of Valsalva and negative stress test, permission to compete can be considered after adequate counseling (Class IIa; Level of Evidence C) [52]. However, when the artery passes between the pulmonary artery and aorta, athletes should be restricted from participation in all competitive sports, except for Class IA sports, before surgical repair, independently from the presence of symptoms (Class III; Level of Evidence B) [52]. If athletes with AAOCA exhibit symptoms, arrhythmias, or signs of ischemia on exercise stress test, they should be restricted from participation in all competitive sports, except for Class IA sports, before a surgical repair (Class III; Level of Evidence C) [52]. The Italian guidelines [53] suggest restricting participation in competitive sports in case of a coronary artery originating from the wrong sinus of Valsalva; conversely, in case of an anomalous The clinical management of patients with AAOCAs is challenging, particularly in asymptomatic subjects practicing sport [44,[47][48][49][50], and discrepant approaches (e.g., surgery vs. follow-up with or without exercise restriction) [49][50][51][52] have been adopted so far. Indeed, athletes with AAOCAs are often asymptomatic [20,21] or with mild symptoms (e.g., chest discomfort, palpitations), and stress testing can be frequently normal (or with minor abnormalities) [9,44]. Mery et al. [47] proposed a protocol where only patients with symptoms ascribed to ischemia (aborted SCD, syncope during or following exercise), or asymptomatic high risk anatomy (intramural, abnormal ostium) or with established perfusion defects undergo cardiac surgery; otherwise, no intervention neither exercise restriction is recommended.
Current American Heart Association/American College of Cardiology (AHA/ACC) guidelines [52] suggest that, in asymptomatic athletes with a coronary artery originating from the wrong sinus of Valsalva and negative stress test, permission to compete can be considered after adequate counseling (Class IIa; Level of Evidence C) [52]. However, when the artery passes between the pulmonary artery and aorta, athletes should be restricted from participation in all competitive sports, except for Class IA sports, before surgical repair, independently from the presence of symptoms (Class III; Level of Evidence B) [52]. If athletes with AAOCA exhibit symptoms, arrhythmias, or signs of ischemia on exercise stress test, they should be restricted from participation in all competitive sports, except for Class IA sports, before a surgical repair (Class III; Level of Evidence C) [52]. The Italian guidelines [53] suggest restricting participation in competitive sports in case of a coronary artery originating from the wrong sinus of Valsalva; conversely, in case of an anomalous origin of the left circumflex coronary (CFx) from the right sinus of Valsalva and demonstrated absence of myocardial ischemia, competitive sports are permitted with a tailored decision with a case-by-case approach [53].

Limitations
Our research was focused on 2D transthoracic echocardiography; therefore, studies evaluating the origin of coronary arteries by three-dimensional and/or transesophageal echocardiography were not included in the present review. There has been limited application [54] of 3D echocardiography for the screening of coronary arteries. Despite 3D echocardiography potentially allowing a better visualization of coronary anomalies in a good acoustic window, 2D echocardiography has greater sensitivity [54].
Studies included in this review were limited, heterogeneous, made use of different technology and expertise, and had differences in terms of study populations. Therefore, data were too heterogeneous to perform a valid meta-analysis. However, interesting data on the feasibility and accuracy of the echocardiographic analysis in this field can be collected and interpreted and may represent food for thought. Data for some defects (particularly a high take-off of LCA) are too limited to draw any type of conclusion regarding the feasibility and the accuracy of echocardiography for diagnosis. Additionally, the projection we propose for LCA high take-off visualization (e.g., modified apical five-chamber view) needs to be validated in wider, prospective studies. Furthermore, some defects (such as myocardial bridge, intramural course) cannot be easily visualized by echocardiography and represent a limitation of transthoracic echocardiographic for the estimation of coronary anomalies in terms of coronary course in the myocardium.

Conclusions
Echocardiographic evaluation of the origins of coronary arteries by transthoracic echocardiography is feasible and accurate. The use of systematic protocols including different acoustic windows in addition to a basic short-axis view (e.g., parasternal, parasagittal views, short axis extended into the ascend ending aorta) is essential for the optimization of imaging for congenital coronary artery anomalies and for improving their detection by echocardiography which, at present, remains suboptimal compared to other imaging modalities such as CCTA or CMR. The definitions of some anomalies (such as a high take-off) need to be standardized and the clinical significance of AAOCAs should be clarified by further research. Since data on LCA high take-off assessed by echocardiography are extremely limited, a method to visualize this defect has been proposed in the present article.

Perspectives
Since abnormal origins of coronary arteries are the second most common cause of sudden cardiac death in apparently healthy athletes [1][2][3]55], the echocardiographic evaluation of origin and proximal course of coronary arteries should be a fundamental part in screening of an athlete. However, the definitions of some coronary artery origin anomalies (such as a high take-off) and their clinical significance have not been completely defined yet. There is a need for recommendations for the definition and the clinical risk classification of AAOCAs as well as for the decision making, including sports eligibility or disqualification. In fact, although concealed life-threatening abnormalities of the coronary arteries should be diagnosed, some defects are benign and it is important to avoid creating unjustified anxiety, using invasive and expensive examinations, or indicating sport restriction when functionally benign defects are demonstrated.
Funding: This research received no external funding.

Institutional Review Board Statement: Not applicable.
Informed Consent Statement: Not applicable.

Data Availability Statement:
The datasets used for the current study are available from the corresponding authors upon reasonable request.

Conflicts of Interest:
The authors declare that they have no competing interests. The authors declare no conflict of interest.