Anomalous origin of coronary arteries is the second most common congenital cause of sudden cardiac death (SCD) in the young, accounting for 17% of young athlete deaths in the United States (14–16). The incidence of coronary anomalies is as high as 1.3% in adults with suspected coronary artery disease, only a fraction considered potentially malignant, but is unknown in the young, asymptomatic, athletic population (22,24). Evaluation of coronary ostia by transthoracic echocardiogram (TTE) remains a standard part of pediatric, although not often adult, echocardiographic protocols and is generally effective in establishing proximal coronary artery anatomy (4–11,20). Some centers have advocated for limited echocardiograms in the screening of college athletes, which generally focus on views capable of evaluating several other common pathologies, such as hypertrophic cardiomyopathy (1,18,21,23). Although efficient with time and resources, such studies may miss important causes of SCD (2). Abnormalities found on TTE may then be more completely evaluated by computed tomography angiography, coronary angiography, or cardiac magnetic resonance imaging (4,20). Many coronary origin anomalies are then amenable to surgical correction (7,17,19).
There are limited data showing successful identification of coronary artery origins of athletes screened and successful identification varied greatly among studies from 71% to 100% (1,13,21). To our knowledge, no published studies have evaluated the ability to identify both coronary origin and proximal course or its interobserver reliability by TTE in the screening of athletes. The purpose of this study was to evaluate the ability of a screening TTE to reliably determine the origin and proximal course of the coronary arteries in intercollegiate athletes.
METHODS
The current athlete preparticipation screening at our institution consists of a two-part evaluation. First, a detailed medical history, family history, and physical examination is performed by a primary care sports medicine physician, which is consistent with the American Heart Association recommended Preparticipation Physical Evaluation (3). A cardiac evaluation follows, consisting of a standard resting 12-lead ECG and a comprehensive congenital TTE. TTE protocols were performed by three registered sonographers: two employed as adult sonographers and one as a pediatric and fetal sonographer, all at our institution. Studies were read real time by a team cardiologist (RWB) with experience in congenital heart disease and sports cardiology. Studies were performed on the portable laptop-sized Philips CX-50 with an S-5n probe. TTE protocol includes the following two-dimensional views, with color, pulse, and continuous Doppler used where indicated: subcostal coronal and sagittal, apical four and five chamber, parasternal long, parasternal short, and suprasternal. The cardiologist documented normal coronary arteries at the time of the study if defined right coronary artery (RCA) and left coronary artery (LCA) ostia from the appropriate aortic sinuses were clearly observed. The TTE images and ECG were saved with the athletes chart for documentation and reference.
An institutional review board–approved retrospective review of all incoming National Collegiate Athletic Association Division-I athletes participating in varsity sports at our institution during the 2013–2014 academic year was performed. Inclusion criteria included the following: current student, age 18 to 24 yr, and performance of a screening TTE from June 1, 2013, through March 1, 2014. Medical and family history, demographics, and sport were recorded along with bedside echocardiographic interpretation. Two board-certified pediatric cardiologists independently reviewed the TTE images of every athlete that met inclusion criteria. Neither cardiologist had previously viewed the images, and each was blinded to the other’s results. Each reviewer documented observation of the RCA and LCA ostia, LCA bifurcation, and measurable course of the LCA and RCA. Measurable course was defined as the maximum continuously observable distance of coronary artery from a clearly delineable ostium from a paused frame or still image from that athlete's TTE. If no ostium is visualized, measurable course was assigned a value of zero. An adequate observation of the proximal course was defined as clear presence of the right ostia with continuous delineation of 1 cm of arterial lumen for the RCA and clear presence of the left ostia with continuous delineation of either 1 cm of arterial lumen or bifurcation for the LCA. The proximal course was reported and evaluated as a discrete variable and measurable course as a continuous variable measured in cm.
The coronary artery measurements of the TTE review were compared among sonographers as well as athlete size, gender, ethnicity, and sport. Athlete demographics were analyzed with simple descriptive statistics. Each reader’s results of the TTE image review were combined and compared. Echocardiogram review results were analyzed for interobserver variability using proportions of agreement and kappa statistic for the categories of ostia and proximal course and paired t-test for measurable length. Results for athlete demographics were compared using the Cochran–Mantel–Haenszel test for ostia and proximal course and repeated measures for measurable length, and results for athlete size were compared using repeated measures for ostia and proximal course and Pearson’s correlation for measurable length. P values <0.05 were considered statistically significant. Statistics were calculated with SAS, version 9.4, with the nondefault Monte Carlo option used for obtaining Fisher’s exact test.
RESULTS
Chart review identified 146 athletes. Sixty-nine (47%) were female and 77 (53%) were male. Twenty-two sports and six ethnicities were represented (see Table 1). Athlete weight varied from 41.2 to 149.5 kg (mean = 78.9 kg, SD = 21.3 kg), height from 147.5 to 208.3 cm (mean = 178.1 cm, SD = 10.6 cm), body surface area (BSA) from 1.4 to 2.8 m2 (mean = 2.0 m2, SD = 0.29 m2), and body mass index (BMI) from 16.0 to 40.9 kg·m−2 (mean = 24.5 kg·m−2, SD = 4.4 kg·m−2). A total of three registered sonographers performed the echocardiograms throughout the academic year. No coronary anomalies were found.
;)
TABLE 1:
Athlete characteristics.
Both readers agreed on and observed the RCA ostium in 143 athletes (98%) and the LCA ostium in 143 athletes (98%); the RCA ostia was not identified by either reader in one athlete (0.7%), and the LCA ostia was not identified in three athletes (2.1%); and the readers disagreed on visualization of the RCA ostia in two athletes (1.4%) and the LCA ostia in none of the athletes. There was 85.6% agreement with a kappa of 0.5 (moderate strength) for the RCA ostia and 100% agreement with a kappa of 1.0 (perfect strength) for the LCA ostia. Both readers agreed on and observed the RCA proximal course in 119 athletes (81.5%) and LCA proximal course in 118 athletes (80.8%); the RCA proximal course was not visualized by either reader in 6 athletes (4.1%), and the LCA proximal course was not identified in 6 athletes (4.1%); and the readers disagreed on visualization of the RCA proximal course in 21 athletes (14.4%) and the LCA proximal course in 22 athletes (15.1%). There was 85.6% agreement with a kappa of 0.30 (fair strength) for the RCA proximal course and 84.9% agreement with a kappa of 0.28 (fair strength) for the LCA proximal course. The readers had much better agreement determining LCA proximal course by a measurable course ≥1 cm (77.4% agreement) than by visualization of the LCA bifurcation (24.7% agreement). The average measurable course was 2.1 cm for the RCA and 1.5 cm for the LCA (see Figs. 1 and 2). There was interobserver variability for RCA measurable course between reader 1 and reader 2 (2.5 and 1.7 cm, respectively), but not for LCA measurable course (1.6 and 1.4 cm, respectively). All sonographers found coronary artery origins in nearly all athletes. The proximal course of either coronary was shown in ≥80% of studies for all sonographers, and the average measurable course of was ≥1.4 cm for either coronary artery (see Table 2).
FIGURE 1:
Right and left coronary origins by screening TTE. An example image from one athlete’s screening echocardiogram illustrating the origin (asterisk) and proximal course (arrow) of the (A) the LCA and (B) RCA.
FIGURE 2:
Scatterplot of the measurable course as read by reader 1 versus reader 2. There was statistical significance for both the RCA (P < 0.001) and the LCA (P = 0.028). Despite the statistical significance, its clinical effect is likely less as both readers visualized 1 cm or more in the majority of athletes.
TABLE 2:
Results by sonographer.
The association of athlete characteristics and the echocardiographic results were evaluated for ostia, proximal course, and measurable course. In each of these three categories, the effect on reader visualization as well as interobserver reliability were calculated for sport, ethnicity, gender, height, weight, BMI, BSA, and sonographer (see Table 3). Overall, there was little effect of athlete characteristics on the visualization and interobserver reliability for ostia or proximal course. Where there was significant effect, the small number of ostia or proximal courses was not visualized, and the small size of the subcategory was likely contributory. In most cases, significance was caused by a subgroup represented by three athletes or less, without which no significance would have been observed for the remaining subgroups. With measurable course, however, there was increased effect on visualization and interobserver reliability. This is particularly notable in regard to LCA measurable course distance, where there was a positive correlation with athlete size by the repeated-measures statistic (see Fig. 3), with correlation values being 0.234 (P < 0.01) for height, 0.031 (P < 0.01) for weight, 0.024 (P = 0.029) for BMI, and 0.048 (P < 0.01) for BSA.
TABLE 3:
Effect of sonographer, athlete demographics, and size on coronary visualization and interobserver reliability.
FIGURE 3:
Scatterplot of the average measurable course of a particular athlete versus his or her BSA. BSA was chosen as it illustrates both athlete height and weight. A positive association was found on LCA measurable course and athlete size by repeated measures, but no association with athlete size and RCA measurable course is seen.
DISCUSSION
The origin and proximal course of the coronary arteries were readily and reliably observed in the overwhelming majority of athletes. Moreover, sport, ethnicity, weight, and height varied greatly and are generalizable to the college-age athlete population. Evaluation of coronary ostia by TTE remains a standard part of pediatric, although not often adult, echocardiographic protocols and is generally effective in establishing proximal coronary artery anatomy (4–11,20). This study illustrates this ability extends into adult ages and sizes in the athletic population. Furthermore, this was accomplished on portable laptop-sized echocardiogram machines in the training room setting. To our knowledge, no published studies have evaluated interobserver reliability of identification of coronary origin or proximal course in the screening of athletes. In comparing studies that previously analyzed visualization of coronary ostia in athlete screening TTE, this study shows similarly high rates of ostia visualization but also compares TTE results between two readers and analyzes both coronary ostia as well as the proximal anatomy of each coronary artery (1,13,21).
There was near perfect interobserver reliability in observing the coronary ostia and good reliability in identifying the proximal course. The visualization of proximal LCA bifurcation showed significant interobserver variability, suggesting this parameter is not as reliable as the measurable length of a coronary artery when evaluating its proximal anatomy. When looking further along the coronary anatomy with measurable course, more substantial differences between reader measurements were observed. Overall, there was fairly little effect on athlete characteristics, including size, for the observation of coronary ostia and proximal course or the associated interobserver reliability. Where athlete characteristics proved statistically significant, the small number athletes whose coronary arteries could not be visualized and subsequent sparse representation in many of the subgroups analyzed were likely contributory. Interestingly, a positive association was found on LCA measurable course and athlete size, including height, weight, BMI, and BSA. This is somewhat counterintuitive to the assumption often seen in adult echocardiography that larger individuals will have worse transthoracic windows. Despite this association, even the smallest athletes showed reasonable coronary artery visualization, with average RCA and LCA measurable course for athletes less than 50 kg being 1.9 and 1.2 cm, respectively. Ultimately, these findings suggest that TTE is capable of visualizing the coronary artery proximal anatomy in the smallest to the largest college-age athletes in a wide variety of sports.
Some centers have advocated for limited echocardiograms in the screening of college athletes (1,18,21,23). These limited TTE protocols varied significantly between studies, and some did not establish the ostia of both coronary arteries. In addition, some limited echocardiogram protocols were not performed or read by a faculty cardiologist. In some limited protocols, images were neither saved nor reviewable. These limited protocols were likely created to increase screening TTE efficiency. Nevertheless, a previous study at this institution found that TTE protocols using the same complete protocols used in this article were completed in approximately 12 min each (12). Although screening TTE is currently not recommended for athletes in the United States, many centers perform limited studies to screen athletes for risk of SCD (1,3,18,21,23). As coronary anomalies are among the most common etiologies of SCD, and this study demonstrates proximal coronary anatomy is readily and reliably observed, the authors believe coronary artery evaluation should always be included when a screening TTE is performed.
The retrospective nature and lack of anomalies diagnosed limits this study’s ability to determine a screening TTE’s predictive capability to diagnose coronary anomalies or reduce SCD. All coronary origins were read as visualized and normal at the time of the athletic screening, so no advanced imaging was obtained. There is a notable discrepancy between 100% of coronary ostia visualized at the screening versus the very few athletes where both readers did not see a particular ostium. As the cardiologist was often bedside for the screening echocardiogram, some images seen live may have not be stored for the readers to view, or there simply could also be a disagreement with the bedside cardiologist and one of the readers. As the duration of each study was not measured, it is not known whether athlete characteristics or sonographer experience affected the time required to visualize each coronary artery. The color flow Doppler of the coronary arteries was not analyzed, which is known to be useful in confirming coronary origins in pediatric echocardiography. The success shown in this article of coronary identification in two-dimensional images suggests that TTE is capable of identifying coronary origins, and successful application of color flow Doppler in college athletes would be likely. In addition, proximal course and measurable course have not been shown to be valuable in confirming normal coronary anatomy or predicting risk of SCD. Instead, those measurements are used here as a marker of quality obtainable by portable TTE in athlete screening and allow for greater statistical analysis and comparison in this study by elevating the required threshold for a positive observation. Although this institution’s varsity athlete population is quite large and diverse compared with many intercollegiate athletic programs, increased number of athletes would enhance the statistical power of the smaller subgroup comparisons. Lastly, our institution’s screening program represents an ideal screening situation with studies being performed and analyzed real time by sonographers and cardiologists experienced in evaluating the hearts of young athletes.
The origin and the proximal course of the coronary arteries were readily and reliably observed in the overwhelming majority of National Collegiate Athletic Association Division-I athletes during screening TTE. As coronary anomalies are among the most common etiologies of SCD, and this study demonstrates proximal coronary anatomy is readily and reliably observed, this study suggests coronary artery evaluation be included in athlete screening TTE protocols.
The authors gratefully acknowledge Ethan Saliba, PhD, ATC, PTC, SCS; Bill Parente, DPT, ATC, CES; Rebecca Vozzo, MEd, ATC, SCSC; Shelley Blakey, MEd, ETC; and all the UVA Athletic Training Staff. They also thank Christine Saunders, MHA, RDCS (AE, PE, FE), RDMS; John Eagle, RDCS, RCS; and Jason Higginson, RDCS, for performing the echocardiograms; and Adam Morrison, MD, for assistance with data collection.
The authors report no conflict of interest; there are no grant, financial, or industry relationships to disclose for any authors. The authors affirm that the results of the study are presented clearly, honestly, and without fabrication, falsification, or inappropriate data manipulation. The results of the present study do not constitute endorsement by the American College of Sports Medicine.
REFERENCES
1. Anderson JB, Grenier M, Edwards NM, et al. Usefulness of combined history, physical examination, electrocardiogram, and limited echocardiogram in screening adolescent athletes for risk for sudden cardiac death.
Am J Cardiol. 2014;114(11):1763–7.
2. Battle RW, Mistry DJ, Malhotra R, MacKnight JM, Saliba EN, Mahapatra S. Cardiovascular screening and the elite athlete: advances, concepts, controversies, and a view of the future.
Clin Sports Med. 2011;30(3):503–24.
3. Bernhardt DT, Roberts WO.
Preparticipation Physical Evaluation. Elk Grove Village (IL): American Academy of Pediatrics; 2010. p. 1.
4. Brown LM, Duffy CE, Mitchell C, Young L. A practical guide to pediatric coronary artery imaging with echocardiography.
J Am Soc Echocardiogr. 2015;28(4):379–91.
5. Clouse M, Cailes C, Devine J, et al. What is the feasibility of imaging coronary arteries during routine echocardiograms in children?
J Am Soc Echocardiogr. 2002;15(10 Pt 2):1127–31.
6. Frommelt PC, Berger S, Pelech AN, Bergstrom S, Williamson JG. Prospective identification of anomalous origin of left coronary artery from the right sinus of valsalva using transthoracic echocardiography: importance of color Doppler flow mapping.
Pediatr Cardiol. 2001;22(4):327–32.
7. Frommelt PC, Frommelt MA, Tweddell JS, Jaquiss RD. Prospective echocardiographic diagnosis and surgical repair of anomalous origin of a coronary artery from the opposite sinus with an interarterial course.
J Am Coll Cardiol. 2003;42(1):148–54.
8. Herlong JR. Congenital anomalies of the coronary arteries. In: Lai WW, Mertens LL, Cohen MS, Geva T, editors.
Echocardiography in Pediatric and Congenital Heart Disease. Oxford (UK): Wiley-Blackwell; 2009. pp. 502–22.
9. Jureidini SB, Marino CJ, Singh GK, Balfour IC, Chen SC. Assessment of the coronary arteries in children: an integral part of each transthoracic echocardiographic study.
J Am Soc Echocardiogr. 2003;16(8):899–900 , author reply 900.
10. Jureidini SB, Marino CJ, Singh GK, Fiore A, Balfour IC. Main coronary artery and coronary ostial stenosis in children: detection by transthoracic color flow and pulsed Doppler echocardiography.
J Am Soc Echocardiogr. 2000;13(4):255–63.
11. Lorber R, Srivastava S, Wilder TJ, et al. Anomalous aortic origin of coronary arteries in the young: echocardiographic evaluation with surgical correlation.
JACC Cardiovasc Imaging. 2015;8(11):1239–49.
12. Malhotra R, Saunders C, Eagle J, et al. Screening elite athletes with congenital ECHO: feasibility and findings in our first year.
J Am Coll Cardiol. 2012;59(13): E1930.
13. Maron BJ, Bodison SA, Wesley YE, Tucker E, Green KJ. Results of screening a large group of intercollegiate competitive athletes for cardiovascular disease.
J Am Coll Cardiol. 1987;10(6):1214–21.
14. Maron BJ, Doerer JJ, Haas TS, Tierney DM, Mueller FO. Sudden deaths in young competitive athletes: analysis of 1866 deaths in the United States, 1980–2006.
Circulation. 2009;119(8):1085–92.
15. Maron BJ, Haas TS, Murphy CJ, Ahluwalia A, Rutten Ramos S. Incidence and causes of sudden death in U.S. college athletes.
J Am Coll Cardiol. 2014;63(16):1636–43.
16. Maron BJ. Sudden death in young athletes.
N Engl J Med. 2003;349(11):1064–75.
17. Moodie DS, Fyfe D, Gill CC, et al. Anomalous origin of the left coronary artery from the pulmonary artery (Bland–White–Garland syndrome) in adult patients: long-term follow-up after surgery.
Am Heart J. 1983;106(2):381–8.
18. Murry PM, Cantwell JD, Heath DL, Shoop J. The role of limited echocardiography in screening athletes.
Am J Cardiol. 1995;76(11):849–50.
19. Schwartz ML, Jonas RA, Colan SD. Anomalous origin of left coronary artery from pulmonary artery: recovery of left ventricular function after dual coronary repair.
J Am Coll Cardiol. 1997;30(2):547–53.
20. Turner II, Turek JW, Jaggers J, Herlong JR, Lawson DS, Lodge AJ. Anomalous aortic origin of a coronary artery: preoperative diagnosis and surgical planning.
World J Pediatr Congenit Heart Surg. 2011;2(3):340–5.
21. Wyman RA, Chiu RY, Rahko PS. The 5-minute screening echocardiogram for athletes.
J Am Soc Echocardiogr. 2008;21(7):786–8.
22. Yamanaka O, Hobbs RE. Coronary artery anomalies in 126,595 patients undergoing coronary arteriography.
Cathet Cardiovasc Diagn. 1990;21(1):28–40.
23. Yim ES, Kao D, Gillis EF, Basilico FC, Corrado GD. Focused physician-performed echocardiography in sports medicine: a potential screening tool for detecting aortic root dilatation in athletes.
J Ultrasound Med. 2013;32(12):2101–6.
24. Zhang LJ, Yang GF, Huang W, Zhou CS, Chen P, Lu GM. Incidence of anomalous origin of coronary artery in 1879 Chinese adults on dual-source CT angiography.
Neth Heart J. 2010;18(10):466–70.
