Author + information
- Received December 15, 2014
- Revision received April 15, 2015
- Accepted April 16, 2015
- Published online November 1, 2015.
- Richard Lorber, MD∗,
- Shubhika Srivastava, MBBS†,
- Travis J. Wilder, MD‡,
- Susan McIntyre, RN‡,
- William M. DeCampli, MD, PhD‡,§∗ ( )(, )
- William G. Williams, MD‡,
- Peter C. Frommelt, MD‖,
- Ira A. Parness, MD†,
- Eugene H. Blackstone, MD∗,‡,¶,
- Marshall L. Jacobs, MD‡,
- Luc Mertens, MD, PhD#,
- Julie A. Brothers, MD∗∗,
- J. René Herlong, MD††,
- AAOCA Working Group of the Congenital Heart Surgeons Society
- ∗Children’s Hospital of San Antonio, Baylor College of Medicine, Houston, Texas
- †Kravis Children’s Hospital at Mount Sinai, New York, New York
- ‡Congenital Heart Surgeons’ Society Data Center, Hospital for Sick Children, Toronto, Ontario, Canada
- §Arnold Palmer Hospital for Children, Orlando, Florida
- ‖Children’s Hospital of Wisconsin, Milwaukee, Wisconsin
- ¶Heart and Vascular Institute, The Cleveland Clinic, Cleveland, Ohio
- #Cardiology, Hospital for Sick Children, Toronto, Ontario, Canada
- ∗∗Children’s Hospital of Philadelphia, Philadelphia, Pennsylvania
- ††Sanger Heart and Vascular Institute-Charlotte Pediatric Cardiology, Charlotte, North Carolina
- ↵∗Reprint requests and correspondence:
Dr. William M. DeCampli, Arnold Palmer Hospital for Children, University of Central Florida College of Medicine, 92 West Miller Street, MP307, Orlando, Florida 32806.
Objectives This study sought to compare findings from institutional echocardiographic reports with imaging core laboratory (ICL) review of corresponding echocardiographic images and operative reports in 159 patients with anomalous aortic origin of a coronary artery (AAOCA). The study also sought to develop a “best practice” protocol for imaging and interpreting images in establishing the diagnosis of AAOCA.
Background AAOCA is associated with sudden death in the young. Underlying anatomic risk factors that can cause ischemia-related events include coronary arterial ostial stenosis, intramural course of the proximal coronary within the aortic wall, interarterial course, and potential compression between the great arteries. Consistent protocols for diagnosing and evaluating these features are lacking, potentially precluding the ability to risk stratify patients based on evidence and plan surgical strategy.
Methods For a prescribed set of anatomic AAOCA features, percentages of missing data in institutional echocardiographic reports were calculated. For each feature, agreement among institutional echocardiographic reports, ICL review of images, and surgical reports was evaluated using the weighted kappa statistic. An echocardiographic imaging protocol was developed heuristically to reduce differences between institutional reports and ICL review.
Results A total of 13%, 33%, and 62% of echocardiograms were missing images enabling diagnosis of intra-arterial course, proximal intramural course, and high ostial takeoff, respectively. There was poor agreement between institutional reports and ICL review for diagnosis of origin of coronary artery, interarterial course, intramural course, and acute angle takeoff (kappa = 0.74, 0.11, –0.03, 0.13, respectively). Surgical findings were also significantly different from those of reports, and to a lesser extent ICL reviews. The resulting protocol contains technical recommendations for imaging each of these features.
Conclusions Poor agreement between institutional reports and ICL review for AAOCA suggests need for an imaging protocol to permit evidence-based risk stratification and surgical planning. Even then, delineation of echocardiographic details in AAOCA will remain imperfect.
Anomalous aortic origin of a coronary artery (AAOCA) is associated with sudden death in the young, especially in athletes <30 years of age (1–6). The anatomic features that are proposed as potential risk factors for ischemia-related events include coronary arterial ostial stenosis, acute angle of takeoff from the aorta, intramural course of the proximal coronary within the aortic wall, and interarterial course resulting in potential compression between the great arteries (7,8). An abnormal coronary artery origin with an anterior, posterior, or intraconal (intraseptal) course is generally considered benign (8,9).
Transthoracic echocardiography (TTE) is the first-line imaging modality for initial diagnosis (8,10,11). Apart from its use in the diagnosis of the abnormal coronary origin, TTE can also be useful in identifying anatomic features that influence management decisions, including referral for surgery (e.g., intramural course, interarterial course). The Congenital Heart Surgeons’ Society (CHSS) initiated a multicenter registry, collecting data from 24 participating institutions beginning in 2009. An echocardiographic imaging core laboratory (ICL) reviewed the submitted images and reports. This study reports the analysis of the echocardiographic and surgical data. Our objectives were to: 1) study the variability in the currently used echocardiographic imaging protocols and reports among the participating CHSS institutions; 2) correlate echocardiographic and surgical findings; and 3) design a standardized imaging and reporting protocol for the echocardiographic evaluation of patients with AAOCA that could be implemented across institutions.
In 2009 the CHSS started a registry of patients diagnosed with AAOCA at age 30 years or younger and evaluated at any of the 24 participating CHSS member institutions. TTE images and reports were available for 159 of the first 243 patients enrolled. Institutional review boards or ethics committees at each participating center reviewed and approved the protocol. Informed consent was obtained before enrollment into the registry. As of October 2014, the registry included 348 patients from 35 institutions. The registry dataset includes copies of video (VHS) and digital echocardiograms and institutional echocardiographic and surgical reports. The echocardiographic studies were uploaded to a central ICL review station (Syngo Dynamics, Siemens, Germany) and 3 readers (R.L., J.H., and S.S.) reviewed the submitted TTEs. The readers assessed all submitted TTEs and recorded findings on an atomization sheet (Online Appendix 1).
Evaluation of submitted echocardiographic images
For the echocardiographic review, the ICL focused on identifying the anatomic features that impact management and surgical planning of patients with AAOCA. These include: 1) location of the ostium in relation to the sinutubular junction (STJ); 2) identification of an interarterial and/or intramural course; 3) identification of intraconal (intraseptal) course of the proximal coronary; and 4) morphologic characterization of the ostium of the anomalous coronary; separate ostium versus single ostium, shape, size, and/or stenosis of the anomalous origin.
The ICL reviews focused primarily on numbers 1, 2, and 3 because most morphologic features of the ostium specified in number 4 cannot be reliably demonstrated echocardiographically. The ICL reviewers evaluated whether the TTE images were adequate for demonstrating features of numbers 1, 2, and 3 and whether these morphologic features of AAOCA were included in the submitted reports. Still frames alone were considered inadequate for making a diagnosis. The data collection form is shown in Online Appendix 1. The submitted TTEs were evaluated for presence of the following imaging views recorded as sweeps or multiple cardiac cycle clips: 1) 2D parasternal short axis (PSSAX) and modified PSSAX demonstrating origin and course of the coronary arteries (Figure 1, Online Video 1); 2) color Doppler PSSAX images demonstrating the origin and course of the coronary arteries with emphasis on color flow mapping of the proximal course of the AAOCA in relation to the aorta and pulmonary trunk (Online Videos 2 and 3); 3) 2D and color Doppler parasternal long axis (PSLAX) imaging and/or modified apical views to demonstrate origin of the AAOCA in relation to the STJ (Online Videos 4A, 4B, 4C, 5, and 6); and 4) 2D and color Doppler PSLAX sweep demonstrating proximal course of the AAOCA (Figure 2, Online Videos 4C and 6).
Definitions used for review
Interarterial course was defined as a proximal course between the aorta and pulmonary trunk on the epicardial surface (Figures 1 and 2). Intramural course is histologically defined as shared tunica media between the coronary and aortic wall. Echocardiographic intramural course is characterized by an acute angulation (<30°) of the coronary origin as it originates from the aorta with diastolic color flow in its proximal course within the aortic wall (8). Most commonly, all intramural courses are interarterial but all interarterial courses are not necessarily intramural (Figures 1, 2, and 3).
Intraconal course is morphologically defined as a course of the proximal left main coronary artery or left anterior descending inferior to the aortic and pulmonary valve annulus through the conal septum to the epicardium. This can be echocardiographically demonstrated by a course inferior to the pulmonary valve annulus with “ground glass” appearance of muscle surrounding this segment (Figures 1 and 4, Online Videos 5 and 7). High ostial takeoff is defined as origin of the coronary artery above the STJ of the aorta (Figure 2).
The ICL reviewers used the available images to identify the ostial location in relation to the STJ and proximal intramural/interarterial course. If images and views were missing, thereby not allowing identification of the features, the study was deemed incomplete.
Evaluation of the completeness of the reports of the echocardiographic data
All institutional reports were examined for the reporting of the following elements: diagnosis of AAOCA, interarterial course, intramural course, intraconal (intramyocardial or intraseptal or subpulmonary) course, and origin of the coronary with respect to the STJ (superior or inferior).
Analytic strategy of echocardiographic correlates of surgical anatomy
Echocardiograms and institution reports were available for 159 of 243 patients and imaging reports were available for 231 of 243 patients (Figure 5). Demographics of all patients are presented in Table 1.
For the purpose of this analysis the description of surgical observations was considered to be the “true” anatomic diagnosis. Among the 82 patients who had a surgical intervention and ICL TTE review, 69 underwent an unroofing procedure. Details of the intraoperative findings were obtained from the surgical reports.
The agreement between anatomy identified during surgery was compared with ICL echocardiogram reviews (REVIEWS) and institutional echocardiogram reports (REPORTS). The agreement between REVIEWS and REPORTS was then compared in 159 patients who had both REVIEWS and REPORTS. Comparison between ICL REVIEWS and surgical reports was used to establish echocardiographic features that might predict the surgical diagnosis of an intramural coronary course.
Data are presented as median with range, mean ± SD, or frequency and percentage where appropriate, with number of missing variables indicated. Continuous variables were compared using Student t test and categorical variables were compared with a chi-square or Fisher exact test when appropriate. All data analyses were performed with SAS version 9.2 (SAS Institute, Inc., Cary, North Carolina).
The strength of agreement between REVIEWS and REPORTS was calculated using a Cohen or weighted kappa (κ) statistic (coefficient) and is measured as a value from 0 to 1. A κ of 1 indicates perfect agreement, whereas a κ of 0 indicates poor agreement. The associated p value tests whether the estimated κ is not caused by chance alone (12).
ICL evaluation of the images
Of 159 TTE reviewed (of these 159 transesophageal echocardiography [TEE] studies, 12 were considered nondiagnostic based on image quality and/or absent diagnostic imaging views; these 12 are included in the percent calculations), 11% (n = 17) did not contain images suitable for evaluation of an acute angle of origin, 33% (n = 53) did not contain color Doppler imaging of the proximal intramural course, and 13% (n = 21) did not contain images suitable for evaluation of proximal interarterial course. Images necessary to diagnose a possible high ostial takeoff (superior to the STJ) were missing in 62% of the studies. The diagnosis of intramural course and high ostial takeoff of the AAOCA could often not be made by ICL review because of the absence of TTE images in PSSAX and PSLAX 2D and color Doppler.
Agreement with surgical findings
Comparisons between institutional reports and surgery and between ICL reviews and surgery indicated an overall stronger agreement between the ICL reviews and surgery than between institution reports and surgery (Table 2). However, both institutional reports and ICL reviews show less agreement with surgical operation reports for measures of more detailed anatomy (Table 2; interarterial course, intramural course, and acute angle of the proximal takeoff).
Echocardiographic criteria for intramural course
The ICL defined an intramural coronary course as an acute angle of origin and diastolic color flow in the proximal course of the AAOCA within the aortic wall. Seventy-three patients who had surgery also had images that could be used to identify all of these criteria. These criteria correctly defined intramural course in 94% patients with an intramural course (64 of 68) and absence of intramural course in 67% (2 of 3) (detailed in Figure 6).
This is the first study to demonstrate a detailed assessment of how different institutions diagnose AAOCA by TTE, highlighting the need and importance of a standardized comprehensive imaging and reporting protocol across pediatric centers. In a prior report, TTE was shown to be diagnostic for intramural coronary course in 92.5% of all patients who had surgery in a single-institution surgical series (13). This 24-institution multicenter study failed to demonstrate a good agreement between institutional reports and “expert” ICL review for characterization of the proximal course of AAOCA. Our study also describes echocardiographic features that can facilitate the diagnosis of an intramural coronary artery. An accurate description of the abnormal origin including high ostial takeoff and identification of an intramural and/or interarterial course are potentially important features of AAOCA that should be identified for each patient. The surgical management of AAOCA with an intramural course usually involves an unroofing procedure (14). Reporting coronary origin with respect to the STJ is important because a high ostial origin superior to the STJ allows for unroofing without the need for commissural resuspension. Adoption of standardized reporting and imaging protocols (Online Appendixes 1 and 2) will allow for the continued study of natural and unnatural history of AAOCA across all institutions (15).
AAOCA is a rare condition that is clinically important because of its association with sudden cardiac death, with a reported prevalence of 0.1% to 0.3%. Anomalous aortic origin of the right coronary artery is 6 to 8 times more common than that of left coronary artery. TTE is the most common initial screening tool for diagnosis of AAOCA in children and often is an incidental finding, because most patients are asymptomatic (14). Because the lesion is uncommon, echocardiography laboratories have developed institution-specific methods of assessing and managing this condition. The therapeutic approach to these lesions is still controversial because of lack of long-term outcome data and wide variability in treatments offered in different centers. The CHSS AAOCA registry was designed to evaluate the natural and unnatural history of AAOCA with data obtained from CHSS member institutions with variable approaches to the diagnosis and management of the abnormality. The CHSS collects all clinical, surgical, and diagnostic information of patients identified in the participating centers and conducts an annual cross-sectional follow-up of enrolled patients. Diagnostic images and their reports are an important component of this data collection. All diagnostic imaging studies including their reports are collected and stored in the Data Center of the CHSS. The AAOCA project is ongoing and, at the time of this writing, includes 380 patients from 35 institutions.
The current study was motivated in part by the need to standardize the diagnostic assessment of AAOCA in this growing cohort. Of importance, we found that even though the diagnosis of AAOCA could be made on the images uniformly, some important anatomic features, such as intramural proximal AAOCA course and high ostial takeoff, were either not demonstrated on the submitted institutional TTEs and/or were not reported on the institutional echocardiography reports. Using the ICL review of the images, the salient morphologic features that may influence the risk for coronary events and potentially improve surgical decision-making were identified based on existing literature and consensus reached among the participating surgical centers. A salient feature that should be part of every evaluation of AAOCA, but was absent in nearly one-third of cases, is the demonstration of color flow mapping images of the proximal course of the AAOCA in relation to the aortic wall from PSSAX TTE views. The identification of an intramural course by color mapping is important, because this not only impacts surgical technique but is also a potential high-risk anatomic feature associated with sudden death. Acute angle takeoff is another anatomic detail that was not mentioned in 120 of 231 (52%) of institutional reports. Based on surgical correlation with echocardiographic features, we now know that an acute angle takeoff in addition to the color flow within the proximal AAOCA lumen can be used to make the diagnosis of a proximal intramural course of the AAOCA.
To overcome these shortcomings, our ICL group proposed a standardized reporting format (Online Appendix 1) and a standardized echocardiography protocol approach to imaging (Online Appendix 2) of AAOCA. An imaging protocol for TEE is also included, even though this study did not focus on evaluation of submitted TEE studies. Representative TEE images are shown in Figure 7. The standardized reporting format and protocol for imaging, along with video clips, can be found within this manuscript and also on the CHSS website (16). In addition, the protocols with video-clip examples referenced as Online Videos 1, 2, 3, 4A, 4B, 4C, 5, 6, 7, 8A, 8B, 8C, 9A, 9B, 10, 11, and 12 will be sent to the participating centers of the CHSS’s AAOCA project for prospective use. Incorporation of the views noted in the imaging protocol inclusive of 2D and color Doppler imaging into institutional imaging protocols will allow for diagnosis and characterization of AAOCA. We believe this approach could be useful for other centers and cardiologists diagnosing this abnormality.
In patients with symptoms suggestive of ischemia and where TTE images are suboptimal to confirm the diagnosis of AAOCA other imaging modalities can be used to confirm the diagnosis. These imaging modalities include TEE, computed tomography angiography (CTA), and cardiac magnetic resonance. TEE, CTA, and cardiac magnetic resonance were also submitted by institutions in some patients, but these were not evaluated for the purpose of this study. Use of CTA to confirm diagnosis (17,18) and define anatomic features that would be considered high risk has been published in small population subsets. For example, Opolski et al. (18) published a series of CTA images in 8,522 patients, out of which only 24 had AAOCA. Turner et al. (13), in their institutional series of 53 patients, demonstrated a higher accuracy of TTE over CTA and magnetic resonance imaging in diagnosis of an intramural AAOCA. Assessment of accuracy and utility of other imaging modalities in diagnosis, in risk stratification, and in planning management strategies is a subsequent multicenter study.
The lack of complete initial TTE studies that were submitted by the institutions to the CHSS, heterogeneity in quality of the submitted TTE studies, and institutional echo and surgical reports available for echo ICL review limited the accuracy and agreement statistics. The diagnosis and description of the intramural proximal course of the AAOCA and its length were also not clear from all surgical reports. This demonstrates lack of a unified surgical approach to diagnosis and reporting of AAOCA. The proposed protocols in this manuscript allow for standardization of imaging and reporting, thereby allowing for more accurate and uniform data collection for the multicenter CHSS registry study, and also to improve the characterization of AAOCA for management of patients with these anomalies.
Ideally, management decisions regarding patients with AAOCA should be guided by evidenced-based risk stratification, which in turn depends on accurate ascertainment of morphologic and functional details of the spectrum of anomalies, and correlation between these features and the outcomes of various treatment strategies. Adoption of uniform imaging and reporting protocols in the diagnosis and assessment of AAOCA can help in risk stratification and appropriate management strategies. The development of a “best practice” protocol for imaging and reporting on AAOCA is the initial step required to further study the natural and unnatural history of children and young adults diagnosed with AAOCA.
COMPETENCY IN MEDICAL KNOWLEDGE: AAOCA can result in SCD, and echocardiography is usually the first imaging or screening modality used to suspect or diagnose AAOCA. Only one-third of the patients may have ischemic symptoms at presentation. This diagnosis is an indication for restriction from sports. Surgical therapy is often necessary. Morphological features of AAOCA by echocardiography can be used for risk stratification.
COMPETENCY IN PATIENT CARE AND PROCEDURAL SKILLS: AAOCA can be diagnosed or suspected by routine echocardiography. Echocardiographic diagnosis is very difficult and imaging protocols to evaluate coronary artery origins have not been standardized. Adoption of these protocols by sonographers and imaging laboratories will allow for improvement in diagnosis and patient outcomes.
COMPETENCY IN PRACTICE-BASED LEARNING: There is no existing standard imaging and reporting protocol that allows all institutions to make the detailed diagnosis and report on individual anatomic features of AAOCA.
TRANSLATIONAL OUTLOOK: Additional longitudinal studies following adoption of the standardized imaging and reporting protocols are needed for studying natural and unnatural history of AAOCA. The CHSS multicenter registry will allow for data collection inclusive of imaging, and clinical and surgical data at presentation and at follow-up of patients diagnosed with AAOCA. Utility of other imaging modalities, such as CT and magnetic resonance imaging in order to allow for risk stratification of AAOCA has not been evaluated. Future studies evaluating application of multimodality imaging for risk stratification and development of management guidelines are needed.
The authors thank the Michael H. Ludwig Memorial Foundation; Maansi Dayal Srivastava for the Coronary Drawings in Figure 3; Children’s Heart Foundation, Chicago; The Cardiac Center at the Children’s Hospital of Philadelphia; Cardiovascular and Critical Care Research Center; Children’s Hospitals and Clinics of Minnesota; Cardiac Surgery Research and Teaching Fund; Hospital for Sick Children, Toronto; and medical illustrations (Figures 1, 2, and 4) by Sharon I. Teal, Indianapolis, Indiana.
For supplemental materials, and videos and their legends, please see the online version of this article.
This study was partially funded by the Michael H. Ludwig Memorial Foundation; Children’s Heart Foundation, Chicago; Cardiac Center at The Children’s Hospital of Philadelphia; Cardiovascular and Critical Care Research Center, Children’s Hospitals and Clinics of Minnesota; and the Cardiac Surgery Research and Teaching Fund, Hospital for Sick Children, Toronto. All authors have reported that they have no relationships relevant to the contents of this paper to disclose. Drs. Lorber and Srivastava have contributed equally to this work.
- Abbreviations and Acronyms
- anomalous aortic origin of coronary artery
- computed tomography angiography
- imaging core laboratory
- parasternal long axis
- parasternal short axis
- sinutubular junction
- transesophageal echocardiography
- transthoracic echocardiography
- Received December 15, 2014.
- Revision received April 15, 2015.
- Accepted April 16, 2015.
- American College of Cardiology Foundation
- Taylor A.J.,
- Rogan K.M.,
- Virmani R.,
- et al.
- Basso C.,
- Maron B.J.,
- Corrado D.,
- et al.
- Virmani R.,
- Chun P.K.C.,
- Goldstein R.E.,
- et al.
- Turner I.I.,
- Turek J.W.,
- Jaggers J.,
- et al.
- Poynter J.A.,
- Williams W.G.,
- McIntyre S.,
- et al.
- ↵Congenital Heart Surgeons' Society. Anomalous Aortic Origin of a Coronary Artery. Available at: http://www.chssdc.org/publications#AAOCA. Accessed September 16, 2015.
- Von Ziegler F.,
- Pilla M.,
- McMullan L.,
- et al.