Anomalous aortic origin of a coronary artery with an interarterial course (AAOCA) is a rare anomaly that carries an increased risk for myocardial ischemia and sudden cardiac death (1- 4) during or just after exercise, notably among otherwise healthy children and young adults (2,5- 7). Anomalous left main coronary artery from the right sinus of Valsalva (ALCA) and right coronary from the left sinus (ARCA) are both associated with sudden death, but the former carries a higher risk (1- 4). The true prevalence of AAOCA is unknown, but estimates range from 0.1% to 0.3% (1,7- 9). Patients are often asymptomatic (7- 8), but when symptoms are present, they usually include: chest pain, pre-syncope, or syncope with exertion (2,10- 13). Because the risk of sudden death is unknown, patient management remains controversial: some children are followed clinically with exercise restrictions, whereas others are referred for surgery with exercise limitations until surgical repair is completed (14).

Data evaluating children for myocardial ischemia after AAOCA repair are limited (10- 12). No studies to date have evaluated children for post-operative ischemia with multiple imaging modalities, including exercise stress test (EST), stress echocardiogram (SE), and myocardial perfusion scan (MPS). The purpose of our study was to prospectively evaluate evidence of rest and stress myocardial ischemia in children with AAOCA after surgical repair.

The study was approved by the Institutional Review Board at The Children’s Hospital of Philadelphia. Patients with AAOCA who underwent surgical repair at our institution between October 2001 and December 2006 were eligible for study inclusion. We excluded patients with other coronary artery anomalies or significant structural heart disease.

After obtaining parental consent and child assent, we reviewed each patient’s medical record for presenting signs and symptoms, pre- and post-operative testing, and surgical repair data. We defined cardiovascular symptoms as chest pain, pre-syncope, or syncope with or just after exertion and cardiovascular signs as ventricular arrhythmia, exercise-induced ischemia, myocardial infarction, or aborted sudden death. Enrolled patients underwent prospective testing as described in the following text.

Subjects exercised to maximal ability with a ramp bicycle protocol (SensorMedics, Yorba Linda, California). Cardiac rhythm was monitored continuously. Four subjects who did not meet the height requirements for cycle ergometry used an incremental treadmill (Marquette Series 2000, Milwaukee, Wisconsin) protocol. Four subjects tested at an outside hospital exercised with a similar treadmill protocol.

Two-dimensional images were obtained with a Sonos 5500 or iE 33 ultrasound system (Phillips, Andover, Massachusetts). To assess regional wall motion, images were obtained in standard parasternal long- and short-axis views and apical 4- and 2-chamber views. The left ventricle was divided into 17 segments to analyze regional myocardial function (15).

Intravenous ^99mTc-sestamibi was administered at rest and peak exercise, and images were obtained at rest and within 30 min of peak exercise. Locations of perfusion defects were noted with the same 17-segment model used for SE (15).

Baseline clinical characteristics and positive post-operative tests for ischemia were described. Descriptive statistics were expressed as mean (± SD) and median (range) for normally and non-normally distributed data, respectively. Two-tailed Fisher exact test was used to compare gender and cardiovascular symptoms with coronary anatomy (i.e., ARCA or ALCA).

Twenty-four of 26 consecutive patients who met inclusion criteria consented to participate. Fifteen were male, and 16 had ARCA. Seven of 8 ALCA patients compared with 8 of 16 ARCA patients were male (p = 0.18). (Table 1) describes presenting signs and symptoms and associated coronary anatomy of the cohort. Of the 12 patients with cardiovascular complaints, 5 had ALCA compared with 7 with ARCA (p = 0.67). Six of 7 ARCA patients who presented with cardiovascular symptoms had chest pain with exercise.

Sixteen patients had pre-operative EST; of these, 9 had presenting symptoms suggestive of ischemia. One ARCA patient had a blunted blood pressure response on EST with no electrocardiogram abnormalities. One ALCA patient with palpitations had frequent premature ventricular contractions on Holter monitor; a second patient had aborted sudden cardiac death after noncompetitive exercise. All other pre-operative ischemia evaluations were normal.

The median age at surgery was 12 (range 5 to 18) years. Surgical repair was performed in 23 patients with the unroofing procedure (16). Coronary reimplantation (17) was used in 1 ARCA patient, owing to the short intramural course of the anomalous vessel. In 12 patients, the aortic commissure was detached before unroofing. All survived the operation. One ALCA patient required emergent surgery on the first post-operative night for an aortic incision disruption that was successfully repaired. Another ALCA patient needed surgical drainage of a loculated pericardial effusion 2 weeks post-operatively. No patients developed significant aortic insufficiency during the follow-up period.

Prospective evaluations for myocardial ischemia were performed a median of 15 (range 2 to 48) months after surgery. All patients had unobstructed neo-coronary ostia by routine echocardiogram. One ALCA and 8 ARCA patients had evaluations suggestive of myocardial ischemia (Table 2). As an example of a positive ischemia evaluation, one ARCA patient underwent EST 6 weeks post-operatively that demonstrated inferior lead ST-segment depression (Figure 1A) with a normal MPS. At 20 months post-operatively, baseline Q waves were noted without ischemic changes during exercise (Figure 1B) with a normal SE and MPS.

Anomalous aortic origin of a coronary artery with an interarterial course is a rare congenital anomaly associated with increased risk of myocardial ischemia and sudden death in children and young adults (1- 6). We present the most comprehensive prospective ischemia evaluation after AAOCA repair on the largest number of pediatric patients to date. We found evidence suggestive of ischemia in 1 ALCA and 8 ARCA patients; these might not have been identified had we used 1 or even 2 testing modalities. All patients remained asymptomatic during testing, calling into question the use of patient complaints in risk assessment for ischemia. Although most patients had symptoms at diagnosis, we classified one-half as cardiovascular in origin. Those with ALCA presented more frequently with cardiovascular complaints than those with ARCA.

What remains controversial is the management of asymptomatic children who seem to have a higher risk of sudden death than those identified later in adulthood (1,6). Serial evaluation with EST might be useful. However, basing a patient’s management on a single test is worrisome, because EST can be falsely negative, especially because ischemia is intermittent with AAOCA (2). Indeed, in our study, 9 of 16 patients who had a pre-operative EST presented with cardiovascular symptoms and only 1 had an abnormal EST.

A limitation of our study was the inability to determine risk of surgery versus observation on myocardial ischemia and sudden death, because this was not a randomized study. Furthermore, because this was a prospective study initiated after surgical repair, we lacked pre-operative ischemia evaluations on 8 patients; this would have made for a more complete assessment of pre- and post-operative ischemia. Last, because the unroofing procedure has only been used fairly recently, we were only able to obtain mid- but not long-term follow-up evaluations.

Subclinical changes suggestive of ischemia might occur after AAOCA surgery despite patent neo-coronary ostia, notably after ARCA repair. The implication of these results on indication for surgery and subsequent sudden death risk is unclear. This study highlights that, although surgery is often performed on children in whom the pre-operative risk of myocardial ischemia is unknown, we currently have suboptimal ways to follow surgical results. Until more accurate testing modalities are found, serial EST, SE, and MPS are essential in evaluating ongoing ischemia risk after AAOCA repair.

Indeed, larger numbers of children need to be followed over time to determine whether the risk of myocardial ischemia after surgery is truly diminished. Angelini et al. (18) and Pelliccia (19) have suggested that a multicenter, multinational database of patients with AAOCA be established. We concur with this and are developing such a database of children and young adults in order to obtain the natural history of those who do not undergo surgery and the “unnatural” history of those who do to assess the utility of surgical intervention in decreasing long-term risk of myocardial ischemia and sudden cardiac death.