Coronary artery anomalies are a rare inborn disease featuring an anomalous origin, course and/or termination of the native coronary vessel [1]. Important and most common coronary artery anomalies are an anomalous origin from the opposite sinus of Valsalva (ACAOS, fig. 1) [2–4]. The course of an ACAOS may be ventral, retro­aortic or interarterial (the anomalous coronary artery running between the aorta and the pulmonary artery, fig. 2) [2]. The latter is a so-called “malignant” variant and is associated with sudden cardiac death (SCD) in young athletes [5], whereas the remaining variants are considered “benign”. Other variants of coronary artery anomalies are coronary fistula, with an abnormal termination of the coronary artery. Bland-White-Garland Syndrome is an anomaly with the origin of a coronary artery from the pulmonary artery, which may result in a “steal-phenomenon” caused by reversed flow from the coronary artery into a pulmonary artery due to the decreased pulmonary artery pressure after birth. Another form, myocardial bridging, is considered a normal variant as it appears in up to 43% of cases in coronary computed tomography angiography (CCTA) studies [6]. Beside the “malignant” variant of ACAOS with an interarterial course of the anomalous coronary artery, other anatomical high-risk features are a slit-like ostium, acute take-off angle, intramural course (the proximal part of the anomalous vessel passing within the tunica media of the aortic wall) with elliptical vessel shape and proximal narrowing of the anomalous vessel (fig. 3) [1, 7]. These anatomical features are considered high risk as it is hypothesised that during exercise aortic expansion may lead to valve-like obstruction of the slit-like ostium, and coronary flow might further be impaired by angulation of the acute take-off of the anomalous vessel. Moreover, the associated aortic dilation during exercise may lead to lateral compression of the proximal and narrowed intramural, elliptical vessel segments. Strenuous sports activity, which results in an increased heart rate and shortened diastolic filling time further contributes to the decreased perfusion of the anomalous coronary artery [7].

In a large registry of the Cleveland Clinic foundation where invasive coronary angiography data from 126 595 patients were analysed, the prevalence of anomalous coronary artery was 1.3% [8]. In a recent study from our centre, we analysed 5634 consecutive CCTAs between March 2007 and July 2015, and found 145 patients with coronary anomalies, which represents a prevalence of 2.6% for all coronary anomalies and a prevalence of 1.2% for ACAOS (table 1) [2]. Due to the increased temporal and spatial resolution of newer generation scanners, the three-dimensional visualisation of anomalies with CCTA is superior to that with other modalities such as echocardiography or invasive coronary angiography (table 2) and it seems that the prevalence also depends on the imaging modality [6, 7]. With the increasing use of noninvasive imaging modalities, a rise in absolute numbers of patients with detected anomalies is expected. Therefore, knowledge about downstream testing options and treatment counselling is gaining importance for family practitioners, cardiologists, heart surgeons and sports medicine physicians alike.

Autopsy series showed that, after hypertrophic cardiomyopathy, ACAOS is the second most common cause of sports-related SCD in young athletes during or shortly after strenuous exercise [5, 9]. In 80% of the cases where athletes died of an ACAOS, the course of the anomalous vessel was interarterial [10]. In another ­series of 6.3 million young US military recruits, who underwent an intensive 8-week boot camp, 64 sudden deaths were considered to be of cardiac origin. One third of the SCD cases showed an ACAOS, in all of which the left coronary artery arose from the right sinus of Valsalva (left-ACAOS), with an interarterial course between the pulmonary artery and aorta [11]. Others also have reported adverse cardiac events associated with right coronary arteries originating from the left coronary sinus (right-ACAOS) [12]. In a Swiss study, where we analysed autopsies of sports-related SCDs in young athletes (<40 years of age), 7% showed an anomalous coronary artery, with a similar prevalence compared with arrhythmogenic right ventricular cardiomyopathy or aortic dissection [13]. The underlying mechanisms by which anomalous coronary arteries lead to adverse cardiac events are not clearly understood. As mentioned above, it is suggested that high-risk anatomical features may all contribute to malperfusion of the myocardium and lead to ischaemia and scarring, especially under strenuous physical exercise conditions (see also fig. 3). As SCD with ACAOS was mainly reported in young athletes, it is unknown whether this holds true for older individuals with newly diagnosed ACAOS, especially in the setting of coronary artery ­disease (CAD) exclusion. We recently published data of 66 middle-aged patients (mean age 56 ± 11 years) with newly detected ACAOS on CCTA and compared their outcome with a cohort matched for age, gender, history of coronary revascularisation and segment stenosis score (a score quantifying CAD burden). Each patient with ACAOS was matched to two controls without ACAOS and was followed-up for 4 years, and major adverse cardiac events (myocardial infarction, revascularisation and cardiac death) were recorded for all patients and controls. Interestingly, the annual event rate of ACAOS patients compared with controls was not significantly different at 4.9 versus 4.8% and a hazard ratio of 0.94 (95% confidence interval [CI] 0.39–2.28, p = 0.89). In the subgroup with ACAOS with an interarterial course (40 patients, 65%), the annual event rate was also not significantly different from their matched controls, at 5.2 and 4.3%, respectively and a hazard ratio of 1.01 (95% CI 0.39–2.58, p = 0.99) (fig. 4) [14]. Thus, in this middle-aged population with newly diagnosed ACAOS and possible concomitant CAD, mid-term outcome was favourable and not statistically different from the matched control cohort without an anomalous coronary artery, regardless of whether or not ACAOS with an interarterial course were present. Whether older patients with ACAOS are less susceptible to adverse cardiac events, or whether a selection bias towards low-risk patients who survived childhood may have influenced our results remains unclear. Whichever is the case, it may be hypothesised that with ageing the risk for cardiovascular morbidity and mortality due to ACAOS moves into the background compared with the increasing risk of CAD-related events [15] and that surgery of ACAOS might not be mandatory for all these incidental findings. Indeed, when a subgroup of this middle-aged population with ACAOS who underwent hybrid imaging with CCTA and single photon emission computed tomography (SPECT) was analysed, myocardial ischaemia due to ACAOS was exceedingly rare and was more likely attributable to concomitant CAD [15].

CCTA has undergone substantial technical advances over the last decade, particularly with regard to spatial resolution and reduction of patients’ exposure to radiation to an average range of 0.21 to 0.5 mSv in daily clinical routine [16]. Therefore in most centres, CCTA has become the first-line imaging modality to assess the anatomy, i.e., the full course, of coronary artery anomalies [2]. Cardiac magnetic resonance imaging (CMR) also offers three-dimensional imaging at high spatial resolution (slightly lower than coronary CCTA), but without radiation exposure, and allows visualisation of the origin and the full course of the ACAOS, ­including its relationship to the great vessels. Furthermore, CMR offers other additional relevant information including valvular function, ventricular function, regional contractility and myocardial viability, all of which could be important considerations during the pre- or postoperative evaluation [17]. Echocardiography is also a valid alternative for assessing primarily the origin of ACAOS.

If an anomalous coronary artery with a high-risk anatomical feature is detected, downstream imaging for assessment of haemodynamic relevance is indicated. As pharmacological (adenosine) testing would not adequately represent strenuous exercise, maximum physical stress imaging, using SPECT (or also physical stress echocardiography, especially in children) is preferred in patients with ACAOS [15, 18, 19]. Alternatively, stress testing with dobutamine – which mimics physical exercise better than adenosine [20] – with use of other imaging modalities can also be considered where, for technical and procedural reasons, adequate maximum intensity exercise stress testing is not possible (e.g., stress-CMR, positron emission tomography [PET]). However, there is no evidence comparing the different imaging modalities in this particular clinical setting. Beside the advantage of the feasibility of physical stress imaging, SPECT also allows fusion of the functional imaging with the anatomical information from CCTA (fig. 5) [15, 21]. In more than half of patients with no coronary anomalies, the so-called standard distribution of myocardial perfusion territories does not correspond with individual anatomy [22], and it is even more challenging in patients with ACAOS to correctly assign territories to the subtending coronary arteries. Therefore, hybrid CCTA/SPECT and hybrid CCTA/PET represents a valuable noninvasive tool for discriminating the impact of ACAOS from that of concomitant CAD on myocardial ischaemia and correctly allocating ischaemia to the subtended anomalous or nonanomalous vessel.

Incorporating current knowledge of the literature of ACAOS, we propose the following imaging evaluation steps, treatment options and sport restriction recommendations (fig. 6) [7]. In “benign” ACAOS with absent high-risk anatomical features, no downstream testing, treatment and follow-up is generally needed. In “malignant” ACAOS with high-risk anatomical features, further imaging evaluation is needed in order to rule out ischaemia. Whether no finding of ischaemia fully reassures the physician and prognosticates absence of ACAOS-related future adverse cardiac events, remains unclear to date.

Although there are no long-term follow-up data showing a benefit of surgery over conservative treatment, it is generally recommended that, especially young patients with left-ACAOS and an interarterial course or documented ischaemia, should undergo surgery [23]. These patients also should be restricted from any competitive sports before surgery, based on the recent American Heart Association / American College of ­Cardiology Task Force 4 recommendations [24]. As an exception, participation in low static or low dynamic (class IA sports, such as bowling, cricket, curling, golf, rifle shooting and yoga) is allowed. This applies to patients with symptomatic or asymptomatic ACAOS diagnosed in either intentional or incidental conditions. This recommendation applies also to patients with right-ACAOS and either symptoms or a positive exercise stress test. In athletes with uncorrected right-ACAOS who exhibit symptoms, arrhythmias, or signs of ischaemia on exercise stress testing, participation in all competitive sports, except for class IA sports, is also not recommended before a surgical repair. For patients with right-ACAOS but no symptoms or ischaemia on an adequately performed exercise stress test, participation in competitive sports can be considered after adequate counselling of the athlete or the athlete’s parents [24].

The operative correction technique most used is so-called unroofing, where the intramural segment in the aorta is opened and a neo-ostium is created. Alternatively, re-implantation of the aberrant coronary artery or bypass surgery can be performed. However, this last technique is usually less effective as the bypass graft is prone to closure because of competing flow in the native vessel [7].

In other haemodynamically relevant coronary artery anomalies, namely Bland-White-Garland Syndrome, an operation is almost always indicated [23]. The primary aim is to re-implant the aberrant coronary artery in the aortic root or to tunnel aortic blood flow through the pulmonary artery to the ostium of the aberrant coronary artery (Takeuchi operation). Large coronary artery fistulas should be corrected with an operation or interventionally [23]. In smaller fistulas, pre-interventional ischaemia, left ventricular dysfunction or arrhythmia should be documented [23].

The evidence on which therapeutic recommendations in patients with anomalous coronary arteries is scarce and mainly based on anecdotal reports, case series and experts’ opinions. There are no prospective, randomised multicentre trials. Mostly, the choice of surgery, sports restriction or no treatment is made on a case-by-case discussion. The combination of clinical symptoms, age, sports behaviour and presence or absence of high-risk features / ischaemia, and individualised discussion between the treating physician, patient and heart surgeon leads to the final decision. In order to adapt current recommendations, more evidence, based on multicentre registries and prospective trials with follow-up studies are needed.