Introduction

The rapid growth in the use of computer tomography angiography (CTA) for the assessment of chest pain, given its non-invasive nature and very high sensitivity (95%) and negative predictive value (nearing 100%) in assessing obstructive coronary stenosis, has resulted in improvement in the management of patients with chest pain. This increased use of CTA has also significantly increased the number of incidental findings detected, including coronary artery anomalies (CAA) [1, 2••, 3, 4].

Many CAA classification schemes have been proposed; however, no single classification scheme has been widely accepted. Despite this, the majority are broadly classified as anomalies of coronary artery origin, course, termination and intrinsic anatomy [5,6,7]. The prevalence of CAA on CTA studies varies widely with prevalence documented between 0.8 and 10.4% of patients presenting for CTA [8,9,10]. Larger observational studies suggest a prevalence between 1 and 3% [11,12,13]. However, that the vast majority of prevalence data is derived from patients undergoing imaging based on clinical symptoms and therefore the true prevalence of these anomalies in the general population remains unknown.

CTA has seen significant recent advances, particularly with improved spatial resolution, temporal resolution, cranio-caudal coverage and reduction in radiation exposure, with newer scanners routinely using very low radiation doses (< 2 mSv) [14, 15]. Trials consistently demonstrate the increased diagnostic yield associated with CTA and its place as a gatekeeper for coronary angiography. The recent randomised controlled trial by Dewey et al. showed cardiac CT reduced the need for invasive coronary angiography from 100 to 14% and was associated with a significantly greater angiographic diagnostic yield from 15 to 75% patients requiring intervention [16], with no significant increase in long-term clinical events in the two groups at median follow-up of 3.3 years. Other studies have also consistently shown that CTA is at least equivalent to other imaging modalities (MRI/coronary angiogram) for detection of CAA [17] and is generally more widely available and cheaper than other anatomical tests. As a result, CTA has a class I indication for the initial assessment of patients either found to have or are suspected of having coronary artery anomalies as it allows a detailed visualisation of anatomical course and subsequent risk stratification [18].

Given the significant uptake in the use of CTA in clinical practice in low and intermediate risk patients presenting with chest pain [19], it is expected that the incidental detection of CAA will increase. It is inevitable that referring cardiologists and physicians will be confronted more commonly with this entity. Describing every coronary artery anomaly is beyond the scope of this review; however, we will outline commonly encountered anomalies, particularly those of clinical significance with CTA pictorial examples to aid CT readers in interpretation. Prevalence and management strategies will be briefly discussed where known.

Normal Coronary Anatomy and Variation

The aortic root represents the outflow tract from the left ventricle and has three coronary cusps. Located anteriorly is the right coronary sinus, giving rise to the right coronary artery (RCA), with the left main coronary artery (LMCA) arising from the left coronary cusp, then bifurcating into the left circumflex (LCx) and left anterior descending (LAD) arteries (Fig. 1). The RCA runs down the right atrioventricular groove with the conal branch arising as the first branch; however, in up to 22% of people, the conus arises directly from the aorta. In approximately 85% of individuals, the RCA gives rise to the posterior descending artery (PDA), known as right dominance. In 7% of patients, the LCx gives rise to the PDA, known as left dominance. In the remaining 8%, the RCA and LCx both provide supply to the PDA, known as co-dominance. The LMCA, usually 1–2 cm in length, arises from the aorta dividing beneath the left atrial appendix. In approximately 30% of cases, the LMCA gives rise to the third vessel between the LAD and LCx called the intermediate artery (or ramus intermedius artery).

Fig. 1
figure 1

Normal electrocardiogram gated multi-detector CT volume rendered reconstruction VRR showing normal coronary artery origins and course

Full size image

By definition, the distinction between coronary artery anomalies and variants is based on their population prevalence. Any anatomic finding found in > 1% of the general population is defined as normal, with a normal variant describing an alternative, relatively usual finding that is found in > 1% of the same population. An anomaly is a morphological feature seen in < 1% of the population [6, 7]. However, recent studies have challenged this strict definition as some particular CAA prevalence rates have been greater than 1% of the selected populations. Normal variants of coronary arteries are considered benign entities with very limited clinical significance [11,12,13, 20, 21]. In contrast, CAAs range from being benign to those associated with a high risk of sudden cardiac death (SCD). In particular, new evidence is emerging that age plays a significant role in prognosis with various studies and authors suggesting patients with CAA under the age of 30–35 are at higher risk of events than those with CAA found later in life incidentally [18, 22••].

Anomalies of Aortic Origin

Single Coronary Artery

SCA is a rare entity with a prevalence ranging between 0.024 and 0.066% in the general population [23] in which the RCA or LMCA is absent from its respective coronary sinus of Valsalva, with one coronary artery providing blood supply to the entire heart (Fig. 2). First described posthumously in 1699 by Fantoni, SCA was only described antemortem in 1967 through the use of coronary angiography [24]. Since that time, there have been multiple classification schemes developed, however Lipton et al. [25••] modified scheme has been widely accepted and divides SCA into three groups according to the location of the ostium, anatomical distribution and the course of its branches resulting in a three letter code. The first letter refers to the originating sinus of Valsalva [i.e. left (L) or right (R)]. Second, a Roman numeral (I, II or III) describes the anatomical course of the artery. In group I, the artery follows the anatomic course of the normal LMCA or RCA. In group II, one coronary artery arises from the proximal segment of the normally located other coronary artery. In group III, the LAD and LCx each arise separately from the normally positioned RCA. The final letter refers to the relation of the coronary arteries to the great vessels (i.e., anterior (A), posterior (P), between the great vessels (B), passing through the interventricular septum (S) and combined (C)) (Supplementary Table 1).

Fig. 2
figure 2

Single left coronary artery. a VRR image of RCA arising from the proximal LMCA (white arrow). b Axial image of single coronary artery (white arrow) arising from the left coronary cusp (LCC). c Axial image showing bifurcation of SCA to RCA (white arrow) and LAD (chevron)

Full size image
Table 1. Lipton Classification of Single Coronary Artery
Full size table

SCA is diagnosed based on two criteria [26]: (1) the presence of a single ostium from a single coronary sinus combined with the absence of an ostium in the opposite sinus and (2) a lack of origination of another coronary artery from an ectopic location. SCA can arise from any of the three coronary sinuses, with the following order of decreasing frequency: right coronary sinus, left coronary sinus and very rarely the non-coronary sinus [27].

Prognosis of persons with SCA is unclear; however, it has been associated with major adverse cardiac events in 15% of patients before the age of 40 [28]. In particular, when the anomalous artery course is interatrial and or traverses between the aorta and pulmonary arterial trunk, this can be associated with SCD and angina [24]. It has also been postulated that given the high arterial flow of a single coronary supplying a large territory, this predisposes to development of atherosclerosis and endothelial injury. Revascularisation is only recommended in patients with significant documented atherosclerosis and medically resistant ischaemia [29]. As there are currently no treatment guidelines, the best management approach in either symptomatic or asymptomatic patients with an interarterial course is unknown and should be individually tailored to patients based on symptoms and reason for presentation. In the majority of cases, SCA remains asymptomatic and the finding is benign. Note that SCA may be dually considered as an anomaly of course, however is only discussed in this section.

Multiple Ostia

Three or more ostia from any of the coronary sinuses is considered multiple. With studies showing a prevalence of between 0.19 and 0.6%, this represents a commonly encountered anomaly [12, 30–31]. This is most commonly seen as the absence of the LMCA with the LAD and LCx and intermediate artery, if present, arising directly from the left sinus of Valsalva (Fig. 3). Additionally, the RCA and conus branch may arise from the right sinus of Valsalva, or an artery may arise from the non-coronary cusp. This anomaly in isolation causes no haemodynamic compromise and is benign with no specific treatment required.

Fig. 3
figure 3

Multiple coronary ostia. a VRR of absent LMCA with LCx (white arrow) and LAD (chevron) arising from separate ostia from the left coronary cusp. b Axial image of LCx (arrow) and LAD (chevron) arising from the left coronary cusp from separate ostia

Full size image

Anomalous Coronary Artery Arising from the Opposite Sinus

ACAOS is a heterogeneous group of anomalies with variable presentation from benign incidental findings to causing SCD, especially in young athletes [32]. Prevalence estimates vary between studies and are likely to be in the range of 0.19–2.16% of all patients referred for non-invasive coronary imaging [21, 33,34,35,36]. A recent systematic review and meta-analysis showed a pooled population prevalence of 0.83%, with CTA revealing the highest pooled prevalence of all modalities assessed (ultrasound, cardiac MRI and catheter angiography) [37].

There are four different recognised types of ACAOS in order of decreasing prevalence: anomalous origin of the RCA from the left coronary sinus (Fig. 4), the LCx or Intermediate arteries from the right coronary sinus (Fig. 5), or the RCA itself, origin of the LMCA from the right coronary sinus (Fig. 6) and either the RCA or LMCA from the non-coronary sinus (Fig. 7) as the rarest variant [38]. Further sub-classification is based on the proximal course of the vessel. Courses include interarterial course (IAC), retro-aortic, trans-septal or subpulmonic and pre-pulmonic course. Only the IAC is haemodynamically significant with a well documented increased risk of SCD [32]. Further, a combination of any of the following anatomical features: acute angle of ostium, slit-like orifice or intramural course, is known to confer an increased risk and are poor prognostic markers (Supplementary Table 2) [39, 40].

Fig. 4
figure 4

a VRR of RCA (black arrow) originating from the left coronary cusp (LCC) anterior to the origin of the left main with proximal portion with an intraarterial course between the aortic and pulmonary trunks. b Axial image of RCA (white arrow) arising from the LCC. c Axial image of proximal RCA showing compression (white arrow), note inferior relation to pulmonary valve (black arrow)

Full size image
Fig. 5
figure 5

a LCx artery (white arrow) arising from the proximal RCA following a benign retro-aortic course occupying the AV groove. b Axial image of LCx (white arrow) with retro aortic course arising from the RCA (chevron)

Full size image
Fig. 6
figure 6

a VRR of anomalous long LMCA (white arrow) arising from the proximal RCA traversing anterior to the aorta and below the RVOT. It then divides into the LAD (black chevron) and LCx (white chevron). b Axial image of LMCA (white arrow) arising from the proximal RCA (chevron)

Full size image
Fig. 7
figure 7

a VRR of anomalous LMCA (black arrow) arising from the non-coronary cusp b. Axial image of anomalous LMCA (white arrow) arising from the NCC

Full size image
Table 2. ACAOS Classification and features associated with high risk of cardiac event
Full size table

There is currently considerable debate regarding the functional assessment and optimal management strategy of patients with either symptomatic or incidentally found ACAOS. Patients with ACAOS with a non-IAC, in the absence of symptoms, typically have an excellent prognosis, and intervention is not warranted. However, assessment and management of a vessel with an IAC are vital (Supplementary Table 3). Consensus in the literature focuses on thorough history taking, risk stratification, symptoms on presentation (syncope, angina or aborted SCD), anatomical factors (e.g. proximal narrowing) and individual patient factors (age, individual preference for management). Notwithstanding this, various protocols have been proposed. These mandate CTA or CMR for anatomical assessment, with varying protocols for the functional dynamic assessment with all involving invasive angiography aided by either IVUS or FFR and intra-coronary stress testing [37, 41•]. If assessed by a heart team to be clinically significant, surgical re-implantation remains the recommended treatment of choice with percutaneous coronary intervention not considered routine [37]. Current guideline recommendations for surgical revascularisation and functional assessment are shown in Supplementary Table 4 (18).

Table 3 Proposed investigation strategy in patients with ACAOS
Full size table
Table 4 Current 2008 ACC/AHA Guidelines for intervention of congenital coronary anomalies of ectopic arterial origin. [43]
Full size table

Despite the aforementioned recommendations, often focused in young patients, a recent study by Grani et al. has cast doubt on the necessity for ACAOS with IAC requiring treatment incidentally found first in middle age. In their small, retrospective matched 1:2 case control study with a mean age of 56 years, there was no significant difference in MACE rates at mean follow-up of 49 months in any of the three groups assessed (ACAOS with IAC, ACAOS without IAC and matched controls) [42•]. Despite its inherent limitations, and the absence of randomised controlled trials, this study suggests that investigation and aggressive corrective treatment should be dependent on the age of presentation or symptoms onset, not simply the anomaly itself. Similar studies and authors have supported this position [37, 43, 44].

Ectopic Origin from Non-coronary Sinus

Either the LMCA or the RCA can arise from the non-coronary sinus of Valsalva (Fig. 8), and this represents a very rare entity with no published prevalence data. Scant case reports have been documented, with contention remaining regarding its clinical significance [45, 46]. In cases found which are non-intramural and asymptomatic, it is likely a benign entity.

Fig. 8
figure 8

LMCA origin from the non-coronary cusp (NCC). a VRR of LMCA (white arrow) arising from the non-coronary cusp. b Axial of LMCA (black arrow) arising from the non-coronary cusp. RCC right coronary cusp, LCC left coronary cusp)

Full size image

Ectopic Origin from Ascending Aorta

Ectopic origin of coronary arteries from the ascending aorta is a rare entity with no published prevalence. It is defined as a vessel having its ostium originating more than 5 mm above the supravalvular ridge, or 10 mm above the sinotubular junction of the proximal ascending aorta (Fig. 9) [7]. This entity does not have any inherent haemodynamic consequence, but may present technical issues in relation to coronary catheterisation (i.e. difficulty with coronary engagement) and cardiothoracic surgery (i.e. cross-clamp placement below the high origin coronary may result in unsuccessful cardioplegia) [5]. Despite the general consensus of its benign nature, others have suggested that the high location of the ectopic ostium decreases coronary perfusion due to the angle at which it leaves the aorta, resulting in a flap like mechanism restricting flow at the ostium which can cause ischaemia [47]. In the absence of symptoms, however, this anomaly should be considered benign.

Fig. 9
figure 9

Ectopic origin of the RCA from the ascending aorta. a VRR of RCA (white arrow) originating above the sino tubular junction, running between the aorta and the RVOT. b Oblique HIP showing RCA (white arrow) arising from the ascending aorta above the sino tubular junction (dotted line)

Full size image

Anomalies of Non-aortic Origin

Anomalous RCA from Pulmonary Artery

ARCAPA is a very rare congenital heart defect with an incidence of 0.002% in the general population, most often diagnosed incidentally [48, 49]. Most are isolated anomalies; however, 25–30% can be associated with other structural heart defects such as ventricular septal defect, aortopulmonary window, tetralogy of Fallot and aortic stenosis [50]. Clinical presentation is variable with the most common presentations being a murmur for investigation, followed by chest pain and symptoms of congestive cardiac failure. Given the lower oxygen demands of the right ventricle, ischaemia and other symptoms are often dependant on collateral supply, coronary dominance and shunt size with the majority remaining asymptomatic (Fig. 10). Despite this, in the paediatric population, surgical correction is indicated even in asymptomatic patients to prevent the development of ischaemia and to reduce the risk of SCD. Treatment in asymptomatic adults is debated; however, most would favour surgical repair despite the paucity of long-term follow-up data [51].

Fig. 10
figure 10

a, c VRR of anomalous right coronary artery arising from the pulmonary artery (PA) (white arrow). b Axial image (white arrow)

Full size image

Anomalous LCA from Pulmonary Artery (Bland-White-Garland Syndrome)

This is a rare congenital coronary anomaly with a reported incidence of 1 in every 300,000 live births and constitutes 0.24–0.46% of all congenital cardiac disease. It is one of the most severe coronary anomalies with infant mortality rates reaching 90% if left untreated [52]. Due to the equivalent blood pressure and oxygen concentration in both left and right systems in utero, effects of this anomaly are not seen until birth, where pulmonary pressure and flow decrease resulting in low LAD blood flow and pressure. At this point, ischaemia stimulates formation of collateral supply between the RCA and LAD, with the left system subsequently draining into the low pressure pulmonary trunk leading to a ‘steal phenomenon’.

Clinical suspicion should remain high as symptoms in the paediatric population are often non-specific with colicky like crying bouts, failure to thrive, feeding difficulties and heart failure symptoms often present. In the 10% that survive to adulthood undetected, symptoms of angina, reduced exercise tolerance, smaller stature, syncope and SCD may be part of the presenting complaint. There are case reports of first diagnosis of ALCAPA in an elderly patient in their seventies [53]; however, this is extremely rare. Surgical intervention is the definitive treatment and is recommended in all, with studies showing long-term survival at 20 years nearing 95% with varying surgical techniques [54].

Anomalous Course Variation

Duplicated LAD

This increasingly recognised anomaly was initially thought to be present in approximately 1% [54] of the population; however, a more recent CTA study has demonstrated a prevalence of 4.18% [55••]. In this anomaly, two distinct segments of the LAD occupy the anterior interventricular sulcus (AIVS) of the heart. Depending on the origin, site of termination and course and location of the arteries in relation to the other anatomical structures, multiple distinct types have been reported. Currently, nine recognised variations have been described as determined by CTA with all having a ‘short’ and ‘long’ LAD. Type 1 (short LAD—origin is LAD proper, terminates high in AIVS, and Long LAD—origin LAD proper, descends LV side of AIVS and re-enters distal AIVS) is the most common variant (Fig. 11), representing 74–86% of cases [56, 57]. Rare case reports have documented origin of one branch of the LAD from the pulmonary artery with other variants possible [58].

Fig. 11
figure 11

VRR of type 1 duplicated LAD. Short LAD (chevron) terminating proximally in the anterior interventricular sulcus (AIS) with the long LAD (long white arrow) descending on the left ventricular side of the AIVS, re-entering the AIVS and terminating distally (short white arrow)

Full size image

There is no documented clinical consequence for this anomaly; however, awareness and recognition of its existence are important. In patients requiring revascularisation, issues may arise regarding surgical planning in regards to avoiding incorrect placement of arteriotomy and may impact on the number of grafts needed in the event of both LADs requiring revascularisation [57].

Myocardial Bridge

A myocardial bridge (MB) is a segment of the myocardium which overlies a coronary artery, while the coronary artery that tunnels referred to as the mural coronary artery (MCA) [59]. If the myocardium does not completely surround the whole circumference of the tunnelled segment, it is termed a ‘superficial’ bridge. If the complete circumference is surrounded, this is described as a ‘complete’ bridge (Fig. 12).

Fig. 12
figure 12

Mid LAD myocardial bridge seen in a VRR (white arrows) and b oblique views (white arrows). Proximal segments to the myocardial bridge are prone to atherosclerosis

Full size image

In the age of CTA, MB detection rates have increased significantly with prevalence estimated to be between 14 and 58%, compared to 0.5–16% on conventional coronary angiography [60]. Symptoms attributed to MB vary. Symptoms consistent with ischaemia may occur; however, the vast majority of patients remain asymptomatic. Multiple studies have demonstrated that atherosclerosis often accompanies MB, in particular in the proximal segment prior to the bridge, thought likely related to increased shear forces predisposing to plaque development [62••]. Recent studies have suggested that the degree of atherosclerosis was not responsible for symptoms and outcomes in patients, but was rather an epiphenomenon with no difference in the prevalence of atherosclerosis found between symptomatic and asymptomatic patients [60].

The most common site for MB is the LAD artery with incidence of between 64 and 79.4%, with the majority of these (39.3–60.4%) occurring in the mid-LAD [60, 61]. The remaining occur either in the diagonal branches of the LAD, LCX including the obtuse marginal branches or RCA, in descending order. In asymptomatic patients, this entity does not require treatment. In those with symptoms, calcium channel blockers or B-blockers are first-line treatment. Nitroglycerin should be used with caution, as reduction of pre-load can worsen symptoms by intensifying systolic compression of the bridged segment and vasodilating segments proximal to the bridge, exacerbating retrograde flow and reducing myocardial ischaemic threshold. In patients not responsive to medical management, MB resection or bypass grafting could be considered, although is rarely required. Percutaneous coronary intervention with drug eluting stents is inferior and is not routinely recommended [62••].

Intracavitary Course

Intracavitary coronary artery (ICCA) is defined as a coronary artery that courses through the myocardium and into a cardiac chamber. CTA studies have estimated a prevalence of 0.33% with two variants described. The majority (0.29%) occur in the mid to distal RCA and 0.039% in the distal LAD [63]. ICCA is thought to be a benign anomaly with no reported clinical symptoms, however does pose a challenge to invasive cardiac procedures. It can seriously complicate coronary artery bypass grafting, pacemaker insertion, ablation procedures and right heart catheterisation.

Anomalies of Termination

Coronary Fistula

Coronary artery fistula (CAF) is a rare anomaly resulting in an abnormal communication between the coronary artery and either the cardiac chambers, superior vena cava, coronary sinus or pulmonary artery or veins, bypassing the myocardial capillary bed. It has an estimated prevalence of 0.9% in CTA studies, significantly more than the previously accepted rate of 0.05–0.25% as found on coronary angiography [64]. CAF is an isolated finding in the majority of cases, however may be associated with other congenital anomalies such as atrial septal defects (ASD)/VSD, tetralogy of Fallot, aortic or pulmonary atresia and patent ductus arteriosus (PDA). The majority of patients with CAF are asymptomatic and are usually detected incidentally (Fig. 13). However, some patients may become symptomatic with the degree of symptoms often dependent on the size of the CAF and the amount of myocardium supplied.

Fig. 13
figure 13

a Arteriovenous fistula (AVF) (white arrow) arising from the mid LAD after the diagonal artery takeoff, traversing anterior to the RVOT as a tortuous complex of vessels and connects to the pulmonary artery (PA), terminating as an aneurysm. b AVF connecting to PA (white arrow) showing contrast jet

Full size image

The incidence of CAF is as follows: both coronary arteries (45.5%), from the left coronary artery (42.4%) and the right coronary artery (12.1%). Further, almost a quarter are associated with aneurysms of the coronary arteries and routinely show increased tortuosity [65]. Regardless of their point of origin, nearly 90% of fistulas drain to the right chambers of the heart. The majority of cases do not require any intervention; however, in those with medically resistant symptoms or existence of a significant right-to-left shunt causing ventricular overload, surgical repair could be considered. If detected in paediatric patients, they are usually repaired to avoid late onset complications, with registries showing very low post operative long-term mortality and morbidity rates [66, 67].

Systemic termination of a coronary artery should be distinguished from CAF. In this related condition, there is no associated enlargement or tortuosity in particular at the terminal end of the affect artery as there is no significant pressure difference between the coronary artery and the systemic artery into which it terminates. In absence of a pressure difference, which can be caused by the presence of atherosclerosis, systemic termination is unlikely to be clinically significant [5].

Coronary Artery Arcade

This is a rare anomaly involving inter-coronary communication between the terminal ends of two coronary arteries resulting in two-way blood flow with prevalence estimated at 0.02% [68]. Two types have been described, with communication between the RCA and LCx in the posterior atrioventricular groove, and communication between the LAD and PDA in the interventricular groove [69]. Unlike collaterals, these are found in otherwise angiographically smooth coronary arteries and have a well-defined muscular layer and are larger in diameter. The functional and clinical importance of this anomaly is not clear, and in the absence of symptoms does not require any intervention or treatment.

Anomalies of Intrinsic Coronary Anatomy

Congenital Coronary Stenosis

The majority of coronary artery stenoses are acquired; however, congenital coronary stenosis has been described. These can either be ostial, due to a valve-like ridge of the aortic wall or fusion of the aortic leaflets and aortic wall, or may be peripheral [70]. There is a paucity of prevalence data and management guidelines in the literature for this entity.

Congenital Coronary Ectasia or Aneurysm Formation

Coronary artery ectasia (CAE) has been reported to have a prevalence of between 0.3 and 5.3% as diagnosed on coronary angiography and is defined as a dilatation of an arterial segment diameter of > 1.5 times that of the adjacent normal coronary anatomy and can be either focal or diffuse (Fig. 14). The main aetiology is believed to be congenital (20–30%) or inflammatory (10–20%) with the remainder likely due to connective tissue disorders such as Ehlers-Danlos syndrome, Kawasaki disease and scleroderma [71].

Fig. 14
figure 14

a VRR showing widespread coronary artery ectasia in the proximal and distal RCA, LAD and LCx (white arrows). b Axial images showing significant RCA ectasia (white arrows) in multiple segments

Full size image

In contrast, coronary artery aneurysm (CAAN) is defined as a diffuse focal arterial dilatation with a diameter of 2 or more times larger than that of its normal contiguous segment. CAAN are termed giant if they are > 4 times the size of the reference vessel diameter. Recent CTA studies have reported a prevalence of 2.31% of CAAN [72]. The most common cause is atherosclerosis (50%), congenital anomalies (20–30%), inflammatory or vasculitis (10–20%) and connective tissue disorders (5–10%) [73]. CAE and CAAN differ in that CAAN often involve shorter segments, their beginning and end points are sharper, and are generally encountered with a larger diameter. Both entities occur most frequently in the RCA with a 2 to 1 ratio compared to the left system [71]. Optimal management is debated with both PCI and surgery an option dependent on size, symptoms and location.

Conclusion

Coronary artery anomalies are being detected with increasing frequency given the more widespread use of CTA, so it is important for physicians reading CTA to have a comprehensive knowledge of these anomalies, in particular malignant variants that may require intervention. Previously, catheter coronary angiography was considered the gold standard for CAA assessment in those with known anomalies; however, CTA has now emerged as the preferred method given its non-invasive nature and its ability to provide excellent spatial and temporal resolution, and more accurate and reproducible anatomical assessment.

As CTA-based prevalence and long-term follow-up data become available, clinicians will be better able to estimate the risk associated with each type of CAA. In addition, given that most clinical management guidelines are based on retrospective registries and expert consensus, appropriately powered randomised controlled trials are required to better define optimal management strategies in all age groups, particularly for anomalies with a malignant course in whom surgery may be safe to avoid. Well-constructed clinical registries with long-term follow-up should also be developed to further add to the evidence base.