A 60-year-old man with a history of cardiac arrest 16 years prior presented via ambulance with substernal pressure of 2 hour duration. An electrocardiogram revealed prominent ST-segment elevation in the inferolateral and apical leads. His blood pressure was 138/73 mmHg, and his heart rate was 53 beats per minute. He had received an aspirin in the ambulance, and intravenous nitroglycerine, heparin, and eptifibatide were initiated in the emergency room.
The medical history was significant for a cardiac arrest that occurred 16 years prior. The patient believed cardiac catheterization at that time revealed a blocked artery “at the bottom of the heart,” but no percutaneous coronary intervention (PCI) was performed. He recalled no further cardiac events since that time. Otherwise, he had a history of treated hypertension and gout. Home medications included extended-release metoproplol 50 mg a day, amlodipine 10 mg a day, and allopurinol 300 mg a day. The patient drank socially and did not smoke.
During emergent cardiac catheterization, initial attempts to engage the left main coronary artery (LM) were unsuccessful. Subsequent right coronary angiography, however, produced simultaneous LM opacification (Figure 1). The right coronary artery (RCA), a small codominant vessel, had a chronic total occlusion (CTO) in the mid segment and a prominent network of bridging collaterals (Figure 2). Selective LM angiography was performed with counter-clockwise torquing of the same Judkins right-4 catheter and revealed a long LM segment. The left circumflex artery (LCx) contained a proximal roller coaster-like 3,60o loop. An ulcerated 95% stenosis was seen at the bifurcation of a large first obtuse marginal branch (OM) (Figure 3). The left anterior descending artery (LAD) contained two 85% mid stenoses and one 95% distal stenosis at the apex (Figure 4). Thrombolysis in myocardial infarction (TIMI)-3 flow was observed in all 3 coronary systems. Although inferoapical hypokinesis was seen on right anterior oblique left ventriculography, the overall ejection fraction was preserved.
The combined scenario of a history of cardiac arrest, potential anomalous LM compression between the great vessels, and triple-vessel disease prompted emergent cardiac surgical consultation. The surgical opinion, however, was to delay bypass surgery beyond the acute event, due to increased operative risk in the setting of possible right ventricular infarction. Thus, due to ongoing angina and ST-segment elevations, the decision was made to proceed with emergent PCI of the LCx.
Guiding roadmap image with an 8-Fr Amplatz right-1 side-hole catheter again demonstrated the 3,60o “roller coaster” proximal LCx loop (Figure 5). Two ATW Marker guidewires (Cordis Corporation, Miami Lakes, Florida) were advanced into the OM and LCx (Figure 6). Each vessel was then serially pre-dilated with a 3.0 x 15 mm Sprinter balloon (<i>Medtronic Corporation, Minneapolis, Minnesota</i>). Given the AMI setting, as well as potential future bypass surgery, we elected to employ bare metal stents in this setting. Simultaneous kissing stents technique was performed using two 3.0 x 18 Driver stents (Medtronic Corp.). These were advanced, with some difficulty, into each coronary limb beyond the stenotic bifurcation. The stents were then simultaneously retracted proximally, with care taken to overlap both proximal edge markers in the mid LCx. Sequential inflations of 15 atmospheres (atm) were then performed in the OM and LCx. Final kissing inflation was performed at 8 atm (Figure 7), with an excellent final angiographic result (Figure 8). The patient’s angina and ST elevations resolved shortly thereafter.
The following day, he underwent cardiac computerized tomographic angiography (CTA) to delineate the anomalous LM course. The LM was seen to arise low within the right aortic cusp, underneath the right ventricular outflow tract (Figure 9), but not entrapped between the great vessels. Transverse images revealed a round and non-compressed lumen, without evidence of flaps or ridges. Additionally, no acute angulations were observed in the left coronary system. The images were reviewed by 2 separate readers. Both felt that, in light of the patient’s age, this was a benign anatomy and did not require surgical correction. After further surgical consultation regarding the details of potential operative risk, the patient and family promptly elected for PCI.
The following day, right transradial PCI (TRPCI) was performed with balloon predilatation of the 2 mid stenoses using a 6 Fr Amplatz right 1 guide catheter, a Whisper wire (Abbott Laboratories, Abbott Park, Illinois) and a 2.0 x 15 mm Maverick balloon (<i>Boston Scientific Corporation, Natick, Massachusetts</i>). The guide catheter support, however, proved inadequate for stent advancement. The Whisper wire was exchanged over a Transit catheter (Cordis Corporation, Miami Lakes, Florida) for an Iron Man extra support wire (Abbott Laboratories), and the guide catheter was “upsized” to a 7 Fr Amplatz right 1 catheter. Two Taxus drug-eluting stents (DES) (<i>Boston Scientific Corporation</i>) (2.5 x 12 mm and 2.5 x 20 mm) were subsequently deployed into the mid-LAD (Figure 10), with an excellent final result (Figure 11). We elected to medically treat the distal LAD lesion at the apical vessel terminus. The patient was discharged on his previous medical regimen with additional daily doses of clopidogrel 75 mg, uncoated aspirin 325 mg, lisinopril 5 mg, and simvastatin 40 mg. A 6-month stress nuclear perfusion scan revealed small, fixed apical and inferior defects only. Although he has remained asymptomatic, we have reemphasized that surgical correction would be required if future symptoms or objective evidence of ischemia developed in the absence of new coronary stenoses.
Epidemiology.</b> Anomalous coronary arteries (ACA) originating from the opposite aortic sinus are a rare congenital anomaly, reported in 0.1–0.3% of all patients undergoing coronary angiography.<sup>1–10</sup> The occurrence of left coronary origin from the right sinus of Valsalva is estimated at 0.6–1.2% of total coronary anomalies.7 Associated defects in coronary anatomy include bicuspid aortic valve, mitral valve prolapse, and ventricular septal defect.3 Due to the relative rarity of these aberrations, however, the overall incidence among sudden death autopsies has been reported to be from 0.6%<sup>3</sup>–1.2%.<sup>11</sup> Kapoor and co-presenters described an interesting case of quadriostial ACA. In this patient, all 3 major coronary vessels, as well as the conus branch, arose from separate ostia within the right sinus.<sup>12</sup>
A major consideration in these patients is the potential for sudden cardiac death. When the ACA courses between the aortic and pulmonary arterial trunks, dilatation of the great vessels during times of increased cardiac output may result in coronary compression.<sup>4</sup> In these individuals, childhood sudden cardiac death can occur with<sup>5–7</sup> or even without physical exertion.<sup>1,2,4</sup> One study noted that 19 of 27 sudden deaths in patients with the left coronary artery originating from the opposite sinus occurred between the waking and physically active hours of 3 to 9 pm.<sup>8</sup> Pre-morbid symptoms occur in only 18–30% of patients;<sup>4</sup> these may include dyspnea on exertion, chest pain, syncope, or dizziness.<sup>2,4</sup>
A review of 51 autopsy cases from the Armed Forces Institute of Pathology revealed no cases of sudden death when the anomalous RCA arose from the left sinus of Valsalva. All patients who died suddenly demonstrated sharp angulation in the ACA course between the great vessels and a narrowed ostium.<sup>5</sup> Sudden cardiac mortality dramatically decreases among individuals free of events through adolescence;1 survival beyond age 50 is associated with a favorable prognosis.<sup>1,2,5,6</sup> One report described a 54-year-old patient with anomalous LM from the right cusp who experienced sudden cardiac death. This ACA, however, unlike that of our patient, had a slit-like compressed ostium.<sup>6</sup>
Roberts and Shirani classified anomalous LM from the right coronary cusp into 4 subtypes. They concluded that myocardial ischemia and sudden cardiac death only resulted from the interarterial course between the great vessels. The remaining ACA locations: anterior to the pulmonary artery, retroaortic, and behind the right ventricular outflow tract (infracristal), were all associated with benign cardiac prognoses.<sup>3</sup> In their pathological series, Taylor and others reported that in individuals with a contralaterally located coronary, those with a single ostium appeared to have a more benign course than those with separate ostia. Moreover, amongst patients with single coronaries, those originating from the right sinus of Valsalva were more likely to experience sudden death, likely reflective of aberrations of the LM or LAD ostia.<sup>11</sup>
Flaps or ridges can sometimes be found in compressed coronary ostia in victims of sudden death.<sup>13</sup> Moreover, these vessels often arise at acute angles, further adding to flow limitations. When the LM arises from the right cusp or RCA, it may initially traverse intramuscularly through the interventricular septum. This finding alone, however, does not appear to result in adverse outcomes.<sup>14</sup> The origin of the LAD from the right sinus may be associated with tetralogy of Fallot; this anomaly is an important surgical consideration to avoid inadvertent transsection of the vessel.<sup>15</sup>
In some cases, the ostial narrowing is related to the artery’s intramural location within the aortic wall.<sup>11</sup> Occasionally, identification of this anatomic variant may only be possible intraoperatively. Angelini and coauthors reported intravascular ultrasonographic findings of intramural ACA and identified proximal segmental intussusception as the main pathophysiologic mechanism. The hypoplasia index, defined as the ratio of the proximal to distal circumference, as well as the asymmetry index, the ratio of the minimal to maximal diameters in the oval intramural segment, were proposed as additional measures of severity.<sup>16</sup> One pathological series, however, found no correlation of differences in length of aortic intramural segment, coronary ostial size/shape, degree of ACA displacement, or vessel angulation with sudden cardiac death. The only predictive factor not resulting in sudden cardiac mortality was an age of > 30 years.<sup>1</sup>
While prodromal symptoms may occur prior to cardiac arrest in susceptible individuals, the complaints are often nonspecific and deceptively benign, especially among the younger population. In a series of 27 young athletes who were sudden death victims, there were 4 cases of previous syncope and 5 cases of chest pain. Unfortunately, all cardiac diagnostic modalities, including rest and stress electrocardiography, as well as transthoracic echocardiography were unrevealing. Of note, all but 2 of these patients succumbed either during or shortly after strenuous physical exertion.<sup>8</sup>
Other workers, however, have reported identification of ACA with both transthoracic and transesophageal echocardiography; optimal imaging is obtained from short-axis view of the aortic root at the coronary sinus level.<sup>1,4,7</sup> In one series, transthoracic echocardiography correctly identified 9 of 10 patients with an intramural aortic ACA course. Color flow Doppler mapping at lowered Nyquist limit of 20 to 40 cm/sec can be useful to visualize the linear diastolic ACA flow.<sup>7</sup>
Coronary angiography can delineate the ACA’s sinus of origin, degree of atherosclerotic disease, as well as ostial compression.<sup>17</sup> Concomitant insertion of a pulmonary artery catheter and coronary angiography from the left lateral projection can aid in identification of the aberrant vessel’s course in relation to the great vessels. The retroaortic “dot” sign in the right anterior oblique projection left ventriculography represents the short axis image of an anomalous LM or LCx. If the “dot” is anterior to the aorta, an interarterial ACA from the contralateral cusp should be suspected. With an anomalous left coronary following an anterior or intramyocardial septal course, right anterior oblique left coronary angiography yields a elliptical shape known as the “eye” sign.<sup>18</sup> Invasive functional evaluation can also be performed. Sayar et al presented a patient with an anomalous RCA as well as a 90% LAD stenosis. Fractional flow reserve interrogation of the ACA was within normal limits. LAD PCI was subsequently performed and surgical intervention was avoided.<sup>19</sup>
Three-dimensional rendering is possible with CTA and magnetic resonance imaging (MRI), today’s gold standards for evaluation of ACA.<sup>1,2,4</sup> These modalities provide clear, non-invasive visualization of the ACA course in relation to surrounding anatomic structures. Additionally, combined pharmacologic stress testing further allows for functional evaluation of any detected coronary stenoses.<sup>20</sup> Van Straalen and coworkers reported 2 interesting cases in which the anomalous LAD courses were delineated by CTA performed in combination with placement of coronary guidewires into the aberrant vessels.<sup>21</sup>
In patients with high-risk clinical and anatomic features, surgical repair is the treatment of choice. While coronary bypass surgery has been the traditional technique, pediatric data suggest that unroofing of the ACA with re-implantation into the appropriate aortic sinus may be associated with improved patency.<sup>2</sup> Frommelt and colleagues reported favorable long-term results with this procedure in their surgical series of patients with intramural ACA. Given the lack of circumferential suture line and the creation of a permanent unobstructed neo-ostium, the authors recommend that even asymptomatic individuals with this anatomic variant undergo this procedure.<sup>7</sup>
Alternatively, in the patient with a benign ACA morphology and favorable clinical history, PCI of in situ atherosclerotic lesions can be safely performed. Even in post-operative ACA patients, PCI can be performed on conduit stenoses.<sup>22</sup> Primary PCI of ACA in patients with AMI has been reported previously.<sup>23–30</sup> Conde-Vela et al as well as Pomar and coauthors, described primary PCI in anomalous RCA with separate and single ostia, respectively.<sup>23,24</sup> Another report featured a 73-year-old patient who underwent primary double-vessel stenting of an anomalous LM originating from the RCA as well as the RCA itself during AMI.<sup>25</sup> Additionally, primary as well as rescue PCI of anomalous LAD, LCx, and single coronary from the right sinus of Valsalva have been described.<sup>25–30</sup> However, based on our Medline Ovid search from 1955 to now, we found no cases of simultaneous kissing stent deployment in ACA patients, either electively or during AMI. This technique has previously proven to be a safe and effective treatment of bifurcation lesions in patients without ACA.<sup>31</sup>
While PCI of the normally located vessel in patients with coexistent ACA usually poses no additional technical challenges, ACA PCI often requires innovative technique and nontraditional equipment selection. Initially, roadmaps involving multiple detailed, orthogonal views of the proximal ACA segment are beneficial to plan the interventional strategy. In 1982, Schwartz and associates described the first balloon angioplasty in an ACA. Proper guide catheter engagement of the anomalous LCx was possible only after manual removal of the catheter tip.<sup>32</sup> In a small primary PCI series of 8 AMI patients with anomalous RCA, all 6 ACA arising from the anterior aortic wall above the sinotubular line required Amplatz guiding catheter support. The remaining 2 from the left sinus were adequately treated with the Judkins left catheter.<sup>33</sup> At times, as in our patient, larger catheter sizes are required for enhanced back-up support to accommodate difficult anatomies.<sup>34</sup> Advancement of 2 stents, as occurred in our case, clearly requires both larger guide catheter lumen and reinforced back-up support.
When extreme difficulty in guide catheter engagement is encountered, initial cannulation with the more flexible and steerable diagnostic catheter can be considered. An exchange-length, extra-support guidewire is then advanced distally into the ACA to facilitate exchange to the appropriate guide catheter. In the case of a retroaortic ACA, the prolonged vessel length may prevent the balloon or stent from reaching the stenosis. In these situations, longer balloon/stent shafts or shorter guiding catheters should be considered.<sup>18</sup>
The frequently tortuous ACA morphology poses challenges to guidewire advancement. Topaz and others propose advancement of the balloon catheter to the guide catheter tip to facilitate guidewire support.35 Furthermore, a stiff guidewire shaft can aid in coaxial guide catheter alignment.<sup>18</sup> When an anomalous LCx arises from the proximal RCA, the exit angle is typically acute. While a sharp guidewire curve is necessary to engage the LCx ostium, any further advancement usually results in wire prolapse into the RCA. We have previously described a double wire technique utilizing the steerable SteerIt (<i>Cordis</i>) guidewire in such anatomies. With one guidewire in the distal RCA, an initial gentle curve is formed on the SteerIt wire. This is then advanced into the proximal RCA, just beyond the LCX ostium. The curvature is then increased to a sharp, almost perpendicular angle; and the wire is slowly withdrawn. The tip subsequently “falls” easily into the LCX ostium. Immediately, the tip is relaxed and the wire is easily advanced into the distal vessel.<sup>36</sup> Due to the complexity of these interventions, the operator must be mindful of potentially unusual complications. In one such case, a fractured guidewire segment was retained in the coronary and could not be retrieved.<sup>37</sup>
Although today’s monorail balloon/stent systems offer excellent tracking ability, over-the-wire catheters are potentially superior in the negotiation of difficult tortuous ACA. Due to unusual anatomies, predilatation should be encouraged and direct stenting avoided in ACA, unless the operator is convinced of its feasibility. In their series of over 20 ACA PCI over a 4-year period, Keelan and Holmes reported procedural success in all but 2 cases. Both failures involved anomalous RCA from the left cusp. Guide catheter engagement was not possible in the first patient, and excessive ACA tortuosity prevented balloon advancement in the other.<sup>18 </sup>
Occasionally, even a simple maneuver such as breath holding may be beneficial. Deep inspiration causes caudal displacement of the diaphragm, resulting in increased distance between the stationary guide catheter in the aortic root and the heart. This technique may facilitate engagement of a superiorly oriented ACA ostium. Difficulty may be encountered during intracoronary device advancement, due to tortuosity, calcification, or previously deployed stent struts. The same caudal cardiac displacement may aid in mild straightening of the vessel anatomy as well as enhance parallel alignment of the advancing device with the stent lumen to avoid strut entanglement.<sup>38</sup>
The efficacy and enhanced patient safety and comfort of TRPCI is well-established.<sup>39–41</sup> A recent study demonstrated significant reductions in transfusion rate and 1-year mortality when compared to traditional transfemoral PCI.<sup>42</sup> This approach has been used in the treatment of other forms of congenital heart disease. We previously presented the first TRPCI in a patient with dextrocardia.<sup>43</sup> TRPCI in patients with anomalous RCA from the contralateral sinus has previously been reported. The authors suggested that the right transradial approach offers improved guide catheter orientation for an ACA from the left aortic cusp.<sup>44</sup> However, based upon our Medline Ovid search from 1955 to present, no cases of LAD TRPCI in a patient with anomalous LM from the right coronary sinus have been described.
Post-PCI follow-up care of these patients bears discussion, especially in individuals with anatomies of borderline significance. Close clinical monitoring for recurrent angina, dizziness, and syncope is crucial. Although stress electrocardiography may be of limited utility in this setting, imaging modalities can be helpful. In addition to de novo diagnosis of an ischemic ACA morphology,<sup>10</sup> post-PCI thallium myocardial scintigraphy has been shown to identify ACA restenosis.<sup>45</sup> In our patient, this evaluation is of particular importance. Although his coronary course had no major high-risk features, the combination of anomalous LM and left dominant system in a patient with treated triple vessel disease warrants vigilant surveillance.
In summary, our case was unique in a number of respects. We have not previously encountered descriptions of PCI in ACA with a 360? “roller coaster” loop. Additionally, there have been no reports of PCI utilizing the simultaneous kissing stent technique in patients with this anomalous coronary anatomy, either with or without concomitant AMI. Other AMI ACA cases have not involved staged TRPCI, and we have found no reports of TRPCI involving an anomalous left coronary system. Accordingly, our case simultaneously illustrates the potential feasibility of both transfemoral and transradial approaches in percutaneous therapy of ACA patients. Like other authors, we employed nontraditional, supportive guide catheters, such as those with larger lumen Amplatz curves. Moreover, we have found hydrophilic wires such as the Whisper wire to be advantageous in negotiation of difficult curves in these settings.
The indications for catheter-based versus surgical intervention in our case are not entirely clear-cut. While we would have preferred emergent bypass surgery during the AMI, this was not possible. Although the LM did not technically traverse between the great vessels, it did pass underneath the right ventricular outflow tract. The symmetric, round ostial and proximal LM lumen, as well as a lack of vessel angulation, were important benign features, as interpreted by 2 independent physicians. Given the previous LCx intervention, the potential long-term benefits of subsequent surgical revascularization with only residual LAD disease would have been diminished. Additionally, bypass surgery would, by necessity, be delayed minimally by 1 month for clopidogrel therapy. We felt that medical therapy for the LAD stenoses during that period would carry significant risk. Alternatively, we proposed LAD PCI with bare metal stents and subsequent surgery the following month. Our surgical colleagues, however, were concerned about graft patency in that setting, given the lack of demonstrable LM flow limitation. Since the decision was made against surgical intervention, DES were deployed in the staged LAD intervention.
Although the exact circumstances surrounding his cardiac arrest 16 years ago were unclear, our patient had undergone extensive cardiac evaluation at that time and was told that he had sustained an infarct at the bottom of his heart. Albeit only codominant, the RCA was confirmed angiographically to be chronically occluded. While the possibility of potential ACA contribution to the cardiac arrest cannot be totally eliminated, his age as well as lack of symptomatology before and after that event speak against it. In the final discussion with the patient, our surgical colleague stated that, in his opinion, the risks of surgery outweighed the remote potential benefit of sudden death prevention.
Modern technical innovations in PCI now allow the interventionalist to broaden the treatment of coronary disease to unusual and difficult anatomies. In the case of ACA, meticulous definition of the anatomic course and angulation, thorough equipment selection, and creative technical adaptations are essential. Additionally, the transradial approach can be a feasible and safe alternative approach. CTA or MRI provide excellent delineation of the anomalous course. When appropriate, a multi-disciplinary approach involving surgical consultation is useful in planning the therapeutic strategy, as exemplified in our case.
Many thanks to Dr. Steven Frohwein for his assistance in the review of CTA.