Isolated Right Ventricular Infarction after Modified Cabrol Operation

Abstract

We report the case of a 27-year-old woman with a rare presentation of right ventricular failure secondary to isolated right ventricular myocardial infarction, 3 weeks after an uncommon surgical procedure, the modified Cabrol operation. Her medical history also included a Ross procedure at the age of 12 years. On the basis of her subacute presentation and a consultation with cardiac surgeons, we decided on medical management. Follow-up echocardiography at 6 months revealed that the right ventricular systolic function remained severely impaired, but the patient was asymptomatic with excellent functional capacity.

We review the surgical techniques of aortic graft replacement and their respective complications. We also discuss the impact of conservative and reperfusion strategies on prognosis and long-term outcomes in the setting of right ventricular infarction.

The Cabrol operation involves replacement of the ascending aorta and aortic valve with a composite graft and subsequent attachment of a Dacron conduit between the coronary arteries and the graft.¹ Kinking and thrombosis of the arterial limbs of the Dacron conduit, although rare, can lead to morbidity. We describe a rare case of isolated subacute right ventricular (RV) infarction leading to RV systolic dysfunction after a modified Cabrol procedure. The RV infarction was the result of thrombosis of the right arterial limb of the Dacron conduit, which was attached to the right coronary artery (RCA). We discuss the various techniques used in composite graft replacement and present the potential mechanisms responsible for the sequelae that arose in this case. We also describe how the resilient nature of the RV influences prognosis and outcome after RV myocardial infarction.

Case Report

A 27-year-old woman presented with a 2-week history of severe lower-extremity edema and a 30-lb weight gain. Three weeks earlier, she had undergone a modified Cabrol operation with use of a Carbomedics composite valve graft (SORIN Group USA, Inc.; Arvada, Colo) for recurrent severe aortic insufficiency associated with markedly dilated sinuses of Valsalva and aortic annulus. At the age of 12 years, she had undergone radiofrequency ablation for recurrent tachyarrhythmia and syncope caused by Wolff-Parkinson-White syndrome, and a Ross procedure to treat aortic insufficiency associated with a bicuspid aortic valve. Before the recent Cabrol operation, coronary angiography had revealed a very small, nondominant RCA and no evidence of atherosclerosis in the coronary arteries.

On presentation, the patient was found to have 3+ pitting edema in both lower extremities, elevated jugular venous pressure of 10 cm H₂O, and a prominent parasternal RV heave. Cardiac biomarkers and electrocardiography revealed no evidence of acute ischemia. After the recent Cabrol procedure, she had been placed on anticoagulation medication because of the mechanical aortic valve; the international normalized ratio was 2.5 at the current admission. Transthoracic echocardiography revealed a left ventricular (LV) ejection fraction of 0.55 to 0.60 and paradoxical motion of the septum consistent with RV volume overload, and a dilated RV with akinesis of the anterior, inferior, and lateral walls of the RV that led to a markedly reduced systolic function with preserved RV apical contraction. Right-sided heart catheterization results were consistent with normal pulmonary capillary wedge pressure and pulmonary artery pressure. A computed tomographic scan showed the right limb of the Dacron conduit attached to the RCA and the left limb attached to the left main coronary artery (Fig. 1). The scan also revealed thrombosis of the right limb of the conduit and patency of the left limb. Left-sided heart catheterization confirmed patency of the left limb (Fig. 2) and occlusion of the right limb (Fig. 3), and showed evidence of left-to-right grade III collateral vessels. No significant stenosis was noted in the left coronary system. On the basis of the subacute presentation and consultation with cardiac surgeons, we decided to continue with medical management. The patient underwent optimal diuresis with achievement of euvolemia before discharge from the hospital. A β-blocker and an angiotensin-converting enzyme inhibitor were also prescribed.

Fig. 1 A computed tomographic scan reveals complete thrombosis of the right arterial limb (white arrow) and patency of the left arterial limb (arrowhead) of the Dacron interposition graft. Note that the Dacron graft is attached to the right side of the aorta (Ao) (black arrow) and it provides 2 synthetic arterial limbs: right arterial limb and left arterial limb, which are attached to the right coronary ostium and left coronary ostium, respectively.

Fig. 2 Angiogram shows patency of the left arterial limb (arrow).

Fig. 3 Angiogram shows complete thrombosis of the right arterial limb (arrow).

Follow-up echocardiography at 6 months revealed that RV systolic function remained severely impaired, with no significant improvement. However, the patient remained completely asymptomatic with excellent functional capacity.

Discussion

Composite aortic graft replacement involves simultaneous replacement of the ascending aorta and the aortic valve. It is the standard procedure for the treatment of aortic annular dilation, ascending aortic aneurysm, dissection, and associated aortic regurgitation. Since the first description of the procedure (classic Bentall technique) by Bentall and DeBono in 1968,² many techniques have been described in an attempt to improve longevity, morbidity, and survival, and lower the rate of complications. These techniques include the modified Bentall, classic Cabrol, and modified Cabrol.

In the classic Bentall approach, the aortic valve and ascending aorta are replaced with a composite graft, the coronary ostia are attached to the graft (the coronary ostia are not mobilized before attachment), and then the aneurysmal wall is wrapped around the graft.² The risks of this approach include tension-induced pseudoaneurysm formation at the anastomotic sites of the coronary ostia or at the distal aortic graft suture line, and inaccessibility of the anastomotic sites, making hemostasis difficult.

Kouchoukos and colleagues³ described the modified Bentall approach, which is the most widely used technique. The modified Bentall approach is also known as the “open” or “button” technique. It involves resection of the aneurysmal wall, composite graft replacement, and then reattachment of the mobilized coronary buttons (coronary ostia with a rim of aortic tissue surrounding them) to the composite graft. Advantages of the button technique include improved accessibility to anastomotic sites, and less tension at the suture sites, which helps prevent pseudoaneurysm formation. Disadvantages include the additional time required to mobilize the coronary ostia, the risk of damaging these vessels, and the remaining possibility of occlusion caused by tension.

The classic Cabrol technique was developed by Cabrol and colleagues in 1981.⁴ This technique involves replacement of the aortic valve and ascending aorta with a composite graft, side-to-side attachment of a Dacron interposition conduit to the aortic graft, and then end-to-end attachment of the coronary ostia to the Dacron conduit. The coronary ostia are not mobilized from the aneurysm before attachment to the conduit, and the aneurysm is resected. This Dacron conduit serves to release the tension of reimplantation of the coronary ostia, preventing the formation of pseudoaneurysms. In the modified Cabrol technique, the coronary ostia are mobilized before reattachment to the interposition conduit and resection of the aneurysm.¹ The Cabrol operation is preferred for reoperations (because of tight adhesions, for example) and in cases of extensively calcified aneurysmal aortas, dissection extending into the coronary ostia, and low-lying coronary ostia in which mobilization of the coronary ostial buttons to the ascending aortic graft would result in tension.¹

The aforementioned techniques can be used for coronary artery reimplantation if the coronary ostium and its distal course are not involved in the dissection. However, when the dissection encircles the affected coronary ostium and extends along the left or right main coronary artery, additional techniques must be considered. In some cases, there is total dehiscence of the affected ostium from the true aortic lumen with intussusception of the inner wall cylinder into the outer coat of the coronary artery. In the absence of intussusception, a longitudinal incision can be made through the affected coronary artery with subsequent patch repair using either a saphenous vein or autologous pericardium. If intussusception is present or if the dissection will require a long patch, bypass grafting can be used. Bypass grafting can also be used when the coronary ostial diameter is small.⁵

The following factors can contribute to arterial graft occlusion after the Cabrol operation: intermittent or persistent kinking of the interposition conduit caused by oversizing of the conduit; a low-flow state caused by small diameter of the coronary ostia or by excessive angulation at the anastomosis of the coronary ostia with the interposition conduit; stenosis or kinking of the anastomosis of the coronary ostia with the interposition conduit; thrombosis of the arterial limbs of the interposition conduit caused by misalignment at the coronary ostium; redissection of the coronary ostium; and pseudoaneurysm formation caused by excessive suture tension at the anastomoses of the interposition conduit to either the aortic graft or the coronary ostia.

Coronary graft occlusion after the Cabrol operation has a reported incidence of 4%⁶ and occurs more often in the right coronary system than in the left.⁷ This finding has been attributed to the 90° diversion of blood flow at the inferior end of the Cabrol graft as it connects to the RCA. Balloon angioplasty with bare-metal stent or drug-eluting stent implantation, and minimally invasive direct coronary artery bypass (MIDCAB), are two of the procedures that have been described for the management of complications associated with stenosis or thrombosis of the interposition conduit after a Cabrol procedure.^8–10

In 2003, Gelsomino and colleagues⁶ published a retrospective analysis, over the longest time frame to that point, consisting of 45 patients (17 of whom had aortic dissection) who underwent the Cabrol operation between 1986 and 2002. The 30-day mortality rate for all patients was 20%; predictors included age, cross-clamp time, cardiopulmonary bypass time, aortic dissection, and emergency operation. It was concluded that the early and long-term complications of the Cabrol technique are such that it should be chosen only when a button technique is not suitable. Bachet and colleagues¹¹ conducted a retrospective review of 203 patients over a 30-year period and also concluded that the various techniques used in composite aortic graft replacement are not equivalent in terms of morbidity and mortality rates, and that the Cabrol technique should be reserved for cases in which the modified Bentall implantation is not feasible. In our analysis of the literature, precise comparisons were impossible because of variations in the indications for surgery. For example, the Cabrol technique was used mostly for aortic dissection and for reoperations.¹¹ No guidelines were provided for technique selection. It is generally agreed that the procedure of choice is the modified Bentall technique, when feasible, because it involves mobilization of the coronary buttons and avoids the use of interposition grafts.

When our patient needed reoperation for recurrent aortic insufficiency 3 weeks before her current presentation, the Cabrol procedure was chosen over bypass surgery. Scarring and adhesions from the previous Ross procedure would have made mobilization of the previously translocated coronary ostia difficult and also would have increased the risk of injury and undesirable tension to the coronary arteries. Furthermore, the RCA was nondominant, meaning that a suitable distal target for bypass grafting was not available. The Dacron conduit of the Cabrol procedure was attached to the best available target, that being the RCA ostium. As is usually recommended, the Dacron conduit was placed on the posterior right side of the aortic graft to avoid kinking and thrombosis of the left limb. Interposition grafts that are more than 8 mm long, or otherwise oversized, are predisposed to kinking and a low-flow state. However, the arterial limbs of the Dacron graft were correctly sized in our patient. Although the exact mechanism for occlusion is unknown in this case, a variety of factors might have contributed to the right arterial limb thrombosis. The computed tomographic scan revealed that the thrombosis of the right arterial limb was slightly distal to the anastomosis between the aortic composite graft and the Dacron conduit, without any evidence of kinking at this site. However, as previously mentioned, the native nondominant RCA in this patient was found to be extremely small at the time of surgery. Therefore, the small diameter of the RCA ostium, which was attached to the Dacron graft, might have caused a low-flow state that led to complete occlusion. Other possible causes of occlusion could have been misalignment, angulation, or intermittent kinking at the anastomosis between the Dacron conduit and the coronary ostia. Nonthrombotic stenosis of the Dacron conduit–coronary ostia junction is unlikely, because the patient presented within 3 weeks of the procedure.

The following unique properties of the RV enable its recovery after a myocardial infarction: it has one sixth as much muscle mass as the LV and, therefore, has a more favorable oxygen supply–demand ratio; it has improved oxygen delivery because of the biphasic nature of coronary blood flow during both systole and diastole; it has the same cardiac output as the LV but performs one fourth of the stroke work because the pulmonary vascular resistance is one tenth of the systemic resistance; it is constantly supplied by a rich system of venous oxygenated blood through the thebesian veins; and the RV has greater likelihood of acute collateral development to the RCA, attributed in part to lower coronary resistance that favors a left-to-right transcoronary pressure gradient.¹² The incidence of isolated RV infarction accounts for less than 3% of all cases of infarction and is typically associated with occlusion of a nondominant RCA. Early reperfusion in patients with acute RV infarction strikingly improves early outcomes. According to a landmark study,¹³ early reperfusion performed within 12.5 hours after the onset of infarction led to a faster recovery of RV function, with significant improvement in RV function observed at 1 hour and a striking reduction in early death. Patients with nonreperfused RVs had a high early mortality rate; however, all survivors had a complete recovery of RV function at 1 month.¹³

Our patient presented just 3 weeks after her Cabrol procedure and had developed grade III collateral vessels, leading us to suspect subacute occlusion of the RCA. Knowing that most patients who survive acute RV infarction eventually recover their RV function spontaneously and that late reperfusion is unlikely to be helpful, we opted to proceed with a conservative strategy for our patient. The Occluded Artery Trial study¹⁴ randomized 2,166 patients with total occlusion of the infarct-related artery more than 24 hours after myocardial infarction to either percutaneous coronary intervention (PCI) or medical therapy. Forty-nine percent of the patients in the study had a totally occluded RCA. The study did not show any benefit of PCI compared with medical therapy. Moreover, in a study conducted by Ketikoglou and colleagues¹⁵ evaluating the evolution of RV infarction, patients presenting with RV infarction had significant improvement in RV systolic function when managed with medical therapy alone. Another study¹⁶ evaluating 65 patients presenting with inferior myocardial infarction with RV involvement found that patients who did not receive early reperfusion had a remarkable improvement in RV function, comparable to improvements seen in the group that underwent early reperfusion. The time for complete RV recovery after acute infarction can range from a few days to a month, but can take up to 4 years.¹⁷ Some reports,^18,19 however, have identified a proportion of patients with RV infarction (up to 18%) who have persistent RV dysfunction over a longer period of time.

We chose medical management for our patient, anticipating an improvement in RV function. If our patient had presented acutely, primary coronary intervention would have been appropriate. Our patient's RV function had not improved at the 6-month follow-up visit. Such lack of improvement, although unusual, has been reported elsewhere.^18,19 When last seen, the patient remained asymptomatic with excellent functional capacity. This patient may belong in the small subset of patients who experience a more gradual improvement in RV function after myocardial infarction.

Conclusion

We report a case of subacute RV failure that occurred after a modified Cabrol operation and was related to thrombotic occlusion of the right limb of the Dacron interposition conduit. Conservative treatment is reasonable for subacute RV infarction, and the RV has several attributes that aid complete recovery over time. Although the Cabrol procedure is less often used in contemporary composite aortic graft replacement, it is imperative that interventional cardiologists be familiar with the anatomy that results from this and other surgical techniques in order to accurately interpret coronary angiograms and plan interventions when indicated. In addition, the use of computed tomography to delineate the altered anatomy is helpful in making a diagnosis and formulating a treatment plan.