Axillo-axillary bypass with polytetrafluoroethylene graft for coronary subclavian steal syndrome

Introduction

The internal mammary artery (IMA) is used with increasing frequency for myocardial revascularization. It has significant improvements in cumulative survival and a reduction in the risk of late myocardial infarction, compared with saphenous vein graft. ^1,2 Most tertiary centers currently utilize a validated carotid artery duplex criteria to evaluate carotid and subclavian artery occlusive disease; therefore, evaluation may include, but not be limited to, preoperative coronary angiography which does not always provide adequate visualization of proximal subclavian arteries. If a proximal subclavian artery occlusive disease exists or develops in a patient who has myocardial revascularization with the IMA, graft malfunction can occur resulting in myocardial ischemia. This phenomenon is known as coronary subclavian steal syndrome. In this condition, subclavian revascularization through endovascular or surgical approaches is indicated.

Endovascular therapy such as percutaneous transluminal angioplasty (PTA) and/or stenting of proximal subclavian artery stenotic lesions provide a less invasive option and have an acceptable result. ³ However, endovascular therapy is not always possible in the presence of proximal subclavian artery occlusion.

In this situation, surgical techniques can be performed to correct the steal and relieve recurrent myocardial ischemia without the risk of redo coronary artery bypass grafting (CABG). During the last few decades, a variety of extra thoracic surgical techniques have been advocated to treat subclavian lesions and to avoid transthoracic approaches because of their higher morbidity and mortality rates. Axillo-axillary bypass (AAB) represents one of these extrathoracic surgical techniques and has been reported to correct the vertebral subclavian steal syndrome. In the present study, we report our results of AAB with a polytetrafluoroethylene (PTFE) graft for the treatment of patients with coronary subclavian steal syndrome due to proximal subclavian artery occlusion.

Methods

The clinical protocol for this study was approved by the Institutional Review Board and Ethics Committee of our hospital and informed consent was obtained from all patients.

From a prospectively maintained database, we retrospectively selected all consecutive patients who underwent AAB with a PTFE graft for coronary subclavian steal syndrome between February 2003 and November 2010 at our hospital. All patients who had other types of extra-anatomical bypasses such as carotid-subclavian bypass (CSB) or carotid-subclavian transposition were excluded from this study. All patients had undergone cardiac catheterization by a cardiologist prior to referral to our department. Digital subtraction arteriography of the aortic arch with selective views of the carotid and subclavian arteries was performed in patients who did not obtain selective views of the left common carotid and subclavian arteries at the time of cardiac catheterization. A patent in situ IMA with reversed flow was identified in all cases. The preoperative functional cardiac status was classified according to New York Heart Association guidelines. ⁴

In all patients, endovascular therapy such as PTA and/or stenting of proximal subclavian artery was attempted first. Briefly, all endovascular procedures were performed under local anesthesia. Arterial access was obtained via the common femoral arteries. Unfractionated heparin (50–70 U/kg) was given after arterial access to maintain an activated clotting time ≥250 seconds. A 6 F or 7 F sheath was used. Fluoroscopic guidance and road mapping were used in all cases. The lesions were crossed using a 0.035-inch hydrophilic wire and predilatation with an angioplasty balloon was not performed routinely. Balloon expandable stent size was selected on the basis of size of the adjacent normal subclavian artery. Post-stenting balloon dilatation with a 1:1 ratio was performed when there was residual stenosis. Completion angiography was performed after stent deployment and residual stenosis less than 30% was deemed as a result of technical success. When endovascular therapy was unsuccessful, patients underwent surgical bypass. All endovascular technical failures were due to the inability to cross the occluded proximal subclavian artery lesions with a guidewire.

All AABs were performed under general anesthesia. The axillary artery was exposed bilaterally by transverse infraclavicular incisions. The incision extended from the medial third of the clavicle towards the clavipectoral groove, and a muscle split technique of the pectoralis major was performed; then the clavipectoral fascia was incised and the pectoralis minor was sectioned in six cases and retracted in five to maximize exposure and to prevent kinking of the graft. A subcutaneous tunnel was created between the two arteries, and an end-to-side anastomosis with a prosthetic graft was performed bilaterally after systemic heparinization. One standard size (diameter, 8 mm) of PTFE graft was uniformly used in this series. To minimize myocardial events, first, systemic heparinization was achieved before left axillary artery clamping; second, we tried clamping the left axillary artery for 10 minutes and simultaneously looked at the monitor to see if there was any question of electrocardiographic changes. If there were no electrocardiographic changes suggestive of myocardial ischemia, then we clamped the axillary artery again and performed the anastomosis. However, if there was any question of electrocardiographic changes after placement of the proximal left axillary artery clamp, further distal exposure of the axillary artery was obtained to allow placement of clamps further distally and simultaneously give intravenous nitroglycerin. In addition, when the clamps were released, arterial flow was restored to the upper extremity first and then to the proximal axillary artery to reduce the chance of embolization into the coronary and vertebral arteries. All patients had perioperative serum cardiac isoenzyme and troponin level tests.

Aspirin (100 mg) was taken indefinitely after treatment. Follow-up information was obtained from direct patient re-examination and non-invasive vascular laboratory data. Graft patency was determined by the presence or absence of peripheral arterial pulses and confirmed by duplex ultrasound which was performed at one, six and 12 months for the first year and annually thereafter. The cumulative patency and survival rate were calculated with the life table method.

Results

The technical success rate of AAB was 100%. No perioperative mortality or stroke occurred. There were no intraoperative electrocardiographic changes suggestive of myocardial ischemia after placement of the left axillary artery clamp. No abnormalities were noted in cardiac isoenzyme and troponin level tests in all patients. There were no cardiac complications postoperatively. The overall perioperative complication rate was 9.1%, with one patient experiencing a minor wound complication. Arm pressure gradient decreased markedly after surgery. The mean post-treatment pressure gradient was 4.3 mmHg (range: −8 to 12 mmHg).

Over a mean follow-up of 36 months (range: 6–81 months), all bypass grafts have remained patent. The patency rates of AAB at one, three and five years were 100, 100 and 100%, respectively. No patient developed recurrent symptoms of myocardial ischemia. Symptoms of vertebrobasilar ischemia and arm claudication disappeared in two patients who had concomitant vertebral subclavian steal syndrome. One patient died from stroke at 31 months. The patient suffered hemorrhagic stroke during catheter-based intra-arterial thrombolysis for acute lower-limb ischemia. The computed tomography scan showed intracerebral hemorrhage with bleeding in the third, fourth and both lateral ventricles. The patient finally died of brain herniation. Nine patients were evaluated after AAB with nuclear imaging technology (dipyridamole-thallium scintigraphy) and were found to have improvement in reversible defects noted preoperatively in the anterior and lateral myocardial regions supplied by the left anterior descending coronary artery.

Discussion

Coronary subclavian steal was first reported by Harjola and Valle ⁵ in 1974 in an asymptomatic patient. Since this report, other investigators have reported numerous cases of symptomatic recurrence of coronary ischemia secondary to coronary subclavian steal syndrome following in situ IMA bypass. ^6–9

Several extrathoracic surgical techniques have been advocated to treat subclavian artery lesions for which endovascular approaches are unsuccessful. These include subclavian to carotid transposition, CSB and AAB. Subclavian to carotid transposition involves only one anastomosis and does not need graft insertion. This procedure has excellent long-term durability, ¹⁰ but it is not possible in all cases and may potentially worsen coronary ischemia during interruption of vertebral and internal mammary blood flow, which makes it a less attractive treatment option in this population.

CSB has been reported to treat coronary subclavian steal syndrome. ⁷ CSB was introduced by Diethrich et al. ¹¹ in 1967 and has excellent long-term durability. Studies of CSB with PTFE have patency rates as high as 95% at 10 years. ^12–14 In a previous study, however, 5- and 10-year patency rates were 66.0 and 40.8%, respectively, for CSB in patients with an associated ipsilateral carotid lesion, because recurrence of carotid stenosis or carotid lesion progression may cause failure of CSB. ¹⁵ In the presence of an ipsilateral carotid artery lesion, a carotid endarterectomy (CEA) may be necessary before or may be done at the time of CSB. However, concomitant ipsilateral CEA significantly increases risk of perioperative stroke. ¹⁶ In our center, we usually perform CSB in patients without concomitant ipsilateral carotid artery disease.

In the present study, all patients had concomitant atherosclerotic occlusive lesions of the ipsilateral common carotid artery and/or internal carotid artery. AAB was first described by Myers et al. ¹⁷ in 1971 at the Marshfield Clinic. As AAB avoids manipulation of a diseased carotid artery, we preferred this procedure to treat subclavian artery lesions in patients with concomitant carotid artery disease, especially ipsilateral carotid artery disease. Graft patency can be easily determined by duplex ultrasound and it is reasonable to expect a 10-year primary patency rate of 90%. ^18–20 AAB has raised several concerns in the literature for its length and subcutaneous course, which might cause thrombosis, infection and skin erosion. In our study, no failed grafts were found during follow-up. Theoretical concerns about the graft infection and skin erosion have not been encountered in this study.

Endovascular treatment such as PTA or stenting is an attractive alternative to bypass grafting and may have several advantages over surgical therapy, including its minimally invasive nature, avoidance of general anesthesia, greater patient acceptance and shorter length of hospital stay. It has a high technical success rate and excellent long-term outcome for patients with localized stenosis of the proximal subclavian artery. ^21,22 The technical success rate was relatively low for occluded proximal subclavian artery, ranging from 50 to 87%. In addition, recurrent restenosis may develop in patients with long-segment lesion during follow-up. However, most recurrent restenosis can be treated with endovascular therapy. This suggests that endovascular management of proximal subclavian artery stenosis may be an excellent alternative to surgical bypass if an adequate surveillance program can be established, enabling early detection and treatment of any significant recurrent restenosis, with comparable long-term patency rates. Today, we routinely select endovascular therapy first for occluded/stenotic proximal subclavian artery. When endovascular therapy is unsuccessful initially or fails due to recurrent restenosis or reocclusion during follow-up, surgical revascularization such as CSB or AAB with a PTFE graft can provide an effective and durable treatment option.

Our study has some limitations. First, it is a retrospective study with a very small sample size and limited follow-up time; therefore, this study has low power and potential type 1 error. Second, all patients undergoing AAB in this study have concomitant ipsilateral carotid artery disease, which can be partially based on selection bias by the surgeons. Lastly, compared with CSB, there are some limitations related to the technique of AAB itself such as two incisions, longer surgery and difficult exposure if future CABG has to be considered. Prospective randomized studies are needed to compare the current different therapy modalities for coronary subclavian steal syndrome.

The results of the present study showed that AAB with a PTFE graft provides an effective and durable treatment option for coronary subclavian steal syndrome when attempted endovascular therapy of occluded proximal subclavian artery is unsuccessful.

Declarations

Conflicts of interest: None.

Funding: None.

Ethical approval: Yes.

Acknowledgments: None.