Outcomes of long left coronary endarterectomy in patients with diffuse coronary artery disease

Introduction

Primary coronary endarterectomy without coronary artery bypass grafting (CABG) was first introduced in 1957 by Bailey for the treatment of acute myocardial infarction^[1-3]. Subsequently, endarterectomy was combined with CABG alone, and later by vein patch and left internal mammary (LIMA) bypass of the left anterior descending artery (LAD)^[4-8]. Initial published reports showed a high incidence of perioperative mortality and ischemia^[9-13]. This led to reluctance in performing coronary endarterectomy. The adverse outcomes have resulted in some patients with severe and diffuse left-sided disease denied complete surgical revascularization. The use of dual antiplatelet therapy (aspirin and clopidogrel) has been shown to improve graft patency. Which led reduce major adverse cardiovascular events. The combination of both may improve survival following CABG^[14,15]. We aim to assess the early and late results of left coronary endarterectomy (LCAE). Which we used as an essential step in CABG. We also combined surgery with dual antiplatelet therapy post-op (aspirin and clopidogrel).

Methods

We included Patients with no option for percutaneous coronary intervention (PCI) and were rejected for conventional CABG. The rejection was because angiographically patients had atheroma involving < 80% of the length of the coronary artery. These patients underwent CABG with LCAE at a university teaching hospital by a single surgeon from February 1999 to September 2007. We identified the patients from a prospectively collected database. We compared this group with a propensity-matched cohort (to allow matching patients characteristics and risk profile) from the same period operated on by the same surgeon in the same hospital. Information regarding pre-operative status was gathered from the clinical database. Also, the clinical notes were reviewed retrospectively for these patients. Post-operative data were obtained from the clinical database, clinic letters and GP surgery data. If the clinical records were not adequate, the patient’s general practitioner was contacted by phone or e-mail. Table 1 summarizes the pre-operative patient demographics of the study patients. The mean follow-up of the patients was for five years (Range 0-10 years).

LCAE was performed on occluded or nearly occluded vessels with multiple and long distal stenoses. Patterns of atherosclerosis were identified pre-operatively, but the decision to perform endarterectomy was made intra-operatively. The surgeon decided on technical consideration in combination with angiographic features. Off-pump CABG cases were included in both the groups (CABG = 28/75, LCAE = 13/87). The methods of myocardial protection techniques were comparable in both groups. Core temperatures were maintained between 30 degrees to 32 degrees during cardiopulmonary bypass.

All coronary endarterectomies were performed manually. A Watson-Cheyne dissector was used to develop a plane between tunica media and the core of the atheromatous plaque. The site for arteriotomy was chosen, and the decision to do endarterectomy was made if the artery was almost entirely or wholly occluded, or if the vessel wall contained heavily calcified plaques. Arteriotomy was extended as required (2.5-9 cm) and Watson-Cheyne dissector used to develop the plane, and gentle graduated traction was used to tease off the atheromatous core from the distal end of the vessel [Figure 1]. If the distal end of the atheroma was not tapered, then the arteriotomy was extended until a satisfactory result was obtained, but no attempt was made to extract the atheromatous core from the individual branches of the artery. A vein patch was used in all the cases. The conduit was anastomosed end to side to the vein patch in all the cases.

Figure 1. endarterectomy

As part of our routine clinical practice, all patients undergoing CABG receive dual antiplatelet therapy (DAPT) post-operatively. Aspirin 300mg is given at 4-6 h post-operatively and 150 mg/day after that which was recommended to continue lifelong. In addition to aspirin, patients received clopidogrel 300 mg at 4-6 h post-operatively and 75 mg/day after that, which was recommended to continue for 1-year. In both groups, DAPT was administered if the bleeding was < 150 mL/h for the first 4 h postoperatively.

Results

The patient demographics were similar in both the groups as presented in Table 1, except for the incidence of Diabetes Mellitus in the CABG group and increased elective surgery numbers in the LCAE group.

Operative data are presented in Table 2. The number of grafts was similar in both groups. The additional time to perform the grafts, including the endarterectomy, resulted in longer cross-clamp time (60.5 min vs. 45.5 min) and longer bypass times (77.2 min vs. 62 min). LIMA use was slightly higher in the CABG group (94% vs. 89%). Of the 87 with diffuse coronary atheroma, the lesion distribution was, 58 patients had isolated LAD endarterectomy, three patients had Cx endarterectomy, and 26 patients had both Cx and LAD endarterectomy. The number of arterial and vein grafts was similar in both groups of patients. All patients undergoing endarterectomy received a vein patch.

Actuarial survival and follow up

There was no significant difference in actuarial survival between the two groups, as shown in Figure 1. After a mean follow-up of 5 years, there were four deaths in the LCAE group when compared to 3 deaths in the CABG group. The deaths in the LCAE group occurred after three years in 2 patients and four years in the other two patients. The three deaths in the CABG only group occurred after 2 years, 4 years and five years respectively. The one-year survival was 98.9%, and the 5-year survival was 95.5% in the LCAE group [Figure 2].

Figure 2. Survival in patients undergoing CABG only (group 0) or CABG combined with coronary endarterectomy (group 1)

The freedom from angina was similar in both groups at the end of one year, as shown in Figure 3. There was no statistical difference in the incidence of angina in the post-op period at the end of 10 years follow up.

Figure 3. Freedom from angina in patients undergoing CABG only (group 0) or CABG combined with coronary endarterectomy (group 1)

Discussion

Despite the excellent results obtained with surgical revascularization, advances in the field of PCI coupled with its less invasive nature have made it the most common interventional treatment strategy for coronary artery disease^[16,17]. However, patients with complex and diffuse disease may be challenging to entirely surgically revascularize. Complete revascularisation and in particular arterial grafting to the left coronary system has a significant impact on long term survival of patients^[18,19]. It is, therefore, essential for surgeons to have an effective strategy for grafting patients with diffuse severe obstructive CAD to achieve complete revascularization. In the presence of diffuse disease, open endarterectomy is a surgical adjunct, which can produce good clinical results in these otherwise inoperable patients.

Various methods of coronary endarterectomy have been described. Initially, a closed traction method with primary closure of the arteriotomy was practised^[1]. Over time the procedure evolved to include patch closure of the arteriotomy combined with LIMA or saphenous vein bypass grafting^[6,7,20,21]. In this study, all endarterectomies were performed under direct vision, exposing the whole arterial lumen and side branches and arteriotomy closure with vein patch with subsequent end to side grafting was done in all patients. This has been an effective method, with only one patient requiring further revascularisation in the immediate post-op period due to poor cardiac output. All patients in the LCAE group had diffuse severe coronary artery disease. The ability to achieve complete revascularization and where possible to use the left internal mammary artery (89% of patients in this study) to bypass the left coronary system has been shown to improve outcomes in patients with CAD. In the LCAE group, complete revascularization likely would not have been possible without the use of adjuvant coronary endarterectomy. Furthermore, in native coronary arteries with multiple sequential obstructive lesions, coronary endarterectomy coupled with vein patching and grafting allows the target vessel to be grafted with a single rather than numerous coronary anastomoses.

The higher incidence of perioperative myocardial ischemia in earlier studies involving technically excellent surgeons was not seen in this study^[5,9,15]. Two critical factors have likely contributed to improved outcomes in this compared to historical studies: firstly advances in techniques of myocardial protection, in particular, the use of retrograde cardioplegia and secondly the use of aggressive post-op anticoagulation with dual antiplatelet therapy. All endarterectomy procedures were performed by an open method, and retrograde cardioplegia was used for flushing out all the debris immediately. During beating heart surgery, the endarterectomy was performed after application of a sling proximally, and retrograde flow used to flush the debris before vein patching and anastomosis. Early and midterm graft failure may be due to thrombosis of the graft itself or of the native vessel. Historically LCE was associated with an increased incidence of graft failure and perioperative myocardial infarction^[22,23]. This is likely due to exposure of the prothrombotic and platelet aggregating promoting components of the vessel wall following endarterectomy. Dual antiplatelet therapy, possibly by reducing early thrombotic events and platelet clumping, has been shown to improve graft patency in off-pump surgery, following acute coronary syndromes, when multiple vein grafts are used and in native vessel disease vessels with reduced runoff. This has translated into improved clinical outcomes in these groups. Besides, aggressive antiplatelet administration immediately post-op with aspirin and clopidogrel has been proven to be safe and effective after coronary artery surgery^[14,24-26]. We commenced all our patients on clopidogrel, and aspirin immediately following surgery and continued on Clopidogrel for 1-year post-op along with lifelong aspirin.

Ideally, angiographic confirmation of graft patency would have been preferred, but the freedom from angina in our study (91.8% at 1year and 79.4% at 10 years) along with a low incidence of perioperative ischaemic events indirectly shows that this strategy is a valuable additional benefit for these patients.

Our results, coupled with those from other centres, suggest that coronary endarterectomy can be accomplished safely and acceptably when applied in a particular way^[27,28]. In this study, we have shown comparable results of endarterectomy with the CABG only group when comparing long term survival and freedom from angina. Although numerically morbidity in the endarterectomy group was slightly higher than control, it did not have a lasting effect on long term outcome. We propose the use of left-sided endarterectomy as a safe adjutant in dealing with diffusely diseased coronary vessel disease, especially when long term prognosis is essential.