Surgery for Chronic Total Occlusion of the Left Main Stem: A 10-Year Experience

Introduction

Chronic total occlusion of the left main stem (LMS) is rare in patients with coronary atherosclerotic disease, with a prevalence of 0.04% to 0.4%. ^{1 –3} Coronary artery bypass grafting (CABG) is the treatment of choice for LMS disease, especially in cases of total occlusion. ^{4 –8} Total LMS occlusion is characterized by no antegrade blood flow to the left anterior descending (LAD) and left circumflex (LCx) arteries. ^9,10 Total occlusion can be either acute or chronic. ¹¹ Acute total occlusion follows an event such as myocardial infarction (MI) or a sudden change in the angina pattern, with electrocardiographic (ECG) changes consistent with the site of occlusion. Chronic occlusion is asymptomatic, with a duration of >3 months. ¹¹ LMS stenosis is usually associated with significant disease elsewhere in the coronary tree, which may lead to symptoms before complete obstruction occurs. ^1,2 LMS stenosis occurs as an isolated lesion in 6%−9% of patients undergoing coronary angiography, whereas multivessel disease is found in 70%−80%. Revascularization for LMS stenosis should be accomplished by CABG rather than stenting. ⁷ Percutaneous coronary interventions (PCI) have been performed in cases of acute total occlusion of the LMS. ^9,11,12 We describe our 10-year experience in patients with chronic LMS total occlusion.

Patients and Methods

This retrospective study was conducted in patients who were treated between June 1998 and June 2008. Of 67,082 coronary angiographies performed during this period, 17 (0.025%) detected chronic total occlusion of the LMS. The characteristics of these 17 patients are given in Table 1. None had a history of transient ischemic attack or cerebrovascular accident; however, carotid Doppler examination revealed 3 patients had 50% stenoses of the right common carotid artery, 1 had 50% stenosis in the left internal carotid artery, and 2 had mild intimal thickening of both carotids. In all patients, the LAD filled retrogradely from the right coronary artery (RCA; Figure 1A,B). In 15 patients, the target vessel was of good quality, 2 had diffuse LAD disease. The LCx filled retrogradely from the RCA in 12 patients, and 12 had >70% stenosis of the RCA and its branches.

OPEN IN VIEWER

Table 1.

Baseline characteristics of 17 patients

Variable	No. of Patients
Age (years)	55.3 ± 6.2
Males	14 (82.4%)
Females	3 (17.6%)
Diabetes mellitus	8 (47%)
Hypertension	6 (35.3%)
Smoking	6 (35.3%)
Dyslipidemia	5 (29.4%)
Carotid artery disease	4 (23.5%)
Stable angina	8 (47.1%)
Unstable angina	6 (35.3%)
Dyspnea	3 (17.6%)

OPEN IN VIEWER

Figure 1.

Angiograms showing (A) chronic total occlusion of the left main stem with no antegrade flow in the left anterior descending artery (LAD) and left circumflex (LCx), and (B) collaterals to the LAD and LCx from a dominant right coronary artery.

All patients underwent CABG on cardiopulmonary bypass with moderate systemic hypothermia at 28°C and a membrane oxygenator. Intermittent crossclamping and fibrillatory arrest was used in 10 patients, 7 had antegrade blood cardioplegia infused into the aortic root and through each vein graft on completion of the distal anastomosis. After a median sternotomy, the left internal mammary artery (LIMA) was harvested and wrapped in gauze soaked in papaverine (papaverine hydrochloride 50 mg in 2 mL; Galen Ilac San, Istanbul, Turkey) to prevent spasm. Saphenous vein was harvested at the same time. The patient was heparinized and cannulated according to standard techniques. In 15 patients, the LIMA was anastomosed to the LAD. Totally occluded vessels in the left system were grafted first (obtuse marginal and diagonals) to reperfuse the left ventricle (LV). Subsequently, the RCA and its branches were grafted, and the LIMA was anastomosed to the LAD. Patients were treated under fibrillation by the same strategy; on completion of each distal anastomosis, the crossclamp was removed, the heart was cardioverted, and proximal anastomosis was performed on the ascending aorta. In 2 patients with severe diffuse LAD disease, endarterectomy was undertaken using a limited incision, the atherosclerotic plaque was dissected proximally and distally and extracted by traction before anastomosing the LIMA directly to the incision site. After completion of anastomoses and rewarming, the patient was weaned off bypass in the usual manner and transferred to the intensive care unit. Aspirin 75 mg was started on postoperative day 1, as well as heparin (Leo Pharmaceutical Products) 5,000 IU subcutaneously 8 hourly. ¹⁴ Patients who had undergone endarterectomy received aspirin 75 mg and clopidogrel 150 mg at 2 h postoperatively, provided there was no excessive drainage (< 100 mL · h⁻¹ for 2 h), followed by aspirin 75 mg and clopidogrel 75 mg daily along with heparin 5,000 IU subcutaneously 8 hourly.

All patients were followed up daily until discharge, and weekly for 30 days postoperatively. All complications were recorded. Hospital mortality was defined as death occurring within 30 days postoperatively. Perioperative MI was defined as persistent ECG changes such as new Q waves, loss of R-wave progression, new intraventricular conduction defects, elevation of creatinine kinase-MB or cardiac troponin, or new echocardiographic evidence of a wall motion abnormality occurring within the first 48 h postoperatively. ¹³ Other hospital complications included postoperative atrial fibrillation, renal failure, cerebrovascular accident, low output syndrome (inotropic support >24 h or new placement of an intraaortic balloon pump to wean off cardiopulmonary bypass or maintain systolic blood pressure >100 mm Hg), and wound infection. ¹³ No postoperative angiography for evaluation of graft patency was offered to these patients.

Statistical analysis was performed using SPSS version 14.0 software for Windows (SPSS, Inc., Chicago, IL, USA). Categorical variables are expressed as frequencies and percentages, while continuous variables are expressed as mean and standard deviation. Hospital mortality and complications are expressed as frequencies and percentages.

Results

The mean aortic crossclamp time was 32.3 ± 9.9 min, and mean cardiopulmonary bypass time was 69 ± 17.1 min. The LIMA was anastomosed to LAD in all except 2 patients who received saphenous vein grafts: one due to advanced age and chronic obstructive pulmonary disease; the other became unstable at sternotomy so cardiopulmonary bypass was instituted and a vein graft was immediately applied to the LAD. The angiographic findings and distribution of grafts are shown in Table 2. Three (17.6%) patients required intraaortic balloon counterpulsation: it was instituted preoperatively due to unstable hemodynamics in 2, and immediately after surgery due to difficulty in weaning from bypass in the patient who underwent LAD endarterectomy. The mean intensive care unit stay was 2.1 ± 1.2 days, and mean hospital stay was 7.1 ± 1.5 days. Two patients had superficial sternal wound infections, 2 had transient atrial fibrillation during the hospital stay, and the patient who underwent endarterectomy had ECG changes consistent with perioperative MI in the LAD territory. One patient with normal carotid arteries had a transient ischemic attack with spontaneous recovery and no residual neurological deficit. There was no operative or hospital death. All patients were discharged on antiplatelet and statin therapy, with no recurrence of symptoms at 30 days, 3, and 6 months postoperatively. None needed any further intervention, and they were referred back to their cardiologists 6 months after the operation.

OPEN IN VIEWER

Table 2.

Coronary angiographic findings and grafts applied in 17 patients

Disease/Grafts	No. of Patients
Left main stem disease	17 (100%)
RCA disease	12 (70.6%)
Ejection fraction	50% ± 11%
Grafts
LIMA to LAD	15 (88.2%)
SVG to LAD	2 (11.7%)
SVG to diagonal	9 (52.9%)
SVG to OM branches	16 (94.1%)
SVG to RCA and branches	12 (70.6%)

LAD = left anterior descending coronary artery, LIMA = left internal mammary artery, OM = obtuse marginal, RCA = right coronary artery, SVG = saphenous vein graft.

Discussion

Chronic total LMS occlusion is a rare presentation of atherosclerotic disease. CABG is the treatment of choice because crossing the occlusion with a wire has a limited success rate of 40%−81% and a higher incidence of restenosis. ³ We observed a 0.025% incidence of chronic total LMS occlusion, which is marginally lower than that in previous series. Initially, we performed angiography in hemodynamically stable patients, but now we undertake angiography and primary PCI of hemodynamically unstable patients in our institute. This could be the reason for no cases of acute total occlusion, as well as the lower incidence of chronic total occlusion in our study.

Ward and colleagues ² described 5 patients with a long history of angina and total LMS occlusion, and highlighted the importance of collateral formation as well as the superiority of CABG over medical therapy. They also observed postoperative improvement in LV function, which we found in 3 patients in our series. Yip and colleagues ¹⁴ carried out primary PCI in 8 patients with total and 10 with subtotal LMS occlusion. There was a 72.2% procedural success rate; 4 patients underwent CABG, 6 died in hospital, and 2 died after discharge. The 10 survivors had more intercoronary collaterals, a dominant RCA, and incomplete LMS occlusion. Acute total LMS occlusion generally presents as pulmonary edema, cardiogenic shock, or sudden death. In our opinion, CABG is the preferred treatment for chronic total LMS occlusion, whereas in acute LMS occlusion, it is one of the therapeutic options along with primary angioplasty which is preferred if the patient has good collaterals and incomplete occlusion. Bare metal stenting for LMS occlusion has a 1-year mortality rate of 3%−28% and a restenosis rate of 20%. ¹⁵ The restenosis rate of drug-eluting stents is 10%. In view of the high restenosis and mortality rates with PCI in this condition, it is suggested that PCI should only be used as a last resort in patients deemed unsuitable for CABG because of prohibitive comorbidities. ¹⁵

A recent trial studied 1,800 patients with left main or 3-vessel coronary artery disease randomly assigned to undergo CABG or PCI with drug-eluting stents. ¹⁶ The CABG group had more severe lesions, including total occlusion, bifurcation lesions, lesions > 20 mm in length, and heavy calcification, than the PCI group; however, patients treated by PCI were more likely to reach the primary end-points of death from any cause, stroke, MI, or repeat revascularization within 12 months. Comparing secondary end-points, the 2 treatment groups had similar rates of death from any cause, stroke, or MI. ¹⁶ Patients undergoing PCI were more likely to require repeat revascularization but less likely to suffer stroke than those undergoing CABG. Hence it was concluded that CABG was preferred for patients with 3-vessel or left main coronary artery disease.

Ipek and colleagues ⁵ studied 7 patients with chronic total LMS occlusion with reference to myocardial preservation options, 3 also had severe RCA lesions, prior MI, and significantly depressed LV function. Five patients had on-pump CABG and 2 had off-pump grafting; all received alternating antegrade/retrograde cardioplegia for myocardial preservation. Postoperative low cardiac output occurred in only one patient who had off-pump surgery, and there was no operative or early mortality. In our study, 12 patients had >70% stenosis in the RCA and its branches, which were successfully grafted with saphenous vein. Intermittent crossclamping and fibrillatory arrest was opted for in 10 patients, while 7 had blood cardioplegia; there was no difference in postoperative events in our 2 groups. We agree with Ipek and colleagues ⁵ on the strategy of revascularization of a totally occluded vessel prior to that giving off collaterals to it, and we also feel that antegrade blood cardioplegia alone is an effective and safe means of myocardial preservation. Only 3 of our 12 patients with RCA disease had depressed LV function, which is lower than in previous reports, possibly because the disease was in distal RCA branches. As chronic total LMS occlusion is a progressive lesion, and patients have a calcified proximal LMS, we do not consider that they are appropriate candidates for surgical angioplasty or reconstruction of the left coronary ostium and left main stem, although it is a good option for isolated ostial LMS stenosis with a normal distal left system. ¹⁷ We agree with Ipek and colleagues ⁵ that priority should be given to retrograde filling vessels of the left system, followed by grafts to the obtuse marginal and diagonal branches, and finally the RCA if diseased. Sugishita and colleagues ⁴ studied 3 patients with chronic total LMS occlusion (2 had atherosclerotic disease, 1 had Kawasaki disease) who underwent successful CABG; all had a dominant RCA with good collaterals to the left system, which are necessary for patients to survive chronic total LMS occlusion with preserved LV function; 2 were followed up for more 10 years after surgery. In our study, all patients had a dominant RCA, but 12 had RCA stenosis requiring bypass grafting. Topaz and colleagues ¹⁸ emphasized the importance of collateral vessels in preserving LV function in 13 cases of chronic total LMS occlusion. We agree as 3 of our 4 patients with proximal RCA disease and poor collaterals had depressed LV function, but we have no explanation for preserved LV function the 4^th patient. Shen and colleagues ¹⁹ found that RCA disease was associated with decreased survival after CABG in 23 patients with chronic total LMS occlusion, whereas we had no deaths among the 12 patients with RCA disease.

We concluded that CABG is a good treatment option for chronic total LMS occlusion, and either intermittent crossclamping and fibrillation or antegrade blood cardioplegia can be used safely.