Percutaneous coronary intervention of totally occluded coronary venous bypass grafts: An exercise in futility?

Abstract

BACKGROUND

Percutaneous coronary intervention (PCI) of diseased saphenous vein grafts (SVG) continues to pose a clinical challenge. Current PCI guidelines give a class III recommendation against performing PCI on chronically occluded SVG. However, contemporary outcomes after SVG intervention have incrementally improved with distal protection devices, intracoronary vasodilators, drug-eluting stents, and prolonged dual antiplatelet therapy.

AIM

To reassess the procedural and long-term outcomes of PCI for totally occluded SVG with contemporary techniques.

METHODS

This was a retrospective observational study conducted at a single university hospital. The study population consisted of 35 consecutive patients undergoing PCI of totally occluded SVG. Post-procedure dual antiplatelet therapy was continued for a minimum of one year and aspirin was continued indefinitely. Clinical outcomes were assessed at a mean follow-up of 1221 ± 1038 d. The primary outcome was freedom from a major adverse cardiac event (MACE) defined as the occurrence of any of the following: death, myocardial infarction, stroke, repeat bypass surgery, repeat PCI, or graft reocclusion.

RESULTS

The study group included 29 men and 6 women with a mean age of 69 ± 12 years. Diabetes was present in 14 (40%) patients. All patients had Canadian Heart Classification class III or IV angina. Clinical presentation was an acute coronary syndrome in 34 (97%) patients. Mean SVG age was 12 ± 5 years. Estimated duration of occlusion was acute (< 24 h) in 34% of patients, subacute (> 24 h to 30 d) in 26%, and late (> 30 d) in 40%. PCI was initially successful in 29/35 SVG occlusions (83%). Total stent length was 52 ± 35 mm. Intraprocedural complications of distal embolization or no-reflow occurred in 6 (17%) patients. During longer term follow-up, MACE-free survival was only 30% at 3 years and 17% at 5 years.

CONCLUSION

PCI of totally occluded SVG can be performed with a high procedural success rate. However, its clinical utility remains limited by poor follow-up outcomes.

Key Words: Coronary artery bypass grafting, Coronary stents, Chronic total occlusion, Percutaneous coronary intervention, Restenosis, Saphenous vein grafts

Core Tip: Cardiovascular guidelines give a Class III recommendation against performing percutaneous coronary intervention (PCI) on chronically occluded saphenous vein grafts (SVG). Given contemporary advances in SVG intervention, the goal of this study was to reassess the outcomes of PCI for totally occluded SVG in 35 consecutive patients. PCI was initially successful in 29/35 (83%) SVG occlusions. However, at 3 years only 30% of patients survived without a major cardiac event. Although PCI of totally occluded SVG can be performed with a high procedural success rate, its clinical utility remains limited by poor follow-up outcomes.

INTRODUCTION

Percutaneous coronary intervention (PCI) of diseased saphenous vein grafts (SVG) continues to be a clinical challenge[1-4]. According to present cardiovascular guidelines, PCI of SVG with chronic total occlusions is discouraged and given a class III recommendation[5]. However, outcomes after SVG intervention have incrementally improved with distal protection devices, intracoronary vasodilators, drug-eluting stents, and prolonged dual antiplatelet therapy (DAPT). The goal of this study was to reassess the procedural and long-term outcomes of PCI for totally occluded SVG in contemporary practice.

MATERIALS AND METHODS

A review of medical records and coronary angiographic films was performed to identify patients with PCI performed on totally occluded SVG. This retrospective study was approved by the Institutional Review Board of the Thomas Jefferson University Hospital. All patients had a 100% occlusion of a venous bypass graft with Thrombolysis In Myocardial Infarction (TIMI) 0 or 1 flow. Patients with subtotal occlusions and TIMI 2 or 3 flow were excluded. PCI of occluded arterial bypass grafts were also excluded.

Detailed procedural information was collected on the type of anticoagulation, use of thrombectomy and distal protection device, the number of stents implanted, total stent length, and type of stent. Procedural success was defined as restored SVG patency with > TIMI 2 flow and < 20% residual stenosis. Post-discharge, all patients with successful PCI were treated with dual antiplatelet therapy (DAPT) for at least 1 year and aspirin indefinitely.

The primary outcome was freedom from a major adverse cardiac event (MACE) during a mean follow- up period of 1121 ± 1038 d. MACE was defined as the occurrence of any of the following: Death, myocardial infarction (MI), stroke, repeat bypass surgery, repeat PCI of the target vessel, or graft reocclusion. Event rates were determined by Kaplan-Meier analysis. Event free survival was also assessed as a function of the duration of SVG occlusion classified as acute (< 24 h), subacute (> 24 h to 30 days), or late (> 30 d). Duration of occlusion was estimated on the basis of the patient’s history of symptom onset. Subgroup analyses were performed based on the duration of occlusion, clinical presentation, and stent type. Fisher’s exact test was used to evaluate the relationship of duration of occlusion and clinical presentation with initial clinical success. Between group comparison of the survival curves was performed using the Mantel-Cox log rank test. P values < 0.05 were considered significant. Descriptive statistics were used to analyze baseline patient and bypass graft characteristics. Continuous variables are reported as mean ± SD.

RESULTS

Baseline characteristics of the patients are presented in Table 1. The study population consisted of 29 men and 6 women. Mean age was 69 ± 12 years. Diabetes was present in 40% of patients. Twenty patients (57%) had a history of prior MI. Mean left ventricular ejection fraction was 43% ± 12%. All patients had Canadian Heart Classification III or IV angina. Clinical presentation included ST-elevation MI (STEMI) in 6 patients (17%), non-STEMI in 14 patients (40%), unstable angina in 14 patients (40%), and stable angina in 1 patient (3%).

Table 1 Patient characteristics.

Characteristic	n = 35
Men	29 (83%)
Age (mean ± SD)	69 ± 12 yr
Hypertension	33 (94%)
Hyperlipidemia	25 (71%)
Diabetes	14 (40%)
Smoking	14 (40%)
Renal insufficiency	14 (40%)
History of CHF	6 (17%)
Prior MI	20 (57%)
Ejection fraction (mean ± SD)	43% ± 12%
Clinical presentation
STEMI	6 (17%)
NSTEMI	14 (40%)
Unstable angina	14 (40%)
Stable angina	1 (3%)
Canadian heart class
I, II	0
III	7 (20%)
IV	28 (80%)

CHF: Congestive heart failure; MI: Myocardial infarction; NSTEMI: Non-ST elevation myocardial infarction; SD: Standard deviation; STEMI: ST elevation myocardial infarction.

Bypass graft characteristics are shown in Table 2. Mean SVG age was 12.1 ± 5.1 years. The grafted native coronary artery was the left anterior descending in 9 patients (26%), left circumflex in 13 patients (37%), right coronary artery in 10 patients (29%), and multiple vessels in 3 patients (9%). Previous PCI had been performed in 10 (29%) of the SVG. The duration of graft occlusion was acute in 12 (34%), subacute in 9 (26%), and late in 14 (40%).

Table 2 Bypass graft characteristics.

Characteristic	n (%)
SVG age (mean ± SD)	12.1 ± 5.1 yr
Prior PCI of SVG	10 (29)
Grafted native vessel
LAD	9 (26)
LCx	13 (37)
RCA	10 (29)
Multiple	3 (9)
Duration of graft occlusion
Acute (< 24 h)	12 (34)
Subacute (> 24 h to 30 d)	9 (26)
Late (> 30 d)	14 (40)

LAD: Left anterior descending; LCx: Left circumflex; m: mean; PCI: Percutaneous coronary intervention; SD: standard deviation; SVG: Saphenous vein graft.

Procedural features and outcomes are presented in Table 3. Glycoprotein IIbIIIa inhibitors were used in 15 (43%) of cases. Thrombectomy was performed in 20 (57%). Distal protection devices were used in 9 (26%). The mean number of stents implanted was 2.1 ± 1.3 and total stent length was 52 ± 35 mm. Drug-eluting stents were used in the majority of patients.

Table 3 Procedural features and outcomes.

Feature	n (%)
Anticoagulation
Heparin only	12 (34)
Bivalirudin	8 (23)
Heparin + GP IIb IIIa inhibitor	15 (43)
Thrombectomy	20 (57)
Distal protection device	9 (26)
Number of stents (mean ± SD)	2.1 ± 1.3
Total stent length (mean ± SD)	52 ± 35 mm
Stent type
Drug-eluting	19 (55)
Bare metal	11 (31)
None	5 (14)
Procedural outcome
Success	29 (83)
Failed	6 (17)
Procedural complications
Distal embolization	1 (3)
No reflow	5 (15)
In-hospital MACE
Death	2 (5.7)
Myocardial infarction	1 (2.9)
Stroke	0
Repeat PCI or CABG	0
Any MACE	3 (8.6)

CABG: Coronary artery bypass grafting; GP: Glycoprotein; m: mean; MACE: Major adverse cardiac event; SD: Standard deviation.

Procedural success in restoring SVG patency was achieved in 29 of the 35 patients (83%). Of the 6 unsuccessful procedures, 2 were due to the failure to cross through the occluded graft with any guidewire; in the remaining 4 cases, successful guidewire passage to the distal native vessel was accomplished but graft patency could not be restored despite balloon angioplasty and thrombectomy. Procedural success was not correlated with either clinical presentation (P = 0.99) or duration of graft occlusion (P = 0.33). Procedural complications included distal embolization or no-reflow in 6 (17%) patients. There were no instances of coronary perforation. In-hospital MACE occurred in 3 (9%) patients including 2 deaths and one nonfatal MI.

Follow-up event free survival is shown in Figure 1. MACE free survival was 58% at 1 year, 30% at 3 years, and 17% at 5 years. Distribution of specific MACE events during follow up is shown in Table 4. Longer term outcome was not related to the initial duration of SVG occlusion (Figure 2) (P = 0.60) or clinical presentation (P = 0.87). There was no difference in MACE free survival for patients with drug-eluting stents compared to bare metal stents (P = 0.97).

Figure 1 Major adverse cardiac event free survival after percutaneous coronary intervention of totally occluded saphenous vein grafts.

Figure 2 Duration of occlusion and major adverse cardiac event free survival.

Table 4 Follow-up major adverse cardiac event.

MACE type	n (%)
Death	5 (14)
Stroke	1 (3)
MI	15 (43)
Repeat bypass surgery	0
Repeat PCI	6 (17)
Graft reocclusionAny MACE	7 (20) 28 (80)

MACE: Major adverse cardiac event; MI: Myocardial infarction; PCI: Percutaneous coronary intervention.

DISCUSSION

The management of patients who develop bypass graft disease after previous coronary artery bypass grafting (CABG) surgery continue to pose a clinical challenge. For post-CABG patients with unstable or medically refractory symptoms, all revascularization options entail heightened complexity and procedural risk. Redo CABG carries an increased risk of morbidity and mortality compared to the initial operation[6,7]. The outcomes of PCI in SVG have been significantly improved by coronary stenting[8]. Nevertheless, compared to native coronary arteries, PCI of SVG continues to be associated with higher risk of both procedural and long-term adverse events[9]. Accordingly, if revascularization is indicated in the presence of SVG disease, it is generally preferable to intervene on the native coronary artery if technically feasible. On the other hand, native coronary arteries long after bypass surgery often have complex, heavily calcified chronic total occlusions (CTO) which result in a significantly augmented risk of PCI failure and major complications[10-12].

Historically, before the introduction of distal protection devices and drug-eluting stents, PCI of totally occluded SVG has been associated with low initial procedural success, a high rate of complications, and frequent follow-up adverse events[13,14]. More contemporary PCI studies of SVG in the setting of acute thrombosis and STEMI while demonstrating improved procedural success still report poor longer term clinical outcomes[15,16]. A few studies have evaluated the outcome of PCI in chronically occluded SVG in the DES era[17-20]. An early small study of 11 patients with chronically occluded SVG suggested a high procedural success rate and favorable longer term outcomes after PCI[17]. Subsequent studies of chronically occluded SVG involving 22 to 34 patients reported PCI procedural success in 68% to 79% of cases with relatively high rates of MACE during medium term follow-up[18-20].

Citing low success rates, high complication rates, and poor long-term patency, ACC/AHA/SCAI PCI guidelines have a Class III: Harm recommendation against PCI for chronic SVG occlusions[5]. However, it is important to note that these guidelines were published over a decade ago. Accordingly, the unanswered question is whether the cumulative advances in SVG intervention have translated into improved outcome for totally occluded grafts. Distal protection devices have been shown to reduce periprocedural MACE, can be deployed with high success rate, and are given a Class I recommendation in PCI guidelines[5,21-23]. Intracoronary vasodilators have been shown to be effective in treating no-reflow which is a common complication during SVG PCI and may have benefits if given prophylactically[24-27]. Drug-eluting stents appear to reduce restenosis in SVG early after PCI although they exhibit a catch-up phenomenon during late follow-up[28-31]. Prolonged DAPT improves outcomes following PCI in high risk patient subsets and has been suggested as a possible strategy to improve the long-term outcome after PCI of SVG[4,32,33]. In addition to these advances in SVG intervention, significant improvements in CTO outcomes have been made through continued technical and procedural innovations[34,35].

Given the aforementioned advances, the goal of the current study was to reassess the procedural and long-term outcomes of occluded SVG PCI in the context of contemporary techniques. The initial procedural success in restoring SVG patency was achieved in 83% of patients. This procedural success rate is quite favorable compared to prior reports of totally occluded SVG. On the other hand, long-term clinical outcomes remained poor even though DAPT was continued for a minimum of one year and DES were used in most patients. MACE free survival was 58% at 1 year, 30% at 3 years, and only 17% at 5 years. There was no relation between the duration of SVG occlusion prior to PCI and long-term outcome. There was also no difference in outcome with drug-eluting compared to bare metal stents. The disappointing long-term results of DES in occluded SVG is consistent with the findings of other recent studies of DES for non-occluded SVG[29-31].

Limitations of the current study should be recognized. This study was retrospective and we do not have information into why PCI of the occluded vein graft was undertaken as opposed to other treatment options. Data is not available on the number of patients who were treated by other means such as PCI of the native vessel, redo CABG, or medical therapy alone. Patients in this study who underwent PCI likely represent a small select group with totally occluded SVG. These patients were treated in the drug-eluting stent era which now spans over the past decade and a half; the study population of this report represents less than 1% of patients undergoing PCI at our institution during this time. Accordingly, the outcomes of these patients could have been affected by the selection bias of the operators.

CONCLUSION

PCI of totally occluded SVG can be performed with a relatively high rate of procedural success. However, the vast majority of patients will experience a major clinical event within a few years following the procedure. Therefore, the clinical utility of PCI for totally occluded SVG continues to be limited due to poor long-term outcome.

ARTICLE HIGHLIGHTS

Research background

Percutaneous coronary intervention (PCI) of diseased saphenous vein grafts (SVG) continues to pose a clinical challenge. Given low success rates, high complication rates, and poor long term patency, current cardiovascular guidelines have a class III recommendation against PCI for chronically occluded SVG.

Research motivation

Contemporary outcomes of SVG intervention have incrementally improved with distal protection devices, intracoronary vasodilators, drug-eluting stents, and prolonged dual antiplatelet therapy. There is a paucity of studies on the outcome of PCI for totally occluded SVG using current techniques.

Research objectives

The goal of this study was to reassess the procedural and long term outcome of PCI for totally occluded SVG with contemporary techniques in the drug-eluting stent era.

Research methods

This was a retrospective observational study of 35 consecutive patients undergoing PCI of totally occluded SVG. The primary outcome was freedom from a major adverse cardiac event (MACE) defined as any of the following: Death, myocardial infarction, stroke, repeat revascularization of the target vessel, or graft reocclusion. Mean follow-up was 1221 ± 1038 d.

Research results

The study group included 29 men and 6 women aged 69 ± 12 years. Mean SVG age was 12 ± 5 years. PCI was initially successful in 29/35 (83%) SVG occlusions. During long term follow-up, MACE-free survival was 30% at 3 years and 17% at 5 years.

Research conclusions

PCI of totally occluded SVG can be performed with a relatively high rate of procedural success. However, the vast majority of patients have a major clinical event within a few years following the procedure. Thus, the clinical utility of PCI for totally occluded SVG continues to be limited by poor long term outcomes.

Research perspectives

Although PCI of totally occluded SVG can be often initially accomplished, the long term clinical outcome remains poor. Future research is required to achieve a more sustained clinical benefit through further innovations in stent design and adjunct pharmacology.

ACKNOWLEDGEMENTS

The authors thank Scott Keith PhD, a biostatistician at the Thomas Jefferson University, for performing a statistical review.