Author + information
- Received March 10, 2003
- Revision received March 10, 2004
- Accepted March 16, 2004
- Published online July 7, 2004.
- Daniel A Caños, MPH*,
- Gary S Mintz, MD, FACC†,
- Chalak O Berzingi, MD*,
- Sue Apple, DNSc, RN*,
- Jun-ichi Kotani, MD*,
- Augusto D Pichard, MD, FACC*,
- Lowell F Satler, MD, FACC*,
- William O Suddath, MD, FACC*,
- Ron Waksman, MD, FACC*,
- Joseph Lindsay Jr, MD, FACC* and
- Neil J Weissman, MD, FACC*,* ()
- ↵*Reprint requests and correspondence:
Dr. Neil J. Weissman, Cardiovascular Research Institute, 110 Irving Street NW, Suite 4B-1, Washington, DC 20010, USA.
Objectives We sought to examine saphenous vein graft (SVG) lesions that fail within the first year after operation.
Background Saphenous vein grafts remain patent for approximately 10 years; however, up to 15% to 20% of SVGs become occluded within the first year.
Methods We studied 100 patients who underwent percutaneous coronary intervention (PCI) for early (<1 year post-implantation) SVG failure lesions and compared them with a diabetes- and hypercholesterolemia-matched cohort of late SVG failures (>1 year). Coronary angiography and intravascular ultrasound images were analyzed.
Results The majority of patients in both groups were males who presented with unstable angina; 36% were diabetic. Graft ages were 6.0 ± 2.9 months and 105.4 ± 50.8 months, respectively. The early SVG failure lesion location was more often ostial or proximal (62% vs. 42%, respectively). Early SVG failures were angiographically smaller than late failures (reference: 2.47 ± 0.86 mm vs. 3.26 ± 0.83 mm, p < 0.001) but had similar lesion lengths. Intravascular ultrasound showed that early failure lesions had smaller proximal and distal reference lumen areas (7.3 ± 6.8 mm2vs. 10.6 ± 3.8 mm2, p = 0.026) and greater reference plaque burden than late failures (52.3% vs. 36.1%, p < 0.001). After PCI, 20.6% of early and 30.6% of late failure lesions had creatine kinase-myocardial band (CK-MB) greater than twice normal.
Conclusions Early SVG failure is mostly proximal or ostial, lesions appear focal, and early SVGs appear smaller than late SVGs. Intravascular ultrasound shows significant reference segment plaque burden, suggesting more severe, diffuse SVG disease.
Coronary artery bypass graft surgery utilizing saphenous vein grafts (SVGs) is a proven technique for the treatment of intractable angina and, in select populations, improves long-term prognosis (1,2). The SVG conduits last ∼10 years (3–6). However, up to 15% fail within the first year, usually resulting in percutaneous coronary intervention (PCI) (7). The etiology of early (within 1 year) graft failure is poorly understood. To better understand this issue, we undertook a retrospective clinical, angiographic, and intravascular ultrasound (IVUS) comparison between patients who presented with early versus late (>1 year) graft failure who were then treated with PCI.
A retrospective analysis of the institution's catheterization database was performed to identify all patients who underwent PCI on one or more SVG lesions within one year of SVG implantation. Between January 1, 1991, and January 1, 2001, 3,775 SVG lesions were treated, of which 100 patients (121 lesions) had treatment within one year of graft implantation and had pre-interventional IVUS. Early failure patients were matched with late SVG failure patients on a 1:2 ratio based on diabetes and hypercholesterolemia status. Restenotic SVG lesions were excluded. The revascularization method was performed at the discretion of the operator. Baseline demographics and in-hospital complications were confirmed by independent hospital chart review.
Angiographic analysis was performed using a computer-assisted, automated edge-detection algorithm (CAAS II, Pie Medical Imaging BV, Maastricht, The Netherlands) independent of clinical and IVUS information. Using the outer diameter of the contrast-filled catheter as the calibration standard, the reference diameter and minimum lumen diameter (MLD) were measured before and after intervention from multiple projections. The results from the “worst” view were recorded, and percent diameter stenosis was calculated. Lesion lengths were measured in the view with the least amount of foreshortening. Degenerated SVGs were those with ectasia or lumen irregularity comprising ≥50% of the SVG shaft length. Ectasia was a lumen >20% larger than the reference segment. Ostial segments were defined <5 mm from the proximal anastomosis.
Intravascular ultrasound imaging and analysis
Imaging with IVUS was performed immediately before PCI, after the administration of intracoronary nitroglycerin, using a motorized pullback and a commercially available scanner (SciMed/Boston Scientific, Maple Grove, Minnesota). After the imaging catheter was advanced at least 10 mm beyond the lesion edge into the distal graft, the transducer was withdrawn at 0.5 mm/s to the aorto-ostial junction.
Lesion site and reference segment SVG wall cross-sectional area (CSA) and lumen CSA were obtained via computerized planimetry (Tape Measure, Indec, Mountain View, California), according to published guidelines (8). The SVG plaque CSA was defined as SVG wall CSA minus lumen CSA. Plaque burden was calculated as SVG plaque CSA divided by SVG wall CSA. Remodeling was calculated as lesion divided by mean reference SVG wall CSA.
Statistical analysis was performed using the SAS statistical package, version 8.0 (SAS Institute, Inc., Cary, North Carolina). Continuous variables are presented as the mean value ± SD and compared using the Student ttest. Discrete variables are presented as percentages and relative frequencies and compared using the chi-square statistic or Fisher exact test.
The clinical characteristics of the early failure (121 lesions) and late failure patients (289 lesions) are displayed in Table 1. The majority were men (70.0% early and 76.5% late, p = 0.22) with a similar age (64.8 ± 9.8 early vs. 66.5 ± 9.6 years, p = 0.17). Most had unstable angina (81.8% early and 73.1% late). The mean SVG age was 6.0 ± 2.9 months in the early group compared with 105.4 ± 50.8 months in the late group.
Procedural and angiographic characteristics
Percutaneous interventions differed in the early versus late groups, including stents (50.9% vs. 68.9%, p < 0.001), balloon alone (25.9% vs. 16.4%, p = 0.030), and laser or atherectomy (42.2% vs. 28.2%, p = 0.007) (Table 2). Early and late failure lesions had a high frequency of creatine kinase- myocardial band (CK-MB) greater than twice normal (20.6% and 30.6%, respectively).
Quantitative coronary angiography (QCA) analysis showed that vein grafts that failed early had a smaller pre-PCI reference diameter (2.47 ± 0.86 mm vs. 3.26 ± 0.83 mm, p < 0.001), smaller pre-PCI MLD (0.80 ± 0.64 vs. 1.08 ± 0.64, p < 0.001), and greater pre-diameter stenosis (71.6 ± 19.0% vs. 66.7 ± 17.7%, p = 0.017). Early graft failure was associated with lower pre-Thrombolysis In Myocardial Infarction (TIMI) flow rates (13.4% vs. 3.6% had TIMI flow grade 0 or 1, p < 0.001). There was no difference between the early and late failure lesion lengths or presence of thrombus (11.6% early vs. 12.6% late, p = 0.8). Grafts that failed late were more likely to be degenerated (43.1% vs. 29.5%, p = 0.015). Post-PCI QCA analysis showed that grafts that failed early had a smaller MLD (2.33 ± 0.96 vs. 2.91 ± 0.85, p < 0.001) and greater stenosis (16.3 ± 24.0% vs. 11.6 ± 18.1%, p = 0.037).
Intravascular ultrasound findings
Pre-intervention quantitative and qualitative results are shown in Table 3. The early failure reference segments had a greater plaque burden (distal: 52.3 ± 0.2% vs. 33.6 ± 0.1%, p < 0.001; proximal: 54.4 ± 0.2% vs. 38.3 ± 0.1%, p = 0.002; lesion: 81.7 ± 0.1% vs. 74.4 ± 0.1%, p = 0.009). There was also a significant difference between the plaque morphology in the early versus late failure groups. Vein grafts that failed early were more likely to be fibrotic (26.7% vs. 3.5%, p = 0.026), whereas grafts that failed late were more likely to be calcific (17.9% vs. 0%, p = 0.021).
Our study found that grafts that fail early are relatively small, less likely to be degenerated, mostly located in the ostial or proximal third portion of the graft, and had an increased frequency of TIMI flow grade 0/1. The diffuse disease (mean reference segment plaque burden >50%) of early failures was reflected in the high rate of procedural CK-MB release.
Previous studies have not focused on the angiographic findings in early graft failure. Savage et al. (9)performed angiographic analysis on 215 SVG lesions before PCI. The mean graft age in their study was 9.7 years, and 7% contained a definite thrombus. The majority of lesions were located in the proximal and mid portions of the grafts. Ahmed et al. (10)studied 340 lesions with a mean graft age of eight years, with similar findings. They reported that 9.6% of the lesions contained a thrombus and 36.3% were degenerated. Finally, Le May et al. (11)studied grafts that failed after an average of 9.9 years. The majority of lesions (75%) were found in the body of the graft, with thrombus in 21%.
This study is in concordance with these previous studies and found that early failure SVGs were angiographically smaller. Intravascular ultrasound analysis suggested that these “angiographically small” early failure grafts were actually diffusely diseased vessels without positive remodeling, resulting in small lumens. The lesion site remodeling index measured 0.97 ± 0.23 in the early failure group (no remodeling). The existence of remodeling within vein grafts is controversial. Nishioka et al. (12)examined 43 SVGs with a mean age of 11 ± 4 years. They found that SVGs do not undergo compensatory enlargement and suggested it as a potential factor affecting progression of SVG stenoses. However, other investigations reported significant positive remodeling in SVG lesions (13–15). These latter studies indicated that diseased SVGs can undergo remodeling responses that prevent lumen compromise, similar to native coronary arteries. Conversely, as in the current report, grafts that become diseased but that do not remodelwill develop small reference lumen areas; this appears to be especially true for early SVG failures.
There are several limitations to this study. First, this study is a retrospective analysis that is subject to its inherent limitations. Second, there may have been a selection bias in that only patients undergoing pre-intervention IVUS were studied. Third, a common presentation of early SVG failure is thrombotic occlusion, and patients with thrombotic occlusion usually do not undergo PCI. As such, the current study is a report of early SVG failure lesions undergoing PCI.
Early SVG failure lesions represent an aggressive, diffuse disease process throughout the graft.
- creatine kinase-myocardial band
- cross-sectional area
- intravascular ultrasound
- minimum lumen diameter
- percutaneous coronary intervention
- quantitative coronary angiography
- saphenous vein graft
- Thrombolysis In Myocardial Infarction
- Received March 10, 2003.
- Revision received March 10, 2004.
- Accepted March 16, 2004.
- American College of Cardiology Foundation
- Davis K.B,
- Chaitman B,
- Ryan T,
- Bittner V,
- Kennedy J.W
- Campeau L,
- Enjalbert M,
- Lesperance J,
- et al.
- Bourassa M.G
- Fitzgibbon G.M,
- Kafka H.P,
- Leach A.J,
- Keon W.J,
- Hooper D,
- Burton J.R
- Motwani J.G,
- Topol E.J
- Mintz G.S,
- Nissen S.E,
- Anderson W.D,
- et al.
- Ahmed J.M,
- Hong M.K,
- Mehran R,
- et al.