A Multicenter, Randomized Trial of Coronary Angioplasty Versus Directional Atherectomy for Patients With Saphenous Vein Bypass Graft Lesions
Background Directional coronary atherectomy and percutaneous transluminal coronary angioplasty have both been used in symptomatic patients with coronary saphenous vein bypass graft stenoses. The relative merits of plaque excision and removal versus balloon dilatation remain uncertain. We compared outcomes after directional coronary atherectomy or angioplasty in patients with de novo bypass graft stenoses.
Methods and Results Fifty-four North American and European sites randomized 305 patients with de novo vein graft lesions to atherectomy (n=149) or angioplasty (n=156). Quantitative coronary angiography at a core laboratory assessed initial and 6-month results. Initial angiographic success was greater with atherectomy (89.2% versus 79.0%), as was initial luminal gain (1.45 versus 1.12 mm, P<.001). Distal embolization was increased with atherectomy (P=.012), and a trend was shown toward more non–Q-wave myocardial infarction (P=.09). Although the 6-month net minimum luminal diameter gain was 0.68 mm for atherectomy and 0.50 mm for angioplasty, the restenosis rates were similar, 45.6% for atherectomy and 50.5% for angioplasty (P=.491). At 6 months, there was a trend toward decreased repeated target-vessel interventions for atherectomy (P=.092); in addition, 13.2% of patients treated with atherectomy versus 22.4% of the angioplasty patients (P=.041) required repeated percutaneous intervention of the initial target lesion.
Conclusions Atherectomy of de novo vein graft lesions was associated with improved initial angiographic success and luminal diameter but also with increased distal embolization. There was no difference in 6-month restenosis rates, although primary atherectomy patients tended to require fewer target-vessel revascularization procedures.
Treatment of patients with coronary artery bypass graft stenoses constitutes an increasingly large part of the practice of interventional cardiology.1 2 3 4 5 6 7 8 Percutaneous transluminal coronary angioplasty (PTCA) has been relatively widely used, but the clinical outcomes have varied, depending on the age of the graft, the location of the stenosis within the graft, and the specific anatomic features.1 2 3 4 5 6 9 10 Restenosis rates in grafted vessels are markedly higher than those in native vessels, particularly in older grafts (>3 years old) and in stenoses involving the aortic ostium or graft body. In addition to restenosis, embolization of atheromatous or thrombotic material during the procedure has been reported in up to 5% of patients.
Directional coronary atherectomy (DCA) has also been used to treat these patients.7 8 11 12 13 14 15 16 In the initial Devices for Vascular Intervention (DVI) Registry (Redwood City, Calif), 17% of procedures involved saphenous vein bypass grafts. Observational studies of DCA have reported restenosis rates of approximately 40% for de novo vein graft lesions.7 13 15 A decrease in restenosis rates could theoretically be achieved by debulking the lesion rather than through dilatation alone. The hypothesis of this randomized, multicenter trial was that atherectomy would result in lower restenosis rates than conventional angioplasty in patients with de novo vein graft stenoses.
Study Sites and Operators
Fifty-four experienced centers enrolled patients (52 in North America and 2 in Europe; see the Appendix). These centers and investigators were selected because they had experience with both coronary atherectomy and angioplasty and had a background of clinical investigation in interventional cardiology.17 To qualify as investigators, individual operators had to have performed more than 400 angioplasty procedures with a success rate exceeding 85% and more than 50 DCA procedures with a success rate exceeding 80%. At each site, the Institutional Review Board approved the protocol.
Patients with prior coronary bypass surgery and de novo vein graft lesions who required revascularization and were suitable for either DCA or PTCA were considered for enrollment. The angiographic inclusion criteria were (1) de novo vein graft lesion; (2) vein graft suitable for ≥6F atherectomy catheter (≥3.0 mm); (3) a subtotal diameter stenosis ≥60% and <100% by visual assessment; and (4) lesion length ≤12 mm. If more than one lesion was present in the vein graft, all had to be amenable to either technique to conform with a single treatment assignment. Patients who had had a myocardial infarction within the previous 5 days were excluded. A log of each atherectomy performed on a vein graft outside of CAVEAT-II was maintained at each site (Table 1⇓). The reasons for not including patients in CAVEAT-II varied, although the most common reason was investigator preference, which occurred in only 36% of the patients. Participation in another investigational study and repeated treatment of lesions were the reasons for exclusion in 10% and 19% of the patients, respectively.
The coordinating center was Duke University in Durham, NC. After informed consent was obtained, the randomization center was contacted by telephone for treatment assignment.
Procedural technical details were published previously for atherectomy and PTCA.17 18 19 Although technical success was defined conventionally as achieving ≤50% stenosis, the goal of revascularization was to achieve an angiographic result of the minimum possible residual stenosis (<20% residual stenosis). Crossover to the other treatment modality was strongly discouraged, but predilatation with a ≤2.0-mm balloon was permitted before atherectomy. The operators prospectively identified patients in whom predilatation would be needed; this was not considered crossover.
Aspirin (≥160 mg) and at least one dose of a calcium channel blocker were administered within 24 hours before the procedure. Activated clotting times were maintained at >350 seconds during the procedure by administering a 10 000-U bolus of heparin; supplemental boluses were given, depending on the size of the patient and the length of the procedure. Femoral access sheaths were removed 4 to 24 hours after the procedure was completed. After the procedure, aspirin (325 mg q.d.) and a calcium channel blocker were given for approximately 1 month. Warfarin was not routinely administered. ECGs were obtained before and within 24 hours after the procedure. Creatine kinase levels with myocardial isoenzymes (MB) were obtained 12 and 24 hours after the procedure.
At the beginning of each procedure, after a dose of 100 to 200 mcg of intracoronary nitroglycerin, coronary angiography was obtained of the target graft in two orthogonal views with a 7F or an 8F diagnostic coronary arterial catheter. These views were repeated at the end of the procedure, again with a 7F or an 8F catheter. This procedure was repeated for the 6-month follow-up angiogram.
The Cleveland Clinic Foundation Angiographic Core Laboratory performed independent, blinded assessment by use of quantitative coronary angiography (QCA) (Image Com). Paired acute and follow-up angiograms were measured by technicians blinded to treatment assignment; the device-containing images were spliced out. The most severe hemiaxial end-diastolic view without foreshortening was used for analysis, although both orthogonal views were analyzed. Preprocedure films were analyzed for extent of coronary artery disease, number of lesions, and lesion complexity and morphology. Each lesion was assessed in all films for vessel caliber, absolute minimum diameter, percent diameter stenosis, and percent stenosis by cross-sectional area.
Quantitative analysis was performed with the use of a validated edge-detection algorithm.20 Vessel edges were determined with the computerized algorithm, and luminal diameters were measured with the empty and contrast-filled catheters as references.
The long-term interobserver variability of the Angiographic Core Laboratory was determined by analyzing 15 cineangiograms on two occasions 8 months apart. Each reviewer independently selected projection angle and frame selection. Standard errors (and correlation coefficient, r) of the measurements for reference diameter and minimum luminal diameter values were 0.25 (.89) and 0.18 (.81) preintervention and 0.23 (.91) and 0.16 (.97) postintervention.
Core Pathology Laboratory
St Elizabeth’s Hospital (Boston, Mass) served as the core pathology laboratory. Tissue specimens from atherectomy were immediately placed in 4% paraformaldehyde/PB5 for 2 hours, stored at 4°C in 30% sucrose/phosphate-buffered saline, and sent to the core laboratory for light microscopy and immunohistochemistry.
Acute end points included procedural success (≤50% diameter stenosis by QCA), major complications, a composite index of complications (death, myocardial infarction, emergency bypass, or abrupt-closure syndrome), abrupt closure, hospital charges, quality of life, and length of hospital stay. The diagnosis of myocardial infarction by each site was defined as creatine kinase MB greater than twice the upper limit of normal. Q-wave changes were recorded. Abrupt closure was defined by the site as angiographically documented Thrombolysis in Myocardial Infarction (TIMI) grade 0 or 1 flow with 100% stenosis and clinical or ECG evidence of ischemia lasting >5 minutes. Distal embolization was defined according to the clinical judgment of the individual investigators and included decreased flow in a previously patent vessel distal to the target lesion in the absence of an occlusion at the treatment site.
End points assessed during follow-up included restenosis (by absolute luminal diameter), major late clinical events (death, myocardial infarction, and coronary bypass surgery), functional capacity, and exercise time. Other follow-up events included angina pectoris; need for repeated intervention; and a composite index of death, myocardial infarction, coronary bypass graft surgery, and repeated intervention.
Determination of Sample Size
The sample size calculation was based on the assumption that the restenosis rate after angioplasty would be approximately 60% compared with approximately 40% after atherectomy.1 2 3 10 13 It was assumed that 15% of patients would have an unsuccessful procedure or crossover, and 15% would not return for angiographic follow-up. Given these assumptions and using an α of .05 and 80% power, we estimated that 300 patients would be required.
Data Management and Statistical Analysis
The research coordinator and investigators prospectively entered the data on a case report form at each site.17 These case report forms were forwarded to the coordinating center and verified by range and consistency checks. Cardiology nurse monitors audited all case report forms. Continuous data are presented as median (25th, 75th percentiles); to test for a difference between treatment groups, we used the Wilcoxon rank-sum test.21 Categorical data are presented as frequency (percentage); we used the χ2 test or Fisher’s Exact Test when comparing treatment groups. Kaplan-Meier survival methods were used to determine the 6-month event rates for clinical outcomes. The event rates were compared between treatment groups using the log-rank test, which incorporated data from a follow-up window that extended through 240 days after enrollment.
Patients with missing data for a given variable were excluded from the calculation of the percentage of patients having that characteristic. This prevented the addition of bias that would result from assuming that patients with missing data were negative for that characteristic.
Relation With Sponsors
The Steering Committee set standards for protocol design and execution that were independent of the sponsors (DVI and Eli Lilly Inc). No member of the Steering Committee or coordinating center was permitted to have any financial equity position with either sponsor. All data were managed at the Duke University Coordinating Center, and no data were accessible to the investigators or sponsors until all 6-month follow-up angiographic data had been analyzed.
From March 12, 1992, to April 16, 1993, 305 patients were randomized: 149 to DCA and 156 to PTCA (Table 2⇓). The majority of patients had unstable angina. Comorbid conditions were frequent; they were seen in 36.9% of the DCA patients and 27.6% of PTCA patients. Rates of infarction within 14 days, congestive heart failure, and cerebrovascular disease were similar in both groups, although peripheral vascular disease occurred more frequently in the DCA group (26.2% versus 15.4%). The grafts being treated were old: 9.5 years old for DCA and 9.9 years old for PTCA (Table 3⇓).
The distributions of vein graft locations and locations of the stenoses within the grafts were similar between the two groups; grafts with single distal insertion sites were most commonly treated. In the DCA patients, a vein graft to the circumflex was most common, while in the PTCA group, an equal number of left anterior descending and circumflex coronary grafts were treated. Typically, the lesion was within the body of the graft (81.9% for DCA; 89.1% for PTCA); only a minority of patients had aorto-ostial lesions treated (14.8% and 9.0% for DCA and PTCA, respectively). As per the protocol, the initial TIMI grade flow of 2 or 3 was predominant in each group (89.5% for DCA; 92.7% for PTCA). A small percentage of patients had decreased flow at baseline.
There was no difference in lesion length between the DCA (10.9 mm) and PTCA (11.0 mm) groups. Adverse specific lesion morphology was common and similar in the two groups (Table 4⇓). Lesion eccentricity was the most common morphology (54.1% for DCA; 58.4% for PTCA). An irregular contour or thrombus often was present.
The initial success rates varied, depending on whether success was determined by the clinical center or the core angiographic laboratory (Table 5⇓). Success was not different in the two groups when site assessment was considered (98.0% for DCA; 97.4% for PTCA). With blinded core laboratory assessment, the success rate was significantly higher for DCA at 89.2% versus 79.0% for PTCA. With QCA, the initial minimum luminal diameter was 0.92 mm for DCA and 1.03 mm for PTCA, and the corresponding diameter stenoses were 73.7% and 71.7% (Table 7⇓). The initial gain achieved by treatment was significantly greater after DCA (1.45 versus 1.12 mm for PTCA, P<.001), so the postprocedure diameter stenosis was also less (31.5% versus 37.6% for PTCA, P<.001).
Predilatation was common in the DCA patients (Table 5⇑). Other adjunctive devices were also used more frequently in these patients: 28.2% versus 14.1% of PTCA patients (P=.003). Patients undergoing DCA also required more radiographic contrast (225 versus 175 mL for DCA and PTCA, respectively) and had longer procedure times.
The in-hospital rates for most major complications were similar in the two treatment groups (Table 6⇓). Mortality was low (2.0% for DCA; 1.9% for PTCA), as was the need for coronary bypass graft surgery. The most important differences were in the rate of acute myocardial infarction and distal embolization. The incidence of Q-wave myocardial infarction was low in each group (1.3% for atherectomy; 1.9% for PTCA). There was a trend toward more non–Q-wave myocardial infarction after DCA (16.1%) than after PTCA (9.6%, P=.09). This usually happened in association with distal embolization, which occurred significantly more often with DCA (P=.012). However, abrupt closure of the treated segments was low in both treatment arms. With the composite adverse end point, there was a trend for a higher rate in the DCA group, mainly because of the excess non–Q-wave myocardial infarctions (P=.059) (Table 6⇓).
The rate of angiographic follow-up for the entire cohort was 80% after a median of 5.9 months (Table 7⇓). The primary end point of angiographic restenosis, >50% stenosis after an initially successful procedure, occurred less often with DCA (43.2% versus 52.1% using site readings and 45.6% versus 50.5% using QCA readings); however, the difference was not significant.
With continuous QCA data, the initial gain with atherectomy was greater: 1.45 versus 1.12 mm with PTCA. At follow-up, the late loss was also somewhat greater (0.62 versus 0.53 mm). At 6 months, the net gain with DCA was still greater (0.68 versus 0.50 mm), but this was not significant (P=.066) and the variability was high.
Figs 1⇓ and 2⇓ show the distribution of lesions. The initial minimum luminal diameter achieved with DCA was significantly larger. By the time of the follow-up angiogram, the minimum lumen diameter with DCA remained larger, but it was no longer significant. The distribution of follow-up stenoses showed that the most common follow-up diameter stenosis for directional coronary atherectomy was 30% to 40%, while with PTCA it was 40% to 50% (Fig 2⇓). Again, this was not significantly different (P=.10).
Clinical follow-up data were available for 300 patients (98%) during a median follow-up of 6.2 months (Table 8⇓). Six-month survival was 95.3% for DCA and 92.3% for PTCA (P=.411). Q-wave infarction was rare in both groups (2.7% for DCA; 4.0% for PTCA), as was the development of stroke. Survival without repeated coronary bypass graft surgery was also excellent: 94.5% and 95.3% for DCA and PTCA, respectively. The survival rate without repeated percutaneous target-vessel intervention was 86.8% for DCA patients versus 77.6% for PTCA patients (P=.041).
This randomized trial of DCA versus PTCA documented that DCA resulted in a higher angiographic success rate and larger initial improvement in graft dimensions for de novo vein graft lesions. Given the large number of patients who have undergone coronary bypass graft surgery with venous conduits and the well-documented continuous rate of attrition in these grafts, treating these patients will continue to be a significant clinical problem.22 23 The underlying pathophysiology in these patients is complex, with degenerated atheromatous lesions and often superimposed thrombus.24 25 26 27 28 29 The results of dilatation depend in part on the age of the graft, the discrete (versus diffuse) nature of the lesion, and the location of the stenosis in the graft. Treatment of older grafts, >3 to 5 years old, and stenosis of the aortic origin or in the body of the graft have been associated with markedly increased restenosis rates.1 2 5 9 10
DCA has been used relatively frequently in venous conduits because of their lack of side branches, their usually straight and nontortuous course, and their large size. In the initial DVI Registry, 17% of cases involved treatment of vein grafts.8 Restenosis rates in these and other series have varied, while embolization has been relatively low.7 8 12 13 14 15 16 30
Although both continuous and discrete dichotomous criteria were used to define restenosis, the latter was chosen as the primary end point using a definition of >50% diameter stenosis at follow-up. Restenosis defined in this manner occurred in a similar proportion of patients who received DCA (45.6%) or PTCA (50.5%, P=.491). This rate of restenosis after DCA is similar to that previously reported by Garratt et al,13 who found a restenosis rate of 42% in patients treated for de novo vein graft lesions in whom there was no deep vessel wall resection. It is lower than other series have documented.31 An equally important finding was that the restenosis rate in these 9.7-year-old vein grafts after PTCA was considerably lower (50%) than that previously recorded.2 10 This contrasts with published series of dilatation in old bypass graft stenoses, which document a restenosis rate of >60% to 70%. Platko et al2 reported a restenosis rate of 83% in patients with vein grafts >3 years old, while Douglas et al10 found a restenosis rate of 64% in patients with vein grafts >5 years of age. These previous studies did not have complete angiographic follow-up, however, so selection bias may explain the differences—more patients with symptomatic restenosis returned for follow-up angiography. Regardless of these considerations, PTCA still resulted in very reasonable intermediate-term outcomes in these patients.
Restenosis was also assessed with continuous QCA criteria. Although the initial gain with DCA was significantly greater, the loss was also somewhat greater. The overall net gain at the end of 6 months did remain larger with DCA (0.68 versus 0.50 mm), but not significantly so (P=.066). The relation between acute gain and late loss has been the subject of intense study.17 18 32 33 34 A critical determinant of subsequent restenosis is the minimum luminal diameter achieved17 32 33 ; new devices may decrease angiographic restenosis by yielding significantly better initial results. The inevitable neointimal hyperplasia is then better tolerated and may not result in clinical restenosis. It is possible that the lack of decreased restenosis with DCA in this trial is related to the fact that a residual diameter stenosis of 31.5% remained.
Although the primary end point of this study was angiographic restenosis, important secondary end points were also assessed. As was true in CAVEAT-I17 and CCAT,18 DCA resulted in improved initial success rates and a larger median initial lumen. In DCA-treated patients in this series, the initial success rate was 89.2% versus 79.0% for PTCA patients, findings very similar to the success rates in CAVEAT-I. The median initial luminal gain was substantially larger: 1.45 mm for DCA versus 1.12 mm for PTCA. There were no differences in in-hospital mortality, need for coronary bypass graft surgery, and Q-wave myocardial infarction, but there was a trend toward increased non–Q-wave myocardial infarction in DCA patients and a significant increase in distal embolization (P=.012).
In CAVEAT-I, there was a moderate decrease in restenosis rates with DCA, but there was no difference in follow-up clinical events or need for repeated intervention.17 In the CCAT trial, there were no differences with respect to repeated intervention.18 In CAVEAT-II, although the 6-month cumulative mortality and Q-wave myocardial infarction rates were similar, there was a trend toward a decreased need for repeated target-vessel intervention or coronary artery bypass graft: 18.6% of DCA patients required repeated target-vessel revascularization versus 26.2% of PTCA patients. The need for any later intervention was also decreased in DCA patients, although the difference was smaller (24.8% versus 31.5% for DCA and PTCA, respectively). There are limited other well-controlled data on follow-up of patients treated for vein graft disease. It is known, however, that vein graft disease is progressive. Longer-term follow-up of these randomized patients in CAVEAT-II will be required to ascertain how long the modestly improved outcome with DCA lasts. It is also possible that this difference in favor of DCA could have resulted partly from the higher rate of distal embolization and non–Q-wave myocardial infarction in this group, leaving patients with a lower likelihood of undergoing symptom-driven repeated interventions.
Other interventional approaches are also being tested in patients with focal vein graft disease, particularly stents.20 35 36 37 There is substantial enthusiasm because of the large initial minimum luminal diameter that can be achieved. The long-term results have not been subjected to a well-controlled trial, although one is now being planned. In a retrospective comparative assessment of a small group of patients, Pomerantz et al30 found no difference in restenosis rates between DCA and one specific stent configuration.
Some limitations should be kept in mind in the interpretation of this study. The first and perhaps most important is that the number of patients was relatively small. The estimated sample size was based on published rates of restenosis of de novo vein graft lesions treated with DCA and PTCA. While the restenosis rate for atherectomy was similar to what had been published, the PTCA restenosis rate was substantially better than expected for these 9.9-year-old grafts. Whether this better-than-expected outcome with PTCA related to patient selection, dilatation performance, or bias in the previous literature cannot be determined. If larger numbers of patients had been randomized, the differences between the two treatment arms might have been more striking. The second limitation relates to the fact that even in the group with the better postprocedure result (DCA), the diameter stenosis was 31.5%. More aggressive DCA or post-DCA dilatation might have resulted in better immediate postprocedure results and improved outcome, although this is controversial.
In this randomized trial, DCA resulted in a higher initial angiographic success rate and a larger initial improvement in graft dimensions than conventional PTCA with the use of QCA techniques. Achievement of this improved success rate was at least partially offset by the moderate initial increase in distal embolization and non–Q-wave myocardial infarction, which are probably the results of passage of the large atherectomy device and active debulking and manipulation of the lesion. There was a trend toward decreased performance of repeated target-vessel intervention at 6 months in patients treated with DCA, but there was no difference in angiographic restenosis rates. Overall, our findings suggest that both forms of revascularization are viable strategies for this complex patient group.
CAVEAT-II Sites and Investigators
Cleveland (Ohio) Clinic Foundation (P. Whitlow, S. Ellis, I. Franco, E. Topol [study chair], D. Debowey [Angiographic Core Laboratory], M. Lincoff); Christ Hospital, Cincinnati, Ohio (D. Kereiakes, C. Abbottsmith); Washington (DC) Cardiology Center (K. Kent, M. Leon, A. Pichard, L. Satler, J. Popma); Sequoia Hospital, Redwood City, Calif (T. Hinohara); St Vincent’s Medical Center, Bridgeport, Conn (E. Kosinski); Carolinas Medical Center & Carolinas Heart Institute, Charlotte (C. Simonton, R.M. Bersin, J. Cedarholm, B. Wilson); Mayo Clinic Foundation, Rochester, Minn (D.R. Holmes, Jr); Midwest Heart Research Foundation, Lombard, Ill (L.S. McKeever); Methodist Hospital, Memphis, Tenn (F. Martin); Riverside Methodist Hospital, Columbus, Ohio (A. Chapekis, B.S. George); Medical College of Virginia, Richmond (M. Cowley); St Vincent’s Hospital, Indianapolis, Ind (C. Pinkerton, T. Peters); St Francis Hospital, Beech Grove, Ind (M. Cohen); Boston (Mass) University Medical Center (A. Jacobs, D.P. Faxon, G. Levine); Maimonides Medical Center, Brooklyn, NY (J. Shani); Maine Medical Center, Portland (M. Kellett, Jr); Emory Hospital, Atlanta, Ga (S. King); Jewish Hospital, Louisville, Ky (R. Masden); Graduate Cardiology Consultants, Philadelphia, Pa (R.S. Gottlieb); Minneapolis (Minn) Heart Institute (M. Mooney); Ochsner Foundation Hospital, New Orleans, La (C.J. White); Klinikum Grosshadern Der Universitat, Munich, Germany (B. Hofling); Rhode Island Hospital, Providence (D. Williams); University of Louisville, Ky (D. Talley); Southwest Cardiology Associates, Albuquerque, NM (H. White); Johns Hopkins Hospital, Baltimore, Md (J. Brinker); Loyola Medical Center, Maywood, Ill (F. Leya); University of Washington, Seattle (D.K. Stewart, J. Chambers); St. Vincent–Portland, Ore (P. Au); Massachusetts General Hospital, Boston (I. Palacios); Beth Israel Hospital, Boston, Mass (R. Kuntz); William Beaumont–Royal Oak (Mich) Hospital (R. Safian); Florida Hospital, Orlando (R. Ivanhoe); Cardiologie/CHU Ranguiel, Toulouse, Cedex, France (J. Puel); Fairfax Hospital, Annandale, Va (B. Raybuck); Montreal Heart Institute, Quebec, Canada (R. Bonan); Walter Reed Army Medical Center, Washington, DC (C. Pearson, J.R. Laird); University of Virginia, Charlottesville (L. Burwell); Mother Frances, Tyler, Tex (R.J. Carney); Sutter Hospitals, Sacramento, Calif (R. Bellinger); Hahnemann University Hospital, Philadelphia, Pa (M. Cohen); Vancouver General Hospital, British Columbia, Canada (D. Ricci); New York Hospital–Cornell Medical Center, New York (A. Spokojny); Henrico Hospital, Fredericksburg, Va (T.E. Martyak); Toronto General Hospital, North York, Ontario, Canada (E. Cohen); Mount Sinai Hospital, Toronto, Ontario, Canada (A. Adelman); Charleston (WV) Medical Center (S. Lewis); St Paul’s Hospital, Vancouver, British Columbia, Canada (J. Webb); Foothills Hospital, Calgary, Alberta, Canada (D. Traboulsi); Presbyterian Hospital, Charlotte, NC (B. Reen, G. Niess); St Lukes–Roosevelt Hospital, New York, NY (J. Slater); Ottawa Heart, Ontario, Canada (J-F. Marquis); Cleveland Clinic Florida, Ft Lauderdale, Fla (H.S. Bush); Lenox Hill Hospital, New York, NY (J.W. Moses); Healthwest Regional Medical Center, Phoenix, Ariz (R. Heuser); Ft Sanders Regional Medical Center, Knoxville, Tenn (M. Ayres); Columbia Presbyterian Medical Center, New York, NY (M.A. Apfelbaum); East Jefferson Hospital, Metairie, La (S. Bleich); University of Alabama, Birmingham (G. Roubin); Sentara Norfolk General Hospital, Norfolk, Va (R. Stein, C.W. Hartman); St Mary’s Hospital, Saginaw, Mich (R. DeNardo); Shadyside Hospital, Pittsburgh, Pa (D. Lindsey); Presbyterian Medical Center, Philadelphia, Pa (W. Corin, B. Unterecker); Medical Center of Delaware, Newark (M. Stillabower); Methodist Hospital of Indiana, Indianapolis (M. Mick); Mt Sinai Medical Center, New York, NY (S. Sharma); St John’s Hospital, Santa Monica, Calif (H. Cohen); Laval Hospital, St-Foy, Quebec, Canada (G. Barbeau); Virginia Beach (Va) General Hospital (J. Griffin); Olympia Fields (Ill) Hospital (A. Arnold); McLaren Regional Medical Center, Flint, Mich (R. DeNardo); and Duke University Medical Center (study coordinating center), Durham, NC (R. Califf).
This study was supported by grants from DVI and Eli Lilly, Inc.
A complete list of investigators and centers is provided in the Appendix.
- Received August 9, 1994.
- Revision received November 2, 1994.
- Accepted November 13, 1994.
- Copyright © 1995 by American Heart Association
Douglas JS, Weintraub WS, Liberman HA, Jenkins M, Cohen CL, Morris DC. Update of saphenous graft (SVG) angioplasty: restenosis and long-term outcome. Circulation. 1991;84(suppl II):II-249. Abstract.
Selmon MR, Hinohara T, Robertson GC, Rowe MH, Vetter JW, Bartzokis TC, Braden LJ, Simpson JB. Directional coronary atherectomy for saphenous vein graft stenoses. J Am Coll Cardiol. 1991;17(suppl A):23A. Abstract.
Garratt KN, Edwards WD, Kaufmann UP, Vlietstra RE, Holmes DR Jr. Differential histopathology of primary atherosclerotic and restenotic lesions in coronary arteries and saphenous vein bypass grafts: analysis of tissue obtained from 73 patients by directional atherectomy. J Am Coll Cardiol. 1991;17:442-448.
Cowley MJ, DiSciascio G. Directional coronary atherectomy for saphenous vein graft disease. Cathet Cardiovasc Diagn. 1993;1:10-16.
Topol EJ, Leya F, Pinkerton CA, Whitlow PL, Hofling B, Simonton CA, Masden RR, Serruys PW, Leon MB, Williams DO, King SB III, Mark DB, Isner JM, Holmes DR Jr, Ellis SG, Lee KL, Keeler GP, Berdan LG, Hinohara T, Califf RM, for the CAVEAT Study Group. A comparison of directional atherectomy with coronary angioplasty in patients with coronary artery disease. N Engl J Med. 1993;329:221-227.
Adelman AG, Cohen EA, Kimball BP, Bonan R, Ricci DR, Webb JG, Laramee L, Barbeau G, Traboulsi M, Corbett BN. A comparison of directional atherectomy with balloon angioplasty for lesions of the left anterior coronary artery. N Engl J Med. 1993;324:228-233.
Safian RD, Gelbfish JS, Erny RE, Schmitt SJ, Schmidt DA, Baim DS. Coronary atherectomy: clinical, angiographic, and histological findings and observations regarding potential mechanism. Circulation. 1990;82:69-79.
Mancini GBJ, Simon SB, McGillem MJ, LeFree MT, Friedman HZ, Vogel RA. Automated quantitative coronary arteriography: morphologic and physiologic validation in vivo of a rapid digital angiographic method. Circulation. 1987;75:452-460.
Matthews DE, Farewell VT. Using and Understanding Medical Statistics. 2nd ed. Basel, Switzerland: Karger; 1988;20-87.
Hamby RI, Aintablian A, Handler M, Voleti C, Weisz D, Garvey JW, Wisoff G. Aortocoronary saphenous vein bypass grafts: long term patency, morphology and blood flow in patients with patent grafts early after surgery. Circulation. 1979;60:901-909.
Hinohara T, Robertson GC, Selman MR, Vetter JW, Rowe MH, Braden LJ, McAuley BJ, Sheehan DJ, Simpson JB. Restenosis after directional coronary atherectomy. J Am Coll Cardiol. 1992; 20:623-632.
Kuntz RE, Safian RD, Carrozza JP, Fishman RF, Mansour M, Baim DS. The importance of acute luminal diameter in determining restenosis after coronary atherectomy or stenting. Circulation. 1992;86:1827-1835.