Author + information
- Received December 31, 1998
- Revision received May 12, 1999
- Accepted June 25, 1999
- Published online October 1, 1999.
- Takahiko Suzuki, MD∗,*,
- Hiroaki Hosokawa, MD†,
- Osamu Katoh, MD‡,
- Tamotsu Fujita, MD§,
- Katsumi Ueno, MD∥,
- Shinichi Takase, MD¶,
- Kenshi Fujii, MD#,
- Hideo Tamai, MD∗∗,
- Tadanori Aizawa, MD††,
- Tetsu Yamaguchi, MD, PhD‡‡,
- Hiroyuki Kurogane, MD§§,
- Mikihiro Kijima, MD∥∥,
- Hirotaka Oda, MD¶¶,
- Etsuo Tsuchikane, MD##,
- Tomoaki Hinohara, MD, FACC∗∗∗,
- Peter J Fitzgerald, MD, PhD, FACC†††,
- for the ABACAS Investigators
- ↵*Reprint requests and correspondence: Dr. Takahiko Suzuki, Toyohashi Heart Center, 21 Aza-Gobutori, Ohyama-cho, Toyhashi, Aichi 441-8071, Japan
This study was conducted to evaluate: 1) the effect of adjunctive percutaneous transluminal coronary angioplasty (PTCA) after directional coronary atherectomy (DCA) compared with stand-alone DCA, and 2) the outcome of intravascular ultrasound (IVUS)-guided aggressive DCA.
It has been shown that optimal angiographic results after coronary interventions are associated with a lower incidence of restenosis. Adjunctive PTCA after DCA improves the acute angiographic outcome; however, long-term benefits of adjunctive PTCA have not been established.
Out of 225 patients who underwent IVUS-guided DCA, angiographically optimal debulking was achieved in 214 patients, then they were randomized to either no further treatment or to added PTCA.
Postprocedural quantitative angiographic analysis demonstrated an improved minimum luminal diameter (2.88 ± 0.48 vs. 2.6 ± 0.51 mm; p = 0.006) and a less residual stenosis (10.8% vs.15%; p = 0.009) in the adjunctive PTCA group. Quantitative ultrasound analysis showed a larger minimum luminal diameter (3.26 ± 0.48 vs. 3.04 ± 0.5 mm; p < 0.001) and lower residual plaque mass in the adjunctive PTCA group (42.6% vs. 45.6%; p < 0.001). Despite the improved acute findings in the adjunctive PTCA group, six-month angiographic and clinical results were not different. The restenosis rate (adjunctive PTCA 23.6%, DCA alone 19.6%; p = ns) and target lesion revascularization rate (20.6% vs. 15.2%; p = ns) did not differ between the groups.
With IVUS guidance, aggressive DCA can safely achieve optimal angiographic results with low residual plaque mass, and this was associated with a low restenosis rate. Although adjunctive PTCA after optimal DCA improved the acute quantitative coronary angiography and quantitative coronary ultrasonography outcomes, its benefit was not maintained at six months.
Directional coronary atherectomy (DCA) was introduced to improve the acute outcome of coronary interventions as well as to reduce restenosis by debulking atheromatous tissue (1). Coronary Angioplasty Versus Excisional Atherectomy Trial (CAVEAT) (2), the first randomized trial to compare DCA and percutaneous transluminal coronary angioplasty (PTCA), failed to demonstrate a significant reduction in restenosis. It was thought that one of the reasons for the relatively high restenosis rate among DCA patients was due to the performance of less than optimal DCA procedures.
There are two approaches to achieve an optimal initial DCA outcome: more aggressive debulking and post-DCA adjunctive balloon dilations. The Optimal Atherectomy Restenosis Study (OARS) demonstrated that optimal DCA followed by adjunctive PTCA resulted in a minimal residual stenosis with a relatively low restenosis rate (3). Balloon Versus Optimal Atherectomy Trial (BOAT), a recent randomized trial, demonstrated that the restenosis rate is significantly lower after “optimal DCA,” usually in conjunction with adjunctive dilation, compared with PTCA (4). In the OARS trial, despite “optimal DCA” by angiographic criteria, the residual plaque area assessed by intravascular ultrasound (IVUS) was significantly high. Some preliminary reports suggest that plaque burden is an important predictor for restenosis, thus more aggressive plaque removal may be required for further reduction of the restenosis rate. By using IVUS-derived information before or during DCA to guide the procedure, more effective debulking may be possible to achieve less plaque burden (5). Adjunctive PTCA seems to improve lumen size further after “optimal” DCA; however, there is no information available regarding the acute as well as long-term effect of adjunctive PTCA after more aggressive IVUS-guided DCA.
The purpose of this study is to evaluate: 1) the effect of adjunctive PTCA after DCA compared with stand-alone DCA, and 2) the outcome of IVUS-guided aggressive DCA.
This was a prospective multicenter randomized trial to evaluate the effect of balloon dilation after optimal DCA. The study was performed between December 1994 and September 1995.
Study site and operators
The trial was conducted at 12 institutions in Japan. These institutions had been well experienced in DCA procedures and other complex interventional therapy. To maintain the quality level of the procedure, individual operators had to have performed over 50 DCA procedures to be qualified for this study.
Patients were selected according to the following inclusion and exclusion criteria.
1. Inclusion criteria:
• symptomatic coronary artery disease
• native coronary artery
• de novo lesion or restenotic lesion (from one prior PTCA)
• lesion stenosis ≧75% and <100% and ≦15 mm in length by visual estimation
• vessel suitable for DCA device size ≧7 F
• acceptable candidate for coronary artery bypass graft (CABG)
• age between 20 and 80-years-old
• signed written informed consent
1. Exclusion criteria:
• culprit lesion for acute myocardial infarction within the last one month
• left main lesion/right coronary artery ostial lesion
• angulated lesion >60°
• diffuse disease
• presence of thrombus at the lesion
• significant tortuosity
• severe peripheral vascular disease
• markedly calcified lesion by angiography and/or superficial calcium >180° detected by IVUS and concomitant significant medical condition.
Initial patient enrollment and DCA procedure
Consecutive patients who were treated with coronary intervention and who met the clinical and angiographic criteria were enrolled into the trial after informed consent was obtained under a protocol approved by the Institutional Review Board at each participating center. The patients then underwent IVUS evaluation. Patients with greater than 180° of superficial calcification were eliminated and deregistered from this study. The initial IVUS information was used for device selection and window orientation based on plaque distribution. DCA was performed using the Simpson Coronary AtheroCath (Guidant, Santa Clara, California). DCA was performed in a routine fashion, typically starting with low balloon pressures (5 to 10 psi). Multiple cuts were obtained using gradually increasing balloon pressures. The goal of optimal DCA by angiographic appearance was a <30% angiographic residual stenosis by visual estimation, and operators felt that no further treatment was necessary based on the angiographic findings. After achievement of angiographically optimal DCA, IVUS was repeated to evaluate the residual plaque burden. When visual estimation of residual plaque was felt to be significant and atherectomy was inadequate, DCA was repeated with a more aggressive approach based on IVUS information; precise orientation of the window towards the plaque using higher balloon pressures and making multiple cuts was generally recommended. After each device insertion, the results were evaluated by both angiography and IVUS. If the debulking still appeared to be inadequate, the procedure was again repeated. When the operator achieved an optimal DCA outcome by angiographic criteria (visually estimated residual stenosis <30%) or by IVUS findings with adequate debulking (visually estimated percent plaque area <50%), or felt that no further debulking could be achieved despite significant residual plaque, the DCA procedure was completed.
Randomization and adjunctive balloon dilation
The criteria for successful DCA for randomization was based on angiographic information. Patients with a residual stenosis of less than 30% by visual estimation without significant angiographic complications were randomized to either DCA treatment alone (no further therapy) or DCA with adjunctive PTCA. Those patients who could not reach those criteria or who had procedural complications during the initial DCA were excluded from randomization and were treated accordingly by physician’s preference. The balloon catheters used for the adjunctive dilation were selected based on visual estimation of the vessel size with a balloon/artery ratio of 1.1 to 1.3, and balloon inflations for 3 min at low pressures (2 to 5 atmospheres) were recommended.
Angiography and IVUS
Coronary angiography was performed after administration of 100 to 200 μg of nitroglycerin before the procedure. Angiograms were taken in two orthogonal views. The angles of these views were recorded. After DCA and PTCA, angiograms were repeated in the same angles. For the IVUS procedures, 2.9 F or 3.2 F CVIS catheters were used. After placing the IVUS catheter distal to the target lesion, intracoronary nitroglycerin was administered. The catheter was slowly pulled back either manually or using the automatic pullback system, and images were recorded in VTR. Images obtained during the procedure were used for case selection and procedure guidance.
Angiograms were analyzed by experienced angiographers at the angiographic core lab at Stanford University Medical Center. The analysis was performed blindly for the treatment arm. Pre- and postangiographic lesion morphology was analyzed using American Heart Association classification. The view that showed the most significant stenosis was used for quantitative coronary angiographic (QCA) analysis. QCA was performed using Stanford’s method, which utilizes the validated edge-detection algorithm. Reference vessel size, minimum lumen diameter (MLD) and diameter stenosis (DS) were measured. Quantitative coronary ultrasonography (QCU) analysis was performed at independent core lab by experienced physicians for IVUS at Stanford University. The lumen/atheroma border and external elastic membrane (EEM) were traced manually. From these tracings, vessel area, lumen area, plaque area and MLD values were automatically measured by computer.
Core pathology laboratory
Gifu University Hospital served as the core pathology laboratory. Tissue samples were weighed and fixed with a 4% formaldehyde solution. Tissue was stained using hematoxylin-eosin and elastic van Gieson, and were analyzed by trained pathologists using an optical microscope.
Patients were followed by investigators at each center. If patients had recurrent ischemic symptoms, they were treated as needed. Patients returned to the investigational centers for clinical and angiographic evaluation at three and six months. Angiograms were obtained after intracoronary nitroglycerin, using the same orthogonal views as before. When a patient had angiographic evidence of restenosis at three or six months, the decision whether to perform a repeat interventional procedure or not depended on the patient’s condition, as well as the preference of the investigator.
Study end points
The primary end point of this study was the angiographic restenosis rate at six months, defined as >50% stenosis by angiography, comparing the group with optimal DCA and the group with optimal DCA followed by PTCA. In addition, QCA and QCU measurements were compared between the two groups. The secondary end point was the clinical event rate (death, myocardial infarction, target vessel revascularization) at six months among these two groups.
Data management and statistical analysis
Investigators at each institution obtained clinical and procedural data as well as follow-up data. The clinical data and QCA and QCU data were collected at Quintils Asia, Inc. (Tokyo, Japan). The size of required sample (200 patients) was based on the assumed rate of angiographic restenosis of 32% in the DCA alone group and 15% in the adjunctive PTCA group (by a two-sided test with an alpha error of 0.05 and a power of 0.80). To compensate for unsuccessful intervention procedures and losses to follow-up, the sample was enlarged by 10% (to 220 patients). For the analysis of continuous data, Wilcoxon test was used to assess differences between the two groups and the results are expressed as means ± SD. Categorical data, which are presented as rates, were compared by chi-square test. For the follow-up clinical data, Wilcoxon test was used for the comparison between the two groups.
In total, 239 patients were initially enrolled in this study based on clinical and angiographic criteria (Fig. 1). Of these, 14 patients were eliminated for the following reasons: 12 patients with calcification greater than 180° by IVUS and two patients with no IVUS image due to equipment malfunction. Thus, 225 patients were enrolled in this study and underwent DCA. Of these, 11 patients were excluded from randomization; complications requiring further treatment occurred in six patients (dissection in four patients, distal embolism in one patient and an aneurysm in one patient), a less than optimal result requiring further treatment in four patients and failure to cross with a DCA device in one patient. These patients were subsequently treated with PTCA (five patients) or stents (six patients). Therefore, the overall success rate achieving angiographic optimal DCA was 95.1%. Overall interventional procedure success rate without significant major complication was 99.5%. None required CABG. There were no deaths, and one patient (0.5%) suffered a Q wave myocardial infarction. Subsequently, a total of 214 cases with successful DCA were randomized into two groups: 106 patients in the DCA alone group and 108 patients in the adjunctive PTCA after DCA group. An example of a case is shown in Figure 2.
Patient and lesion characteristics
Patient characteristics for each group are summarized in Table 1. Despite randomization, males and hypertension were more frequently observed in the DCA alone group. Although there was a trend for more unstable angina in the DCA alone group, these differences were not statistically significant. Target vessel and lesion characteristics are summarized in Table 2. The majority of lesions were de novo, and more than half of the target vessels were LAD. Approximately 80% of the lesions were type B, and none of them were type C. There were no significant differences in lesion characteristics among these two groups.
None of the lesions were predilated with a balloon before DCA. DCA procedure results are summarized in Table 3. There were no significant procedural differences between these two groups. All patients were treated with either a 7 F device or a 7 F graft device. In approximately one-third of the patients, DCA was repeated after initial post-DCA IVUS evaluation (two cycles or more). Among adjunctive balloon dilation in the randomized group, mean balloon/artery ratio was 1.2, mean balloon inflation pressure was 6 atmospheres and mean duration was 312 s.
Use of IVUS information for DCA
After each cycle of DCA, IVUS information was obtained and used to decide if further treatment with DCA was needed or not. After initial angiographically optimal DCA, repeat IVUS was performed, and in 62% of patients, DCA was completed (one cycle). Twenty-eight percent of patients underwent a second-cycle DCA procedure after IVUS evaluation, and 10% of patients had three or more cycles of DCA. The relationship between the number of cycles of DCA treatment and procedural factors is summarized in Table 4. The total number of cuts increased with DCA treatment cycles. With three cycles of the DCA treatment, there was a trend toward more tissue removal and a higher incidence of deep cuts (either resection of media or adventitia). The final residual plaque area was similar regardless of the number of cycles.
The incidence of complications for each randomized group is summarized in Table 5. There were no deaths or coronary bypass surgery in either group. Myocardial infarction was observed in two patients (1.8%) in the DCA-alone group and one patient (0.9%) in the DCA/PTCA group. Abrupt occlusion was observed in one patient in each group, and one patient had extravasation of contrast after balloon dilation. Overall, there was no difference in the incidence of complications between the two randomized groups.
The acute and follow-up results of the QCA data are summarized in Table 6. There was no difference in the vessel size and preprocedural stenosis. Postprocedural MLD was significantly larger and the percent DS was significantly lower for the adjunctive PTCA group compared with the DCA-alone group. Three-month angiograms were available in 191 patients (89%), and six-month angiograms were available in 199 patients (93%). There was no significant difference in MLD and percent DS between the two groups at three and six months. Cumulative QCA curves are shown in Figure 3. Loss index was similar in both groups: 51.3% for DCA alone and 55.6% for adjunctive PTCA. The six-month angiographic binary restenosis was defined as >50% stenosis at six months or sooner, including patients who underwent target vessel revascularization within six months. The overall restenosis rate of this study was 21.4%. The restenosis rate for the DCA alone group was 19.6% and 23.6% for the DCA/PTCA group, resulting in no significant difference between the two groups.
The acute results of the QCU data are summarized in Table 7. Postprocedural MLD, vessel area and lumen area were all significantly larger for the adjunctive PTCA group compared with the DCA-alone group, while all the parameters were not different between the groups before the procedure and also before adjunctive PTCA. Postprocedure plaque burden was low in both groups but significantly lower in the DCA/PTCA group (DCA alone, 45.6% vs. DCA/PTCA, 42.6%).
Six-month clinical follow-up was available in 212 patients (99%) with a mean follow-up period of 6.9 ± 1.2 months. In addition, 12-month clinical follow-up was available in 210 patients (98%). The clinical cumulative events, including acute procedural events, are summarized in Table 8. At six months, there were no cardiac deaths in either group. Major adverse cardiac event rates (cardiac death, Q wave MI, CABG, target lesion revascularization [TLR]), including in-hospital events, were 17.1% for DCA alone and 20.6% for DCA with adjunctive PTCA (p = NS). Although the rate for TLR was slightly higher in the DCA with adjunctive PTCA group (15.2% for DCA alone and 20.6% for DCA with adjunctive PTCA), the difference was not statistically significant. At 12 months, there were no cardiac deaths in either group. The requirement for TLR after six months was rare (none for the DCA alone group and only one patient in the DCA with adjunctive PTCA group).
This study demonstrated two important DCA outcomes. The first was that careful and aggressive DCA using IVUS information by experienced operators provided excellent acute results with not only low residual stenosis by angiogram but also low residual plaque mass by IVUS without an increased risk of complications. These initial optimal results were associated with a low incidence of restenosis at six months. The second was that adjunctive PTCA after aggressive DCA further improved lumen size without increasing the risk of complications; however, there was no improvement in either angiographic or clinical outcome at six months compared with the DCA-alone group.
Effects of adjunctive PTCA
Previous studies (OARS, BOAT) (3,4)have demonstrated that adjunctive PTCA after angiographically optimal DCA further increased MLD and decreased residual percent DS. Because postinterventional lumen size is the most important predictor for restenosis described by Kuntz et al. (6)and by subsequent studies, improvement of the lumen size with adjunctive PTCA after optimal DCA may improve long-term outcome. Based on this information, most patients were treated with adjunctive PTCA after DCA.
In this randomized study, the effect of adjunctive PTCA was evaluated. Adjunctive PTCA further improved lesion lumen size despite more aggressive initial DCA without an increased risk of complications, as expected from other previous studies. However, there were no significant differences in MLD and percent DS or restenosis rate at six months. The reasons for the deviation from previously described large lumen concept are not clear and could be multifactorial. The improvement by adjunctive PTCA after IVUS-guided DCA is relatively small and may not be significant enough to sustain the initial benefit at six months. Also, adjunctive PTCA after aggressive DCA may increase the tissue response with neointimal hyperplasia and enhance the process of remodeling of the vessel wall, or may improve the lumen size only temporarily. This randomized study clearly demonstrated that adjunctive PTCA is not necessary when adequate results are obtained with IVUS-guided DCA.
DCA is performed under angiographic guidance. As demonstrated in BOAT, this approach provides angiographically optimal DCA without a significant risk of perforation, however, to achieve more effective debulking, IVUS may be necessary to guide more precise DCA (5,7,8). It had been demonstrated that angiograms often mislead the distribution of plaque, and IVUS is the only currently available method to evaluate plaque distribution and extensiveness of the disease accurately. In this study, IVUS was used to select appropriate cases and to guide the DCA procedure. Based on IVUS findings, appropriately sized devices were selected. Many cases were performed with larger balloon devices (7 F graft devices), much higher than used in the OARS trial. The device window orientation was based on IVUS findings; the plaque orientation was determined by the relationship between plaque and side branches, and window orientation was guided by the relative relationship with the side branch. After initial DCA with optimal angiographic results, IVUS was repeated to evaluate the effectiveness of debulking. When the debulking was inadequate by IVUS findings, despite optimal angiographic results, more aggressive DCA with higher balloon pressures, larger devices and multiple cuts were attempted. More than one-third of patients were treated with two or more cycles of the DCA procedure based on post-DCA IVUS findings. In the OARS trial, in which IVUS was used in all cases, post-DCA IVUS was obtained mostly to evaluate the DCA results and not necessarily to guide more aggressive subsequent DCA. In this study, the most remarkable finding is the low residual plaque mass after DCA; residual plaque mass was only 42.6%. In OARS, despite better angiographic results (residual stenosis OARS 7%, ABACAS 13%), the residual plaque mass by IVUS was 58%, much higher than ABACAS.
The one major concern of aggressive DCA, particularly under angiographic guidance alone, is perforation (9). Although the incidence is low, obviously perforation needs to be prevented because of potentially significant clinical events. In this study, aggressive DCA guided by IVUS was safe without significantly increasing the risk of perforation; clinically nonsignificant contrast extravasation without tamponade was observed in 0.5%. Meticulous and precise orientation of the window of the device towards the plaque probably prevented clinically significant perforations; however, histologic evaluation demonstrated a high incidence of deep cuts (either excision of media or adventitia). These deep cuts may be difficult to avoid with current approaches because of sequential use of IVUS and the orientation of the window are not as precise as needed to avoid deep cuts. Although clinical significance of “deep cuts” remained controversial (10–12), cutting adventitia needs to be avoided to prevent perforation and late pseudoaneurysm formation (13).
In this study, it was demonstrated that adequate debulking can be achieved by IVUS-guided DCA; however, there are certain limitations with the current approach. One of the most significant limitations is that information by IVUS is sequential and not simultaneous. Although plaque distribution can be determined by the relationship to side branches, particularly in the left anterior descending artery using a diagonal branch or septal, the window orientation relative to plaque distribution is still limited with less accuracy. To achieve more effective debulking or to avoid deep cuts, simultaneous IVUS image capability during the cutting process is required. One of the other major limitations is that the procedure becomes much more complex and time consuming. To evaluate with IVUS frequently during the DCA procedure may not be practical in most of the busy interventional practices for routine use.
The restenosis rate after IVUS-guided DCA in this study was one of the lowest rates among published data in the past. In OARS, the restenosis rate was 30%, higher than ABACAS, despite slightly better angiographic results (residual stenosis 7% in OARS, 13% in ABACAS). One of the significant differences in the procedural aspects of these two studies is the residual plaque mass after angiographically optimal DCA; residual plaque mass was much lower in ABACAS (45.6% for DCA alone and 42.6% for DCA with adjunctive PTCA) compared with OARS (58.5%). Figure 4shows the relationship between angiographic residual stenosis and restenosis rate of previously published DCA trials. In ABACAS, the restenosis rate seems to be discordant compared with other studies. These findings are also somewhat different from previously published data of “large lumen” concept, in which final MLD is the most important prognostic factor. Some previous studies have demonstrated that plaque burden is one of the significant predictors for restenosis after intervention (14). The low restenosis rate in ABACAS may be due to the low residual plaque burden compared with previous other studies. Optimal debulking may be important to reduce the restenosis rate. Comparison of ABACAS data to others needs to be careful and in perspective, because these studies were done in a different set, with different operators and were not designed to compare with other studies. In this study, the mechanism of restenosis was not studied, and it is not clear what is the effect of plaque burden on the restenosis process. Plaque burden may influence the process of remodeling of the artery and intimal hyperplasia, as shown in some of the previous studies with IVUS analysis (15,16,1718,19). In this study, some patients had follow-up IVUS studies, and analysis of these data in the future may provide some insights into the restenosis process.
With IVUS guidance, aggressive DCA yielded effective debulking with low residual plaque mass without increasing the risk of complications. This aggressive DCA was associated with a low angiographic restenosis rate. After aggressive DCA, adjunctive PTCA further enlarged lumen size without risk of complication; however, there was no significant angiographic or clinical benefit achieved by adjunctive PTCA.
We thank E. Nishimura, N. Minagawa and K. Hashimoto for their assistance with the preparation of this manuscript.
☆ This study was supported in part by a grant from Getz Bros. Co., Ltd., Tokyo Japan.
- Adjunctive Balloon Angioplasty After Coronary Atherectomy Study
- Balloon Versus Optimal Atherectomy Trial
- coronary artery bypass graft
- Coronary Angioplasty Versus Excisional Atherectomy Trial
- directional coronary atherectomy
- diameter stenosis
- intravascular ultrasound
- minimum lumen diameter
- Optimal Atherectomy Restenosis Study
- percutaneous transluminal coronary angioplasty
- quantitative coronary angiography
- quantitative coronary ultrasonography
- target lesion revascularization
- Received December 31, 1998.
- Revision received May 12, 1999.
- Accepted June 25, 1999.
- American College of Cardiology
- Hinohara T,
- Selmon M.R,
- Robertson G.C,
- Braden L,
- Simpson J.B
- Simonton C.A,
- Leon M.B,
- Baim D.S,
- et al.
- Baim D.S,
- Cutlip D.E,
- Sharma S.K,
- et al.
- Yock P.G,
- Fitzgerald P.J,
- Linker D.T,
- Angelsen B.A.J
- Kuntz R.E,
- Gibson C.M,
- Nobuyoshi M,
- Baim D.S
- Garratt K.N,
- Holmes D.R,
- Bell M.R,
- et al.
- Holmes D.R,
- Garratt K.N,
- Isner J.M,
- et al.
- Mintz G.S,
- Chuang Y.C,
- Popma J.J,
- et al.
- Mintz G.S,
- Popma J.J,
- Pichard A.D,
- et al.
- Mintz G.S,
- Popma J.J,
- Pichard A.D,
- et al.