Stents in Total Occlusion for restenosis Prevention

Transcription

1 European Heart Journal (2000) 21, doi: /euhj , available online at on Stents in Total Occlusion for restenosis Prevention The multicentre randomized STOP study C. Lotan 1, Y. Rozenman 1, A. Hendler 2, Y. Turgeman 3, O. Ayzenberg 4, R. Beyar 5, R. Krakover 2, T. Rosenfeld 1 and M. S. Gotsman 1 for the Israeli Working Group for Interventional Cardiology 1 Hadassah University Hospital, Jerusalem, Israel; 2 Assaf Harofeh Medical Center, Tzerifin, Israel; 3 HaEmek Medical Center, Afula, Israel; 4 Kaplan Hospital, Rehovot, Israel; 5 Rambam Medical Center, Haifa, Israel Aims This multicentre randomized study set out to evaluate whether coronary stenting improves the results of successful balloon angioplasty for chronic total occlusion. Balloon angioplasty for chronic total occlusion has a high restenosis rate. Several reports have suggested that coronary stenting may decrease the likelihood of restenosis and reocclusion. Methods and Results Patients with total coronary artery occlusions who had an optimal PTCA result were randomized either to no further treatment or additional stent implantation. The AVE microstent was used and all patients were scheduled for a 1-, 3-, and 6-month clinical follow-up. Repeat coronary angiography to assess the rate and pattern of restenosis was performed at 6 months or earlier if clinically indicated. Ninety-six patients were enrolled in this study. The mean age was years and 15 were females. Forty-eight patients were randomized to the stent arm, receiving 52 stents (lengths mm). Stent implantation was successful in all and there were no major procedure-related complications. Sixty-nine patients (72%) were restudied after 6 months. The binary restenosis rates (50%), in the PTCA arm were 70 9% with a minimal lumen diameter of mm compared to 42 1% in the stent arm with a minimal lumen diameter of mm (P=0 034). Reocclusion occurred in 7 9% in the stent group compared to 16 1% in the PTCA group. Restenosis in the PTCA group was focal in 88% of patients and occurred at the point of total obstruction (within 5 mm), compared to diffuse instent restenosis, which occurred in 54% of the patients in the stent group. Conclusion Coronary stenting can significantly decrease the rate of restenosis and reocclusion of total occlusions. As restenosis in the stent group was more diffuse, care should be taken to implant short stents at the site of occlusion. (Eur Heart J 2000; 21: , doi: /euhj ) 2000 The European Society of Cardiology Key Words: Total occlusion, stents, restenosis. See page 1911 for the Editorial comment on this article Introduction Coronary angioplasty has been used successfully to recanalize totally occluded coronary arteries (TIMI grades 0 and 1 flow). The primary success rate has varied Revision submitted 11 May 2000, and accepted 17 May This study was supported in part by AVE who supplied the stents and the Israel Society for Interventional Cardiology who provided the resources for core lab analysis. See Appendix for the list of participating centres and principal investigators. Correspondence: Chaim Lotan MD, FACC, Director of Interventional Cardiology, Hadassah University Hospital, Jerusalem, Israel X/00/ $35.00/0 and in different series ranges from 50 73% [1 7]. With recent advances in technology [8,9], and increased operator experience, results have improved and angioplasty has been used even in chronic total occlusion with bridging collateral vessels [10]. The initial success rate of opening occlusions is high, but the long-term results have been disappointing with a restenosis rate of about 50 70% [1,4,5,11,12], with a considerable percentage of reocclusion. This is associated with a significant increase in the rate of clinical events and the need for revascularization, compared with angioplasty in subtotal stenoses [12]. Preliminary data and several recently published randomized trials have shown a significant decrease in restenosis rates after using the Palmaz Schatz or WIKTOR stents for chronic total occlusion [13 18]. Nevertheless, the role of other stent 2000 The European Society of Cardiology

2 STOP study 1961 types as well as the pattern of restenosis and the need for routine stenting, has not been established. As provisional coronary stenting is well accepted for non-optimal results, we elected to assess whether coronary stenting offers an additional benefit after successful balloon angioplasty for chronic total occlusion. Therefore, this prospective randomized study compared optimal balloon angioplasty alone to angioplasty plus stenting using the AVE microstent. Methods The STOP study is a prospective multicentre randomized study, of balloon angioplasty vs balloon angioplasty plus coronary stenting in patients undergoing angioplasty for totally occluded native coronary arteries. The primary hypothesis was that in patients undergoing balloon angioplasty of totally occluded coronary arteries the long-term outcome of an optimal result could be improved by the implantation of a stent. This should decrease the rate of restenosis and reocclusion by the 6-month angiogram. The population base consisted of patients hospitalized for stable/unstable angina or after myocardial infarction where the infarct related artery was occluded for more than 10 days and in whom there was an indication for balloon angioplasty. Inclusion and exclusion criteria Patients were eligible for inclusion if they were aged 21 or over, had a totally occluded native coronary artery, had successful balloon angioplasty with acceptable results (without the use of stents) and had a reference diameter of >2 75 mm. Eligible patients were included if they were able and willing to undergo repeat angiography at 6 months and to sign a consent form. Patients were excluded from the study if the angioplasty failed or they needed a stent for suboptimal angioplasty. Interventional procedure and randomization The selection of equipment for the angioplasty procedure was left to the discretion of the investigator. Provisional coronary stenting was recommended in cases of residual stenosis with a quantitative coronary angiography of >30%, dissection type C (Dorros classification) [19] and early recoil (defined as a loss of 50% of the gain within 10 min). These patients were excluded from the study. As the intention of the study was to assess the additional value of coronary stenting, randomization was performed only after successful angioplasty (when the operator was ready to leave the room), using consecutively numbered sealed envelopes. The stent used was the AVE microstent available in lengths of 18, 24, 30 and 39 mm, and operators were free to choose the length needed to cover the lesion. At the time of the study, high pressure inflation of a non-compliant balloon was recommended after stent implantation. Post-angioplasty, all patients received aspirin and heparin for 24 h. Ticlodipine (250 mg. day 1 )], was added for 1 month in the stented patients. Number of patients The sample size estimation was based on an expected restenosis rate of 60% after angioplasty as shown in previous studies [5,11,15,16]. In order to demonstrate a 35% reduction in restenosis with stents, approximately 50 patients are required in each treatment arm (α=0 05, β=0 20, one tailed). Moreover, as previously mentioned, if minimal lumen diameter was considered as a continuous variable, this would increase the power of the study and compensate for losses of patients at follow-up. End-points The primary end-point was defined as the reduction of restenosis/reocclusion at the 6 month angiogram. Secondary end-points included the assessment of procedural success and complications, as well as the assessment of major adverse cardiac events and the need for revascularization during the 6 month follow-up. Clinical follow-up Patients were scheduled for clinical visits at 1, 3 and 6 months following the initial procedure. A stress test was recommended at 1 and 6 months. Angiographic analysis Quantitative coronary angiography (QCA) was performed in a central core laboratory (Hadassah Hospital) by two experienced independent investigators. Measurements were performed using mechanical calipers in a method previously validated and reported [20,21]. In short, using a single end-diastolic frame and by means of a high magnification projector, the edge of the contrast medium filled guiding catheter was used as a scaling device, in a view best showing the lesion. The proximal lumen diameter, the final lumen diameter and the occlusion length (when possible) were measured at the end of the procedure. The interpolated reference luminal diameter was determined as the mean of proximal and distal diameters. All measurements were performed after intracoronary injection of nitroglycerin. All lesions were assessed in at least two orthogonal views and the projection showing the smallest diameter was used for

3 1962 C. Lotan et al. Table 1 Baseline and clinical characteristics of patients PTCA group (n=48) Stent group (n=48) Age Gender Males Females 8 7 Risk Factors Hypertension 20 (41 7%) 22 (45 8%) Smoking 22 (45 8%) 21 (43 8%) Diabetes 12 (25%) 12 (25 0%) Dyslipidaemia 18 (37 5%) 19 (39 6%) Family history 9 (18 8%) 11 (22 9%) History of previous MI 34 (70 8%) 28 (58 3%) History of previous PTCA to target artery 8 (16 7%) 4 (8 3%) Angina pectoris 32 (66 7%) 33 (68 8%) Data are presented as mean SD, or number (%) of patients. quantitative coronary angiography analysis, and views with the least foreshortening were used for measuring the length of the occlusion. A similar method of measurement was applied for the follow-up angiography. Using the same final projections of the first procedure, measurements were taken after intracoronary nitroglycerin injection. Angiographic restenosis was defined as 50% stenosis relative to the reference lumen diameter on follow-up angiography. To better evaluate the nature of the restenotic process, we defined a restenosis as focal if the segment length was <10 mm, and as diffuse if the stenotic segment was more than 10 mm in length. Other angiographic parameters which were evaluated included total vs functional obstruction (TIMI grade 0 vs grade 1 flow), the presence or absence of visible thrombus (manifested as a filling defect), and collateral flow (either antegrade by bridging collaterals, or retrograde from other vessels.) Statistical analysis Data are presented as mean SD. Categorical variables were compared using the Fisher exact test or chisquared, and continuous variables were compared by the Student t-test. Multivariate analysis using logistic regression was performed to assess categorical and continuous variables. A P value of <0 05 was considered significant. Statistical analysis was performed using the SPSS version 8 for Windows. Results During the study period between October 1996 and September 1997, PTCA for totally occluded coronary arteries was attempted in 420 patients in 11 centres in Israel. Most of the patients were excluded from the study because they had undergone provisional stenting for unsatisfactory results, as mentioned in the methods section. The final study group consisted of 96 patients who fulfilled the inclusion/exclusion criteria and were willing to sign a consent form. Their mean age was and 81 were males. Forty-eight patients were assigned to the PTCA-only group and 48 to the stent group. Baseline and clinical characteristics are summarized in Table 1. Although a history of previous myocardial infarction and previous PTCA to the target artery were more prevalent in the PTCA group, these differences were not statistically significant. The median age of the occlusion time for the PTCA group was 101 days compared to 122 days for the stent group (P=ns). Other clinical characteristics were similar in the two study groups. Angiographic features The angiographic features are shown in Table 2. Major baseline variables were similar in each group. TIMI 0 flow was observed in most of the patients (70 8% in the PTCA group and 75% in the stent group). Thrombus was evident in 17/96 (17 7%) patients, with a similar distribution in both groups. Final balloon size was larger in the stent group, (3 08 mm vs 2 98 mm P=0 05), as was mean inflation time (11 9 min vs 9 6 min respectively). These differences were attributed to the high pressure deployed at the time of the study. Stent deployment was successful in all patients. The average stent size was 3 1 mm and the average length was mm. Four patients (8 3%) needed more than one stent to cover the entire lesion. Angiographic follow-up was performed in 69 patients (72%). A summary of the quantitative angiographic data from the initial and follow-up angiogram is shown in Table 3. The reference lumen diameter was similar in

4 STOP study 1963 Table 2 Angiographic and procdural variables 100 PTCA (n=48) Stent (n=48) Artery involved LAD/D 18 (35 4%) 21 (43 7%) LCX/M 15 (31 3%) 11 (23%) Right coronary artery 15 (31 3%) 16 (35 3%) TIMI grade Grade 0 34 (70 8%) 36 (75%) Occlusion morphology Retrograde flow 21 (43 8%) 30 (62 5%) Thrombus 10 (20 8%) 7 (14 6%) Estimated lesion length* (mm) Implantation parameters Max. balloon size (mm) ** Max. balloon/artery ratio Maximal pressure (atm.) ** Inflation time (min) ** *When antegrade or retrograde flow was apparent. **P<0 05. LAD/D=left anterior descending or main diagonal; CX/M= circumflex artery or marginal; RCA=right coronary artery. Data are presented as mean+sd, or number (%) of patients. each group as was the final minimal lumen diameter after PTCA (2 41 mm in the PTCA group vs 2 45 mm in the stent group before randomization). After stenting, the minimal lumen diameter in the stent group increased to 3 11 mm, accounting for an additional gain of 0 66 mm in this group. The angiogram at the 6 months follow-up revealed a significant decrease in minimal lumen diameters in both groups compared with the post-procedural results, but a significantly larger minimal lumen diameter in the stent group (1 63 mm in the stent group vs 1 01 mm in the PTCA group; P=0 034). This result was obtained despite a somewhat larger late luminal loss in the stent group (1 53 vs 1 34 mm; P=ns). These finding are Cumulative % of patients Follow-up Postprocedure MLD (mm) Figure 1 Cumulative distribution of post-procedural and 6-month follow-up mean luminal diameter (mm) in the two treatment groups. The post-procedural acute gain in the stent group was well maintained throughout the follow-up period.....=ptca; =stent. shown in Fig. 1, which demonstrates the cumulative distribution of pre- and post-procedural minimal lumen diameters. The post procedural acute gain achieved in the stent group was well maintained throughout the follow-up period. Similarly, significant differences were also found in the mean diameter stenosis (54% in the stent group vs 71 6% in the PTCA group, (P=0 039), with reocclusion seen in 7 9% in the stent group compared with 16 1% in the PTCA group (P=ns). Categorical restenosis, (defined as 50% diameter stenosis), was observed in 42 1% in the stent group vs 70 9% in the PTCA group (P=0 032). When analysing patients with angiographic restenosis only (excluding total reocclusion), restenosis was focal (<10 mm length) in 15/17 (88%) in the PTCA arm compared to 6/13 (46%) in the stent arm (P<0 05). Thus, 54% of the patients in the stent arm had diffuse instent restenosis. Moreover, restenosis after PTCA occurred invariably at the point of previous total obstruction. Table 3 Quantative angiographic data PTCA (n=48) Stent (n=48) P Pre-PTCA Reference diameter (mm) Post-PTCA MLD (mm) Post-stenting Final MLD (mm) Acute gain (due to stent) Final % stenosis Follow-up (6 month) (n=69) (n=31) (n=38) Reference diameter (mm) MLD (mm) % Stenosis Total occlusion 5 (16 1%) 3 (7 9%) ns Restenosis ( 50%) 22 (70 9%) 16 (42 1%) Late loss (mm) ns Loss index ns

5 1964 C. Lotan et al. Table 4 Clinical events during follow-up Discussion PTCA Logistic regression analysis of several baseline and angiographic variables revealed that the final minimal lumen diameter at the end of the initial procedure (either after balloon alone or after stenting) was the only significant predictor for minimal lumen diameter at the 6 month follow-up (odds ratio 9 9, P=0 014). When minimal lumen diameter at follow-up was categorized as either <1 5 mmor 1 5 mm, we found the assignment to the stent group to be to be the most powerful predictor for restenosis (relative risk 4 05, 95% CI , P=0 04). To further assess whether coronary stenting offers an additional benefit if optimal results were achieved after PTCA, we analysed minimal lumen diameter at follow-up as a function of the results after the initial PTCA. Patients were categorized as having post-ptca minimal lumen diameter of either >2 75 mm (n=22), or <2 75 mm (n=47). In the group with post-ptca minimal lumen diameter of >2 75 mm, minimal lumen diameter at follow-up was not statistically different between the two groups ( mm in the PTCA group vs mm in the stent group, P=0 84). In contrast, in patients with post-procedural minimal lumen diameter of <2 75 mm, minimal lumen diameter at follow-up in the PTCA group ( mm), was significantly lower than minimal lumen diameter in the stent group ( mm; P=0 046). Clinical events during follow-up Stent Death 0 0 Recurrent AP 8 (16 7%) 7 (14 5%) MI Q wave 1 (2%) 0 MI/AP 9 (18 7%) 7 (14 5%) Repeat PTCA 19 (39 6%) 10 (21%) CABG 1 (2%) 2 (4%) TLR (PTCA+CABG) 20 (41 6%) 12 (25%) MACE 29 (60 4%) 19 (39 6)* *P= AP=angina pectoris; MI=myocardial infarction; TLR=target lesion revascularization; MACE=major adverse cardiac events. Clinical events during the follow-up period are summarized in Table 4. Two patients in the stent group needed early catheterization due to chest pain: one had an open stent and the other an occluded stent, which was reopened. Three patients in the PTCA group who had early catheterization, had restenosis and needed re-ptca. The need for target lesion revascularization at 6 months in the PTCA group (41%) was higher compared to target lesion revascularization rates in the stent group (25%). Similarly, the major adverse cardiac events at 6 months were higher in the PTCA compared to the stent group (60 4% vs 39 6%, respectively, P=0 067). Several studies have long noted a significant increase in restenosis rates after successful balloon angioplasty for totally occluded coronary arteries [1,4,5,11,12,14]. The MARCATOR (Multicenter American Research trial with cilazapril after Angioplasty to prevent transluminal coronary Obstruction and Restenosis) investigators [12] showed that in patients with total occlusion treated with cilazapril, angioplasty was followed by a restenosis rate of 49%. In view of the initial success of coronary stents to prevent restenosis, the authors raised the need for prospective randomized studies in this subgroup of patients. The data from the two landmark studies, the STRESS and the BENESTENT studies, showed that coronary stenting significantly decreased the rate of restenosis: this improved late clinical outcome and reduced the target lesion revascularization during follow-up [22,23]. The use of coronary stenting in patients undergoing angioplasty for chronic total occlusion has partly been addressed. In a retrospective analysis of registry data, Goldberg et al. [14], reported angiographic restenosis of 20% with only one reocclusion in 59 patients who underwent coronary stenting following angioplasty in chronic total occlusions. Two randomized studies, the SICCO [15] and the GISSOC [16], both using the Palmaz Schatz stent, have suggested that coronary stenting may improve longterm angiographic and clinical results after PTCA for totally occluded arteries. The results of the current study corroborate earlier findings that implantation after achieving optimal balloon angioplasty can improve long-term angiographic and clinical outcome. In the current study, using the AVE microstent, coronary stenting resulted in an additional gain of 0 66 mm (from 2 45 mm after PTCA to 3 11 mm after stenting), similar to the 0 57 mm gain reported by the SICCO investigators [15]. This gain, which was well maintained after 6 months, resulted in a mean minimal lumen diameter of mm in the stent group compared to only mm in the PTCA group (p=0 03). This was translated to a binary restenosis rate of 71 0% in the PTCA group compared with 42 1% after stenting, with a significant reduction in target lesion revascularization. The importance of coronary stenting was noticed mainly in those patients with post-ptca minimal lumen diameters of 2 75 mm, while in those with minimal lumen diameter of >2 75 mm after PTCA, follow-up minimal lumen diameter was not statistically different between the two treatment groups. The relatively high restenosis rates in the PTCA group (70 9%), observed in this study, are comparable with the 68% stenosis observed in the GISSOC and the 74% in the SICCO trials. It is of interest to note that in all these randomized studies the restenosis rates appear to be higher than reports published earlier [4,11,12,14]. Possible explanations might be related to the different nature of these studies (well controlled prospective randomized

6 STOP study 1965 studies compared to retrospective observational studies). Alternatively, the study design which requires an optimal result after PTCA (before randomization), could have been related to longer inflation times (9 6 min compared to 3 min in BENESTENT study) with more possible damage to the vessel wall. The restenosis rate observed in the stent group (42 1%) was significantly higher than the restenosis rate after dilating a subtotal stenosis. It was comparable, although slightly higher than that observed in the GIS- SOC and SICCO studies (32% restenosis). These changes in restenosis rates might be related to a different stent length or design. As the AVE microstent used in the current study was available in 18, 24, 30 and 39 mm, the mean stent length used in this study (21 97 mm) was significantly longer than in the previous studies a finding that has been related to increased risk for restenosis [24,25]. Furthermore, tissue prolapse and less than optimal coverage of this particular stent might also play a role in the higher restenosis rate, especially in lesions with chronic total occlusion that have the highest plaque burden. Possible advantages of coronary stenting The high restenosis rates observed after angioplasty for total occlusion has several potential explanations. Angioplasty of totally occluded vessels carries the risk of creating false channels and dissections in an artery with an abnormal endothelium and this may impair blood flow and increase the likelihood of thrombosis and reocclusion [12,26]. Additionally, in the setting of decreased flow due to scarred myocardial tissue and decreased oxygen demand, early recoil of an undefiled artery is more likely to occur. Thus, in view of this set-up, it is easy to understand why coronary stenting with its inherent scaffolding properties may prevent early recoil and by sealing dissections, improve blood flow a major determinant in the pathophysiology of restenosis. An interesting finding relates to the nature of restenosis in the two arms. Restenosis in the PTCA group was focal (<10 mm) in 88% of patients, mostly at the occlusion site, while diffuse in-stent restenosis occurred in 54% of patients in the stent group. This observation is of great clinical importance. As several studies have shown a direct relationship between stent length and an increased risk of restenosis [24,25], some operators prefer to implant short stents to cover only the most severe lesion an approach that has been called spot stenting [27]. This approach might be problematic in cases of total occlusion where it is difficult to assess lesion length. This has led to the implantation of longer stents in an attempt to bridge the occlusion point with the site of the retrograde flow. The finding that restenosis occurs primarily at the occlusion site and is mostly focal, will guide implantation of short stents at the site of occlusion. Study limitations The intention of the study was to assess the additional benefit of coronary stenting after achieving optimal results with balloon angioplasty. Thus, a significant number of patients, who had provisional stenting for suboptimal results were excluded from the study. It is difficult to predict the rate of restenosis in this group. On the one hand, several studies in patients with subtotal stenosis have shown an increase in restenosis rates after stenting for acute complications or suboptimal results. On the other hand, attempts to optimize the results may have contributed to higher restenosis in the PTCA group, as previously discussed. The role of the optimal anticoagulation protocol after PTCA or stenting after angioplasty for total occlusion is not well established. In contrast with previously reported studies, which have used warfarin in the stent group [13,23,24], only ticlopidine was added for 1 month in the stented patients. Although it is impossible to completely rule out an effect of a different antiplatelet therapy on long-term restenosis, other studies using ticlopidine or IIb/IIIa inhibitors, have not shown favourable effects of these agents on restenosis [28,29]. The sample size of this study was calculated according to the primary end-point, but it is too small for subgroup analysis. Thus, this study is not powered to detect other possible predictors for restenosis such as time since occlusion or other clinical or angiographic findings. Also, as the number of clinical events (other than target vessel revascularization) was low, it was difficult to detect potential differences with regards to major clinical events. Conclusions Coronary stenting can significantly decrease the rate of restenosis and reocclusion after optimal PTCA in chronic total occlusions and thus decrease the need for target vessel revascularization. It appears that using long stents for total coverage of the entire gap may increase restenosis rates in the stented group. Since restenosis was mostly found at the exact site of total occlusion, it seems that perhaps short/spot stenting at this site may decrease restenosis without adding the risk of diffuse instent restenosis in long stents. We are grateful to all the principal investigators and nurse coordinators in all the centres, Mrs Sigal Shafran-Tiqwa, RN and Mrs Mibi Solomon, RN, the study coordinators, for their dedicated work in data collection and finally to Dr Bella Adler, PhD for the statistical analysis. References [1] Serruys PW, Umans V, Heyndrickx GR et al. Elective PTCA of totally occluded coronary arteries not associated with acute myocardial infarction: short-term and long-term results. Eur Heart J 1985; 6: 2 12.

7 1966 C. Lotan et al. [2] Holmes DR Jr, Vlietstra RE, Reeder GS et al. Angioplasty in total coronary artery occlusion. J Am Coll Cardiol 1984; 3: [3] Meier B. Total coronary occlusion: a different animal? J Am Coll Cardiol 1991; 17: 50B 7B. [4] Safian RD, McCabe C, Sipperly ME, McKay RG, Baim DS. Initial success and long-term follow-up of percutaneous transluminal coronary angioplasty in chronic total occlusions versus conventional stenoses. Am J Cardiol 1988; 61: 23G 8G. [5] Bell MR, Berger PB, Bresnahan JF, Reeder GS, Bailey KR, Holmes DR Jr. Initial and long-term outcome of 354 patients following coronary balloon angioplasty of total coronary artery occlusions. Circulation 1992; 85: [6] Stone GW, Rutherford BD, McConahay DR et al. Procedural outcome of angioplasty for total coronary artery occlusion: an analysis of 971 lesions in 905 patients. J Am Coll Cardiol 1990; 15: [7] Maiello L, Colombo A, Gianrossi R et al. Coronary angioplasty of chronic occlusions: factors predictive of procedural success. Am Heart J 1992; 124: [8] Corcos T, Guerin Y, Favereau X. Stiff hydrophilic glidewire: a useful tool for percutaneous insertion of intraaortic balloon counterpulsation catheters. Catheter Cardiovasc Interv 1999; 46: 497. [9] Hamburger JN, Serruys PW, Scabra-Gomes R et al. Recanalization of total coronary occlusion using a laser guidewire (the European TOTAL Surveillance Study). Am J Cardiol 1997; 80: [10] Kinoshita I, Kathoh O, Nariyama J et al. Coronary angioplasty of chronic total occlusion with bridging collateral vessels. Immediate and follow up outcome from a large single center experience. J Am Coll Cardiol 1995; 26: [11] Violaris AG, Melkert R, Serruys PW. Long-term luminal renarrowing after successful elective coronary angioplasty of total occlusions: a quantitative angiographic analysis. Circulation 1994; 91: [12] Berger PB, Holmes DR, Ohman EM et al. Restenosis, reocclusion and adverse cardiovascular events after successful balloon angioplasty of occluded Versus Non-occluded Coronary Arteries. J Am Coll Cardiol 1996; 27: 1 7. [13] Medina A, Melian F, Suarez de Lezo J et al. Effectiveness of coronary stenting for the treatment of chronic total occlusion in angina pectoris. Am J Cardiol 1994; 73: [14] Goldberg S, Colombo A, Maiello L, Borrione M, Finci L, Almagor Y. Intracoronary stent insertion after balloon angioplasty of chronic total occlusions. J Am Coll Cardiol 1995; 26: [15] Sirnes PA, Golf S, Myreng Y et al. Stenting in chronic coronary occlusion (SICCO): a randomized, controlled trial of adding stent implantation after successful angioplasty. J Am Coll Cardiol 1996; 28: [16] Rubartelli P, Niccoli L, Verna E et al. for the Gruppo Italiano di Studio sullo Stent Occlusion: Coronariche (GISSOC). Stent implantation versus balloon angioplasty in chronic coronary occlusions: results from the GISSOC Trial. J Am Coll Cardiol 1998; 32: [17] Buller CE, Dzavik V, Carere RG et al. Primary stenting versus balloon angioplasty in occluded coronary arteries: the Total Occlusion Study of Canada (TOSCA). Circulation 1999; 100: [18] Hoher M, Wohrle J, Grebe OC et al. A randomized trial of elective stenting after balloon recanalization of chronic total occlusions. J Am Coll Cardiol 1999; 34: [19] Dorros G, Cowley MJ, Simpson J et al. Percutaneous transluminal coronary angioplasty: report of complications from the National Heart, Lung, and Blood Institute PTCA Registry. Circulation 1983; 67: [20] Rozenman Y, Gilon D, Welber S et al. Influence of coronary angioplasty on the progression of coronary atherosclerosis. Am J Cardiol 1995; 76: [21] Rozenman Y, Sapoznikov D, Mosseri M et al. Long term angiographic follow up of coronary balloon angioplasty in patients with Diabetes Mellitus. A clue to the explanation of the BARI study. J Am Coll Cardiol 1997; 30: [22] Fischman DL, Leon MB, Baim DS et al. A randomized comparison of coronary-stent placement and balloon angioplasty in the treatment of coronary artery disease. N Engl J Med 1994; 331: [23] Serruys PW, De Jaegere P, Kiemeneij F et al. A comparison of balloon-expandable-stent implantation with balloon angioplasty in patients with coronary artery disease. N. Engl J Med 1994; 331: [24] Bauters C, Hubert E, Prat A et al. Predictors of restenosis after coronary stent implantation. J Am Coll Cardiol 1998; 31: [25] Kornowski R, Mehran R, Hong MK et al. Procedural results and late clinical outcome after placement of three or more stents in single coronary lesions. Circulation 1998; 97: [26] Sanborn TA. Recanalization of arterial occlusions: pathologic basic and contributing factors. J Am Coll Cardiol 1989; 13: [27] De Gregorio J, Colombo A. Spot stenting with IVUS guidance for the treatment of long lesions and small vessels. Intl J Cardiovasc Interven 1998; 1: [28] Schomig A, Neumann FJ, Kastrati A et al. A randomized comparison of antiplatelet and anticoagulant therapy after the placement of coronary artery stents. N Engl J Med 1996; 334: [29] Topol EJ, Ferguson JJ, Weisman HF et al. Long-term protection from myocardial ischemic events in a randomized trial of brief integrin beta3 blockade with percutaneous coronary intervention. EPIC Investigator Group. Evaluation of Platelet IIb/IIIa Inhibition for Prevention of Ischemic Complication. JAMA 1997; 278: Appendix List of centres and principal investigators Assaf Harofe Hendler A, Bikur-Cholim Ben-Chorin J, Hadassah Rozenman Y, Ha emek Turgeman Y, Hillel- Jaffe Peled B, Kaplan Swissa M, Meir Baxt A, Rambam Beyar R, Sha arei Zedek Almagor Y, Tel-Ha shomer Di-Segni E, Woolfson Tamari I.