Comparison of angiographically guided direct stenting technique with direct stenting and optimal balloon angioplasty guided with intravascular ultrasound. The multicenter, randomized trial results

Comparison of angiographically guided direct stenting technique with direct stenting and optimal balloon angioplasty guided with intravascular ultrasound. The multicenter, randomized trial results Robert J. Gil, MD, PhD, FESC,a Tomasz Pawyyowski, MD, PhD,b Dariusz Dudek, MD, PhD,b Grzegorz Horszczaruk, MD,b Krzysztof Zmudka, MD, PhD,b Maciej Lesiak, MD, PhD,b Adam Witkowski, MD, PhD,b Andrzej Ochayya, MD, PhD,b and Jacek Kubica, MD, PhD,a on behalf of Investigators of Direct Stenting vs Optimal Angioplasty Trial (DIPOL) Warsaw, Poland

Aim The primary objective of the trial was to test the hypothesis that intravascular ultrasound (IVUS) guidance for coronary angioplasty is superior to the quantitative coronary angiography approach both during stenting and plain balloon angioplasty. Methods Two hundred fifty-nine patients (70 females, 189 males; mean age, 54 ± 9) were enrolled into our study. They were randomized into 3 groups: group 1—direct stenting guided with quantitative coronary angiography; group 2—direct stenting guided with IVUS; group 3—optimal balloon angioplasty guided with IVUS. At 6-month follow-up, we recorded evidence of major adverse cardiac events (death, myocardial infarction, repeat coronary revascularization). Results Procedural success was achieved in 95% of cases. At 6-month follow-up, the evidence of composite end point (major adverse cardiac events) was 16.2%, 7.3%, and 21.8% in groups 1, 2, and 3, respectively (P b .05). Use of IVUS led to 55% reduction of the primary end point between group 1 and group 2. The same when compared to the balloon angioplasty group assumed 66% reduction. Conclusion

Ultrasound guidance for direct stenting is the most effective for long-term outcome in comparison with other strategies. It is very likely that bigger lumen gain due to appropriate device sizing is responsible for better outcome. Direct stenting guided with IVUS might be an alternative option for patients requiring coronary revascularization and have contraindications to prolonged antiplatet therapy as is mandatory for drug-eluting stent implantation. (Am Heart J 2007;154:669-75.)

The most popular techniques for percutaneous coronary interventions (PCI) are coronary stenting and balloon angioplasty. The question of optimal guidance for them was open for several years. Neither quantitative coronary angiography (QCA) nor intravascular ultrasound (IVUS) nor functional assessment (ie, Doppler or pressure wires) has been proven as an ideal tool for interventional cardiologists. On the one hand, there is a conviction that physiologic guidance might provide similar results like angiography,1,2 but on the other hand, there is some evidences that IVUS may be superior to QCA.3,4 There is no consensus regarding superiority of one kind of guidance for PCI over another, especially using bare From the aMedical Research Center, Polish Academy of Science and bDepartment of Invasive Cardiology, Central Clinic Hospital, Warsaw, Poland. All investigators are listed in Appendix A. Submitted December 10, 2006; accepted June 17, 2007. Reprint requests: Robert J. Gil, MD, PhD, FESC, Department of Invasive Cardiology, Central Clinic Hospital, Woloska 137 str, 02-507 Warsaw, Poland. E-mail: [email protected] 0002-8703/$ - see front matter © 2007, Mosby, Inc. All rights reserved. doi:10.1016/j.ahj.2007.06.017

metal stents. Introduction of drug-eluting stents (DESs) gave hope that the problem of restenosis would be completely eliminated; however, reality showed that it is a dream.5 At the time when we planned our trial, the question of optimal guidance for PCI was open. Therefore, we decided to conduct a multicenter, randomized trial to test different techniques for guidance of percutaneous interventions.

Methods Study design, objectives, and end points The Direct Stenting versus Optimal Angioplasty (DIPOL) trial was designed in 2000 as a prospective, randomized, multicenter trial to compare 4 different strategies of coronary angioplasty. The 7 most experienced Polish interventional centers were included in the trial. Sequential envelope system located in each center was used for blinded randomization. The patients were randomly divided into 4 groups: group 1— direct stenting technique guided with QCA (DS-QCA), group 2—direct stenting guided with IVUS (DS-IVUS), group 3—balloon angioplasty with IVUS guidance (POBAIVUS), and group 4—balloon angioplasty guided with fractional flow reserve (POBA-FFR).

670 Gil et al

Table I. Inclusion and exclusion criteria for DIPOL trial Inclusion criteria • Stable angina pectoris • Age between 18 and 70 y • 1 or 2 de novo vessel disease • Vessel RD bigger than 2.75 mm • Lesion length up to 25 mm Exclusion criteria • Recent myocardial infarction or unstable angina • Large calcifications seen on angiography • Presence of large (N2 mm in diameter) side branch in segment to be stented • Chronic total occlusion

The primary objective of the trial was to test the hypothesis that IVUS guidance for PCI is superior to the QCA and FFR approaches both during the stenting and plain balloon angioplasty. The primary end point of the study was a composite end point calculated as major adverse cardiac events (MACE—death, myocardial infarction, and repeat coronary revascularization [RCR]) that occurred at 6 months. Repeat coronary revascularization was assumed as target lesion revascularization (TLR), target vessel revascularization (TVR), or any revascularization due to atherosclerosis progression independently by angioplasty or bypass surgery. Myocardial infarction was defined as the presence of a new Q wave on electrocardiogram in at least 2 contiguous leads and a significant elevation of creatine kinase MB fraction. In the absence of pathologic Q waves, the diagnosis of myocardial infarction was based on an increase of the MB fraction of creatine kinase to more than 3 times than the upper level of limit of the reference range. Target lesion revascularization was defined as coronary revascularization of initially treated lesion plus 5-mm segments of accompanying lesion in response to recurrence of angina or documented ischemia on noninvasive tests (clinically driven revascularization). Target vessel revascularization was defined as any (percutaneous or surgical) revascularization of initially treated vessel in other site than the target lesion. Based on the mentioned criteria, the index of atherosclerosis progression was calculated (atherosclerosis progression index). It was defined as the difference of RCR and TLR. The secondary end point was the achievement of the postprocedural target criteria in each group. The study protocol was approved by ethics committees of each investigational center, and all patients provided written informed consent.

Patients selection Two hundred fifty-nine consecutive patients who fulfilled all inclusion criteria and none of exclusion criteria (Table I) entered the study. The inclusion criteria covered stable angina pectoris, age between 18 and 70 years, 1 or 2 “de novo” vessel disease, vessel reference diameter (RD) greater than 2.75 mm, and lesion length up to 25 mm. The exclusion criteria were defined as recent myocardial infarction or unstable angina, significant left main stenosis large calcifications seen on angiography, presence of large (N2 mm in diameter) side branch in segment to be stented, and chronic total occlusion.

American Heart Journal October 2007

Interventional procedures and concomitant medications All procedures were performed in a standard way via femoral (96.5%) or brachial access (3.5%) using guiding catheters of 6F and 7F diameter. After the insertion of the arterial sheath, every patient received unfractioned heparin (70-100 IU/kg). Additional bolus was given to maintain an activated clotting time of N200 seconds. All patients received acetylsalicylic acid (300 mg per 24 hours) and ticlopidine (500 mg per 24 hours) 72 hours before PCI. Use of glycoprotein IIb/IIIa inhibitors was left to the operator's discretion. Per protocol, there was an intention to use only bare metal stents. In group 1, stent size was selected by the operator based on balloon/artery ratio, which should be between 1.1 and 1.3. The calculations were based on online QCA. The length of the stent was to be 2 to 4 mm longer than the lesion. The procedure was assessed as successful (target criteria) when the residual percent diameter stenosis was lower than 10% with angiographic signs of full apposition of the stent. In all cases of group 2, preprocedural ultrasound imaging was obligatory. In this group, stent was selected to achieve its size close to media-to-media diameter on IVUS examination. Stent length was selected in a similar way as in group 1. The procedural target criteria were minimal lumen area (LA) larger than 80% of average of proximal and distal reference LAs or minimal stent area bigger than 7.5 mm2 with full stent apposition. The group 3 procedure protocol assumed that balloon size should be selected closely due to media-to-media diameter on IVUS (balloon/media ratio should be 0.9:1.0). The angioplasty procedure was completed when there were no significant (disturbing coronary flow) dissection and minimal lumen larger than 65% of average proximal and distal reference areas or minimal LA bigger than 6.0 mm2. Coronary stenting (ie, crossover) was allowed when the above-mentioned criteria were not met. The stent size selection was based on the same manner as in group 2 and was completed successfully when procedural criteria adopted from group 2 were achieved. In group 4, the procedure was guided by FFR. The angioplasty was continued until an FFR value greater than 0.9 was reached and no significant dissections on angiography was observed. In opposite fashion, coronary stenting was performed until an FFR value of 0.94 was reached.

Quantitative coronary angiography and IVUS protocols All coronary angiograms were recorded after intracoronary administration of 200 μg of nitroglycerin. Two orthogonal views were chosen to visualize the target lesion. Using an online quantitative system, available at participating centers (ACA-DCI; Philips, Eindhoven, The Netherlands) or CASS II (Pie Data Medical, Maastricht, Netherlands; edge detection algorithm), the minimal lumen diameter (MLD), average RD, and percentage of diameter stenosis (%DS) were calculated before and after procedure. The difference between post- and preprocedural MLD was calculated as acute gain (in mm). The IVUS examinations were performed before intervention and after the last inflation in groups 2 and 3 (also in case of coronary stenting). After angioplasty wire placement in the

American Heart Journal Volume 154, Number 4

target vessel, the ICUS catheter (Avanar, Endosonics Co, Rancho Cordova, CA) was advanced more than 3 cm distal to the lesion and was pulled back at a speed of 0.5 mm/s until the guiding catheter was reached.

Intravascular ultrasound core laboratory analysis All IVUS recordings were sent to the trial's core laboratory (Szczecin, Poland). Offline quantitative IVUS analysis was performed by 2 independent investigators (T.P., R.J.G.) who were unaware of the QCA measurements. Data from ICUS examinations were presented as means and included minimal LA, total vessel area boarded by external elastic membrane, and plaque area. The results were shown for the stented segment and were obtained from serial crosssections measurements at every 1 mm.

Follow-up assessment Clinical follow-up was obtained at 6 months after initial procedure based on visit in the outpatient clinics. The data regarding composite end point were collected. The clinical course was examined for MACE, such as death, myocardial infarction, and any RCR, assumed as target lesion, vessel revascularization, or any revascularization due to atherosclerosis progression. Coronary angiography at follow-up was not imposed by the protocol and was left to the discretion of the operator. The investigators were asked to perform angiography in case of recurrence of angina, positive results of stress tests, and history of acute coronary syndrome. Restenosis was detected when percentage of lumen stenosis (%DS) was larger than 50% at follow-up angiography.

Statistical analysis The assumptions used for power calculation were 9% of 6 months MACE in group 2 (DS-IVUS) versus 25% of 6 months MACE in either group 1 or group 3. The assumptions were based on the estimation of unpublished data regarding IVUS use during direct stenting in primary investigator center and the mean rate of MACE different studies with direct stenting and plain balloon angioplasty. Using 2-sided for differences in independent binomial proportions with α level of .05, we calculated the total number of 172 patients randomized into 2 groups to achieve a power of 80% to detect the presented difference. The tests were performed independently for comparisons of group 1 versus 2 and group 2 versus 3. Continuous variables are presented as means and SD. The categorical variables were presented as number of percentage (%). Continuous variables were estimated using the unpaired Student t test. The χ2 test or Fisher exact test were used to compare categorical variables. Multivariate analysis was performed using the stepwise logistic regression method. P values b.05 was considered significant. Statistical tests were performed with STATISTICA Package, version 5.5 for Windows (StatSoft Inc).

Results Patient and procedure characteristics From September 2000 to December 2002, 279 patients were enrolled into the DIPOL trial. Group 1 consisted of 80 patients (mean age, 54 ± 8 years); group 2, 83 patients (mean age, 56 ± 8 years); and group 3,

Gil et al 671

Table II. Baseline clinical and angiographic characteristics of patients DS-QCA (n = 80) Age (y) 54 ± 8 Female (%) 27 Smoking history (%) 52 Previous infarction (%) 40 Diabetes (%) 11 Dyslipidemia (%) 40 Ejection fraction (%) 48 ± 10 Vessel treated and ACC/AHA class (n) LAD 37 Cx 19 RCA 24 Class A 49 Class B1 21 Class B2 10 Class C 0 Mean balloon 3.36 ± 0.39 diameter (mm) Stent length (mm) 16.5 ± 3.1 Total inflation time (s) 21 ± 6 Maximal inflation 14.3 ± 2.1 pressure (atm) Balloon predilatation (%) 0.0 BA/RD ratio 1.12 ± 0.35 Stent redilatation with 26.2 bigger balloon and higher pressure (%) IIb/IIIa inhibitors use (%) 0

DS-IVUS (n = 83)

POBAIVUS (n = 96)

56 ± 8 29 47 44 10 47 52 ± 9

52 ± 9 25 50 39 13 50 54 ± 8

34 22 27 48 22 13 0 3.74 ± 0.43 *

36 26 34 55 27 14 0 3.14 ± 0.55

15.9 ± 2.9 22 ± 8 13.7 ± 1.2

17.7 ± 4.7 86 ± 18 y 13.5 ± 1.3

1.2 1.29 ± 0.31 * 39.7 *

100 1.38 ± 0.42 28.4

0

11.4

LAD, Left anterior descending artery; Cx, circumflex artery; RCA, right coronary artery; BA/RD, balloon/RD ratio. ⁎P b .05 for group 2 versus groups 1 and 3. yP b .05 for group 3 versus groups 1 and 2.

96 patients (mean age, 52 ± 9 years). Group 4 was consisted of patients treated with POBA-FFR. However, after completion of 20 patients in group 4, the randomization was stopped for this arm because of technical problems in the center of the principal investigator, where most of cases were planned. Thus, the final number of patients that underwent analysis was 259. There were no differences between groups in terms of baseline clinical and procedural characteristics (Table II). The biggest angioplasty balloons were used in group 2 as compared with other groups (P b .05). The maximal pressure used for inflation of balloons and stents deployment did not differ significantly among the groups; however, the total inflation time was significantly longer in group 3 (P b .05).

In-hospital phase There were no history of death and only a few cases of nonfatal myocardial infarction in the studied population during the hospitalization time (listed in Table III). The rates of procedural success (the secondary end point) were similar in group 1 (97.5%) and group 2 (96%). The need for balloon predilation was observed only in 1 case of

American Heart Journal October 2007

672 Gil et al

Table III. Evidence of MACE and TVR, TVF in all studied groups during in-hospital phase and at 6 months time point DS-QCA (n = 80) In-hospital phase Death, n (%) Myocardial infarction, n (%) RCR, n (%) 6 mo follow-up Death, n (%) Myocardial infarction, n (%) RCR, n (%) CABG, n (%) PTCA, n (%) Composite MACE, n (%) Stent thrombosis, n (%) TVR, n (%) CABG, n (%) PTCA, n (%) TLR, n (%) CABG, n (%) PTCA, n (%) Nontarget revascularization, n (%) CABG, n (%) PTCA, n (%)

DS-IVUS (n = 83)

0 (0) 1 (1.2)

0 (0) 2 (2.0)

0 (0)

1 (1.2)

1 (1.0)

1 (1.2) 4 (5.0)

1 (1.2) 1 (1.2)

2 (2.0) 3 (3.1)

(10.0) (3.7) (6.2) (16.2) (0) (1.2) (1.2) (0) (7.5) (1.2) (6.2) (1.2)

1 (1.2) 0 (0)

4 2 2 6 0 1 1 0 3 1 2 0

(4.8) (2.4) (2.4) (7.3) y (0) (1.2) (1.2) (0) (3.6) (1.2) (2.4) (0)

0 (0) 0 (0)

DS-QCA (n = 80)

POBA-IVUS (n = 96)

0 (0) 1 (1.2)

8 3 5 13 0 1 1 0 6 1 5 1

Table IV. Quantitative coronary angiography results

16 5 11 21 0 2 2 0 12 2 10 2

(16.6) ⁎ (5.2) (11.4) ⁎ (21.8) ⁎ (0) (2.0) y (2.0) (0) (12.7) ⁎ (2.0) (10.4) ⁎ (2.0)

1 (1.0) 1 (1.0)

CABG, Coronary-artery based grafts; PTCA, percutaneous transluminal coronary angioplasty; TVF, taget vessel failure. ⁎ P b .05 for group 3 versus group 2. yP b .05 for group 2 versus groups 1 and 3.

group 2 (1.2%). Worse results were found in group 3, where only 29.1% of patients reached planned target criteria. In the last group, the crossover to coronary stenting was 70.8% due to suboptimal reason of balloon angioplasty. The main reason (75%) was the presence of long coronary dissection (N15 mm on angiography).

Angiographic results Table IV shows QCA results obtained during primary procedure. There were no differences in reference vessel diameter before procedure among studied groups. In all patients, the MLD increased significantly after PCI, but the biggest lumen gain was achieved in the DS-IVUS group (P b .01). However, the smallest %DS at the end of procedure was found in the DS-IVUS group (P b .05). Quantitative ultrasound results There were no significant differences in lumen and vessel dimensions obtained before procedure in groups 2 and 3. In the DS-IVUS group, the LA increased significantly (P b .01) after stenting procedure (3.69 ± 1.67 to 9.8 ± 2.88 mm2) with significant (P b .05) reduction of plaque burden (from 68.7% ± 15.4% to 46.3% ± 11.2%). In group 3, wherein patients were treated with plain balloon angioplasty, the LA also increased significantly (from

MLD pre (mm) MLD post (mm) Acute gain (mm) RD pre (mm) RD post (mm) %DS pre (mm) %DS post (mm) ⁎

0.95 ± 3.06 ± 2.11 ± 3.19 ± 3.36 ± 70.2 ± 8.9 ±

0.32 0.52 0.43 0.59 0.59 11.4 5.4

DS-IVUS (n = 83) 0.97 ± 3.34 ± 2.37 ± 3.21 ± 3.32 ± 69.7 ± 3.4 ±

0.33 0.55 ⁎ 0.51 ⁎ 0.64 0.70 14.2 2.9 ⁎

POBA-IVUS (n = 96) 1.02 2.94 1.92 3.28 3.26 68.9 9.8

± 0.28 ± 0.62 ± 0.41 ± 0.62 ± 0.48 ± 15.3 ± 10.5

P b .05 for group 2 versus groups 1 and 3.

4.14 ± 2.27 to 6.51 ± 3.15 mm2, respectively, pre vs post; P b .05). Additional analysis has shown that LA has increased significantly from 3.89 ± 1.54 to 9.86 ± 3.49 mm2 in patients of group 3, in whom crossover to stenting was performed.

Clinical and angiographic follow-up During follow-up, all treated patients (n = 259) were screened for clinical events at 6 months time point. Table III refers to all events that occurred during hospitalization and at long-term follow-up. There were no significant differences in the incidence of death (n = 4) among the studied groups during followup. One case of death in group 1 was noncardiac and was associated with lung cancer, and one case in group 2 due to myocardial infarction of nontarget lesion. Two patients of group 3 died during follow-up, one because of chronic leukemia and the other due to acute heart failure (pulmonary edema). There was a nonsignificant trend to higher incidence of myocardial infarction in the DS-QCA group compared with the other groups. At 6 months, the evidence of composite end point (MACE) was 16.2%, 7.3%, and 21.8%, respectively, in groups 1, 2, and 3. Use of IVUS led to 55% reduction of the primary end point between group 1 and group 2. The same when compared with the balloon angioplasty group assumed 66% reduction (Table III; Figures 1 and 2). The potential predictors of MACE were put into the multivariable analysis model covering IVUS guidance, final stent area by IVUS, previous myocardial infarction, location of lesion, and demographic characteristic. The results were presented in Table V. The only predictors of MACE were final stent diameter and IVUS-measured LA and presence of diabetes (P b .05). Target lesion revascularization rate in group 2 (DS-IVUS) was significantly lower (P b .05) than in group 3, and the trend to lower number was observed when compared with group 1. The rate of TLR in the whole group 3 was the highest of all the groups (12.7%), mainly treated with repeated angioplasty. Of note, because of atherosclerosis progression in other (not index) vessels, 4 patients underwent bypass surgery. Table III and Figure 3 present data regarding RCR, TLR, and TVR.

American Heart Journal Volume 154, Number 4

Gil et al 673

Figure 1

Example case of DS-IVUS. A, Preprocedure ultrasound imaging shows minimal lumen area of 3.5 mm2 and minimal vessel diameter of 3.6 mm. The 3.5 × 13-mm stent was chosen and deployed at 14 atm. B, Incomplete expansion of stent (minimum lumen cross-sectional area, 6.2 mm2) despite good angiographic results. C, There are results of redilatation with noncompliant balloon (3.5 × 12 mm; 16 atm)—the minimal stent area increased to 8.1 mm2. At the follow-up, the patient was fine, and no MACE occurred; however, there was no control angiography.

Although the analysis performed in group 3 was based on intention to treat, some differences between balloon angioplasty–alone patients and stented patients in end points were presented in Figure 3 (analysis as treated). Using developed formula of atherosclerosis progression, the atherosclerosis progression index was calculated for all groups as follows: 2.5% in group 1, 3.7% in group 2, and 3.9% in group 3. There was no significant difference between the studied groups. Angiographic follow-up was obtained in 217 patients (83.7% in total, 83.7% in group 1, 87.9% in group 2, and 80.2% in group 3). At 6 months follow-up, the restenosis occurred significantly (P b .01), rarely in group 2 than in group 1 or 3 (26.6%, 9.6%, 20.8% in groups 1, 2, and 3, respectively; P b .05).

Figure 2

Discussion

Evidence of RCR, TLR, and TVR in all treated groups at 6 months follow-up. Group 3 presented all enrolled patients. For the rate of angiographic control, see text.

The DIPOL trial (multicenter, prospective, and randomized) has faced the hypothesis that optimal coronary stenting results can be decreased by using appropriate stenting technique and imaging support. We could hypothesize that late results are related to favorable acute results of the procedures. The trial has proven the high efficacy of ultrasound guidance procedure for coronary stenting. Namely, there was a significant reduction of MACE at 6 months follow-up in the ultrasound-guided direct stenting group. Simply, our results confirmed the well-known formula “bigger is better.”6 The success rate during plain balloon angioplasty was very low. In more than 70% of cases, there was a need for a crossover to stenting. The criteria for suboptimal results of the procedure were adopted from previous studies,7-10 but in none of them was the crossover rate so high even if the protocol was as

aggressive as ours (1.38 ± 0.42 mm for balloon/artery ratio). In addition, the mean values of maximal pressure used to deploy balloons were comparable.8,9 We used the similar aggressive protocol as used Schiele et al10 for balloon angioplasty, but in the mentioned trial, there were no differences in terms of MACE and restenosis at 6 months follow-up even if angioplasty alone and crossover stenting (17.5 vs 12.0%, respectively) were compared, although the trend to higher restenosis rate after balloon angioplasty was observed. One can conclude that the optimalization of POBA is not worth doing, mainly because of low procedural success and restenosis rate comparable to elective stenting. The results of these trials could force the

American Heart Journal October 2007

674 Gil et al

Table V. Multivariable analysis of MACE predictors Variable Diabetes Hypertension Previous MI Smoking history IVUS guidance LAD vs RCA location Final MLD by QCA (mm) Final stent LA by IVUS

Odds ratio 12.6 0.44 0.67 0.54 0.91 0.78 3.65 2.56

Figure 3

95% CI

P

5.4-29.3 0.21-1.11 0.35-1.35 0.25-0.98 0.44-1.88 0.34-1.34 1.79-9.34 1.12-5.01

.003 .25 .71 .70 .11 .29 .024 .016

CI, Confidence interval.

operators to use the strategy of elective stenting with or without guidance and finally to forget about optimal POBA strategy. It is reasonable to speculate that IVUS may improve outcome during percutaneous interventions. First, as we know from comparative studies, angiography can underestimate the real dimensions of the lumen of the vessel.11 Second, IVUS may help to identify vessel calcifications, plaque relation to side branches, and vessel remodeling what subsequently can modify the strategy of the procedure. The IVUS probe used as a device for procedure optimization can push the operator to redilatation, improving final stent lumen and subsequently late outcome as was shown by the CRUISE trial when in 36%, operators were forced to redilate the stent.3 The similar rate (39.7%) was achieved in the presented trial. Planning this trial, we hypothesized that any additional guidance for percutaneous intervention can provide better outcome than angiography guidance alone. When compared with angiography-controlled intervention, DS-IVUS should be better at 6 months follow-up. In the past studies, there were ambiguous results: for example, in the CRUISE trial,3 IVUS-guided stenting led to better late clinical outcome and provided 44% reduction of events at 9 months time point. Similarly, in the TULIP trial, reduction both in MACE and restenosis was observed.4 Our findings (ie, 55% reduction of MACE) can confirm investigators' thesis in both trials that more effective stent expansion may reduce less favorable outcome. On the other hand, OPTICUS and RESIST trials12,13 did not show any superiority of IVUS guidance in this term. However, as mentioned above, we can speculate that the reason for this could be the very low percentage of MUSIC criteria fulfilled in OPTICUS trial. This is in opposite with our trial wherein operators pursed to a goal of procedural end points with very high rate of success (more than 90%). So, we stand that patience of the operator during IVUS-guided procedure leads to better patient's outcome. It is remarkable that angiographic follow-up was performed in over 80% of patients. The trial was

Evidence of MACE and type of revascularization in group 3 divided into nonstented and stented patients at 6 months follow-up (as treated analysis).

not intended to test all patients during follow-up. The restenosis rates within the groups have similar trends as rates of major adverse events in our trial. These findings clearly demonstrate that the problem of restenosis can be directly related to late patient's outcome. The question of less vessel injury during direct stenting was opened for several years.14 Randomized trials did not prove the efficiency of this technique in terms of restenosis rate.15,16 Our results showed that the use ultrasound guidance can improve the late outcome of a patient when direct stenting is used. Compared with stenting with angiographic guidance, those patients received larger stents deployed with higher pressures. This approach should translate into benefit during followup, especially when IVUS is used before the stent implantation such as in this trial. The new “wave” in interventional cardiology, that is, DESs, gives the opportunity to treat coronary artery disease more effectively. The first experience with DESs has shown a significant reduction of major adverse events (mainly due to restenosis) in comparison with bare metal stents.5,17 Especially the TAXUS-IV trial results (7.9% of restenosis) can be compared with our findings in the DS-IVUS group. In addition, the rate of TLR in the mentioned trial and in our study can be comparable (3.0% and 3.6%, respectively). The same can be said regarding the SIRIUS trial,5 in which TLR was 4.1% at 9 months. At this moment, one can raise the question of the well-known Achilles' heel of DES use—late stent thrombosis. Of note, there were no cases with stent thrombosis in the presented trial, and dual antiplatelet therapy was prolonged only for a few weeks. The findings of our trial can provide hope that “old-fashioned” interventional cardiology using bare metal stent, properly implanted, is a still an attractive

American Heart Journal Volume 154, Number 4

option in treatment of patients with coronary artery disease. Practical implementation of the presented trial is related with a subgroup of patients in which cessation of dual antiplatelet therapy is mandatory in the followup (ie, planned noncardiac surgery).

Limitations Although the study was randomized, both the operators and patients were unblinded to type of procedure performed and their immediate results. The follow-up angiography was not performed in all patients. Conclusions Ultrasound guidance for direct stenting is the most effective for long-term outcome in comparison with other strategies. It is very likely that bigger lumen gain due to appropriate device sizing is responsible for better outcome. Direct stenting guided with IVUS might be an alternative option for patients requiring coronary revascularization and have contraindications to prolonged antiplatet therapy as is mandatory for DES implantation.

References 1. Di Mario C, Moses J, Anderson T, et al. Randomized comparison of elective stent implantation and coronary balloon angioplasty guided by online quantitative angiography and intracoronary Doppler. Circulation 2000;102:2938-44. 2. Bech GJ, Pijls NH, De Bruyne B, et al. Usefulness of fractional flow reserve to predict clinical outcome after balloon angioplasty. Circulation 1999;99:883-8. 3. Fitzgerald PJ, Oshima A, Hayase M, et al, for the CRUISE Investigators. Final results of the Can Routine Ultrasound Influence Stent Expansion (CRUISE) study. Circulation 2000;102:523-30. 4. Oemrawsingh PV, Mintz GS, Schalij MJ, et al. Intravascular ultrasound guidance improves angiographic and clinical outcome of stent implantation for long coronary artery stenoses: final results of a randomized comparison with angiographic guidance (TULIP study). Circulation 2003;107:62-7. 5. Moses J, Leon M, Popma J, et al. Sirolimus-eluting stents versus standard stents in patients with stenosis in a native coronary artery. N Engl J Med 2003;349:1315-23. 6. Kuntz R, Gibson C, Nobuyoshi M, et al. Generalized model of restenosis after conventional balloon angioplasty, stenting and directional atherectomy. J Am Coll Cardiol 1993;21:15-25. 7. Abizaid A, Pichard AD, Mintz GS, et al. Acute and long-term results of an IVUS-guided PTCA/provisional stent implantation strategy. Am J Cardiol 1999;84:1381-4. 8. Frey AW, Hodgson JM, Muller C, et al. Ultrasound-guided strategy for provisional stenting with focal balloon combination catheter: results from the randomized Strategy for Intracoronary Ultrasoundguided PTCA and Stenting (SIPS) trial. Circulation 2000;102: 2497-502. 9. Colombo A, De Gregorio J, Moussa I, et al. Intravascular ultrasound guided percutaneous transluminal coronary angioplasty with provi-

Gil et al 675

10.

11.

12.

13.

14. 15.

16.

17.

sional spot stenting for treatment of long coronary lesions. J Am Coll Cardiol 2001;38:1427-33. Schiele F, Meneveau N, Gilard M, et al. Intravascular ultrasoundguided balloon angioplasty compared with stent immediate and 6-month results of the multicenter, randomized Balloon Equivalent to Stent Study (BEST). Circulation 2003;107:545-51. Moussa I, Kobayashi Y, Adamian M, et al. Characteristics of patients with a large discrepancy in coronary artery diameter between quantitative angiography and intravascular ultrasound. Am J Cardiol 2001;88:294-6. Mudra H, di Mario C, de Jaegere P, et al, for the OPTICUS (OPTimization with ICUS to reduce stent restenosis) Study Investigators. Randomized comparison of coronary stent implantation under ultrasound or angiographic guidance to reduce stent restenosis (OPTICUS study). Circulation 2001;104:1343-9. Schiele F, Meneveau N, Vuillemenot A, et al. Impact of intravascular ultrasound guidance in stent deployment on 6-month restenosis rate: a multicenter, randomized study comparing two strategies—with and without intravascular ultrasound guidance. RESIST Study Group. REStenosis after Ivus guided STenting. J Am Coll Cardiol 1998;32: 320-8. Rogers C, Tseng D, Squire J, et al. Balloon—artery interactions during stent placement. Circ Res 1999;84:378-83. Carrie D, Khalife K, Citron B, et al, for the BET (Benefit Evaluation of Direct Coronary Stenting) Study Group. Comparison of direct coronary stenting with and without balloon predilatation in patients with stable angina pectoris. Am J Cardiol 2001;87:693-8. Martınez-Elbal E, Ruiz-Nodar J, Zueco J, et al, for the DISCO Investigators. Direct coronary stenting versus stenting with balloon pre-dilation: immediate and follow-up results of a multicentre, prospective, randomized study the DISCO trial. Eur Heart J 2002;23: 633-40. Stone G, Ellis S, Cox D, et al. A Polymer-based, paclitaxel-eluting stent in patients with coronary artery disease. N Engl J Med 2004;350:221-31.

Appendix A This study was supported by the research grant number 4P05B10019 from the National Committee of Science. We would like to thank all the investigators participating in the DIPOL study. 1. Robert J. Gil, Tomasz Pawlowski, Jarosƚaw Gorący— Catherization Laboratory, Institute of Cardiology, Szczecin; 2. Dariusz Dudek, Jacek Legutko, Krzysztof Żmudka, Andrzej Gackowski—Catherization Laboratory, Collegium Medicum, Cracow; 3. Grzegorz Horszczaruk, Janusz Kochman—Department of Cardiology, Warsaw; 4. Maciej Lesiak, Wƚodzimierz Skorupski—Department of Cardiology, Poznań; 5. Jacek Kubica, Marek Radomski—Department of Cardiology, Bydgoszcz; 6. Adam Witkowski—Institute of Cardiology, Warsaw; 7. Andrzej Ochaƚa—Department of Cardiology, Katowice.