Author + information
- Received January 23, 2001
- Revision received April 27, 2001
- Accepted June 8, 2001
- Published online September 1, 2001.
- Milena Adamian, MD, PhD†,
- Antonio Colombo, MD, FACC∗,* (, )
- Carlo Briguori, MD, PhD∗,
- Takahiro Nishida, MD∗,
- Federica Marsico, MD∗,
- Carlo Di Mario, MD, PhD, FACC∗,
- Remo Albiero, MD∗,
- Issam Moussa, MD† and
- Jeffrey W Moses, MD, FACC†
- ↵*Reprint requests and correspondence: Dr. Antonio Colombo, EMO Centro Cuore Columbus, Via M. Buonarroti 48, 20145 Milan, Italy
The purpose of the study was to determine whether cutting balloon angioplasty (CBA) has advantages over other modalities in treatment of in-stent restenosis (ISR).
Controversies exist regarding optimal treatment for ISR. Recently, CBA emerged as a tool in management of ISR.
A total of 648 lesions treated for ISR were divided into four groups according to the treatment strategy: CBA, rotational atherectomy (ROTA), additional stenting (STENT), and percutaneous transluminal coronary angioplasty (PTCA). Following the matching process, 258 lesions were entered into the analysis.
Baseline clinical and angiographic characteristics were similar among the groups (p = NS). Acute lumen gain was significantly higher in the STENT group (2.12 ± 0.7 mm), whereas in the CBA group the gain was similar to one achieved following ROTA and following PTCA (1.70 ± 0.6 vs. 1.79 ± 0.5 mm and 1.56 ± 0.7 mm, respectively; p = NS). The lumen loss at follow-up was lower for the CBA versus ROTA and versus STENT (0.63 ± 0.6 vs. 1.30 ± 0.8 mm and 1.36 ± 0.8 mm, respectively; p < 0.0001), yielding a lower recurrent restenosis rate (20% vs. 35.9% and 41.4%, respectively; p < 0.05). By multivariate analysis, CBA (odds ratio [OR] = 0.17; confidence interval [CI], 0.06 to 0.51; p = 0.001) and diffuse restenosis type at baseline (OR = 2.07; CI, 1.15 to 3.71; p = 0.02) were identified as predictors of target lesion revascularization.
We conclude that CBA is a safe and efficient technique for treatment of ISR, with immediate results similar to atheroablation and better clinical and angiographic outcomes at follow-up. This approach might be implemented as a viable option in management of focal ISR and to prepare diffuse ISR for brachytherapy treatment.
With the expanded application of coronary stenting beyond the criteria for inclusion in the randomized trials, stent restenosis emerged as a new “disease entity.” With over 800,000 stents a year in the U.S., and a stent restenosis incidence of 20% to 50% of lesions, it becomes clear that in-stent restenosis (ISR) is a significant problem (1). The most common modality to treat ISR is percutaneous transluminal coronary angioplasty (PTCA), but long-term results are unsatisfactory, with a high recurrence rate, particularly after treatment of diffuse (lesion length >10 mm) ISR (2). Additional stent implantation (STENT) may have merit because it reduces acute recoil and helps tissue extrusion; however, this acute benefit did not translate into improved long-term outcome (3–5). Atheroablative therapies (i.e., rotational atherectomy [ROTA] and excimer laser angioplasty) have been used in an attempt to decrease the rate of recurrence. However, the beneficial effect of these devices was not proven (6). Recent reports suggested that cutting balloon angioplasty (CBA) might be of benefit in the treatment of ISR, decreasing the need for repeat revascularization procedures (7,8).
The aim of our study was to compare acute and long-term angiographic and clinical outcomes between the various techniques that have been used to treat ISR in matched lesion subsets.
Patient and lesion characteristics
Between January 1997 and February 1999, a total of 1,410 patients (1,980 lesions) underwent successful coronary stent implantation at Centro Cuore Columbus, Milan, Italy. Out of this population, 684 lesions (38%) had ISR. These lesions were divided into four groups based on the ISR treatment modality: PTCA, STENT, ROTA and CBA. The groups were matched according to clinical and angiographic characteristics: presence of diabetes mellitus, reference vessel size, minimal lumen diameter (MLD), lesion length and type of ISR. Following the matching process and the random selection of only one lesion per patient, the final patient population totaled 258 (258 lesions). The numbers of patients/lesions in each group were: CBA, 57; ROTA, 48; STENT, 79; and PTCA, 74.
Angiographic follow-up was obtained in 85% of the eligible patients at an average duration of 6.2 ± 3.2 months. Clinical follow-up was obtained through outpatient clinic visits, by direct telephone interview and with the referring physician when additional information was necessary. Clinical follow-up was obtained in 100% of patients at an average of 11 ± 8 months.
Before angioplasty, 70 U/kg of intravenous heparin and 0.2 mg nitroglycerin intracoronary were administered. All patients received aspirin 300 mg q.d. and ticlopidine 250 mg b.i.d. starting three days before the procedure when feasible. Each operator selected a particular procedure according to his judgment. In general, we can state that ROTA, PTCA and stenting were used more frequently in the early period while CBA was used more recently. Each patient was encouraged to return for angiographic follow-up between the fifth and sixth month following the procedure.
Cutting balloon angioplasty
Cutting balloon (InterVentional Technologies, San Diego, California) is a conventional balloon catheter 10 or 15 mm in length with three (on smaller balloon sizes) or four microblades (on balloon sizes >3.5 mm in diameter). These blades, mounted longitudinally on the surface of balloon, are ∼0.25 mm in height and three to five times sharper than conventional surgical blades (9). During dilation, the device produces three or four endovascular surgical incisions. The CBA was performed with multiple inflations, increasing the balloon pressure, usually up to 12 atm.
ROTA with adjunct balloon angioplasty
The techniques used for ROTA have been described elsewhere (6,10). Rotablator or Rotalink devices (Boston Scientific, Scimed, Maple Grove, Minnesota) were used with an incremental burr size approach to achieve a burr/artery ratio ≥0.7. After ROTA, adjunct PTCA was routinely performed with balloon-to-artery ratio of 1:1 utilizing medium to high pressure (12 to 16 atm).
Additional stent implantation
Restenotic lesions treated by additional stent implantation were predilated using a traditional balloon angioplasty. If necessary, multiple stents were used to cover the entire lesion. Postdilation was performed with an appropriately sized balloon at high pressure (≥14 atm).
The PTCA procedure was done using noncompliant or semicompliant balloons that matched the size of the final balloon used at the time of stent implantation. Single or multiple high-pressure balloon inflations (≥12 atm) were generally performed with the goal of achieving a near 0% residual stenosis. A larger balloon (0.5 mm) than the original postdilation balloon was used at the operator’s discretion if the stent was considered undersized compared to the reference vessel size.
Intravascular ultrasound evaluation
Intravascular ultrasound (IVUS) was performed only in situations where a possible stent underexpansion was thought to be present. On average, IVUS was used in 21%, 22%, 24%, and 34% in CBA, ROTA, STENT and PTCA groups, respectively (p = NS).
In-stent restenosiswas defined as >50% diameter stenosis by quantitative coronary angiography (QCA) inside the stent or within 5 mm from the edges of a stented lesion presenting at least eight weeks after the stent implantation. In addition, ISR was defined as focal (<10 mm in length) or diffuse (>10 mm in length).
Major adverse cardiac events(MACE) were defined as death, any myocardial infarction (MI) and/or repeat revascularization (coronary artery bypass graft surgery [CABG]/repeat PTCA). A diagnosis of Q-wave myocardial infarction(QMI) was made when there was documentation of new pathologic Q-waves (>0.04 s) on an electrocardiogram in conjunction with an elevation of total creatine kinase (CK) more than twice the normal value, with a concomitant elevation of creatine kinase-myocardial band. A diagnosis of non–Q-wave myocardial infarction(NQMI) was made when an elevation of total CK to greater than twice the upper limit of normal value, with a concomitant elevation of CK-MB, was documented without development of new pathologic Q-waves. Procedural successwas defined as achievement of a residual diameter stenosis <30% at the lesion site without the occurrence of death, bypass surgery, or QMI. Recurrent restenosis at follow-up angiography was defined as ≥50% diameter stenosis by quantitative coronary angiography (QCA) occurring not earlier than eight weeks after successful treatment of ISR.
For each lesion, the single view showing the most severe degree of stenosis was used for QCA with a computer-assisted automated edge detection algorithm (QCA-CMS version 4.0, MEDIS, Leiden, the Netherlands). Both absolute reference and MLD in millimeters were determined using the guiding catheter filled with contrast for calibration. The lesion length was measured as the distance from shoulder to shoulder. In complex lesions with the involvement of adjacent segments proximal and distal, a user-defined reference lumen diameter of a proximal and distal angiographically normal-appearing segment was chosen.
Acute lumen gainwas defined as the MLD immediately after the procedure minus the MLD before the procedure. Late lumen losswas defined as the MLD after the procedure minus the MLD at follow-up. The late-loss indexwas defined as the late lumen loss divided by the acute lumen gain.
Matching was performed by a computerized program and was based on principles of the matching process derived from Umans et al. (11). The database was reviewed sequentially; for each restenotic lesion treated with the cutting balloon, the first lesion encountered in each of the three different treatment groups (ROTA, STENT, CBA) that satisfied the matching parameters was chosen. The matching parameters in order of sequential selection were: 1) diabetes; 2) reference diameter ± 0.3 mm; 3) baseline MLD ± 0.1 mm; 4) lesion length ± 1 mm; and 5) type of restenosis. The order we selected to perform the lesion selection for the matching process influenced the parameter, which could not always find a corresponding lesion with the same parameter.
Statistical analysis was performed using the StatView statistical package (StatView 5, SAS Institute, Cary, North Carolina). To remove any confounding factor due to the behavior of different lesions in the same patient, we used the patients as the unit of the analysis. For patients with multiple lesions, only one randomly selected lesion was included in the analysis according to the following randomization scheme; each lesion was assigned a random number between 0 and 1. The lesion corresponding to the smallest number was included in the analysis.
Comparison among the groups was performed by the χ2test (or the Fisher exact test) for categorical data. A value of p < 0.05 was considered statistically significant. The analysis of variance test was performed for continuous data. Multiple comparisons were accounted for by using the Bonferroni-Dunn method to preserve the overall significance level. Results were expressed as mean value ± SD. Independent risk factors were determined using a multivariate logistic regression model. Results were presented as the odds ratio (OR) and 95% confidence interval (CI) for each variable. Values with a significance level of p < 0.05 were accepted as significant.
Patient characteristics and procedural outcome
There was no difference in the baseline clinical and lesion characteristics among the four groups (Tables 1 and 2). ⇓Procedural success was achieved in all patients in the CBA group, whereas there were one NQMI in the ROTA group, two QMIs in the STENT group, and three events in the PTCA group (1 death; 1 emergency CABG; and 1 QMI). However, the difference in incidence of in-hospital MACE among the four groups did not reach statistical significance.
Quantitative angiographic measurements, both pre- and postintervention and at follow-up, are summarized in Table 2. No significant differences were seen in baseline angiographic measurements among the four groups. After intervention, the greatest MLD was achieved in the STENT group, without statistical significance, compared to the ROTA and CBA groups. Lesions treated with PTCA alone had the smallest final MLD. In the STENT group, the balloon was inflated at a higher pressure (15.7 ± 3.9 atm) than in the CBA group (10.7 ± 2.8 atm; p < 0.0001). A significantly higher late loss at follow-up was observed in the ROTA, STENT, and PTCA groups (1.30 ± 0.8 mm vs. 1.36 ± 0.8 mm vs. 1.07 ± 0.8, respectively) compared to the CBA group (0.63 ± 0.7 mm; p < 0.0001). In addition, the loss index was significantly lower in the group of lesions treated with CBA (0.38 ± 0.3 mm) than in the other groups.
Long-term clinical outcome
The mean duration of clinical follow-up was 11 ± 8 months. Death, CABG and MI occurred with equal proportions in all groups (Table 3). Recurrent ISR rate was significantly lower in the CBA group compared to other groups (CBA, 20%; ROTA, 35.9%; STENT, 41.4%; and PTCA, 45.2%; p = 0.04) with a significantly lower target lesion revascularization (TLR) rate (CBA, 15.8%; ROTA, 31.9%; STENT, 35.5%; and PTCA, 37.8%; p = 0.03). A diffuse pattern of recurrence was more common in lesions treated with PTCA, whereas focal pattern of recurrence was more common in the CBA group (Table 3).
Predictors of TLR
By univariate analysis, the presence of final MLD, diffuse ISR at baseline, and CBA were all identified as risk factors of TLR. However, in the multivariate model, CBA (OR 0.33; p = 0.003) and diffuse pattern of ISR at baseline (OR 2.07; p = 0.02) were distinguished as predictors of repeat revascularization procedures at follow-up (Table 4).
Using logistic regression analysis we also looked for factors triggering the occurrence of diffuse recurrent restenosis. Using multivariate analysis, the diffuse character of the initial ISR and treatment with PTCA were identified as independent predictors of recurrence of diffuse ISR.
The results of our study indicate that CBA can be safely used to treat ISR with acceptable acute angiographic and clinical outcome. Moreover, CBA significantly decreases the rate of repeat revascularization procedures at follow-up compared to other commonly used techniques for management of ISR.
Current treatment of ISR
In contrast to angioplasty, where restenosis is predominantly caused by elastic recoil and vascular remodeling, ISR is almost exclusively due to neointimal hyperplasia (12,13). Although the optimal treatment of ISR has not yet been well defined, three treatment approaches are commonly used: 1) PTCA; 2) atheroablation (ROTA, excimer laser angioplasty [ELCA] and directional coronary atherectomy); and 3) STENT. Recently, brachytherapy has been effectively used to prevent angiographic and clinical recurrence following optimal mechanical dilation (14). Most of the studies comparing different treatment strategies for ISR were not randomized or matched comparisons. Of importance is that not all ISR lesions have a similar risk of recurrence (15).
The simplest way to treat ISR is repeat dilation using a conventional PTCA balloon. Mechanism of lumen enlargement after PTCA for ISR is controversial. Although Gordon et al. (16)attributed it to neointimal tissue compression and extrusion of out-of-stent struts, an IVUS study by Mehran et al. (17)showed that tissue extrusion and additional stent expansion contributed equally to lumen improvement. In the Mehran et al. study, there was 32% of residual neointimal tissue of the stent area following PTCA.
In general, recurrence after PTCA treatment of ISR has been reported to range from 37% to 50% (angiographic restenosis) and from 14% to 30% (clinical restenosis) (2,17–19).
In our series, PTCA treatment gave the lowest acute gain (1.56 ± 0.7 mm) with a late loss of 0.89 ± 0.6 mm at follow-up associated with the highest rates of angiographic and clinical recurrence (45.2% and 41.4%, respectively). In the past we reported better long-term results following PTCA for ISR with a target vessel revascularization of 11% (19). The discordance between the results of the current study and our former report can be explained by the larger reference vessel size, the shorter lesion length, and the very low angiographic follow-up rate reported in that study.
Atheroablative techniques have been used in an attempt to improve the acute results of PTCA by increasing lumen dimensions through ablation of neointimal tissue (20,21). Clinical recurrence after ROTA has ranged from 28% to 50%, with an angiographic restenosis rate up to 45% (10,22–24). In one study (25), 28% of TLRs occurred after ROTA with adjunctive PTCA versus 46% with PTCA alone.
In another study, vom Dahl et al. (6)recently reported that the occurrence of re-restenosis correlated with length of the primarily stented lesion and was as high as 49% (TLR, 35%) for long diffuse lesions treated with ROTA and adjunctive balloon angioplasty. A subsequent randomized study confirmed the lack of additional benefit of ROTA and adjunctive PTCA versus simple angioplasty (26).
Preliminary observations indicate that additional stent implantation achieves the best acute results (QCA diameter stenosis of 10%) by recovering all the lumen area of the original stenting procedure, primarily via neointimal tissue extrusion out of the stent, with some additional stent expansion (5,18). However, angiographic recurrence after additional stent implantation has ranged from 30% to 35%, with clinical recurrence from 17% to 40% (3,4). The highest acute gain and the largest acute MLD were obtained after STENT. However, these acute results were offset by a high late lumen loss (about 47%) at follow-up. Different amounts of net gain obtained with PTCA, STENT and ROTA may explain the different angiographic and clinical restenosis rates. A comparison of PTCA, ELCA, ROTA and STENT in a large series of 821 restenotic lesions was concluded with the statement that “all interventional strategies are disappointing,” demonstrating similar late clinical outcome independent of device choice (27).
The cutting balloon catheter is a relatively novel device. The concept of CBA rests on the presence of microblades intended to incise the atherosclerotic plaque or neointimal tissue at the beginning of balloon inflation and to develop a controlled fault line along which dilation will occur (9). The main application of CBA was in noncalcified lesions with concentric plaque; however, this technique was subsequently applied effectively to treat several types of lesions including ISR (8,28). The Japanese Multicenter Registry of CBA for ISR reported data on 194 lesions treated. Angiographic restenosis occurred in 29% and target lesion revascularization in 22% of the lesions (29). These findings are consistent with those reported by Chevalier et al. (7)demonstrating better acute and follow-up angiographic results in treatment of ISR when CBA is compared to PTCA. The acute gain obtained was significantly bigger in the CBA group compared to the PTCA group (2.1 ± 0.47 mm vs. 1.74 ± 0.58 mm; p < 0.05). The TLR at nine months was 12% in the CBA group (7). In our series, despite the fact that lumen size immediately after CBA was not as large as after additional stent implantation or ROTA with adjunctive PTCA, follow-up MLD in the CBA group was significantly larger compared to those obtained with other techniques (Fig. 1). Both late lumen loss and loss index were significantly lower in the CBA group compared to the other techniques, resulting in a lower angiographic restenosis rate (20%) and need for TLR (15.7%).
The possible mechanisms for dilation with CBA for ISR are probably related to plaque extrusion through the stent’s struts (30,31)compared to other techniques, possibly, with less tissue injury. Figure 2demonstrates a typical IVUS finding following CBA for ISR. Another added advantage of the technique of CBA for ISR is the lack of occurrence of the “watermelon seeding effect.” This fact could become a problem when contemplating additional brachytherapy, a setting where there is the need to carefully control the boundaries of the injured segment.
Predictors of recurrent ISR
Various investigators reported a recurrence rate following treatment of ISR of between 31% and 85% with relation to the focal or diffuse type of original restenosis (2,6,22).In our study, use of CBA predicted a lower recurrence rate of restenosis but did not predict a lower occurrence of diffuse ISR at the time of the second recurrence. For this reason, and, more importantly, for the high frequency of focal ISR at baseline, we caution to state that CBA could be a viable solution to the problem of diffuse ISR. The important and positive features of the CBA approach for diffuse ISR are to provide a reasonable large final lumen even in large vessels, with no need to use a large guiding catheter; with a lower risk of distal embolization compared to ROTA or directional atherectomy; with the possibility to select carefully the length of the injury segment, and with very limited need for additional stent implantation. When dealing with a diffuse or proliferative ISR, the CBA approach seems the most logical way to prepare the lesion for a more definite therapy such as brachytherapy (32).
There are several limitations to the present study. It is a retrospective study and therefore inherently contains all the disadvantages of such a comparative analysis. The matching process was performed only for a certain set of variables, and this technique may suffer from biases of retrospective evaluations. There were no data regarding IVUS assessment of the treated lesions so as to understand better the mechanism of CBA in ISR.
Finally, CBA seems to be a reasonable approach for patients with ISR. This technique provides immediate results superior to simple angioplasty and quite similar to more complex and expensive atheroablative techniques. Follow-up results do not seem to penalize the CBA approach by a higher late loss compared to other approaches.
We believe that this technique can be considered a reasonable first line and possible definite approach for most focal ISRs and for the preparation of diffuse ISR to other more definite treatment modalities. To evaluate better the definite value of CBA versus PTCA to treat ISR, a randomized study has recently been launched in Europe.
☆ No financial support was received for this study.
- coronary artery bypass graft surgery
- cutting balloon angioplasty
- confidence interval
- creatine kinase
- creatine kinase-myocardial band
- in-stent restenosis
- intravascular ultrasound
- major adverse cardiac events
- myocardial infarction
- minimal lumen diameter
- non–Q-wave myocardial infarction
- odds ratio
- percutaneous transluminal coronary angioplasty
- quantitative coronary angiography
- Q-wave myocardial infarction
- rotational atherectomy
- additional stenting
- target lesion revascularization
- Received January 23, 2001.
- Revision received April 27, 2001.
- Accepted June 8, 2001.
- American College of Cardiology
- Eltchaninoff H,
- Koning R,
- Tron C,
- Gupta V,
- Cribier A
- Chevalier B,
- Guyon P,
- Glatt B
- Radke P.W,
- Klues H.G,
- Haager P.K,
- et al.
- Hoffmann R,
- Mintz G.S,
- Dussaillant G.R,
- et al.
- Kornowski R,
- Mintz G.S,
- Kent K.M,
- et al.
- Kastrati A,
- Schomig A,
- Elezi S,
- et al.
- Mehran R,
- Dangas G,
- Abizaid A.S,
- et al.
- Gordon P.C,
- Gibson C.M,
- Cohen D.J,
- Carrozza J.P,
- Kuntz R.E,
- Baim D.S
- Bauters C,
- Banos J.L,
- Van Belle E,
- McFadden E.P,
- Lablanche J.M,
- Bertrand M.E
- Reimers B,
- Moussa I,
- Akiyama T,
- et al.
- Mehran R,
- Dangas G,
- Mintz G.S,
- et al.
- Sharma S.K,
- Duvvuri S,
- Dangas G,
- et al.
- Mehran R
- Kurbaan A.S,
- Kelly P.A,
- Sigwart U
- Nakamura M,
- Matsubara T
- Kinoshita Y,
- Matsuno Y,
- Ohta T,
- et al.
- Teirstein P.S,
- Massullo V,
- Jani S,
- et al.