Rotablator® Versus Cutting Balloon® for the <br />
Treatment of Long In-Stent Restenoses

In comparison to conventional balloon dilatation, intracoronary stents reduce the rate of restenoses and peri-interventional complications for specific vessel segments and in vessels with a diameter of > 3 mm. The lower rate of restenoses after stent placement is due to a larger diameter gain and smaller recoil than with just balloon dilatation. Both factors result in a larger net lumen, despite increased intimal proliferation. Depending on the characteristics of the original lesion, one can expect in-stent restenosis (ISR) in 10–40% of cases after stent implantation. One-third of such ISR are focal, and two-thirds are diffuse, extending throughout the stent. While the treatment of focal ISR with conventional balloon angioplasty has acceptable results, the management of diffuse ISR by PTCA is characterized by restenosis rates of up to 70%, making such lesions an unsolved problem for the interventional cardiologist.1 The final clinical value of intracoronary brachytherapy — especially with respect to its long-term outcome — remains undetermined, despite encouraging initial results. Numerous attempts have been made to treat diffuse ISR with other interventional modalities. Based on theoretical considerations, primarily ablative technologies such as atherectomy, laser angioplasty and Rotablator® (RA) (Boston Scientific, Redmont, Washington) appeared suitable to re-open the lumen sufficiently despite the thickened neo-intimal layer. For example, RA was shown to have a low complication rate when used to treat ISR.2 When the Cutting Balloon® (CB) (Boston Scientific), a non-ablative technique, was evaluated for the management of diffuse ISR, the primary results were good and the complication rate was low.3,4 Several studies have compared both RA and CB with conventional balloon angioplasty, however, studies comparing alternative treatments for ISR are scarce.5,6 In order to compare the short- and long-term results of these treatment modalities, we performed a longitudinal study. Initially, a number of consecutive patients presenting with symptomatic, diffuse ISR were managed with the RA. During the second phase, all patients with identical presentation were treated with the CB. In both groups, the restenosis rate was determined angiographically, the frequency of major coronary events (MACE) was assessed, and the rate of required reinterventions (at the TL) was determined. Materials and methods Patient population. A total of 127 patients (Table 1) presenting with symptomatic, long (> 10 mm) ISR were treated (89 males, 38 females). From August 1996 to May 1998, a total of 51 consecutive patients were treated with the RA; from June 1998 to June 2000, there were 76 consecutive patients managed using the CB. During each treatment period, all patients with symptomatic long ISR (as determined by online QCA) were treated with the corresponding treatment technique without any further selection except of the time point of treatment. There were no crossovers between treatment groups. Both patient groups were not different with respect to age, gender, vessel affected, and the ratio of diabetics and history of myocardial infarction (Table 1). Regarding the characteristics of the coronary lesions, such as length, minimal luminal diameter (MLD), stent length, stent size, etc. (Table 2), the treatment groups were also not significantly different. Just the spectrum of stents found to be restenosed was somewhat different, because of the different time periods. Multilink® stents (Guidant Corp., Indianapolis, Indiana) and GFX® stents (Medtronic Inc., Minneapolis, Minnesota) were the most frequently found stents in both groups. Treatment Methods RA. Rotational angioplasty was performed using the Rotablator® system. Before the intervention, all patients received 10.000 IU of heparin iv and several boluses of ic nitroglycerin, as tolerated by blood pressure, into the target vessel. For interventions on the RCA, a transvenous pacemaker lead was positioned in the apex of the right ventricle. All patients were on an oral medication of 100 mg of aspirin od, which was continued after the procedure. An average of 1.22 rotablator burrs were used per case; the average size was 2.14 mm. Depending on the maximum burr size, 8 French (Fr) or 9 Fr guiding catheters were selected. Following rotablation, balloon dilatation completed all procedures. The average balloon size was 3.09 mm and the mean inflation pressure 4.8 ATM (71 psi). The burr-to-stent ratio was 0.73, the burr-to artery ratio was 0.84 and the balloon-to-artery ratio was 1.23. In all cases, the rotation rate was between 150,000–170,000/minute. The initial success rate was 100% (51 of 51). No major acute complications were seen. CB. The Cutting Balloon® was used. At the beginning of the procedure all patients received a bolus of 10.000 IU of heparin iv and a bolus of nitroglycerin into the target vessel. All patients were on an oral medication of 100 mg of aspirin od, which was continued after the procedure. In order to position the CB inside the stent, four cases required pre-dilatation with a small balloon. Only one CB was needed in each case. CB size was based on stent size (CB-to-stent ratio = 1.01). The average CB was 3.05 mm in size (CB-to-artery ratio = 1.18). Mean inflation pressure was 6.12 ATM (90 psi). In 42 cases (55%), a 10 mm CB was utilized, a 15 mm CB in 34 cases (45%). Aiming to improve the angiographic result, the vessel was further dilated with a conventional balloon of identical size (mean inflation pressure = 6.4 ATM; 94 psi) in 11 cases. The primary rate of success was 100% in this group as well, and no significant acute complication occurred. In accordance with institutional guidelines, informed consents were obtained from all patients. Quantitative coronary angiography. Treated vascular segments were analyzed with quantitative coronary angiography, using the QUANTCOR system (Siemens, Erlangen, Germany), which is based on the sufficiently validated CAAS II system. Data analysis was performed offline, after completion of the follow-up studies, and the observer was blinded. Angiographic data had been obtained after vasodilatation with intracoronary nitroglycerine, imaging the segments with the least possible axial distortion and overlap. For calibration, the native guiding catheters (no contrast agent) were used as a reference. Identical projections to the ones utilized for the initial intervention were used during follow-up. The following parameters were assessed: reference segment diameter, length of lesion, MLD and percent stenosis. Lesion length was measured from “shoulder to shoulder”. Statistical analyses. All parameters lumped in each group are expressed as mean values ± standard deviation, or — when appropriate — as median values. For comparative statistics of categorical data (restenosis), Fisher’s exact test was used. Data with p-values of Acute interventional results. In all cases, the treatment was primarily successful (i.e., resulting in less than a 50% residual stenosis and not causing any acute complications). The analysis of the study parameters before and after the interventions are summarized in Tables 2 and 3. It is evident that none of the pre-interventional QCA measurements differ significantly between the groups. The initial conditions with respect to the degree of stenosis, MLD and reference size were identical; only the lesions were slightly longer in the RA group (11.10 mm vs. 14.07 mm). Five percent of patients in the RA group and four percent in the CB group had chronic total occlusions. By QCA, post-intervention MLD is remarkably larger in the CB group (2.23 mm vs. 2.04 mm; p = 0.03). As the pre-interventional MLD and reference values were identical, this translates into significant differences with respect to luminal gain (MLD, post-MLD, pre), which was 1.58 mm in the CB group vs. 1.42 mm in the RA group (p = 0.03). Follow-up angiography after 6 months. Re-angiography was planned for all cases; it was performed in 55 of the 76 patients in the CB group, and in 44 of the 51 patients in the RA group. The angiographic follow-up rate was 72.4% and 86.3%, respectively. The median follow-up interval was 181 (CB) and 180 days (RA). At the 6-month follow-up examination, QCA results showed a significantly larger MLD (1.59 mm; p = 0.03) in the CB groups than in the RA group (1.15 mm). At the 6-month follow-up examination, QCA results showed a significantly larger MLD (1.59 mm; p = 0.03) in the CB group than in the RA group (1.15 mm) leading to a highly significant group difference for percent diameter stenosis (p Complications and major events. Immediately after the intervention (within 24 hours), no major events were observed in either group. Puncture site hematomas not requiring any intervention and mild CK elevations — up to 2.5 times the normal value — occurred in both groups with the expected frequencies. While in the hospital on the second day after the intervention, one CB group patient developed signs of an acute myocardial infarction due to an acute occlusion of the target lesion. Immediate PTCA resolved the problem without adverse consequences; CK levels remained less than 2.5 times normal. Two patients expired within the 6-month follow-up period between the intervention and the scheduled angiographic check-up (one in each group). The cause of the CB patient’s death remains unclear; the death in the RA group was most likely cardiac. Within the follow-up period, two patients in the CB group and 3 in the RA group experienced acute myocardial infarctions. In one case in each group, the target lesion under observation for this study was responsible. The rate of major cardiac events (excluding TLR) was 7.3% in the CB group and 9.1% in the RA group; therefore, within the expected range for this population. Discussion Achieving good long-term results when treating long ISR is the most important unresolved question in the field of interventional cardiology. Long ISR represent a new disease entity. Because of the unacceptable restenosis rate, conventional balloon angioplasty is not the proper treatment modality for this particular disease process.1,7,8 The real value of intracoronary brachytherapy remains controversial, although encouraging initial results have been accomplished with b- or g-rays. The long-term outcome is undetermined, though, and the method is logistically challenging. Three processes are known to contribute to restenoses: vascular remodeling, elastic recoil, and intimal proliferation. For ISR formation, only intimal proliferation is of any quantitative significance. From a pathophysiological point of view, the use of ablative technologies, such as directional coronary atherectomy, laser angioplasty, or RA, would appear favorable for treating ISR. So far, only RA has been evaluated in a study with sufficient patient enrollment (ARTIST trial), not counting case reports. In addition to other (non-ablative) techniques, CB has also been proposed and evaluated by some authors as a possible treatment option for patients with long ISR. It was shown that either RA or CB could be used in ISR with successful outcomes and low complication rates.4,9–13 However, long-term data and figures on the frequency of restenosis are lacking, especially data comparing both methods. Only Briguori et al.5 and Adamian6 compared matched groups and showed that CB resulted in lower restenosis rates than conventional balloon angioplasty and RA. Colombo’s study did not asses the differences between short and long ISR. Our study was designed to retrospectively compare the frequency of restenoses, MACE, and TLR at 6 months, between groups treated with either technique. Although the study design was retrospective and not randomized, the study groups were shown to be comparable with respect to their demographic and angiographic characteristics (Table 1). For each treatment period, the only selection criterion was the time point of the intervention, which excluded any other selection bias for either treatment option. The type, length, and size of the restenosed stents were also not significantly different throughout the study period (Table 2). However, when analyzing the QCA data obtained right after the intervention, it is evident that CB treatment has significantly better immediate results with respect to MLD and acute luminal gain (Table 4A). This may be surprising at first, as CB represents a non-ablative technique. CB appeared to extrude the plaque material better through the stent struts after the segmentation of the plaque by its blades than did conventional PTCA following ablation using RA, even if the average inflation pressure was somewhat lower in the CB group (Tables 3A and 3B). The follow-up rates of 70% and 85% were acceptable in both groups. At the scheduled follow-up, there was a trend for patients belonging to the RA group to be more symptomatic and have more frequent objective markers for ischemia related to the vessel originally targeted. In comparing QCA studies obtained at the 6-month follow-up, it was shown that the original MLD difference, which favored CB, did not only persist, but in fact increased. Figure 2 depicts graphically how a significantly larger net luminal gain results from a significant initial luminal gain augmented by a relatively smaller late luminal loss. As elastic recoil is certainly not a significant contributor for restenosis when treating ISR, one can only conclude that the lower loss of MLD (late loss) after CB is secondary to less intimal proliferation. This conforms to the hypothesis that CB treatment results in less neointimal proliferation because of less intimal trauma.14 In the CB group, reference segment data are unchanged upon follow-up, the average length of the stenoses is somewhat smaller, but the changes in percent stenoses and MLD’s differ significantly (Table 4A). Plots of the frequency distribution of percent stenoses and MLD (Figures 3 and 4) clarify the better initial gain and late loss in the CB group. This translates into a significantly lower restenosis rate of 27.3% in patients treated with the CB vs. 63.6% (Figure 1A). At first sight, a restenosis rate of over 60% in the RA group seems to be surprisingly high, but it confirms the result of the ARTIST trial, which represents the largest controlled trial of the RA in the treatment of in-stent restenosis.15 In this study, RA therapy showed a restenosis rate of 65%. Because clinical symptoms as well as objective ischemia were less frequent in the CB group, the groups are also significantly different with respect to any need to revascularize the targeted segment (18.7% vs. 43.2%) (Figure 1B). In CB patients, TLR occurred with the same order of magnitude as in coronary brachytherapy. During the study period, the rate of major adverse events (MACE without TLR) was identical in both groups (CB = 7.3%; RA = 9.1%) and conformed to our expectations (Table 5). Even when the methodological disadvantages inherent in non-randomized studies are considered, the clear-cut differences between the groups seem to indicate that CB treatment of long ISR is associated with a better long-term prognosis than RA. The comparable good results for brachytherapy and CB with respect to ISR suggest that both techniques need to be compared in a controlled study. As CB treatment is associated with much lower logistic, technical, and financial expenditure, it may very well turn out to be a valid alternate technique for some subgroups which still need to be identified. On the other hand, if the above proposed hypothesis holds true that CB treatment leads to less intimal proliferation as compared to conventional balloon angioplasty, pretreatment of diffuse in-stent restenosis before brachytherapy may be a preferable treatment option. Conclusion Because of poor long-term results, conventional balloon dilatation is not suitable for treating long ISR. In this retrospective comparative study, CB resulted in a significantly better outcome after 6 months with respect to rate of restenosis and TLR. The low restenosis rates and lower need for TLR suggest that a comparative study on CB vs. coronary brachytherapy would make sense, especially because of the much lower expenditure needed for treating long ISR with CB.

1. Eltchaninoff H, Koning R, Tron C, et al. Balloon angioplasty for the treatment of in-stent restenosis. J Am Coll Cardiol 1998;32:980–984. 2. vom Dahl J, Radke PW, Haager PK. Clinical and angiographic predictors of recurrent restenosis after percutaneous transluminal rotational atherectomy for treatment of diffuse in-stent restenosis. Am J Cardiol 1999;83:862–867. 3. Chevalier B, Guyon P, Glatt B. Treatment for in-stent restenosis. J Am Coll Cardiol 1999;33:(62A). 4. Kurbaan AS, Foale RA, Sigwart U. Cutting balloon angioplasty for in-stent restenosis. Cathet Cardiovasc Intervent 2000;50:480–483. 5. Briguori C, Yamashita T, De Gregorio, et al. Cutting balloon for treatment of in-stent restenosis: A matched comparison with conventional angioplasty and rotational atherectomy. Circulation 1999;100(Suppl I):1595. 6. Adamian M, Colombo A, Briguori C, et al. Cutting balloon angioplasty for the treatment of in-stent restenosis. J Am Coll Cardiol 2001;38:672–679. 7. Reifart N, Schwarz F, Hofmann M, et al. Balloon angioplasty of stent restenosis. Z Kardiol 1998;87(Suppl 3):79–80. 8. Haude M, Welge D, Baumgart D, et al. Arguments against conventional balloon dilatation of recurrences within the stent. Z Kardiol 1998;87(Suppl 3):72–77. 9. Bottner RK, Hardigan KR. High-speed rotational ablation for in-stent restenosis. Cathet Cardiovasc Intervent 1997;40:144–149. 10. Anzai H, Nakamura M, Hara H, et al. Research on the efficacy of cutting balloon for in-stent restenosis. Japanese Intervention Meeting Abstract 1997;I-M030. 11. Kinoshita S, Suzuki T, Hosokawa H, et al. A comparison of initial and long-term results of cutting balloon and POBA in the treatment of in-stent restenosis. Japanese Intervention Meeting Abstract. 1997;I-M029. 12. Albiero R, Nishida T, Karvouni E, Corvaja N. Cutting balloon angioplasty for treatment of in-stent restenosis. Cathet Cardiovasc Intervent 2000;50:452–459. 13. Schiele F, Meneveau N, Vuillemenot A, et al. Treatment of in-stent restenosis with high speed rotational atherectomy and IVUS guidance in small 3