In-stent restenosis

TUESDAY 9/24/02 9:00 –10:00

In-Stent Restenosis Tuesday, September 24, 2002 9:00 –10:00 AM 3:00 – 4:00 PM Main Lobby (Abstract nos. 195–219)

TCT-195 The Relation Between Low-Density Lipoprotein Cholesterol and Neointimal Hyperplasia After Stent Implantation: Intravascular Ultrasound Study. K. Matsushita, H. Hirayama, M. Nanasato, T. Muramatsu, K.Unno, M. Shimano, M. Takefuji, N. Inoue, Y. Yoshida, K. Hasegawa, H. Takezawa, N. Tsuboi, T. Itou. Nagoya Daini Red Cross Hospital Cardiovascular Center, Nagoya, Japan.

P O S T E R A B S T R A C T S

Background: Although hyperlipidemia, especially low-density lipoprotein cholesterol (LDL-C) is strongly associated with coronary atherosclerosis, no articles showed the relation between LDL-C and restenosis after percutaneous coronary intervention (PCI) without stent. The correlation between LDL-C and restenosis after PCI with stent is unclear. The goal of this study was to evaluate the correlation between LDL-C and restenosis after stent implantation. Methods: This study comprised 63 lesions from 57 patients who underwent successful stent implantation for symptomatic coronary artery disease. They underwent follow-up angiography and intravascular ultrasound (IVUS) at 6 months. Patients were divided into 2 groups according to serum level of LDL-C at follow-up study, with respective range of ⬍100 mg/dL (Group 1; 19 lesions from 17 patients), ●I˙100 mg/dL (group 2; 44 lesions from 42 patients). Stent area (SA) and lumen area (LA) were measured by IVUS at minimum lumen area. Neointimal area (NIA) and %NIA were calculated as: NIA ⫽ SA C¨ ´ 100. Restenosis rate, NA, and %NIA were LA, %NIA ⫽ NIA/SA●A evaluated between the 2 groups. Results: Baseline clinical characteristics of patients were not different between the 2 groups. Reference diameter before PCI was ` 0.20 mm vs 2.70●A ` 0.08 mm, p ⫽ NS). similar in 2 groups (2.67●A Postprocedural lumen diameter also was similar in the 2 groups ` 0.13 mm vs 2.95●A ` 0.09 mm, p ⫽ NS). The angiographic (2.85●A restenosis rate was 7.7% in Group 1 and 36.0% in Group 2 (p ⬍0.05). ` 0.31 NIA and %NIA were significantly smaller in Group 1 (2.43●A ` 0.29 mm2, p ⬍0.005; 23.6●A ` 2.6% vs 44.0●A ` 3.1%, p mm2 vs 4.12●A ⬍0.005, respectively). Conclusion: The serum level of LDL-C affected the restenosis rate in PCI with stent. As LDL-C plays an important role in neointimal hyperplasia after stent implantation, aggressive lipid-lowering therapy may prevent angiographic restenosis after coronary stenting.

TCT-196 Stent-Based Delivery of Everolimus Leads to Complete Vessel Wall Healing Without Toxicity in a 90-Day Porcine Model. T. Honda1, S. Kar1, H. Honda3, K. Takizawa3, D. McClean1, M.C. Fishbein2, N. Eigler1, F. Litvack1. 1Cedars-Sinai Medical Center,

80H

The American Journal of Cardiology姞

AM

and 3:00 – 4:00

PM

(Main Lobby)

Los Angeles, California, USA; 2UCLA, Los Angeles, California, USA; 3Sendai Kosei Hospital, Sendai, Japan. Background: Everolimus is a new macrocyclic drug that can be locally delivered to the coronary vessel wall using a polymer-coated stent. Our preliminary data have shown that the Everolimus eluting stent prevents restenosis at 30 days after implantation in pig coronary arteries. However, longer term biocompatibility of this drug and polymer remains unknown. We investigated the long-term effects of this agent on vessel wall healing in a porcine coronary model of restenosis. Methods: Five bare stents and 5 Everolimus eluting stents were implanted in 10 coronary arteries of 5 pigs. The pigs were killed at 90 days. A graded histologic analysis quantifying healing and inflammation was used. Results: There were no premature deaths. The histologic area stenosis was similar between both groups. Histopathology showed persisting voids around the stent struts present in Everolimus stents that are not seen on the bare metal stents, consistent with the presence of surface polymer. Despite this, the vessels showed evidence of complete healing with no difference in intimal inflammation, fibrin deposition, or matrix formation between the 2 groups There was near complete endothelialization in all sections. Conclusion: The presence of complete vessel healing without evidence of toxicity demonstrates the biocompatibility of Everolimus eluting stents at 90 days. Long-term clinical trials are now warranted to further explore the role of this new agent in preventing coronary restenosis.

TCT-197 Evaluation of Perfusion Balloon Compared with Cutting Balloon for In-Stent Restenosis (EPCUT): A Randomized Study. T. Hayashi, T. Takaya, A. Takarada, H. Ueno, J. Shite, S. Yamada, K. Yamashiro, K. Mizutani, T. Kajiya. Department of Cardiology, Himeji Cardiovascular Center, Himeji, Japan. Background: Previously we reported that the lumen dilatation mechanism of perfusion balloon (PB) for in-stent restenosis (ISR) evaluated by IVUS was similar to that of cutting balloon (CB) and PB had less expansive effects on vessels than CB. We randomly compared the effectiveness of PB with CB for ISR treatment. Methods: A total of 65 patients with ISR were randomly assigned to the PB group (n ⫽ 35) or CB group (n ⫽ 30). Procedural protocols were as follows: balloon size was determined by angiographic assessment (balloon/artery ratio ⫽ 1:1.2), with balloon inflation pressure within 10 atm. Durations of total balloon inflation were 10 minutes in the PB group and 5 minutes in the CB group. Repeat coronary angiography was performed at 6 months after intervention. PB (n ⴝ 21) Lesion length (mm) 11.275.2 Pre RD (mm) Pre MLD (mm) Pre %DS Post MLD (mm) F/U MLD (mm) F/U %DS

CB (n ⴝ 21)

p Value

2.36±0.49 0.60±0.49 72.9±21.2 2.08±0.45 1.13±0.65 50±26

0.19 0.74 0.65 0.36 0.02 0.03

10.2±5.6 0.60 2.33±0.55 0.65±0.48 69.7±24.6 2.20±0.39 1.60±0.64 34±21

%DS ⫽ percent diameter stenosis; F/U ⫽ follow-up; MLD ⫽ minimal lumen diameter, RD ⫽ reference diameter. Results: There was no difference in the baseline clinical and lesion characteristics of the 2 groups. The diameter (PB, 3.2 ⫾ 0.3 mm; CB,

SEPTEMBER 24, 2002

TCT ABSTRACTS/Poster

TUESDAY 9/24/02 9:00 –10:00

AM

and 3:00 – 4:00

3.1 ⫾ 0.3 mm; p ⫽ NS) and length (PB,17.0 ⫾ 3.7 mm; CB,16.9 ⫾ 4.0 mm; p ⫽ NS) of implanted stent at lesion were similar in both groups. Three cases of unsuccessful delivery of balloon to the lesion were noted in the CB group (p ⫽ 0.06) Complete angiographic results were obtained in 42 lesions (PB, n ⫽ 21; CB, n ⫽ 21). Target lesion revascularization rate was 2.0% in the PB group and 21.8% in the CB group (p ⬍0.01). Conclusion: These data demonstrate that PB has more clinical effectiveness compared with CB for the treatment of in-stent restenosis.

TCT-198 Can We Defer Intervention for Intermediate In-Stent Restenosis Based on Intravascular Ultrasound? H. Taguchi, H. Okura, K. Yamamoto, E. Enya, T. Shibutani, I. Toda, M. Morishita. Bell Land General Hospital, Sakai, Japan. Background: Intravascular ultrasound (IVUS) studies showed that coronary lesion with minimal lumen area (MLA) of 3 to 4 mm2 may not require coronary interventions. Additionally, recent studies have demonstrated long-term (1 to 3 years) in-stent neointimal regression after stent implantation. The purpose of this study was to investigate whether IVUS can be used for deferral of coronary interventions for intermediate in-stent restenosis lesions. Methods: A total of 124 patients with 28 lesions were enrolled in this study. Angiographic follow-up was performed at 6 months following successful stent implantation. IVUS imaging was performed in lesions with intermediate stenosis at the time of 6-month follow-up. Intervention was deferred if MLA was ⬎3.5 mm2 by IVUS. Clinical follow-up was obtained at 1 year after the index procedure. Results: In-stent restenosis (%DS? ⫽ 50?) was observed in 31 of 135 lesions (23%) at 6 months follow-up. Interventions were deferred based on IVUS in 7 lesions from 7 patients. At 1 year (mean, 13.8 months) after the initial stent implantation, no cardiac events and TLR were documented in these patients. Neointimal regression was observed in 2 cases in whom IVUS imaging was repeated at 1 year. Conclusion: Deferral of intervention for intermediate in-stent restenosis lesion may be possible based on IVUS imaging.

TCT-199 Long-Term Effects of Stent-Based Delivery of Sirolimus in the Porcine Model. MA1, PST1, GK2, ACY1, FT3, JD2, RF2, AJC4. 1 Stanford University, Palo Alto, California, USA; 2Cordis, Inc, Warren, New Jersey, USA; 3University of Texas, San Antonio, Texas, USA; 4Providence/St. Vincent’s Medical Center, Portland, Oregon, USA. Background: Sirolimus (SRL) is a cytostatic inhibitor of cell cycle progression and suppresses cytokine-mediated T-cell proliferation. Stent-based delivery of SRL has shown profound reduction in neointimal hyperplasia and restenosis after 2 years in humans. The purpose of this study was to evaluate the chronic vascular response to the SRL-eluting stent on regulatory proteins of the cell cycle, to analyze the expression of inflammatory cytokines, and to provide histologic correlation in a porcine coronary model. Methods: A total of 53 pigs underwent placement of 129 oversized stents (bare metal, n ⫽ 54; 150 to 200 ␮g SRL fast release, n ⫽ 20; 150 to 200 ␮g SRL slow release, n ⫽ 55) in the coronary arteries with histologic analysis and Western blot (proliferating cell nuclear antigen [PCNA], p27kip1, interleukin [IL]–2, IL-6, monocyte cheomotactic protein [MCP]–1) analysis at 3, 30, 90, or 180 days. Results: At 3 days, the mean neointimal area was similar for bare metal (0.38 ⫾ 0.19 mm2) and SRL slow-release (0.29 ⫾ 0.09 mm2) groups. After 30 days, the mean neointimal area was significantly less for the SRL slow-release (1.40 ⫾ 0.35 mm2) versus the bare metal

The American Journal of Cardiology姞

PM

(Main Lobby)

stents (2.94 ⫾ 1.28 mm2, p ⬍ 0.001). At 90 and 180 days, the mean neointimal area was similar for the SRL slow release (3.03 ⫾ 0.92 mm2 and 3.34 ⫾ 0.99 mm2), and SRL fast release (2.86 ⫾ 1.11 mm2 and 3.87 ⫾ 0.88 mm2) as compared with bare metal stents (3.45 ⫾ 1.09 mm2 and 3.65 ⫾ 1.23 mm2), resulting in equivalent lumen area and percent in-stent stenosis. Vascular segments treated with SRL-eluting stents demonstrated elevated p27kip1 levels at 30 and 90 days (p ⫽ 0.05 at 90 days) as compared with bare metal stents. However, this was accompanied by increased PCNA levels for SRL versus bare metal stents at 90 days (p ⫽ 0.003). Local production of inflammatory cytokines IL-2, IL-6, and MCP-1 was not detected by Western blot analysis at 90 days. Conclusion: SRL-eluting stents favorably modulate neointimal formation for 30 days in the porcine coronary model. Long-term inhibition of neointimal hyperplasia is not sustained despite evidence of persistent CDK upregulation, suggesting that other factors, such as the vascular response to altered shear stress, contribute to normalization of the vessel lumen.

TCT-200 Intravascular Ultrasound–Guided Histologic Measurements for Evaluation of In-Stent Restenosis in Porcine Coronary Artery. C. Strehblow1, M. Gyongyosi1, W. Sperker1, M. Shirazi1, U. Windberger2, S. Marlovits3, D. Glogar1. University of Vienna, 1 Division of Cardiology, 2Department of Biomedical Research, and 3 Department of Traumatology, Vienna, Austria. Background: We evaluated the usefulness of intravascular ultrasound (IVUS) in in-stent neointimal hyperplasia, comparing results of macroscopic measurements with standard histomorphometric analysis. Methods: Coronary artery stenting was performed in 37 left coronary arteries of 32 domestic pigs (33.4 ⫾ 8.5 kg, 52% male) using 14 Tenax (Biotronik Gmbh & Co, Berlin, Germany), 10 bare Genius (Eurocor, Bonn, Germany), 8 polymer-coated Genius, and 5 Biodivysio Matrix LO (Biodivysio Ltd., Farnham, Surrey, UK) stents. After 4 weeks, coronary angiography and IVUS with automatic pullback were performed. IVUS images were analyzed by 3-dimensional (3-D) analysis (EchoPlaque 2; INDEC Systems Inc., California, USA). The stented arteries were formalin-fixed, embedded in Technovit 9100, and cut to 4- to 8-␮m thick slides. The most diseased in-stent segment was 4.49 ⫾ 4.54 mm away from the distal stent edge assessed by IVUS. Sections of these segments were stained for quantitative histomorphometric analysis. Results: A significant correlation was found between IVUS-guided histomorphometric and 3-D IVUS measurements of maximal intimal thickness (r ⫽ 0.6611, p ⬍0.001) and area (r ⫽ 0.6460, p ⬍0.001). Macroscopic measurements resulted in a trend (p ⬎0.05) to larger maximal intimal thickness (0.94 ⫾ 0.36 mm vs 0.73 ⫾ 0.38 mm by IVUS and computerized planimetry, respectively) and area (4.68 ⫾ 1.79 mm2 vs 3.14 ⫾ 1.38 mm2 by IVUS and computerized planimetry, respectively) as compared with histomorphometry. Although the stent length, diameter, stent-balloon inflation pressure, and time and injury score did not differ between the groups, implantation of bare Genius stents resulted in significantly smaller neointimal hyperplasia expressed as neointimal volume after 4 weeks compared with Tenax, polymercoated Genius, and Biodivysio stents: 33.8 ⫾ 18.9 mm3 vs 67.4 ⫾ 34.3 mm3, 69.8 ⫾ 16.2 mm3 and 84.2 ⫾ 42.6 mm3, respectively (p ⬍0.05). Conclusion: The significant correlation between IVUS-guided histomorphometric analysis and IVUS measurements confirm the usefulness of IVUS in evaluation of experimental in-stent restenosis. Implantation of bare Genius stents resulted in significantly lower neointimal hyperplasia as compared with polymer-coated Genius, phosphorylcholine-coated Biodivysio, or Tenax stents.

SEPTEMBER 24, 2002

TCT ABSTRACTS/Poster

81H

P O S T E R A B S T R A C T S

TUESDAY 9/24/02 9:00 –10:00

TCT-201 Lesion Length is Independent Predictor of Recurrent Restenosis After Cutting Balloon Angioplasty for In-Stent Restenosis. K.S. Cha, D.I. Kim, Dl Kim1, J.W. Kim2, S.Y. Lee3, K.H. Cho1, J.Y. Oh3, M.H. Kim, T.H. Ahn4, W Kim5, MH Jeong5, Y.D. Kim, D.S. Kim1, J.C. Kang5, E.K. Shin4, J.S. Kim. Dong-A University Medial Center, Pusan, Pusan Paik Hospital, Pusan1, Ulsan University Hospital, Ulsan2, Masan Samsung Medical Center, Masan3, Ghil University Hospital, Incheon4, Chonnam University Hospital, Gwangju5, South Korea Background and Purpose: Cutting balloon angioplasty (CBA) is a reasonable first-line option in management of in-stent restenosis (ISR). However, recurrent restenosis after CBA for ISR was reported as 20% ⬃ 34% in overall cases and up to 50% in diffuse ISR. This study was designed to find independent predictors for recurrent restenosis after CBA for ISR. Methods: Sixty-two patients (69 lesions) with first ISR (focal in 40 lesions, diffuse in 26, occlusive in 3) underwent CBA and follow-up (FU) angiography systematically. A cutting balloon of equal size or 0.25 ⬃ 0.5 mm larger than stent diameter was chosen and multiple inflations were performed. Multivariate logistic regression analysis was performed using clinical and angiographic characteristics, and procedure-related factors. Results: Results are as follows.

P O S T E R

Diffuse or occlusive ISR* Stent diameter (mm) Stent length (mm) Ref. vessel diameter (mm) Pre-procedural diameter stenosis (DS,%)* Pre-procedural MLD (mm)* Lesion length (mm)* CB diameter (mm) Max. inflation pressure (atm) Inflation frequency Post-procedural DS (%)* Post-procedural MLD (mm)* FU-DS (%) FU-MLD (mm)

Restenosis (n ⴝ 15)

No restenosis (n ⴝ 54)

13 (87%) 3.2 ⫾ 0.3 23.5 ⫾ 9.9 3.1 ⫾ 0.2 90.3 ⫾ 7.9

28 (52%) 3.1 ⫾ 0.3 17.0 ⫾ 3.8 3.0 ⫾ 0.3 78.7 ⫾ 10.8

0.4 ⫾ 0.3 21.9 ⫾ 8.8 3.3 ⫾ 0.2 8.2 ⫾ 1.0 3.3 ⫾ 1.4 12.1 ⫾ 10.2 2.6 ⫾ 0.3 80.6 ⫾ 11.5 0.6 ⫾ 0.4

0.7 ⫾ 0.4 13.7 ⫾ 5.4 3.3 ⫾ 0.3 8.0 ⫾ 1.1 3.0 ⫾ 1.1 5.0 ⫾ 4.5 2.9 ⫾ 0.3 18.6 ⫾ 11.7 2.5 ⫾ 0.4

* p ⬍0.05.

A B S T R A C T S

By univariate analysis, ISR morphology, lesion length, pre-procedural diameter stenosis (DS) and MLD, and post-procedural DS and MLD were significant variables for restenosis. However, by multivariate logistic regression analysis, independent predictor of recurrent restenosis was only lesion length (OR 12.2, 95% CI: 1.3–115.2, p ⫽ 0.0286). Conclusions: Independent predictor of recurrent restenosis after CBA for ISR is lesion length. More definite treatment modalities, such as brachytherapy or atheroablation, need to be combined with bigger size cutting balloon in diffuse or occlusive ISR or ISR with heavy neointima burden.

AM

and 3:00 – 4:00

PM

(Main Lobby)

Circulatory Disease, College of Medicine, Ulsan University, Gangneung Asan Hospital, Gangneung, Korea; 3Department of Internal Medicine, College of Medicine, Yonsei University, Seoul, Korea; 4Department of Aerospace Engineering, Chosun University, Gwangju, Korea; 5Department of Mechanical Engineering, Soongsil University, Seoul, Korea. Background: The present study in angulated coronary stenosis was to evaluate the influence of flow velocity and wall shear stress (WSS) on coronary atherosclerosis, the changes of hemodynamic indices after coronary stenting, as well as their effect of evolving in-stent restenosis using human in vivo hemodynamic parameters and computed simulation qualitatively and quantitatively. Methods: Initial and follow-up coronary angiographies in the patients with angulated coronary stenosis were performed (n ⫽ 60). Optimal coronary stenting in angulated coronary stenosis had 2 models: ⬍50% angle change (Model 1, n ⫽ 33), ⬎50% angle change group (Model 2, n ⫽ 27) according to percent change of vascular angle between pre- and poststenting. Flow-velocity wave obtained from in vivo intracoronary Doppler study data was used for in vitro numerical simulation. Spatial and temporal patterns of flow-velocity vector and recirculation area were drawn throughout the selected segment of coronary models. WSS of pre/post intracoronary stenting were calculated from 3-dimensional computer simulation. Results: Follow-up coronary angiogram demonstrated significant difference in the percent of diameter stenosis between 2 groups (Group 1, 40.3 ⫾ 30.2 vs Group 2, 25.5 ⫾ 22.5%, p ⬍0.05). Negative shear area on 3-dimensional simulation, which is consistent with recirculation area of flow vector, was noted on the inner wall of poststenotic area before stenting. The negative WSS disappeared after stenting. High spatial and temporal WSS before stenting fell within physiologic WSS after stenting. Conclusion: The present study suggests that hemodynamic forces exerted by pulsatile coronary circulation, termed WSS, might affect the evolution of atherosclerosis within the angulated vascular curvature. The local recirculation area, which has low or negative WSS, might lead to progression of atherosclerosis. Moreover, geometric characteristics, such as angular difference between pre and post intracoronary stenting, might define optimal rheologic properties. Optimal rheologic properties may induce favorable vascular repair following stenting.

TCT-203 Timing of In-Stent Restenosis in Patients with Heart Transplant Coronary Arteriopathy. A.H. Doing, P.J. Casterella, J.S. Osborn, K.J. Walsh, S.C. Horton, S.G. Sorensen, J.R. Revenaugh, J.L. Burke, J.B. Muhlestein. Division of Cardiology, LDS Hospital and University of Utah, Salt Lake City, Utah, USA.

Hemodynamic Effects on Restenosis in Angulated Coronary Stenosis After Stenting. B.K. Lee1, J. Lee2, B.K. Hong3, D. Kim3, H.M. Kwon3, H.-W. Roh4, M.-T. Cho5, S.H. Suh5, H.-S. Kim3. 1 Department of Internal Medicine, College of Medicine, Inje University, Sanggye Paik Hospital, Seoul, Korea; 2Department of

Background: Transplant coronary arteriopathy (TCA) is the major cause of death beyond the first year after cardiac transplantation and is associated with high rates of restenosis and in-stent restenosis (ISR). Although reinnervation occurs rarely, due to cardiac denervation, TCA seldom presents with classic angina. Diagnosis of ISR is normally made with scheduled angiography because noninvasive testing is known to be unreliable. In nontransplanted arteries, ISR typically occurs within 4 to 6 months of stent implantation. In patients with TCA the timing of ISR is not known. Methods: Beginning in 1995, an aggressive percutaneous coronary intervention (PCI) strategy for the treatment of TCA was initiated at our institution. Yearly angiography was performed, and whenever possible, patients with TCA were treated with PCI and surveillance angiography was then performed at 3 months. A total of 28 consecutive TCA patients undergoing 123 separate PCI procedures (stented lesions, n ⫽ 46; nonstented de novo lesions, n ⫽ 37; nonstented restenotic

82H

SEPTEMBER 24, 2002

TCT-202

The American Journal of Cardiology姞

TCT ABSTRACTS/Poster

TUESDAY 9/24/02 9:00 –10:00

AM

and 3:00 – 4:00

lesions, n ⫽ 40) were analyzed. One-year angiographic restenosis (⬎50% lesion stenosis) was obtained. Results: Restenosis occurred in 29 of 46 (64%) stented lesions at 1 year. Fourteen patients (48%) had ISR by the third month, 8 patients (28%) had ISR between 4 and 6 months, and 7 patients (24%) had ISR between months 7 and 12 post implantation. Of note is that restenosis was also seen in 2 stents in the 14th and 1 stent in the 22nd month after implantation. Conclusion: In this large series of TCA patients treated with a strategy of aggressive percutaneous intervention with follow-up angiography, ISR rates were markedly higher than reported rates in nontransplant patients. TCA ISR occurs early (⬍3 months) in 45%, but also occurs late (⬎6 months) in 31% of the TCA ISR population. While optimal therapy for TCA remains to be determined, aggressive early and continued surveillance angiography appears warranted.

TCT-204 Paclitaxel Local Delivery to Prevent Diffuse, Recurring In-Stent Restenosis. P. Buszman, A. Zurakowski, A. Gruszka, B. Bialkowska. Silesian Medical School, Katowice, Poland. Background: It has been proved in animal models that paclitaxel has a cytotoxic influence on intimal cells. The aim of this study is to test whether locally delivered paclitaxol can prevent recurrence of diffuse in-stent restenosis. Methods: Five patients with ⱖ2 episodes of in-stent restenosis within the same coronary segments were included in the study. Eleven restenotic segments underwent predilatation with Remedy balloon catheter. At the end of first dilatation, 100 ␮g of paclitaxel diluted in 2 mL of 0.9% NaCl was delivered under the pressure of 2 to 3 atm during 60 seconds in each place. Postdilatation with a larger balloon was performed to gain optimal result. No additional stenting was required in any case. Results: There were no major acute coronary events during hospitalization or 6 months’ observation. Blood tests performed 1 week after the procedure did not reveal any abnormalities that could be connected to cytostatic treatment. Follow-up angiographic evaluation showed late loss of 0.4 ⫾ 0.5 mm and restenosis in 3 segments (27.2%). Repeat angioplasty with delivery of increased dose of paclitaxel (200 ␮g in 2 mL) was performed in 3 patients with restenosis. The late results will be known after the next 6 months. Conclusions: Locally delivered paclitaxel might be a valuable treatment for in-stent restenosis. However, further evaluation of a proper dose of delivered drug is warranted.

TCT-205 Implications of the Watermelon Seeding Phenomenon During Treatment of Patients with In-Stent Restenosis. M. GomezRecio1, I. Calvo2, J.A. Bullones2, J.M. Hernandez2, J.R. LopezMinguez2, R. Lezaun2, M.J. Perez-Vizcayno2, R. Melgares2, C. Moris2, J.M. Auge2, F. Alfonso2. 1Hospital de la Princesa, Madrid, Spain; 2Hospital Clinico San Carlos, Madrid, Spain. Background: Treatment of patients (pts) with in-stent (ST) restenosis (RE) remains a challenge. Although most of the pts undergoing repeat coronary interventions in this setting obtain a good acute result, recurrence rates remain high. Technical factors and suboptimal procedural results may be implicated in the adverse long-term outcome. In this study we analyzed the occurrence and implications of the watermelon seeding (WMS) phenomenon during treatment of pts with in-ST RE. Methods: Pts were included in the RIBS randomized study (Restenosis Intra-Stent: Balloon Angioplasty Versus Elective Stenting); 224 pts allocated to ST and 226 to balloon angioplasty (BA) were analyzed. Angiographic readings (centralized core lab) included qual-

The American Journal of Cardiology姞

PM

(Main Lobby)

itative and quantitative coronary angiography (QCA) analyses. The presence of WMS was determined after carefully reviewing the angiograms, dilation material, and the technical difficulties reported on case report forms. Results: WMS was demonstrated in 42 pts (9%), 16 (7%) in the ST arm and 26 (12%) in the BA arm (p ⫽ NS). In the ST arm WMS was seen during BA predilation, never during ST deployment. WMS occurred more frequently in diffuse RE (⬍15 mm length) (13% vs 8%, p ⫽ 0.08; relative risk [RR], 1.4; 95% confidence interval, 0.99 –1.97), and more severe RE: % stenosis (79 ⫾ 12% vs 76 ⫾ 16%, p ⫽ 0.08); minimal lumen diameter (MLD) (0.6 ⫾ 0.4 mm vs 0.7 ⫾ 0.4 mm, p ⫽ 0.1), and TIMI⬍3 (21% vs 8%, p ⫽ 0.01). Pts with WMS required longer dilations (185 ⫾ 116 sec vs 150 ⫾ 106 sec, p ⬍0.05) and had poorer final results: percent diameter stenosis (20 ⫾ 11% vs 16 ⫾ 11%, p ⫽ 0.06); MLD (2.3 ⫾ 0.5 mm vs 2.5 ⫾ 0.5 mm, p ⫽ 0.03). In addition, pts experiencing WMS also had poorer angiographic results at follow-up including MLD (1.3 ⫾ 0.7 mm vs 1.6 ⫾ 0.7 mm, p ⫽ 0.007), percent diameter stenosis (55 ⫾ 24 vs 44 ⫾ 24%, p ⫽ 0.006), and recurrent RE (56% vs 37%, p ⫽ 0.02; RR, 1.5; 95%CI, 1.1–2). Finally, in the BA arm, independent predictors of recurrent RE included diabetes, time to RE, MLD, crossover, residual dissection, and WMS. Conclusion: WMS is a relatively common phenomenon during treatment of pts with in-ST RE. Long and severe RE predispose pts to WMS. WMS is associated with complex procedures, suboptimal results, and higher recurrent RE rates.

TCT-206 In-Stent Neointimal Regression and Peri-Stent Plaque Changes Between 6 and 12 Months After Stent Implantation: A Volumetric Intravascular Ultrasound Study. H. Taguchi, H. Okura, K. Yamamoto, E. Enya, T. Shibutani, I. Toda, M. Morishita. Bell Land General Hospital, Sakai, Japan. Background: Previous studies have demonstrated in-stent neointimal regression at 1 to 3 years after stent implantation. The purpose of this study was to investigate both in-stent neointimal and peristent plaque changes between 6 and 12 months after stenting. Methods: Six stented lesions that did not require repeat revascularization at 6 months follow-up were enrolled. Intravascular ultrasound (IVUS) imaging was obtained at 6 months and repeated at 12 months after stent implantation using an automated pullback system (0.5 mm/sec). Stent, vessel, and lumen volume were computed by the Simpson method. In-stent neointimal volume was calculated as stent minus lumen volume. Persistent plaque volume was calculated as vessel minus stent volume. The mean values of these indices were calculated to standardize stent length. Results: Results are presented in the Table.

Mean persistent plaque volume (mm2) Mean in-stent neointimal volume (mm2)

6 Mo

12 Mo

p Value

7.9 ⫾ 1.4

7.6 ⫾ 1.1

NS

4.2 ⫾ 1.1

3.8 ⫾ 1.1

0.03

Conclusion: In-stent neointimal regression occurred between 6 and 12 months after stent implantation without significant peristent plaque changes.

TCT-207 Oral Sirolimus to Prevent Stent Restenosis. C. Costantini, D. Zanuttni, S. Tarbine, R. Darwich, M. Freitas, O.C. Costantini, J. Lazarte, M. Maranha˜ o, O. Garcia, M. Bubna, M. Barbosa, L. Sabattini, M. De Oliveira, M. Andrade, G. Yared.

SEPTEMBER 24, 2002

TCT ABSTRACTS/Poster

83H

P O S T E R A B S T R A C T S

TUESDAY 9/24/02 9:00 –10:00 Background: Experimental animal data showed systemic rapamicin to significantly decrease the restenosis rate after stent implantation in de novo lesions. Lately, the sirolimus-coated stent has shown similar efficacy in clinical studies. The aim of this study was to assess the safety and efficacy of oral rapamycin in patients (pts) after stent implantation in de novo lesions. Methods: From November 2001 through June 2002, 18 de novo coronary lesions were treated with intravascular ultrasound (IVUS)– guided stenting in 17 consecutive pts. The stent used was the BX Velocity. Oral sirolimus was administered at a dose of 6 mg 2 hours before the procedure followed by 2 mg/day during 14 days. Complete blood laboratory and clinical examinations were performed before procedure, and at 10 and at 30 days after drug initiation. IVUS and angiographic follow-up were planned at 4 months after stent implantation. Results: Demographics were as follows: male sex, 70%; diabetes (DM), 29%; HTA, 40%; Disl., 60%. Reference vessel diameter was 2.97 mm, lesion length 8.7 mm, pre–minimal lumen diameter (MLD) was 1.07 ⫾ 0.41 mm, and pre-direct stenting (DS%) 67.4% ⫾ 11.7%. Angiographic success was 100%. Final MLD was 3.13 ⫾ 0.36 mm, and final DS% was 4.02% ⫾ 4.81%. Mean stent diameter was 3.39 ⫾ 0.48 mm, mean stent length was 16.2 ⫾ 4.37 mm, and the mean pressure was 17.1 ⫾ 2.19 atm. In hospital and late major cardiac events were absent. At the moment, angiographic and IVUS follow-up was performed in 47% (n ⫽ 8) of patients, and the results have shown restenosis in 2 patients. Only 1 restenosis was in-stent; the other was in-segment but outside the stent. There were no hematologic or clinical adverse reactions related to the rapamicin administration. Conclusion: Administration of oral sirolimus in this group of patients proved to be safe and did not present any clinical or hematologic complication. There were no major cardiac events. Efficacy remains to be assessed. A complete clinical, angiographic, and IVUS follow-up will be available for TCT.

P O S T E R A B S T R A C T S

TCT-208 Cutting Balloon Angioplasty Versus Rotational Atherectomy for Treatment of Diffuse In-Stent Restenosis: Multicenter Registry in Japan. S. Nakamura1, E. Saito1, T. Miyauchi1, J. Yokoyama1, A. Kanazawa1, Y. Hayama1, H. Nakamura2, K. Yamamoto2, K. Hozawa2, S. Nakamura2, J. Koyama3, N. Makishima3, T. Keida4, H. Okumura4, S. Matsukawa4, H. Ohira4. 1New Tokyo Hospital, Matsudo-shi, Japan, 2Kasori Hospital, Chiba-shi, Japan, 3Kobari General Hospital, Noda-shi, Japan, 4Edogawa Hospital, Edogawaku, Japan. Background: Diffuse in-stent restenosis (ISR) is still a challenging problem and optimal treatment has not been established.

Procedural success (%) MACE at 30 days (%) Lesion length, mm (mean) Reference diameter, mm (mean) Post MLD, mm (mean) Follow-up MLD, mm (mean) Restenosis (%) TLR (%)

CB (n ⴝ 234)

Rota (n ⴝ 288)

p Value

98.3 0 22.6 2.72 2.20 1.55 41.8 32.8

98.6 0 24.5 2.83 2.73 2.03 27.8 20.6

NS — NS NS ⬍0.05 ⬍0.05 ⬍0.05 ⬍0.05

MACE ⫽ major adverse cardiac event (death/coronary artery bypass graft/myocardial infarction); MLD ⫽ minimum lumen diameter; NS ⫽ not significant; TLR ⫽ target lesion revascularization.

AM

and 3:00 – 4:00

PM

(Main Lobby)

angioplasty (Rota) for the treatment of ISR, we studied 522 patients (586 lesions) with a first ISR; 288 patients (322 lesions) received Rota and 234 patients (264 lesions) received CB. Basic characteristics of both groups were similar. Results: See Table for immediate and long-term clinical outcomes between CB and Rota. Conclusion: These data show that Rota seems to offer a better long-term clinical outcome than CB. A randomized trial is needed.

TCT-209 Percutaneous Transluminal Coronary Angioplasty Versus Cutting Balloon for In-Stent Restenosis: Acute and 24-Hour Angiographic and Intravascular Ultrasound Evaluation. P. Montorsi, S. Galli, G. Teruzzi, D. Trabattoni, P. Ravagnani, F. Fabbiocchi. A.L. Bartorelli. Institute of Cardiology, Centro Cardiologico Monzino, Milan, Italy. Background: To investigate the angiographic and intravascular ultrasound (IVUS) changes after cutting balloon (CB) and percutaneous transluminal coronary angioplasty (PTCA) treatment in in-stent restenosis (ISR). Methods: A group of 50 consecutive patients (pts) with focal ISR were randomized to CB (n ⫽ 25) or PTCA (n ⫽ 25). CB and PTCA balloon size was selected by IVUS 1:1 device-to-stent ratio. Devices were inflated at 8 atm until a ⬍30% angiographic diameter stenosis (DS) was achieved. Quantitative coronary angiography (QCA) and IVUS (both planar and volumetric) evaluation were carried out before, immediately after, and 24 hours after treatment. Measurements included minimal lumen diameter (MLD) and DS for QCA and external elastic membrane area (EEMA), stent area (SA), minimal lumen area (MLA), restenosis area (RA ⫽ SA ⫺ MLA), and plaque plus media area (PMA ⫽ EEMA ⫺ SA) for IVUS. Results: A similar MLD increase (1.20 ⫾ 0.46 mm vs 1.37 ⫾ 0.57 mm) and DS decrease (⫺37% ⫾ 15% vs ⫺45% ⫾ 14%) was detected after PTCA and CB, respectively. No significant difference was found in MLA increase (4.98 ⫾ 2 mm2 vs 5.5 ⫾ 2.2 mm2) and RA decrease (⫺3.9 ⫾ 1.5 vs ⫺4.3 ⫾ 1.9 mm2) in PTCA and CB, respectively. SA, EEMA, and PMA did not significantly change after treatment in both groups. Of the total mean lumen enlargement after PTCA and CB, 20% was due to additional stent expansion and 80% was due to RA decrease. At 24 hours, a greater MLD (⫺0.28 ⫾ 0.37 mm vs ⫺0.05 ⫾ 0.26 mm, 2 p ⫽ 0.05) and MLA loss (⫺2 ⫾ 1.6 mm vs ⫺0.3 ⫾ 0.9 mm2, p ⫽ 2 2 0.001) and RA increase (1.8 ⫾ 1.5 mm vs 0.35 ⫾ 0.57 mm , p ⫽ 0.001) was detected in PTCA vs CB. No stent or vessel recoil occurred. Volumetric changes paralleled planar IVUS variations. Conclusion: PTCA and CB share similar mechanisms of lumen enlargement in ISR. In-stent tissue is mainly redistributed along a larger stent rather than being extruded out of the stent struts (as suggested by the unchanged PMA). At 24 hours, a significant lumen loss occurred only in PTCA. This may be due to a more stable and uniform in-stent tissue shift after CB that could translate into a better long-term result.

TCT-210 A Novel Mechanism Explaining Acute Recoil After Balloon Dilatation for the Treatment of In-Stent Restenosis. J.S. Mun˜ oz, M. Albertal, A. Abizaid, F. Feres, R. Staico, L.A. Mattos, R. Graebin, A.G.M.R. Sousa, J.E. Sousa. Instituto Dante Pazzanese de Cardiologia, Sa˜ o Paulo, Sa˜ o Paulo, Brazil.

Methods: To compare the efficacy and safety of cutting balloon angioplasty (CB) with rotational atherectomy with adjunctive balloon

Background: Whether the mechanism of acute luminal loss after balloon dilatation for the treatment of in-stent restenosis (ISR) is due to (1) prolapsation of already extruded neointimal hyperplasia (NIH), (2)

84H

SEPTEMBER 24, 2002

The American Journal of Cardiology姞

TCT ABSTRACTS/Poster

TUESDAY 9/24/02 9:00 –10:00

AM

and 3:00 – 4:00

stent (St) recoil, or (3) decompression of the remnant NIH is a matter of debate. To determine the mechanism of acute luminal loss after balloon dilatation for the treatment of ISR, we prospectively analyzed the changes observed in plaque (P) volume behind the implanted St and at the reference sites as well as the degree of early reaccumulation of NIH volume after balloon dilatation. Methods: A total of 18 patients underwent intravascular ultrasound imaging (IVUS) before (Pre) and after balloon angioplasty (Post1) for ISR and again after a delay of approximately 25 minutes from the last balloon dilatation (Post2). IVUS volumetric analysis was performed at the proximal, middle, and distal stented segments and at the reference sites. Results: Volumetric analysis is shown in the Table. Volume Index (mm2)

Ref P 7.17 ⫾ 2.3

Pre Post 1 Post 2

Proximal P/St

Middle NIH 2.2*†

9.4 ⫾ 2.8/ 6.19 ⫾ 10.6 ⫾ 2.9*† 7.52 ⫾ 2.3 9.5 ⫾ 2.8/ 3.84 ⫾ 1.7 11.5 ⫾ 3.1 7.41 ⫾ 2.3 9.4 ⫾ 2.7/ 4.94 ⫾ 2.1* 11.3 ⫾ 3.1

P/St

Distal NIH 2.2*†

9.5 ⫾ 2.7/ 6.5 ⫾ 10.6 ⫾ 2.9*† 9.5 ⫾ 2.8/ 4.0 ⫾ 1.7 11.5 ⫾ 3.0 9.4 ⫾ 2.6/ 5.2 ⫾ 2.0* 11.3 ⫾ 3.1

P/St

NIH

9.5 ⫾ 2.7/ 6.4 ⫾ 2.4* 10.6 ⫾ 2.*† ‡ 9.6 ⫾ 2.8/ 4.0 ⫾ 1.8 11.4 ⫾ 3.0 9.5 ⫾ 2.6/ 5.3 ⫾ 2.2* 11.3 ⫾ 3.0

Ref ⫽ interpoled references sites; volume index ⫽ P, St and NIH/ length. *p ⬍0.05 vs Post 1. † p ⬍0.05 vs Post 2. Conclusion: The most plausible explanation for the acute loss observed after treatment of ISR was decompression of remnant NIH that was already compressed during balloon dilatation. The process of compression and decompression appears to be homogenous throughout the St length. No significant St recoil was observed. A lack of P changes behind the St and at the reference sites argues against the presence of significant tissue prolapsed or NIH extrusion/redistribution.

TCT-211 A Simple Angiographic Method for Predicting the Outcome of Treatment of In-Stent Restenosis. M. Boccalatte, N.T. Mulvihill, I. Bossi, R. Gottilla, B. Farah, J. Fajadet, J. Marco. Clinique Pasteur, Toulouse, France. Background: In-stent restenosis (ISR) remains the principal limiting factor to the long-term success of coronary stenting. The incidence of ISR varies between 10% to 58%, depending on specific lesion and patient characteristics. The outcome of treatment for focal ISR remains good with target lesion revascularization (TLR) rates of 20% to 30%. However, the results of treating more diffuse ISR lesions are disappointing, and vascular brachytherapy (VBT) currently offers the only effective treatment. We sought to develop a simple angiographic classification system to predict clinical outcome after treatment for ISR and to allow for appropriate selection of patients for expensive novel therapies. Methods: We treated 257 lesions of ISR in 234 patients with balloon angioplasty. No patients were treated with adjunctive VBT. Group A (n ⫽ 83) had an ISR length ⬍50% of the total stented length. Group B (n ⫽ 173) had an ISR length ⱖ50% of total stent length and also included proliferative and occlusive ISR. Results: The average stent length was 30emm vs 25emm, p ⫽ 0.05 respectively. The original postprocedural minimal lumen diameter was 3.4e.4 mm vs 3.2e.3 mm, p ⫽ 0.08 respectively. Patients were followed up for a mean of 459 days (range, 286 – 693 days). Adverse events were defined as death, nonfatal myocardial infarction, and target vessel revascularization. The adverse event rates for Groups A and B were 18% and 29% respectively, p ⫽ 0.05.

The American Journal of Cardiology姞

PM

(Main Lobby)

Conclusion: Classifying the length of ISR as a function of total stent length allows for prediction of clinical outcome and offers the potential for appropriate patient selection for more expensive novel antirestenosis therapies including VBT and potentially drug-eluting stents.

TCT-212 Comparative Results of Cutting Balloon Angioplasty and Rotational Atherectomy in Long Diffuse In-Stent Restenosis. C.Y. Fang, C.J. Wu, Y.K. Hsieh, J.S. Yang, H.K. Yip, B.F. Guo. Chang Gung Memorial Hospital Kaohsiung Medical Center, Taiwan. Background: Rotational atherectomy (ROTA) has frequently been used in the management of long diffuse in-stent restenosis (ISR), but controversies remain. Recent studies have suggested that cutting balloon angioplasty (CBA) can be an alternative for the treatment of simple ISR. It is not clear, however, whether CBA is also useful in long, diffuse ISR. Methods: Between September 1997 and April 2002, a registered, nonrandomized observational study was performed to determine whether CBA had an advantage over ROTA in the treatment of long, diffuse ISR (lesion length ⬎ 20 mm). Intravascular ultrasound during the intervention was used in a substudy in 40 patients (25 CBA, 15 ROTA). We compared the immediate and long-term results in 100 patients, 40 patients undergoing CBA and 60 patients undergoing ROTA. Results: Both groups had similar clinical and lesion characteristics except for longer lesion length and smaller vessel size in the patients undergoing ROTA. CBA and ROTA have similar angiographic and clinical results as well as significant recurrent restenosis rates (Table). By multivariate analysis, lesion calcification, multiple stents implantation, and impaired ventricular systolic function were identified as predictors of recurrent restenosis. Conclusion: CBA can achieve an efficacy similar to ROTA in treating long, diffuse ISR, but recurrent restenosis remains a challenging issue.

TCT-213 Human Immunodeficiency Viral Infection Increases Clinical Restenosis Rate After Coronary Stent Implantation in Patients with Insulin-Requiring and Non–Insulin-Requiring Type 2 Diabetes Mellitus. S. Kokolis, S. Mezitis, J. Andrieni, R. Kokolis, G. Dangas. New York, New York, USA; Lenox Hill Hospital, New York, New York, USA; Cardiovascular Research Foundation, New York, New York, USA. Background: This study measured the increased in-stent restenosis after treatment with coronary angioplasty in human immunodeficiency virus (HIV)–positive diabetes patients compared with non–HIV-positive patients with non–insulin- and insulin-requiring type 2 diabetes mellitus. Previous studies have demonstrated that patients with diabetes have increased restenosis after coronary angioplasty. These rates were attributed to accelerated neointimal tissue growth after stent implantation. HIV has been shown to cause hypercholesterolemia, increase insulin resistance, and consequently decrease the effect of diabetes medication. Methods: We assessed 15 HIV-positive patients with diabetes who received coronary stents and were subsequently followed for 6 to 18 months with repeat catheterization. The patients continued on their antidiabetic regimen. The control group included 108 HIV-negative patients who also received stents and had follow-up catheterization 6 to 18 months. In-stent restenosis was considered when there was a luminal diameter stenosis ⬎ 50% of the stented lesion by arteriography. The groups had similar baseline characteristics: similar body mass index; mean 20 smoking pack-years; mean hypercholesterolemia, 195 mg/dL;

SEPTEMBER 24, 2002

TCT ABSTRACTS/Poster

85H

P O S T E R A B S T R A C T S

TUESDAY 9/24/02 9:00 –10:00 mean systolic blood pressure, 145 mm Hg; diastolic blood pressure, 85 mm Hg; mean ejection fraction, 0.50; average hemoglobin A1C, 7.8; age range, 44 to 81 years. All patients received 75 mg of clopidogrel (Plavix) for 4 weeks and 325 mg of aspirin indefinitely after stenting. Results: HIV appeared to be a significant factor in increased in-stent restenosis between the 2 groups with diabetes (n ⫽ 15; odds ratio [OR], 9.5; 95% confidence interval [CI], 2.0 – 43; p ⬍0.05). In the subgroup analysis, the HIV-positive non–insulin-requiring patients appeared to have further increased in-stent restenosis compared with the non–insulin-requiring patients without HIV (OR, 14; 95% CI, 1.7–114; p ⬍0.05). The HIV-positive insulin-requiring patients also appeared to have a higher rate of in-stent restenosis compared with the insulinrequiring diabetes patients without HIV (OR, 5; p ⫽ NS). Conclusion: This retrospective case-control study demonstrated that HIV–positive patients with diabetes have increased in-stent restenosis compared with their HIV-negative counterparts.

TCT-214 Human Immunodeficiency Virus Infection Increases In-Stent Restenosis Rate. S. Kokolis, R. Mehran, J. Andrieni, R. Hanna, S. Polena, R. Kokolis, M. Amheriene, J. Moses, M.B. Leon, N. Kipshidze, G. Dangas. Lenox Hill Hospital, Lenox Hill Heart and Vascular Institute, New York, New York, USA.

P O S T E R A B S T R A C T S

Background: We evaluated the outcome of patients with human immunodeficiency virus (HIV) infection who had percutaneous coronary interventions (PCIs) with stent implantation. HIV and antiretroviral therapy may propagate intimal proliferation and inflammation, thereby accelerating in-stent restenosis via these cascades. The atherogenic effects of comorbid conditions such as HIV and coronary artery disease (CAD) have never been studied angiographically. It is hypothesized that this comorbidity increases in-stent restenosis. Methods: Patients with HIV (n ⫽ 30) who had PCI with stents from 1997 to 2002 were analyzed. All patients had PCI with repeat angiography 6 to 36 months after initial PCI. In-stent restenosis was considered when there was a luminal diameter stenosis ⬎50% of the stented lesion by angiography. Results: The patients had similar baseline characteristics and procedural variables. Restenosis rates were as follows: All HIV patients, 74% (22 of 30); HIV patients with CD4 ⬍350, 100% (20 of 20); HIV patients with no HIV therapy, 67% (8 of 12); HIV patients with HIV therapy, 78% (14 of 18); protease inhibitor, 75% (6 of 8); nucleoside reverse transcriptase inhibitor, 87.5% (7 of 8); non-nucleoside reverse transcriptase inhibitor, 89% (8 of 9) (Table).

All HIV patients HIV with CD4 ⬍ 350 cells/cubic millimeter HIV with viral load ⬎ 50,000 copies/cc HIV with no anti-retroviral therapy HIV with anti-retroviral therapy Protease inhibitor Nucleoside reverse transcriptase inhibitor Non-nucleoside reverse transcriptase inhibitor

n

Restenosis

30 15 20 18 12 8 8 9

74% (22/30) 100% (15/15) 100% (20/20) 78% (14/18) 67% (8/12) 75% (6/8) 87.5% (7/8) 89% (8/9)

AM

and 3:00 – 4:00

PM

(Main Lobby)

New York, New York, USA; Cardiovascular Research Foundation, New York, New York, USA. Background: The aim of this study was to determine whether thiazolidinedione (TZD) therapy with rosiglitazone or pioglitazone reduces the clinical recurrence after treatment of coronary in-stent restenosis with angioplasty in patients with non–insulin-requiring type 2 diabetes mellitus. Methods: We assessed 100 patients treated with TZDs. They received coronary stents and were subsequently followed for 6 to 18 months with repeat catheterization or thallium stress testing. The control group included 165 patients who received stents and had follow-up catheterization 6 to 18 months later and who were continued on their oral diabetic regimen that did not include any TZDs. In-stent restenosis was considered when there was a luminal diameter stenosis ⬎50% of the stented lesion by arteriography or a thallium-documented moderate or severe ischemia in the arterial territory previously treated with a stent. The groups had similar baseline characteristics. All patients received 75 mg of clopidogrel (Plavix) for 4 weeks and 325 mg of aspirin indefinitely after stenting. Results: Clinical restenosis was 28% and 42% in the TZD group and the non–TZD group, respectively. The odds ratio (OR) was 0.52 times the likelihood of decreasing in-stent restenosis with the group that was treated with the TZDs compared with the control group (95% confidence interval [CI], 0.3– 0.87; p ⬍0.05). In the subgroup analysis, rosiglitazone had an OR 0.55 times the likelihood of decreasing in-stent restenosis in the TZD group compared with control (n ⫽ 65; 95% CI, 0.3– 0.98; p ⬍0.05). Pioglitazone had an OR 0.45 times the likelihood of decreasing in-stent restenosis (n ⫽ 35; 95% CI, 0.2– 0.97; p ⬍0.05). With respect to diabetic treatment intensity, 20% of the rosiglitazone group were treated with suboptimal doses between 2 to 4 mg/day on a medication whose maximal dosing is 8 mg/day. In the pioglitazone group, 96% of the patients were treated with the recommended doses of 30 to 45 mg/day. Conclusions: This case-control study demonstrated that TZDs provided a statistically significant benefit in decreasing clinical recurrence of in-stent restenosis in this group of non–insulin-requiring type 2 diabetes patients. These results are concordant with the findings of Awazi et al., who found decreased intimal in-stent hyperplasia in non–insulin-requiring type 2 diabetes patients treated with troglitazone.

TCT-216 Predictors of Clinical Restenosis in De Novo Lesions. A.D. Kugelmass1, D.J. Cohen2, M.J. Mack3, F.D. Houser3, S.L. Battaglia3, L.G. Tarkington3, A.W. Simon4, E.R. Becker5, S.D. Culler5. 1Henry Ford Health System, Detroit, Michigan, USA; 2Beth Israel-Deaconess, Boston, Massachusetts, USA; 3HCA, Inc., Nashville, Tennessee, USA; 4Cardiac Data Solutions, Inc., Indianapolis, Indiana, USA; 5Emory University, Atlanta, Georgia, USA.

Thiazolidinediones Reduce Clinical Restenosis Rate After Coronary Stent Implantation in Patients with Non–InsulinRequiring Type 2 Diabetes Mellitus. S. Kokolis, S. Mezitis, J. Shao, J. Andrieni, R. Kokolis, G. Dangas. Department of Medicine, Divisions of Cardiology and Endocrinology, Lenox Hill Hospital,

Background: The primary purpose of this study is to identify those factors associated with the likelihood of repeat percutaneous coronary intervention (PCI) within 1 year on a target lesion after stent implantation. Methods: Use and outcome information were prospectively collected on all patients who had a single de novo lesion PCI procedure involving a stent in 1 of 4 community hospitals between July 1, 1999 and September 30, 2000 (N ⫽ 1,685). The primary outcome was target lesion repeat PCI during a 1-year follow-up period. A logistic regression equation was estimated to identify which of the 25 control factors were significantly (p ⬍0.05) associated with clinical restenosis of the target lesion.

86H

SEPTEMBER 24, 2002

Conclusion: HIV patients have a very high in-stent restenosis rate. A low CD4 count appears to correlate with the highest restenosis rate.

TCT-215

The American Journal of Cardiology姞

TCT ABSTRACTS/Poster

TUESDAY 9/24/02 9:00 –10:00

AM

and 3:00 – 4:00

Results: Overall, 6.1% of the patients enrolled in the study had an additional target lesion PCI performed. The Table reports the estimated odds ratios for those factors significantly (p ⬍0.05) associated with a repeat PCI procedure. Only 1 clinical variable (diabetes) trended toward significance. However, having a minimum stent diameter of ⬍3.0 cm increased the likelihood of a repeat revascularization by 4.3 times (if diameter was ⬍2.5 cm) and 3.3 times (if diameter was between 2.6 and 3.0 cm) compared with patients who have a minimum stent diameter of ⬎3.0 cm. Conclusion: The minimum diameter of the stent implanted appears to be strongly associated with the likelihood of repeat revascularization of the target lesion. Additional research is needed to examine whether drug-eluting stents can reduce the target lesion revascularization rate in patients with small vessel size.

TCT-217 Clinical Course of Patients with Chest Pain After Successful Implantation of a Stent in Acute Coronary Syndromes. J.J. Arango, M.F. Villegas, J.G. Vela´ squez, C. Arana, A. De la Pava, M. Badiel, F. Rosso. Laboratorio de Cardiologı´a Intervencionista. Fundacio´n Valle del Lili, Cali, Colombia. Background: Coronary stenting improved the safety of percutaneous coronary interventions and decreased early complications in acute coronary syndromes (ACS). Nevertheless, the presence of chest pain (CP) after successful stent implantation creates uncertainty in the cardiologist because of the possibility of an acute or subacute stent thrombosis. Our goal was to show the frequency and clinical course of CP after a recently successful stent implantation. Methods: Between January and December of 2001 in the interventional cardiac laboratory in Fundacio´ n Valle del Lili, Cali, Columbia, approximately 1,800 procedures were performed. We selected patients with a diagnosis of ACS who had implanted intracoronary stents as treatment. All patients received aspirin and heparin before the procedure. At the beginning of the procedure, clopidogrel was administred. Use of glycoprotein IIb/IIIa was at the discretion of the cardiologist. Demographics data, presence of CP during the hospitalization and 30 days’ follow-up, complications, and target vessel revascularization (TVR), or new catheterization diagnosis were collected. Results: A total of 187 patients met the selection criteria; data were obtained in 163 (73.5% men). The incidence of CP was 17.2% (n ⫽ 28), only 27.3% (n ⫽ 6) needed a new diagnosis catheterization, and in 1 patient there was a need for repeat TVR. At 30 days’ follow-up, 9 of 28 (32.1%) patients presenting with pain during hospitalization had a repeat symptom. At 30 days, 11.85% (16 of 135) of those who did not present with an episode of CP during hospitalization presented with CP in follow-up, but only half of them (8 of 16) required a new catheterization diagnosis. Compared with those who did not present with CP during their hospitalization, CP was more frequent in patients with diabetes (39.3% vs 20%, p ⬍0.03) and dyslipidemia (75% vs 57.7%, p ⫽ 0.07). In patients without CP after PCI, there was a greater proportion of smokers (44.4% vs 25%, p ⫽ 0.05). Conclusion: Although the incidence of CP after stent implantation is high (17.2%), the clinical course seems to be at goal because only 3.57% (1 of 28) required a repeat TVR. CP presented in patients with diabetes more frequently.

TCT-218 Total Occlusion Due to Diffuse In-Stent Restenosis: Is Brachytherapy a Solution? E. Nikolsky, L. Gruberg, A. Huber, E. Rosenblatt, E. Grenadier, M. Boulos, Z. Bernstein, A. Roguin, M.

The American Journal of Cardiology姞

PM

(Main Lobby)

Suleiman, R. Gitman, R. Bar-Deroma, W. Markiewicz, R. Beyar. Department of Invasive Cardiology and Brachytherapy Unit, Rambam Medical Center and Faculty of Medicine, Technion-Israel Institute of Technology, Haifa, Israel. Background: Randomized and observational trials consistently demonstrated the effectiveness of intracoronary brachytherapy (ICBT) in preventing recurrence after in-stent restenosis. However, few data are available to suggest that ICBT is likewise effective in a subset of patients with totally occluded in-stent restenosis lesions. Our aim was to assess the long-term outcome of patients treated with intracoronary ␥-radiation for totally occluded (TO) in-stent restenosis lesions. Methods: Percutaneous coronary intervention and subsequent catheter-based irradiation with 192Ir was performed in 100 patients (mean age, 61 ⫾ 10 years), in 103 vessels with diffuse in-stent restenosis. At baseline, TO of the target vessel at the site of stent (TIMI flow 0-1) was present in 15 patients (14.5%). Follow-up data were collected during hospital visits and from telephone interviews. Repeat coronary angiography was performed in patients with clinical restenosis. Results: Patients with TO of the target vessel at baseline compared with patients without TO had a trend for higher rates of unstable angina at the index procedure (66.7% vs 41.4%, respectively; p ⫽ 0.12). There was a more frequent use of longer radiation source (14 seeds, covering a length of 55 mm) in patients with TO versus patients without TO (53.3% vs 24.1%; p ⫽ 0.04). With mean follow-up of 47.5 ⫾ 24.0 months, major adverse cardiac events (MACE) were observed in 9 of 15 patients (53.3%) with TO compared with 26 of 88 patients (29.5%) without TO. According to multivariate analysis, TO of target vessel at baseline was the single independent predictor of MACE at 1-year follow-up (relative risk, 16.2; 95% confidence interval, 4.2– 62.9; p ⬍0.0001). Conclusion: The efficacy of ␥-radiation in totally occluded in-stent restenosis lesions is significantly worse compared to the non-occlusive lesions. TO is an independent predictor of MACE at follow-up.

TCT-219 C-Reactive Protein: A Clinical Restenosis Predictor. A. Damonte, G. Zapata, F. Kozak, S. Diangelo, Y. Arzani, E. Picabea. Instituto Cardiovascular de Rosario. Rosario, SF, Argentina. Background: Recent reports have demonstrated elevation of C-reactive protein (CRP) after elective and successful coronary stenting. However, little is known about the relation between CRP elevation after stent deployment and clinical restenosis. The purpose of this study was to evaluate whether the CRP level at 48 hours post stenting is a predictor of clinical restenosis at 6 months’ follow-up. Methods: A total of 64 consecutive patients with chronic stable angina or stabilized unstable angina and single-vessel disease, undergoing successful single-lesion stenting, were prospectively included in this study. Plasma CRP concentration was determined by the immunoturbidimetric method, before and 48 hours after stenting. The normal upper reference value for CRP with this method is up to 1 mg/dL. Patients were stratified in 3 terciles by the CRP concentration at 48 hours: ⬍1.2 mg/dL, 1.2–1.9 mg/dL, and ⬎1.9 mg/dL. Clinical restenosis at 6 months was defined as death, myocardial infarction related to the treated artery, target lesion revascularization, recurrent angina, and/or positive stress test.

SEPTEMBER 24, 2002

TCT ABSTRACTS/Poster

87H

P O S T E R A B S T R A C T S

TUESDAY 9/24/02 9:00 –10:00

AM

and 3:00 – 4:00

PM

(Main Lobby)

Results: CRP concentration increased from 0.8 ⫾ 0.77 mg/dL (95% confidence interval [CI], 0.7–1.08) before to 2.24 ⫾ 1.94 mg/dL (95% CI, 1.71–2.77) after procedure (p ⫽ 0.03). At 6 months, clinical restenosis occurred in 14 of 64 patients (21.8%). According to the 3 terciles of CRP level at 48 hours, clinical restenosis was: ⬍1.2 mg/dL (2 of 22, 9.09%); 1.2–1.9 mg/dL (4 of 20, 20%; relative risk [RR], 2.2;

95% CI, 0.45–10.7; ⬎1.9 mg/dL (8 of 22, 36.6%; RR, 4.0; 95% CI, 0.95–16.7); trend test, p ⫽ 0.029. Conclusion: Our finding suggests that higher levels of CRP 48 hours after a successful stenting procedure are predictive of clinical restenosis at 6 months’ follow-up. Inflammation appears to play a role in the development of clinical restenosis post procedure.

88H

SEPTEMBER 24, 2002

P O S T E R A B S T R A C T S

The American Journal of Cardiology姞

TCT ABSTRACTS/Poster