Clinical and Angiographic Outcomes after Tsunami‚Ñ¢ Coronary Stent Placement

ABSTRACT: Study aim: The aim of this study was to determine six months restenosis rate measured by off-line quantitative coronary angiography, and to assess the efficacy and safety of newly developed six cells Tsunami™ stent system for the treatment of symptomatic, de novo, native coronary artery lesions. Study design and methods: Between January and August 2001, 100 patients with 120 lesions were included at four clinical sites in this prospective, non-randomised, single arm study. In 35% of the patients two or more stents were implanted, 15% of the lesions were stented with more than one stent and 46% of the lesions were in small vessels (1–8 The importance of the stent design has been recognised even with the most advanced technology of drug eluting stents.⁹ Recently, a new, six-cell stent system, the Tsunami™ (Terumo Corp., Tokyo, Japan) has been introduced in the European market. Recognising the importance of careful clinical evaluation of any new product we conducted a multicenter prospective, single arm study to evaluate restenosis rate, safety and efficacy of this stent, at six months clinical and angiographic follow-up. Methods Patient population. From January to August 2001, 100 patients with up to two de novo coronary artery lesions, were treated with Tsunami™ stents at four European clinical sites. Criteria for inclusion comprised age >= 20 years, angina pectoris or objective evidence of ischemia, target lesions suitable to be treated with available stent lengths (10, 15, 20 and 30 mm) and diameters (3.0, 3.5, and 4.0 mm), target lesion stenosis >= 50%. Myocardial infarction within three days of the procedure, pregnancy, known allergies to aspirin, clopidogrel or heparin, unprotected left main coronary disease, angiographic evidence of thrombus that cannot be resolved by therapy, occupying >= 50% of the target vessel, ejection fraction Stent procedure. Coronary angiography and intervention were performed according to the standard clinical practice in the investigation sites. Choice of femoral or radial approach and predilatation were left to the investigator’s discretion. Patients received Aspirin 80–300 mg per day, and 300 mg of clopidogrel per day, starting at least 12 hours before the procedure. During coronary intervention, patients received an initial intravenous bolus of heparin (70–100 IU/kg), supplemented by additional heparin administration, if needed, to maintain activated clotting times (ACT) > 300 seconds. It was recommended to discontinue heparin immediately following the procedure to allow early sheath removal. Immediately after the procedure and for at least four weeks thereafter, patients were medicated with clopidogrel, 75 mg, once daily, and aspirin, 80–150 mg per day for at least six months. Intravenous administration of GP IIb/IIIa inhibitors was at the discretion of the physician. The stent implantation procedure was performed using predilatation with the balloon or by means of a direct stenting technique. Post-dilatation was recommended if an optimal stent vessel ratio of 1.1:1 was not achieved. Definitions. The following definitions were used in this registry to describe treatment outcomes: Angiographic success: final result 10 A non-Q-wave MI was defined as creatinine kinase (CK) enzyme elevation more than twice the upper normal limit with concomitant elevation of CK-MB. All deaths were to be considered cardiac unless otherwise documented. Bleeding was considered major if it lead to lethal outcome or permanent disability, when it produced a fall in haemoglobin of at least 3 mmol/L and required transfusion of two or more units of red cells or whole blood, and if peripheral vascular surgery was required. Stent thrombosis was defined as angiographic documentation of occlusion of the stented lesion. It was defined as acute if it took place during the procedure, and subacute when occurring between the removal of the guiding catheter and one month following the procedure. Restenosis in this study was defined as in-stent restenosis. Quantitative coronary angiography (QCA). Quantitative coronary angiograms obtained at baseline, at the completion of stent procedure and at six months follow-up, were submitted to the independent core laboratory (Cardialysis, Rotterdam, The Netherlands) and were analysed by the Cardiovascular Angiography Analysis System. Baseline angiographic variables included reference vessel (RVD) and minimal lumen diameter (MLD), percent diameter stenosis, lesion length, and complexity by American Heart Association/American College of Cardiology (AHA/ACC) classification. Postprocedural, luminal diameter stenosis, RVD, MLD, acute gain (final MLD minus pre MLD) and the presence of the dissection were noted. Follow-up angiographic variables included MLD, in-stent percent diameter stenosis, incidence of the binary (>= 50%) restenosis, late loss (final MLD minus follow-up MLD) and loss index (calculated ratio of late loss and acute gain). Endpoints of the registry. Primary endpoint was the incidence of the binary in-stent restenosis at six months follow-up as determined by off-line QCA. Secondary endpoints were effectiveness and safety. Endpoints of effectiveness were angiographic and procedural success, duration of hospitalisation, follow-up MLD, late loss, total gain, percent diameter stenosis, target vessel failure (TVF) and target vessel revascularisation (TVR). Endpoints of safety were in-hospital and out of hospital complications, MACE at 30 days and six months, bleeding and vascular complications, stent thrombosis, and survival rate at 30 days and six months by Kaplan-Meier method. Statistical analysis. Statistical analysis was performed with SAS for Windows version 8.02 (SAS Institute, Cary, North Carolina). All the analyses were based on the intention to treat population. Continuous variables were presented as means with standard deviation (SD) or median and range. Categorical variables were presented as numbers and rates and were compared using chi-square test, Fishers exact test or Cochron-Mantel Haenszel test. QCA data were presented for all patients with matched angiographic data, pre- and post-procedure and at follow-up. Logistic regression analysis was used to assess clinical and angiographic predictors of angiographic restenosis and the occurrence of the target vessel failure (TVF). Wilcoxon Rank Sum test was used to compare two groups when variables were not normally distributed. If more than two groups were compared, the Kruskal-Wallis test was used. For all performed analyses differences were considered statistically significant at p Procedural characteristics. Tsunami™ stent implantation was not successful in five cases due to the following reasons (categories not mutually exclusive): guidewire did not cross the lesion (1), stent device did not cross the lesion (2), dissection/occlusion not repaired by bail-out procedure (1), or stents dislodged from the balloon in an attempt to cross tortuous and calcified lesions (2). Also, the attempts with non-study stents failed and the lesions were treated with balloons only. The average RVD for those lesions was 2.1 mm (range 1.94–2.29 mm), while the smallest available stent in this study was 3 mm. Two patients experienced Q wave MI, both with the stents implanted into small vessels (RVD 2.28 mm and 2.68 mm). One MI was caused by subacute stent thrombosis, shortly after the patient had been transferred to the hospital ward. Due to the small size of the artery, underexpansion of the stent was considered as a possible cause of this event, but no IVUS was available to confirm it. This patient did not receive intravenous glycoprotein IIb/IIIa inhibitors. The occlusion was successfully treated by balloon angioplasty, and the patient left hospital without anginal symptoms. The second MI occurred in a patient with multivessel disease during the passage of the guide wire through the first stented lesion in an attempt to treat the second lesion. A total occlusion occurred distally to the first implanted stent. This patient did not receive clopidogrel because of allergy. The patient was successfully treated by the implantation of a second stent in the occluded segment. One patient experienced major bleeding with haematoma on the puncture site requiring blood transfusion and vascular intervention. There were no further in hospital events. In 35% of the patients more than one stent was implanted, and 15% of the lesions were treated with more than one stent. Fifty six percent of the stents were implanted by direct stenting, which was successful in 87% of attempts. Postdilatation was required in 27% of the procedures, 11% after direct stenting and 16% in the procedures that followed predilatation. Postdilatation was performed with Tsunami™ balloon in 10% of the cases and in 17% with an additional balloon. Angiographic success rate was achieved in 100% of the stented lesions (residual diameter stenosis Angiographic characteristics. The baseline angiographic characteristics are given in Table 2. Although it had not been anticipated at the start of the study, we observed that 46% of the lesions were in the small vessels (RVD Angiographic follow-up results. Angiographic follow-up at six months was available for 81 patients, 94 lesions. Angiographic in-stent restenosis rate was 13.8% (Table 2). Late luminal loss, loss index and restenosis decreased by vessel size (Table 3). A separate analysis was performed for single stented lesions, after their categorical assignment into different groups by RVD and stent length (Tables 4a, 4b and 4c). The restenosis rate for single stented lesions was 7.4% and we observed a trend toward reduced restenosis by increasing RVD. Acute gain was significantly lower and loss index significantly higher in small ( 2.75) vessels. Restenosis rate in long stents (20 and 30 mm) was 7.7% and did not significantly differ from restenosis rate in short (10 mm) stents (4.3%). Loss index in short and long stents was similar. Clinical follow-up. No further clinical events were observed between the procedure and the one-month follow-up, thus, by cumulative frequency, 98% of the patients remained MACE free (Table 5). At six months, 96% of the patients were available for clinical follow-up. There were no deaths in this registry. No additional myocardial infarctions occurred during the observed period (clinical and angiographic follow-up intervals were 189 ± 33 and 186 ± 32 days, respectively). Seven Re-PTCA of the target lesion were performed, five out of seven (71%) in the small vessel sub-population. Four patients underwent elective CABG procedure for target vessel, three out of four (75%) in small vessels. Two out of four patients with CABG were diabetics, and one patient did not have a stent implanted. The latter patient was included into the analysis since it was done on the intention to treat basis. Target vessel failure in small vessels was 16.3% (including CABG patient who did not receive stent), while in vessels larger than 2.75 mm, the TVF rate was 4.6%. At six months clinical follow-up, overall 88% of the patients remained MACE free and 84.4% of the patients were angina-free. Logistic regression analysis.By multivariable modelling, in a logistic regression, we identified lesion length, multiple stenting, RVD and hypertension as independent predictors of restenosis. There were four independent predictors of target vessel failure identified in the logistic regression model. The most powerful determinant was RVD, followed by multiple stenting, diabetes, and lesion length (Table 6). Discussion The Tsunami™ stent system in our study proved to be a safe and efficient tool for the percutaneous treatment of coronary artery disease. The primary end-point of binary restenosis rate was reached in 13.8% of the lesions for the total population and 7.4% for the lesions treated with a single stent. These findings are well in line with other published series,^11–19 particularly considering that more than 46% of the stented lesions were in the small vessels (2,3 Angiographic restenosis results, however, must be interpreted with caution because angiographic follow-up was not available for all the patients (81% of patients, 78% of lesions). It has been reported that angiographic restenosis tends to be overestimated when the angiographic follow-up compliance is below 80%,²⁰ therefore, we consider it unlikely that results would be worse if the missing patients could be assessed by angiogram. No mortality and only two patients with MI highlight the good safety profile of the Tsunami™ stent system. Both MI patients were successfully treated and remained angina free at six months follow-up. Several reports^21–25 have identified small vessels as a risk factor for MACE and in our group of patients most of the MACE occurred in this subpopulation. The subanalysis by vessel size revealed that, at 210 days, only 3 (4.6%) out of 65 lesions within the vessels larger than 2.75 mm required any revascularisation, resulting in 95.4% freedom from TVF. Six months event free survival in the small vessel subgroup in the present study is comparable to the other studies enrolling patients with similar lesion characteristics.^21–25 In this study, 7% of all the patients received intravenous GPIIb/IIIa inhibitors. Based on the recent reports^22,26–28 showing very good short and long term outcomes when small and complex lesion were treated in combination with GPIIb/IIIa receptor antagonists, we can speculate that our results could further be improved if applying this approach to the higher risk patients. Critical factors to predict restenosis and target vessel failure identified in our study are in agreement with previously published series.^25,29–32 They include lesion length multiple stenting of a single lesion, RVD and hypertension, for restenosis, and RVD, multiple stenting, diabetes and lesion length, for target vessel failure. Unsuccessful implantation of Tsunami™ stent in this registry was 4.6% and it could be related to the tortuosity and calcification of the lesions located in very small vessels (average RVD of 2.1 mm), whose crossing was attempted with a relatively large stent (3 mm). Other, non-study, stents were not able to cross those lesions either, indicating that they might be unsuitable for this technique. High flexibility and trackability were the main characteristics of this stent recorded during the trial. The thin struts of the Tsunami™ stent may have some beneficial long-term effects, but on the other hand they may have compromised visibility during the procedure. Low crossing profile, and the balloon, well matched to the stent length, allowed direct stenting in most of the lesions selected for this technique. Study limitations. This is a single-arm study, with only a relatively small number of patients enrolled. Other limitations include the low percentage of diabetic patients and type C lesions, which could have an impact on the final results, and the absence of IVUS data, considered today as a useful tool for optimal stent placement and outcome analysis. Conclusion. Tsunami™ stent provides a safe and effective option for the treatment of coronary artery disease. The restenosis rate in this study is an encouraging finding that requires further confirmation in larger studies with more versatile patient/lesion population.