Cobalt-Chromium Stents in Long Lesions of Large Vessels

Abstract

Although drug-eluting stents have reduced the restenosis rate, some patients are not good candidates for prolonged double-antiplatelet therapy. Our goal was to learn the results of implanting cobalt-chromium stents in long lesions of large vessels.

All consecutive patients with ≥1 lesion treated with a cobalt-chromium stent ≥28 mm in length and ≥3.5 mm in diameter during a 12-month period in 2 centers were monitored clinically and angiographically to determine the clinical status, restenosis rate, pattern of restenosis, and need for revascularization.

The series comprised 78 patients with 81 lesions, mean age, 62.6 ± 15.2 yr; diabetes mellitus,19.2%; and primary or rescue intervention, 20.6%. Target vessels were the left anterior descending coronary artery, 12 (14.8%); left circumflex coronary artery, 12 (14.8%); and right coronary artery, 57 (70.4%). The mean lesion length was 35.18 ± 12.65 mm. The proximal reference diameter after percutaneous coronary intervention was 3.64 ± 0.54 mm; the distal, 3.2 ± 0.43 mm; and the mean, 3.42 ± 0.44 mm. The mean stent length was 38.05 ± 12.78 mm (range, 28–90 mm). The binary restenosis rate was 23% (15/64), with pattern IB in 2 cases, IC in 5, II in 7, and IV in 1. Revascularization was needed in 6 patients.

In treating long lesions of large vessels, we found that cobalt-chromium stents achieved a moderate rate of target-vessel restenosis and a low rate of repeat revascularization. The pattern of restenosis was focal in almost half of the cases, and, in most remaining cases, restenosis affected short segments of the vessel.

Drug-eluting stents (DESs) have achieved a dramatic reduction in the restenosis rate.¹ Although data provided by meta-analysis suggest a DES safety profile similar to that of bare-metal stents (BMSs) in long-term follow-up,^2,3 there are still some concerns about late stent thrombosis⁴ and the need for long-term double-antiplatelet therapy. Some patients are not good candidates for this prolonged therapy, because of planned noncardiac surgical procedures, a history of bleeding, a need for oral anticoagulation therapy, or predicted poor compliance.

The main predictors of in-stent restenosis are the final size of the area covered by the stent and the length of the stent. Kasaoka and colleagues⁵ showed in 1998 that the restenosis rate associated with BMSs diminished in increments of 21% per mm² of the final area covered by the stent. In another study,⁶ each increment of 10 mm in the length of the vessel covered by the stent increased the restenosis rate by 7.7%.

Although there are numerous reports on the outcomes of BMS placement in large vessels⁷ and in long lesions,⁸ there are, to date, few reported data on BMS placement in vessels with both characteristics. Moreover, the studies of BMS outcomes have in most cases bee n performed with stainless-steel stents. There have been hypotheses that stent design^9–12 and strut thickness^13,14 play roles in the restenosis rate. Drug-eluting stents and cobalt-chromium stents (CCSs) emerged within a few months of one another, so DES outcomes have overshadowed CCS outcomes.

The purpose of this study was to evaluate the clinical and angiographic results of CCSs in long lesions of large vessels, as well as the pattern of in-stent restenosis.

Methods

Design. This was a prospective study of a cohort of patients, which was performed in 2 different centers within a 12-month period, from January 2005 through November 2006.

Patients and Lesions. We included in the study all consecutive patients who received at least one CCS of ≥28 mm in length and ≥3.5 mm in diameter. The stents used were the 28-mm length VISION® (Abbott Vascular, part of Abbott Laboratories; Redwood City, Calif) and the 30-mm length Driver® (Medtronic, Inc; Minneapolis, Minn).

Due to the policy of the regional government where these 2 hospitals are located, the percentage of DES implantations could not exceed 40% of all stent implantations during the period of study; further, DESs were used in treating in-stent restenosis, long lesions in vessels ≤3 mm in diameter, and stenoses in ostial locations, vein grafts, left main coronary arteries, and bifurcations. All patients gave informed consent for percutaneous coronary intervention (PCI) and follow-up angiography. Currently, in the areas where these 2 hospitals are located, the maximum percentage of DES implantations cannot exceed 60% of all stent implantations.

Procedure. Decisions concerning radial or femoral access, glycoprotein IIb/IIIa inhibitors, and stent technique were made by the operator. In cases of residual stenosis, we performed postdilation of greater than 20% with noncompliant balloons; when more than 1 stent was used, the procedure was completed with balloon inflation in the overlapped segment. In accordance with the protocols of both hospitals, an electrocardiogram was performed at the end of the procedure, and troponin tests were performed 6 and 24 hours after PCI. In both centers, antiplatelet therapy consisted of clopidogrel 75 mg/d for 1 month and aspirin 100 mg/d indefinitely.

Follow-Up. All patients were examined in the cardiology office 6 months after PCI, and there was subsequent follow-up by telephone at 12 months. Angiographic follow-up was performed 9 months after PCI, unless it had been done previously due to recurrence of symptoms. If restenosis was found in an asymptomatic patient, a noninvasive test was indicated to determine the need for new revascularization. Symptomatic patients who refused angiographic follow-up were studied with noninvasive tests.

Quantitative Coronary Analysis. This analysis was performed off-line after the initial procedure and follow-up angiography. It was centralized in 1 of the 2 centers, and all measurements were performed by 1 physician. A contrast-filled catheter was used as the calibration standard, in conjunction with the previously validated Medis® system (Medis Medical Imaging Systems BV; Leiden, The Netherlands). The reference segment diameter was taken as the average diameter of user-defined, 5-mm-long angiographically normal segments proximal and distal to the stenosis, in the projection with the least foreshortening.

Definitions. Acute myocardial infarction was diagnosed when creatine kinase levels rose to ≥3 times the normal value or when troponin levels rose to ≥10 times normal. Angiographic success was defined as residual stenosis of <20%, with Thrombolysis in Myocardial Infarction (TIMI) 3 flow. Procedural success was defined as angiographic success without serious sequelae, such as death or Q-wave myocardial infarction. Restenosis, in follow-up angiography, was defined as ≥50% stenosis either in-stent or in the adjacent proximal or distal 10-mm segments. It was considered focal, diffuse, or proliferative, in accordance with Mehran and associates' classification.¹⁵ Acute gain was defined as the difference between minimal lumen diameter before and after PCI. Late loss was defined as the difference between minimal lumen diameter after PCI and at follow-up angiography. Poor candidates for prolonged double-antiplatelet therapy were patients with any limitation to follow-up treatment with aspirin and clopidogrel for more than 1 month, due to planned noncardiac surgical procedures, a history of bleeding, a need for oral anticoagulation therapy, or predictable poor compliance.

Statistical Analysis

Continuous variables with normal distribution are expressed as mean ± SD and qualitative variables as absolute value and percentage. The results were analyzed with SPSS version 15.0 software (SPSS Inc., an IBM company; Chicago, Ill).

Results

The series comprised 78 patients with 81 lesions. The mean age of those patients was 62.6 ± 15.2 yr, and 15 of them (19.2%) had diabetes mellitus. Primary or rescue PCI was the indication for PCI in 16 patients (20.5%), and acute coronary syndrome with or without ST elevation accounted for PCI in 56 patients (72%). Fifty-six patients (72%) had some limitation on prolonged antiplatelet therapy. Table I provides an overview of the clinical characteristics.

TABLE I. Clinical Characteristics of the 78 Patients

The right coronary artery was the treated vessel in 70.4% of all patients. The mean lesion length was 35.18 ± 12.65 mm, and the mean reference-vessel diameter was 3.17 ± 0.57 mm, with a mean difference of 0.5 mm between proximal and distal reference diameters. An angiographic image compatible with thrombus was present in 31% of the lesions. Chronic total occlusions made up 7.3% of the lesions, but TIMI 0 flow at the beginning of PCI was present in 29%, due to ST-elevation acute coronary syndrome in many of the patients. Table II shows the lesions' characteristics.

TABLE II. Characteristics of the 81 Lesions

Direct stenting was achieved in 27% of the lesions and the mean number of stents per patient was 1.59 ± 0.72. The mean reference-vessel diameter postintervention was 3.42 ± 0.44 mm, and the mean length of the vessel covered by stent was 38.05 ± 12.78 mm. All the interventions, including radial-access procedures, were performed with 6F guiding catheters. Table III shows the characteristics of the procedures. Figure 1 shows a lesion in the right coronary artery and Figure 2 the angiographic results after two 4 × 30-mm and one 4.5 × 30-mm Driver stents were deployed.

TABLE III. Procedural Characteristics in Repairing 81 Lesions

Fig. 1 Angiogram shows a long lesion in the proximal, mid, and distal segments of the right coronary artery.

Fig. 2 Angiographic result in the same patient, after implantation of two 4 × 30-mm and one 4.5 × 30-mm Driver stents.

There were 4 deaths during the 12-month follow-up period. Two occurred during the admission period—both during the 1st week after rescue PCI and both in elderly patients who had experienced prolonged delays in transfer from a secondary center. The 3rd death occurred 2 months after hospital discharge in a patient who had incomplete revascularization of 3-vessel disease, severe left ventricular dysfunction, and acute pulmonary edema. The last death occurred 8 months after the procedure due to a new myocardial infarction secondary to a new occlusion of a different vessel. No cases of definite or probable thrombosis were registered. The patient who died 2 months after hospital discharge could have experienced stent thrombosis, although his death, alternatively, could have been secondary to left ventricular dysfunction or to incomplete revascularization, as we have speculated above.

A new percutaneous intervention, with no surgical revascularization, was required in 6 cases (7.7%). Sixty-four lesions were angiographically reviewed, which represented 79% of all treated lesions, or 81% if we exclude the lesions of patients who died prematurely. The mean reference diameter was 3.42 ± 0.41 mm, and the mean stenosis at angiographic follow-up was 39.2% ± 22.72%; binary restenosis was found in 15 lesions (23.4%). In 7 of the 15 lesions, the pattern of restenosis was IB (Mehran's classification), while there were 7 pattern II cases and only 1 pattern IV. Finally, 4 of the 7 pattern II cases had affected segments shorter than 15 mm—despite their being considered diffuse because the restenotic segment was longer than 10 mm. Table IV shows the clinical and angiographic results, and Table V shows the pattern of restenosis. Among the 14 patients who refused angiographic follow-up due to the absence of symptoms, 8 underwent a stress test with a negative result, and the remaining 6 underwent stress echocardiography, also with a negative result.

TABLE IV. Clinical and Angiographic Results upon Follow-Up

Discussion

To our knowledge, this is the 1st series that studies specifically the clinical and angiographic results of CCS placement in long lesions of large vessels. The main findings are these: 1) the clinical results were satisfactory; 2) despite a relatively high degree of hyperplasia at follow-up and a moderate rate of binary restenosis, the need for repeat target-vessel revascularization remained low due to the large vessel diameter; and 3) in almost half of the binary restenotic lesions, restenosis took the form of pattern I.

Clinical Results. It has been suggested that clinical criteria are more useful than angiographic criteria as primary endpoints in the study of coronary angioplasty results.¹⁶ In our study, the incidence of stent thrombosis was very low. Only 1 case could have met the Academic Research Consortium's criteria for stent thrombosis, and this patient had other possible causes of death. Because the stented segments in our study were long, we might have expected higher rates of thrombosis. Probably the absence of antiproliferative drugs and the large diameter of the stents contributed to this low rate. The rate of repeat target-vessel revascularization due to clinical symptoms or positive noninvasive testing at follow-up in our series was only 7.7%.

Although in previous reports¹⁷ clinical restenosis represented one half of that seen upon angiography, this rate might be lower in large vessels. It has been shown that late loss is homogeneous all along the stent,¹⁸ and, for a given percentage of stenosis at follow-up, the minimal lumen diameter in large vessels might be large enough to avoid a significant compromise of coronary flow.¹⁹ Even in the event of a late loss of 1 mm, which is unusual with CCSs, a vessel of 3.5 mm in diameter would retain 2.5 mm of unobstructed lumen.⁷ Ormiston and colleagues,⁸ in a study that included only vessels of ≥3 mm in diameter and of >20 mm in lesion length, found an angiographic restenosis rate of 32%. The mean lesion length was 30.1 ± 13.5 mm, and the mean stent length was 35.8 ± 14.6 mm. Because only 8% of the lesions had angiographic restenosis ≥70%, the need for repeat revascularization did not exceed 12%.

Previous reports have confirmed that outcomes are satisfactory after conservative treatment of patients whose restenosis is <70%, and it has been proved that some lesions can even regress.^20–22 Consequently, the late-loss gap between CCS and DES might diminish in some lesions, especially in large vessels. In 1 study,²³ a sirolimus stent was compared with a CCS in vessels of 3 mm in diameter and lesion length of ≤28 mm; despite the fact that late loss was significantly lower in the DES, target-vessel failure at 12 months was similar. In fact, in comparison with BMSs, DESs behave as cost-effective devices only in vessels of <3 mm in diameter.7,18,24,25

Angiographic Results. The angiographic binary restenosis rate in our series was 23%. It has to be emphasized, however, that the rate probably was lower, because 14 completely asymptomatic patients refused angiographic follow-up. All of these patients were studied with noninvasive testing that did not induce ischemia. The mean late loss of luminal diameter in our study group as a whole was 0.84 ± 0.58 mm, which is 3 times higher than that reported for DES intervention with drugs such as sirolimus or everolimus, and twice as high as that reported for DES intervention with paclitaxel. In most cases, the large luminal diameter of the target vessels in our patients enabled this degree of hyperplasia to exist without clinical symptoms.

Two factors might have facilitated these satisfactory results. First, the proper design of the CCS might have played a role. The effect of strut thickness on restenosis rate has been shown repeatedly,^9–14,26 and the replacement of stainless steel by cobalt-chromium alloy has enabled a reduction in thickness to 80 µm in the Vision stent and to 91 µm in the Driver stent. In comparison with stainless-steel stents, cobalt-chromium stents have reduced the restenosis rate; and late loss found in comparisons of CCS with DES is much lower than late loss in comparisons of DES with stainless-steel stents.^23,27 In the initial studies of Vision and Driver,^28,29 the angiographic restenosis rate was 15%, and the rate of repeat target-vessel revascularization was only 6%.

The 2nd factor that might have contributed to successful results in our series is that we tried to adjust, as much as possible, stent length to lesion length. The restenosis rate is directly related to lesion length³⁰ and stent length, and the restenosis rate increases by 4% for each 10 mm of mismatch between these lengths.⁶ In previous studies, the implantation of excessively long stents (in relation to the lesions they cover)^28,31 might have increased the restenosis rate.⁶

No previous series has concentrated specifically upon the restenosis rate in long lesions of large vessels treated with BMSs, nor (even more especially) those treated with CCSs. However, 2 groups of investigators have published data provided by prediction models developed with variables that can influence the restenosis rate of stainless-steel stents. In an analysis using data from 4 studies of Multilink stents, Kereiakes and colleagues²⁸ published their model, which included the minimal lumen diameter postintervention, stent length, tobacco use, and diabetes; in stents ≥35 mm in length, the restenosis rate would be 16.1% for stents of 3.5 mm in diameter and 12.1% for those of 3.75 mm. No data were included for stents >3.75 mm. Using a model developed with clinical information from the MUSIC, WEST-II, ERASER, ESSEX, and TRAPIST studies, de Feyter and associates³⁰ related stent length and minimal lumen diameter after PCI to the expected restenosis rate in vascular segments covered by stents up to 60 mm in length and between 3 and 12 mm² in area.

Clinical Usefulness. The benefit of DESs in reducing the restenosis rate is unquestionable and is manifest especially in application to long lesions, where the advantages of DESs in comparison with BMSs are more visible. However, some patients should not undergo prolonged double-antiplatelet therapy due to planned noncardiac surgical procedures, prior history of bleeding, predicted poor compliance, or need of oral anticoagulation. In the last group, combining double-antiplatelet therapy with oral anticoagulation raises the risk of major bleeding during the 1st month to 2% and raises it an additional 2% to 3% per year thereafter, which amounts to a risk 5 times higher than the risk imparted by double-antiplatelet therapy alone.³² On the other hand, the benefits of DESs are lower in large vessels,^24,33 and the results of BMSs in lesions with low risk of restenosis are satisfactory.³⁴ The findings of our study may be relevant because they underscore the low restenosis rate of CCSs in these lesions. Moreover, many patients with restenosis and limitations on the use of DESs could benefit from a new PCI with CCS. As shown in Table V, almost half of our cases involved affected segments that were <10 mm in length and, in most of the remaining cases, the segments were <20 mm.

TABLE V. Patterns of Restenosis^*

Limitations of the Study

This study has some limitations. First, this was an observational study of a nonrandomized cohort from a 2-center registry, which subjects it to the possibility of bias. The results should be judged cautiously because of the small number of study participants and the small number of events. The lack of a control group comprising patients who received DESs prevents us from comparing the 2 types of devices. However, the purpose of the study was never to question the results of DESs; it was to determine whether the small percentage of patients with this type of lesion who cannot follow a double-antiplatelet therapy for prolonged periods of time might do well with CCS implantation as an alternative to surgical revascularization.

Although angiographic follow-up was not achieved in all cases, the restenosis rate was probably lower than it appeared, because the patients who refused repeat angiography were asymptomatic and presumably had no restenosis.

Finally, previous reports have shown the usefulness of intravascular ultrasound in achieving correct apposition, especially in placing long stents. Intravascular ultrasound was not used in our study, and our restenosis rate might have been higher as a consequence.

Conclusions

In treating long lesions of large vessels, cobalt-chromium stents achieved a moderate rate of target-vessel restenosis and a low rate of repeat revascularization. The pattern of restenosis was focal in almost half of the lesions, and, in most of the remaining cases, restenosis affected short segments of the vessel. These long stents may be considered useful in large vessels, especially in application to patients who should not undergo prolonged periods of antiplatelet therapy.