Keywords
INTRODUCTION
One of the main challenges facing coronary interventions is the restenosis rate which entails the consequent need for revascularization during follow-up. In our context, percutaneous reinterventions account for 9%-10%1 of total interventions. This figure is an underestimation of clinical restenosis, since it does not include patients who are sent for revascularization surgery after repeat or non-treatable restenosis, nor those who present recurrence of angina symptoms but are not sent for recatheterization.
In particular groups of patients, such as diabetics,2 and in certain specific lesions such as diffuse lesions,3,4 lesions in small vessels,59 total occlusions1013, and in-stent restenosis,14 the need for a new intervention during follow-up is much higher, amounting to around 10%-30%.
However, treatment of in-stent restenosis continues to be a problem, since standard strategies, such as the use of balloons and stents, have a relatively high incidence of restenosis.14 The technique that seems to yield better results in this context, brachytherapy, is not generally available due to its logistic and operational complexity.15,16
Sirolimus- and paclitaxel-eluting stents have proven to strongly reduce restenosis rates in lesions with a light to moderate risk of restenosis.17-20 Little information exists regarding their safety and the benefits these stents can offer in higher-risk lesions, although some results have been reported in a small series of in-stent restenosis patients.21,22 In addition, there is some data related to their use in the clinical context of acute coronary syndrome, including ST-elevation acute myocardial infarction.23,24
We evaluated the impact of stent use on clinical restenosis in a group of patients who had lesions with a high risk of restenosis.
PATIENTS AND METHODS
From 1 June 2002 onwards, we included all patients sent to our unit with clinical indications for coronary angiography who were candidates for percutaneous revascularization, and in whom some or all of the lesions fulfilled at least one of the following criteria:
1. In-stent restenosis.
2. Total occlusion, however old.
3. Diffuse lesion (>20 mm).
4. Lesion in small vessel (2.5 mm).
Patients undergoing primary angioplasty and those in cardiogenic shock were excluded. If the same patient presented other target lesions that did not fulfill these criteria, a sirolimus-eluting stent was also used. Whether to implant the stent directly or not was left to the discretion of the surgeon, although a direct implant was never attempted in cases of total occlusion, where there was a need for a stent >23 mm and in vessels with evident calcification or marked angulation or tortuosity.
The procedure was done via the femoral artery and in all cases vascular closing devices were utilized. The use of glycoprotein IIb/IIIa inhibitors remained the surgeon's decision. Angiography was considered successful when a residual stenosis <25% with TIMI III flow was achieved. Serial determination of enzymes was not performed systematically, unless there was electrocardiographic or clinical suspicion of necrosis following percutaneous coronary transluminal angioplasty (PTCA) (complications during procedure, ischemia and post-PTCA symptoms). All patients received 300 mg clopidogrel orally immediately after the procedure and, subsequently, 100 mg acetylsalicylic acid and 75 mg clopidogrel in combination for 3 months.
Clinical follow-up was done by checking medical records and a telephone call to all patients. They were monitored by their respective clinical cardiologists, who requested ischemia tests or coronary angiography in those patients for whom they considered it advisable.
Statistical Analysis
Continuous variables are expressed as mean±SD and categorical variables as percentages. We obtained a Kaplan-Meier curve for event-free survival. Cox regression analysis was done to identify event predictor variables during follow-up. A value of P<.05 was considered statistically significant. The statistical package SPSS 11.0 was used.
RESULTS
Between June 2002 and December 2002, 100 consecutive patients were included in the study in whom 154 lesions were treated. This series forms 28% of the total number of patients treated in the unit during this period. The clinical characteristics of these patients are presented in Table 1 where the high rates of previous revascularization (31% PTCA, 11% surgery) stand out. The angiographic characteristics of the 154 lesions are described in Table 2. Of these, 109 (71%) fulfilled some or several of the selection criteria and, as required by the criteria, every patient had at least one of these high-risk lesions. Among the 45 target lesions that did not fulfill the criteria, there were some (3 ostial lesions and 6 bifurcational) that could also be regarded as having a high risk of restenosis, although these were not the focus of the study. With regard to total occlusions, of the 31 lesions, 10 were recent, according to clinical data (recent acute coronary syndrome, no primary angioplasty), and the remaining 21 were chronic. Aspects relating to the revascularization procedure and sizes of the stents used are detailed in Table 3. The total length of stents implanted per patient was quite high at 33±16 mm. The abciximab use rate (16%) was similar to that in our experience with standard stents. We attempted to apply stents directly in 55 lesions, were successful in 52 (94.5%) cases and three needed predilatation. Angiographic success was achieved in 98% of the patients. Of the remaining two cases, one failed due to post-stent TIMI II flow (poor distal bed and competition with collateral circulation), and the other to residual stenosis >25% (fibrocalcified lesion).
Procedural Complications and Hospitalization
Two non-Q wave infarctions occurred (2%). One was due to occlusive retrograde dissection after implanting a stent in the medial right coronary artery and was resolved with an additional stent. The other was caused by the occlusion of a small diagonal branch (diameter <2 mm) after implanting a stent in the anterior descending artery. A transitory ischemic cerebral attack (1%) occurred in a hypertensive 66-year-old male patient (Table 4).
Clinical Follow-Up
During the follow-up period of 8.5±2 months (6-12 months), during which no patient was lost, two late thromboses occurred (2%) one of which caused an infarction. Three target lesion revascularizations were done (3%), two of which corresponded to the late thromboses mentioned before (Table 4). At the end of follow-up, only 2 patients (2%) presented stable angina class II symptoms but were not sent for further tests due to effective medical control. The event-free survival curve (death, infarction or revascularization of the target lesion) is shown in the Figure.
Fig. 1. Kaplan-Meier curve of event-free survival (death, infarction or revascularization of the target lesion).
With regard to late thrombosis, one occurred in a 76-year-old male patient who suffered previous occlusive and diffuse restenosis with a standard stent placed in the proximal-medial part of the anterior descending artery. The distal bed of the vessel was diffusely diseased and was visualized perfectly through the large collateral vessels from the posterior interventricular branch of the right coronary artery. After predilatation, a Cypher® stent was implanted (length, 33 mm; diameter, 2.75 mm), without ever achieving adequate flow (TIMI II). At 3 months, and immediately after suspending clopidogrel use, the patient began to present unstable angina and high troponin I levels. Coronary angiography revealed total occlusion of the stent, with the vessel distal to the stent filled because of excellent collateral circulation. Surgical revascularization from the left internal mammary artery to the anterior descending artery was successfully done without extracorporeal circulation.
The other case also involved a 76-year-old patient and occurred 7 months after an initial procedure in which a 33-mm stent was implanted in a de novo lesion in the proximal-medial anterior descending artery. Coronary angiography showed an occlusive lesion located immediately distal to the stent (which was 50% in the initial study) in the medial anterior descending artery. This had produced a thrombosis extending toward the stent. After the thrombotic material was extracted with an X-Sizer® device, an angiography revealed a certain degree of focal restenosis in the stent. A stent was implanted again following angioplasty.
A 45-year-old diabetic patient treated with insulin presented progressive effort angina 5 months after the procedure. She initially presented three-vessel disease with vascular beds in poor condition for surgery. Percutaneous revascularization was therefore performed with Cypher® stents implanted in the anterior descending artery (3/33 mm and 3/8 mm), in the circumflex artery (2.5/33 mm), and in the right artery (3/18 mm). Coronary angiography carried out 5 months later due to recurrence of symptoms showed two non-marginal in-stent restenotic lesions, severe and focal (<5 mm) in the anterior descending artery and the circumflex artery. Cutting-balloon angioplasty was done in both lesions. Four months after the second procedure, the patient again presented progressive effort angina. Coronary angiography showed focal restenosis in the same locations in the circumflex artery stent, as well as others more distally located in the anterior descending artery stent which were treated by the in-stent implantation of short Cypher® stents.
Finally, 3 patients needed a new coronary angiogram for recurrent symptoms. In one patient, progression was observed in a previously moderate lesion in the target vessel but far from the stent (>10 mm), and in the other 2 patients there were no changes in the coronary tree. A new angioplasty was done in the first patient. The complete absence of in-stent and stent margin restenosis was striking in these patients, which is a relatively infrequent situation with standard stents.
Regarding the 15 patients who had an initial diagnosis of silent ischemia (tests strongly positive for ischemia under treatment, but without associated symptoms), tests for ischemia were done during follow-up with negative results in most cases (13 patients). In 2 patients a far smaller degree of positivity without medication was detected which did not constitute an indication for coronary angiography. According to clinical criteria, the patients presented a certain degree of diffuse atheromatosis and this could explain the mild residual ischemia.
DISCUSSION
Percutaneous revascularization has yielded good results due to technical and pharmacological progress in recent years, both in terms of safety and mid- and long-term clinical benefits. The basic restriction of this therapy is the appearance of restenosis and the consequent need for new revascularization in the mid-term. With the current rate of stent implantation (80%), the total percentage of angioplasty reinterventions is around 9%-10%.1 This figure refers to clinical restenosis, which is always lower than the rate of angiographic restenosis, but it is this which really determines the impact of this event in clinical practice. Nevertheless, this does not include the proportion of patients who undergo surgical reinterventions, only those undergoing percutaneous ones. In our experience, 14% of patients with clinical restenosis will have to undergo surgical procedures, either after the first restenosis or the second. In some lesions, such as diffuse ones,3,4 lesions in small vessels,5-9 total occlusion,10-13 and especially, in-stent restenosis,14 clinical restenosis is significantly higher (10%-30%).
Thus, intense research has recently been geared toward developing methods for combating restenosis. Basic lines of work have focused on brachytherapy, pharmacology and coated stents. Brachytherapy has demonstrated its efficacy,15,16 although aspects relating to its safety (late-stent thrombosis, edge restenosis, etc), application, and logistics have strongly limited its becoming more widespread. Some systemic drugs, such as tranilast, although having undergone promising pilot studies have not demonstrated their efficacy in larger tests.25 Finally, drug-free stents offer the greatest potential and have already yielded excellent results. Nevertheless, the large studies done to date with sirolimus- and paclitaxel-eluting stents have only evaluated lesions with a light- to moderate risk of restenosis and have excluded lesions >30 mm, total occlusion, ostial lesions, lesions requiring stents <2.5 mm or in-stent restenosis.18-20 The efficacy of coated stents remains unknown in these lesions, although small studies with a short follow-up have been published that suggest good results, specifically for in-stent restenosis.21,22 In view of this, from the moment the Cypher® stent (sirolimus-eluting) was made available, we began a study to evaluate clinical restenosis in those patients who had lesions which presented characteristics more prone to restenosis--i.e., diffuse lesions, those in small vessels, total occlusion and in-stent restenosis. In some of these cases, especially diffuse in-stent restenosis, it is probable that in the absence of these stents the patients would have undergone revascularization surgery. Thus, we studied a high-risk group with a high rate of multiple lesions and multivessel revascularization, and who had undergone previous procedures. Seventy-one percent of the lesions treated in these patients fulfilled at least one of the inclusion criteria that involved a greater risk of restenosis.
No incidence of acute thrombosis or subacute thrombosis was observed. The first episode of late thrombosis involved a patient with a non-optimal initial result (TIMI II flow, competition from direct collateral circulation) and in whom the occlusion occurred after stopping clopidogrel use (3 months after the procedure); however, due to excellent collateral circulation, it manifested as unstable angina and not as infarction. The other case, described as a late thrombosis that appeared as an infarction at 7 months, raises certain doubts. This case appeared to be a late thrombosis, but we believe that, given the angiographic image of the lesion distal to the stent and the elapsed time (7 months), it could also be explained by a complication involving the distal lesion and consequent thrombosis retrograde to the occlusion that would have affected the stent.
Finally, the last case involved clear clinical restenosis in a diabetic patient treated with insulin with multivessel and severe diffuse coronary disease. Initially, a 92-mm-long stent was implanted. Based on our previous experience, we believe that, given the clinical and anatomical characteristics of the patient, the restenosis would have been more diffuse and severe had we used standard stents. Nevertheless, this case shows that restenosis can in some cases continue to be a problem despite advances in new therapies, although it is a far less frequent problem and easier to manage (focal restenosis).
Without doubt, the 3% of patients who underwent new target lesion revascularizations during follow-up (including thrombosis cases)--with 2% of patients with class II stable angina at the end of follow-up--is a spectacular result in a group of patients with lesions having the characteristics described above. Furthermore, these would be good results even for a population with non-selected lesions.
The group of 26 patients in which 31 in-stent restenoses were treated presented the most striking results. Only one patient (3.8%) presented events during follow-up (the late thrombosis case mentioned previously). Based on our previous experience, approximately one-third of patients treated with standard stents or balloons for in-stent restenotic lesions are readmitted with new clinical restenosis. In the RIBS study, in which 450 cases of non-occlusive in-stent restenosis were randomized to treatment with balloons or stents, the revascularization rate of the target vessel per year was 19.6% in the stent group and 24.3% in the balloon group.14
The antiplatelet therapy period varies from study to study: this was 2 months in the RAVEL study,18 3 months in the SIRIUS study,19 2-4 months in the Rotterdam group for in-stent restenosis study21 and 3 months in the acute coronary syndrome study (6 months if there were patients treated with more than 3 stents, >36 mm stents, bifurcational or in-stent restenosis).23,24 In our case, treatment was carried out for 3 months but, based on empirical evidence, a 6-month period might be advisable in cases of in-stent restenosis, long stents, multiple stents, etc. The two late thrombosis cases described above may support this position which is the one we currently adopt.
The possible drawbacks of prolonging combined acetylsalicylic acid and clopidogrel therapy for periods of 3 months or more need addressing. Regarding safety, we did not observe significant hemorrhages or allergic or hematological reactions. However, it should be mentioned that more than half of the revascularized patients previously presented anginal instability and we know that with or without revascularization these patients benefit from a prolonged period (>3 months) of combined antiplatelet treatment.26,27 On the other hand, more recent studies demonstrate that combined antiplatelet treatment for a year leads to the reduction of ischemic events in patients who have undergone angioplasty with or without diagnosed unstable angina.28 Thus, the benefits of combination therapy are doubled with the prevention of stent thrombosis being added to the reduction of new events in different coronary lesions.
Limitations
This was an observational study describing the results obtained in a selected consecutive nonrandom series of patients which, accordingly, lacks a control group. In addition, angiographic follow-up was not done. Although this detracts from the value from the study, it should be mentioned that the study was set within the context of clinical practice. As such, our interest lay in investigating the impact of the use of these stents, their safety (incidence of thrombosis), reducing the need for new reinterventions, and the clinical status of the patient at the end of a reasonable clinical follow-up period.
These striking and unprecedented results are obviously in contrast with all the previous experience reported with standard stents in similar lesions3-14 and with our own experience, thus making it possible to draw very positive conclusions.
It could be argued that using the need for new revascularization as the deciding criterion introduces subjectivity into the result. However, the direct and independent evaluation of the clinical status of the patients at the end of follow-up confirms that this aim was realistic, since only two patients presented stable angina class II at the end of this period.
CONCLUSIONS
The use of sirolimus-eluting stents in lesions with a high risk of restenosis (diffuse, vessel 2.5 mm, total occlusion and in-stent restenosis) appears to be safe and very effective. Taking the limitations of this study into account, a larger series of patients with random comparisons with standard stents is needed to consolidate these findings with greater certainty.
Correspondence: Dr. J.M. de la Torre Hernández.
Unidad de Hemodinámica y Cardiología Intervencionista.
Hospital Universitario Marqués de Valdecilla.
Avda. Valdecilla, s/n. 39008 Santander. España.
E-mail: hemodinamica@humv.es