Keywords
INTRODUCTION
Randomized studies have shown that coronary stenting reduces the incidence of restenosis and improves the short- and long-term prognosis of balloon angioplasty (percutaneous transluminal coronary angioplasty, PTCA).1-3 More than 70%3 of percutaneous procedures performed in the United States4,5 and 45%6 to 77%7 of those done in Europe involve stent placement.
Stent implantation without predilation (direct stenting) is possible in selected lesions8,9 and results in a reduction in the use of consumables and contrast material, a decrease in radiation exposure, and a shorter duration of the procedure.10,11 The main limitation of this technique is the need for prior selection of candidate lesions. Almost 40% of the stents placed after predilation exhibit suboptimal intravascular ultrasound (IVUS) results, even when high pressure is used.12 This problem could be even more important with direct stenting, particularly if the favorable results extend the indications for this procedure.
Before the use of direct stenting is recommended for more complex lesions, it is essential to assess the results of this technique in terms of stent expansion. We conducted a randomized prospective study, initially postulating that no differences would be found in expansion between direct stenting and implantation with predilation. The primary aim was to compare the most well-validated IVUS expansion parameters observed between both techniques. The secondary aim was to assess potential clinical and angiographic differences in the follow-up.
PATIENTS AND METHODS
Selection of Lesions
For the purpose of obtaining study results similar to those from clinical practice, the interventional radiologist's opinion on the patient's eligibility for direct stenting in lesions with ≥70% stenosis, as estimated visually, was the only inclusion criteria. This selection was based on current knowledge of the technique, and was intended to avoid lesions with severe calcification or evident proximal tortuosity. Once the decision was made concerning eligibility for direct stenting, the patient was randomized by the sealed envelope method to implantation with predilation or to direct stenting. If a patient had two or more lesions treatable by direct stenting, all lesions were assigned to the same treatment group. The exclusion criteria were saphenous graft lesion, restenotic lesion, ostial location, myocardial infarction <24 h, lesion length >25 mm, impossible to treat with a ≥2.5 mm stent, left main coronary artery disease, scheduled heart surgery, allergy to aspirin, renal failure and follow-up impossible. All patients gave written informed consent to participate in the study, which was conducted in accordance with the Declaration of Helsinki.
For department-related logistical reasons, the study was planned for 100 consecutive lesions (50 per group) and the inclusion period was restricted to cases scheduled in the afternoon. At our hospital, the characteristics of patients treated in the afternoon are similar to those of patients treated in the morning.
Procedure
The procedures were done by three experienced interventional radiologists (>300 angioplasties per radiologist and year). Prior to surgery, all patients received aspirin (200 mg/day) and intravenous heparin to achieve an activated coagulation time above 250 s. The use of glycoprotein IIb/IIIa-receptor inhibitors was decided by the radiologist. All patients were given systemic ticlopidine (loading dose of 20 mg during the procedure and 250 mg/12 h over the next four weeks) or clopidogrel (loading dose of 300 mg and 75 mg/day for four weeks after the procedure). The use of post-dilation and the decision to end the procedure were based on angiographic criteria. Only second- and third-generation tubular stents (Multilink Tri-Star, Duet or Tetra, Guidant Inc., Temecula, California, USA; NIR Primo, and SOX, Scimed, Boston Scientific, Maple Grove, Minnesota, USA) were used. Lesions that could not be treated with stents of 2.5 to 4 mm diameter were excluded. Furthermore, only lesions allowing the use of stents ≤25 mm were included. The balloon type and size in the predilation group and the stent implantation pressure were selected by the radiologist.
Ultrasound (IVUS) Analysis
Two commercially available catheters (30 MHz, 3.2 French UltraCross and 40 MHz, 2.5 French Atlantis, Boston Scientific Corp., Watertown, Massachusetts) were used. Intravascular ultrasound study was not allowed at baseline. After the administration of intracoronary nitroglycerin, the transducer was advanced >10 mm distal to the lesion. The images were recorded (Super-VHS) with an automatic pullback at 0.5 mm/s along the entire length of the stent until the aorto-ostial segment of the artery being examined. When the angiographic result was satisfactory, an IVUS study was performed by another interventional cardiologist (blinded) different from the person doing the PTCA. Blinding was unmasked only when the interventional cardiologist in charge of the IVUS perceived potential danger for the patient. The expansion parameters analyzed were the minimum stent cross-sectional area (CSA), percentage of residual stenosis per reference area (minimum stent CSA divided by the mean reference lumen CSA), percentage of distal residual stenosis per reference area (minimum stent CSA divided by the distal reference lumen CSA), stent symmetry index (minimum stent diameter divided by maximum stent diameter) and good apposition (defined as sufficiently close contact to preclude blood flow between the stent mesh and the artery wall).
Follow-up
Creatine kinase (CK) and creatine kinase MB isoenzyme (CK-MB) levels and the electrocardiogram (ECG) were systematically recorded immediately after the procedure and at 6, 12 and 18 h post-procedure, as well as every 6 h thereafter if the patient had chest pain. In-hospital events, including death, nonfatal myocardial infarction and new revascularization (PTCA or surgery) of lesions, whether included in the study or not, were recorded during the 12 months post-procedure. Routine follow-up angiography was performed between 7 and 9 months after the procedure. During the follow-up, binary restenosis was defined as >50% stenosis on quantitative analysis.
Angiographic Measurements and Statistical Analysis
Two expert interventional radiologists performed measurements after the procedure using an automatic contour detection system (CAAS II, V4.1.1, Pie Medical Imaging, Maastricht, Netherlands). Clinical and revascularization events were analyzed on an intent-to-treat basis. Comparisons between the stent expansion IVUS parameters were also based on an intent-to-treat principle, according to the actual treatment received. Continuous variables are shown as mean ± standard deviation. Differences between groups were calculated using Student's t test for continuous variables and the chi-square test for proportions. A P-value of .05 was considered statistically significant. SPSS for Windows, version 11.0, was used for the statistical analysis.
RESULTS
From 1 May 2000 to 15 July 2001, a total of 247 patients with 299 new lesions were treated by stent placement in the afternoon schedule at our hospital. Of the 299 lesions, 100 (33.4%) in 82 patients consecutively met the inclusion criteria (43 patients in the predilation group and 39 in the direct stenting group). The difference in the use of direct stenting between the morning and afternoon scheduled procedures was not statistically significant (33.4% vs 30.2%; P=.091). We retrospectively analyzed all patients treated in the afternoon schedule between 1 May 2000 and 15 July 2001 to investigate inclusion bias. Two patients were not available for follow-up and were not included in the study. All the other patients who met the inclusion criteria were included consecutively.
Clinical and Angiographic Characteristics at Baseline
There were no significant differences in the clinical characteristics of the 2 groups including age, sex, hypertension, diabetes, hypercholesterolemia, active smoker, prior infarction, prior revascularization, ejection fraction, number of diseased vessels, and indication for the procedure (Table 1). Among the patients studied, 25 additional lesions that did not meet the inclusion criteria or that presented exclusion criteria were treated in the predilation group and 14 in the direct stenting group. These lesions were not included in the angiographic or IVUS study. Angiographic characteristics at baseline were similar in both groups. No differences were observed in calcification, lesion length, reference diameter or percent stenosis (Table 2).
Procedure Characteristics and Immediate Results
All procedures in both treatment groups were performed satisfactorily (Table 3). In all cases of predilation, only a conventional balloon was used. There were no cases of in-hospital mortality, acute thrombosis of the stent or emergency surgery. One (2%) and two (5%) non-Q-wave AMI episodes were observed in the predilation and direct stenting groups, respectively (P=NS). Direct stenting was unsuccessful in four patients (crossover to predilation group). There were no significant differences in the diameter, implantation pressure or use of additional stents between the two groups (Table 2). The stents used in the predilation group were significantly longer than those used in direct stenting group. Additional post-dilation was used in four (8%) lesions in each group in order to achieve optimal angiographic results in the operator's opinion, with these results similar in both groups. There were no cases of stent loss.
Ultrasound Assessment
Intravascular ultrasound study was possible in 99 of the 100 lesions. In the remaining case in the predilation group, the catheter did not advance through a circumflex artery with a large kink, hindering satisfactory study of an obtuse marginal artery stent. Ultrasound results are shown in Table 4. There were no differences in the reference or expansion parameters between the two groups. Only 17% of the stents achieved a stent CSA >9 mm (45.7% in stents ≥3.5 mm). Poor apposition was observed in one lesion after placement at high pressure; in this case, a larger diameter balloon had been used despite the excellent angiographic results (Figure 1).
Fig. 1. A: Poor apposition of stent mesh in the vessel wall in lesion number 22 treated by direct stenting. The white arrows indicate the area between the vessel wall and the mesh. B: Final ultrasound result after post-dilation with a larger diameter balloon.
Clinical and Angiographic Follow-up
Clinical follow-up was performed in 100% of the patients, with no differences found between the 2 groups after 12 months (Table 3). Scheduled or symptom-guided angiographic follow-up was possible in 43 lesions (86%) per group. There were no significant differences in binary restenosis (23% vs 20%) or late loss index (0.92 [0.81] vs 0.88 [0.60]) between the 2 groups (Table 4 and Figure 2).
Fig. 2. Cumulative distribution curves of the percentage of stenosis found on quantitative angiographic analysis at baseline after stent implantation and during follow-up, for direct stenting and implantation with predilation.
DISCUSSION
The present study, designed to replicate normal direct stenting activity, showed no differences in IVUS expansion parameters as compared to those obtained with the predilation technique. Furthermore there were no differences in the clinical or angiographic follow-up, and similar results were observed in the incidence of restenosis at 6 months.
The current frequent use of direct stenting is supported by numerous studies demonstrating the safety of this technique,8,9,11,13-23 with clinical and angiographic results that are similar11,13,20,24,25 or even better26,27 than those obtained with traditional implantation. Loss or incomplete expansion could be two potential risks of direct stenting. Because of the progressive advances in stent design and fixation by the balloon, the incidence of stent dislodgement or embolization is low or non-existent (<1%)10,20,21,28 (0% in our study).
Correct deployment of the stent is related not only to the stent design and implantation pressure; the characteristics of the lesion also represent an important conditioning factor. In cases of severe calcification unidentified before direct stenting, correct deployment may be impossible despite elevated implantation pressures. Suboptimal expansion has been associated with a risk of acute occlusion and restenosis.29,30 In addition, angiography is limited to ensure correct expansion of the stent.12,31,33 A high percentage of stents with excellent angiographic results present poor expansion as visualized by IVUS. Up to now, there is little available information on the degree of expansion achieved with direct stenting. In a descriptive study conducted in the early years of this technique, correct expansion was found in 66% of the patients.34 Two substudies,19,20 one with 3D-IVUS in a small number of patients35 and a recent randomized study with IVUS,36 showed no differences in expansion between the two implantation techniques. Our work shows two main differences with respect to previous studies: it was specifically designed to analyze direct stent expansion with IVUS and it was intended to replicate the usual practice in stent placement. In order to ensure the latter point, baseline IVUS was not allowed and the IVUS data were not used to optimize the results.
The balloon-to-artery ratio in patients with predilation was slightly below 1 (0.9). This undersizing in the predilation, along with the high pressures used in both subgroups when placing the stent, may have contributed to the absence of differences between the 2 techniques in terms of expansion, decreasing or minimizing the predilation balloon value. The use of undersized balloons for predilation is currently common practice, in an era in which the stent is elective in virtually all lesions. We believe the results observed in our study in this regard are representative of those seen in daily practice at any hospital in which this type of lesion is treated.
Although high inflation pressures are used, the percentage of lesions with a minimum stent CSA >9 mm in our study was lower than that found in previous studies with an optimized technique according to sonographic criteria.37,38 The observational, blind use of IVUS and implantation of up to 24% of stents <3.0 mm might have been the main causes. In many cases, calcification not visualized on IVUS could have caused suboptimal expansion. Consistent with lower expansion, the incidence of restenosis at 8 months was higher than in previous studies with IVUS optimization.30
There was only one case (2%) of poor stent apposition in the direct stenting group. This potential cause of acute thrombosis led to unmasking of IVUS blinding and the use of IVUS for stent optimization with a larger diameter balloon. No conclusions can be drawn from this single case. However, it suggests that IVUS follow-up of lesions treated by direct stenting might be beneficial in complex anatomies.
The results of the present study cannot be extrapolated to other types of lesions. Forty percent of direct stenting procedures require considerable effort to select the lesions, although this could reflect a figure similar to the number of lesions that can be treated by direct stenting.39 Based on the results of our study, it is impossible to predict correct expansion in lesions with severe calcification, in bifurcations and small vessels, in ostial lesions, long lesions, restenosis and other complex lesions. In one recently published study40 with 128 patients treated by direct stenting in long lesions (>18 mm), the authors found 2.3% of acute occlusion and 9.4% of non-Q-wave infarction despite optimal angiographic results in 99% of the cases. Optimal expansion on angiography was not confirmed by IVUS in that series. Inadvertent stent underexpansion could have been the cause of these complications. Until randomized IVUS studies in more complex lesions are conducted, it is impossible to ensure that no expansion differences between the 2 techniques exist in these types of lesions.
Limitations
The present study was conducted in a single hospital and the only inclusion criterion was the decision that direct stenting was possible. Therefore, although the percentage of lesions included was similar to the percentage seen in daily practice at our hospital during the study period and similar to that of other published series,7,25 it is evident that the interventional radiologists performing the procedures selected the lesions considered optimal for direct stenting. As mentioned, the results cannot be compared to series with a greater use of direct stenting, to stenting in long lesions or small vessels, or to lesions treated in hospitals with less experience.
Although the sample size was not calculated and the number was chosen for logistic reasons, the number of lesions included was higher than in previous studies on stent expansion in direct implantation.19,20,35,36 Moreover, it is unlikely that a larger number of lesions would have led to clinically relevant differences in the expansion parameters, since no trend toward differences was observed with the number of patients in our study. Nevertheless, the sample size was clearly insufficient to study differences in the incidence of restenosis and the clinical events, as is evident when our study is compared with others specifically oriented toward these objectives.11,19,20
CONCLUSIONS
Direct stenting in the lesions normally selected for this technique was associated with expansion parameters similar to those obtained with conventional placement after predilation. Both implantation techniques showed similar rates of angiographic restenosis and clinical events at 1 year of follow-up.
The generalized use of direct stenting in more complex lesions should be based on future studies showing that the expansion parameters and the long-term clinical and angiographic results are similar to those obtained with predilation.
Full English text available at: www.revespcardiol.org
ABBREVIATIONS
CSA: cross-sectional area.
ECG: electrocardiogram.
IVUS: intravascular ultrasound.
PTCA: percutaneous transluminal coronary angioplasty.
Work partially presented as an oral communication at the 37th National Congress of the Sociedad Española de Cardiología, Barcelona, Spain, 2001.
Correspondence: Dr. R. López-Palop.
Ricardo Gil, 20, 3.o dcha. 30002 Murcia. España.
E-mail: mlopezs@meditex.es