Keywords
INTRODUCTION
The benefits of stents in small vessels is still a controversial matter. Initial observational studies have suggested an increased rate of acute thrombosis,1,2 whereas more recent ones showed a decrease.3,4 Three random studies (ISAR-SMART, BESMART, and SISA) presented disparate results in the rate of restenosis.5-7 Thus, not all of the conclusions derived from studies on vessels >3 mm can be extrapolated to smaller vessels.
Direct stenting (DS) has brought about changes in percutaneous coronary interventions (PCI) leading to savings in costs, imaging, radiation exposure, and procedural time.8,9 Earlier research studied the predictors of the primary success rates of DS.10,11 Most of these studies only included stents ≥3 mm. We evaluated the influence of stent diameter on the primary success rates of DS.
PATIENTS AND METHODS
Based on our laboratory's interventions register, we identified those patients treated by DS, as defined below, from among the PCI carried out between September 1998 and April 2001. Lesions were divided into 2 groups: group 1, nominal stent diameter 2.5 mm and group 2, ≥2.75 mm. Two cardiac surgeons independently examined the angiograms to describe angulation and tortuosity without knowing the result of the intervention, whereas the other variables were determined prospectively. The percentage of stenosis was determined visually and the methodology followed was as previously described.11
Definitions
Primary success is defined as a stent implant without predilatation with a balloon or arterectomy, and secondary success as an implant with previous dilatation with a balloon following DS failure. The calcification density image visible on fluoroscopy was classified as moderate or severe. Tortuosity and angulation were classified as moderate or severe by subjective evaluation. Location was divided into two groups: proximal (segments proximal and medial to the principal arteries) and distal (distal segments and lateral branches).
Statistical Analysis
Continuous variables were expressed as mean±SD and categories as absolute value and percentage. Means and proportions were compared by Student t test and χ2, respectively. Multivariable analysis was done to determine the predictors of the primary success rates of DS. A value of P=.05 was considered significant. Data were analyzed with SPSS 10.0 for Windows.
RESULTS
During the study period, 987 lesions in 773 patients were treated by DS. Lesions were divided into 2 groups: group 1, 237 lesions (21% of DS out of a total treated with a 2.5-mm stent in the same period) and group 2, the remaining 750 lesions (33% of DS out of a total treated with a stent ≥2.75 mm). Patient characteristics are shown in Table 1 and lesions in Table 2. Small vessel lesions were associated with age, female sex, diabetes, distal location, proximal tortuosity and location in the circumflex artery. The type of stent is shown in Table 3.
Results of the Procedure
The primary success rate of DS was lower in the smaller vessel group and postdilatation was more frequent in group 2, with no differences in vessel dissection rate (Table 4).
Stent diameter was a predictor of the primary success rate of DS, as were location of the most distal segment, calcification, proximal tortuosity, angulation and stenosis. Age and stent length acted as confounding factors (Table 5). The predictive power of the model under the ROC curve was 0.87 (confidence interval [CI] 95%, 0.82-0.92).
DISCUSSION
This is the first work to analyze the influence of stent diameter on the primary success rate of DS. The most important finding is that stents of 2.5 mm are associated with a lower primary success rate, even after adjusting this for predictors found in previous studies.10,11
Primary Success
Three factors may have contributed to the lower rate of success in the first group:
1. Despite correcting for variables which acted as predictors for DS failure in previous studies, less favorable, unknown characteristics not included in the analysis may have led to this poor result. Specifically, the lower sensitivity of angiography to detect calcification compared to ultrasound12 could have played a role. This is more marked in small vessels.13
2. The use of smaller diameter stents was associated with more distal locations, which also act as predictors for implant failure.
3. Finally, the stents used in smaller vessels may not have behaved like larger diameter stents. A design specific to small vessels should adapt to the marked tortuosity and angulation often associated with these vessels. On the other hand, the nominal diameter is not proportional to the profile, and that of small vessels has a higher nominal profile/diameter relationship than large vessels. Consequently, with a certain percentage of stenosis, it would be more difficult to cross the lesion in small vessels. Given the period in which PCI was performed, an appreciable percentage of the stents used for small vessels were not specifically designed for them. Thus, we think that current trends aimed at designing platforms for small-caliber vessels, especially those made from the new alloys, will probably help to improve the results.
Results of the Procedure
In addition to the difference in primary success rates, post-dilatation is required more frequently in large vessels. During our study period, the smallest-diameter stent was 2.5 mm. Thus, stents of 2.5 mm might have been implanted in vessels with a reference diameter less than the nominal stent diameter, leading to an acceptable degree of residual stenosis in small vessels. No differences were found in the percentage of vessel dissection rates.
Clinical Profile
Group 1 lesions were associated with patients with the least favorable clinical profiles. Although the aim of this study was to analyze the immediate results of the procedure and not to describe clinical evolution, both groups displayed differences regarding factors associated with poor prognosis, such as age, female sex, and diabetes.14 These differences could have led to poor results.
LIMITATIONS
In this study, the percentage of lesions treated with DS is similar to that in contemporary studies, but less than in current use.15,16 In our center, similar to others, the percentage of DS increased from 11% in 1998 to 40% in 2001.17 The percentage of primary success has decreased since 1998 to 90.1%, probably because of a less stringent selection of lesions. The period analyzed and the types of stent used could limit, to certain extent, the applicability of the conclusions to current techniques.
Some variables, such as tortuosity, calcification and angulation, are qualitative, and thus a subjective component cannot be ruled out from their description.
This study lacks angiographic quantification of the lesions. However, we think that the differentiation into 2 groups according to the size of the stent allows the lesions to be divided into 2 well-defined groups, thus facilitating the main aim of the study.
CONCLUSIONS
Direct implantation of a 2.5-mm stent is associated with a significantly lower percentage of primary success than that obtained with stents with a bigger diameter. In addition to the unfavorable anatomy of lesions in small vessels, the actual diameter of the stent acts as an independent predictor of implant failure.
Correspondence: Dr. I. Lozano.
2132 Carretera Piles-Infanzón. La Redonda. Somió.
33203 Gijón. Asturias. España.
E-mail: IMLML9@hotmail.com