Clinical Outcomes Following IVUS-Guided Stent Deployment in a Community Hospital

ABSTRACT: Objectives. A discrepancy exists in the medical literature as to what effect intravascular ultrasound (IVUS)-guided stent deployment has on target vessel revascularization (TVR) at 6 months. The major endpoints of this study are the need for TVR, defined as clinically driven repeat interventional or surgical therapy of the index vessel at 6 months and major adverse cardiac events. Methods. One hundred interventional stent cases (50 IVUS-guided, 50 non-IVUS guided) were randomly selected in a 6-month period (January to June 2001) for review by measurement of minimal luminal diameter (MLD) pre- and post-intervention. Seventy males and 30 females were distributed among the 2 groups. There were a total of 135 lesions (70 IVUS-guided, 65 non-IVUS guided) in the 2 groups. A 6-month follow-up chart review was performed following the initial stenting. Results. At 6-month follow-up, there were 2 deaths in the IVUS-guided group and 3 deaths in the non-IVUS guided group (p = NS). All deaths were cardiovascular in nature. Post-procedure MLD was 3.58 ± 0.08 mm for the IVUS-guided group and 2.88 ± 0.09 mm for the non-IVUS guided group [t = 5.7 (df, 133); p 1 IVUS studies demonstrated that incomplete deployment of stents occurred in up to 80% of patients at nominal pressures [8–12 atmospheres (atm)]. This helped to usher in the use of high pressure (> 12 atm) techniques and emphasized the need for careful attention to maximizing target segment expansion.² IVUS was first utilized to guide stent deployment and help gain maximal stent-vessel wall apposition and maximal intraluminal diameters. A discrepancy exists in the medical literature as to what effect IVUS-guided stent deployment has on 6-month target vessel revascularization (TVR). The CRUISE study, along with the AVID trial, showed decreased rates of TVR.^2,3 In contrast, the RESIST study and the OPTICUS trial showed essentially no difference in restenosis rates at 6 months.^4,5 This retrospective study was conducted to help determine whether IVUS-guided stent deployment is necessary when placing either native coronary artery or coronary graft stents in a community hospital setting. The primary endpoint was the need for TVR, defined as clinically-driven repeat interventional or surgical therapy of the index vessel, at 6 months. Secondary endpoints were major adverse cardiac events (MACE) and the post-procedure measurement of minimal luminal diameter (MLD) by angiography. Methods Study design. This was designed as a retrospective study, with randomly selected cases in order to determine the 6-month outcomes of the effect of IVUS-guided stent deployment on TVR and MACE. Data were derived from patients admitted to the Charleston Area Medical Center (CAMC) in Charleston, West Virginia, from January 1, 2001 to June 30, 2001. CAMC is the major tertiary care referral center for Southern West Virginia. Patient selection. During the selected study period, there were 1,297 primary coronary intervention (PCI) cases performed at CAMC. Of these, a total of 173 (13%) utilized IVUS guidance. Fourteen cardiologists utilized IVUS technology during this period. A single cardiologist performed 96 of the 173 IVUS-driven cases (54%). No patients were excluded from the study. Patients were selected randomly by comparing the last three digits of their medical record number to a readily available table of random numbers. The patients were divided into two separate groups consisting of angiography-guided and IVUS-guided cases. Imaging analysis. An independent analyst at CAMC blinded to the guidance method analyzed cineangiograms of both the angiography and IVUS-guided groups. Cine frames, from only the angiographic portion of each case, were selected before intervention and after final stent deployment and balloon inflation. Measurements were taken using standard calipers and comparisons were made to standard guiding catheter size. MLD was recorded pre- and post-intervention for both the angiography and IVUS-guided cases. Stents were deployed in all lesions. Retrospectively, procedural outcomes for optimal stent deployment were consistent with American College of Cardiology/American Heart Association (ACC/AHA) standards of achieving a minimum stenosis diameter reduction to less than 20%.6 In addition, IVUS criteria for stent and balloon sizing, by taking tomographic views of the reference and diseased segments proximally and distally and utilizing the average diameter from mid-media to mid-media of the reference vessel, were consistent with ACC/AHA guidelines.1 All patients were discharged on aspirin and clopidogril. Clinical follow-up. A 6-month chart review was conducted after stent implantation for the occurrence of TVR and MACE. TVR was defined as clinically-driven repeat revascularization of the initially treated target vessel. Only the hospital record was available for review. If there were no readmissions, then a death record was sought. If no death record was found, then the case was recorded as negative for TVR or MACE. Death records were requested from the appropriate state office and were also compared with results from a national genealogy tracking website. If readmission had occurred after the closure of the 6-month study period, then it was assumed there had been no TVR or MACE. CAMC is the dominant cardiovascular referral center in southern West Virginia. It is highly expected that if a patient underwent PCI at CAMC, they would return for follow-up intervention if medically necessary (likewise for adverse events). Statistical analysis. Categorical data were compared using Chi Square and Fisher exact tests. Continuous variables were compared using the student t-test. Predictors of MACE were examined by use of multivariate logistic regression. All comparisons were made using SPSS, version 10.1. Results Patient characteristics. During the 6-month study period (January 1, 2001 to June 30, 2001), a total of 100 patients were randomly selected. Fifty were in the angiography-guided group and 50 were in the IVUS-guided group. Baseline demographics and clinical data are shown in Table 1. The clinical profile showed that a significantly higher percentage of people in the angiography group had a history of congestive heart failure (CHF) and a significant number of patients in the IVUS group had previous PCI. The IVUS-guided group consisted of 70 lesions and the angiography-guided group consisted of 65 lesions. Stents were deployed in all lesions. A significant number of left anterior descending (LAD) lesions were seen in the IVUS group. The angiographic group showed statistical significance for circumflex and vein graft lesions. Baseline angiographic data are seen in Table 2. Angiographic and IVUS results. Angiographic and IVUS results are listed in Table 2. For all lesions, the post-procedure MLD for the IVUS group was significantly larger (3.58 ± 0 .08 mm for IVUS versus 2.88 ± 0.09 mm for angiography; p Clinical outcome at 6 months. Clinical outcomes at 6 months are shown in Table 3 and illustrated in Figure 1. There were no differences in the incidence of death and myocardial infarction during follow-up. However, the incidence of TVR was 9.7% higher in the IVUS-guided group (4.6% versus 14.3%; p = 0.057), which approached statistical significance. MACE was not found to be statistically different between the two groups. The potential predictors of TVR and MACE were entered into both univariate and multivariate regression models. The results are shown in Table 4. Table 5 and Figures 2 and 3 show post-MLD values by type of outcome (MACE) versus no event and by type of procedure (angiography versus IVUS). Figure 2 represents all lesions and Figure 3 only accounts for native coronary arteries. Discussion The ideal method to study whether restenosis after PCI can be prevented with the use of IVUS is a prospective, randomized trial. CAMC is one of the top 6 busiest heart centers in the United States. Private physicians, from multiple single specialty group practices, cover the interventional procedures. Therefore, to conduct an ideal study at this center is not practical at this time. Moreover, this study describes the “everyday situation” of interventional cardiology, answering the question of the clinical outcomes following IVUS-guided stent deployment. The difficulty in studying this “everyday situation” in a retrospective fashion is the introduction of selection bias. At our center, IVUS guidance was utilized in 13% of the interventional cases in the 6-month study period. One operator accounted for approximately half of these cases. However, when removing this operator from the statistical analysis, although the overall “n” becomes smaller, the statistical trend remains the same. The propensity for the selection of IVUS among patients with LAD lesions and prior PCI is also likely secondary to selection bias. Review of the multivariate regression analysis shows that despite this inherent selection bias, no patient characteristic, including a LAD lesion or prior PCI, made any statistically significant difference in the overall clinical outcome. Several possible reasons have been presented as to why IVUS would tend to decrease restenosis rates at 6 months and are based primarily upon maximization of lumen size and optimization of stent geometry. Table 5, Figure 2 and Figure 3 show that despite the anticipated effect of obtaining a larger final MLD, there was no statistical difference in the outcomes of MACE at 6 months when comparing the IVUS-guided group to the angiography group. This parallels similar results shown by the RESIST and OPTICUS study groups. In-stent restenosis is caused by neointimal proliferation, which occurs in association with macrophage accumulation and extensive neorevascularization, suggesting a role for the organization of mural thrombus.⁷ Cutlip et al. described in a pooled analysis of 6,186 patients that there were several independent predictors of clinical restenosis: diameter stenosis greater that 70% on follow-up angiography, small pretreatment and smaller final MLD, longer stent length, diabetes, unstable angina and hypertension; smoking and MI were negative predictors.⁸ Wexberg et al. further proved how little is truly understood regarding the dynamics of the coronary endothelium. In a study of 244 patients with stable angina, those who had constrictive remodeling (a smaller arterial lumen than expected from the plaque size alone) prior to PTCA had higher in-hospital complication rates including MI, fatal arrhythmia, death, etc. Those with adaptive remodeling (the presence of a large lumen despite a substantial atherosclerotic plaque size) had a higher rate of MACE at 7.7 month follow-up.⁹ In this single center, retrospective study, although IVUS-guided stent implantation resulted in an overall larger MLD, the IVUS-guided group had a 9.7% higher rate of TVR (p = 0.057). There were no statistical differences between the numbers of death and MI in the two groups. The results of this study suggest that in the everyday situation, the use of angiography alone is sufficient to obtain an acceptable rate of clinically-driven repeat revascularization procedures despite the use of IVUS guidance to help maximize final MLD. The preliminary results with coated stents show that the phenomenon of neointimal hyperplasia is extremely complex. The basis of rapamycin therapy is the potent antiproliferative and antimigratory effect.¹⁰ The long-term outcomes of drug-eluting stents are yet to be seen, but the immediate- and short-term results help prove that the biochemistry of the endothelium will simply not be overcome by a mere change in vessel geometry. Therefore, this study, which proposes that the use of IVUS-guided stent deployment does not help to reduce the need for clinically-driven repeat TVRs at 6 months, helps to expand the literature. The additional cost of IVUS-guidance does not appear to be warranted. Study limitations. The major limitation of this study was the introduction of selection bias. This would be best avoided by conducting a prospective, randomized trial. Given the current practice patterns at our facility, it was not feasible to proceed with this type of study. A second limitation was found in the measurement of the MLD. Although each cineangiogram was reviewed, equipment was not available to perform quantitative coronary angiography in a retrospective fashion, thereby limiting the measurement of the MLD to a measurement compared to the known size of the guiding catheter. A third limitation was that patients were only followed for a clinical outcome. Since a retrospective study was conducted, it was not applicable to perform a repeat catheterization at 6 months. A better method would be to conduct a prospective trial with repeat angiography and documentary IVUS at 6 months. A fourth limitation also exists in determining all negative outcomes. Since the data were derived from chart and mortality review only, it is conceivable that a patient did not return to our center for treatment of a major cardiac event. Since the study period was for only 6 months, it is possible that if the study period were extended that there may be less statistical difference between the two groups regarding need for TVR. The small size of this trial limits the power to detect small differences among the two groups. It also limits the ability to interpret data regarding the outcomes of the saphenous vein graft lesions. The inability to prospectively define the criteria for IVUS and stent/balloon sizing and optimal stent deployment introduces the variable of residual confounding. Despite adjustments using regression analysis, this specific category of confounding is broad and there may be unaccounted confounders. Finally, data to further help define predictors of restenosis, including reference vessel diameter, lesion length and others, were not obtained. With these data, further characterization may have been capable. Conclusion. In this population, IVUS-guided stent deployment does not appear to reduce the need for clinically-driven repeat TVR at 6 months or MACE. The added expense of IVUS does not appear to be warranted.