- Email: [email protected]
Achieving optimal results with standard balloon angioplasty: can baseline and angiographic variables predict stent-like outcomes?
Journal of the American College of Cardiology © 2001 by the American College of Cardiology Published by Elsevier Science Inc.
Vol. 37, No. 7, 2001 ISSN 0735-1097/01/$20.00 PII S0735-1097(01)01244-X
Achieving Optimal Results With Standard Balloon Angioplasty: Can Baseline and Angiographic Variables Predict Stent-Like Outcomes? Warren J. Cantor, MD,* Anne S. Hellkamp, MS,† Eric D. Peterson, MD, MPH,† James P. Zidar, MD,† Patricia A. Cowper, PHD,† Michael H. Sketch JR, MD,† James E. Tcheng, MD,† Robert M. Califf, MD,† E. Magnus Ohman, MD† Toronto, Canada and Durham, North Carolina To predict which patients might not require stent implantation, we identified clinical and angiographic characteristics associated with repeat revascularization after standard balloon angioplasty. BACKGROUND Stents reduce the risk of repeat revascularization but are costly and may lead to in-stent restenosis, which remains difficult to treat. Identification of patients at low risk for repeat revascularization may allow clinicians to reserve stents for patients most likely to benefit. METHODS Data from five interventional trials (5,146 patients) were pooled for analysis. We identified patients with optimal angiographic results (final diameter stenosis ⱕ30% and no dissection) after balloon angioplasty and determined the multivariable predictors of repeat revascularization. RESULTS Optimal angiographic results were achieved in 18% of patients after angioplasty. The repeat revascularization rate at six months was lower for patients with optimal results (20% vs. 26%, p ⬍ 0.001) but still higher than observed in stent trials. Independent predictors of repeat revascularization were female gender (odds ratio [OR] 1.67, p ⫽ 0.01), lesion length ⱖ10 mm (OR 1.62, p ⫽ 0.03) and proximal left anterior descending coronary artery lesions (OR 1.62, p ⫽ 0.03). For the 8% of patients with optimal angiographic results and none of these risk factors, the repeat revascularization and target vessel revascularization rates were 14% and 8% respectively, similar to rates after stent implantation. Cost analysis estimated that $78 million per year might be saved in the U.S. with a provisional stenting strategy using these criteria compared with elective stenting. CONCLUSIONS A combination of baseline characteristics and angiographic results can be used to identify a small group of patients at very low risk for repeat revascularization after balloon angioplasty. Provisional stenting for these low risk patients could substantially reduce costs without compromising clinical outcomes. (J Am Coll Cardiol 2001;37:1883–90) © 2001 by the American College of Cardiology OBJECTIVES
Coronary stents have been shown to reduce rates of clinical and angiographic restenosis in selected patients (1,2). Stent implantation is costly, however, adding more than $2,000 in direct costs to each interventional case (3). Given these cost considerations, provisional stenting (stenting in selective cases as opposed to all cases) has been proposed as an alternative (4). The selection criteria for provisional stenting, however, have not been fully elucidated. We hypothesized that the same low rate of repeat revascularization might be achieved without stent implantation when optimal angiographic results are obtained from balloon angioplasty alone. Specifically, using a large registry from five clinical trials, we investigated the degree to which angiographic results, patient clinical characteristics or both From *St. Michael’s Hospital, Toronto, Canada; and †Duke Clinical Research Institute, Durham, North Carolina. Sponsored in part through a research fellowship grant jointly funded by the Royal College of Physicians and Surgeons of Canada, the Medical Research Council of Canada, and Novartis Pharmaceuticals Corp., East Hanover, New Jersey. Presented in part at the 48th Annual Scientific Session of the American College of Cardiology, March 9, 1999, New Orleans, Louisiana, and the 72nd Scientific Sessions of the American Heart Association, November 10, 1999, Atlanta, Georgia. Manuscript received August, 22, 2000; revised manuscript received January 30, 2001, accepted February 13, 2001.
could be used to identify patients with low repeat revascularization rates. After defining these angiographic and clinical criteria, we determined the potential economic effects of applying these criteria to limit elective stent usage.
METHODS Patient population. Data were pooled from five randomized trials of percutaneous coronary interventions (5–9). The protocols and study populations of the five trials and the methods used to pool the data have been described in detail (10). The Coronary Angioplasty Versus Excisional Atherectomy Trial (CAVEAT) randomized 1,012 patients to undergo angioplasty or directional coronary atherectomy. Patients who underwent atherectomy were excluded from the analysis. In both the Evaluation of 7E3 for the Prevention of Ischemic Complications (EPIC) (2,099 patients) and Integrilin to Minimize Platelet Aggregation and Coronary Thrombosis (IMPACT-II) (4,010 patients) studies, patients undergoing percutaneous interventions were randomly allocated to receive placebo or an intravenous glycoprotein IIb/IIIa inhibitor—abciximab or eptifibatide, respectively. The Multicenter American Research trial with
1884
Cantor et al. Stent-Like Outcomes With Balloon Angioplasty
Abbreviations and Acronyms BENESTENT ⫽ BElgium-NEtherlands Stent CAVEAT ⫽ Coronary Angioplasty Versus Excisional Atherectomy Trial CK-MB ⫽ creatine kinase-MB EPIC ⫽ Evaluation of 7E3 for the Prevention of Ischemic Complications EPISTENT ⫽ Evaluation of Platelet IIb/IIIa Inhibitor for Stenting IMPACT II ⫽ Integrelin to Minimize Platelet Aggregation and Coronary Thrombosis II LAD ⫽ left anterior descending coronary artery MARCATOR ⫽ Multicenter American Research trial with Cilazapril after Angioplasty To prevent coronary Obstruction and Restenosis MI ⫽ myocardial infarction OPUS ⫽ Optimum Percutaneous transluminal coronary angioplasty compared with roUtine Stent PBC ⫽ Perfusion Balloon Catheter study QCA ⫽ quantitative coronary angiography SLR ⫽ stent-like result TVR ⫽ target vessel revascularization
Cilazapril after Angioplasty To prevent coronary Obstruction and Restenosis (MARCATOR) assessed the effect of cilazapril on restenosis in 1,436 patients undergoing angioplasty. The Perfusion Balloon Catheter (PBC) study compared prolonged versus short inflations using the perfusionballoon catheter in 478 patients. The data in these trials were collected during various periods, ranging from April 1989 to November 1995. Patients with acute myocardial infarction (MI) were excluded before pooling the data. The pooled data contain detailed clinical and angiographic data for 8,726 patients. After excluding patients who had stents implanted (n ⫽ 159), vein-graft intervention (n ⫽ 322), left-main intervention (n ⫽ 130), atherectomy or laser treatment (n ⫽ 1,469), failed or partially successful angioplasty (final diameter stenosis ⱖ50% in any attempted lesion, n ⫽ 1,047) and incomplete angiographic data or six-month status (n ⫽ 453), there were 5,146 patients for this analysis. Interventional procedures. Before the procedure, all patients received aspirin (160 to 325 mg/day). Patients in the PBC trial also received dipyridamole. Angioplasty procedures were performed using standard techniques, with the exception of perfusion-balloon catheters used in the PBC trial. Stenting was unusual in these trials (and excluded from this analysis), being performed in ⬍2% of cases. Intravenous glycoprotein IIb/IIIa inhibitors were used only in the active treatment arms of EPIC and IMPACT-II, accounting for 2,099 (41%) of the patients in this analysis. During the procedure, all patients received intravenous heparin to reach a target activated clotting time of ⬎300 s (⬎350 s for CAVEAT). After intervention, creatine kinase (CK)-MB
JACC Vol. 37, No. 7, 2001 June 1, 2001:1883–90
levels were obtained routinely over the next 24 h. All patients were discharged on aspirin 325 mg/day. Data collection. For each trial, the participating sites collected data prospectively on case report forms and sent them to data coordinating centers. Verification of the data was performed using consistency checks and double data entry. Sites were queried for missing or questionable data. The dataset for each trial was verified by retrospective review of patient charts. Complete copies of the individual datasets from the five trials were combined into a single database. Clinical follow-up. All five trials followed patients for at least six months for the occurrence of death, MI, repeat angioplasty or bypass surgery. For repeat angioplasty, involvement of the target vessel was not recorded in MARCATOR but was specified in the other trials. Angiographic analysis. For all five trials, cineangiograms were sent for review to independent core angiographic laboratories blind to treatment assignment. For EPIC and IMPACT-II, diameter stenosis was measured using calipers without catheter calibration. Except for 860 patients in the IMPACT-II angiographic substudy, absolute lumen dimensions were not obtained for patients in these two trials. For the other three trials and for the IMPACT-II angiographic substudy, quantitative coronary angiography (QCA) was performed using automated edge-detection algorithms as described previously (11,12). Definitions. Myocardial infarction was defined as new Q-waves on the electrocardiogram or elevation of CK or CK-MB. Enzymatic criteria for MI were values ⱖ2 times the upper limit of normal for MARCATOR and the PBC trial and three times the upper limit for the other trials. All events were adjudicated by independent committees blind to randomization. Repeat revascularization included all repeat percutaneous coronary interventions and bypass operations within the first six months, whether or not the target lesion was involved. To estimate the rate of target vessel revascularization (TVR), repeat percutaneous interventions without target vessel data were assumed to have the same proportion of target vessel involvement as the population for whom these data were available. All surgical procedures were assumed to have involved the target vessel. Multilesion interventions were considered “stent-like” results only if the stent-like criteria were achieved in all lesions attempted. Statistical analysis. Baseline, procedural and angiographic characteristics are summarized as medians with 25th and 75th percentiles for continuous measures and as percentages for discrete measures. Event rates are summarized as percentages with 95% CI. For comparisons between groups, the likelihood ratio chi-square test was used for discrete variables and the Wilcoxon rank-sum test was used for continuous variables. Sensitivity and specificity for predicting six-month survival free of repeat revascularization were calculated for stent-like result (SLR) criteria based on absence of dissection and residual diameter stenosis levels of ⱕ10%, ⱕ20%, ⱕ25%, ⱕ30%, ⱕ40% and ⱕ50%. These
JACC Vol. 37, No. 7, 2001 June 1, 2001:1883–90
criteria excluded all dissections, on the basis of a survey that indicates that most interventional cardiologists use stents routinely for any angiographically visible dissection (Cantor et al., unpublished data). We defined the best threshold for residual diameter stenosis to be the level that included the most patients yet maintained ⱖ80% specificity to predict survival free of repeat revascularization at six months. Multivariable logistic regression was performed using a stepwise, backward-elimination algorithm to determine independent predictors of repeat revascularization at six months in cases with SLRs. Variables were included in the multivariable model if they were significant in univariable analysis or if considered clinically important. The independent predictors were used to construct a classification and regression tree and classify patients with stent-like angiographic results as high or low risk for repeat revascularization. We validated the CART model by dividing the data into training and test sets to do the development and testing separately. We developed the classification scheme on half of the patients with SLR and tested it on the other half. Once agreement between the two samples was established, we then applied the scheme to the full sample using SLR as the first split in the tree. Cost-effectiveness analysis. The six-month costs and clinical outcomes of elective and provisional stenting strategies were compared using decision-analysis techniques (DATA 3.5, Treeage Software, Williamstown, Massachusetts). The rates of stenting for the provisional stent strategies and the probability of repeat revascularization after balloon angioplasty were based on the results of the analysis described in the preceding text. The cost estimates for angioplasty, stent procedures and bypass surgery were based on a cost analysis of patients undergoing percutaneous coronary intervention at Duke University Medical Center (3). This previous analysis was based on the Transition 1 Accounting System (Transition Systems, Inc., Boston, Massachusetts), which
Cantor et al. Stent-Like Outcomes With Balloon Angioplasty
1885
Table 1. Freedom From Repeat Revascularization at Six Months Final Diameter Stenosis*
n (%)
Sensitivity
Specificity
ⱕ10% ⱕ20% ⱕ25% ⱕ30% ⱕ35% ⱕ40% ⱕ50%
64 (1.2%) 319 (6.2%) 575 (11.2%) 938 (18.2%) 1,351 (26.3%) 1,793 (34.8%) 2,329 (45.3%)
1.4% 6.7% 11.9% 19.3% 27.3% 35.4% 61.2%
99.2% 95.5% 90.9% 85.1% 76.9% 67.0% 55.7%
*With no angiographically visible dissection.
contains detailed cost records at the intermediate product level within each department (for example, number of specific diagnostic tests, hours of care by in-patient unit or time spent in catheterization laboratory). The acquisition cost of a stent was $1,595. For this analysis, indirect costs associated with stent and angioplasty procedures were assumed to be equal. The repeat revascularization rate for stented patients was estimated from randomized trial and registry results (2,13,14). Sensitivity analysis was performed to examine the effects of varying estimated costs and event rates over a wide range of values.
RESULTS Determination of angiographic diameter stenosis criteria for SLR. A total of 5,146 patients were included in the analysis. The numbers of patients with final diameter stenoses at or below various thresholds are summarized in Table 1. Only 1% of cases achieved a residual diameter stenosis ⱕ10% with no dissection as assessed by the core lab. The event rates for the composite end point of death, MI or any repeat revascularization for the different criteria of SLR are shown in Figure 1. Although there was no apparent relation between the final diameter stenosis and
Figure 1. Clinical event rates at six months for four different thresholds of postintervention diameter stenosis. DS ⫽ diameter stenosis; MI ⫽ myocardial infarction; QCA ⫽ quantitative coronary angiography.
1886
Cantor et al. Stent-Like Outcomes With Balloon Angioplasty
JACC Vol. 37, No. 7, 2001 June 1, 2001:1883–90
Table 2. Six-Month Clinical Outcomes With and Without Stent-Like Results
Death Myocardial infarction Repeat angioplasty Bypass surgery Any repeat revascularization Composite of above
Stent-Like Result (n ⴝ 938)
No Stent-Like Result (n ⴝ 4,208)
p Value
0.4% (0.0–0.9) 3.6% (2.4–4.8) 17.1% (14.7–19.5) 4.4% (3.1–5.8) 20.0% (17.5–22.6) 22.0% (19.3–24.6)
0.6% (0.3–0.8) 5.6% (4.9–6.3) 21.0% (19.7–22.2) 6.6% (5.8–7.3) 25.5% (24.1–26.8) 27.6% (26.3–29.0)
0.6 0.009 0.007 0.11 ⬍ 0.001 ⬍ 0.001
Data are presented as percentage (95% CI).
the rates of death and MI, the repeat revascularization rates increased with increasing diameter stenosis thresholds. The sensitivity and specificity for the different thresholds of postangioplasty residual diameter stenosis (core laboratory determination) in predicting freedom from repeat revascularization are shown in Table 1. To maintain ⱖ80% specificity, a diameter stenosis threshold ⱕ30% would need to be used as criteria for a SLR. The clinical event rates for the final diameter stenosis thresholds of 20%, 25% and 30% were similar. Therefore, we defined a SLR as a final diameter stenosis ⱕ30% without dissection. Only 938 eligible patients (18%) met these criteria. Clinical outcomes with SLRs. The six-month rates of death, MI and repeat revascularization for the two groups are summarized in Table 2. Patients with SLR had significantly lower rates of MI (4% vs. 6%; p ⫽ 0.009) and repeat revascularization (20% vs. 26%; p ⬍ 0.001). In the SLR group, the target vessel was involved in 64 (65%) of the repeat angioplasty cases in which vessel location was known. The estimated TVR rate for the SLR group was 14.5%. Predictors of repeat revascularization. Univariate and multivariable regression analyses were performed for the SLR cases to identify predictors of repeat revascularization at six months (Table 3). The univariate predictors of repeat revascularization were female gender, proximal left anterior
descending coronary artery (LAD) lesion, lesion length ⱖ10 mm and no prior MI. The same four variables were independent predictors of repeat revascularization in the multivariate model. Subgroups of the SLRs. Figure 2 shows the classification and regression tree analysis for patients who achieved stent-like angiographic results, stratified by presence or absence of three of the independent predictors of repeat revascularization. In the lowest risk subgroup (men with lesions ⬍10 mm long and no proximal LAD lesions) the repeat revascularization rate at six months was 14% (95% CI, 10% to 18%). The estimated TVR rate in this group was 8.1%. In contrast, the repeat revascularization rate of the highest risk subgroup (women with lesions ⱖ10 mm long and proximal LAD lesions) was 46% (95% CI, 20% to 68%). About 45% of patients with an SLR (excluding cases for which lesion length was missing) were in the lowest risk subgroup, representing about 8% of all eligible patients. Economic implications of the two provisional stenting strategies. We evaluated the expected repeat revascularization rates and costs of two provisional stenting strategies and an elective stenting strategy for stent use based on event rates described in the preceding text. In the elective stenting strategy, all patients would undergo stent implantation irrespective of the angiographic result after balloon angio-
Table 3. Predictors of Repeat Revascularization at Six Months Univariable
Multivariable
Factor
Odds Ratio (95% CI)
p Value
Age (per 10-year increase) Female gender Diabetes Congestive heart failure Unstable angina Prior infarction Multivessel disease Proximal left anterior descending lesion Ostial lesion Total occlusion Calcification Length ⱖ10 mm Reference vessel diameter* Preprocedural minimum luminal diameter* Preprocedural percent diameter stenosis* Postprocedural minimum luminal diameter* Postprocedural percent diameter stenosis*
1.12 (0.97–1.30) 1.88 (1.34–2.64) 1.14 (0.77–1.69) 0.63 (0.26–1.53) 1.16 (0.80–1.66) 0.71 (0.52–0.99) 1.09 (0.78–1.53) 1.46 (1.01–2.12) 0.99 (0.58–1.70) 0.79 (0.35–1.82) 1.26 (0.86–1.86) 1.61 (1.05–2.47) 1.10 (0.76–1.61) 1.02 (0.54–1.93) 1.00 (0.99–1.01) 1.03 (0.61–1.73) 1.02 (0.99–1.04)
0.13 ⬍ 0.001 0.50 0.31 0.44 0.04 0.62 0.04 0.97 0.59 0.24 0.03 0.61 0.95 0.95 0.91 0.14
*Per one-unit increase in the covariate value. †Variables “forced” into multivariate model.
Odds Ratio (95% CI)
p Value
1.67 (1.12–2.48)
0.01
0.68 (0.46–1.01)
0.05
1.62 (1.06–2.47) 0.87 (0.50–1.52)†
0.03 0.63
1.11 (0.71–1.73)† 1.62 (1.04–2.51)
0.66 0.03
1.02 (0.99–1.04)†
0.27
Cantor et al. Stent-Like Outcomes With Balloon Angioplasty
JACC Vol. 37, No. 7, 2001 June 1, 2001:1883–90
Figure 2. Stratification of risk for repeat revascularization at six months for patients with optimal (stent-like) angiographic results. Using three baseline characteristics (gender, lesion length and lesion location), patients are classified as low risk (repeat revascularization rate of 14%) or high risk (repeat revascularization rate of 25%). DS ⫽ diameter stenosis; LAD ⫽ left anterior descending coronary artery.
plasty. Repeat revascularization rates for this strategy were assumed to be 14% (2,13,14). In the second strategy (provisional stenting guided by baseline characteristics and angiographic results), patients would not undergo stenting if they had a stent-like angiographic result (residual diameter stenosis ⱕ30%, no dissection) and none of the three predictors of repeat revascularization (female gender, lesion length ⱖ10 mm or proximal LAD lesion). With this strategy, 8% of patients would be treated with angioplasty alone and the remainder of patients would receive stents. The repeat revascularization rate of 14% in this low risk subgroup is equal to the repeat revascularization rate after stent implantation (2,13,14). In the third strategy (provisional stenting guided by angiography only), patients who achieve a stent-like angiographic result with angioplasty
(18% of patients) were not stented. The repeat revascularization rate for patients treated with angioplasty alone in this strategy is 20%. On the basis of the modalities of repeat revascularization observed in our database, we assumed that percutaneous revascularization would be used in 75% of repeat interventions, bypass surgery in 17% and both in the remaining 8%. Patients undergoing repeat percutaneous intervention were assumed to receive a stent if they were treated initially with angioplasty only, and vice versa. Table 4 summarizes the clinical outcomes and costs of the three strategies. In the elective stenting strategy and the provisional stenting strategy guided by both baseline characteristics and angiographic results, the expected rate of repeat revascularization is 14%. The expected rate of repeat revascularization in the provisional stenting strategy guided by angiographic criteria alone is 1% higher (15%). Both provisional stenting strategies result in cost savings, with the lowest cost in the angiographic-guided provisional stenting strategy. In this least-cost strategy, the savings from using fewer stents are minimally offset by the costs associated with the slightly higher repeat revascularization rate. Although expected costs per patient do not vary greatly among these strategies, when considered from a health system perspective, substantial savings could be achieved by following the provisional stenting strategies. To assess the sensitivity of results to model assumptions, we varied key parameters in the model. The cost savings associated with provisional stenting guided by both baseline characteristics and angiography declined with the cost of stents. If stent costs fell to 50% of the baseline value, aggregate savings would fall to $37 million. However, the provisional strategy did not become more expensive than elective stenting until stent costs fell to ⬍$200. The cost savings were also sensitive to the rate of repeat revascularization after stenting. For example, if the repeat revascularization rate were 10% rather than 14%, the aggregate savings would fall to $53 million. The cost savings with a repeat revascularization rate of 5% after stenting are more modest at $23 million. The provisional stenting strategy guided by angiography alone became more expensive than the more selective provisional stenting strategy with very small changes in stent costs and repeat revascularization rates. This sensitivity is
Table 4. Six-Month Clinical Outcomes and Costs for Stenting Strategies
Strategy Elective stenting for all patients Provisional stenting strategies Angiographic‡ and clinical§ criteria Angiographic‡ criteria only
1887
Cases Stented
Expected Repeat Revascularization Rate (%)*
Overall Six-Month Cost
Overall Savings in U.S./Year†
100%
14%
$16,396
—
92% 82%
14% 15%
$16,240 $16,231
$78 million $83 million
*Repeat revascularization rate after stenting assumed to be 14% for all strategies. †Assumes 500,000 stent cases performed per year in the U.S. ‡Stent if final diameter stenosis ⬎30% or dissection present. §Stent if female, lesion length ⱖ10 mm or lesion in proximal left anterior descending artery.
1888
Cantor et al. Stent-Like Outcomes With Balloon Angioplasty
due to the relatively high rate of repeat revascularization among the 18% of patients not initially stented. This strategy is therefore both clinically and economically unattractive.
DISCUSSION Major findings. We have shown that the combination of angiographic residual diameter stenosis and baseline patient and lesion characteristics can be used to identify a subset of patients at low risk for repeat revascularization after standard angioplasty. Combining angiographic results with baseline characteristics improves the sensitivity for predicting repeat revascularization compared with reliance upon angiographic results alone. The rate of repeat revascularization in the low risk cohort is similar to that observed with stent implantation (2,13,14). Although only a small proportion of patients undergoing angioplasty fall into this low risk category, the cost savings achieved by avoiding unnecessary stent use in this subgroup are substantial. Provisional stenting. Randomized trials have shown lower restenosis and repeat revascularization rates with stenting than with balloon angioplasty (1,2). On the basis of these encouraging results, many institutions advocate and practice universal stenting (15). When restenosis does occur within stents, however, it is often difficult to manage and refractory to conventional percutaneous interventions (16 –18). Previous attempts to identify patients with favorable outcomes after standard angioplasty have focused on the postprocedural residual diameter stenosis. The BElgiumNEtherlands Stent (BENESTENT) trial investigators showed that patients with stent-like angiographic results, defined as a residual stenosis ⱕ30% with no major dissection, had clinical and angiographic outcomes similar to those of patients who underwent stenting (19). This finding has been validated in other stent trials (20,21) and angioplasty registries (22,23). The Optimum Percutaneous transluminal coronary angioplasty compared with roUtine Stent strategy (OPUS) study was the first randomized trial of provisional stenting based on angiographic results (24). A total of 479 patients were randomized to a strategy of elective stent implantation or initial balloon angioplasty, with provisional stenting for a residual stenosis ⱖ20% by visual assessment (or ⱖ30% by QCA), significant dissection or threatened closure. At six months, patients randomized to elective stenting had lower rates of TVR. The use of visual estimation rather than QCA may account in part for the discrepancy with previous studies. However, these results are consistent with our observation that angiographic criteria alone may be inadequate to identify patients at low risk for restenosis and repeat intervention. Two small trials have shown the feasibility of a provisional stenting strategy guided by repeating angiography 20 to 30 min after balloon angioplasty to assess for early recoil (25,26). The use of intravascular ultrasound and physiologic
JACC Vol. 37, No. 7, 2001 June 1, 2001:1883–90
measurements, including coronary and fractional flow reserve, may help discriminate which patients will have favorable clinical outcomes with balloon angioplasty alone (27–31). However, these modalities prolong procedure times and carry additional costs. In contrast, the use of angiographic results and baseline characteristics to guide stent use is simple and inexpensive. Our analysis focused primarily on the incidence of repeat revascularization. A meta-analysis of trials comparing stenting and balloon angioplasty demonstrated that the primary clinical benefit of stent implantation is a reduction in the need for repeat revascularization, with no difference in the incidence of death or MI (32). The other potential advantage of stent implantation is a reduction in the incidence of abrupt occlusion. In this study, the rates of death, MI and repeat revascularization within 48 h after intervention for patients with optimal angiographic results were low (0%, 2% and 3% respectively) and similar to the rates reported after stent implantation in a recent large randomized trial (33). Nevertheless, we cannot exclude the possibility that early ischemic complications may occur more often with provisional stenting. Glycoprotein IIb/IIIa inhibitors were used in 41% of the cases in this study. In the Evaluation of Platelet IIb/IIIa Inhibitor for Stenting (EPISTENT) study, the combination of stenting and glycoprotein IIb/IIIa inhibitors was associated with significantly lower rates of death, MI and TVR at one year than either therapy alone (34). However, stents were used in the angioplasty group only for reduced coronary flow, long dissections or a residual diameter stenosis ⬎70% and were implanted in only 19% of patients. The outcomes of the EPISTENT angioplasty group were likely related in part to the patients with suboptimal angiographic results who did not receive stents. In the provisional stenting described in this study, stents were used in ⬎90% of patients and only the lowest risk patients were treated with angioplasty alone. However, a direct comparison of this provisional stenting strategy with routine stenting in the background of intravenous glycoprotein IIb/IIIa inhibition is needed to validate these findings. Cost implications. Although ⬍10% of all eligible patients fall into the low risk category, our cost analysis shows a potentially substantial cost savings of $78 million on a national basis achieved by avoiding unnecessary stenting in this small subgroup. Prior cost analyses of stenting strategies have shown significantly higher in-hospital costs with stenting, but the difference in costs is offset somewhat by the lower rate of repeat revascularization procedures (3,35–37). Our findings suggest that in a select group of patients, treatment with balloon angioplasty alone is associated with repeat revascularization rates similar to those seen with stenting, and the initial cost savings achieved by avoiding stent use are maintained over six months. In addition to the economic benefits, these strategies lower the rates of instent restenosis.
JACC Vol. 37, No. 7, 2001 June 1, 2001:1883–90
Predictors of repeat revascularization. The multivariate model in this analysis identified three independent predictors of repeat revascularization among patients with optimal angiographic results: female gender, proximal LAD lesion and lesion length ⱖ10 mm. Absence of prior MI was also associated with the need for repeat revascularization, although with borderline statistical significance. Numerous prior studies have identified predictors of angiographic restenosis. Longer lesion length (38 – 40), proximal LAD lesion location (38,41– 43) and the absence of previous MI (44) have been shown to be associated with higher rates of angiographic restenosis in previous studies. With respect to gender differences, early studies reported higher restenosis rates in men (42,44,45). However, subsequent studies showed no difference between genders, and it has been postulated that the difference observed in earlier studies may have been related to inadequate lumen enlargement in larger vessels attributable to the limitations of older angioplasty equipment (46). Women have higher rates of acute closure after angioplasty, which may account for the more frequent need for repeat percutaneous intervention during the initial hospitalization (47,48). Diabetes has been shown to be a significant predictor of restenosis in some (39,43), but not all (40,42,49), prior studies. In this analysis, the slightly higher rate of repeat revascularization in patients with diabetes was not statistically significant (21.8% vs. 19.6%; p ⫽ 0.5). It is possible that patients with diabetes have higher rates of asymptomatic restenosis that may not lead to repeat revascularization. Alternatively, diabetes may not be an important risk factor for restenosis when optimal angiographic results are achieved. Higher restenosis rates have also been reported for total occlusions, ostial lesions and multivessel intervention (50). In this study, relatively few patients with optimal angiographic results had these characteristics (5%, 13% and 7% respectively), and the analysis was therefore insufficient to evaluate the association between these factors and repeat revascularization rates. Study limitations. The patients in this analysis were enrolled in clinical trials with strict inclusion and exclusion criteria, and complex lesions may therefore be underrepresented compared with clinical practice. Furthermore, saphenous vein graft interventions were excluded from the analysis. No direct comparison with coronary stenting was undertaken in this study. The low rate of repeat revascularization after standard balloon angioplasty in the low risk cohort compares favorably with the rates seen after stenting in prior studies. However, our findings require further validation, including prospective comparison with patients undergoing stenting. It is possible that patients with favorable clinical and angiographic characteristics who undergo stenting have significantly lower rates of repeat revascularization than the overall rate used for this analysis. Conclusions. Optimal or stent-like angiographic results (defined as residual diameter stenosis ⱕ30% and no dissection) are achieved in about 20% of patients and are associated with lower rates of repeat revascularization and MI
Cantor et al. Stent-Like Outcomes With Balloon Angioplasty
1889
than cases with non-SLRs. Further stratification of these cases based on gender, lesion length and lesion location allows identification of a group of patients at low risk for repeat revascularization. Men with stent-like angiographic results after angioplasty in short lesions located outside the proximal LAD are unlikely to require stent implantation and may be best managed with angioplasty alone. A provisional stent strategy based on the use of final angiographic results and preprocedural characteristics can provide substantial cost savings, with clinical outcomes equivalent to a strategy of universal stenting. Acknowledgment The authors thank Pat French for her help in reviewing and editing this manuscript. Reprint requests and correspondence: Dr. Warren J. Cantor, St. Michael’s Hospital, Division of Cardiology, 30 Bond Street, Toronto, Ontario, Canada M5B 1W8. E-mail: cantorw@smh. toronto.on.ca.
REFERENCES 1. Fischman DL, Leon MB, Baim DS, et al. A randomized comparison of coronary stent placement and balloon angioplasty in the treatment of coronary artery disease. N Engl J Med 1994;331:496 –501. 2. Serruys PW, De Jaegere P, Kiemeneij F, et al. A comparison of balloon-expandable-stent implantation with balloon angioplasty in patients with coronary artery disease. N Engl J Med 1994;331:489 –95. 3. Peterson ED, Cowper PA, DeLong ER, Zidar JP, Stack RS, Mark DB. Acute and long-term cost implications of coronary stenting. J Am Coll Cardiol 1999;33:1610 – 8. 4. Narins CR, Holmes DR, Topol EJ. A call for provisional stenting: the balloon is back. Circulation 1998;97:1298 –305. 5. Topol EJ, Leya F, Pinkerton CA, et al. A comparison of directional atherectomy with coronary angioplasty in patients with coronary artery disease. The CAVEAT Study Group. N Engl J Med 1993;329:221–7. 6. The EPIC Investigators. Use of a monoclonal antibody directed against the platelet glycoprotein IIb/IIIa receptor in high-risk coronary angioplasty. N Engl J Med 1994;330:956 – 61. 7. The IMPACT-II Investigators. Randomised placebo-controlled trial of effect of eptifibatide on complications of percutaneous coronary intervention: IMPACT-II. Integrelin to Minimise Platelet Aggregation and Coronary Thrombosis-II. Lancet 1997;349:1422– 8. 8. Faxon DP. Effect of high dose angiotensin-converting enzyme inhibition on restenosis: final results of the MARCATOR study, a multicenter, double-blind, placebo-controlled trial of cilazapril. The Multicenter American Research Trial With Cilazapril After Angioplasty to Prevent Transluminal Coronary Obstruction and Restenosis (MARCATOR) Study Group. J Am Coll Cardiol 1995;25:362–9. 9. Ohman EM, Marquis JF, Ricci DR, et al. A randomized comparison of the effects of gradual prolonged versus standard primary balloon inflation on early and late outcome. Results of a multicenter clinical trial. Circulation 1994;89:1118 –25. 10. Anderson RD, Ohman EM, Holmes DR, Jr, et al. Prognostic value of congestive heart failure history in patients undergoing percutaneous coronary interventions. J Am Coll Cardiol 1998;32:936 – 41. 11. Strauss BH, Escaned J, Foley DP et al. Technologic considerations and practical limitations in the use of quantitative angiography during percutaneous coronary recanalization. Prog Cardiovasc Dis 1994;36: 343– 62. 12. Mancini GB, Simon SB, McGillem MJ, LeFree MT, Friedman HZ, Vogel RA. Automated quantitative coronary arteriography: morphologic and physiologic validation in vivo of a rapid digital angiographic method. Circulation 1987;75:452– 60. 13. Carrozza JP, Jr, Schatz RA, George CJ, et al. Acute and long-term outcome after Palmaz-Schatz stenting: analysis from the New Ap-
1890
14. 15. 16. 17. 18. 19.
20. 21.
22.
23. 24. 25. 26.
27.
28. 29.
30.
31. 32.
Cantor et al. Stent-Like Outcomes With Balloon Angioplasty
proaches to Coronary Intervention (NACI) registry. Am J Cardiol 1997;80:78K– 88K. Kastrati A, Schomig A, Elezi S, et al. Predictive factors of restenosis after coronary stent placement. J Am Coll Cardiol 1997;30:1428 –36. Corcos T, Favereau X, Gue´rin Y. A call for professional stenting: the balloon is dead and buried! Circulation 1998;98:2644 –5. Gershlick AH, Baron J. Dealing with in-stent restenosis. Heart 1998;79:319 –23. Mintz GS, Hoffmann R, Mehran R, et al. In-stent restenosis: the Washington Hospital Center experience. Am J Cardiol 1998;81:7– 13E. Mehran R, Dangas G, Mintz GS, et al. In-Stent Restenosis: “The great equalizer”— disappointing clinical outcomes with ALL interventional strategies (abstr). J Am Coll Cardiol 1999;33:63–A. Foley DP, Serruys PW. Provisional stenting—stent-like balloon angioplasty: evidence to define the continuing role of balloon angioplasty for percutaneous coronary revascularization. Semin Interv Cardiol 1996;1:269 –73. Carere RG, Barbeau G, Dzavik V, et al. Do superior balloon angioplasty results provide stent-like outcomes in coronary occlusions (abstr)? Circulation 1998;98:I284. Shah V, Haude M, Erbel R, et al. Long-term follow-up of postPTCA results (ⱕ 30% residual diamter stenosis) in the restenotic lesions: results of the REST trial (abstr). J Am Coll Cardiol 1997;29: 76A. Holmes DR, Kip KE, Yeh W, Kelsey SF, Detre KM, Williams DO. Long-term analysis of conventional balloon angioplasty and an initial “stent-like” result: the NHLBI PTCA Registry. J Am Coll Cardiol 1998;32:590 –5. Espinola-Klein C, Rupprecht H-J, Trautmann S, et al. Ten year follow-up of patients with optimal “stent-like,” suboptimal or failed coronary angioplasty (abstr). Eur Heart J 1997;18:24. Weaver WD, Reisman MA, Griffin JJ, et al. Optimum percutaneous transluminal coronary angioplasty compared with routine stent strategy trial (OPUS-1): a randomised trial. Lancet 2000;355:2199 –203. Dangas G, Ambrose JA, Rehman D, et al. Balloon Optimization versus Stent Study (BOSS): provisional stenting and early recoil after balloon angioplasty. Am J Cardiol 2000;85:957– 61. Rodrı´guez A, Ayala F, Bernardi V, et al. Optimal Coronary Balloon Angioplasty with provisional Stenting versus primary stent (OCBAS): immediate and long-term follow-up results. J Am Coll Cardiol 1998;32:1351–7. Serruys PW, di Mario C, Piek J, et al. Prognostic value of intracoronary flow velocity and diameter stenosis in assessing the short- and long-term outcomes of coronary balloon angioplasty: the DEBATE study (Doppler Endpoints Balloon Angioplasty Trial Europe). Circulation 1997;96:3369 –77. Serruys PW, de Bruyne B, Carlier S, et al. Randomized comparison of primary stenting and provisional balloon angioplasty guided by flow velocity measurement. Circulation 2000;102:2930 –7. Lafont A, Dubois-Rande´ JL, Steg PG, et al. The French randomized optimal stenting trial: a prospective evaluation of provisional stenting guided by coronary velocity reserve and quantitative coronary angiography. J Am Coll Cardiol 2000;36:404 –9. Di Mario C, Moses JW, Anderson TJ, et al. Randomized comparison of elective stent implantation and coronary balloon angioplasty guided by online quantitative angiography and intracoronary Doppler. Circulation 2000;102:2938 – 44. Bech GJW, Pijls NHJ, de Bruyne B, et al. Usefulness of fractional flow reserve to predict clinical outcome after balloon angioplasty. Circulation 1999;99:883– 8. Kong DF, Hasselblad V, Tcheng JE, Ohman EM, Topol EJ, Califf RM. Clinical outcome improvements from coronary stenting: a systemic overview (abstr). Eur Heart J 1998;19:571.
JACC Vol. 37, No. 7, 2001 June 1, 2001:1883–90 33. The ESPRIT Investigators. The ESPRIT study: a randomised, placebo-controlled trial of a novel dosing regimen of eptifibatide in planned coronary stent implantation. Lancet 2000;356:2037– 44. 34. Topol EJ, Mark DB, Lincoff AM, et al. Outcomes at 1 year and economic implications of platelet glycoprotein IIb/IIIa blockade in patients undergoing coronary stenting: results from a multicentre randomised trial. Lancet 1999;354:2019 –24. 35. Cohen DJ, Breall JA, Ho KK, et al. Economics of elective coronary revascularization. Comparison of costs and charges for conventional angioplasty, directional atherectomy, stenting and bypass surgery. J Am Coll Cardiol 1993;22:1052–9. 36. Cohen DJ, Krumholz HM, Sukin CA, et al. In-hospital and one-year economic outcomes after coronary stenting or balloon angioplasty: results from a randomized clinical trial. Circulation 1995;92:2480 –7. 37. Dick RJ, Popma JJ, Muller DWM, Burek KA, Topol EJ. In-hospital costs associated with new percutaneous coronary devices. Am J Cardiol 1991;68:879 – 85. 38. Hirshfeld JW, Jr, Schwartz JS, Jugo R, et al. Restenosis after coronary angioplasty: a multivariate statistical model to relate lesion and procedure variables to restenosis. The M-HEART Investigators. J Am Coll Cardiol 1991;18:647–56. 39. Rensing BJ, Hermans WR, Vos J, et al. Luminal narrowing after percutaneous transluminal coronary angioplasty. A study of clinical, procedural, and lesional factors related to long-term angiographic outcome. Coronary Artery Restenosis Prevention on Repeated Thromboxane Antagonism (CARPORT) Study Group. Circulation 1993;88:975– 85. 40. Bourassa MG, Lesperance J, Eastwood C, et al. Clinical, physiologic, anatomic and procedural factors predictive of restenosis after percutaneous transluminal coronary angioplasty. J Am Coll Cardiol 1991;18: 368 –76. 41. Halon DA, Merdler A, Shefer A, Flugelman MY, Lewis BS. Identifying patients at high risk for restenosis after percutaneous transluminal coronary angioplasty for unstable angina pectoris. Am J Cardiol 1989;64:289 –93. 42. Leimgruber PP, Roubin GS, Hollman J, et al. Restenosis after successful coronary angioplasty in patients with single-vessel disease. Circulation 1986;73:710 –7. 43. Weintraub WS, Kosinski AS, Brown CL, King SB. Can restenosis after coronary angioplasty be predicted from clinical variables? J Am Coll Cardiol 1993;21:6 –14. 44. Holmes DR, Jr, Vlietstra RE, Smith HC, et al. Restenosis after percutaneous transluminal coronary angioplasty (PTCA): a report from the PTCA Registry of the National Heart, Lung, and Blood Institute. Am J Cardiol 1984;53:77C– 81C. 45. Vandormael MG, Deligonul U, Kern MJ, et al. Multilesion coronary angioplasty: clinical and angiographic follow-up. J Am Coll Cardiol 1987;10:246 –52. 46. Califf RM, Fortin DF, Frid DJ, et al. Restenosis after coronary angioplasty: an overview. J Am Coll Cardiol 1991;17:2B–13B. 47. Ellis SG, Roubin GS, King SB, III, et al. Angiographic and clinical predictors of acute closure after native vessel coronary angioplasty. Circulation 1988;77:372–9. 48. Cantor WJ, Kramer JM, Miller JM, et al. Higher mortality in women following angioplasty explained by differences in body size, not vessel size (abstr). Circulation 1999;100:I780. 49. MacDonald RG, Henderson MA, Hirshfeld JW, Jr, et al. Patientrelated variables and restenosis after percutaneous transluminal coronary angioplasty—a report from the M-HEART Group. Am J Cardiol 1990;66:926 –31. 50. Miller JM, Ohman EM, Moliterno DJ, Califf RM. Restenosis: the clinical issues. In: Topol EJ, ed. Textbook of Interventional Cardiology. Philadelphia, PA: W.B. Saunders, 1999:393.
Vol. 37, No. 7, 2001 ISSN 0735-1097/01/$20.00 PII S0735-1097(01)01244-X
Achieving Optimal Results With Standard Balloon Angioplasty: Can Baseline and Angiographic Variables Predict Stent-Like Outcomes? Warren J. Cantor, MD,* Anne S. Hellkamp, MS,† Eric D. Peterson, MD, MPH,† James P. Zidar, MD,† Patricia A. Cowper, PHD,† Michael H. Sketch JR, MD,† James E. Tcheng, MD,† Robert M. Califf, MD,† E. Magnus Ohman, MD† Toronto, Canada and Durham, North Carolina To predict which patients might not require stent implantation, we identified clinical and angiographic characteristics associated with repeat revascularization after standard balloon angioplasty. BACKGROUND Stents reduce the risk of repeat revascularization but are costly and may lead to in-stent restenosis, which remains difficult to treat. Identification of patients at low risk for repeat revascularization may allow clinicians to reserve stents for patients most likely to benefit. METHODS Data from five interventional trials (5,146 patients) were pooled for analysis. We identified patients with optimal angiographic results (final diameter stenosis ⱕ30% and no dissection) after balloon angioplasty and determined the multivariable predictors of repeat revascularization. RESULTS Optimal angiographic results were achieved in 18% of patients after angioplasty. The repeat revascularization rate at six months was lower for patients with optimal results (20% vs. 26%, p ⬍ 0.001) but still higher than observed in stent trials. Independent predictors of repeat revascularization were female gender (odds ratio [OR] 1.67, p ⫽ 0.01), lesion length ⱖ10 mm (OR 1.62, p ⫽ 0.03) and proximal left anterior descending coronary artery lesions (OR 1.62, p ⫽ 0.03). For the 8% of patients with optimal angiographic results and none of these risk factors, the repeat revascularization and target vessel revascularization rates were 14% and 8% respectively, similar to rates after stent implantation. Cost analysis estimated that $78 million per year might be saved in the U.S. with a provisional stenting strategy using these criteria compared with elective stenting. CONCLUSIONS A combination of baseline characteristics and angiographic results can be used to identify a small group of patients at very low risk for repeat revascularization after balloon angioplasty. Provisional stenting for these low risk patients could substantially reduce costs without compromising clinical outcomes. (J Am Coll Cardiol 2001;37:1883–90) © 2001 by the American College of Cardiology OBJECTIVES
Coronary stents have been shown to reduce rates of clinical and angiographic restenosis in selected patients (1,2). Stent implantation is costly, however, adding more than $2,000 in direct costs to each interventional case (3). Given these cost considerations, provisional stenting (stenting in selective cases as opposed to all cases) has been proposed as an alternative (4). The selection criteria for provisional stenting, however, have not been fully elucidated. We hypothesized that the same low rate of repeat revascularization might be achieved without stent implantation when optimal angiographic results are obtained from balloon angioplasty alone. Specifically, using a large registry from five clinical trials, we investigated the degree to which angiographic results, patient clinical characteristics or both From *St. Michael’s Hospital, Toronto, Canada; and †Duke Clinical Research Institute, Durham, North Carolina. Sponsored in part through a research fellowship grant jointly funded by the Royal College of Physicians and Surgeons of Canada, the Medical Research Council of Canada, and Novartis Pharmaceuticals Corp., East Hanover, New Jersey. Presented in part at the 48th Annual Scientific Session of the American College of Cardiology, March 9, 1999, New Orleans, Louisiana, and the 72nd Scientific Sessions of the American Heart Association, November 10, 1999, Atlanta, Georgia. Manuscript received August, 22, 2000; revised manuscript received January 30, 2001, accepted February 13, 2001.
could be used to identify patients with low repeat revascularization rates. After defining these angiographic and clinical criteria, we determined the potential economic effects of applying these criteria to limit elective stent usage.
METHODS Patient population. Data were pooled from five randomized trials of percutaneous coronary interventions (5–9). The protocols and study populations of the five trials and the methods used to pool the data have been described in detail (10). The Coronary Angioplasty Versus Excisional Atherectomy Trial (CAVEAT) randomized 1,012 patients to undergo angioplasty or directional coronary atherectomy. Patients who underwent atherectomy were excluded from the analysis. In both the Evaluation of 7E3 for the Prevention of Ischemic Complications (EPIC) (2,099 patients) and Integrilin to Minimize Platelet Aggregation and Coronary Thrombosis (IMPACT-II) (4,010 patients) studies, patients undergoing percutaneous interventions were randomly allocated to receive placebo or an intravenous glycoprotein IIb/IIIa inhibitor—abciximab or eptifibatide, respectively. The Multicenter American Research trial with
1884
Cantor et al. Stent-Like Outcomes With Balloon Angioplasty
Abbreviations and Acronyms BENESTENT ⫽ BElgium-NEtherlands Stent CAVEAT ⫽ Coronary Angioplasty Versus Excisional Atherectomy Trial CK-MB ⫽ creatine kinase-MB EPIC ⫽ Evaluation of 7E3 for the Prevention of Ischemic Complications EPISTENT ⫽ Evaluation of Platelet IIb/IIIa Inhibitor for Stenting IMPACT II ⫽ Integrelin to Minimize Platelet Aggregation and Coronary Thrombosis II LAD ⫽ left anterior descending coronary artery MARCATOR ⫽ Multicenter American Research trial with Cilazapril after Angioplasty To prevent coronary Obstruction and Restenosis MI ⫽ myocardial infarction OPUS ⫽ Optimum Percutaneous transluminal coronary angioplasty compared with roUtine Stent PBC ⫽ Perfusion Balloon Catheter study QCA ⫽ quantitative coronary angiography SLR ⫽ stent-like result TVR ⫽ target vessel revascularization
Cilazapril after Angioplasty To prevent coronary Obstruction and Restenosis (MARCATOR) assessed the effect of cilazapril on restenosis in 1,436 patients undergoing angioplasty. The Perfusion Balloon Catheter (PBC) study compared prolonged versus short inflations using the perfusionballoon catheter in 478 patients. The data in these trials were collected during various periods, ranging from April 1989 to November 1995. Patients with acute myocardial infarction (MI) were excluded before pooling the data. The pooled data contain detailed clinical and angiographic data for 8,726 patients. After excluding patients who had stents implanted (n ⫽ 159), vein-graft intervention (n ⫽ 322), left-main intervention (n ⫽ 130), atherectomy or laser treatment (n ⫽ 1,469), failed or partially successful angioplasty (final diameter stenosis ⱖ50% in any attempted lesion, n ⫽ 1,047) and incomplete angiographic data or six-month status (n ⫽ 453), there were 5,146 patients for this analysis. Interventional procedures. Before the procedure, all patients received aspirin (160 to 325 mg/day). Patients in the PBC trial also received dipyridamole. Angioplasty procedures were performed using standard techniques, with the exception of perfusion-balloon catheters used in the PBC trial. Stenting was unusual in these trials (and excluded from this analysis), being performed in ⬍2% of cases. Intravenous glycoprotein IIb/IIIa inhibitors were used only in the active treatment arms of EPIC and IMPACT-II, accounting for 2,099 (41%) of the patients in this analysis. During the procedure, all patients received intravenous heparin to reach a target activated clotting time of ⬎300 s (⬎350 s for CAVEAT). After intervention, creatine kinase (CK)-MB
JACC Vol. 37, No. 7, 2001 June 1, 2001:1883–90
levels were obtained routinely over the next 24 h. All patients were discharged on aspirin 325 mg/day. Data collection. For each trial, the participating sites collected data prospectively on case report forms and sent them to data coordinating centers. Verification of the data was performed using consistency checks and double data entry. Sites were queried for missing or questionable data. The dataset for each trial was verified by retrospective review of patient charts. Complete copies of the individual datasets from the five trials were combined into a single database. Clinical follow-up. All five trials followed patients for at least six months for the occurrence of death, MI, repeat angioplasty or bypass surgery. For repeat angioplasty, involvement of the target vessel was not recorded in MARCATOR but was specified in the other trials. Angiographic analysis. For all five trials, cineangiograms were sent for review to independent core angiographic laboratories blind to treatment assignment. For EPIC and IMPACT-II, diameter stenosis was measured using calipers without catheter calibration. Except for 860 patients in the IMPACT-II angiographic substudy, absolute lumen dimensions were not obtained for patients in these two trials. For the other three trials and for the IMPACT-II angiographic substudy, quantitative coronary angiography (QCA) was performed using automated edge-detection algorithms as described previously (11,12). Definitions. Myocardial infarction was defined as new Q-waves on the electrocardiogram or elevation of CK or CK-MB. Enzymatic criteria for MI were values ⱖ2 times the upper limit of normal for MARCATOR and the PBC trial and three times the upper limit for the other trials. All events were adjudicated by independent committees blind to randomization. Repeat revascularization included all repeat percutaneous coronary interventions and bypass operations within the first six months, whether or not the target lesion was involved. To estimate the rate of target vessel revascularization (TVR), repeat percutaneous interventions without target vessel data were assumed to have the same proportion of target vessel involvement as the population for whom these data were available. All surgical procedures were assumed to have involved the target vessel. Multilesion interventions were considered “stent-like” results only if the stent-like criteria were achieved in all lesions attempted. Statistical analysis. Baseline, procedural and angiographic characteristics are summarized as medians with 25th and 75th percentiles for continuous measures and as percentages for discrete measures. Event rates are summarized as percentages with 95% CI. For comparisons between groups, the likelihood ratio chi-square test was used for discrete variables and the Wilcoxon rank-sum test was used for continuous variables. Sensitivity and specificity for predicting six-month survival free of repeat revascularization were calculated for stent-like result (SLR) criteria based on absence of dissection and residual diameter stenosis levels of ⱕ10%, ⱕ20%, ⱕ25%, ⱕ30%, ⱕ40% and ⱕ50%. These
JACC Vol. 37, No. 7, 2001 June 1, 2001:1883–90
criteria excluded all dissections, on the basis of a survey that indicates that most interventional cardiologists use stents routinely for any angiographically visible dissection (Cantor et al., unpublished data). We defined the best threshold for residual diameter stenosis to be the level that included the most patients yet maintained ⱖ80% specificity to predict survival free of repeat revascularization at six months. Multivariable logistic regression was performed using a stepwise, backward-elimination algorithm to determine independent predictors of repeat revascularization at six months in cases with SLRs. Variables were included in the multivariable model if they were significant in univariable analysis or if considered clinically important. The independent predictors were used to construct a classification and regression tree and classify patients with stent-like angiographic results as high or low risk for repeat revascularization. We validated the CART model by dividing the data into training and test sets to do the development and testing separately. We developed the classification scheme on half of the patients with SLR and tested it on the other half. Once agreement between the two samples was established, we then applied the scheme to the full sample using SLR as the first split in the tree. Cost-effectiveness analysis. The six-month costs and clinical outcomes of elective and provisional stenting strategies were compared using decision-analysis techniques (DATA 3.5, Treeage Software, Williamstown, Massachusetts). The rates of stenting for the provisional stent strategies and the probability of repeat revascularization after balloon angioplasty were based on the results of the analysis described in the preceding text. The cost estimates for angioplasty, stent procedures and bypass surgery were based on a cost analysis of patients undergoing percutaneous coronary intervention at Duke University Medical Center (3). This previous analysis was based on the Transition 1 Accounting System (Transition Systems, Inc., Boston, Massachusetts), which
Cantor et al. Stent-Like Outcomes With Balloon Angioplasty
1885
Table 1. Freedom From Repeat Revascularization at Six Months Final Diameter Stenosis*
n (%)
Sensitivity
Specificity
ⱕ10% ⱕ20% ⱕ25% ⱕ30% ⱕ35% ⱕ40% ⱕ50%
64 (1.2%) 319 (6.2%) 575 (11.2%) 938 (18.2%) 1,351 (26.3%) 1,793 (34.8%) 2,329 (45.3%)
1.4% 6.7% 11.9% 19.3% 27.3% 35.4% 61.2%
99.2% 95.5% 90.9% 85.1% 76.9% 67.0% 55.7%
*With no angiographically visible dissection.
contains detailed cost records at the intermediate product level within each department (for example, number of specific diagnostic tests, hours of care by in-patient unit or time spent in catheterization laboratory). The acquisition cost of a stent was $1,595. For this analysis, indirect costs associated with stent and angioplasty procedures were assumed to be equal. The repeat revascularization rate for stented patients was estimated from randomized trial and registry results (2,13,14). Sensitivity analysis was performed to examine the effects of varying estimated costs and event rates over a wide range of values.
RESULTS Determination of angiographic diameter stenosis criteria for SLR. A total of 5,146 patients were included in the analysis. The numbers of patients with final diameter stenoses at or below various thresholds are summarized in Table 1. Only 1% of cases achieved a residual diameter stenosis ⱕ10% with no dissection as assessed by the core lab. The event rates for the composite end point of death, MI or any repeat revascularization for the different criteria of SLR are shown in Figure 1. Although there was no apparent relation between the final diameter stenosis and
Figure 1. Clinical event rates at six months for four different thresholds of postintervention diameter stenosis. DS ⫽ diameter stenosis; MI ⫽ myocardial infarction; QCA ⫽ quantitative coronary angiography.
1886
Cantor et al. Stent-Like Outcomes With Balloon Angioplasty
JACC Vol. 37, No. 7, 2001 June 1, 2001:1883–90
Table 2. Six-Month Clinical Outcomes With and Without Stent-Like Results
Death Myocardial infarction Repeat angioplasty Bypass surgery Any repeat revascularization Composite of above
Stent-Like Result (n ⴝ 938)
No Stent-Like Result (n ⴝ 4,208)
p Value
0.4% (0.0–0.9) 3.6% (2.4–4.8) 17.1% (14.7–19.5) 4.4% (3.1–5.8) 20.0% (17.5–22.6) 22.0% (19.3–24.6)
0.6% (0.3–0.8) 5.6% (4.9–6.3) 21.0% (19.7–22.2) 6.6% (5.8–7.3) 25.5% (24.1–26.8) 27.6% (26.3–29.0)
0.6 0.009 0.007 0.11 ⬍ 0.001 ⬍ 0.001
Data are presented as percentage (95% CI).
the rates of death and MI, the repeat revascularization rates increased with increasing diameter stenosis thresholds. The sensitivity and specificity for the different thresholds of postangioplasty residual diameter stenosis (core laboratory determination) in predicting freedom from repeat revascularization are shown in Table 1. To maintain ⱖ80% specificity, a diameter stenosis threshold ⱕ30% would need to be used as criteria for a SLR. The clinical event rates for the final diameter stenosis thresholds of 20%, 25% and 30% were similar. Therefore, we defined a SLR as a final diameter stenosis ⱕ30% without dissection. Only 938 eligible patients (18%) met these criteria. Clinical outcomes with SLRs. The six-month rates of death, MI and repeat revascularization for the two groups are summarized in Table 2. Patients with SLR had significantly lower rates of MI (4% vs. 6%; p ⫽ 0.009) and repeat revascularization (20% vs. 26%; p ⬍ 0.001). In the SLR group, the target vessel was involved in 64 (65%) of the repeat angioplasty cases in which vessel location was known. The estimated TVR rate for the SLR group was 14.5%. Predictors of repeat revascularization. Univariate and multivariable regression analyses were performed for the SLR cases to identify predictors of repeat revascularization at six months (Table 3). The univariate predictors of repeat revascularization were female gender, proximal left anterior
descending coronary artery (LAD) lesion, lesion length ⱖ10 mm and no prior MI. The same four variables were independent predictors of repeat revascularization in the multivariate model. Subgroups of the SLRs. Figure 2 shows the classification and regression tree analysis for patients who achieved stent-like angiographic results, stratified by presence or absence of three of the independent predictors of repeat revascularization. In the lowest risk subgroup (men with lesions ⬍10 mm long and no proximal LAD lesions) the repeat revascularization rate at six months was 14% (95% CI, 10% to 18%). The estimated TVR rate in this group was 8.1%. In contrast, the repeat revascularization rate of the highest risk subgroup (women with lesions ⱖ10 mm long and proximal LAD lesions) was 46% (95% CI, 20% to 68%). About 45% of patients with an SLR (excluding cases for which lesion length was missing) were in the lowest risk subgroup, representing about 8% of all eligible patients. Economic implications of the two provisional stenting strategies. We evaluated the expected repeat revascularization rates and costs of two provisional stenting strategies and an elective stenting strategy for stent use based on event rates described in the preceding text. In the elective stenting strategy, all patients would undergo stent implantation irrespective of the angiographic result after balloon angio-
Table 3. Predictors of Repeat Revascularization at Six Months Univariable
Multivariable
Factor
Odds Ratio (95% CI)
p Value
Age (per 10-year increase) Female gender Diabetes Congestive heart failure Unstable angina Prior infarction Multivessel disease Proximal left anterior descending lesion Ostial lesion Total occlusion Calcification Length ⱖ10 mm Reference vessel diameter* Preprocedural minimum luminal diameter* Preprocedural percent diameter stenosis* Postprocedural minimum luminal diameter* Postprocedural percent diameter stenosis*
1.12 (0.97–1.30) 1.88 (1.34–2.64) 1.14 (0.77–1.69) 0.63 (0.26–1.53) 1.16 (0.80–1.66) 0.71 (0.52–0.99) 1.09 (0.78–1.53) 1.46 (1.01–2.12) 0.99 (0.58–1.70) 0.79 (0.35–1.82) 1.26 (0.86–1.86) 1.61 (1.05–2.47) 1.10 (0.76–1.61) 1.02 (0.54–1.93) 1.00 (0.99–1.01) 1.03 (0.61–1.73) 1.02 (0.99–1.04)
0.13 ⬍ 0.001 0.50 0.31 0.44 0.04 0.62 0.04 0.97 0.59 0.24 0.03 0.61 0.95 0.95 0.91 0.14
*Per one-unit increase in the covariate value. †Variables “forced” into multivariate model.
Odds Ratio (95% CI)
p Value
1.67 (1.12–2.48)
0.01
0.68 (0.46–1.01)
0.05
1.62 (1.06–2.47) 0.87 (0.50–1.52)†
0.03 0.63
1.11 (0.71–1.73)† 1.62 (1.04–2.51)
0.66 0.03
1.02 (0.99–1.04)†
0.27
Cantor et al. Stent-Like Outcomes With Balloon Angioplasty
JACC Vol. 37, No. 7, 2001 June 1, 2001:1883–90
Figure 2. Stratification of risk for repeat revascularization at six months for patients with optimal (stent-like) angiographic results. Using three baseline characteristics (gender, lesion length and lesion location), patients are classified as low risk (repeat revascularization rate of 14%) or high risk (repeat revascularization rate of 25%). DS ⫽ diameter stenosis; LAD ⫽ left anterior descending coronary artery.
plasty. Repeat revascularization rates for this strategy were assumed to be 14% (2,13,14). In the second strategy (provisional stenting guided by baseline characteristics and angiographic results), patients would not undergo stenting if they had a stent-like angiographic result (residual diameter stenosis ⱕ30%, no dissection) and none of the three predictors of repeat revascularization (female gender, lesion length ⱖ10 mm or proximal LAD lesion). With this strategy, 8% of patients would be treated with angioplasty alone and the remainder of patients would receive stents. The repeat revascularization rate of 14% in this low risk subgroup is equal to the repeat revascularization rate after stent implantation (2,13,14). In the third strategy (provisional stenting guided by angiography only), patients who achieve a stent-like angiographic result with angioplasty
(18% of patients) were not stented. The repeat revascularization rate for patients treated with angioplasty alone in this strategy is 20%. On the basis of the modalities of repeat revascularization observed in our database, we assumed that percutaneous revascularization would be used in 75% of repeat interventions, bypass surgery in 17% and both in the remaining 8%. Patients undergoing repeat percutaneous intervention were assumed to receive a stent if they were treated initially with angioplasty only, and vice versa. Table 4 summarizes the clinical outcomes and costs of the three strategies. In the elective stenting strategy and the provisional stenting strategy guided by both baseline characteristics and angiographic results, the expected rate of repeat revascularization is 14%. The expected rate of repeat revascularization in the provisional stenting strategy guided by angiographic criteria alone is 1% higher (15%). Both provisional stenting strategies result in cost savings, with the lowest cost in the angiographic-guided provisional stenting strategy. In this least-cost strategy, the savings from using fewer stents are minimally offset by the costs associated with the slightly higher repeat revascularization rate. Although expected costs per patient do not vary greatly among these strategies, when considered from a health system perspective, substantial savings could be achieved by following the provisional stenting strategies. To assess the sensitivity of results to model assumptions, we varied key parameters in the model. The cost savings associated with provisional stenting guided by both baseline characteristics and angiography declined with the cost of stents. If stent costs fell to 50% of the baseline value, aggregate savings would fall to $37 million. However, the provisional strategy did not become more expensive than elective stenting until stent costs fell to ⬍$200. The cost savings were also sensitive to the rate of repeat revascularization after stenting. For example, if the repeat revascularization rate were 10% rather than 14%, the aggregate savings would fall to $53 million. The cost savings with a repeat revascularization rate of 5% after stenting are more modest at $23 million. The provisional stenting strategy guided by angiography alone became more expensive than the more selective provisional stenting strategy with very small changes in stent costs and repeat revascularization rates. This sensitivity is
Table 4. Six-Month Clinical Outcomes and Costs for Stenting Strategies
Strategy Elective stenting for all patients Provisional stenting strategies Angiographic‡ and clinical§ criteria Angiographic‡ criteria only
1887
Cases Stented
Expected Repeat Revascularization Rate (%)*
Overall Six-Month Cost
Overall Savings in U.S./Year†
100%
14%
$16,396
—
92% 82%
14% 15%
$16,240 $16,231
$78 million $83 million
*Repeat revascularization rate after stenting assumed to be 14% for all strategies. †Assumes 500,000 stent cases performed per year in the U.S. ‡Stent if final diameter stenosis ⬎30% or dissection present. §Stent if female, lesion length ⱖ10 mm or lesion in proximal left anterior descending artery.
1888
Cantor et al. Stent-Like Outcomes With Balloon Angioplasty
due to the relatively high rate of repeat revascularization among the 18% of patients not initially stented. This strategy is therefore both clinically and economically unattractive.
DISCUSSION Major findings. We have shown that the combination of angiographic residual diameter stenosis and baseline patient and lesion characteristics can be used to identify a subset of patients at low risk for repeat revascularization after standard angioplasty. Combining angiographic results with baseline characteristics improves the sensitivity for predicting repeat revascularization compared with reliance upon angiographic results alone. The rate of repeat revascularization in the low risk cohort is similar to that observed with stent implantation (2,13,14). Although only a small proportion of patients undergoing angioplasty fall into this low risk category, the cost savings achieved by avoiding unnecessary stent use in this subgroup are substantial. Provisional stenting. Randomized trials have shown lower restenosis and repeat revascularization rates with stenting than with balloon angioplasty (1,2). On the basis of these encouraging results, many institutions advocate and practice universal stenting (15). When restenosis does occur within stents, however, it is often difficult to manage and refractory to conventional percutaneous interventions (16 –18). Previous attempts to identify patients with favorable outcomes after standard angioplasty have focused on the postprocedural residual diameter stenosis. The BElgiumNEtherlands Stent (BENESTENT) trial investigators showed that patients with stent-like angiographic results, defined as a residual stenosis ⱕ30% with no major dissection, had clinical and angiographic outcomes similar to those of patients who underwent stenting (19). This finding has been validated in other stent trials (20,21) and angioplasty registries (22,23). The Optimum Percutaneous transluminal coronary angioplasty compared with roUtine Stent strategy (OPUS) study was the first randomized trial of provisional stenting based on angiographic results (24). A total of 479 patients were randomized to a strategy of elective stent implantation or initial balloon angioplasty, with provisional stenting for a residual stenosis ⱖ20% by visual assessment (or ⱖ30% by QCA), significant dissection or threatened closure. At six months, patients randomized to elective stenting had lower rates of TVR. The use of visual estimation rather than QCA may account in part for the discrepancy with previous studies. However, these results are consistent with our observation that angiographic criteria alone may be inadequate to identify patients at low risk for restenosis and repeat intervention. Two small trials have shown the feasibility of a provisional stenting strategy guided by repeating angiography 20 to 30 min after balloon angioplasty to assess for early recoil (25,26). The use of intravascular ultrasound and physiologic
JACC Vol. 37, No. 7, 2001 June 1, 2001:1883–90
measurements, including coronary and fractional flow reserve, may help discriminate which patients will have favorable clinical outcomes with balloon angioplasty alone (27–31). However, these modalities prolong procedure times and carry additional costs. In contrast, the use of angiographic results and baseline characteristics to guide stent use is simple and inexpensive. Our analysis focused primarily on the incidence of repeat revascularization. A meta-analysis of trials comparing stenting and balloon angioplasty demonstrated that the primary clinical benefit of stent implantation is a reduction in the need for repeat revascularization, with no difference in the incidence of death or MI (32). The other potential advantage of stent implantation is a reduction in the incidence of abrupt occlusion. In this study, the rates of death, MI and repeat revascularization within 48 h after intervention for patients with optimal angiographic results were low (0%, 2% and 3% respectively) and similar to the rates reported after stent implantation in a recent large randomized trial (33). Nevertheless, we cannot exclude the possibility that early ischemic complications may occur more often with provisional stenting. Glycoprotein IIb/IIIa inhibitors were used in 41% of the cases in this study. In the Evaluation of Platelet IIb/IIIa Inhibitor for Stenting (EPISTENT) study, the combination of stenting and glycoprotein IIb/IIIa inhibitors was associated with significantly lower rates of death, MI and TVR at one year than either therapy alone (34). However, stents were used in the angioplasty group only for reduced coronary flow, long dissections or a residual diameter stenosis ⬎70% and were implanted in only 19% of patients. The outcomes of the EPISTENT angioplasty group were likely related in part to the patients with suboptimal angiographic results who did not receive stents. In the provisional stenting described in this study, stents were used in ⬎90% of patients and only the lowest risk patients were treated with angioplasty alone. However, a direct comparison of this provisional stenting strategy with routine stenting in the background of intravenous glycoprotein IIb/IIIa inhibition is needed to validate these findings. Cost implications. Although ⬍10% of all eligible patients fall into the low risk category, our cost analysis shows a potentially substantial cost savings of $78 million on a national basis achieved by avoiding unnecessary stenting in this small subgroup. Prior cost analyses of stenting strategies have shown significantly higher in-hospital costs with stenting, but the difference in costs is offset somewhat by the lower rate of repeat revascularization procedures (3,35–37). Our findings suggest that in a select group of patients, treatment with balloon angioplasty alone is associated with repeat revascularization rates similar to those seen with stenting, and the initial cost savings achieved by avoiding stent use are maintained over six months. In addition to the economic benefits, these strategies lower the rates of instent restenosis.
JACC Vol. 37, No. 7, 2001 June 1, 2001:1883–90
Predictors of repeat revascularization. The multivariate model in this analysis identified three independent predictors of repeat revascularization among patients with optimal angiographic results: female gender, proximal LAD lesion and lesion length ⱖ10 mm. Absence of prior MI was also associated with the need for repeat revascularization, although with borderline statistical significance. Numerous prior studies have identified predictors of angiographic restenosis. Longer lesion length (38 – 40), proximal LAD lesion location (38,41– 43) and the absence of previous MI (44) have been shown to be associated with higher rates of angiographic restenosis in previous studies. With respect to gender differences, early studies reported higher restenosis rates in men (42,44,45). However, subsequent studies showed no difference between genders, and it has been postulated that the difference observed in earlier studies may have been related to inadequate lumen enlargement in larger vessels attributable to the limitations of older angioplasty equipment (46). Women have higher rates of acute closure after angioplasty, which may account for the more frequent need for repeat percutaneous intervention during the initial hospitalization (47,48). Diabetes has been shown to be a significant predictor of restenosis in some (39,43), but not all (40,42,49), prior studies. In this analysis, the slightly higher rate of repeat revascularization in patients with diabetes was not statistically significant (21.8% vs. 19.6%; p ⫽ 0.5). It is possible that patients with diabetes have higher rates of asymptomatic restenosis that may not lead to repeat revascularization. Alternatively, diabetes may not be an important risk factor for restenosis when optimal angiographic results are achieved. Higher restenosis rates have also been reported for total occlusions, ostial lesions and multivessel intervention (50). In this study, relatively few patients with optimal angiographic results had these characteristics (5%, 13% and 7% respectively), and the analysis was therefore insufficient to evaluate the association between these factors and repeat revascularization rates. Study limitations. The patients in this analysis were enrolled in clinical trials with strict inclusion and exclusion criteria, and complex lesions may therefore be underrepresented compared with clinical practice. Furthermore, saphenous vein graft interventions were excluded from the analysis. No direct comparison with coronary stenting was undertaken in this study. The low rate of repeat revascularization after standard balloon angioplasty in the low risk cohort compares favorably with the rates seen after stenting in prior studies. However, our findings require further validation, including prospective comparison with patients undergoing stenting. It is possible that patients with favorable clinical and angiographic characteristics who undergo stenting have significantly lower rates of repeat revascularization than the overall rate used for this analysis. Conclusions. Optimal or stent-like angiographic results (defined as residual diameter stenosis ⱕ30% and no dissection) are achieved in about 20% of patients and are associated with lower rates of repeat revascularization and MI
Cantor et al. Stent-Like Outcomes With Balloon Angioplasty
1889
than cases with non-SLRs. Further stratification of these cases based on gender, lesion length and lesion location allows identification of a group of patients at low risk for repeat revascularization. Men with stent-like angiographic results after angioplasty in short lesions located outside the proximal LAD are unlikely to require stent implantation and may be best managed with angioplasty alone. A provisional stent strategy based on the use of final angiographic results and preprocedural characteristics can provide substantial cost savings, with clinical outcomes equivalent to a strategy of universal stenting. Acknowledgment The authors thank Pat French for her help in reviewing and editing this manuscript. Reprint requests and correspondence: Dr. Warren J. Cantor, St. Michael’s Hospital, Division of Cardiology, 30 Bond Street, Toronto, Ontario, Canada M5B 1W8. E-mail: cantorw@smh. toronto.on.ca.
REFERENCES 1. Fischman DL, Leon MB, Baim DS, et al. A randomized comparison of coronary stent placement and balloon angioplasty in the treatment of coronary artery disease. N Engl J Med 1994;331:496 –501. 2. Serruys PW, De Jaegere P, Kiemeneij F, et al. A comparison of balloon-expandable-stent implantation with balloon angioplasty in patients with coronary artery disease. N Engl J Med 1994;331:489 –95. 3. Peterson ED, Cowper PA, DeLong ER, Zidar JP, Stack RS, Mark DB. Acute and long-term cost implications of coronary stenting. J Am Coll Cardiol 1999;33:1610 – 8. 4. Narins CR, Holmes DR, Topol EJ. A call for provisional stenting: the balloon is back. Circulation 1998;97:1298 –305. 5. Topol EJ, Leya F, Pinkerton CA, et al. A comparison of directional atherectomy with coronary angioplasty in patients with coronary artery disease. The CAVEAT Study Group. N Engl J Med 1993;329:221–7. 6. The EPIC Investigators. Use of a monoclonal antibody directed against the platelet glycoprotein IIb/IIIa receptor in high-risk coronary angioplasty. N Engl J Med 1994;330:956 – 61. 7. The IMPACT-II Investigators. Randomised placebo-controlled trial of effect of eptifibatide on complications of percutaneous coronary intervention: IMPACT-II. Integrelin to Minimise Platelet Aggregation and Coronary Thrombosis-II. Lancet 1997;349:1422– 8. 8. Faxon DP. Effect of high dose angiotensin-converting enzyme inhibition on restenosis: final results of the MARCATOR study, a multicenter, double-blind, placebo-controlled trial of cilazapril. The Multicenter American Research Trial With Cilazapril After Angioplasty to Prevent Transluminal Coronary Obstruction and Restenosis (MARCATOR) Study Group. J Am Coll Cardiol 1995;25:362–9. 9. Ohman EM, Marquis JF, Ricci DR, et al. A randomized comparison of the effects of gradual prolonged versus standard primary balloon inflation on early and late outcome. Results of a multicenter clinical trial. Circulation 1994;89:1118 –25. 10. Anderson RD, Ohman EM, Holmes DR, Jr, et al. Prognostic value of congestive heart failure history in patients undergoing percutaneous coronary interventions. J Am Coll Cardiol 1998;32:936 – 41. 11. Strauss BH, Escaned J, Foley DP et al. Technologic considerations and practical limitations in the use of quantitative angiography during percutaneous coronary recanalization. Prog Cardiovasc Dis 1994;36: 343– 62. 12. Mancini GB, Simon SB, McGillem MJ, LeFree MT, Friedman HZ, Vogel RA. Automated quantitative coronary arteriography: morphologic and physiologic validation in vivo of a rapid digital angiographic method. Circulation 1987;75:452– 60. 13. Carrozza JP, Jr, Schatz RA, George CJ, et al. Acute and long-term outcome after Palmaz-Schatz stenting: analysis from the New Ap-
1890
14. 15. 16. 17. 18. 19.
20. 21.
22.
23. 24. 25. 26.
27.
28. 29.
30.
31. 32.
Cantor et al. Stent-Like Outcomes With Balloon Angioplasty
proaches to Coronary Intervention (NACI) registry. Am J Cardiol 1997;80:78K– 88K. Kastrati A, Schomig A, Elezi S, et al. Predictive factors of restenosis after coronary stent placement. J Am Coll Cardiol 1997;30:1428 –36. Corcos T, Favereau X, Gue´rin Y. A call for professional stenting: the balloon is dead and buried! Circulation 1998;98:2644 –5. Gershlick AH, Baron J. Dealing with in-stent restenosis. Heart 1998;79:319 –23. Mintz GS, Hoffmann R, Mehran R, et al. In-stent restenosis: the Washington Hospital Center experience. Am J Cardiol 1998;81:7– 13E. Mehran R, Dangas G, Mintz GS, et al. In-Stent Restenosis: “The great equalizer”— disappointing clinical outcomes with ALL interventional strategies (abstr). J Am Coll Cardiol 1999;33:63–A. Foley DP, Serruys PW. Provisional stenting—stent-like balloon angioplasty: evidence to define the continuing role of balloon angioplasty for percutaneous coronary revascularization. Semin Interv Cardiol 1996;1:269 –73. Carere RG, Barbeau G, Dzavik V, et al. Do superior balloon angioplasty results provide stent-like outcomes in coronary occlusions (abstr)? Circulation 1998;98:I284. Shah V, Haude M, Erbel R, et al. Long-term follow-up of postPTCA results (ⱕ 30% residual diamter stenosis) in the restenotic lesions: results of the REST trial (abstr). J Am Coll Cardiol 1997;29: 76A. Holmes DR, Kip KE, Yeh W, Kelsey SF, Detre KM, Williams DO. Long-term analysis of conventional balloon angioplasty and an initial “stent-like” result: the NHLBI PTCA Registry. J Am Coll Cardiol 1998;32:590 –5. Espinola-Klein C, Rupprecht H-J, Trautmann S, et al. Ten year follow-up of patients with optimal “stent-like,” suboptimal or failed coronary angioplasty (abstr). Eur Heart J 1997;18:24. Weaver WD, Reisman MA, Griffin JJ, et al. Optimum percutaneous transluminal coronary angioplasty compared with routine stent strategy trial (OPUS-1): a randomised trial. Lancet 2000;355:2199 –203. Dangas G, Ambrose JA, Rehman D, et al. Balloon Optimization versus Stent Study (BOSS): provisional stenting and early recoil after balloon angioplasty. Am J Cardiol 2000;85:957– 61. Rodrı´guez A, Ayala F, Bernardi V, et al. Optimal Coronary Balloon Angioplasty with provisional Stenting versus primary stent (OCBAS): immediate and long-term follow-up results. J Am Coll Cardiol 1998;32:1351–7. Serruys PW, di Mario C, Piek J, et al. Prognostic value of intracoronary flow velocity and diameter stenosis in assessing the short- and long-term outcomes of coronary balloon angioplasty: the DEBATE study (Doppler Endpoints Balloon Angioplasty Trial Europe). Circulation 1997;96:3369 –77. Serruys PW, de Bruyne B, Carlier S, et al. Randomized comparison of primary stenting and provisional balloon angioplasty guided by flow velocity measurement. Circulation 2000;102:2930 –7. Lafont A, Dubois-Rande´ JL, Steg PG, et al. The French randomized optimal stenting trial: a prospective evaluation of provisional stenting guided by coronary velocity reserve and quantitative coronary angiography. J Am Coll Cardiol 2000;36:404 –9. Di Mario C, Moses JW, Anderson TJ, et al. Randomized comparison of elective stent implantation and coronary balloon angioplasty guided by online quantitative angiography and intracoronary Doppler. Circulation 2000;102:2938 – 44. Bech GJW, Pijls NHJ, de Bruyne B, et al. Usefulness of fractional flow reserve to predict clinical outcome after balloon angioplasty. Circulation 1999;99:883– 8. Kong DF, Hasselblad V, Tcheng JE, Ohman EM, Topol EJ, Califf RM. Clinical outcome improvements from coronary stenting: a systemic overview (abstr). Eur Heart J 1998;19:571.
JACC Vol. 37, No. 7, 2001 June 1, 2001:1883–90 33. The ESPRIT Investigators. The ESPRIT study: a randomised, placebo-controlled trial of a novel dosing regimen of eptifibatide in planned coronary stent implantation. Lancet 2000;356:2037– 44. 34. Topol EJ, Mark DB, Lincoff AM, et al. Outcomes at 1 year and economic implications of platelet glycoprotein IIb/IIIa blockade in patients undergoing coronary stenting: results from a multicentre randomised trial. Lancet 1999;354:2019 –24. 35. Cohen DJ, Breall JA, Ho KK, et al. Economics of elective coronary revascularization. Comparison of costs and charges for conventional angioplasty, directional atherectomy, stenting and bypass surgery. J Am Coll Cardiol 1993;22:1052–9. 36. Cohen DJ, Krumholz HM, Sukin CA, et al. In-hospital and one-year economic outcomes after coronary stenting or balloon angioplasty: results from a randomized clinical trial. Circulation 1995;92:2480 –7. 37. Dick RJ, Popma JJ, Muller DWM, Burek KA, Topol EJ. In-hospital costs associated with new percutaneous coronary devices. Am J Cardiol 1991;68:879 – 85. 38. Hirshfeld JW, Jr, Schwartz JS, Jugo R, et al. Restenosis after coronary angioplasty: a multivariate statistical model to relate lesion and procedure variables to restenosis. The M-HEART Investigators. J Am Coll Cardiol 1991;18:647–56. 39. Rensing BJ, Hermans WR, Vos J, et al. Luminal narrowing after percutaneous transluminal coronary angioplasty. A study of clinical, procedural, and lesional factors related to long-term angiographic outcome. Coronary Artery Restenosis Prevention on Repeated Thromboxane Antagonism (CARPORT) Study Group. Circulation 1993;88:975– 85. 40. Bourassa MG, Lesperance J, Eastwood C, et al. Clinical, physiologic, anatomic and procedural factors predictive of restenosis after percutaneous transluminal coronary angioplasty. J Am Coll Cardiol 1991;18: 368 –76. 41. Halon DA, Merdler A, Shefer A, Flugelman MY, Lewis BS. Identifying patients at high risk for restenosis after percutaneous transluminal coronary angioplasty for unstable angina pectoris. Am J Cardiol 1989;64:289 –93. 42. Leimgruber PP, Roubin GS, Hollman J, et al. Restenosis after successful coronary angioplasty in patients with single-vessel disease. Circulation 1986;73:710 –7. 43. Weintraub WS, Kosinski AS, Brown CL, King SB. Can restenosis after coronary angioplasty be predicted from clinical variables? J Am Coll Cardiol 1993;21:6 –14. 44. Holmes DR, Jr, Vlietstra RE, Smith HC, et al. Restenosis after percutaneous transluminal coronary angioplasty (PTCA): a report from the PTCA Registry of the National Heart, Lung, and Blood Institute. Am J Cardiol 1984;53:77C– 81C. 45. Vandormael MG, Deligonul U, Kern MJ, et al. Multilesion coronary angioplasty: clinical and angiographic follow-up. J Am Coll Cardiol 1987;10:246 –52. 46. Califf RM, Fortin DF, Frid DJ, et al. Restenosis after coronary angioplasty: an overview. J Am Coll Cardiol 1991;17:2B–13B. 47. Ellis SG, Roubin GS, King SB, III, et al. Angiographic and clinical predictors of acute closure after native vessel coronary angioplasty. Circulation 1988;77:372–9. 48. Cantor WJ, Kramer JM, Miller JM, et al. Higher mortality in women following angioplasty explained by differences in body size, not vessel size (abstr). Circulation 1999;100:I780. 49. MacDonald RG, Henderson MA, Hirshfeld JW, Jr, et al. Patientrelated variables and restenosis after percutaneous transluminal coronary angioplasty—a report from the M-HEART Group. Am J Cardiol 1990;66:926 –31. 50. Miller JM, Ohman EM, Moliterno DJ, Califf RM. Restenosis: the clinical issues. In: Topol EJ, ed. Textbook of Interventional Cardiology. Philadelphia, PA: W.B. Saunders, 1999:393.