There is currently no established standard treatment for in-stent restenosis (ISR) after the implantation of a drug-eluting stent (DES). The aim of this study was to investigate the efficacy of DES versus balloon angioplasty (BA) for the treatment of DES ISR in a multicenter registry cohort. After matching propensity scores of 805 patients with DES ISR treated with either DES (n = 422) or BA (n = 383), 268 matched pairs were selected and analyzed for major adverse cardiac events, a composite of death, myocardial infarction, and target-vessel revascularization, as the primary end point. Baseline clinical and lesion characteristics of the matched pairs were similar. Survival free of major adverse cardiac events at 2 years was higher with DES compared to BA (88.9% vs 78.7%, p <0.001), mainly because of higher TVR-free survival (92.4% vs 81.0%, p <0.001). Among various baseline variables, BA (hazard ratio 2.546, 95% confidence interval 1.412 to 4.593, p = 0.002) was the most important independent risk factor for recurrent target vessel revascularization, followed by acute coronary syndromes as the clinical presentation of DES ISR, and previous implantation of a sirolimus-eluting stent. Survival free of death, myocardial infarction, or stent thrombosis did not differ between the 2 groups. Whereas there was no significant difference in survival free of target vessel revascularization between DES and BA for focal ISR lesions, DES was superior to BA in diffuse ISR lesions (94.3% vs 75.2% at 2 years, p <0.001). In conclusion, compared to BA, the implantation of DES was safe and more effective in the treatment of DES ISR.
Implantation of drug-eluting stents (DES) has significantly reduced the incidence of restenosis and the need for repeat revascularization compared to bare-metal stents. However, restenosis still occurs after implantation of DES. In high-risk patients and for complex lesions, the restenosis rate of DES has been reported to be >20%. With wider use of DES, clinicians encounter an increasing number of patients with DES-related in-stent restenosis (ISR) in daily practice. However, to date, there is no standard therapy established for ISR developing after the implantation of DES. Several studies have reported the outcomes of various treatment modalities for DES ISR. In general, previous studies observed relatively higher rates of repeat revascularization after the implantation of DES in DES ISR lesions compared to de novo lesions. However, most of the studies had either relatively small study populations or were focused on ISR lesions after the implantation of specific DES types. Therefore, the purpose of this study was to investigate the clinical outcomes of DES reimplantation compared to balloon angioplasty (BA) in a relatively large patient cohort of DES ISR.
Methods
In this retrospective, multicenter registry study, 832 patients with 937 DES ISR lesions from 20 Korean percutaneous coronary intervention (PCI) centers who underwent repeat PCI for DES ISR lesions from January 2003 to June 2008 were enrolled. After the exclusion of 20 patients with 40 lesions treated with both DES and BA and 7 patients with 7 DES ISR lesions treated with the implantation of bare-metal stents, 422 patients with 457 lesions treated with DES and 383 patients with 433 lesions treated with BA were investigated. There were no exclusion criteria except for a life expectancy <1 year because of co-morbidities. Between the 2 patient groups treated with either DES or BA, there were significant differences in the frequency of acute coronary syndromes as the clinical presentation of DES ISR, previous chronic total occlusion, bifurcation, and infarct-related lesions. Therefore, we performed propensity score matching to adjust significant differences in characteristics of the patient groups. After matching propensity scores of 422 patients treated with DES and 383 patients treated with BA, 268 matched pairs were selected for the comparison analysis. Written informed consent for PCI was obtained from all patients. The institutional review board at each of the participating centers approved the study protocol and waived requirements for informed consent for this retrospective analysis.
PCI was performed using standard techniques via the femoral or radial approach. All patients were treated with aspirin 300 mg and a 300- to 600-mg loading dose of clopidogrel before PCI if not already receiving maintenance doses. Generally patients received lifelong aspirin 100 mg/day and clopidogrel 75 mg/day for ≥12 months. During the intervention, all patients received unfractionated heparin 100 to 140 IU/kg. The choice of the PCI method to treat DES ISR, predilation of the lesion, and the use of platelet glycoprotein IIb/IIIa inhibitors were at the discretion of the operator.
Baseline clinical and angiographic data, including age, gender, traditional coronary risk factors, medical history, and clinical presentation, were collected for all patients. Follow-up included visits to the outpatient clinic at 30 days and 3 months after the procedure and every 3 to 6 months thereafter. Patients lost to regular clinical follow-up were contacted by telephone and asked about their clinical status. Angiographic follow-up was clinically driven or scheduled at the operator’s discretion.
DES ISR was defined as a luminal diameter stenosis >50% by quantitative coronary angiography located within a DES or within 5 mm of the stent edges. Focal ISR pattern was defined as ≤10 mm in length and positioned at the body of the stent, the proximal or distal margin, or a combination of these sites, whereas diffuse ISR pattern was defined as >10 mm in length.
We defined the primary end point as major adverse cardiac events, a composite of all-cause death, nonfatal myocardial infarction (MI), and target vessel revascularization (TVR) compared between DES and BA for the treatment of DES ISR. MI was defined as an elevation in cardiac troponin above the 99th percentile of the upper reference limit or an elevation in creatine kinase-MB ≥2 times the upper normal value in combination with ≥1 of the following criteria: ischemic symptoms, new electrocardiographic changes indicative of new ischemia (ST-T changes or left bundle branch block), development of pathologic Q waves on electrocardiography, or imaging evidence of new regional wall motion abnormality. TVR included any percutaneous or surgical revascularization of the stented epicardial vessel. Target lesion revascularization (TLR) was defined as either PCI or coronary artery bypass grafting surgery because of restenosis or stent thrombosis of the target lesion that included the proximal and distal edge segments and the ostia of side branches. Stent thrombosis was classified as definite, probable, or possible according to definitions proposed by the Academic Research Consortium. Subgroup analyses included comparisons of outcomes between sirolimus-eluting stents (SES) and paclitaxel-eluting stents (PES) and between same versus different DES for the patients treated with reimplantation of DES and between plain balloons versus cutting balloons for the patients treated with BA. Same or different DES refers to the reimplanted DES type that was same or different as the treated restenotic DES type.
We performed propensity score matching to reduce treatment bias and potential confounding factors. We estimated the propensity score by fitting a logistic regression model using the following variables for treatment with DES versus BA: gender, age, diabetes mellitus, hypertension, hypercholesterolemia, acute coronary syndromes, current smoking, chronic renal failure, heart failure, multivessel disease, previous restenotic lesion, previous bifurcation lesion, previous infarct-related lesion, previous DES diameter and length, and ISR pattern (focal or diffuse). Using propensity scoring, patients treated with DES were matched 1 to 1 with patients treated with BA with similar baseline characteristics. The model discrimination and calibration were assessed by means of C statistics and Hosmer-Lemeshow statistic. The goodness-of-fit statistic was 8.75 (p = 0.364) and the C statistic for the model was 0.75. Continuous variables are reported as mean ± SD and categorical variables as frequencies and percentages. Continuous variables were compared using the unpaired Student’s t test for normally distributed values; otherwise, the Mann-Whitney U test was used. Categorical variables were compared using the chi-square test or Fisher’s exact test when the expected frequency was <5. Event-free survival was estimated using Kaplan-Meier analysis and compared using the log-rank test. Cox proportional-hazards analysis was used to assess the influence of relevant baseline variables on the risk for TVR. Multivariate Cox analysis was used to determine independent risk factors for TVR. All reported p values are 2 sided, and p values <0.05 were considered statistically significant. Statistical analyses were performed using SAS version 9.2 (SAS Institute Inc., Cary, North Carolina) and SPSS version 18.0 (SPSS, Inc., Chicago, Illinois).
Results
Baseline clinical characteristics and procedural data of the study population before and after propensity score matching are listed in Tables 1 and 2 , respectively. The baseline clinical and procedural data between the patient groups treated with DES and with BA were similar after matching. Among 268 patients who underwent repeated DES implantation for DES ISR, 164 (61.2%) were treated with SES, 68 (25.4%) with PES, and 36 (13.4%) with other DES such as zotarolimus-eluting stents (n = 20) and everolimus-eluting stents (n = 12). Among 268 patients treated with BA for DES ISR, 163 patients (60.8%) were treated with plain balloons and 105 (39.2%) with cutting balloons.
| Variable | Before Match (n = 805) | After Match (n = 536) | ||||
|---|---|---|---|---|---|---|
| DES | BA | DES | BA | p Value | ||
| (n = 422) | (n = 383) | p Value | (n = 268) | (n = 268) | ||
| Age (years) | 62.0 ± 10.8 | 62.8 ± 9.6 | 0.264 | 62.4 ± 10.7 | 63.2 ± 9.7 | 0.387 |
| Men | 66.8% | 70.2% | 0.298 | 66.0% | 70.1% | 0.308 |
| Diabetes mellitus | 39.8% | 39.4% | 0.911 | 41.0% | 38.1% | 0.480 |
| Insulin requiring | 7.3% | 6.3% | 0.544 | 6.3% | 5.2% | 0.579 |
| Hypertension | 61.4% | 56.1% | 0.131 | 60.1% | 54.9% | 0.221 |
| Hypercholesterolemia ⁎ | 39.8% | 35.0% | 0.158 | 39.6% | 34.0% | 0.179 |
| Renal failure † | 7.1% | 7.3% | 0.912 | 8.2% | 7.1% | 0.626 |
| Hemodialysis | 1.4% | 2.6% | 0.227 | 1.9% | 3.0% | 0.400 |
| Current smoker | 23.7% | 22.2% | 0.613 | 22.4% | 23.5% | 0.758 |
| Previous stroke | 6.6% | 5.2% | 0.398 | 6.3% | 3.7% | 0.167 |
| Previous coronary bypass | 0.7% | 1.8% | 0.153 | 0.7% | 2.2% | 0.154 |
| Clinical presentation ‡ | ||||||
| Acute coronary syndromes | 29.6% | 21.7% | 0.010 | 26.5% | 23.9% | 0.489 |
| Acute MI | 9.5% | 8.4% | 0.577 | 10.8% | 7.8% | 0.224 |
| Stent thrombosis | 4.7% | 5.7% | 0.522 | 4.5% | 4.5% | 1.000 |
| Left ventricular ejection fraction <45% | 8.8% | 8.6% | 0.939 | 10.1% | 7.5% | 0.285 |
| Multivessel disease | 44.8% | 51.7% | 0.050 | 51.5% | 48.5% | 0.490 |
| Left main disease | 2.1% | 2.9% | 0.501 | 2.6% | 3.4% | 0.612 |
⁎ Total cholesterol level >200 mg/dl or previous use of statins.
† Serum creatinine level >1.5 mg/dl.
‡ Clinical presentation at the time of index PCI for DES ISR.
| Variable | Before Match (n = 805) | After Match (n = 536) | ||||
|---|---|---|---|---|---|---|
| DES | BA | p Value | DES | BA | p Value | |
| Lesion number | 457 | 433 | 289 | 287 | ||
| In-stent restenosis pattern | 0.630 | 0.876 | ||||
| Focal | 54.3% | 52.7% | 57.4% | 56.8% | ||
| Diffuse | 45.7% | 47.3% | 42.6% | 43.2% | ||
| Previous chronic total occlusion | 10.8% | 10.2% | 0.750 | 11.1% | 10.5% | 0.810 |
| Previous bifurcation lesion | 7.2% | 15.2% | <0.001 | 10.7% | 10.1% | 0.807 |
| Previous infarct-related lesion | 19.5% | 33.5% | <0.001 | 27.0% | 30.0% | 0.429 |
| Previous restenotic lesion | 7.5% | 1.8% | <0.001 | 2.1% | 31.0%) | 0.504 |
| Previous DES diameter (mm) | 3.1 ± 0.4 | 3.0 ± 0.3 | <0.001 | 3.1 ± 0.3 | 3.1 ± 0.3 | 0.749 |
| Previous DES length (mm) | 25.2 ± 6.4 | 27.3 ± 7.2 | <0.001 | 25.9 ± 6.4 | 26.4 ± 6.4 | 0.359 |
| Previous DES type | 0.235 | 0.284 | ||||
| SES | 42.1% | 38.1% | 45.7% | 40.8% | ||
| PES | 38.9% | 44.6% | 40.5% | 47.0% | ||
| Zotarolimus-eluting stent | 19.0% | 17.3% | 13.8% | 12.2% | ||
| DES used for ISR treatment | ||||||
| SES | 62.8% | 61.2% | ||||
| PES | 20.8% | 26.0% | ||||
| Zotarolimus-eluting stent | 11.6% | 6.9% | ||||
| Everolimus-eluting stent | 4.8% | 4.5% | ||||
| Use of cutting balloon | 45.7% | 40.0% | ||||
Patients were clinically followed for a median duration of 937 days (range 1 to 2,335). Among all patients, 94.8% (n = 508) completed 1-year clinical follow-up, and 25.7% (n = 138) underwent follow-up coronary angiography within 1 year. The cumulative rate of survival free of major adverse cardiac events was significantly higher in the DES group than in the BA group (94.6% vs 87.8% at 1 year and 88.9% vs 78.7% at 2 years, log-rank p = 0.001). This was due mainly to higher TVR-free (96.9% vs 88.1% at 1 year and 92.4% vs 81.0% at 2 years, log-rank p <0.001) and TLR-free survival rates (96.9% vs 88.5% at 1 year and 92.4% vs 81.8%, log-rank p = 0.001) in the DES group than in the BA group. The rates of survival free of death, MI, or stent thrombosis did not differ between the 2 patient groups ( Figure 1 ) .
When the patient subgroup with nonfocal pattern of DES ISR (n = 227) was evaluated separately, the TVR-free survival rates (97.1% vs 84.1% at 1 year and 94.3% vs 75.2% at 2 years, log-rank p <0.001) were higher in the DES group (n = 111) than in the BA group (n = 116) ( Figure 2 ). However, in the patient subgroup with focal ISR pattern (n = 309), there was no significant difference in the TVR-free survival rates (96.7% vs 91.2% at 1 year and 91.1% vs 85.4% at 2 years, log-rank p = 0.292) between DES (n = 157) and BA (n = 152) ( Figure 2 ).
