Long-Term Outcome of Patients Treated With Repeat Percutaneous Coronary Intervention After Failure of γ-Brachytherapy for the Treatment of In-Stent Restenosis
Background— Although 192Ir intracoronary brachytherapy has been demonstrated to dramatically reduce the recurrence of in-stent restenosis, up to 24% of these patients will still require repeat target-vessel revascularization. The short- and long-term outcomes of repeat percutaneous intervention in this population have not been characterized.
Methods and Results— Analysis was performed of all patients enrolled in the GAMMA-I and GAMMA-II brachytherapy trials who underwent repeat percutaneous target lesion revascularization (TLR) because of restenosis. Subjects were divided into 2 cohorts: those who had received 192Ir brachytherapy and those randomized to placebo. Forty-five (17.6%) of a total of 256 patients whose index treatment was intracoronary radiation therapy and 36 (29.8%) of 121 patients whose index treatment was placebo required repeat percutaneous TLR. The mean time to this first TLR was 295±206 days in the irradiated group and 202±167 days in the placebo group (P=0.03). Acute procedural success occurred in 100% of irradiated patients and 94% of placebo controls (P=0.19). After the first TLR, a subsequent TLR was required in 15 (33.3%) of 45 brachytherapy patients versus 17 (47.2%) of 36 placebo failure patients (P=0.26). There was no significant difference in time to second TLR between the 2 groups. Other long-term major adverse event rates in both groups were comparable to those of other contemporary angioplasty/stenting series.
Conclusions— In those patients who “fail” 192Ir intracoronary brachytherapy for in-stent restenosis, treatment with 192Ir delays the time to first TLR. Additionally, repeat percutaneous intervention in these patients is safe and efficacious in the short term, with acceptable long-term results.
Received May 3, 2002; revision received July 29, 2002; accepted August 21, 2002.
Repeat coronary angioplasty for patients with in-stent restenosis is associated with subsequent restenosis rates of ≈50%.1–4⇓⇓⇓ Diffuse and occlusive patterns of in-stent restenosis recur in an especially aggressive manner, with 1-year target-vessel revascularization (TLR) rates of up to 83%.5Although debulking strategies and repeat stenting have been used as an adjunct to PTCA for the treatment of in-stent restenosis, none have been proven to lower the rate of recurrent angiographic or clinical restenosis compared with PTCA alone.6–9⇓⇓⇓
Four randomized clinical trials of patients with in-stent restenosis have demonstrated a reduction in recurrent restenosis by at least 50% with the use of 192Ir γ-radiation compared with nonradiation PTCA or plaque-debulking therapies.10–13⇓⇓⇓ Although these favorable results were tempered by the occurrence of late stent thrombosis in ≈5% of patients, optimization of antiplatelet regimens and avoidance of additional stent implantation have largely overcome this complication.14
Despite the efficacy of 192Ir in the treatment of restenosis, up to 24% of patients will “fail” intracoronary brachytherapy and require repeat TLR for restenosis.10–12⇓⇓ The choice of which revascularization technique to use is not well defined; although CABG may represent a more definitive choice, repeat PTCA has significant appeal in terms of reduced morbidity, mortality, and length of hospitalization.
The primary objective of this analysis was to examine the clinical outcome of patients who required percutaneous TLR for recurrent in-stent restenosis after receiving 192Ir intracoronary brachytherapy (as the index procedure for in-stent restenosis) compared with control patients who had undergone percutaneous intervention for recurrent in-stent restenosis without brachytherapy.
Analysis was performed of all patients enrolled in the GAMMA-I and GAMMA-II trials who underwent repeat percutaneous TLR because of restenosis. Subjects were divided into 2 cohorts: those who received 192Ir brachytherapy (patients randomized to the active treatment arm of GAMMA-I, and all patients enrolled in the GAMMA-II registry) and placebo (those randomized to placebo in GAMMA-I).
The methods and results of GAMMA-I have been described previously.11 To summarize, GAMMA-I was a randomized, double-blind trial comparing 192Ir versus placebo in patients with myocardial ischemia and in-stent restenosis within a native coronary artery that was 2.75 to 4.0 mm in diameter and ≤45 mm in length. The lesion was treated at the operator’s discretion by conventional interventional techniques, with additional stenting reserved for those cases with residual stenoses >30%, significant dissection, or restenosis that extended beyond the original stent borders.
Immediately after successful coronary intervention, a rapid-exchange, closed-end-lumen, noncentered, 4F radiation catheter (Cordis, a Johnson & Johnson Company) was inserted over the intracoronary guidewire. Then, a 0.76-mm ribbon (Best Industries) that contained randomly assigned sealed sources of either 192Ir or placebo was inserted into the delivery catheter. Ultrasound data and the specific activity of the radiation source were used to determine the time required to deliver 8 Gy to the target farthest from the radiation source, provided no more than 30 Gy was delivered to the target closest to the source. Of 252 eligible patients, 131 were randomly assigned to receive 192Ir and 121 were assigned to placebo.
GAMMA-II enrolled 125 patients, all of whom were assigned active treatment with a slightly higher fixed dose (14 Gy at 2 mm) of 192Ir.15 Inclusion criteria and treatment methods were otherwise identical to those described for GAMMA-I.
In both trials, patients underwent a 6-month protocol-mandated angiogram. Patients returning with evidence of recurrent myocardial ischemia at other times underwent coronary angiography at the operator’s discretion. TLR or vessel revascularization either by percutaneous intervention or CABG was also performed at the operator’s discretion.
Both trials complied with the Declaration of Helsinki regarding investigation in humans, and all investigational sites received approval from their local hospital institutional review boards.
Data Collection and Core Laboratory Analysis
Detailed case report forms were completed by the clinical coordinators at each site, monitored by independent study monitors, and submitted to the data-coordinating center (Cardiovascular Data Analysis Center, Harvard Clinical Research Institute, Boston, Mass.). Clinical follow-up was mandated at 1, 6, and 9 months, and telephone contacts were mandated at 2, 24, 36, 48, and 60 months. An independent Clinical Events Committee classified all events. Patients adjudicated as having repeat TLR because of stent thrombosis (at any time after the index procedure) were excluded from this analysis.
Angiograms were obtained during the procedure, at 6-month follow-up, and at the time of repeat TLR(s) and were submitted to the angiographic core laboratory (Cardiovascular Research Foundation, Washington Hospital Center, Washington, DC). All angiograms were analyzed with a computer-based edge detection system (Medis). The contrast-filled injection catheter was used as the reference standard to obtain quantitative measures (in millimeters) from the average proximal and distal reference segments and minimal lumen diameter.
The initial intervention with adjunct brachytherapy/placebo was defined as the index procedure. Patients undergoing a repeat percutaneous intervention after the index procedure for restenosis (first TLR) compose the study cohort. Second TLR refers to a subsequent percutaneous intervention for restenosis (if required) in this group of patients. Major adverse cardiac events were a composite of death, Q-wave or non–Q-wave myocardial infarction, emergent CABG, or TLR. Myocardial infarction was classified as follows. Q-wave myocardial infarction was defined as development of new, pathological Q waves in 2 or more contiguous leads (as assessed by the ECG core laboratory) with postprocedural creatine kinase (CK) or CK-MB levels elevated above normal. Non–Q-wave myocardial infarction was defined as elevation of postprocedural CK levels to >2 times normal, with CK-MB elevated above normal in the absence of new, pathological Q waves. Acute procedural success was defined as attainment of <50% residual diameter stenosis and no in-hospital major adverse cardiac events.
Binary variables are presented as percentages, and continuous variables are presented as mean±SD. Binary variables were compared with χ2 analysis or Fisher’s exact test when appropriate, and continuous variables were computed with the Student’s t test. To identify factors that might be related to the occurrence of repeat TLR, logistic regression was used. Kaplan-Meier survival analysis was performed to assess freedom from TLR. A 2-sided probability value <0.05 was considered significant. All statistical analyses were performed with the SAS System (version 6.12, SAS Institute).
Forty-five (17.6%) of a total of 256 patients assigned to intracoronary radiation therapy and 36 (29.8%) of the 121 patients who received placebo required repeat percutaneous TLR after mean follow-up times after the index procedure of 501±285 and 688±307 days, respectively. Total occlusions were present in 9.4% of irradiated patients versus 3.1% of placebo controls (P=0.61).
The 2 groups were well matched in terms of demographics and angiographic characteristics at the time of the index procedure (Tables 1 and 2⇓). Whereas the irradiated cohort had a higher prevalence of left anterior descending coronary artery disease and the placebo group had a higher proportion of patients with angina at rest, the groups were well matched for other variables more strongly associated with TLR risk, such as diabetes and vessel diameter.
The mean time to first TLR was 295±206 days in the irradiated group and 202±167 days in the placebo group (P=0.03). There were no significant differences in treatment modality at the time of the first TLR. Eighty-nine percent of patients in the placebo group underwent balloon angioplasty compared with 98% in the irradiated group. Debulking techniques were used in both groups, with a trend to higher use in the placebo arm: rotablation, directional coronary atherectomy, or excimer laser angioplasty was performed in 19% of patients in the irradiated group and 37% of those assigned to placebo (P=0.07). The rate of stenting was relatively high in both groups, with 66% of irradiated patients and 43% of placebo patients receiving at least 1 additional stent (P=0.07) at first TLR. Overall, acute procedural success occurred in 100% of irradiated patients and 94% of placebo controls (P=0.19).
Of the 45 irradiated patients who required repeat TLR by percutaneous methods, 15 (33.3%) required a second TLR: 8 underwent CABG, and 7 had repeat percutaneous intervention. In the placebo cohort, 17 (47.2%) of 36 patients required a second TLR, 8 by CABG and 9 by PTCA. The difference in rates of second TLR between the 2 cohorts did not reach statistical significance (P=0.26). Kaplan-Meier plots depicting freedom from a second TLR for the 2 cohorts are shown in the Figure and also demonstrate no difference between the 2 groups (P=0.72).
In those who received brachytherapy, mean time to second TLR was 197±146 days compared with 311±261 days in those patients treated with placebo (P=0.14). No univariate predictors of a second TLR were identified by logistic regression analysis (Table 3). Factors such as vessel diameter, lesion length, left anterior descending artery location, and diabetes were not predictive of need for second TLR in this model. Use of brachytherapy and the length of the radiation source did not have an effect on requirement for a second TLR.
The rate of major adverse cardiac events that occurred from the first TLR throughout the follow-up period was relatively high in both groups (42.2% in brachytherapy patients versus 50.0% in placebo patients, P=0.51). However, this was predominantly accounted for by the rates of second TLR (Table 4). One patient in the placebo group had a vascular complication that required surgical repair. There were no other serious complications (cardiac perforation, major bleeding, hematological dyscrasia, or stroke) in either group.
Four randomized clinical trials have demonstrated the superiority of intracoronary 192Ir γ-brachytherapy plus PTCA or stenting over balloon angioplasty or stenting alone for the prevention of recurrent restenosis.10–13⇓⇓⇓ However, despite the dramatic reduction in restenosis in patients undergoing adjunct brachytherapy, up to one quarter will still require repeat TLR.10–12⇓⇓ It is anticipated this rate will be reduced by optimization of antiplatelet regimens (and hence a reduction in stent thrombosis), adjustments to the dose prescription, and more meticulous attention to irradiating the entire balloon-injured segment (to reduce the occurrence of “geographic miss”).16 However, among the 150 000 patients who undergo treatment for in-stent restenosis annually in the United States, there remains a population of patients who will encounter recurrent restenosis despite optimal intracoronary brachytherapy.
Interestingly, when patients fail γ-radiation, the present results indicate the time to first TLR is delayed compared with that for placebo patients. Patients in the failed irradiation cohort had a longer mean time to first TLR (295±206 versus 202±167 days, P=0.03). This suggests γ-brachytherapy has a delaying as well as an inhibitory effect on in-stent restenosis, presumably through inhibition of the cellular division and/or matrix formation in the media and adventitia that are responsible for the late proliferative response to PTCA.17–19⇓⇓ It is also interesting to note that none of the traditionally accepted risk factors for clinical restenosis20 were predictive of need for additional TLR.
Although repeat percutaneous intervention is an appealing treatment strategy, the results after this procedure in patients who fail brachytherapy have not been categorized previously. In the present series, repeat percutaneous intervention after failed γ-brachytherapy was associated with a high (100%) short-term procedural success rate. Furthermore, patients who underwent repeat percutaneous intervention for recurrent restenosis after failed brachytherapy had acceptable long-term clinical outcomes, with one third of patients requiring yet another TLR.
The second TLR rate of 33.3% in the irradiated cohort is relatively low given the aggressively restenotic nature of this patient population. This rate is also acceptable when compared with the rate of first TLR after γ-brachytherapy (23% to 24%) or that in patients treated with percutaneous intervention without brachytherapy for in-stent restenosis (41% to 63%).11,12⇓ Major adverse cardiac events (excluding TLR) in the irradiated group were also comparable to those seen in both brachytherapy/restenosis series and de novo PTCA/stenting trials,6,21–24⇓⇓⇓⇓ especially when the long follow-up period of the present patient population is taken into account.
Notably, previous exposure to irradiation does not appear to convey a “memory” effect. Irradiated patients who required a first TLR subsequently required a second TLR with nearly the same frequency as placebo patients (33.3% versus 47.2%, P=0.26), and by univariate analysis, placebo (versus brachytherapy) as the index procedure was not found to be a predictor of a second TLR.
This is a retrospective analysis and is therefore subject to the limitations of such studies. Also, this series reflects the early experience with 192Ir brachytherapy, and different outcomes could be anticipated with current techniques, which include better coverage of stent margins with the brachytherapy source and much lower rates of additional stent implantation for in-stent restenosis than seen in the GAMMA-I and -II trials. Additionally, contemporary management of recurrent restenosis, with less use of ablative techniques and additional stents and higher utilization of cutting balloon angioplasty, may yield different results. Another limitation is the relatively small sample size. This is in part, however, a reflection of the relatively low number of patients who have recurrent restenosis after 192Ir brachytherapy. Also, the analysis was deliberately confined to patients in the GAMMA-I and -II trials because of their similar design and inclusion criteria and rigorous long-term follow-up.
The present series demonstrates that in patients with recurrent restenosis after 192Ir intracoronary brachytherapy as the index procedure for in-stent restenosis, repeat percutaneous intervention is associated with a high degree of periprocedural safety and efficacy and acceptable rates of long-term major adverse cardiac events, including need for further revascularization. Percutaneous strategies for the treatment of in-stent restenosis continue to be refined. Drug-eluting stent implantation is currently undergoing evaluation for the treatment of in-stent restenosis and may also have a role in the management of patients who have failed brachytherapy for in-stent restenosis. To conclude, in the absence of other indications for CABG, repeat percutaneous intervention should be considered as a first-line therapy for patients re-presenting with symptomatic restenosis after 192Ir brachytherapy.
Drs Moses, Tripuraneni, Jani, and Leon are consultants to Cordis, a Johnson and Johnson Company. Drs Jani and Tripuraneni are also consultants to Best Industries. Dr Reilly has received honoraria for speaking from Cordis. Dr Teirstein serves as a consultant for and receives research grants from several companies working in the field of vascular radiotherapy. He also receives royalties from the sale of radiation-delivery devices. J.-A. Giorgianni is an employee of Cordis.
Guest Editor for this article was Dr Donald S. Baim, Brigham and Women’s Hospital, Boston, Mass.
- ↵Bauters C, Banos JL, Van Belle E, et al. Six-month angiographic outcome after successful repeat percutaneous intervention for in-stent restenosis. Circulation. 1998; 97: 318–321.
- ↵Elezi S, Kastrati A, Neumann FJ, et al. Vessel size and long-term outcome after coronary stent placement. Circulation. 1998; 98: 1875–1880.
- ↵Mehran R, Dangas G, Abizaid AS, et al. Angiographic patterns of in-stent restenosis: classification and implications for long-term outcome. Circulation. 1999; 100: 1872–1878.
- ↵Mehran R, Dangas G, Mintz GS, et al. Treatment of in-stent restenosis with excimer laser coronary angioplasty versus rotational atherectomy: comparative mechanisms and results. Circulation. 2000; 101: 2484–2489.
- ↵Waksman R, White RL, Chan RC, et al. Intracoronary gamma-radiation therapy after angioplasty inhibits recurrence in patients with in-stent restenosis. Circulation. 2000; 101: 2165–2171.
- ↵Teirstein PS, Moses JW, Leon MB, et al. Prolonged antiplatelet therapy and reduced stenting eliminates late thrombosis after radiation: the Scripps III trial. J Am Coll Cardiol. 2002; 39: 65A.
- ↵Wong SC, Teirstein PS, Moses JW, et al. Nine-month clinical outcomes after intravascular radiation therapy for in-stent restenosis: a report from the GAMMA-II registry. Am J Cardiol. 2000; 86 (supp1): 8A.
- ↵Sabate M, Costa MA, Kozuma K, et al. Geographic miss: a cause of treatment failure in radio-oncology applied to intracoronary radiation therapy. Circulation. 2000; 101: 2467–2471.
- ↵Cutlip DE, Chauhan M, Rizzitano C, et al. Predictors of clinical restenosis after coronary stenting. Circulation. 1998; 98: I-2289.
- ↵Popma JJ, Califf RM, Topol EJ. Clinical trials of restenosis after coronary angioplasty. Circulation. 1991; 84: 1426–1436.
- ↵Teirstein PS, Massullo V, Jani S, et al. Three-year clinical and angiographic follow-up after intracoronary radiation: results of a randomized clinical trial. Circulation. 2000; 101: 360–365.