SES vs Brachytherapy for Bare Metal In-stent Restenosis
A total of 384 patients were enrolled into the study from February 12, 2003, through July 27, 2004, at 26 centers and were randomized to treatment with SES (n = 259) or VBT (n = 125). Clinical follow-up compliance at 5 years (1,800 days) was similar in both groups at 90.3% for SES (234/259) and 89.6% for VBT (112/125).
Baseline demographics and clinical characteristics showed that patients in the VBT group had a mean age of 63.5 years, 65.6% (82/125) were men, 29.6% (37/125) had diabetes, and 83.2% (104/125) had hypertension. In the SES group, the mean age was 62.7% years, 67.9% (176/259) were men, 33.2% (86/259) had diabetes, and 80.3% (208/259) had hypertension (Table I).
Using the Kaplan-Meier analysis, the survival free for target lesion revascularization was 72.9% vs 66.4% (log-rank P = .08) (Figure 1) and 64.4% vs 61.3% (log-rank P = .349) (Figure 2) for TVF. It must be reemphasized that the trial was neither designed nor powered to examine safety and efficacy end points beyond the 9-month primary end point period and at 5 years, the absolute rates of TLR narrowed. Although TLR with DES was still improved, the differences were nonsignificant, with SES group having a TLR rate of 24.7% compared with 31.2% for VBT (95% CI −16.3% to2.8%, P = .179) (Table II). Similarly, TVF and overall MACE rate at 5 years remained improved with SES, but the differences were not statistically significant. Target vessel failure defined as cardiac death, MI, or target vessel revascularization at 5 years postprocedure was 33.6% vs 36.8% (95% CI −13.5% to 6.7%, P = .568) for SES compared with VBT. The overall MACE rate at 5 years postprocedure was 34.0% vs 36.8% (95% CI −13.1% to 7.1%, P = .648) for SES compared with VBT (Table II).
(Enlarge Image)
Figure 1.
Kaplan-Meier curves and survival table for the end point of TLR to 5 years.
(Enlarge Image)
Figure 2.
Kaplan-Meier curves and survival table for the end point of TVF to 5 years.
Although examination of the MACE rates revealed no statistically significant differences in the rates of overall death or cardiac death at 1,800 days (Table II), there were, however, differences in the rates of Q-wave and non–Q-wave MI at 5 years. The rates of MI at 5 years between SES and VBT were 10.8% vs 3.2% (95% CI 1.9% to 12.4%, P = .010), respectively; this difference was seen in the rates of Q-wave and non–Q-wave MI between the 2 groups at 3.5% for SES and 0% for VBT (95% CI 0.1% to 6.5%, P = .034) and at 8.9% vs 3.2% (95% CI, 0.1% to 10.2%, P = .054) for Q-wave MI vs non–Q-wave MI. This difference between groups is present for both target vessel and nontarget vessel MI (Table III).
A closer examination of the restenosis patterns in the 2 groups revealed that most (65%) had type 1 (focal) lesions, followed by type 3 (proliferative; 15%); whereas type 2 (diffuse) and type 4 (total occlusion) were 12% and 8%, respectively. A third (31.2%) of patients in the VBT group underwent TLR, which was mainly PTCA with insertion of a DES (87%); whereas the remainder had coronary artery bypass graft (CABG). On the other hand, 24.7% of patients in the SES group underwent TLR, which was mainly PTCA with insertion of a DES (91% of cases); whereas the remainder had CABG. Based on the available data, it is difficult to tease out if patients with brachytherapy had a similar or different pattern of restenosis.
Overall, 9.1% (35/384) of patients had multiple TLR at 5 years, higher in patients undergoing VBT (13.6%; 17/125) compared with patients undergoing DES placement (6.9%; 18/259). The univariate and multivariate predictors of repeat TLR in all patients were increasing age, the presence of preprocedure thrombus, and increasing lesion complexity based on American College of Cardiology/American Heart Association lesion classification.
Although not the primary focus of this study, an examination of differences in outcomes (9-month TVF) between β vs γ forms of radiation was examined. Of note, most patients received β-radiation, and the rates of MACE were higher in this group (39% vs 21%) compared with patients receiving γ-radiation. The numbers are quite small to make a definitive conclusion.
Stent thrombosis did not differ between the 2 groups at 1 year: 1.5% (4/259) of patients had late-stent thrombosis in the SES group, whereas 0.8% (1/125) of patients had late-stent thrombosis in the VBT group (P = 1.0). The rate of very late thrombosis was different but not statistically significant between the 2 groups, 4.2% for SES and 2.4% for VBT (P = .562). There were no significant differences in the rates of definite/probable stent thrombosis (5.9% vs 2.5%) between the 2 groups (Figure 3). The overall rates of "Definite or Probable ARC stent thrombosis" at 1,800 days was 5.4% (14/259) for SES and 2.4% (3/125) for VBT (P = .289) (Table IV).
(Enlarge Image)
Figure 3.
Cumulative incidence of definite/probable stent thrombosis to 5 years.
At the time of initiation of the SISR study, only 6 (VBT) and 3 (SES) months of dual-antiplatelet therapy were recommended. Most patients in the study, however, continued dual-antiplatelet therapy for a longer period. At 5 years, 54.9% (53.9% SES and 56.9% VBT) of patients were on dual-antiplatelet therapy (Figure 4); 40% received either aspirin or a thienopyridine agent as single-antiplatelet therapy, whereas 23.7% of patients were not on any form of antiplatelet therapy. The rates of any major bleeding complications at 5 years was low in both groups, 1.9% (5/259) for SES and 0.8% (1/125) for VBT.
(Enlarge Image)
Figure 4.
Dual-antiplatelet therapy use to 5 years.
Results
A total of 384 patients were enrolled into the study from February 12, 2003, through July 27, 2004, at 26 centers and were randomized to treatment with SES (n = 259) or VBT (n = 125). Clinical follow-up compliance at 5 years (1,800 days) was similar in both groups at 90.3% for SES (234/259) and 89.6% for VBT (112/125).
Baseline demographics and clinical characteristics showed that patients in the VBT group had a mean age of 63.5 years, 65.6% (82/125) were men, 29.6% (37/125) had diabetes, and 83.2% (104/125) had hypertension. In the SES group, the mean age was 62.7% years, 67.9% (176/259) were men, 33.2% (86/259) had diabetes, and 80.3% (208/259) had hypertension (Table I).
Using the Kaplan-Meier analysis, the survival free for target lesion revascularization was 72.9% vs 66.4% (log-rank P = .08) (Figure 1) and 64.4% vs 61.3% (log-rank P = .349) (Figure 2) for TVF. It must be reemphasized that the trial was neither designed nor powered to examine safety and efficacy end points beyond the 9-month primary end point period and at 5 years, the absolute rates of TLR narrowed. Although TLR with DES was still improved, the differences were nonsignificant, with SES group having a TLR rate of 24.7% compared with 31.2% for VBT (95% CI −16.3% to2.8%, P = .179) (Table II). Similarly, TVF and overall MACE rate at 5 years remained improved with SES, but the differences were not statistically significant. Target vessel failure defined as cardiac death, MI, or target vessel revascularization at 5 years postprocedure was 33.6% vs 36.8% (95% CI −13.5% to 6.7%, P = .568) for SES compared with VBT. The overall MACE rate at 5 years postprocedure was 34.0% vs 36.8% (95% CI −13.1% to 7.1%, P = .648) for SES compared with VBT (Table II).
(Enlarge Image)
Figure 1.
Kaplan-Meier curves and survival table for the end point of TLR to 5 years.
(Enlarge Image)
Figure 2.
Kaplan-Meier curves and survival table for the end point of TVF to 5 years.
Although examination of the MACE rates revealed no statistically significant differences in the rates of overall death or cardiac death at 1,800 days (Table II), there were, however, differences in the rates of Q-wave and non–Q-wave MI at 5 years. The rates of MI at 5 years between SES and VBT were 10.8% vs 3.2% (95% CI 1.9% to 12.4%, P = .010), respectively; this difference was seen in the rates of Q-wave and non–Q-wave MI between the 2 groups at 3.5% for SES and 0% for VBT (95% CI 0.1% to 6.5%, P = .034) and at 8.9% vs 3.2% (95% CI, 0.1% to 10.2%, P = .054) for Q-wave MI vs non–Q-wave MI. This difference between groups is present for both target vessel and nontarget vessel MI (Table III).
A closer examination of the restenosis patterns in the 2 groups revealed that most (65%) had type 1 (focal) lesions, followed by type 3 (proliferative; 15%); whereas type 2 (diffuse) and type 4 (total occlusion) were 12% and 8%, respectively. A third (31.2%) of patients in the VBT group underwent TLR, which was mainly PTCA with insertion of a DES (87%); whereas the remainder had coronary artery bypass graft (CABG). On the other hand, 24.7% of patients in the SES group underwent TLR, which was mainly PTCA with insertion of a DES (91% of cases); whereas the remainder had CABG. Based on the available data, it is difficult to tease out if patients with brachytherapy had a similar or different pattern of restenosis.
Overall, 9.1% (35/384) of patients had multiple TLR at 5 years, higher in patients undergoing VBT (13.6%; 17/125) compared with patients undergoing DES placement (6.9%; 18/259). The univariate and multivariate predictors of repeat TLR in all patients were increasing age, the presence of preprocedure thrombus, and increasing lesion complexity based on American College of Cardiology/American Heart Association lesion classification.
Although not the primary focus of this study, an examination of differences in outcomes (9-month TVF) between β vs γ forms of radiation was examined. Of note, most patients received β-radiation, and the rates of MACE were higher in this group (39% vs 21%) compared with patients receiving γ-radiation. The numbers are quite small to make a definitive conclusion.
Stent thrombosis did not differ between the 2 groups at 1 year: 1.5% (4/259) of patients had late-stent thrombosis in the SES group, whereas 0.8% (1/125) of patients had late-stent thrombosis in the VBT group (P = 1.0). The rate of very late thrombosis was different but not statistically significant between the 2 groups, 4.2% for SES and 2.4% for VBT (P = .562). There were no significant differences in the rates of definite/probable stent thrombosis (5.9% vs 2.5%) between the 2 groups (Figure 3). The overall rates of "Definite or Probable ARC stent thrombosis" at 1,800 days was 5.4% (14/259) for SES and 2.4% (3/125) for VBT (P = .289) (Table IV).
(Enlarge Image)
Figure 3.
Cumulative incidence of definite/probable stent thrombosis to 5 years.
At the time of initiation of the SISR study, only 6 (VBT) and 3 (SES) months of dual-antiplatelet therapy were recommended. Most patients in the study, however, continued dual-antiplatelet therapy for a longer period. At 5 years, 54.9% (53.9% SES and 56.9% VBT) of patients were on dual-antiplatelet therapy (Figure 4); 40% received either aspirin or a thienopyridine agent as single-antiplatelet therapy, whereas 23.7% of patients were not on any form of antiplatelet therapy. The rates of any major bleeding complications at 5 years was low in both groups, 1.9% (5/259) for SES and 0.8% (1/125) for VBT.
(Enlarge Image)
Figure 4.
Dual-antiplatelet therapy use to 5 years.
Source...