Use of intracoronary radiation to prevent restenosis

INTRODUCTION — Historically, restenosis has been one of the principal limitations to the long-term success of percutaneous coronary revascularization. The rate of angiographic restenosis following balloon angioplasty alone (defined as 50 percent or greater reduction in luminal diameter at follow-up angiography) is 40 to 50 percent [1]. About half of the patients with angiographic restenosis manifest clinical restenosis with recurrent ischemia that leads to repeat revascularization of the targeted vessel [2].

Intracoronary stents, by virtue of their ability to prevent elastic recoil and constrictive remodeling, initially reduced the frequency of angiographic restenosis to approximately 20 percent and the need for repeat revascularization to approximately 10 to 12 percent (figure 1) [3,4]. Even better results have been obtained with drug-eluting stents, which are the stents of choice for most coronary interventions currently (figure 2). (See "Percutaneous coronary intervention with intracoronary stents: Overview".)

Prior to the availability of drug-eluting stents, one major focus of investigation to treat restenosis was the use of intracoronary radiation. Based upon the data presented below, the United States Food and Drug Administration (FDA) has approved both gamma and beta intracoronary radiation for use in the treatment of in-stent restenosis [5]. This technique is applied immediately after mechanical treatment (usually balloon dilatation) of the stent restenosis. However, the increased efficacy of drug-eluting stents for in-stent restenosis has markedly limited the role of intracoronary radiation for this indication.

Data do not support the use of intracoronary radiation for the prevention of restenosis after treating de novo lesions with balloon angioplasty alone, and the FDA has not approved this technique for this indication. (See 'Comparison to drug-eluting stents' below.)

Another radiation approach, insertion of radioactive stents, has been demonstrated to cause an increase in restenosis at the edges of the stent and may delay rather than prevent restenosis within the stent itself. The development of these stents has been discontinued.

GENERAL PRINCIPLES — The rationale of using ionizing radiation to prevent restenosis is derived from the observed benefit of radiation in reducing exuberant wound healing following surgery. In porcine models, application of endovascular radiation inhibited neointimal formation after balloon injury [6,7]. These benefits may be mediated by apoptosis (programmed cell death), inhibition of the first wave of cellular proliferation, and prevention of adventitial fibrosis [7].

Radiation can be delivered over the course of 3 to 30 minutes (depending upon the isotope and its activity) via percutaneously inserted catheters containing radioactive wires, balloons, ribbons, or seeds. Most studies have used gamma radiation from an Ir-192 source, which is highly penetrating in human tissue and falls off according to the inverse square law. Other studies have used shorter range beta particles from a Sr-90, P-32, or I-125 source, which have a range of only a few millimeters in tissue and yield lower whole-body dose exposure for a corresponding dose to the arterial wall. Because of this intense local absorption, beta brachytherapy can be delivered in much shorter dwell times and with much less shielding of the laboratory staff.

Radiation dose — Long-term results after radiotherapy may be influenced by the radiation dose, the homogeneity of dose distribution, the complexity of the stenotic lesion, and vessel injury at the time of radiation delivery [8-11]. In early studies, the effect of increasing the beta-emitter dose was evaluated in a report of 130 patients undergoing percutaneous coronary intervention (PCI) without stenting who were randomly assigned to receive 9, 12, 15, or 18 Gy of radiation [8]. At six months, the restenosis rate was 28, 17, 16, and 4 percent, respectively. The rate of repeat revascularization with a dose of 9 and 18 Gy was 18 and 6 percent, respectively and there was a dose-dependent enlargement of the vessel lumen.

Intracoronary radiation using a fixed dose of 15 Gy is less effective in longer in-stent restenotic lesions, due to greater eccentricity and variability in lesion geometry and longer source-to-target distances [9]. However, efficacy is improved with the use of higher doses (18 Gy at 2 mm from the source) [10].

Scattering after stenting — A concern with the use of radiotherapy after stenting is the potential for absorption and scattering, especially of short-range beta particles, by stent struts that may cause significant variations in the uniformity and magnitude of the radiation dose. Thinner strut stents result in less radiation dose perturbation than previously popular thicker strut stents [12]. However, the clinical significance of these differences is uncertain. As described below, trials using beta-emitters for in-stent restenosis have shown benefit that appears roughly similar to that seen in trials of more penetrating gamma radiation. (See "Intracoronary stent restenosis".)

EFFICACY — Brachytherapy is effective in treating restenosis occurring after either balloon angioplasty or stenting. Some long-term data from clinical trials suggest that radiation may delay, rather than prevent, restenosis [13,14].

Use for in-stent restenosis — Low dose brachytherapy has been approved by the United States Food and Drug Administration (FDA) as adjunctive treatment for in-stent restenosis; it appears to act by inhibiting recurrent exuberant neointimal formation within the stent, without any effect at the stent edge [15].

The benefit and clinical outcomes with intracoronary radiation are independent of the primary method used to treat in-stent restenosis (balloon angioplasty, excimer laser angioplasty, rotational atherectomy, or additional stent implantation) [16,17]. In addition, intracoronary radiation is effective in all vessel sizes, with benefit greatest in small vessel (<2.5 mm) and all lesion lengths, especially in diffuse lesions [18]. (See "Intracoronary stent restenosis".)

Although both beta and gamma radiation have similar benefit in preventing recurrence of in-stent restenosis, most laboratories prefer the beta systems because of their shorter treatment times (typically 3 versus 30 minutes) and the lack of requirement for significant lead shielding to protect the operator and staff.

A significant clinical benefit has been noted with intracoronary radiation in a number of published trials using a gamma emitter (Gamma-One, WRIST, Long WRIST) [19-22] or a beta emitter (INHIBIT, Beta WRIST, and START) [23-25]. The following observations illustrate the range of findings.

●The Gamma-One trial evaluated 252 patients undergoing percutaneous therapy for in-stent stenosis in native coronary arteries; the patients were randomly assigned to a gamma emitting or nonradioactive intracoronary ribbon [19]. At nine months, the primary end point (death, myocardial infarction, or need for target lesion revascularization) was significantly lower with intracoronary gamma radiation (28 versus 44 percent for placebo), due entirely to a reduction in target lesion revascularization.

●The WRIST trial included 130 patients with in-stent restenosis who underwent a successful percutaneous coronary intervention (PCI) and were then randomly assigned to either intracoronary radiation with a gamma emitter or placebo [20]. At six months, radiation was associated with a significant reduction in the incidence of angiographic restenosis (19 versus 58 percent for placebo), target lesion revascularization (14 versus 63 percent), and freedom from major cardiac events (29 versus 68 percent).

At five years, the major cardiac event rate continued to be significantly lower with intracoronary radiation (46 versus 69 percent with placebo) [14]. However, in the interval between six months and five years, radiation was associated with significantly more target lesion revascularization (22 versus 5 percent), suggesting that radiation may have delayed, but not prevented, the development of restenosis.

●The Long WRIST trial demonstrated the benefit of intracoronary radiation in 240 patients with long in-stent lesions (36 to 80 mm) [22]. The first 120 patients were randomly assigned to either gamma irradiation at a dose of 15 Gy or placebo; subsequently, an additional 120 patients were treated with 18 Gy. At six months, patients assigned to high dose radiation (18 Gy), low dose radiation (15 Gy), or placebo had target lesion revascularization rates of 20, 39, and 62 percent, respectively.

The benefit of radiotherapy was more pronounced in diabetic patients in the WRIST trial and in a retrospective review of 749 patients at a single center, 43 percent of whom were diabetic [21]. Although the outcomes were similar in nondiabetics and diabetics treated with radiation, higher rates of restenosis occurred in the diabetics in the placebo group [21]. (See "Coronary artery revascularization in stable patients with diabetes mellitus".)

●The INHIBIT trial randomly assigned 332 patients with diffuse in-stent lesions up to 47 mm long to intracoronary beta radiation or placebo after successful PCI [26]. At nine months, radiotherapy significantly reduced the incidence of in-lesion restenosis (16 versus 48 percent for placebo), in-vessel restenosis (26 versus 52 percent), and major adverse cardiac effects (death, Q wave MI, and target lesion revascularization; 15 versus 31 percent). These relative benefits were maintained during a two-year follow-up.

Comparison to drug-eluting stents — Two randomized trials (TAXUS V ISR and SISR) compared the efficacy of intracoronary radiation to a paclitaxel or sirolimus stent in patients with in-stent restenosis that developed after prior bare metal stenting. Both had significantly better outcomes in terms of angiographic restenosis and target lesion revascularization with drug-eluting stents (6.3 versus 13.9 percent at nine months in TAXUS V ISR and 8.5 versus 19.2 percent at nine months in SISR) [27,28]. At three years, freedom from target lesion revascularization was significantly better with sirolimus-eluting stents (SES) than radiation in the SISR trial (81 versus 72 percent) [29]. (See "Intracoronary stent restenosis", section on 'Choice of device'.)

In recurrent drug-eluting stent in-stent restenosis — Typically, restenosis after drug-eluting stent implantation is treated with either percutaneous transluminal coronary angioplasty (PTCA) or debulking (eg, laser or rotational atherectomy), followed by repeat drug-eluting stenting (so-called "sandwich stenting"). This approach is believed to diminish the risk of recurrent restenosis. Should restenosis recur following "sandwich stenting," additional stent implantation may have limited efficacy. At some centers, intracoronary brachytherapy is utilized to treat the stenosis in the sandwich stents. Although no randomized clinical trials have been conducted supporting this technique, a recent meta-analysis of six observational studies reported favorable long-term outcomes using intracoronary brachytherapy for this condition, with an overall incidence of target lesion revascularization of 14.1 percent at one year and 22.7 percent at two years [30].

Saphenous vein grafts — The role of intracoronary gamma radiation as a treatment of in-stent restenosis in saphenous vein grafts was evaluated in the SVG WRIST trial of 120 patients who underwent balloon PTCA, additional stenting, or atherectomy [31]. At six months, the incidence of angiographic restenosis was significantly lower with intracoronary radiation (21 versus 44 percent for placebo). At 12 months, intracoronary radiation was associated with a lower rate of target lesion revascularization (17 versus 57 percent) and a lower rate of major cardiac events (32 versus 63 percent). Similar benefits were noted in another study in which intracoronary gamma radiation was as effective in saphenous vein grafts as in native artery lesions [32]. (See "Coronary artery bypass graft surgery: Prevention and management of vein graft stenosis".)

De novo stenting — The benefit of intracoronary brachytherapy was less evident in the BRIDGE study of de novo coronary lesions treated with stenting [33]. Of 112 patients, 54 were randomly assigned to radiation after stent placement. At six months, minimal lumen diameter was significantly greater and volume loss was significantly less with brachytherapy. However, despite efforts to optimize procedural performance, edge restenosis and late occlusions resulted in a higher rate of target vessel revascularization with radiation (20 versus 12 percent).

Use for restenosis after PTCA — Intracoronary radiotherapy also appears to be effective treatment for restenosis occurring after PCI without stent placement [13,34-36]. The potential benefit was illustrated in the original SCRIPPS trial of 55 patients with a restenotic coronary artery that was a candidate for a stent (38 percent) or already contained a stent [34]. The patients were treated with PCI (balloon angioplasty and stent placement as needed) and then randomly assigned to intracoronary Ir-192 (a gamma-emitter) or placebo; angiographic restenosis at six months was much lower in the radiation group (17 versus 54 percent).

The benefit of radiation persisted at three and five years [13,35]. Target lesion revascularization was significantly less common among patients receiving intracoronary radiotherapy compared to placebo (15 versus 48 percent at three years and 23 versus 48 percent at five years). The composite end point of death, myocardial infarction, or target lesion revascularization was also significantly lower in the radiation group (23 versus 55 percent at three years and 38 versus 65 percent at five years) (figure 3). However, more late loss in minimal luminal diameter was seen in the radiation group.

However, intracoronary brachytherapy is rarely used for restenosis after PTCA in the current era, particularly with the availability of small diameter drug-eluting stents.

Use after angioplasty alone — Data are limited on the routine use of intracoronary radiation after balloon angioplasty alone, as a prophylactic measure to prevent restenosis. This issue has become less important, since the majority of percutaneous coronary interventions currently performed include stent placement.

Small case series and a small clinical trial, PREVENT, that was not limited to patients receiving angioplasty alone suggested a reduction in postangioplasty restenosis with intracoronary radiotherapy, using either a gamma or beta emitter [36-38]. In contrast, a much larger study of 1455 patients with either a single de novo lesion or unstented restenosis found no significant decline in clinical restenosis with radiotherapy with a beta-emitter compared to placebo [39]. Although radiotherapy reduced the incidence of angiographic restenosis within the lesion segment, it also increased restenosis at the lesion edges.

SAFETY — Coronary brachytherapy appears to be relatively safe. However, a number of concerns specific to this technology have been raised. These include possible injury to angiographically normal vessels, late thrombosis, edge stenosis, delayed healing of dissections, and acute coronary spasm. In addition, restenosis remains a problem.

Effect on normal coronary segments — There is no evidence of any permanent deleterious effect of irradiation on angiographically normal coronary artery segments, as illustrated by the following observations:

●Intracoronary ultrasound performed at six months in the WRIST trial showed no effect on normal reference segments [40].

●Late follow-up in the SCRIPPS trial showed that the rate of nontarget-vessel revascularization at five years was the same in the radiation and placebo groups [35].

Late thrombosis — Late thrombosis, occurring more than 30 days after radiation therapy, is an infrequent complication that may be associated with myocardial infarction [19,41,42]. Potential causes of late thrombosis include [43-45]:

●Delayed re-endothelialization, due to the inhibition of neointima over the newly placed stents, which leaves them exposed to blood; the struts become a potential nidus for thrombosis, well beyond the two- to four-week time for endothelialization seen after stenting without radiation

●Fibrin deposition and enhanced platelet recruitment

●Impaired vasoreactivity and spasm

●Tissue erosion around the stent

●Unhealed dissection

●Impaired resolution of intramural hemorrhage

The frequency with which late thrombosis occurs was assessed in a report of 473 patients who presented with in-stent restenosis and were enrolled in various radiation protocols, whether randomized to placebo versus radiation or entered into registries [42]. After a follow-up of at least six months, 9 percent presented with a late thrombotic coronary occlusion associated with a myocardial infarction, unstable angina, or no symptoms; late thrombotic occlusion was much less common (1.2 percent) in patients treated with placebo.

With multivariate analysis, implantation of a new stent was the main predictor of late thrombotic occlusion [42]. Similar findings were noted in the Gamma-One trial in which late thrombosis occurred in 5.3 percent of patients receiving radiation at nine months [19]. It was seen only in patients who had received a new stent at the time of radiation treatment and only after the discontinuation of oral antiplatelet therapy with ticlopidine or clopidogrel.

Restenosis — As noted in the trials above, intracoronary radiotherapy for in-stent restenosis lowers the rate of subsequent restenosis compared to no radiotherapy. However, recurrent restenosis still occurs in 15 to 29 percent at nine months [19,20,26,46]. Delayed restenosis is also possible. In a five-year follow-up from the original SCRIPPS trial, there were two occurrences of very late restenosis in the treated segment [35].

The clinical features of late failure after intracoronary radiation were evaluated in a series of 961 patients from the WRIST trials, 282 of whom (29 percent) failed at a mean follow-up of 494 days; the failure rate was the same for beta and gamma radiation [46]. The main risk factor for late failure was repeat stenting. The lesions were primarily focal with a mean length of 11.9 mm. Most patients were treated with percutaneous coronary intervention (PTCA in 61 percent, restenting in 26 percent, and atherectomy in 11 percent) or, less often, bypass surgery. Although the patients responded well to revascularization, 17 percent required repeat target vessel revascularization at six months.

A limitation to these observations is that they were made before the need for prolonged combined antiplatelet therapy, as described above, was appreciated. In addition, approximately 10 to 15 percent of failures were determined by angiography, not by symptoms or stress testing, which may have overestimated the true failure rate.

Similar findings were noted in a smaller study of 45 patients with restenosis after intracoronary irradiation in the Gamma-One and Gamma-Two trials [47]. Repeat percutaneous coronary intervention (PCI) was acutely successful in all patients but subsequent restenosis occurred in 15 (33 percent), requiring yet another revascularization procedure, either surgical or percutaneous.

Prolonged dual antiplatelet therapy — The association of late thrombosis with a new stent and discontinuation of ticlopidine or clopidogrel led the FDA to recommend avoidance of placement of new stents when brachytherapy is used, if possible, and to continue dual antiplatelet therapy (aspirin plus clopidogrel or ticlopidine) for a minimum of six months after brachytherapy and for at least one year if a new stent is implanted [5]. The potential efficacy of prolonged dual antiplatelet therapy has been demonstrated by the following observations:

●In the WRIST PLUS trial of 120 patients with diffuse in-stent restenosis, six months of therapy with clopidogrel and aspirin was associated with the need for a new stent in 28 percent of patients and an incidence of total occlusion and late thrombosis of 5.8 and 2.5 percent, respectively [48]. These rates were lower than those seen with radiation therapy in the WRIST and LONG WRIST trials in which antiplatelet therapy was given for only one month and were similar to those in the placebo group in these trials.

●In the WRIST 12 trial, 120 patients with diffuse in-stent restenosis were treated with clopidogrel and aspirin for one year; the results were compared to those in WRIST PLUS [49]. Although there was no difference in late thrombosis at 15 months in the two groups (3.3 and 4.2 percent), prolonged antiplatelet therapy was associated with a significant reduction in target vessel revascularization (20 versus 35 percent). Only four patients in these two studies developed late thrombosis while still taking clopidogrel.

These findings have led many clinicians to continue dual antiplatelet therapy for one year after intracoronary radiation therapy.

Edge stenosis — Edge restenosis, which was originally described with the use of radioactive stents, also occurs with intracoronary radiation [50,51]. In the WRIST trial, for example, the incidence of edge stenosis was 10 percent with intracoronary radiation compared to 4.7 percent with placebo after successful treatment of in-stent restenosis with PCI [50].

The development of edge stenosis is enhanced by "geographic miss" [50]. This term refers to failure of the radiation source to adequately cover the entire segment of artery injured during the PCI that is performed in conjunction with the radiation therapy (figure 4). Edge stenosis can be avoided by use of a radiation source at least 10 mm longer than the mechanically treated segment and careful positioning of both dilating balloons and the radiation source.

Delayed healing of dissections — Mild to moderate dissections are not uncommon after angioplasty, and usually resolve completely. Only limited data are available on the effect of intracoronary radiation on the healing of a dissection. One report reviewed 44 patients with postangioplasty dissections who underwent intracoronary ultrasonography at six months [52]. At this time, residual dissection was present in only 1 of 28 patients who did not receive intracoronary radiation compared to 7 of 16 who did (3.5 versus 44 percent). In the latter group, however, there was a significant improvement in the dissection and an increase in lumen area.

Coronary artery spasm — Animal and clinical studies have demonstrated an incidence as high as 67 percent of coronary artery spasm after high-dose (24 Gy) beta irradiation [53]. Spasm occurred immediately, proximal and distal to the stent, and is due to endothelial dysfunction. Although the spasm was usually minor or moderate, ECG changes or hemodynamic instability were seen in almost 30 percent of cases. Repetitive administration of nitroglycerin is an effective therapy in this situation.

SUMMARY

●Background – Brachytherapy using either gamma or beta radiation sources became a common treatment for in-stent restenosis prior to the availability of drug-eluting coronary stents, since it substantially reduced recurrent restenosis. It can be particularly helpful in treating long lesions, which are more subject to recurrent restenosis. The procedure has a favorable safety profile, especially if dual antiplatelet therapy is continued for an extended period (up to one year) to reduce occurrences of late thrombosis [54].

●Drug-eluting stents are still preferred over intracoronary radiation as initial treatment of in-stent restenosis. (See 'Comparison to drug-eluting stents' above and "Intracoronary stent restenosis", section on 'Choice of device'.)

●Possible indication – Brachytherapy may be considered in certain instances of in-stent restenosis. Typically, it may be considered after recurrent in-stent restenosis after initial treatment with a drug-eluting stent has failed.