Vascular brachytherapy using beta and gamma radiation was the standard for percutaneous reintervention in patients with in-stent restenosis.^1–4 While its beneficial effect has been excellent, the development of new lesions at the proximal and distal edges of the irradiated segments — the “edge effect” — necessitates additional care during such interventions.^5,6 Clinical trials utilizing beta sources have shown that at least a 10 mm or more safety margin of radiation from the balloon-injured segment is required to minimize this phenomenon.⁷ However, this strategy requires irradiation of uninjured, often “normal” areas of the artery, with little known about the potential damage of radiation to “normal” coronary artery tissue. The following case describes the development of new and significant disease in proximal, normal-appearing coronary vascular segments irradiated during a procedure for more distal in-stent restenosis. The mechanism likely involved angiographically unapparent injury by the guiding catheter in combination with the radiation treatment. This adverse effect of coronary brachytherapy has not been previously reported.

Case report. A petite (4 ft. 11 in., 133 lbs.) 57-year-old female with hyperlipidemia and a history of smoking, presented with unstable angina in August 2002. She underwent percutaneous coronary intervention with a 2 mm x 18 mm stent placed in the mid-left anterior descending (LAD) in combination with balloon angioplasty of a first diagonal artery. Four months later, she presented with increasing exertional angina; angiography revealed high-grade in-stent and peri-stent restenosis. A significant focal proximal stenosis in a small first obtuse marginal artery, present previously, was also noted (Figure 1).
A 7 Fr L3.5 guide catheter was advanced to the left coronary artery and a 2.0 x 20 mm balloon was used for dilatation of the lesions, with inflations up to 12 atmospheres. A 2.5 x 15 mm balloon was required for further dilatation to permit passage of the 32 x 2.5 mm Galileo III Centering Catheter (Guidant, Houston, Texas). The centering catheter was initially positioned from the middle of the stent to the more distal LAD (Figure 2A). After the prescribed radiation dose of 20 Gy at 1 mm radial distance from the center of the source was delivered, the centering catheter was pulled back to adequately cover and irradiate the more proximal portion of the injured area in a similar manner to provide adequate safety margins (Figure 2B). The dwell time was 69 seconds for the distal segment and 73 seconds for the proximal segment. Inflation of the centering balloon was not used due to the small size of the vessel and concern about injury on either side. Following this, a 2.0 x 13 mm Pixel stent (Guidant Inc., Santa Clara, California) was deployed in an overlapping fashion at 11 atmospheres distal to the previous stent. Final angiography revealed a satisfactory result with no significant residual stenosis (Figure 3). The patient was discharged on beta-blocker, ACE inhibitor, and statin, in addition to dual antiplatelet therapy.
She returned eight months later with recurrence of severe and frequent anginal symptoms. Left coronary cannulation with a 5 Fr catheter resulted in prompt damping of pressure. Due to significant left main stenosis, the study was limited, but angiography revealed severe left main and ostial LAD disease and a high-grade ostial lesion of the left circumflex vessel (Figure 4). There appeared to be only moderate restenosis within the previously stented area, however, more definitive data are lacking due to the limited study.

Discussion. The limited length of radiation sources and the need to cover the entire treated area to avoid “edge effect” has led to the development of tandem source positioning using either manual or automated stepping systems. Thus, it is not unusual for sizable areas of angiographically normal-appearing and non-intervened segments being exposed to irradiation.⁸ While generally regarded as innocuous when balloon injury has not accompanied the radiation of the segment, our case demonstrates that this approach may not be entirely safe and may include areas inadvertently affected that are remote from areas of usual concern.
The edge phenomenon is thought to be due to the stimulatory effect of a lower dose of radiation delivered by the source ends at the lesion extremity, causing neointimal proliferation in the balloon-injured artery.⁹ During an interventional procedure, the injury to the vessel wall may not be limited to the intervened area. In fact, non-intervened vascular segments may be inadvertently injured due to the use of interventional equipment such as guidewires, balloons, cutting devices and even the centering or delivery catheter tip itself. Furthermore, as in this report, the guiding catheter may also serve as a source of injury in small vessels or when inadvertent or aggressive intubation has been necessary. In our patient, the combination of such injury to the proximal vascular segment and low-dose radiation in the same area was the likely precursor for the development of significant stenoses.
Other deleterious effects of radiation, such as adventitial fibrosis and atherosclerosis, are not evident until four to five years after the treatment.^10–11 Therefore, it is unlikely that in this case the development of new lesions in the short span of eight months is secondary to these phenomena. Similarly, de novo atherosclerosis in the affected segments was unlikely due to the short time span. Data about the biological effects of intravascular radiation on normal vessels are limited.¹² Studies evaluating beta and gamma radiation for in-stent restenosis have mainly focused on the original lesion length, including immediate proximal and distal subsegments of the artery in determining the restenosis rate. Therefore, none of these trials give any insight to the outcome of untouched, but irradiated remote vascular segments.
This case illustrates that while it is important to adequately irradiate the edges to avoid restenosis, special care should also be taken to minimize the radiation exposure in normal vascular segments remote from any intervened coronary segments.