Serial intravascular ultrasound analysis of the impact of lesion length onthe efficacy of intracoronary gamma-irradiation for preventing recurrent in-stent restenosis

Background-The relation between lesion length and effectiveness of brachyth erapy is not well studied. Methods and Results-We compared serial (postintervention and follow-up) int ravascular ultrasound findings in 66 patients with native coronary artery i n-stent restenosis (ISR) who were treated with Ir-192 (15 Gy delivered 3 mm away from the radiation source). Patients were enrolled in the Washington Radiation fur In-Stent Restenosis Trial (WRIST; ISR length, 10 to 47 mm; n= 36) or Long WRIST (ISR length, 36 to 80 mm; n=30). External elastic membran e, stent, lumen, and intimal hyperplasia (IH; stent minus lumen) areas and source-to-target (intravascular ultrasound catheter to external elastic mem brane) distances were measured. Postintervention stent areas were larger in WRIST and smaller in Long WRIST patients (P<0.0001). At follow-up, maximum IH area significantly increased in both WRIST and Long WRIST patients (P<0 .0001 for both), but this increase was greater in Long WRIST patients (P=0. 0006). Similarly, minimum lumen cross-sectional area significantly decrease d in both WRIST and Long WRIST patients (P<0.05 and P<0.0001, respectively) , but this decrease was more pronounced in Long WRIST patients (P=0.0567), The maximum source-to-target distance was longer in Long WRIST than in WRIS T, and it correlated directly with ISR length (r=0.537, P<0.0001), Overall, the change in minimum lumen area and the change in maximum TH area correla ted with the maximum source-to-target distance (r=0.352, P=0.0038 and r=0.5 23, P<0.0001 for WRIST and Long WRIST, respectively). The variability (maxi mum/minimum) in IH area at follow-up also correlated with the maximum sourc e-to-target distance (r=0.378, P<0.0001). Conclusions-Brachptherapy may be less effective in longer ISR lesions becau se of the greater variability and longer source-to-target distances in diff use ISR.