Localization of neurovascular bundles on pelvic CT and evaluation of radiation dose to structures putatively involved in erectile dysfunction after prostate brachytherapy

Purpose: To (a) locate neurovascular bundles (NVB) on pelvic CT and (b) retrospectively evaluate relationships between radiation dose to structures putatively involved in prostate brachytherapy-induced erectile dysfunction (ED) and incidence of postbrachytherapy ED. Methods and Materials: (a) Right/left NVB were identified on nine prostate MRIs. Structures visible on MRI and CT were cross-referenced. Cross-sectional area of each NVB was measured. (b) All patients treated with implant alone and whose treatment was planned on Variseed (Varian Medical Systems, Palo Alto, CA), with follow-up of > 12 months were included; n = 41. Median follow-up was 20 months. All patients were potent (+/- sildenafil) before implant (erection sufficient for intercourse). The right/left NVB (using results from part "a"), penile bulb, and right/left crus were outlined on postimplant CT. Volumes and doses to these structures were calculated. Results: (a) On prostate MRI, NVB was consistently located where the prostate border bends away from the levator ani, at the gland's smallest radius of curvature. Average area of the circle best encompassing the NVB = 0.27 cm(2); diameter was 0.58 cm. (b) 11 of 41 (27%) patients had ED; 30 of 41 were potent (15 with sildenafil). There was no significant difference between potent/impotent patients in isotope, age, diabetes, hypertension, follow-up, or volume of prostate, bulb, right/left NVB, or right/left crus. There was a relationship between smoking and ED (p = 0.05). There was a relationship between bulb %D90 and ED: >10% 67% (4 of 6) vs. <10% 20% (7 of 35) (p = 0.03), which remained when controlling for smoking. There was no relationship between dose to left NVB and potency. There was paradoxical decreased risk of ED with right NVB %V100 >60% (p = 0.019), and right NVB %D60 >100% (p = 0.003). There was no relationship between dose to right/left crus and ED. Conclusions: A reliable method for localizing NVB on CT is demonstrated. There is no increased risk of prostate brachytherapy-induced ED with increasing dose to crus or NVB at the doses given in this study. There is a possible dose-response relationship between dose to the bulb and risk of ED. (C) 2004 Elsevier Inc.