Abstract: Purpose/Objective: To evaluate the effect of TURP cavity size on urinary symptom severity and incontinence following prostate brachytherapy Materials/Methods: 52 patients with a history of prior TURP were treated with prostate brachytherapy for T1-T2 prostate cancer from 10/93 — 12/01. TURP cavity volume was determined from intraoperative ultrasound by measurement of the maximum height and diameter of the cavity in transverse. The cavity volume was estimated using the prolate ellipsoid formula-4/3πhr2. Prospective data was collected using the International Prostate Symptom form (IPSS) and by recording presence of stress incontinence. Post-implant dosimetry was performed at 1 month by CT DVH analysis. Doses to 90% of the prostate (D90), 30% of the urethral volume (U30) and the % of prostate covered by the prescription dose (V150) were reported. Associations were tested by chi-square method. Differences in means were tested by Wilcoxon sign rank. Results: Mean patient age was 72 years (range 60-84). Prostate volume ranged from 12.7-85 cc (mean 31.6 cc). Implant doses were D90: 141-225 Gy (mean 172 Gy), V150: 28-93% (mean 59.7%) and U30: 127-398 Gy (mean 235 Gy). The median time from TURP to implant was 6.51 years (range 3 months to 22.4 years). TURP cavity size ranged from 0-9 cc (mean 2.1) 30/52 (57.7%) had a visible TURP cavity of which 15 (28.8%) had a cavity 2cc or greater. Mean follow-up was 4 years (1-9.5). 1/37 (2.3%) required a post-implant urinary catheter for 3 days. Mean pretreatment total IPSS increased from 7 to 12 at 1 month (p<0.001) and to 10 at last follow-up (p=0.027). Mean pretreatment QOL increased from 1.5 to 2.3 at 1 month (p=0.002) and remained at 2.3 at last follow-up (p=0.002). Patients with TURP cavities >2cc had a worsening of urinary IPPS QOL compared to patients with a defect <2cc (p=0.03). Incontinence occurred in 4/52 (7.7%) and was significantly associated with size of the TURP cavity. 4/15 (27%) of patients with a TURP cavity greater than 2cc developed incontinence versus 0/37 with a TURP cavity less than 2cc (p=0.001). The risk of incontinence associated with a defect > 2cc was 4.4 (95% CI 2.6-7.3, p=0.001). There were no associations with patient age, D90, V150, U30 or prostate size and the development of incontinence. Conclusions: TURP cavity size greater than 2cc at the time of implant is highly associated with the risk of late urinary incontinence. The occurrence of incontinence in TURP patients does not appear to be influenced by prostate or urethral doses. Estimation of TURP cavity size should be included when considering such patients for prostate brachytherapy.