<h3>Purpose/Objective(s)</h3> Salvage radiotherapy (SRT) is the only curative option for recurrent prostate cancer after radical prostatectomy. Urinary bladder is the major organ at risk in this setting. Different bladder filling protocols are used to reduce radiation dose to the bladder and thus limit genitourinary (GU) side effects. However, constant bladder volume is hard to maintain throughout the treatment course. A pronounced inter-fractional variation in bladder volume is a well-recognized problem in SRT. The purpose of this study was to investigate if there is a relationship between actual bladder volumes measured on daily cone-beam computer tomography (CBCT) images and GU adverse effects. <h3>Materials/Methods</h3> Consecutive patients treated with SRT at Karolinska university hospital, Sweden, in 2019-2020 (n=126) were included. Patients received either 35 × 2 Gy (n=103) or 35 × 2.2 Gy and/or pelvic lymph node irradiation (n=23) with IMRT (n=82) or four-field box-technique (n=44). All patients were treated with image-guided radiotherapy (IGRT) including daily cone-beam CT (CBCT). Bladder volumes were graded (visual analogue scale) on CBCT images (n=4410) at each treatment fraction. Internal validation of bladder volume measurements was performed in 22 patients by manually delineating bladders on daily CBCTs (n=770). Clinician-reported outcome (ClinRO) and patient-reported outcome measures (PROM) were available in 103 (83%) and 80 (66%) cases, respectively, ClinRO and PROM combined in 68 (54%) patients. Baseline bladder volumes, within-patient volume variation as well as average of 35 CBCT bladder volumes for the entire treatment course were compared between groups of patients with and without GU adverse events during 12 months of follow-up. <h3>Results</h3> Manually delineated and visually graded bladder volumes were strongly correlated (r=0.869, p<0.001). ClinRO and PROM were significantly associated (Chi-square =33.1, p<0.001). A repeated-measures ANOVA determined that mean bladder volumes did not differ across 35 CBCT measurements (p=0.16). Average actual bladder volumes were significantly smaller than baseline volumes (Wilcoxon S-R test Z=-2.68, p<0.007). Bladder volumes were significantly smaller in patients with GU adverse events compared to asymptomatic patients, measured with both ClinRO (p=0.006) and PROM (p=0.003). <h3>Conclusion</h3> In conclusion, our detailed analysis indicates that patients with GU adverse effects had significantly smaller bladder volumes during SRT. Clinical use of bladder filling protocol to increase bladder size during SRT is thus justified.