Ultrasound-Based Bladder Volume Assessment in Gynecological Radiotherapy: Agreement with CT/CBCT and Longitudinal Variation

Background: Consistent bladder filling is important in postoperative pelvic radiotherapy because changes in volume can alter the relationship between pelvic targets and adjacent organs at risk. CBCT provides three-dimensional verification but adds treatment-room time and imaging dose. Ultrasound offers a rapid, portable, non-ionizing method for checking bladder volume before treatment-room imaging. Objective: To evaluate agreement between investigator-performed ultrasound (US) and reference planning CT/cone-beam CT (CBCT) bladder-volume measurements in postoperative gynecological radiotherapy, and to characterize longitudinal bladder-volume change during treatment. Methods: This prospective, single-institution method-comparison study included 29 postoperative patients receiving adjuvant pelvic radiotherapy (14 with cervical and 15 with endometrial cancer). A total of 406 paired US-reference measurements were obtained at 14 scheduled time points from simulation to treatment day 25. Patients followed a standardized protocol of bladder emptying followed by 500 mL of water 40 minutes before imaging. US was performed with a GE LOGIQ F8 Expert system and curvilinear/convex transducer; the automated volume function used three orthogonal bladder dimensions. Three consecutive scans were obtained and the median was analyzed. Planning CT was the baseline reference and CBCT was the treatment reference. Agreement was assessed using a two-way mixed-effects, absolute-agreement, single-measure ICC and Bland-Altman methods adapted for repeated observations within patients, with patient-level bootstrap confidence intervals. Longitudinal change was evaluated with a linear mixed model. Results: Mean bladder volume was 195.5 ± 71.6 mL by US and 199.2 ± 71.6 mL by reference CT/CBCT. The overall correlation was strong (r = 0.951, p < 0.001), and the single-measure absolute-agreement ICC was 0.950 (95% CI, 0.927-0.965). Repeated-measures Bland-Altman analysis showed a mean bias of -3.69 mL (95% CI, -6.48 to -1.27 mL), with limits of agreement of -47.52 mL (95% CI, -54.29 to -41.03) and 40.14 mL (95% CI, 36.47 to 44.26). Baseline US-planning CT bias was -2.28 mL, while treatment US-CBCT bias was -3.80 mL. In exploratory agreement-threshold analyses, 364/406 pairs (89.7%) were within 20%, 396/406 (97.5%) within 30%, 400/406 (98.5%) within 50 mL, and 405/406 (99.8%) within 100 mL. Mean reference volume declined from 249.6 ± 86.6 mL at baseline to 164.4 ± 49.3 mL on day 25; the mean participant-level percentage decline was 28.0% ± 25.5%. Conclusion: Investigator-performed US showed close average agreement with planning CT/CBCT for bladder-volume estimation, but individual differences approaching approximately 50 mL remained within the repeated-measures limits of agreement. US may therefore be useful as a rapid preparation-screening adjunct, but it cannot verify postoperative target position or the full pelvic anatomy and should not replace clinically indicated image guidance. Patient-specific or plan-specific implementation thresholds require prospective validation against anatomy, dosimetry, workflow outcomes, and clinical endpoints.