BACKGROUND:Non-invasive measurement of bladder volume could help patients determine proper timing for voiding or catheterization. A wearable, bladder ultrasound device (DFree®, Triple W Japan Inc.) was introduced in 2018 but has not been validated for efficacy and safety in children. We sought to determine whether the DFree® device could be used to monitor real-time bladder volume in children with urologic conditions by comparing device readings to instilled volume during urodynamic study (UDS).
METHODS:We prospectively enrolled 40 children (6 months to 18 years old) who were undergoing UDS (10/2023 to 02/2024). Exclusion criteria included high-grade (IV or V) vesicoureteral reflux on prior voiding cystourethrogram and pelvic fluid collections on ultrasound. The device was taped to the suprapubic region and volume measurements were obtained in real-time through a mobile application on a Bluetooth-enabled iPad. Measurements were displayed as an integer on a 0 to 10 scale corresponding with an empty to full bladder. Measurements after leak/voiding events were excluded from analysis. Primary outcome was accurate measurement of bladder capacity at last measurement, defined as device measurement >8, when instilled volume was >80% of cystometrogram-determined capacity. Mixed effects linear regression was used to model the relationship between individual device measurements and percentage of bladder capacity after adjusting for clinical variables.
RESULTS:Median age was 7.2 years, 20 (50%) were male, 27 (67.5%) were CIC-dependent and 30 (75%) had spina bifida (Table 1). Although we excluded high-grade reflux, 5 (12.5%) had low-grade reflux (I to III). 9 (22.5%) had a history of prior bladder surgery and 13 (32.5%) had ultrasound-detected bladder abnormalities. The device accurately measured capacity in 46.2% of patients with a sensitivity of 0.70, specificity of 0.13, positive predictive value of 0.53, and negative predictive value of 0.22. Certain patients appeared to have device readings which tracked closely with actual instilled volume, while others correlated poorly (Figure 1). No significant difference in patient characteristics was noted regarding accurate versus inaccurate measurements. After adjusting for other variables, every 10% increase in device reading was associated with an average increase of 10% (95% CI: 9 to 12%, p<0.001) in bladder volume (Table 2).
CONCLUSIONS:The DFree® device performed poorly for determining bladder capacity in this cohort of children with urologic conditions. We were unable to identify patient characteristics that predicted device accuracy. However, we did observe certain individuals in whom the device appeared relatively accurate; this suggests that repeated measurements during a trial period might identify such patients. Additional studies are needed to test the device’s reliability in this hypothetical group of patients over multiple datapoints in real-world environments.
