Evidence review

Clinical and technical evidence

Regulatory bodies

A search of the Medicines and Healthcare Products Regulatory Agency (MHRA) website revealed no manufacturer Field Safety Notices or Medical Device Alerts for this device. There were 77 adverse events identified from a search of the US Food and Drug Administration (FDA) database: Manufacturer and User Device Facility Experience (MAUDE) from 2007 to the present. One report related to the BladderScan BVI 3000 model, 29 reports did not clearly state the model, and the remaining 17 were on the BladderScan BVI 9400model (FDA 2013). All adverse events were listed as device malfunctions, including device lock‑up and false readings (for example, reporting zero when the bladder was not empty). None were reported to result in patient harm.

Clinical evidence

Of 18 papers identified, 13 were excluded from further assessment because they reported an intervention or outcomes that were not relevant to this briefing (for example, if a study investigated device repeatability but not accuracy), included an overlapping study population, had a small sample size (fewer than 10 people) or were not clinical studies. Studies were only included if they used the BladderScan BVI 9400. No studies in adults met the inclusion criteria for this briefing. Three fully published studies and 1 abstract in paediatric populations are summarised in this briefing.

The Bevan et al. (2011) study was set in Australia. The reproducibility and accuracy of the BladderScan BVI 9400 was compared with conventional real‑time ultrasound (RTUS; ACUSON S2000, Siemens) for measuring bladder volume in 61 healthy children aged 0 to 24 months. Children were scanned with both the BladderScan and RTUS on 2 separate occasions, 1 hour apart. The BladderScan readings were done by 2 senior paediatric doctors. On each occasion 6 BladderScan measurements were taken for each child, 3 by each doctor. In 36 children, the measurements were repeated using a second BladderScan BVI 9400 machine in case of machine variability. The BladderScan measurements were immediately followed by RTUS measurement. All RTUS measurements were done by a single senior paediatric sonographer who was blinded to the BladderScan measurements. The BladderScan and RTUS measurements were compared by Bland−Altman analysis to determine the reproducibility and the limits of clinical agreement (as a measure of accuracy). The authors describe 'poor' reproducibility within the repeated BladderScan measurements at each time point for both BladderScan BVI 9400 devices used. BladderScan measurements ranged from 10 ml to 86 ml. The overall reproducibility coefficient within BladderScan readings was 20 ml, indicating that 95% of repeated BladderScan measurements were within 20 ml of each other. The authors report 'poor' accuracy when comparing the BladderScan measurements to RTUS. By Bland–Altman analysis, the 95% limits of agreement between BladderScan and RTUS were −31 to +19 ml, indicating that 95% of BladderScan readings were between −31 and +19 ml of the corresponding RTUS measurement. The authors concluded that the BladderScan BVI 9400 does not appear to be a reliable method for assessing bladder volumes in children aged 0 to 24 months. Details of the study are shown in table1.

The Buntsma et al. (2012) study was set in Australia. The success rate of suprapubic aspiration (SPA) was determined for 60 children aged 0 to 24 months when assisted by BladderScan BVI 9400. There was no comparator measurement. The children had presented to A&E needing urine collection by SPA. The audit showed an overall success rate of 53% (32/60; 95% confidence interval 41–66%), which the authors describe as low. The BladderScan‑assisted SPA success rate was higher in children with readings ≥20 ml (70%). The authors also note that this rate is lower than previous success rates reported in studies using RTUS. (See table 2 for details.)

The study by Marciano et al. (2013) was available as an abstract. The study was set in Italy. The authors used the 'child mode' of the BladderScan BVI 9400 to evaluate the accuracy of bladder volume measurement in 59 children who were scheduled for diagnostic evaluation or treatment under general anaesthesia (or deep sedation). There was no comparator measurement in this study. The average age was 7.1 years (range 1–19 years). Bladder volume was measured using BladderScan immediately after bladder emptying. Saline was then infused into the bladder of each child using a transurethral catheter. The bladder volume was measured again after total volumes of 20 ml, 50 ml and 100 ml of saline had been infused. The BladderScan measurement was repeated 3 times by the same operator for each volume infused. No significant differences were found among the repeated BladderScan measurements. Bladder volumes were negligible after bladder emptying. The bladder volumes measured using BladderScan were statistically significantly lower than the amount of saline infused when the children were considered as a single group. However, when the data were stratified according to both the age of the child and the amount infused, only the volumes measured in children aged 1 to 6 years were significantly lower than the amount of saline infused. No significant differences were seen for infusion volumes above 20 ml in the children aged 7 to 12 years or at any volume in the children over 12 years. The authors concluded that the BladderScan BVI 9400 provides an accurate and reliable measure of bladder volume in children, but that use should be limited to children older than 6 years. (See table 3 for details.)

The Rowe et al. (2014) study was set in New Zealand. The 'child mode' of BladderScan BVI 9400 was used to evaluate the accuracy of bladder volume measurement compared with catheterisation in 50 children who were scheduled for urodynamics (assessment of bladder and urethra functioning) or surgery that included urethral catheterisation. The BladderScan volumes ranged from 0 ml to 513 ml (mean=79; median=34) and the catheterised volume from 0 ml to 500 ml (mean=81; median=31). The correlation between the BladderScan and catheterised urine volume was 0.96 (95% confidence interval 0.92 to 0.97); the mean difference between the volumes was −2.0±21 ml. In 12 children aged less than 36 months, the BladderScan volumes ranged from 0 ml to 39 ml (mean=14 ml; median=13 ml) and the catheterised from 0 ml to 40 ml (mean=16 ml, median=14 ml). Correlation between the BladderScan and catheterised volume in children aged less than 36 months was not as strong (rho=0.82) with a mean volume difference of −2.6 ml. The authors concluded that BladderScan BVI 9400 was accurate when compared with catheterised volume but should be used with caution in children under 36 months. (See table 4 for details.)

Recent and ongoing studies

No ongoing or in‑development trials using the BladderScan BVI 9400 for bladder volume measurement were identified.

Costs and resource consequences

The BladderScan BVI 9400 can be used instead of diagnostic catheterisation to help diagnose urological conditions non‑invasively. It can also be used to assess whether therapeutic catheterisation would otherwise be needed and could, therefore, reduce the number of unnecessary catheterisation procedures. No other additional facilities or technologies are needed alongside the technology. The manufacture recommends training and offers an educational training pack and an in‑service session to train staff, both of which are included in the cost of the machine. The manufacturer states that a sonographer is not needed to operate the BladderScan BVI 9400.

No published evidence on the resource consequences of the BladderScan BVI 9000 series was identified.

Strengths and limitations of the evidence

All 4 studies considered in this briefing were prospective cohort studies in children, which limits their generalisability to adults. No randomised controlled trials were identified. The BladderScan BVI 9400 model was used in all 4 studies. All studies were carried out in secondary or tertiary care settings, which may limit their generalisability to the primary care setting. In addition, none of the studies were carried out within an NHS setting or within the UK. Clinical practice and patient characteristics may differ between countries and care must be taken when relating them to the NHS context.

The BladderScan BVI 9400 is intended to be used for both adults and children; however, no clinical evidence investigating the accuracy and reproducibility of this model for the assessment of bladder volume in adults met the inclusion criteria for this briefing. Evidence in adults exists for the previous versions of BladderScan, however no studies have compared previous models to the current model and, therefore, it is unclear whether results obtained from previous models are generalisable to the 9400 model.

Rowe et al. (2014) was the only study that reported a sample size calculation. Consequently, it is unclear whether the other studies were adequately powered to detect differences in the outcomes.

A potential source of performance bias is the training received to use the device. The manufacturer states that a sonographer is not needed to carry out the scanning, and that the BladderScan should only be used by healthcare professionals who have had training and authorisation by the appropriate healthcare provider. Training of scanner operators was mentioned in the Buntsma et al. (2012) study, but not described in the other studies. Skill levels and scanning methodology may differ among operators.

Marciano et al. (2013) was published as an abstract only so has limited detail compared with the other studies. It is not clear whether the bladders were emptied between each infusion of saline or whether additional volumes were added sequentially to give total volumes of 50 ml and 100 ml. It is also not clear whether the investigators were blinded to the volumes of saline infused. The fact that the bladder was infused with saline may be a source of bias, because this is not the same as natural urine production. It is unclear whether this would affect BladderScan readings. Additionally, the underlying conditions of the children were not reported, which may be another potential source of bias. The infusion of saline solution into a child's bladder may carry ethical implications, and it is unclear whether these were addressed in the study.

The 2 studies assessing intra‑observer reproducibility of BladderScan measurement reported varied results. Marciano et al. (2013) found no significant intra‑observer differences between repeated BladderScan measurements. In contrast, Bevan et al. (2011) reported lower levels of intra‑observer reproducibility. There was variation in the statistical analyses used in these studies and so care should be taken when comparing the magnitude of outcome significance between studies.

There was variation in the measures used to show BladderScan accuracy. Two studies looked at the accuracy of the BladderScan compared with known bladder volume (Marciano et al. 2013 using a saline infusion; Rowe at al. 2014 using catheterisation) and 1 investigated accuracy compared with real‑time ultrasound (RTUS; Bevan et al. 2011). The remaining study (Bunstma et al. 2012) assessed the success rate of a suprapubic aspiration procedure when aided by the BladderScan. In Buntsma et al. (2012) there was no within‑study comparator, but the authors compared results with success rates from conventional RTUS in previous studies. The limitation with this comparison is that the setting was likely to be different between the previous and current studies and therefore there were likely to be confounding factors. For accuracy studies, the optimal comparator is bladder volume determined using bladder catheterisation, however, this is not possible for ethical reasons in children not already due to undergo procedures involving catheterisation. As mentioned above, it is unclear in the Marciano et al. (2013) study whether the ethical implications of catheterising children to determine bladder volume were addressed.

Age should be taken into account when considering the accuracy or reproducibility of the BladderScan device for children. Bevan et al. (2011) reported that the BladderScan may not be a reliable or accurate method for assessing bladder volume in children aged between 0 months and 24 months. Marciano et al. (2013) reported that although BladderScan produced accurate readings overall, there was less accuracy in children under 6 years, and suggested that use should be limited to children older than 6 years. Rowe et al. (2014) concluded that the BladderScan was accurate when compared with catheterised volume but should be used with caution in children younger than 36 months.

Verathon provided the BladderScan device in the Rowe et al. (2014) study. None of the other studies reported any manufacturer involvement. Manufacturer involvement in the project may have potential for introducing bias in the reporting of outcomes.