Ultrasound Obstet Gynecol 2015; 46: 233–238 Published online in Wiley Online Library (wileyonlinelibrary.com). DOI: 10.1002/uog.14678
How can we measure bladder volumes in women with advanced pelvic organ prolapse? ´ M. ESPUNA-PONS†, ˜ J. CASSADO*, H. DI´AZ-CUERVO‡ and P. REBOLLO‡ , on behalf of the GISPEM Group *Hospital Universitari Mutua Terrassa, Terrassa, Spain; †Hospital Clinic, Universidad de Barcelona, Barcelona, Spain; ‡LA-SER Analytica, Oviedo, Spain
K E Y W O R D S: bladder; pelvic organ prolapse; ultrasound imaging
ABSTRACT Objectives To compare bladder volumes determined by three different formulae using measurements obtained from two-dimensional translabial ultrasound (2D-US), with true bladder volumes, in women with advanced pelvic organ prolapse (POP). Methods This was a prospective observational multicenter study of consecutive women on the waiting list for prolapse surgery in 24 gynecology departments. All women had a symptomatic genital prolapse Stage 2 or higher according to the Pelvic Organ Prolapse Quantification System (POP-Q). Bladder volumes were calculated before and after spontaneous voiding by 2D-US, and true bladder volumes were determined by micturition and catheterization. Volumes determined by US were calculated using three formulae (Haylen, Dietz and Dicuio). Correlation was calculated between the volume determined by US measurement before micturition and the true volume, and also between the volume determined by US measurements after micturition and the true volume. Correlations (Spearman’s rho) and concordance (intraclass correlation coefficient (ICC)) were estimated for each of the three formulae considered. Results One-hundred and eighty-six women with POP were included in the study. A total of 349 bladder volumes (186 before micturition and 163 after micturition) were obtained. Good correlation (rho, 0.818–0.849) and concordance (ICC, 0.827–0.898) were found between total measured volume (volume of spontaneous bladder voiding + volume obtained from catheterization) and the volume determined by US using the three different formulae, as well as between the post-void residual volume measured by catheterization and the post-void volume calculated by US using the three formulae (rho, 0.739–0.777; ICC, 0.840–0.877).
Conclusions Bladder volumes in women with advanced POP can be measured easily by 2D-US. Volumes determined using the three different formulae show good correlations and concordance with true bladder volume. Copyright © 2014 ISUOG. Published by John Wiley & Sons Ltd.
INTRODUCTION The basic diagnostic preoperative assessment of women with advanced pelvic organ prolapse (POP) commonly includes determination of the degree of prolapse (on anterior, posterior and middle compartments), a stress test with or without POP reduction and measurement of the post-void residual urine volume (PVR). There are two circumstances in which it is important to determine bladder volume: when performing the stress test and for determination of PVR. The cough stress test is usually performed at a fixed bladder volume (usually 300 mL) or at the maximum cystometric capacity if the test is performed during a urodynamic examination. Both methods require urethral catheterization to fill the bladder. If the evaluation procedure does not include a urodynamic study and the diagnosis of stress incontinence is based exclusively on the clinical assessment, the result of the stress test is key for defining treatment. If no facilities are available to fill the bladder during the pelvic examination, the stress test can be performed with spontaneous bladder filling, asking the patient to cough when she feels she has a full bladder. To prevent false negatives as a result of low bladder volume, bladder volume must be assessed before the stress test. This can be done easily if an ultrasound (US) device with a transvaginal or translabial/perineal transducer is available at the outpatient clinic. Determination of PVR is another important part of the assessment of women with POP. Evaluation of bladder
´ Hospital Universitari Mutua Terrassa, Terrassa, Spain (e-mail: [email protected]) Correspondence to: Dr J. Cassado, Accepted: 19 September 2014
Copyright © 2014 ISUOG. Published by John Wiley & Sons Ltd.
ORIGINAL PAPER
234
Cassado´ et al.
volume after spontaneous micturition can be assessed by catheterization or by US (transabdominal1 , transvaginal2 or perineal/ translabial3 ). Different formulae have been used to determine bladder volumes, using bladder diameters measured by two-dimensional (2D) US. The aim of the present study was to evaluate, specifically in women with advanced POP in whom anatomical distortion is present, the performance of the currently available formulae for the estimation of bladder volume in order to assess their applicability in usual practice, both to guarantee an adequate setting for the performance of stress tests without the need for bladder filling and to assess PVR without the need for catheterization.
METHODS This was an observational and multicenter prospective study on consecutive women on the waiting list for prolapse surgery. The study involved the participation of 24 specialized pelvic floor units in Spain that are ´ en disfunciones de part of the Grupo de Investigacion Suelo P´elvicoen la Mujer (GISPEM). All women included in the study were over 18 years of age and presented with symptomatic genital prolapse Stage 2 or higher according to the Pelvic Organ Prolapse Quantification System (POP-Q); those with any previous pelvic surgery were excluded. Women were recruited and data were collected during a single visit between July 2012 and March 2013. All participants were fully informed about the nature and objectives of the study before enrolment and gave their written informed consent to enter the study. The study protocol was approved by the Clinical Research Ethics Committee of Hospital Clinic i Provincial in Barcelona, Spain. At the study visit, all women underwent a urogynecological assessment, with a comprehensive history obtained, including validated symptom questionnaires and physical examination. The degree of prolapse was assessed in women with an empty bladder using the ICS-IUGA (International Continence SocietyInternational Urogynecological Association) POP-Q staging system. Bladder dimensions were measured by translabial 2D-US using 3.5–6-MHz curved-array transducers, commonly available in most gynecological and urological units. Bladder volumes were determined before and after spontaneous voiding; all women had been encouraged to attend with a full bladder. Initially, a first translabial US was performed before micturition. Immediately afterwards, the women were asked to urinate in complete privacy (without POP reduction) and the volume voided was collected and measured. Immediately after voiding, bladder dimensions were redetermined by a second translabial US, performed by the same observer, and the PVR was measured subsequently by catheterization using a disposable short catheter (12 F or 14 F). Catheterization was performed by the same observer who carried out the US examination. The true bladder volume was calculated as the sum of the spontaneous micturition volume and
Copyright © 2014 ISUOG. Published by John Wiley & Sons Ltd.
Figure 1 Two-dimensional translabial ultrasound images of the bladder (B) in sagittal (a) and transverse (b) planes, illustrating measurement of height (h), depth (d) and maximum transverse diameter (t) of the bladder. PB, pubis; U, urethra.
the volume obtained by catheterization. Bladder volumes of patients who failed to void spontaneously or who emptied their bladder completely, were excluded from the post-micturition assessment because the volume was 0 mL (empty bladder) or was the same as the premicturition volume. Maximal longitudinal diameter (height) and maximal anteroposterior diameter (depth) of the bladder were measured, in cm, on the sagittal plane of the US, viewing the entire urethra. Maximum transverse diameter of the bladder was also measured in the transverse plane (Figure 1). All measurements obtained included only the anechoic contents, avoiding the pubic bone shadow. If the prolapsed organ passed through the introitus it had to be reduced first (either by digital pressure or by the transducer), as the bladder needed to be in the vagina to avoid air interference. All measurements were obtained at rest. Bladder volumes were determined by US using the following three formulae: Haylen’s formula2 (volume (mL) = height (cm) × depth (cm) × 5.9–14.6); Dietz’s formula3 (volume (mL) = height (cm) × depth
Ultrasound Obstet Gynecol 2015; 46: 233–238.
Bladder volumes in POP
RESULTS A total of 186 women with POP underwent assessment. Average (SD) patient age was 64.2 ± 9.84 (range, 32.50–86.44) years, mean parity was 2.83 ± 1.27 (range, 0–8) and mean body mass index was 26.64 ± 3.58 (range, 18.73–35.96) kg/m2 . Figure 2 shows the distribution of women included, according to the type and degree of POP (ICS-IUGA POP-Q). A total of 349 bladder volumes were obtained (186 before micturition and 163 after micturition). Twenty-three volumes were excluded from the post-micturition analysis: seven from women who were unable to void their bladder spontaneously; and 16 from those who did so completely. The median true bladder volume measured immediately before micturition (calculated by measuring spontaneous micturition volume plus the volume obtained by catheterization) was 260 (range, 32–930) mL. The median pre-micturition volume, determined by US, was: 268.13 (range, −6.34 to 852.70) mL with Haylen’s formula; 268.35 (range, 7.84–823.20) mL with Dietz’s formula; and 176.02 (range, 3.92–1102.50) mL with Dicuio’s formula. Only 16 (9%) patients voided completely without residual urine. Forty-eight per cent of women presented PVR volumes ranging between 1 and 50 mL, and
Copyright © 2014 ISUOG. Published by John Wiley & Sons Ltd.
60 50 Women (%)
(cm) × 5.6); and Dicuio’s formula4 (volume (mL) = height (cm) × depth (cm) × transverse diameter (cm) × 0.52). Demographic and US data were recorded using an electronic case-report form and were exported to the Stata 10.0 data analysis and statistical software (StataCorp LP, College Station, TX, USA). Only patients with data for all relevant measurements (all US assessments and micturition and catheterization volumes) were included in the analysis. Continuous variables were described using means of measures of central tendency (mean and standard error, and median) and dispersion (SD, and minimum and maximum values). For categorical variables, frequencies were assessed. The number of patients under each descriptive represents the valid data for each variable; no imputation of missing data was performed. Correlation (Spearman’s rho) was calculated between the volume determined by US measurement before micturition and the true volume (voided + catheterization), and also between the volume determined by US measurements after micturition and the true volume after micturition (catheterization). Correlations were calculated for each of the three formulae considered. The intraclass correlation coefficient (ICC) was calculated for each type of volume obtained and formula used to assess the concordance of the results. Bland–Altman graphs are presented representing differences between the volumes obtained using each formula and the measured volume (pre-void and post-void) for each patient, using the means of calculated and measured values, together with the observed average agreement and the corresponding 95% limits of agreement.
235
40 30 20 10 0
Anterior
Middle
Posterior
No prolapse
2.15%
4.84%
31.18%
Grade I
2.69%
8.60%
33.87%
Grade II
17.20%
31.72%
21.51%
Grade III
54.30%
37.63%
9.14%
Grade IV
23.66%
17.20%
4.30%
Figure 2 Distribution of women in the study cohort according to type (anterior, middle or posterior) and degree (no prolapse or Grade I–IV) of pelvic organ prolapse. , no prolapse; , Grade I; , Grade II; , Grade III; , Grade IV.
27% of patients had a PVR volume of over 100 mL. The median PVR measured by catheterization was 40 (range, 1–320) mL. The median PVR determined by US was: 38.50 (range, −12.83 to 421.53) mL with Haylen’s formula; 50.40 (range, 1.68–413.95) mL with Dietz’s formula; and 15.46 (range, 0.05–332.64) mL with Dicuio’s formula. The complete description of US measured bladder diameters and their corresponding estimated volumes, as well as micturition measured volumes and their subsequent estimated total bladder volume, are presented in Table 1. The correlation and concordance between true total bladder volume (spontaneous + catheterization) and the estimated volume determined by the three different formulae, as well as the correlation between the volume measured by catheterization and the post-void volume determined by the three different formulae, are shown in Table 2. Bland–Altman plots for each of the formulae compared with pre- and post-void measured volumes is shown in Figure 3.
DISCUSSION Sonographic imaging provides a possible alternative method of estimating bladder volume in a non-invasive manner. This study demonstrates that bladder volumes in women with advanced POP can be measured easily by translabial 2D-US. According to our results, translabial US is a non-invasive method suitable for measuring PVR in women with advanced POP and it can be performed with 3.5–6-MHz curved-array transducers, commonly used and available in most gynecological/urological units. Bladder volume can be estimated accurately with 2D-US using transabdominal, transvaginal or translabial approaches. However, both underestimation and overestimation have been reported when using 2D-US5,6 . Evidence has been presented to justify that both 2D and
Ultrasound Obstet Gynecol 2015; 46: 233–238.
Cassado´ et al.
236
Table 1 Bladder diameters in women with pelvic organ prolapse obtained on two-dimensional translabial ultrasound (US) and calculated volumes using predefined formulae, compared with bladder volumes determined by spontaneous voiding and catheterization Pre-micturition (n = 186) Variable US measurement Longitudinal diameter (height, in cm) Anteroposterior diameter (depth, in cm) Transverse diameter (cm) Calculated volume Haylen’s formula (mL) Dicuio’s formula (mL) Dietz’s formula (mL) True bladder volume Spontaneous voiding (mL) Catheterization (mL) Total volume (mL)
Post-micturition* (n = 163)
Mean ± SD
Median (range)
Mean ± SD
Median (range)
8.21 ± 2.29 5.76 ± 2.07 7.87 ± 2.19
8.35 (2.30 to 15.00) 5.65 (0.50 to 12.00) 8.07 (2.00 to 15.00)
4.03 ± 2.11 2.72 ± 1.66 4.20 ± 2.02
3.89 (0.50 to 11.20) 2.40 (0.20 to 7.90) 4.00 (0.20 to 10.38)
279.54 ± 157.88 211.40 ± 156.87 279.19 ± 149.85
268.13 (−6.34 to 852.70) 176.02 (3.92 to 1102.50) 268.35 (7.84 to 823.20)
62.70 ± 82.66 35.75 ± 51.45 73.37 ± 78.46
38.50 (−12.83 to 421.53) 15.46 (0.05 to 332.64) 50.40 (1.68 to 413.95)
235.22 ± 162.12 62.12 ± 69.25 297.34 ± 176.56
200 (0 to 900) 35 (0 to 320) 260 (32 to 930)
— 64.52 ± 67.00 —
— 40 (1 to 320) —
*Twenty-three women who were either unable to micturate (n = 7) or did so completely (n = 16) were excluded. Table 2 Correlation between pre-void and post-void true bladder volumes and volumes estimated using three different formulae, using intraclass correlation coefficient (ICC) and Spearman’s rho Pre-micturition volumes (n = 186)
Post-micturition volumes (n = 163)*
Formula
ICC (95% CI)
rho
P
ICC (95% CI)
rho
P
Haylen Dietz Dicuio
0.898 (0.864–0.924) 0.895 (0.858–0.922) 0.827 (0.454–0.921)
0.849 0.849 0.818
< 0.001 < 0.001 < 0.001
0.876 (0.831–0.909) 0.877 (0.831–0.910) 0.840 (0.513–0.926)
0.739 0.739 0.777
< 0.001 < 0.001 < 0.001
*Twenty-three women who were either unable to micturate (n = 7) or did so completely (n = 16) were excluded.
three-dimensional (3D) US devices overestimate bladder volumes at lower fillings (bladder volume < 160 mL), and underestimate volumes at higher fillings7 . These authors also found that the 2D device had better reproducibility, particularly at lower bladder volumes. In general, errors are estimated in around 25% of the voided volume measurements, and the reasons for overestimation or underestimation include difficulties in measuring the height and depth of the bladder because of its irregular shape and being unable to view the entire bladder in a single scan. In women with pelvic floor dysfunction, transvaginal or translabial 2D-US are easy to perform and the results compare favorably with those obtained by abdominal US5,7 . The most commonly used formula for use with transvaginal US is Haylen’s formula2 . However, this formula has the limitation of yielding negative results at low bladder volumes. A new formula was published recently, by Dietz et al.3 , designed for determining PVR by translabial US. The measurement of a third dimension (transverse diameter) and calculation with a specific formula was used by Dicuio, but no significant improvement in accuracy was found by this method4 . Hwang et al. proposed an algorithm that takes into account the anatomical variants of the population (sex and bladder anatomy, e.g. diverticula) but it is not suitable for cases of advanced prolapse8 . In women with advanced POP, anatomical distortion may alter US measurements as a result of paravaginal defects9 . A defect in the lateral suspension of the bladder could lead to an enlarged transverse diameter, and thus
Copyright © 2014 ISUOG. Published by John Wiley & Sons Ltd.
volume estimation using the available formulae may not be accurate enough to be used in clinical practice for these women. On the other hand, that the bladder content is anechogenic and the bladder structure deformable could also imply that there should be no calculation differences with regard to patients without prolapse when the bladder is pushed and relocated ‘in situ’. Taking into account that patients with advanced POP usually present high post-void residual volume10 and that there was no evidence to guarantee a good correlation between US and catheterization in patients with advanced POP, a prospective study was necessary in patients with advanced prolapse to ascertain the correlation between US and catheterization measurements. In this study, we assessed the performance of Haylen’s, Dietz’s and Dicuio’s formulae, specifically in women with advanced POP. We found that the formula proposed by Haylen yields slightly lower correlations than those originally reported by the authors in pre-micturition volumes (rho, 0.849 vs 0.94) and in PVR (rho, 0.739 vs 0.94), respectively2 . We also found negative values for low bladder volumes. With Dietz’s formula, we found very similar correlations and avoided obtaining negative values in patients with very low bladder volume; the strength of our correlations was also lower than those published previously for pre-micturition volumes (rho, 0.849 vs 0.96) and for PVR (rho, 0.739 vs 0.96)3 . Using Dicuio’s formula, we found a slightly lower linear correlation for pre-micturition volumes (rho, 0.818) and a higher
Ultrasound Obstet Gynecol 2015; 46: 233–238.
Bladder volumes in POP (a)
237 400
400
Difference
Difference
200 0
200
0
–200 –200
–400 0
200
600
400
800
0
1000
100
Mean (b)
300
400
300
400
300
400
400
400
Difference
200 Difference
200 Mean
0
200
0
–200 –200
–400 0
200
600
400
800
0
1000
100
(c)
200 Mean
Mean 400
400
Difference
Difference
200 0 –200
200
0
–400 –200 0
200
400
600
800
1000
Mean
0
100
200 Mean
Figure 3 Bland–Altman plots showing agreement of bladder volumes obtained using Haylen’s (a), Dietz’s (b) or Dicuio’s (c) formula with pre-void (left; volumes obtained after spontaneous voiding + catheterization volume) and post-void (right; catheterization volume) measured , observed average difference; , 95% limits of agreement. volumes.
correlation for post-micturition residuals (small volumes) (rho, 0.777) compared with the other formulae. We thought, however, that measuring concordance between the measured and estimated volumes was a more adequate approach to evaluate the performance of the different formulae. Using ICC, we found that the performance of all three formulae was good, with very similar performances for Haylen’s and Dietz’s formulae and slightly worse results for Dicuio’s formula, further pointing to the suitability of using translabial 2D-US to estimate total and PVR bladder volumes in women with advanced POP. We also presented the most accurate analysis for study agreement, in this case Bland–Altman graphs. These corroborate the adequacy and similar performances of Haylen’s and Dietz’s formulae, and we observed a tendency for underestimation using Dicuio’s formula, for both pre-void and post-void volumes.
Copyright © 2014 ISUOG. Published by John Wiley & Sons Ltd.
The originality of our study compared with those first describing the different formulae2 – 4 lies in the fact that the target population were specifically women with symptomatic advanced prolapse, not the general population of a urogynecology unit. Another contribution is the wide range of volumes tested (with desire to void and post-void), allowing identification of the formula(e) that best adapt to both small and large bladder volumes. A limitation of this study was possible delay in catheterization after the US assessment of bladder volume, even if the investigators were instructed to minimize this source of error. A second limitation may be associated with larger bladder volumes, in which imaging of the cranioventral aspects of the bladder may be impaired by acoustic shadowing as a result of a calcified interpubic disc. This may introduce potential error, particularly in older women in whom higher post-void residual volumes
Ultrasound Obstet Gynecol 2015; 46: 233–238.
Cassado´ et al.
238
are likely and relevant. Finally, these results were obtained in a population with advanced POP on the waiting list for surgery and may not be representative of the general female population with POP. However, the population tested is probably typical for the patient group in whom such testing is most likely to be necessary in the context of a urogynecological evaluation. In conclusion, we found that bladder volumes, both large and small, can be determined by translabial 2D-US in women with advanced POP using any of three published formulae, and all yield good correlation with actual bladder volume. Dietz’s formula avoids the estimation of negative volumes, found when using Haylen’s formula, when estimating post-void residuals.
ACKNOWLEDGMENT The authors would like to acknowledge all investigators of the GISPEM group who were involved in data collection. The study received funding from Astellas Pharma.
Copyright © 2014 ISUOG. Published by John Wiley & Sons Ltd.
REFERENCES 1. Mainprize TC, Drutz HP. Accuracy of total bladder volume and residual urine measurements: comparison between real-time ultrasonography and catheterization. Am J Obstet Gynecol 1989; 160: 1013–1016. 2. Haylen BT. Verification of the accuracy and range of transvaginal ultrasound in measuring bladder volumes in women. Br J Urol 1989; 64: 350–352. 3. Dietz HP, Velez D, Shek KL, Martin A. Determination of postvoid residual by translabial ultrasound. Int Urogynecol J 2012; 23: 1749–1752. 4. Dicuio M, Pomara G, Menchini FF, Ales V, Dahlstrand C, Morelli G. Measurements of urinary bladder volume: comparison of five ultrasound calculation methods in volunteers. Arch Ital Urol Androl 2005; 77: 60–62. 5. Marks LS, Dorey FJ, Macairan ML, Park C, deKernion JB. Three-dimensional ultrasound device for rapid determination of bladder volume. Urology 1997; 50: 341–348. 6. Riccabona M, Nelson TR, Pretorius DH, Davidson TE. Distance and volume measurement using three-dimensional ultrasonography. J Ultrasound Med 1995; 14: 881–886. 7. Schnider P, Birner P, Gendo A, Ratheiser K, Auff E. Bladder volume determination: portable 3-D versus stationary 2-D ultrasound device. Arch Phys Med Rehabil 2000; 81: 18–21. 8. Hwang JY, Byun SS, Oh SJ, Kim HC. Novel algorithm for improving accuracy of ultrasound measurement of residual urine volume according to bladder shape. Urology 2004; 64: 887–891. 9. Hosni MM, El-Feky AE, Agur WI, Khater EM. Evaluation of three different surgical approaches in repairing paravaginal support defects: a comparative trial. Arch Gynecol Obstet 2013; 288: 1341–1348. 10. Haylen BT, Lee J, Logan V, Husselbee S, Zhou J, Law M. Immediate postvoid residual volumes in women with symptoms of pelvic floor dysfunction. Obstet Gynecol 2008; 111: 1305–1312.
Ultrasound Obstet Gynecol 2015; 46: 233–238.
How can we measure bladder volumes in women with advanced pelvic organ prolapse? ´ M. ESPUNA-PONS†, ˜ J. CASSADO*, H. DI´AZ-CUERVO‡ and P. REBOLLO‡ , on behalf of the GISPEM Group *Hospital Universitari Mutua Terrassa, Terrassa, Spain; †Hospital Clinic, Universidad de Barcelona, Barcelona, Spain; ‡LA-SER Analytica, Oviedo, Spain
K E Y W O R D S: bladder; pelvic organ prolapse; ultrasound imaging
ABSTRACT Objectives To compare bladder volumes determined by three different formulae using measurements obtained from two-dimensional translabial ultrasound (2D-US), with true bladder volumes, in women with advanced pelvic organ prolapse (POP). Methods This was a prospective observational multicenter study of consecutive women on the waiting list for prolapse surgery in 24 gynecology departments. All women had a symptomatic genital prolapse Stage 2 or higher according to the Pelvic Organ Prolapse Quantification System (POP-Q). Bladder volumes were calculated before and after spontaneous voiding by 2D-US, and true bladder volumes were determined by micturition and catheterization. Volumes determined by US were calculated using three formulae (Haylen, Dietz and Dicuio). Correlation was calculated between the volume determined by US measurement before micturition and the true volume, and also between the volume determined by US measurements after micturition and the true volume. Correlations (Spearman’s rho) and concordance (intraclass correlation coefficient (ICC)) were estimated for each of the three formulae considered. Results One-hundred and eighty-six women with POP were included in the study. A total of 349 bladder volumes (186 before micturition and 163 after micturition) were obtained. Good correlation (rho, 0.818–0.849) and concordance (ICC, 0.827–0.898) were found between total measured volume (volume of spontaneous bladder voiding + volume obtained from catheterization) and the volume determined by US using the three different formulae, as well as between the post-void residual volume measured by catheterization and the post-void volume calculated by US using the three formulae (rho, 0.739–0.777; ICC, 0.840–0.877).
Conclusions Bladder volumes in women with advanced POP can be measured easily by 2D-US. Volumes determined using the three different formulae show good correlations and concordance with true bladder volume. Copyright © 2014 ISUOG. Published by John Wiley & Sons Ltd.
INTRODUCTION The basic diagnostic preoperative assessment of women with advanced pelvic organ prolapse (POP) commonly includes determination of the degree of prolapse (on anterior, posterior and middle compartments), a stress test with or without POP reduction and measurement of the post-void residual urine volume (PVR). There are two circumstances in which it is important to determine bladder volume: when performing the stress test and for determination of PVR. The cough stress test is usually performed at a fixed bladder volume (usually 300 mL) or at the maximum cystometric capacity if the test is performed during a urodynamic examination. Both methods require urethral catheterization to fill the bladder. If the evaluation procedure does not include a urodynamic study and the diagnosis of stress incontinence is based exclusively on the clinical assessment, the result of the stress test is key for defining treatment. If no facilities are available to fill the bladder during the pelvic examination, the stress test can be performed with spontaneous bladder filling, asking the patient to cough when she feels she has a full bladder. To prevent false negatives as a result of low bladder volume, bladder volume must be assessed before the stress test. This can be done easily if an ultrasound (US) device with a transvaginal or translabial/perineal transducer is available at the outpatient clinic. Determination of PVR is another important part of the assessment of women with POP. Evaluation of bladder
´ Hospital Universitari Mutua Terrassa, Terrassa, Spain (e-mail: [email protected]) Correspondence to: Dr J. Cassado, Accepted: 19 September 2014
Copyright © 2014 ISUOG. Published by John Wiley & Sons Ltd.
ORIGINAL PAPER
234
Cassado´ et al.
volume after spontaneous micturition can be assessed by catheterization or by US (transabdominal1 , transvaginal2 or perineal/ translabial3 ). Different formulae have been used to determine bladder volumes, using bladder diameters measured by two-dimensional (2D) US. The aim of the present study was to evaluate, specifically in women with advanced POP in whom anatomical distortion is present, the performance of the currently available formulae for the estimation of bladder volume in order to assess their applicability in usual practice, both to guarantee an adequate setting for the performance of stress tests without the need for bladder filling and to assess PVR without the need for catheterization.
METHODS This was an observational and multicenter prospective study on consecutive women on the waiting list for prolapse surgery. The study involved the participation of 24 specialized pelvic floor units in Spain that are ´ en disfunciones de part of the Grupo de Investigacion Suelo P´elvicoen la Mujer (GISPEM). All women included in the study were over 18 years of age and presented with symptomatic genital prolapse Stage 2 or higher according to the Pelvic Organ Prolapse Quantification System (POP-Q); those with any previous pelvic surgery were excluded. Women were recruited and data were collected during a single visit between July 2012 and March 2013. All participants were fully informed about the nature and objectives of the study before enrolment and gave their written informed consent to enter the study. The study protocol was approved by the Clinical Research Ethics Committee of Hospital Clinic i Provincial in Barcelona, Spain. At the study visit, all women underwent a urogynecological assessment, with a comprehensive history obtained, including validated symptom questionnaires and physical examination. The degree of prolapse was assessed in women with an empty bladder using the ICS-IUGA (International Continence SocietyInternational Urogynecological Association) POP-Q staging system. Bladder dimensions were measured by translabial 2D-US using 3.5–6-MHz curved-array transducers, commonly available in most gynecological and urological units. Bladder volumes were determined before and after spontaneous voiding; all women had been encouraged to attend with a full bladder. Initially, a first translabial US was performed before micturition. Immediately afterwards, the women were asked to urinate in complete privacy (without POP reduction) and the volume voided was collected and measured. Immediately after voiding, bladder dimensions were redetermined by a second translabial US, performed by the same observer, and the PVR was measured subsequently by catheterization using a disposable short catheter (12 F or 14 F). Catheterization was performed by the same observer who carried out the US examination. The true bladder volume was calculated as the sum of the spontaneous micturition volume and
Copyright © 2014 ISUOG. Published by John Wiley & Sons Ltd.
Figure 1 Two-dimensional translabial ultrasound images of the bladder (B) in sagittal (a) and transverse (b) planes, illustrating measurement of height (h), depth (d) and maximum transverse diameter (t) of the bladder. PB, pubis; U, urethra.
the volume obtained by catheterization. Bladder volumes of patients who failed to void spontaneously or who emptied their bladder completely, were excluded from the post-micturition assessment because the volume was 0 mL (empty bladder) or was the same as the premicturition volume. Maximal longitudinal diameter (height) and maximal anteroposterior diameter (depth) of the bladder were measured, in cm, on the sagittal plane of the US, viewing the entire urethra. Maximum transverse diameter of the bladder was also measured in the transverse plane (Figure 1). All measurements obtained included only the anechoic contents, avoiding the pubic bone shadow. If the prolapsed organ passed through the introitus it had to be reduced first (either by digital pressure or by the transducer), as the bladder needed to be in the vagina to avoid air interference. All measurements were obtained at rest. Bladder volumes were determined by US using the following three formulae: Haylen’s formula2 (volume (mL) = height (cm) × depth (cm) × 5.9–14.6); Dietz’s formula3 (volume (mL) = height (cm) × depth
Ultrasound Obstet Gynecol 2015; 46: 233–238.
Bladder volumes in POP
RESULTS A total of 186 women with POP underwent assessment. Average (SD) patient age was 64.2 ± 9.84 (range, 32.50–86.44) years, mean parity was 2.83 ± 1.27 (range, 0–8) and mean body mass index was 26.64 ± 3.58 (range, 18.73–35.96) kg/m2 . Figure 2 shows the distribution of women included, according to the type and degree of POP (ICS-IUGA POP-Q). A total of 349 bladder volumes were obtained (186 before micturition and 163 after micturition). Twenty-three volumes were excluded from the post-micturition analysis: seven from women who were unable to void their bladder spontaneously; and 16 from those who did so completely. The median true bladder volume measured immediately before micturition (calculated by measuring spontaneous micturition volume plus the volume obtained by catheterization) was 260 (range, 32–930) mL. The median pre-micturition volume, determined by US, was: 268.13 (range, −6.34 to 852.70) mL with Haylen’s formula; 268.35 (range, 7.84–823.20) mL with Dietz’s formula; and 176.02 (range, 3.92–1102.50) mL with Dicuio’s formula. Only 16 (9%) patients voided completely without residual urine. Forty-eight per cent of women presented PVR volumes ranging between 1 and 50 mL, and
Copyright © 2014 ISUOG. Published by John Wiley & Sons Ltd.
60 50 Women (%)
(cm) × 5.6); and Dicuio’s formula4 (volume (mL) = height (cm) × depth (cm) × transverse diameter (cm) × 0.52). Demographic and US data were recorded using an electronic case-report form and were exported to the Stata 10.0 data analysis and statistical software (StataCorp LP, College Station, TX, USA). Only patients with data for all relevant measurements (all US assessments and micturition and catheterization volumes) were included in the analysis. Continuous variables were described using means of measures of central tendency (mean and standard error, and median) and dispersion (SD, and minimum and maximum values). For categorical variables, frequencies were assessed. The number of patients under each descriptive represents the valid data for each variable; no imputation of missing data was performed. Correlation (Spearman’s rho) was calculated between the volume determined by US measurement before micturition and the true volume (voided + catheterization), and also between the volume determined by US measurements after micturition and the true volume after micturition (catheterization). Correlations were calculated for each of the three formulae considered. The intraclass correlation coefficient (ICC) was calculated for each type of volume obtained and formula used to assess the concordance of the results. Bland–Altman graphs are presented representing differences between the volumes obtained using each formula and the measured volume (pre-void and post-void) for each patient, using the means of calculated and measured values, together with the observed average agreement and the corresponding 95% limits of agreement.
235
40 30 20 10 0
Anterior
Middle
Posterior
No prolapse
2.15%
4.84%
31.18%
Grade I
2.69%
8.60%
33.87%
Grade II
17.20%
31.72%
21.51%
Grade III
54.30%
37.63%
9.14%
Grade IV
23.66%
17.20%
4.30%
Figure 2 Distribution of women in the study cohort according to type (anterior, middle or posterior) and degree (no prolapse or Grade I–IV) of pelvic organ prolapse. , no prolapse; , Grade I; , Grade II; , Grade III; , Grade IV.
27% of patients had a PVR volume of over 100 mL. The median PVR measured by catheterization was 40 (range, 1–320) mL. The median PVR determined by US was: 38.50 (range, −12.83 to 421.53) mL with Haylen’s formula; 50.40 (range, 1.68–413.95) mL with Dietz’s formula; and 15.46 (range, 0.05–332.64) mL with Dicuio’s formula. The complete description of US measured bladder diameters and their corresponding estimated volumes, as well as micturition measured volumes and their subsequent estimated total bladder volume, are presented in Table 1. The correlation and concordance between true total bladder volume (spontaneous + catheterization) and the estimated volume determined by the three different formulae, as well as the correlation between the volume measured by catheterization and the post-void volume determined by the three different formulae, are shown in Table 2. Bland–Altman plots for each of the formulae compared with pre- and post-void measured volumes is shown in Figure 3.
DISCUSSION Sonographic imaging provides a possible alternative method of estimating bladder volume in a non-invasive manner. This study demonstrates that bladder volumes in women with advanced POP can be measured easily by translabial 2D-US. According to our results, translabial US is a non-invasive method suitable for measuring PVR in women with advanced POP and it can be performed with 3.5–6-MHz curved-array transducers, commonly used and available in most gynecological/urological units. Bladder volume can be estimated accurately with 2D-US using transabdominal, transvaginal or translabial approaches. However, both underestimation and overestimation have been reported when using 2D-US5,6 . Evidence has been presented to justify that both 2D and
Ultrasound Obstet Gynecol 2015; 46: 233–238.
Cassado´ et al.
236
Table 1 Bladder diameters in women with pelvic organ prolapse obtained on two-dimensional translabial ultrasound (US) and calculated volumes using predefined formulae, compared with bladder volumes determined by spontaneous voiding and catheterization Pre-micturition (n = 186) Variable US measurement Longitudinal diameter (height, in cm) Anteroposterior diameter (depth, in cm) Transverse diameter (cm) Calculated volume Haylen’s formula (mL) Dicuio’s formula (mL) Dietz’s formula (mL) True bladder volume Spontaneous voiding (mL) Catheterization (mL) Total volume (mL)
Post-micturition* (n = 163)
Mean ± SD
Median (range)
Mean ± SD
Median (range)
8.21 ± 2.29 5.76 ± 2.07 7.87 ± 2.19
8.35 (2.30 to 15.00) 5.65 (0.50 to 12.00) 8.07 (2.00 to 15.00)
4.03 ± 2.11 2.72 ± 1.66 4.20 ± 2.02
3.89 (0.50 to 11.20) 2.40 (0.20 to 7.90) 4.00 (0.20 to 10.38)
279.54 ± 157.88 211.40 ± 156.87 279.19 ± 149.85
268.13 (−6.34 to 852.70) 176.02 (3.92 to 1102.50) 268.35 (7.84 to 823.20)
62.70 ± 82.66 35.75 ± 51.45 73.37 ± 78.46
38.50 (−12.83 to 421.53) 15.46 (0.05 to 332.64) 50.40 (1.68 to 413.95)
235.22 ± 162.12 62.12 ± 69.25 297.34 ± 176.56
200 (0 to 900) 35 (0 to 320) 260 (32 to 930)
— 64.52 ± 67.00 —
— 40 (1 to 320) —
*Twenty-three women who were either unable to micturate (n = 7) or did so completely (n = 16) were excluded. Table 2 Correlation between pre-void and post-void true bladder volumes and volumes estimated using three different formulae, using intraclass correlation coefficient (ICC) and Spearman’s rho Pre-micturition volumes (n = 186)
Post-micturition volumes (n = 163)*
Formula
ICC (95% CI)
rho
P
ICC (95% CI)
rho
P
Haylen Dietz Dicuio
0.898 (0.864–0.924) 0.895 (0.858–0.922) 0.827 (0.454–0.921)
0.849 0.849 0.818
< 0.001 < 0.001 < 0.001
0.876 (0.831–0.909) 0.877 (0.831–0.910) 0.840 (0.513–0.926)
0.739 0.739 0.777
< 0.001 < 0.001 < 0.001
*Twenty-three women who were either unable to micturate (n = 7) or did so completely (n = 16) were excluded.
three-dimensional (3D) US devices overestimate bladder volumes at lower fillings (bladder volume < 160 mL), and underestimate volumes at higher fillings7 . These authors also found that the 2D device had better reproducibility, particularly at lower bladder volumes. In general, errors are estimated in around 25% of the voided volume measurements, and the reasons for overestimation or underestimation include difficulties in measuring the height and depth of the bladder because of its irregular shape and being unable to view the entire bladder in a single scan. In women with pelvic floor dysfunction, transvaginal or translabial 2D-US are easy to perform and the results compare favorably with those obtained by abdominal US5,7 . The most commonly used formula for use with transvaginal US is Haylen’s formula2 . However, this formula has the limitation of yielding negative results at low bladder volumes. A new formula was published recently, by Dietz et al.3 , designed for determining PVR by translabial US. The measurement of a third dimension (transverse diameter) and calculation with a specific formula was used by Dicuio, but no significant improvement in accuracy was found by this method4 . Hwang et al. proposed an algorithm that takes into account the anatomical variants of the population (sex and bladder anatomy, e.g. diverticula) but it is not suitable for cases of advanced prolapse8 . In women with advanced POP, anatomical distortion may alter US measurements as a result of paravaginal defects9 . A defect in the lateral suspension of the bladder could lead to an enlarged transverse diameter, and thus
Copyright © 2014 ISUOG. Published by John Wiley & Sons Ltd.
volume estimation using the available formulae may not be accurate enough to be used in clinical practice for these women. On the other hand, that the bladder content is anechogenic and the bladder structure deformable could also imply that there should be no calculation differences with regard to patients without prolapse when the bladder is pushed and relocated ‘in situ’. Taking into account that patients with advanced POP usually present high post-void residual volume10 and that there was no evidence to guarantee a good correlation between US and catheterization in patients with advanced POP, a prospective study was necessary in patients with advanced prolapse to ascertain the correlation between US and catheterization measurements. In this study, we assessed the performance of Haylen’s, Dietz’s and Dicuio’s formulae, specifically in women with advanced POP. We found that the formula proposed by Haylen yields slightly lower correlations than those originally reported by the authors in pre-micturition volumes (rho, 0.849 vs 0.94) and in PVR (rho, 0.739 vs 0.94), respectively2 . We also found negative values for low bladder volumes. With Dietz’s formula, we found very similar correlations and avoided obtaining negative values in patients with very low bladder volume; the strength of our correlations was also lower than those published previously for pre-micturition volumes (rho, 0.849 vs 0.96) and for PVR (rho, 0.739 vs 0.96)3 . Using Dicuio’s formula, we found a slightly lower linear correlation for pre-micturition volumes (rho, 0.818) and a higher
Ultrasound Obstet Gynecol 2015; 46: 233–238.
Bladder volumes in POP (a)
237 400
400
Difference
Difference
200 0
200
0
–200 –200
–400 0
200
600
400
800
0
1000
100
Mean (b)
300
400
300
400
300
400
400
400
Difference
200 Difference
200 Mean
0
200
0
–200 –200
–400 0
200
600
400
800
0
1000
100
(c)
200 Mean
Mean 400
400
Difference
Difference
200 0 –200
200
0
–400 –200 0
200
400
600
800
1000
Mean
0
100
200 Mean
Figure 3 Bland–Altman plots showing agreement of bladder volumes obtained using Haylen’s (a), Dietz’s (b) or Dicuio’s (c) formula with pre-void (left; volumes obtained after spontaneous voiding + catheterization volume) and post-void (right; catheterization volume) measured , observed average difference; , 95% limits of agreement. volumes.
correlation for post-micturition residuals (small volumes) (rho, 0.777) compared with the other formulae. We thought, however, that measuring concordance between the measured and estimated volumes was a more adequate approach to evaluate the performance of the different formulae. Using ICC, we found that the performance of all three formulae was good, with very similar performances for Haylen’s and Dietz’s formulae and slightly worse results for Dicuio’s formula, further pointing to the suitability of using translabial 2D-US to estimate total and PVR bladder volumes in women with advanced POP. We also presented the most accurate analysis for study agreement, in this case Bland–Altman graphs. These corroborate the adequacy and similar performances of Haylen’s and Dietz’s formulae, and we observed a tendency for underestimation using Dicuio’s formula, for both pre-void and post-void volumes.
Copyright © 2014 ISUOG. Published by John Wiley & Sons Ltd.
The originality of our study compared with those first describing the different formulae2 – 4 lies in the fact that the target population were specifically women with symptomatic advanced prolapse, not the general population of a urogynecology unit. Another contribution is the wide range of volumes tested (with desire to void and post-void), allowing identification of the formula(e) that best adapt to both small and large bladder volumes. A limitation of this study was possible delay in catheterization after the US assessment of bladder volume, even if the investigators were instructed to minimize this source of error. A second limitation may be associated with larger bladder volumes, in which imaging of the cranioventral aspects of the bladder may be impaired by acoustic shadowing as a result of a calcified interpubic disc. This may introduce potential error, particularly in older women in whom higher post-void residual volumes
Ultrasound Obstet Gynecol 2015; 46: 233–238.
Cassado´ et al.
238
are likely and relevant. Finally, these results were obtained in a population with advanced POP on the waiting list for surgery and may not be representative of the general female population with POP. However, the population tested is probably typical for the patient group in whom such testing is most likely to be necessary in the context of a urogynecological evaluation. In conclusion, we found that bladder volumes, both large and small, can be determined by translabial 2D-US in women with advanced POP using any of three published formulae, and all yield good correlation with actual bladder volume. Dietz’s formula avoids the estimation of negative volumes, found when using Haylen’s formula, when estimating post-void residuals.
ACKNOWLEDGMENT The authors would like to acknowledge all investigators of the GISPEM group who were involved in data collection. The study received funding from Astellas Pharma.
Copyright © 2014 ISUOG. Published by John Wiley & Sons Ltd.
REFERENCES 1. Mainprize TC, Drutz HP. Accuracy of total bladder volume and residual urine measurements: comparison between real-time ultrasonography and catheterization. Am J Obstet Gynecol 1989; 160: 1013–1016. 2. Haylen BT. Verification of the accuracy and range of transvaginal ultrasound in measuring bladder volumes in women. Br J Urol 1989; 64: 350–352. 3. Dietz HP, Velez D, Shek KL, Martin A. Determination of postvoid residual by translabial ultrasound. Int Urogynecol J 2012; 23: 1749–1752. 4. Dicuio M, Pomara G, Menchini FF, Ales V, Dahlstrand C, Morelli G. Measurements of urinary bladder volume: comparison of five ultrasound calculation methods in volunteers. Arch Ital Urol Androl 2005; 77: 60–62. 5. Marks LS, Dorey FJ, Macairan ML, Park C, deKernion JB. Three-dimensional ultrasound device for rapid determination of bladder volume. Urology 1997; 50: 341–348. 6. Riccabona M, Nelson TR, Pretorius DH, Davidson TE. Distance and volume measurement using three-dimensional ultrasonography. J Ultrasound Med 1995; 14: 881–886. 7. Schnider P, Birner P, Gendo A, Ratheiser K, Auff E. Bladder volume determination: portable 3-D versus stationary 2-D ultrasound device. Arch Phys Med Rehabil 2000; 81: 18–21. 8. Hwang JY, Byun SS, Oh SJ, Kim HC. Novel algorithm for improving accuracy of ultrasound measurement of residual urine volume according to bladder shape. Urology 2004; 64: 887–891. 9. Hosni MM, El-Feky AE, Agur WI, Khater EM. Evaluation of three different surgical approaches in repairing paravaginal support defects: a comparative trial. Arch Gynecol Obstet 2013; 288: 1341–1348. 10. Haylen BT, Lee J, Logan V, Husselbee S, Zhou J, Law M. Immediate postvoid residual volumes in women with symptoms of pelvic floor dysfunction. Obstet Gynecol 2008; 111: 1305–1312.
Ultrasound Obstet Gynecol 2015; 46: 233–238.
