Has the true prevalence of voiding difficulty in urogynecology patients been underestimated?
- First Online:
- Cite this article as:
- Haylen, B.T., Krishnan, S., Schulz, S. et al. Int Urogynecol J (2007) 18: 53. doi:10.1007/s00192-006-0094-x
- 88 Views
Voiding difficulty has been relatively overlooked as a diagnosis. Previous estimates of its prevalence have generally been no more than 14% with one exception at 24%. The aim of this study is to determine the true prevalence and associations of voiding difficulty using a validated definition [urine flow rate under 10th centile of the Liverpool Nomograms and/or residual urine volume (by transvaginal ultrasound) more than 30 ml]. This study involved 592 women referred for an initial urogynecological assessment including urodynamics. Data were separated according to the presence or absence of voiding difficulty. The prevalence of voiding difficulty was 39%, far higher than previous estimates. It is the third most common urodynamic diagnosis behind urodynamic stress incontinence (USI—72%) and uterine/vaginal prolapse (61%) and ahead of the overactive bladder (13%). Voiding difficulty significantly increased in prevalence with age and increasing grades of all types of uterine/vaginal prolapse. Prolapse appeared to be the main factor in the age deterioration. Other significant positive relationships with voiding difficulty were prior hysterectomy and prior continence surgery, whilst USI and the symptom and sign of stress incontinence had significant inverse relationships.
KeywordsUrinary incontinenceUrodynamicsVoiding difficultyUterine/vaginal prolapse
The condition of voiding difficulty is defined as abnormally slow and/or incomplete micturition . Voiding difficulty has been relatively overlooked as a diagnosis in comparison with urodynamic stress incontinence (USI), overactive bladder (OAB) and uterine and/or vaginal prolapse. The presence of voiding difficulty may impact on the management of the first two of the other three conditions.
There has been no analysis of a large cohort of urogynecology patients specifically targeting voiding difficulty and all its relations. Previous estimates for the prevalence of voiding difficulty amongst women with lower urinary tract symptoms vary between 6.5 and 14% [2–4]. Recently, Costantini et al., in a study on uroflowmetry in female voiding disturbances, estimated the prevalence of voiding difficulty in a urogynecological population at 24% .
The diagnosis of voiding difficulty requires the presence of an abnormal urine flow rate and/or an abnormal residual urine volume (RUV) . Effective levels of normality for both parameters have already been determined. In the case of urine flow rates, the Liverpool nomograms  provide clear normal ranges for maximum and average urine flow rates in women over a wide range of voided volumes, allowing for the strong dependency of urine flow rates on voided volumes. The 10th centile has been determined by two [7, 8] studies as the most sensitive and specific lower limit of normal. By definition then, as a minimum in the following study, 10% of symptomatic women must have voiding difficulty due to an abnormal urine flow rate alone.
Costantini et al.  do acknowledge the presence of an alternate “arbitrary” set of criteria to diagnose voiding difficulty: maximum urine flow rate less than 15 ml/s and/or RUV greater than 50 ml, with a minimum total bladder volume of 150 ml before voiding . Neither of these criteria have been validated. This minimum total bladder volume might significantly limit useful data as the voided volume for best “free” uroflowmetry averaged only 144 ml in one series .
RUVs in normal women have been determined  and compared with those in symptomatic (urogynecology patients) women . An upper limit of normal of 30 ml has been derived, with the RUVs measured by vaginal ultrasound. Under this level lie 95% of RUVs in asymptomatic women  and 87%  of RUVs in symptomatic women. A RUV more than 30 ml has long been considered to have the potential for causing recurrent UTI . This finding was re-confirmed recently in a large study .
This prospective study is then a systematic analysis of a large cohort of urogynecology patients specifically targeting voiding difficulty and all its relations.
Materials and methods
The study involved 592 women, all of whom were referred for an initial urogynecological assessment including urodynamics, because of symptoms of lower urinary tract dysfunction. After a comprehensive history, a full examination was performed. Initial testing for clinical stress leakage was by coughing in the supine and standing positions with the presenting bladder volume. Prolapse assessment followed voiding, to eliminate the effect of a full bladder restricting the full extent of the prolapse . Assessments were made for the presence and grade of any uterine prolapse, anterior vaginal wall (cystocoele), posterior vaginal wall (rectocoele) and vaginal vault (enterocoele) prolapse. The patients were examined in the left lateral position using a Sims’ speculum and at maximal strain. Grades 1 and 2 were equivalent to the respective International Continence Society (ICS) gradings ; ICS grades 3 and 4 are combined as grade 3 for tabulation.
Each woman, having been encouraged to attend for the assessment with a comfortably full bladder, voided for free uroflowmetry in complete privacy over a Dantec Urodyn 1000 uroflowmeter. The maximum and average urine flow rates and the respective voided volumes were recorded. Residual urine volumes were measured by transvaginal ultrasound (Aloka UST-657-5 probe) generally within 60 s of voiding [15, 16]. Where the RUV was more than 175 ml (the upper limit of the transvaginal ultrasound technique), measurement of RUV was by urethral catheterisation using a 14 Fr short plastic catheter. The initial residual was used for subsequent analysis, although where the initial residual was more than 10 ml, the patient voided for a second time over a conventional toilet (rather than a uroflowmetry commode) for repeat void residual urine volume analysis. The lowest value was taken as the final residual.
Further urodynamics testing included multichannel filling and provocative subtraction cystometry, followed by voiding cystometry.
Urine flow rates need to be corrected to allow for their strong dependency on voided volume . Each urine flow rate and the respective voided volume were, thus, applied to the equation form of the Liverpool Nomograms  to obtain a centile ranking. The flow data from 94 women fell outside the range of interpretation of the Liverpool Nomograms (voided volumes 15–600 ml), leaving the data from 498 urogynecology patients in the study.
Voiding difficulty is abbreviated as VD. Groups VD-1 (maximum urine flow rate) and VD-2 (average urine flow rate) were created where the respective urine flow rate data were under the 10th centile of the equivalent Liverpool Nomogram and/or RUV was more than 30 ml (some women were unable to void for free uroflowmetry). Groups NON VD-1 (maximum urine flow rate) and NON VD-2 (average urine flow rate) were created where the respective urine flow rate data were 10th centile or more on the equivalent Liverpool Nomogram and RUV was 30 ml or under.
The prevalences of voiding difficulty for both the maximum (VD-1) and average (VD-2) urine flow rate data were determined. Relations were sought for age, parity, presenting symptoms, prior hysterectomy, prior continence surgery, menopause/ hormone replacement therapy (HRT) use, recurrent urinary tract infections (UTI—defined as two or more urinary tract infections in the previous 12 months), all grades of the different types of prolapse as well as all the final diagnoses (after completion of all urodynamics studies including cystometry).
The statistical significance of differences between urine flow rates in the different groups were assessed using non-parametric tests (Wilcoxon or Kruskal–Wallis tests as appropriate) All terminology was in line with that prescribed by the International Continence Society .
The median age for the entire group was 57 (range 16–98). The median age for the women with voiding difficulty was 65 (range 27–90).
The prevalence of voiding difficulty was identical at 39% in both VD-1 and VD-2 groups. The prevalence for the other diagnoses were USI (72%), prolapse (61%) and OAB (13%).
Symptoms of voiding difficulty (p<0.001) and prolapse (p<0.001) were significantly linked to VD-1, VD-2 with the symptom (p<0.001) and sign (p<0.001) of stress incontinence significantly linked to NON VD-1, NON VD-2 (absence of voiding difficulty).
There was a significant relationship of voiding difficulty to prior hysterectomy (VD-1, p=0.006; VD-2, p=0.053) and prior continence surgery (VD-1, p=0.008; VD-2, p=0.010) respectively. There was a greater increase in VD-1, VD-2 with vaginal hysterectomy compared with abdominal hysterectomy, but the difference did not reach significance. There was more VD-1, VD-2 in women with more than one prior continence procedure compared with just one prior surgery, but due to small numbers, significance was again not reached.
Data initially showed that menopausal women on combined oestrogen/progesterone therapy had significantly less voiding difficulty than those not taking HRT. However, this effect was lost when age was taken into account.
There is a significant relationship of voiding difficulty to increasing grades of uterine prolapse, anterior vaginal wall (cystocoele), posterior vaginal wall (rectocoele) and vaginal vault (enterocoele), respectively, (p<0.001 for VD-1 and VD-2 for all types of prolapse). The relationship between the retroverted uterus and voiding difficulty was non-significant (p=0.250).
There was no significant relationship between voiding difficulty [VD-1 (p=0.213), VD-2 (p=0.213)] and recurrent UTI.
There were significant relationships of VD-1 and VD-2 to the final diagnoses of prolapse (p<0.001) and voiding difficulty (p<0.001) (this diagnosis incorporates additional cystometry data) with a strong inverse relation to the diagnosis of USI (p=0.001). There was a lesser inverse relationship of voiding difficulty to the diagnosis of detrusor overactivity [VD1 (<0.139); VD2 (<0.021)].
Women presenting for urogynecological assessment have a 39% prevalence of voiding difficulty. This is much more frequent than previously determined. It is the third most common diagnosis, after USI (72%) and prolapse (61%) and ahead of OAB (13%). Failure to allow for voiding difficulty in the management of USI and/or OAB may lead to the risk of exacerbating the voiding difficulty with the possibility of acute or chronic retention.
There is a significant increase in voiding difficulty with age (Fig. 1). In a previous article , prolapse, prior hysterectomy and lower urinary tract changes were thought to be the main reasons for this increase. In the present study, prolapse appeared to be the most important factor (Fig. 2).
The proposed mechanism of genital prolapse is the distortional or kinking effect of the prolapse on the urethra to create a bladder outflow obstruction. Pressure-flow studies necessary to prove this point are difficult owing to the obstructive nature of the pressure catheter . A review  summarises the results of four studies by stating that women with high-grade genital prolapse have increased urethral closure pressure and pressure/transmission ratios, which decrease after the prolapse is reduced. It can be a clear aim of prolapse surgery to reduce or eliminate the RUV with its accompanying contribution to a diagnosis of voiding difficulty. It has yet to be conclusively proven that prolapse surgery is effective in this task.
The previous finding that prior hysterectomy might have contributed to higher incidence of voiding difficulty  is confirmed in the current study. One probable mechanism for this is nerve dysfunction . There was no significant difference between the vaginal and abdominal hysterectomy. Prior continence surgery is an additional significant association of voiding difficulty.
The inverse relationship between VD-1, VD-2 and the symptom of stress incontinence, the sign of clinical leakage and the final diagnosis of USI, is reassuring. This finding was also seen in Costantini’s  series. USI is the main urodynamic diagnosis present in 427 (72%) women in this series. Effective minimally invasive surgical techniques are available to cure USI should the severity of symptoms so warrant.  It would appear that candidates for surgical management of USI are less likely to have pre-existing voiding difficulty.
Voiding difficulty as a diagnosis can be easily made with a non-invasive voiding study, involving free uroflowmetry and RUV measurement by ultrasound, all lasting only a few minutes. Suggestions that “the best method of analysing voiding function quantitatively is the pressure-flow study of micturition, with simultaneous recording of abdominal, intravesical and detrusor pressures and flow rate”  are inappropriate in the first instance of screening. High failure to void rates of 41% can occur during voiding cystometry .
The voiding study has simple and clear parameters for which the normal and abnormal values have been clearly outlined. This study highlights that, because of the high frequency with which it occurs, voiding difficulty can no longer be the “Cinderella”  diagnosis of urogynecology. It will tend to embarrass those who ignore it.
Voiding difficulty is the third most common urodynamic diagnosis with a prevalence of 39%. It increases with age, with increasing uterine/vaginal prolapse appearing to be the most important factor. It also increases in prevalence with both prior hysterectomy and prior continence surgery. There is a significant inverse relationship between voiding difficulty and USI which implies any surgical treatment of USI is less likely to be compromised by the presence of pre-existing voiding difficulty.