NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.

National Collaborating Centre for Women's and Children's Health (UK). Urinary Incontinence: The Management of Urinary Incontinence in Women. London: RCOG Press; 2006 Oct. (NICE Clinical Guidelines, No. 40.)

  • This publication is provided for historical reference only and the information may be out of date.

This publication is provided for historical reference only and the information may be out of date.

Cover of Urinary Incontinence

Urinary Incontinence: The Management of Urinary Incontinence in Women.

Show details

3Assessment and investigation

3.1. Introduction

Initial assessment when a woman comes into first contact with a health professional is important. It forms the basis for counselling, ongoing management and treatment. Categorisation of UI by symptom profile may direct the patient to the most appropriate and effective resources. Further evaluation of severity of the condition, and ultimately the impact of treatment on that severity, will enable the healthcare professional to deliver the optimum care. Co-existing conditions (e.g. prolapse, diabetes, heart failure) or treatments (e.g. drug therapy) must be recognised and clinicians must be aware of their interaction with UI. Investigation should be used appropriately, taking into account the nature of the condition.

Studies considered for the assessment and investigation section

For history taking and physical examination, where primary research data were not available, published consensus statements and narrative reviews that discussed these issues were used as a basis for the GDG’s statements and recommendations.31,33,34

For each investigation used in the assessment of UI in women, up to five questions were asked (refer to the assessment matrix in Appendix B):

  • Does the investigation direct the woman to an alternative pathway?
  • What is the diagnostic accuracy of the investigation?
  • What is the test–retest reliability of the investigation?
  • Does the use of the investigation affect outcomes?
  • Does the use of the investigation predict outcomes of treatment?

For questions of diagnostic accuracy, the ideal study is one that makes blind comparison of the test with a validated reference standard in a sample of women reflective of the population to whom the test would apply. Within the clinical area of UI, there is no agreement as to what the reference standard is for the diagnosis of UI and thus studies that consider accuracy are limited by the standard against which they are compared.

3.2. History taking and physical examination

History taking of women with UI or OAB guides the investigation and management by evaluating symptoms, their progression and the impact of symptoms on lifestyle. Taking a history also allows the assessment of risk factors associated with the possible diagnoses. The relevant elements of history follow.

Urinary symptoms

In order to reach a clinical diagnosis, a urinary history is taken to determine storage and voiding patterns and symptoms. The major symptoms to consider include:

  • Storage symptoms:

    frequency (daytime), nocturia, urgency, urge UI

    stress UI

    constant leakage (which may rarely indicate fistula).

  • Voiding symptoms:

    hesitancy, straining to void, poor or intermittent urinary stream.

  • Post-micturition symptoms:

    sensation of incomplete emptying, post-micturition dribbling.

Accompanying symptoms that may indicate the possibility of a more serious diagnosis and which require referral, such as haematuria, persisting bladder or urethral pain, or recurrent urinary tract infection (UTI), can also be identified when taking a urinary history.

How do urinary symptoms compare with urodynamic findings?

We found no studies in which clinical outcomes in women with UI diagnosed by clinical history alone were compared with those in women with UI diagnosed using urodynamics. However, several studies have evaluated the accuracy of the symptom of stress or urge UI relative to findings on urodynamic (UD) investigations in women undergoing assessment of their urinary symptoms. Most of these studies have been considered in two reviews and a health technology assessment of diagnostic methods for UI.46–48 Two of the publications included studies of women with symptoms of stress, mixed or urge UI46,48 and one included only studies evaluating women with stress UI.47 The reviews that included women with stress, mixed or urge UI calculated and combined sensitivity and specificity data for the symptom of stress (be it with or without mixed symptoms) and for the symptom of urge UI (be it with or without mixed symptoms). The GDG considered that the mixed ‘symptom’ should be considered separately (because in practice women are categorised into those with stress, mixed or urge UI) and that the important question in relation to the comparison of urinary history with urodynamic findings is whether urodynamics gives additional information to that obtained from the history alone. In considering this question, the GDG took the approach that a clinical history would be taken for every woman, and that a positive history for a particular type of UI would always be followed by treatment appropriate to that type of UI.

Overall, 25 relevant studies that compared the diagnosis based on history with urodynamic findings were considered by the GDG. These studies used cystometry as the reference standard for diagnosis of UI and therefore assumed that history taking had a lower diagnostic value in comparison. Fourteen studies included women with stress, mixed or urge UI, and 11 presented raw data in a way that allowed sensitivity, specificity, positive predictive value (PPV) and negative predictive values (NPV) to be calculated.49–62 Two of these studies only reported accuracy data for stress and mixed UI.61,62 Five studies only investigated how a history of urge UI or OAB compared with urodynamic findings of detrusor overactivity (DO).63–67 Six studies only investigated how a history of stress UI compared with the finding of urodynamic stress incontinence,68–73 four of which provided some but not all accuracy data.

Multichannel cystometry (with or without uroflowmetry, urethral pressure profilometry, or cystourethrography) was the urodynamic method used in 24 studies. The remaining study used single-channel cystometry for women with urge UI (and suspected DO) and multichannel cystometry for women with stress UI.50 All except four studies62–64,70 stated that terminology used for urodynamic findings conformed to ICS standards.

With the exception of one study,49 which involved primary and secondary care, all studies were conducted in secondary or tertiary care.

The GDG focused on the 11 studies that provided diagnostic accuracy data for stress, mixed and urge UI. Confidence intervals were calculated for each value, as this was considered to be more appropriate than pooling data from individual studies. Pooling the available data (by meta-analysis) or generating receiver operating characteristic curves was not considered to be appropriate because the population in each study varied in terms of the relative proportions of stress, mixed or urge UI, the methods used to obtain a history varied, and the studies were considered to be of poor quality in terms of defining diagnostic accuracy because of unblinded urodynamic testing.

The diagnostic accuracy data for the studies is summarised in Table 3.1. For further details refer to Appendix C.

Table 3.1. Summary of diagnostic accuracy data.

Table 3.1

Summary of diagnostic accuracy data.

The overall conclusions are that the available studies comparing history of stress, mixed or urge UI with findings of stress UI and/or DO on multichannel cystometry have poor internal and external validity. We consider that the NPV is of particular interest in terms of assessing whether urodynamic testing provides additional information compared with clinical history, because this quantity summarises the extent to which a negative history is associated with a negative finding on urodynamics (i.e. whether diagnosis based on urodynamics would alter the findings for women with no history of a particular type of UI). In addressing the question of whether urodynamics gives additional information to that obtained from history alone, with the limitations of the studies in mind, the following conclusions can be drawn:

  • If a woman does not report mixed UI (i.e. if she reports pure stress UI or pure urge UI), the probability of finding urodynamic stress incontinence (USI) plus DO on cystometry is small (around 10%), therefore urodynamic testing might be said to offer little additional diagnostic value. It is acknowledged that urodynamic investigation is not simply used to distinguish USI and DO, and that further information may be obtained about other elements of lower urinary tract function, such as the voiding pattern.
  • If a woman does not report pure urge UI, the probability of finding DO on cystometry is small (again around 10%), therefore urodynamic testing offers little added diagnostic value.

The situation for pure stress UI is less clear-cut. Here 15–51% (median 31%) of women who do not report pure stress UI may nevertheless be found to have USI on cystometry. However, the lack of consistency between the NPVs in the available studies together with the lack of detailed information about the method of obtaining a history and the poor quality of the studies limit the extent to which the evidence would support urodynamic testing for women who do not report stress UI. However, a limitation of dealing with stress, mixed and urge UI as three separate entities is that the analysis ignores the interdependence between the different diagnoses.

History taking is regarded as the cornerstone of assessment of UI. Current practice is that women with UI are categorised according to their symptoms into those with stress, mixed or urge UI; women with mixed UI are treated according to the symptom they report to be the most troublesome. In the absence of evidence that urodynamic testing improves the outcome of women treated conservatively (see Section 3.11), and without robust evidence that urodynamic testing provides additional valuable information to the history alone in the initial assessment of women with UI, the GDG concluded that urodynamic testing is not required before initiating conservative treatment.

Bowel symptoms

Constipation or problems with defecation may predispose to UI and adversely affect the outcome of any continence surgery. Straining can contribute to loss of bladder control by weakening pelvic floor muscles (refer to lifestyle interventions, Section 4.1). Faecal incontinence in association with UI or OAB may suggest the presence of cognitive impairment, neurological and/or anatomical damage. Women with faecal incontinence may require referral for management of that condition (a NICE guideline on faecal incontinence is in development, with an expected date of issue of June 2007).

Medical history

Conditions that may exacerbate or co-exist with UI or OAB can be identified by taking a general history and are important contributory factors to exclude. These include mental health, cognitive impairment and disorders of the:

  • neurological system (e.g. multiple sclerosis, spinal cord injury, Parkinson’s disease, cerebrovascular accident, cauda equina syndrome, pelvic plexus injury)
  • metabolic system (e.g. diabetes)
  • cardiorespiratory system
  • renal system.

Surgical history

Previous surgery for UI or for POP may complicate treatment and make diagnosis more difficult because of its interference with the normal support mechanisms of the vagina and urethra. Any surgery that might have interfered with normal nerve supply to the bladder or urethra may also be relevant; this could include low spinal surgery, radical hysterectomy, low rectal surgery, sympathectomy or complex pelvic surgery.

Obstetric and gynaecological history

The number and type of deliveries and their outcome would normally be documented. The woman’s desire for further childbearing should also be established as this may have implications for the most appropriate treatment options. The menstrual history and menopausal status should be determined, and enquiry made into symptoms of uterovaginal prolapse. The woman’s sexual function and her expectations from this point of view should also be considered.

Drug history

Some medications may be associated with UI and their use may need to be reviewed. These include drugs that affect:

  • the central nervous system, for example sedatives, hypnotics, anxiolytics and smooth muscle relaxants
  • the autonomic nervous system, for example drugs with antimuscarinic action, sympathomimetics and sympatholytics
  • fluid balance, for example diuretics and alcohol.

A drug history should consider previous medication for UI symptoms, and any known allergies, which may affect future treatment choices.

Does history taking affect outcome?

No evidence was identified that addressed this question. Nevertheless, the fundamental importance of history taking within all aspects of clinical practice cannot be overemphasised.

Test–retest reliability of history taking

No evidence was identified that addressed this question.

General assessment

Assessment of the social and functional impact of UI, desire for treatment, expectations and motivation are important as these help to establish the woman’s goals, and may influence the type and degree of intervention offered.

Social circumstances to consider include home environment, personal relationships, occupational history and lifestyle factors such as smoking and body mass index (BMI). Adjustment of these lifestyle factors may form part of the management of the condition in some women (refer to Section 4.1).

Functional assessment, which may include consideration of access and ease of use of toileting aids, mobility and dexterity, is important. Assessment of the home environment may be undertaken, for example by an occupational therapist.

Physical examination

Physical examination is carried out to guide the diagnosis and management of incontinence and the identification of any underlying, modifying or serious conditions that require treatment outside the scope of this guideline.

The assessment of cognitive impairment allows the effect of disease to be taken into account and allows modification of treatment. The Abbreviated Mental Test Score (AMTS) and the Mini Mental State Examination (MMSE) should be undertaken for women aged over 75 years with complex comorbidities. These scales should also be considered for younger women if clinically appropriate.

Abdominal examination can detect a significantly enlarged bladder or palpable pelvic mass. A palpable bladder may indicate the presence of chronic urinary retention. Palpation may detect a volume of 300 ml or more.74 Urinary incontinence may occur in association with urinary retention (often called overflow incontinence).

Pelvic assessment is important and should include vaginal examination, and possibly also rectal examination if clinically indicated. Vaginal examination can assess POP and identify atrophic changes, infection and excoriation. Uterine and ovarian enlargement may be determined by bi-manual examination. When rectal examination is undertaken, it is used to further evaluate posterior vaginal wall prolapse and, where indicated by a history of constipation, prolapse or faecal incontinence. Assessment of pelvic floor, prolapse and residual urine are considered in more detail in Sections 3.3, 3.4 and 3.6.

Neurophysiology

Neurophysiological tests include assessments of nerve conduction and electromyography (EMG). The former include sacral reflex latencies, pudendal terminal motor latencies and evoked potentials, which test the integrity of nerve pathways relating to voiding and continence. Abnormal results might indicate underlying neurological dysfunction. Electromyography tests the end organ function of somatic muscles of the pelvic floor, or sphincter complexes, but cannot be used to record activity from smooth muscle. No evidence was identified that addressed diagnostic accuracy of neurophysiological testing in relation to idiopathic UI. Where history suggests evidence of neurological disease, examination of lower limbs together with sacral sensation and sacral reflexes is required.

Evidence statements for history taking and physical examination

The reporting of stress, urge or mixed UI is commonly used to direct treatment decisions. [EL = 4]

The diagnostic value of history taking has been compared with urodynamic testing in women with UI or OAB. In general, there is a low level of agreement between a history of urinary symptoms and urodynamic findings. [EL = DS III]

However, women who do not report mixed or urge UI are unlikely to have findings of mixed UI or DO on urodynamics. [EL = DS III]

Recommendations for history taking and physical examination

At the initial clinical assessment, the woman’s UI should be categorised as stress UI, mixed UI, or urge UI/OAB. Initial treatment should be started on this basis. In mixed UI, treatment should be directed towards the predominant symptom. [D (GPP)]

The clinical assessment should seek to identify relevant predisposing and precipitating factors and other diagnoses that may require referral for additional investigation and treatment. [D (GPP)]

3.3. Pelvic floor muscle assessment

Methods used to assess pelvic floor muscle contraction include digital palpation, EMG and perineometry. For digital palpation, grading scales have been used to quantify the strength of contraction, for example the Oxford grading system.31

Does pelvic floor muscle assessment affect outcome?

In RCTs that evaluated pelvic floor muscle training (PFMT) for the treatment of UI, both those studies that did and those that did not assess pelvic floor muscle contraction prior to treatment showed efficacy of active treatment compared with control (refer to physical therapies, Section 4.2). [EL = 3] No further evidence was identified in relation to the effects of undertaking pelvic floor assessment on outcomes of women with UI.

Test–retest reliability

Two case series evaluated the test–retest reliability of grading systems for digital assessment.75,76 One assessed the test–retest and inter-rater reliability of a four-item pelvic floor muscle rating scale (covering pressure, duration of contraction and displacement; n = 37, about two-thirds of whom had urinary symptoms). Retest was done after 1–4 weeks. Significant inter- and intra-rater correlations were reported for the rating scale, and for EMG, which was also performed.75 [EL = 3] The second case series evaluated test–retest reliability of digital assessments of pelvic floor muscle power and endurance in women with UI, with power being assessed on a modified Oxford grading system. Retest was undertaken after 2–5 weeks. Significant correlations between test and retest results were reported for both parameters, with agreement seen in nine cases (n = 20).76 [EL = 3]

Two case series considered the inter-rater (not test–retest) reliability of a modified Oxford grading system. In the first study (n = 20, seven of whom had stress UI), poor inter-rater reliability was found for the grading system undertaken by physiotherapists, with agreement in nine cases (r = 0.7, Kappa score 0.37).77 [EL = 3] The second case series in women with UI compared results of three clinicians’ assessments with that of an expert clinician’s assessment. No agreement was found between one of the clinicians and the expert prior to training. After training, the percentage agreement between the three clinicians and the expert was 70%, 78% and 80% (n = 30).78 [EL = 3]

Evidence statements for pelvic floor muscle assessment

There is a lack of evidence as to whether digital assessment of pelvic floor muscle contraction affects the outcome of PFMT in women with UI. [EL = 3] Inter- and intra-observer reliability of grading systems for digital assessment of pelvic floor muscle contraction is poor. [EL = 3]

From evidence to recommendations

The GDG recognises that there is a lack of evidence for clinical utility of digital pelvic floor muscle assessment. However, expert opinion is that the determination of whether a woman can contract pelvic floor muscles will direct treatment decisions. [EL = 4]

Recommendation for pelvic floor muscle assessment

Routine digital assessment of pelvic floor muscle contraction should be undertaken before the use of supervised pelvic floor muscle training for the treatment of UI. [D (GPP)]

Research recommendation for pelvic floor muscle assessment

The role of clinical pelvic floor muscle assessment prior to PFMT should be investigated to determine whether it enhances the therapeutic effect of the intervention.

3.4. Assessment of prolapse

Several grading systems for POP have been described, including the Baden and Walker halfway method and the Pelvic organ prolapse quantification (POP-Q) system. The POP-Q system was designed to measure the type and severity of prolapse. In practice, the extent to which symptoms are bothersome is also an important factor to consider.

Women who present with symptoms of prolapse and UI should have an abdominal examination to exclude other pathology. Inspection of the genitalia may reveal the degree of uterine descent. On vaginal examination, the position of the cervix is determined and, at this point, the clinician can make an assessment of the strength of the pelvic floor. Bimanual examination will reveal any pelvic masses. The woman can be asked to cough and bear down, which may demonstrate stress UI or the presence of cystocele or rectocele.

Diagnostic accuracy

In considering the question of diagnostic accuracy in relation to POP assessment, the GDG considered that the important issue is how symptoms correlate with the prolapse grading systems available. Three cross-sectional studies considered the correlation of POP symptoms with the degree of pelvic organ support.79–81 POP-related symptoms and/or their bother factor were not significantly associated with the stage of prolapse, as defined by the POP-Q classification system in one study. However, women reported a mean of more than one symptom when the prolapse extended beyond the hiatus, compared with a mean of less than one symptom when the leading edge of the prolapse did not reach the hiatus (n = 477).79 [EL = 3] In a survey of women attending a gynaecological examination, the prevalence of POP was 31%, with 2% having prolapse that reached the introitus. Although no significant differences in the prevalence of certain symptoms (sense of heaviness in the abdomen, difficulty voiding or emptying bowel) were found in women with or without POP, cystocele or rectocele, respectively, the symptom prevalence was higher in women with prolapse (n = 487).80 The third study found that seeing or feeling a bulge was reported by all women who had prolapse beyond the hiatus. However 21% of women who had this level of prolapse did not report this symptom.81 [EL = 3]

Does assessment of prolapse affect outcome?

No evidence was identified that addressed this question.

Evidence statement for assessment of prolapse

Prolapse visible at or beyond the introitus is commonly associated with symptoms. [EL = 3]

Recommendation for assessment of prolapse

Women with UI who have symptomatic prolapse that is visible at or below the vaginal introitus should be referred to a specialist. [D (GPP)]

3.5. Urine testing

Urinalysis is used to detect infection, protein, blood and glucose in the urine. Protein may indicate infection and/or renal impairment, blood may indicate infection or malignancy, and glucose may indicate diabetes mellitus. Some findings on urine testing indicate referral (see Section 3.7).

Diagnostic accuracy of urine testing for urinary tract infection

One study evaluated the accuracy of urine reagent strips for the diagnosis of UTI in women with UI. Identification of leucocytes and/or nitrites on a reagent strip was indicative of a positive test for infection. Using urine culture as the reference standard, the reagent strips had sensitivity of 29%, specificity of 99%, and positive and negative predictive values of 82% and 92%, respectively (n = 265).82 [EL = DS II]

Does urine testing affect outcome?

No evidence was identified that addressed this question. The GDG also considered whether treating a UTI affects UI symptoms. While no evidence was found in relation to treating an infection, one study in nursing home residents considered the impact of treating bacteriuria on UI. Half the patients had bacteriuria at baseline, which was removed in 81% of patients by antibiotic treatment. Eradication of bacteriuria appeared to have no effect on incontinence (n = 191; 71% women).83 [EL = 2+]

Evidence statements for urine testing

Urine dipstick testing for leucocytes and nitrites has low sensitivity and high specificity for the diagnosis of UTI in women with UI. A negative urine dipstick test therefore excludes a UTI with a high degree of certainty. Only one-third of positive tests are associated with bacteriologically proven UTIs. [EL = DS II]

From evidence to recommendations

Although urine testing for leucocytes and nitrites has a high specificity, false negatives can arise when infection occurs with organisms that do not convert nitrates to nitrites. Hence, in symptomatic patients with a negative test, the clinician may wish to consider prescribing antibiotics once a urine sample has been collected but before results of midstream urine culture are available.

In patients with symptoms of UTI and a positive urine test for leucocytes and nitrites, infection is highly likely and immediate prescription of antibiotics is therefore justified. In single uncomplicated UTI it might be argued that this should be done without sending urine for laboratory analysis, depending on local knowledge of antibiotic resistance. In recurrent or complicated cases, however, this would not be appropriate. We therefore recommend that a midstream urine specimen be sent for culture and antibiotic sensitivities before prescribing in all cases.

Recommendations for urine testing

A urine dipstick test should be undertaken in all women presenting with UI to detect the presence of blood, glucose, protein, leucocytes and nitrites in the urine. [D (GPP)]

Women with symptoms of UTI whose urine tests positive for both leucocytes and nitrites should have a midstream urine specimen sent for culture and analysis of antibiotic sensitivities. An appropriate course of antibiotic treatment should be prescribed pending culture results. [D (GPP)]

Women with symptoms of UTI whose urine tests negative for either leucocytes or nitrites should have a midstream urine specimen sent for culture and analysis of antibiotic sensitivities. The healthcare professional should consider the prescription of antibiotics pending culture results. [D (GPP)]

Women who do not have symptoms of UTI, but whose urine tests positive for both leucocytes and nitrites, should not be offered antibiotics without the results of midstream urine culture. [D (GPP)]

Women who do not have symptoms of UTI and whose urine tests negative for either leucocytes or nitrites are unlikely to have UTI and should not have a urine sample sent for culture. [B (DS)]

3.6. Assessment of residual urine

Some findings on physical examination or from history taking in relation to emptying may indicate referral because of suspected voiding dysfunction. Abdominal examination can detect a significantly enlarged bladder, which may indicate the presence of chronic urinary retention. Palpation may detect a volume of 300 ml or more.74 Large post-void residual urine may indicate the presence of underlying bladder outlet obstruction, neurological disease or detrusor failure. These would be a reason for referral to a specialist rather than progression through a path of conservative treatments. Large residual urine – in effect chronic retention of urine – may also present with renal failure although this is much less likely in women than in men. However, there is no accepted definition for what constitutes a high or large residual volume in women with UI. Residual urine volumes may vary and thus repeat measurements may be required.

Methods used to measure post-void residual urine are abdominal palpation, ultrasound scanning and catheterisation.

Diagnostic accuracy

The diagnostic accuracy of bladder ultrasound for the measurement of post-void residual urine was evaluated in three studies, using a bladder scanner (portable bladder ultrasound).84–86 The accuracy of bimanual examination was evaluated in one study.87 Catheterisation was used as the reference standard in each. The residual urine volume indicative of a positive test result ranged from 50 to 200 ml across the studies.

Portable ultrasound (bladder scanner) versus catheterisation

Two studies enrolled men and women.84,86 The first study, in nursing home residents, reported sensitivities of 90–95% with a portable bladder ultrasound device at residual volume cut-off of less than 50 or 100 ml, and 59–69% for residual volumes of more than 100, 150, or 200 ml. Specificity values were 63–71% and 95–99%, respectively. It was not possible to calculate positive and negative predictive values from the data given (n = 201; 74% women).86 [EL = DS Ib] The second study reported sensitivities of 90% for a post-void residual of 100 ml or more, and 92% for 200 ml or more. Specificities were 88% and 83%, PPV 91% and 76%, and NPV 86% and 95%, respectively (n = 46; 74% women).84 [EL = DS Ib]

One study evaluated only women with UI. Based on a residual urine volume of 100 ml being a positive result, ultrasound had a sensitivity of 67% and specificity of 97%. It was not possible to calculate PPV or NPV from the data given (n = 95).85 [EL = DS II]

Bimanual examination versus catheterisation

The diagnostic accuracy of bimanual examination relative to catheterisation in women was assessed in one study. Based on a residual urine volume of 50 ml being a positive result, bimanual examination had a sensitivity of 14%, specificity of 67%, PPV 7% and NPV 82% (n = 47).87 [EL = DS Ib]

Does assessment of residual urine affect outcome?

No evidence was identified that addressed this question.

Test–retest reliability

One study reported test–retest reliability of portable ultrasound scanner measurements, with good correlation reported for both observers (r = 98 for one [187 pairs measured] and r = 97 for the second [143 pairs]).86 [EL = DS Ib]

Evidence statements for assessment of residual urine

The sensitivity and specificity of ultrasound (using a bladder scanner) in the detection of post-void residual urine volume, in comparison with catheterisation, is within clinically acceptable limits. [EL = DS II] The former is less invasive with fewer adverse effects. [EL = 4] The sensitivity of bimanual examination to detect small post-void residual volumes is poor. [EL = DS Ib]

From evidence to recommendations

The GDG considers that the lack of evidence for what constitutes a clinically significant residual volume in women with UI precludes making a recommendation other than in women who have signs or symptoms suggestive of voiding dysfunction.

Recommendations for residual urine

The measurement of post-void residual volume by bladder scan or catheterisation should be performed in women with symptoms suggestive of voiding dysfunction or recurrent UTI. [B (DS)]

A bladder scan should be used in preference to catheterisation on the grounds of acceptability and lower incidence of adverse events. [D (GPP)]

Women who are found to have a palpable bladder on bimanual or abdominal examination after voiding should be referred to a specialist. [D (GPP)]

3.7. Referral

Evidence statement

Certain signs or symptoms on assessment indicate referral for further investigation by an appropriate specialist. The NICE guideline on referral for suspected cancer covers some indications for referral that are relevant to this guideline.23 [EL = 4]

From evidence to recommendations

There are other women for whom referral for further advice or specialist intervention may be considered, because of either co-existent conditions or a history of prior interventions. The GDG recognises that not all such women may wish to be referred. However, referral should be considered in these cases. Owing to variations in service configuration, it is not possible to state to which service or healthcare professional women should be referred. Similarly, timescales are not specified for referral priorities because NICE recommends that trusts should work to local definitions of maximum waiting times. [EL = 4]

Recommendations for referral

Women with UI who have any of the following should receive an urgent* referral:

  • microscopic haematuria in women aged 50 years and older
  • visible haematuria
  • recurrent or persisting UTI associated with haematuria in women aged 40 years and older
  • suspected malignant mass arising from the urinary tract.

Indications for referral are:

  • symptomatic prolapse that is visible at or below the vaginal introitus (Section 3.4)
  • the finding of a palpable bladder on bimanual or abdominal examination after voiding (Section 3.6). [D (GPP)]

In women with UI, further indications for consideration for referral to a specialist service include:

  • persisting bladder or urethral pain
  • clinically benign pelvic masses
  • associated faecal incontinence
  • suspected neurological disease
  • symptoms of voiding difficulty
  • suspected urogenital fistulae
  • previous continence surgery
  • previous pelvic cancer surgery
  • previous pelvic radiation therapy
*

NICE’s Referral Guidelines for Suspected Cancer (www​.nice.org.uk/CG027) define urgent referral as the patient being seen within the national target for urgent referrals (currently 2 weeks).

Haematuria visible with the aid of a microscope.

3.8. Symptom scoring and quality of life assessment

Symptom and quality of life scoring is used to give some quantification of the impact of urinary symptoms and provides a measure that can be used to assess outcomes of treatment at a later stage. The International Consultation on Incontinence (ICI) uses three grades of recommendation for symptom scoring and QOL scales, based on the evidence available to support their use, as listed below:31

  • highly recommended – validity, reliability and responsiveness established with rigour in several data sets (‘Grade A’), or in one dataset with UI (‘Grade Anew’)
  • recommended (‘Grade B’) – validity, reliability and responsiveness indicated but not with rigour; validity and reliability established with rigour in several data sets
  • with potential (‘Grade C’) – questionnaires in early development.

The test–retest reliability of ICI ‘Grade A’ or ‘Grade Anew’ condition-specific scales for use in women are considered within this guideline because these are the questionnaires validated to the highest level. These are:31

Test–retest reliability of symptom scoring and quality of life scales

Test–retest reliability was generally reported within validation studies for these questionnaires, although in the case of CONTILIFE,88 and the short forms of UDI and IIQ (UDI-6, IIQ-7),89 no test–retest reliability data were presented. The available studies generally quoted correlation of test and retest findings, with some publications reporting actual scores or differences in test–retest scores. [EL = 3]

The results are shown in Table 3.2. For ICIQ, BFLUTS, I-QOL, SUIQQ, UISS, SEAPI-QMM, ISI and KHQ, either significant agreement or significant correlation between test and retest scores was reported (correlation was assessed using Spearman’s, Pearson’s, Cronbach’s alpha, or the intraclass correlation coefficients). Few studies reported the actual test and retest scores. The available evidence for UDI, IIQ and OAB-q showed either a significant difference between test and retest scores and/or poor correlation between test and retest scores. [EL = 3]

Table 3.2. Test–retest reliability of quality of life and symptom scoring scales.

Table 3.2

Test–retest reliability of quality of life and symptom scoring scales.

Evidence statement for symptom scoring and quality of life

The test–retest reliability of ICIQ, BFLUTS, I-QOL, SUIQQ, UISS, SEAPI-QMM, ISI and KHQ is good. For other scores, the evidence is weak or absent. [EL = 3]

Recommendation for symptom scoring and quality of life

The following incontinence-specific quality of life scales are recommended when therapies are being evaluated: ICIQ, BFLUTS, I-QOL, SUIQQ, UISS, SEAPI-QMM, ISI and KHQ. [D (GPP)]

3.9. Bladder diaries

Bladder diaries are used to document each cycle of filling and voiding over a number of days and can provide information about urinary frequency, urgency, diurnal and nocturnal cycles, functional bladder capacity and total urine output. They also record leakage episodes, fluid intake and pad changes and give an indication of the severity of wetness. They may also be used for monitoring the effects of treatment.

Test–retest reliability of bladder diaries

Five studies evaluated the test–retest reliability of bladder diaries or frequency–volume charts. One study evaluated a 1 day diary,103 one a 3 day diary,104 and three studies evaluated a 7 day diary.105–107 Data captured in the charts were: frequency and voided volume;103 frequency, leakage and urgency episodes, and voided volume;104 frequency and leakage episodes;105,106 and leakage episodes only.107

1 day diary

A case–control study reported some reproducibility data for the 1 day charts in women with stress UI. The 95% limits of agreement between the first and second days of measurement lay between 0.5 and 2.1 for frequency, and for total, mean and largest single voided volume measures (n = 80).103 [EL = 3]

3 day diary

In a case series of men and women with stress, mixed or urge UI, or OAB, the retest reliability of a 72 hour bladder diary and pad test was evaluated after an interval of 1 week (n = 106; 84% women). The authors’ predefined minimum correlation coefficient for test–retest reliability of 0.7 was met for overall frequency, day frequency, leakage episodes, urgency episodes and mean voided volume (correlation coefficient 0.70–0.87), but not for night frequency (0.605). Correlation, but not actual results, was also reported for 24 and 48 hour results; the correlation appeared to improve at 72 hours compared with 24 hours for urgency episodes. Compliance with recording voided volume data fell with time.104 [EL = 3]

Diaries of 7 days or longer

A case series evaluated the reproducibility of leakage episode and frequency data from a 7 day diary repeated after 4 weeks in women with urodynamic stress UI (n = 138). Minimal differences in test–retest results were seen for leakage episodes (mean 1.7/week, r = 0.906) and frequency (mean 0.03/(24 hours), r = 0.831). Correlation coefficients for the first 3 and last 4 days of the diary were also reported, which were 0.887 for leakage episodes and 0.908 for frequency.105 [EL = 3]

A second case series reported test–retest variability and correlations of leakage episodes, and diurnal and nocturnal frequency, based on a 7 day diary in women aged 55 years or over (n = 50; 68% USI, 32% DO with or without stress UI). The test–retest reliability was reported to be significant for the three parameters, with correlation coefficients of 0.86–0.91. No significant differences were seen between any test and retest results.106 [EL = 3]

The third case series investigated the reliability of a 14 day diary for the measurement of leakage episodes in women. Significant correlation was reported between the diary findings of weeks one and two, with 5 days’ recording being necessary for internal consistency, for women with predominant urge UI and 7 days for women with predominant stress UI (n = 214).107 [EL = 3]

Does the use of bladder diaries affect outcome?

A considerable placebo effect has been reported in many placebo-controlled trials evaluating the effectiveness of conservative interventions for the treatment of UI or OAB, with this placebo effect being reflected in self-reported changes in voiding pattern, using bladder diaries. This placebo effect usually decreases over time. Several investigators suggest that completion of these diaries, together with the close monitoring, placebo medication, and therapeutic attention via interaction between them and their patients, induces a bladder training/retraining effect. However, none of the trials were designed to examine whether bladder diaries affect outcomes of women with UI or OAB, and therefore the conclusion that the diaries are the cause of a placebo effect is not proven. In addition, sufficient duration will be needed to allow the initial ‘beneficial’ placebo response to run its course so that true effects of interventions can be observed.

Evidence statements for bladder diaries

Bladder diaries are a reliable method of quantifying urinary frequency and incontinence episodes. [EL = 3] They are useful as a measure of outcome of treatments. The optimum duration of bladder diaries is unclear. [EL = 4]

Recommendation for bladder diaries

Bladder diaries should be used in the initial assessment of women with UI or OAB. Women should be encouraged to complete a minimum of 3 days of the diary covering variations in their usual activities, such as both working and leisure days. [D (GPP)]

3.10. Pad testing

Pad tests are used to detect and quantify urine loss. Pad tests of varying duration have been evaluated: 1 hour or less, 24 hours, and 48 hours or longer.

Does pad testing affect outcome?

No studies were identified that addressed this question.

Test–retest reliability

Short pad tests

Fourteen studies considered the test–retest reliability of pad testing: eight studies considered short pad tests (1 hour or less), and six considered tests of longer duration (24–72 hours). The data reported generally were correlation of test and retest findings, with some publications reporting actual scores or differences in test–retest scores. Correlation was assessed using Spearman’s, Pearson’s or Lin’s concordance correlation coefficients in five studies; the remaining nine studies did not state which test was used.

The results for the short pad tests are shown in Table 3.3. One study considered the reliability of a test of 12–15 minutes’ duration.108 The standardised 1 hour pad test was used in four studies.109–113 Another three studies used a modified version where the test differed in the method of filling the bladder, or in the quantity of fluid instilled or consumed.114–116 Across these studies, significant correlation or agreement was reported between test and retest urine loss, although the differences between test and retest results across studies varied widely (means of 2–23 g).

Table 3.3. Test–retest reliability of short pad tests.

Table 3.3

Test–retest reliability of short pad tests.

24, 48 and 72 hour pad tests

Six case series reported test–retest reliability for pad tests of 24–72 hours.104,117–121 The results are shown in Table 3.4. These studies also reported significant correlation between test and retest results. One study noted that the number of pad-test days required for optimal reliability was 3 days (data only shown in graph in publication).119 Another found that correlation appeared to improve at 72 hours compared with 24 hours, but compliance fell.104 [EL = 3]

Table 3.4. Test–retest reliability of pad tests of at least 24 hours’ duration.

Table 3.4

Test–retest reliability of pad tests of at least 24 hours’ duration.

Evidence statements for pad testing

The evidence supporting the use of pad testing is contradictory and of poor quality. However, there appears to be good group correlation with test–retesting even though the amounts leaked may differ in individuals. It is possible that pad tests of longer duration (24 hours or longer) are more sensitive for measuring incontinence than a short standardised 1 hour test. There is a lack of evidence in relation to whether pad testing in the assessment of women with UI affects outcomes. [EL = 3]

While there is no evidence of diagnostic value or clinical utility for pad testing, the GDG’s view is that pad tests may be useful for evaluating therapies for incontinence. [EL = 4]

Recommendation for pad testing

Pad tests are not recommended in the routine assessment of women with UI. [D]

3.11. Urodynamic testing

The term ‘urodynamics’ encompasses a number of varied physiological tests, of bladder and urethral function, which aim to demonstrate an underlying abnormality of storage or voiding. The term is often used loosely to mean multichannel cystometry.

Cystometry is the measurement of intravesical pressure, which can be carried out through a single recording channel (simple cystometry) or, more commonly, by multichannel cystometry, which involves the synchronous measurement of both bladder and intra-abdominal pressures by means of catheters inserted into the bladder and the rectum or vagina. The aim is to replicate the woman’s symptoms by filling the bladder and observing pressures changes or leakage caused by provocation tests.

Uroflowmetry entails a free-flow void into a recording device, which provides the practitioner with information about the volume of urine passed and the rate of urine flow.

There are also numerous tests of urethral function, including urethral pressure profilometry and leak point pressure measurement. These are used to derive values that reflect the ability of the urethra to resist urine flow, expressed most commonly as maximum urethral closure pressure (MUCP), or as abdominal, cough or Valsalva leak point pressures (ALPP, CLPP, VLPP).

Videourodynamics involves synchronous radiographic screening of the bladder with multichannel cystometry and is so called because originally the information was recorded to videotape. Ambulatory urodynamics involves multichannel cystometry carried out with physiological bladder filling rates and using portable recording devices, which enable the woman to remain ambulant during the test.

Diagnostic accuracy

The agreement between urinary history and urodynamic findings is considered in Section 3.2.

Does urodynamic testing affect outcome?

A systematic review considering this question has been published but it only included one fully published RCT.122 The RCT was considered alongside other relevant, fully published data.123

The RCT compared conservative treatment tailored to urodynamic findings (PFMT or bladder retraining) with a multicomponent conservative treatment regimen (PFMT and bladder retraining), without prior urodynamic investigation in women with symptoms of UI or OAB (n = 60; 48 analysed). After 3 months of treatment, no significant differences were seen between groups in any outcome (leakage episodes, frequency, nocturia, subjective assessment, short pad test). Since the uninvestigated trial arm received both treatments, the outcomes from treatment may be uninformative about the value of urodynamics.123 [EL = 1−]

Two observational studies reported outcomes of continence surgery for stress UI in women who had preoperative urodynamic investigations (61% or 71%), compared with those who did not. Neither found significant differences between groups in cure/success rates or symptom severity scores at mean follow-up of 25 months (n = 109)124 [EL = 2−] or at 1 year (n = 279).125,126 [EL = 2+]

Do preoperative urodynamic findings predict post-surgical outcomes?

In a few case series of surgical interventions for stress UI in women, authors retrospectively explored whether certain preoperative findings on urodynamic testing (urethral pressure profilometry or uroflowmetry) predicted surgical success or complications. These studies cover a range of surgical procedures (suspension procedures, slings, intramural bulking agents), with some studies evaluating more than one procedure. Data were presented in various ways, with some studies comparing the mean urethral pressures in successful or failed groups and others considering the success or failure rates above certain urethral pressure thresholds. Generally, small numbers of women were involved in the individual studies (n range 45–375, with most studies including fewer than 100 women). Cure rates were reported at varying durations of follow-up, ranging from 3 to 26 months (most under 1 year). The impact of possible confounding factors was not generally considered within the study reports. The studies were neither designed nor powered to show differences in the outcomes evaluated. [EL = 3]

The majority of the identified studies considered whether urethral closure pressures or leak point pressures predicted success of surgical procedures.127–139 Other studies considered whether urine flow rates or urethral closure pressures predicted complications (voiding dysfunction or de novo DO).129,140–144

Success

Five of eight case series reported that the urethral closure pressure was statistically significantly lower in women who failed surgery (colposuspension, tension-free vaginal tape [TVT], vaginal wall sling), or that the failure rate was higher in women with maximum urethral closure pressure (MUCP) of 20 cmH2O or less.127,131–133,139 Two case series found no significant association between preoperative MUCP and surgical success or failure.128,129 One case series reported varying results according to the procedure undertaken, with mean MUCP significantly lower in women who failed colporrhaphy, with no differences found for MUCP in women who had successful or unsuccessful needle suspension or colposuspension procedures.134

A further two studies considered different pressure measurements. One reported that preoperative opening detrusor pressure and urethral pressure at closure were significantly lower in women who had objective failure of colposuspension.129 Another reported that the preoperative ‘index of urethral relaxation at stress’ (ratio of highest intraurethral pressure between coughs in the stress urethral pressure profilometry to the MUCP at rest) in women undergoing a suspension procedure was significantly lower in cases of objective failure.128

Four studies considered success according to abdominal or Valsalva leak point pressures (VLPP). Three reported no difference in success rates after surgery (vaginal wall sling, polypropylene sling, polytetrafluoroethylene bulking) according to baseline leak point pressures.130,131,135 One of these studies reported that the failure rate was significantly higher in women with both VLPP less than 50 cmH2O and MUCP less than 30 cmH2O, compared with both values above these thresholds.131 The fourth case series reported a significantly lower cure rate with TVT in women with low VLPP (less than 60 cmH2O).138

Complications

The preoperative maximum urine flow rate was significantly lower in women who had delayed voiding in three of four studies that considered this.140,142,144 One reported that a maximum flow rate of less than 20 ml/second was associated with delayed voiding.140 The third study found no significant association between preoperative peak urine flow rate or residual volume and delayed voiding.141 The procedures undertaken were a fascia lata sling, TVT and colposuspension.

No association was found between preoperative MUCP or VLPP values and voiding dysfunction in the studies that considered this.140,142,143

Two studies considered factors associated with the development of de novo DO, following colposuspension. Preoperative opening detrusor pressure, urethral pressure at closure and acceleration of flow rate were significantly higher in women with de novo DO in one study, while preoperative MUCP was not found to be associated with de novo DO (n = 209).129 The second study did not report an association between uroflowmetry and de novo DO (n = 77).144 [EL = 3]

Different methods of urodynamic investigation

Single-channel versus multichannel cystometry

The findings of single-channel (‘simple’) and multichannel cystometry were compared in four studies (n range 70–179).145–148 Two studies included elderly men and women, with data reported separately for women.145,146 The tests were conducted on the same day in three studies146–148 (in random order in one147) and after an interval of 1–4 weeks in the fourth study.145 Assessments or interpretation of the traces were performed blind in two of the studies146,147 [EL = DS II] but not in the other two.145,148 [EL = DS III] The diagnostic accuracy results, for simple compared with multichannel cystometry, for a diagnosis of DO (or detrusor hyperreflexia145) in the four studies were: sensitivity range 59–100%, specificity 68–89%, PPV 17–84%, NPV 79–100%.

A fifth study reported accuracy of single-channel cystometry with cough stress test, relative to multichannel cystometry, for a diagnosis of stress UI. The sensitivity, specificity, PPV and NPVs found were 84%, 84%, 87% and 81%, respectively (n = 145).149 [EL = DS III]

Stress test versus multichannel urodynamics

Three studies evaluated the accuracy of a simple stress test for a diagnosis of stress UI.150–152 In the first, the stress test was conducted with a fixed bladder volume (which requires catheterisation). Compared with videocystometry and leak point pressure findings, the stress test had sensitivity of 94%, specificity 90%, PPV 97% and NPV 82%.150 [EL = DS II] In the second study, the stress test was conducted with an empty bladder and found sensitivity of 49%, specificity 95%, PPV 98% and NPV 29% compared with multichannel cystometry. These values changed to 65%, 76%, 66% and 76%, respectively, when the stress test was compared with MUCP of 20 cmH2O or less.151 [EL = DS III]

The third study compared the accuracy of urethral closure pressure profilometry during multi-channel cystometry for a diagnosis of stress UI, relative to a diagnosis based on a clinical stress test. Urethral closure pressure profilometry had sensitivity of 93%, specificity of 83%, PPV 92% and NPV 86% (n = 981).152 [EL = DS III]

Ambulatory versus conventional multichannel urodynamics

Six case series compared ambulatory urodynamics with conventional multichannel cystometry or videocystometry (n range 20–22).153–158 In three studies, the populations evaluated were those in whom a diagnosis had not been reached on conventional cystometry or whose symptoms did not match the cystometric findings.153,156,158 Three of the studies included men and women, the majority being women.153,156,157

The studies differed in the duration of ambulatory monitoring (3–24 hours) and in the interval between tests (from 1 week to a mean of 37 weeks in those that reported this). The studies considered agreement between the two methods with none reporting data in a way that allows calculation of sensitivity, specificity, PPV or NPV. Two studies reported the agreement for any type of UI,153,154 and four for urge UI or DO.155–158

The studies reported the following:

  • 63% had additional findings on ambulatory urodynamics153
  • significant difference in the proportions with DO or with normal findings on ambulatory versus conventional urodynamic testing154
  • more patients were found to have DO on ambulatory than conventional urodynamic testing155–158 [EL = 3]

Videocystourethrography versus other methods

Videocystourethrography (VCU) was compared with multichannel cystometry in one study (n = 159). VCU had sensitivity of 61%, specificity 70%, PPV 56% and NPV 74% for a diagnosis of stress UI, and 14%, 97%, 87% and 45%, respectively, for urge UI.159 [EL = DS III]

Compared with clinical assessment (n = 37), the accuracy of VCU was: 160

  • sensitivity of 74%, specificity 78%, PPV 78% and NPV 74% for stress UI
  • 0, 91%, 0 and 91%, respectively, for mixed UI (zero sensitivity and PPV because no women had both a clinical and urodynamic finding of mixed UI)
  • 50%, 89%, 20% and 97%, respectively, for urge UI.160 [EL = DS III]

No studies were identified that compared the accuracy of leak point pressures with MUCP for the diagnosis of intrinsic sphincter deficiency.

Test–retest reliability of urodynamic testing

One case series evaluated the intra- and inter-observer reliability of voiding measurements (pressure flow parameters) in women. Repeat cystometry was done after 1 week. Differences in intra- and inter-rater findings for the parameters measured (opening and closure detrusor pressure, maximum flow rate, detrusor pressure at maximum flow rate) were reported to be small although no statistical analysis was reported (n = 554).161 [EL = 3]

No studies were identified in relation to the test–retest reliability of the filling phase of cystometry.

In a series of men and women with OAB, cystometry was performed at baseline and repeated after 2–4 weeks’ placebo treatment within an RCT. All parameters (volume at first desire to void, volume at first involuntary contraction, and maximum pressure of involuntary contraction) increased significantly at the second measurement, and therefore it seems this study evaluated the effects of placebo on cystometric parameters rather than reproducibility (n = 30; 40% women).162 [EL = 3]

Health economics of urodynamic testing

Resource scarcity provides the rationale for undertaking any health economic analysis. Finite resources mean that expenditure on preoperative urodynamic testing, or anything else for that matter, carries an opportunity cost – that is, other possible uses of those resources and benefits from them are foregone. The efficiency issue is then whether that expenditure represents the best use of those scarce resources: could greater patient benefit be obtained if the resources used for preoperative urodynamic testing were employed elsewhere? The health economics of preoperative urodynamic testing is especially important to consider because its impact on outcomes has been questioned and yet it currently represents routine clinical practice, using actual NHS resources. Therefore, an economic analysis of preoperative urodynamic testing in women who failed conservative treatment has been undertaken for this guideline and the details are given in Appendix D.

Evidence statements for urodynamic testing

There is often inconsistency between the clinical history and the urodynamic findings. [EL = DS III] Multichannel cystometry, when it reproduces the woman’s symptoms, may reveal the underlying pathophysiological explanation of incontinence. Single-channel cystometry is less reliable, although a simple clinical stress test may be as accurate as multichannel cystometry in the diagnosis of stress UI. [EL = 3] Although videocystourethrography has the benefit of simultaneous structural and functional assessment, it is not clear whether this adds any relevant diagnostic accuracy, compared with multichannel cystometry. Ambulatory monitoring demonstrates functional abnormalities more often than multichannel cystometry, but the significance of this is unclear. [EL = 3]

There is no evidence that pretreatment multichannel cystometry will improve the outcomes of treatments for incontinence. [EL = 2−] Although some urodynamic parameters have been found to correlate with adverse outcomes of surgery such as voiding difficulty and OAB, no test has been shown to reliably predict beneficial or adverse outcomes of surgery. [EL = 3] Nevertheless, it is recognised that preoperative urodynamic testing is firmly established in clinical practice and widely believed to contribute to improved patient counselling with regard to the likely outcomes of surgery. [EL = 4]

Economic modelling shows the cost effectiveness of preoperative urodynamic testing to be highly sensitive to the proportion of women with pure stress incontinence who failed conservative treatment, but this is not clearly established.

From evidence to recommendations

The GDG considers that urodynamic testing does not assist in the assessment of a woman prior to conservative treatment. The GDG also maintains the view that urodynamics investigation is not essential in every woman prior to primary surgery for stress UI, and therefore is not routinely recommended.

While the evidence suggests that preoperative urodynamics are not necessary for women with pure stress incontinence, the GDG accepts that these tests may help where the clinical diagnosis is not clear or in those women where initial surgical therapy has failed. Complex reconstructive urological procedures such as augmentation cystoplasty have been developed for use in specific urodynamic abnormalities; they should only be undertaken where these abnormalities are shown to be present.

Recommendations for urodynamic testing

The use of multichannel cystometry, ambulatory urodynamics or videourodynamics is not recommended before starting conservative treatment. [D]

For the small group of women with a clearly defined clinical diagnosis of pure stress UI, the use of multichannel cystometry is not routinely recommended. [D]

Multichannel filling and voiding cystometry is recommended in women before surgery for UI if:

  • there is clinical suspicion of detrusor overactivity, or
  • there has been previous surgery for stress incontinence or anterior compartment prolapse, or
  • there are symptoms suggestive of voiding dysfunction.

Ambulatory urodynamics or videourodynamics may also be considered in these circumstances. [D (GPP)]

Research recommendation for urodynamic testing

Further research is needed to answer the question of whether the use of urodynamics, prior to initial or subsequent treatments, affects the outcomes and cost effectiveness of interventions in women with UI or OAB.

3.12. Other tests of urethral competence

Other than urodynamic studies, the Q-tip, POP-Q, Bonney, Marshall and Fluid-Bridge tests can assess urethral competence (hypermobility of the urethrovesical junction). The Q-tip test involves placing a sterile Q-tip in the urethra and the woman is asked to bear down. If the Q-tip moves more than 303, the test is considered positive. The Bonney and Marshall tests involve pressing either the index and middle finger of the examiner’s hand (Bonney test) or the jaws of a forceps (Marshall test) against the anterior vaginal wall, without pressing on the urethra. The stress provocation test is repeated and if no leakage occurs the Bonney or Marshall test is said to be positive. The Fluid-Bridge test is designed to test bladder neck competence by testing for the presence of fluid within the urethra, by demonstrating continuity between two channels of a pressure-recording catheter (one in the bladder and the other in the urethra).

Diagnostic accuracy

Q-tip test

One study compared the accuracy of the Q-tip test for evaluating urethrovesical junction mobility against ultrasound, as the reference standard in women with prolapse or UI (93% UI). It reported that the Q-tip test (change of 303 or more between rest and straining angles from the horizontal) had sensitivity of 25%, specificity 78%, PPV 67% and NPV 37%, relative to a positive test on ultrasound (more than 10 mm movement, n = 114).163 [EL = DS III]

Three studies compared visual assessment of the urethrovesical junction (POP-Q) with the Q-tip test.164–166 One of these studies reported the accuracy of the POP-Q system for diagnosing urethral hypermobility, using the Q-tip test as the reference standard, in women with symptoms of prolapse (70%) and/or UI (30%). Results were presented for different cut-off points of Aa descent. As the Aa point became more distal, specificity and PPV of visual assessment increased (from 36% to 100% and 80% to 100%, respectively), and sensitivity and NPV fell (from 94% to 2% and 67% to 26%, respectively) (n = 111).164 [EL = DS III] The other studies evaluated the correlation between Q-tip and POP-Q measurements in women who had urethral hypermobility (maximum straining angle of 303 or more on Q-tip test). The proportions with urethral hypermobility for each stage of the POP-Q classification were:

Bonney and Marshall tests

Three studies reported findings of urodynamic testing with or without the Bonney test in women with stress UI.167–169 None were truly diagnostic accuracy studies in that they did not report sensitivity, specificity, PPV or NPV.

One case series found that none of the women who had demonstrable leakage during urodynamic investigation leaked during the Bonney test (hence all had a positive Bonney test). Urethral closure pressure was approximately three-fold higher when conducting the test than when not (n = 61).167 [EL = 3]

In a similar study, outcomes were measured at rest, when conducting the Bonney test, and with direct compression of the urethra and bladder neck. Urethral closure pressure, pressure profile area and cough pressure profile area increased significantly compared with resting profiles, when conducting the Bonney test, and when urethra and bladder neck were directly compressed. No significant differences in pressures were seen between the Bonney test and urethral compression. All women had urine leakage when coughing at rest; none leaked during the Bonney test or with urethral/bladder neck compression (n = 12).168 [EL = 3]

A third study did not report complete numerical results for the outcomes assessed (urethroscopic observations, proximal urethral pressure changes, pressure transmission ratio). The authors noted that there was no overlap between the Bonney test and direct urethral compression in proximal urethral pressure changes (n = 37).169 [EL = 3]

One study reported urethral closure pressure profile and cough pressure profile measurements in women with stress UI when undertaking the Marshall test, and with intentional urethral occlusion.

The urethral closure and cough pressure profiles were not significantly different during the Marshall test or with urethral occlusion (n = 16).170 [EL = 3]

Fluid-Bridge test

Three studies evaluated the accuracy of the Fluid-Bridge test for the diagnosis of stress UI in women.171–173 In each study, fluid reaching a point 0.5 cm from the urethrovesical junction indicated a positive test. In two studies, the reference standards used were a clinical and a urodynamic diagnosis of stress UI.171,172 The test was conducted with women in the supine position in both studies, and additionally in the erect position in one.172 With women in the supine position, the Fluid-Bridge test had:

With women in the erect position, the Fluid-Bridge test had:

A third study evaluated the accuracy of a ‘modified’ Fluid-Bridge test relative to a diagnosis of stress UI due to bladder neck incompetence based on findings on history or on the Marshall test (n = 66). The Fluid-Bridge test had:

Urethral pressure profilometry had sensitivity 50%, specificity 88%, PPV 69% and NPV 78% relative to the Marshall test.173 [EL = DS III]

Do tests of urethral competence predict outcome?

Only one study reported any data relevant to this question. In a case series of women undergoing colposuspension or needle suspension, the failure rate was significantly higher in women with a negative Q-tip test, who formed 4% of the study population (n = 406).174 [EL = 3]

Evidence statements for tests of urethral competence

The Q-tip, Bonney, Marshall and Fluid-Bridge tests have been developed to evaluate the mobility or competence of the urethrovesical junction. [EL = DS III] However, there is no evidence to support their role in the clinical assessment of UI. [EL = 4]

Recommendations for tests of urethral competence

The Q-tip, Bonney, Marshall and Fluid-Bridge tests are not recommended in the assessment of women with UI. [D]

3.13. Cystoscopy

Cystoscopy is the direct visualisation of the bladder and urethral lumen using either a rigid or flexible cystoscope. Examination is used to identify areas of inflammation, tumour, stones and diverticula, all of which are findings that will require management within a different clinical pathway.

Alternative pathway

One study aimed to determine whether multichannel cystometry, in combination with urethrocystoscopy, improved the ‘diagnostic accuracy’ of cystometry alone in women, 93% of whom presented with UI, the remainder with prolapse (6%) or retention (1%). The women’s history as reported did not indicate that cystoscopy was necessary. Urethrocystoscopy indicated a new diagnosis, of different pathology, in six women (n = 84).175 [EL = 3]

Diagnostic accuracy

Two studies evaluated the accuracy of dynamic urethroscopy (urethroscopy with simultaneous supine cystometry) relative to multichannel cystometry with or without urethral pressure profilometry, for the diagnosis of stress UI (one study)176 or DO (one study).148 For a diagnosis of stress UI, urethroscopy had sensitivity of 60%, specificity 79%, PPV 75% and NPV 66% (n = 99).176 [EL = DS III] For a diagnosis of DO, urethrocystoscopy had sensitivity of 25%, specificity 94%, PPV 65% and NPV 74% (n = 218).148 [EL = DS III]

Evidence statement for cystoscopy

The available evidence does not support the role of cystoscopy in the assessment of women with UI. [EL = 3]

From evidence to recommendation

The GDG felt that cystoscopy may be of value in women with pain or recurrent UTI following previous pelvic surgery, or where fistula is suspected; its place in recurrent stress UI without these additional features is less clear.

Recommendation for cystoscopy

Cystoscopy is not recommended in the initial assessment of women with UI alone. [D (GPP)]

3.14. Imaging

Imaging techniques that can be used in the assessment of the urinary tract include ultrasonography, X-ray, computed tomography (CT) and magnetic resonance imaging (MRI). Confirmation of alternative pelvic pathology, by means of cross-sectional imaging or ultrasound, would be an indication for referral to a specialist. In addition, imaging may be used to characterise the extent and anatomical contents of a POP, especially in the standing position with MRI.

Diagnostic accuracy

The use of ultrasound for the diagnosis of post-void residual urine is considered in Section 3.6.

Diagnosis of UI

No evidence was identified that considered the use of MRI or CT scanning in the assessment of women with UI. Studies considering the use of ultrasound and X-ray imaging are described below.

The sensitivity and specificity of ultrasound and the Q-tip test, relative to a urodynamic finding of stress UI in women, was reported in one case series. Ultrasound (a positive test defined as a 1 cm or greater drop in urethrovesical junction) had a sensitivity and specificity of 86% and 91%, respectively. The values were 90% and 55%, respectively, for a positive Q-tip test (change in angle of 353 or more) (n = 67).177 [EL = DS II]

Other studies have investigated whether certain parameters that could be measured by imaging might be used in the assessment of women with UI. These parameters include bladder wall thickness, bladder neck positioning, specific urethral measurements and the posterior urethrovesical angle.

Bladder wall thickness for DO diagnosis

Two studies focused on bladder wall thickness measured by transvaginal ultrasound for the diagnosis of DO.178,179 One reported significantly greater bladder wall thickness in women with DO than with any other diagnosis, and that bladder wall thickness of more than 5 mm had sensitivity of 84%, specificity 89% and PPV 94% for diagnosing DO, using videocystourethrography with or without ambulatory urodynamics as the reference standard (n = 180).178 [EL = 3] The second study investigated bladder wall thickness in women in whom urodynamic findings and clinical diagnoses were equivocal. Compared with women with stress UI, and compared with women without UI on urodynamic testing, the bladder wall thickness in women with DO appeared to be significantly greater (n = 128).179 [EL = 3]

Studies investigating correlation of anatomical shape or movement with stress UI

Another four studies considered whether anatomical shape or movement correlates with reporting of UI, or with urodynamic findings, but the clinical significance of the findings reported in these studies is not clear. The studies considered the following (see evidence tables for findings).

  • A case series investigated whether urethral measurements, taken by intraurethral ultrasonography, could distinguish women with intrinsic sphincter deficiency (ISD) from those with urodynamic stress UI (n = 39).180
  • Two studies considered whether the posterior urethrovesical angle measured using bead chain urethrocystography could be used to diagnose stress UI.181,182 One of the studies aimed to determine how bladder neck descent and posterior urethrovesical angle correlated with urodynamic findings (n = 84).181 [EL = 3] The other study considered the prevalence of several parameters, including posterior urethrovesical angle of 1153 or more, in continent and incontinent groups (n = 59).182 [EL = 3]
  • One study evaluated two parameters measured on bead chain cystography (the urethra at the most dependent position in the bladder, and descent of the urethrovesical junction below the posterior edge of the symphysis pubis) compared with a 1 cm or greater drop in urethrovesical junction, measured on ultrasound in women with stress UI (n = 85).183 [EL = 3]

Does imaging affect women’s outcomes?

No evidence was identified that addressed this question.

Evidence statement for imaging

There is a lack of evidence regarding the use of MRI or CT scanning in the assessment of women with UI. The available data do not support the use of ultrasound or X-ray imaging in the assessment of UI. The correlation between anatomy and function is unclear. [EL = 3]

Recommendation for imaging

Imaging (magnetic resonance imaging, computed tomography, X-ray) is not recommended for the routine assessment of women with UI. Ultrasound is not recommended other than for the assessment of residual urine volume. [D]

Research recommendation for imaging

Further studies are required to clarify the role of ultrasound for the assessment of OAB.

3.15. Information provision

No evidence was identified in relation to whether providing information to a woman has an impact in terms of her satisfaction with the outcomes of treatment for UI or OAB.

While there is a lack of evidence in relation to the information given to women with UI, women need the right information, at the right time, with the support they need to use it. It is well recognised in the healthcare community that clear communication, the involvement of service users and the provision of timely evidence-based information are key elements in moving towards a genuinely patient-centred service. Improving information for patients was a commitment in the NHS Plan184 and part of the recommendations in the Kennedy Report.185 There are NHS guidelines on the importance of patient information and a toolkit to help develop these can be downloaded from www.nhsidentity.nhs.uk/patientinformationtoolkit/index.htm.

Patients’ desire for information may be underestimated in the majority of cases, although it is also recognised that individual patients’ desire for information varies.186 Information ‘seekers’ may cope better with more information, and information ‘avoiders’ cope better with less.

Women presenting with symptoms of UI need information that helps them to understand the various types of UI, their symptoms, investigations and the treatments recommended. They need to feel confident that the information provided is based on valid, systematic research into which clinical procedures, drug therapies and medical devices are most effective. However, patients and their families need information that is both scientifically valid and understandable. Since patients make important medical decisions with their clinicians (not separately from them), information provided must be designed for use by patients with their clinicians.

Women with UI should also be given information on where else to go for help and support, for example patient organisations such as Incontact (www.incontact.org) and the Continence Foundation (www.continence-foundation.org.uk).

Copyright © 2006, National Collaborating Centre for Women’s and Children’s Health.

No part of this publication may be reproduced, stored or transmitted in any form or by any means, without the prior written permission of the publisher or, in the case of reprographic reproduction, in accordance with the terms of licences issued by the Copyright Licensing Agency in the UK [www.cla.co.uk]. Enquiries concerning reproduction outside the terms stated here should be sent to the publisher at the UK address printed on this page.

The use of registered names, trademarks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant laws and regulations and therefore for general use.

Bookshelf ID: NBK57199

Views

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...