Does preoperative urodynamics improve outcomes for women undergoing surgery for stress urinary incontinence? A systematic review and meta-analysis

Authors


Abstract

Background

Urodynamics is widely used in the investigation of urinary incontinence. The existing evidence questions its add-on value in improving the outcome of surgical treatment for stress urinary incontinence (SUI).

Objectives

To compare the surgical outcomes in women with SUI or stress-predominant mixed urinary incontinence (MUI) based on urodynamic diagnoses compared with diagnoses based on office evaluation without urodynamics.

Search strategy

We searched Cochrane, MedLine, Embase, CINAHL, LILACS, metaRegister of Controlled Trials (mRCT) and Google Scholar databases from inception until March 2013.

Selection criteria

We included randomised controlled trials (RCTs) comparing surgical outcomes in women investigated by urodynamics and women who had office evaluation only.

Data collection and analysis

Two independent reviewers (S.R. and P.L.) extracted the data and analysed it using review manager (revman) 5.2 software.

Main results

Of the 388 articles identified, only four RCTs met our criteria. The data from one study are as yet unpublished. In the other three RCTs, the women with SUI or stress-predominant MUI were randomised either to office evaluation and urodynamics (= 388) or to office evaluation only (= 387). There was no statistical difference in the risk ratio (RR) of subjective cure in the two groups (RR 1.02, 95%CI 0.90–1.15, P = 0.79, I2 = 45%), objective cure (RR 1.01, 95%CI 0.93–1.11, = 0.28, I2 = 20%) or complications such as voiding dysfunction (RR 1.54, 95%CI 0.61–3.89, = 0.27, I2 = 18%) or urinary urgency (RR 0.80, 95%CI 0.28–2.3, = 0.19, I2 = 40%).

Authors' conclusions

In women undergoing primary surgery for SUI or stress-predominant MUI without voiding difficulties, urodynamics does not improve outcomes – as long as the women undergo careful office evaluation.

Introduction

Stress urinary incontinence (SUI) is the involuntary leakage of urine during increased intra-abdominal pressure, in the absence of a detrusor contraction.[1] Urodynamics has been universally accepted as the gold standard investigation for SUI. Urodynamics includes noninvasive uroflowmetry followed by multichannel filling cystometry. Urodynamics is an expensive, invasive and uncomfortable investigation carrying a risk of urinary tract infections (3–5%).[2] Office evaluation for SUI includes focused history-taking, quality of life (QoL) questionnaires, bladder diaries, pelvic examination, cough test, uroflowmetry, measurement of postvoid residual (PVR) urine volume and urine analysis.[3]

The role of urodynamics in predicting the success of surgical treatment or postoperative complications has been a matter of debate. Clinicians may be concerned that in women who do not undergo urodynamics, discordant diagnoses such as provoked detrusor overactivity (DO) and voiding dysfunction may be missed, to the detriment of a good outcome of surgery for SUI. DO is the occurrence of involuntary contractions which may be spontaneous or provoked during the filling phase of cystometry and may be seen in 20% of patients with a history of pure SUI.[4]

The National Institute for Health and Care Excellence (NICE) guideline (CG171) recommends that urodynamics not be performed in the small group of women where pure SUI is diagnosed based on a detailed clinical history and examination.[5] According to the Cochrane review on urinary incontinence, urodynamics does not predict surgical failure or postoperative DO.[6] Despite this evidence, a recent survey of urologists and urogynaecologists concluded that 89% would arrange urodynamics for women with SUI or stress-predominant mixed urinary incontinence (MUI). However, the majority of respondents recognised the need for further research to address this question.[7]

The objective of this review was to assess whether the performance of urodynamics altered the outcomes of cure or complications in women undergoing surgery with isolated SUI or stress-predominant MUI symptoms.

Methods

We carried out a systematic review after formulating a protocol a priori using recommended methods. We included only randomised controlled trials (RCT) in our systematic review and meta-analyses.

Our four-part clinical questionnaire for this systematic review was as follows:

  1. Population: women with pure SUI or stress-predominant MUI requesting surgical treatment.
  2. Intervention: baseline office evaluation (clinical history and examination including cough test and PVR on bladder scan).
  3. Comparison: urodynamics.
  4. Outcome: cure or improvement of SUI.

Search strategy

We conducted an extensive search in Embase, MedLine, MEDION, Google Scholar, CENTRAL (the Cochrane Library, Issue 1, 2013), the metaRegister of Controlled Trials, the National Library for Health, the National Research Register and Google Scholar, LILACS and CINAHL (from database inception until March 2013). The keywords and MESH headings used in the search were: stress urinary incontinence, office evaluation, history, clinical examination, women AND urodynamics/cystometry/pressure flow studies. We did not apply any methodological filter or language restrictions. We adapted our search strategy to suit each database. We also searched the bibliographies of all relevant primary articles, IUGA (2005–2013), AUGS (2005–2013), UKCS (2009–2013) and ICS (2005–2013) websites.

Study selection

We scrutinised the abstracts identified by the electronic searches and obtained full manuscripts of all the citations that were thought to have met the inclusion criteria.

Data extraction and statistical analysis

The authors (S.R. and P.L.) independently examined the manuscripts to determine their eligibility for inclusion in the review and assessed their quality. We extracted data from the studies using a structured proforma. We have contacted the authors of trials with inadequate or unreported data but were unable to secure the missing data. We extracted the study data on patient characteristics, quality criteria, subjective and objective surgical outcomes and postoperative complications (see Table 1).

Table 1. Characteristics of included studies.
Author, year, country, study designSettingInclusion criteriaExclusion criteriaMethods and duration of follow upIntervention (E) and comparison (C)Outcome measures and follow up
  1. IIQ, Incontinence Impact Questionnaire; ICIQ-SF, International Consultation on Incontinence Questionnaire short form; ICIQ, International Consultation of Incontinence Questionnaire; I-QoL, Incontinence Quality of Life Questionnaire; ITT, intention to treat; MESA, Medical, Epidemiological and Social Aspects of Ageing Questionnaire; MUI, mixed urinary incontinence; PFMT, pelvic floor muscle training; PGI, patient global impression; PGI-I, patient global impression of improvement; POP, pelvic organ prolapse; POP-Q, pelvic organ prolapse quantification system; PVR, postvoid residual scan; QoL, quality of life; Rx, treatment; SUI, stress urinary incontinence; UDI, Urogenital Distress Inventory; UDS, urodynamics; UI, urinary incontinence; UPP, urethral pressure profilometry.

van Leijsen et al. (2012)[9] Netherlands.

Multicentre RCT, noninferiority (VUSIS study)

One academic and nine nonacademic hospitals (initial plan was to randomise 260 women – stopped prematurely due to slow recruitment)Pure SUI or stress predominant MUI, previous failed conservative management; planned surgery and written informed consent givenPrevious SUI surgery. Grade ≥ 3 on POP-Q, PVR >150 ml

Follow up 6 weeks, 6, 12 and 24 months.

Analysis: ITT

(E) Office evaluation: history, physical examination, voiding diary, 48-h pad test, PVR scan, UDI-6 and urinalysis (= 28) (C) Office evaluation + UDS and optional UPP (= 31)

Subjective cure of SUI was defined as a no leakage reported during physical activity (UDI). Objective cure of SUI was defined as a negative stress test by physical examination

Outcome measures: Physical examination (stress test, PVR scan), questionnaires (UDI-6, PGI-Improvement Scale), voiding diary and a pad test

Primary outcome: subjective cure. Mean difference in UDI-6 scale 1 year after surgery was 14 in favour of the group (E) indicating noninferiority of (E)

Absence of SUI and/or discordant urodynamic findings present in 20/31 (65%) in the group (C)

Nager et al. (2012),[10] USA. Multicentre RCT, noninferiority

(VALUE study)

Eleven centres in USA; 90% of the surgeons were fellowship-trained>21 years of age, duration of SUI for at least 3 months; MESA score for SUI > score on this questionnaire for urgency; PVR <150 ml; negative urinalysis or urine culture; clinical assessment of urethral mobility; desire for surgery for SUI and a positive stress testPrevious surgery for UI; a history of pelvic irradiation; pelvic surgery within the previous 3 months; anterior apical POP of 1 cm or more distal to the hymen

Follow up: 3 and 12 months.

Analysis: primary per protocol and secondary by ITT

(E) Office evaluation: SUI symptoms by MESA, stress test, PVR, urine dipstick, UDI,PGI-I, assessment of urethral mobility by Q-Tip test, visualisation, palpation, point Aa on POP-Q exam (= 315) (C) Office evaluation + UDS and optional UPP (= 315)

Primary outcome: UDI and PGI-I scores (subjective)

Success defined as reduction in UDI score by ≥ 70% at the end of 1 year –200/259 (77.2%) in (E) versus 203/264 (76.9%) in (C)

Secondary: incontinence severity index, MESA questionnaire, IIQ, Medical Outcomes Study 12-item Short Form Health Survey, PGI – severity, stress test at 12 months. No significant differences between the study groups

Voiding dysfunction at >6 weeks identical – 6/315 (1.9%) versus 6/315 (1.9%).

The diagnosis of SUI, as made by office evaluation was confirmed urodynamically in 97% in the UDS group

Hilton et al. (2011),[8] UK.

Multicentre RCT-(INVESTIGATE-1) (mixed methods pragmatic multicentre pilot study)

Six centres in the UKInclusion: SUI or stress-predominant MUI; have undergone PFMT (± other nonsurgical treatment for urge symptoms) with inadequate resolution of symptoms, woman and her treating clinician agree that surgery is the required treatmentExclusion: symptomatic POP requiring Rx; previous surgery for UI or POP; UDS in the last 3 years; neurological disease causing UI; current involvement in any competing research studiesFollow up: 6 months after surgery(E) Office evaluation: basic clinical assessment along with frequency/volume charting, midstream urine culture, urine flow rate and PVR measurement (= 110) (C) Office evaluation + UDS with optional ambulatory UDS (= 112)

Outcome measures: rate of recruitment, randomisation, data completion

Qualitative interviews of women regarding their willingness to participate in randomisation; national survey of clinicians on their willingness to randomise for a definitive trial; qualitative interviews of clinicians on how they use UDS to make decisions

Feasibility: RCT results still unavailable

Romero Maroto et al. (2010),[11]

Spain.

Conference abstract.

Noninferiority RCT

Not clear whether single-centre or multicentre Exclusion: patients younger than 18; previous radiotherapy; previous incontinence procedure. Women had adjustable tapes as the surgical procedureSurgical intervention: adjustable mesh implant. Follow up: 49 months(E) Office evaluation: history, physical examination with full bladder, flowmetry, PVR, I-QoL, ICIQ-SF (= 44) (C) Office evaluation + UDS (= 42)

Outcome measures: history, cough test with full bladder, flowmetry with PVR, I-QoL, ICIQ-SF and PGI-I scores

Primary outcome: at 49 months' follow up 36/44 (81%) (E) versus 38/42 (90%) (C) showed improvement on cough test and 56 versus 59% on ICIQ-SF.

Voiding difficulty was not mentioned in the postoperative complications

All data were analysed using review manager (revman) 5.2 software, provided by the Cochrane Collaboration (http://ims.cochrane.org/revman). Heterogeneity between trial results was tested with a standard Chi-squared test. Forest plots were generated to illustrate the study-specific effect sizes along with a 95% confidence interval (CI).

Methodological quality assessment

The authors (S.R. and P.L.) evaluated studies according to the description of the study population, the methodology of patient selection, diagnostic tests, random allocation, blinding and the description of withdrawals and dropouts. We ascertained the risk of bias on each of the included studies independently.

Results

Figure 1 shows the process we used to select studies for this systematic review. Of the 388 articles identified in the search, only four RCTs met our inclusion criteria. One of them was a pilot study to test the feasibility of a future definitive RCT to address the question of whether invasive urodynamics alters treatment decisions and outcomes of SUI surgery, its clinical and cost-effectiveness.[7] We contacted the authors for their data but were advised that they had not been analysed to date (P. Hilton, pers. comm.).[8] Three of the RCTs had a noninferiority design (= 775) and excluded women with preexisting voiding dysfunction, advanced prolapse and previous incontinence surgery as well as those with high PVR volume. The women underwent surgery after being randomised to either baseline office evaluation of history, bladder scan for PVR volume, bladder diaries and clinical examination for stress test (= 338) or for preoperative urodynamics (= 345).

Figure 1.

Flow diagram depicting the study selection process in the meta-analysis of effectiveness of office evaluation versus urodynamics for cure following surgery for stress urinary incontinence.

Table 1 gives the details of the included studies with regards to the populations, settings, study methods, details of office evaluation and outcome measures. All the studies excluded patients with PVR >150 ml and prolapse > grade 2. Figure 2 summarises the meta-analyses of cure and complications in the two groups that were compared.

Figure 2.

Meta-analysis of outcomes of cure and postoperative complications.

Excluded studies

We excluded those studies (see Supporting Information Table S1) that did not address the four-part clinical question given above or those that were not RCTs.[12-17]

Risk of bias

Two of the included studies (VALUE and VUSIS) were of good methodological quality (see Table 2). Blinding of participants is not possible in trials with invasive interventions such as urodynamics. Selection bias was minimised by allocation concealment in two of the three studies included.[9, 10] There was detection bias in two of the included studies, as the outcome assessor was not blinded to the diagnostic interventions used.[9, 11] The third RCT was published as a conference abstract and hence a full quality assessment could not be carried out.[11] Data on the fourth study were unavailable, so its quality could not be assessed.[8] We did not assess publication bias as the number of included studies was too small to construct a funnel plot.

Table 2. List of studies excluded from the systematic review of urodynamics compared to office evaluation in the surgical outcomes of women with stress urinary incontinence
 Design/Study characteristicsReason for exclusion
Black et al.,[12]

Prospective multicentre cohort study

285 women – 164 underwent office evaluation +UDS and 103 women had only office evaluation. Percentage of women cured were similar in both the groups: 24% versus 26% (P = 0.8). Severe urgency improved from 41% to 24% after surgery

It was not an RCT
Ramsay et al.,[13]RCTThe research question was on conservative management of urinary incontinence and not on surgical treatment of SUI
Khullar et al.,[14]RCT. There seems to be a high rate of attrition bias as the study started with 105 women and only 66 were available for follow up at 3 years. Also the number of women in ambulatory UDS (n = 42) are twice of those in office evaluation group (n = 22). Outcomes of assessment were not clearResearch question was on the role of UDS in lower urinary tract symptoms (LUTS) rather than the specific condition of SUI
Holtedahl et al.,[15]RCT in GP practices in NorwayThe research question was on conservative management of urinary incontinence and not on the surgical treatment of SUI
Anger et al.,[16]Women who had preoperative UDS compared to those who did not have UDS had more frequent diagnoses of urgency incontinence after surgery (21.9% versus 12.7%, = 0.0001). Women did not undergo UDS preoperatively had a lower rate of denovo urgency incontinence (12.2% versus 20.4%, respectively). There were no significant differences in surgical complications or repeat incontinence procedures between both the groupsRetrospective cohort study. Selection bias of patients with significant urgency component of mixed incontinence having preoperative UDS made the findings of this study unreliable
Nager et al.,[23]RCTResearch question was on comparison between Burch colposuspension versus pubovaginal sling and not on UDS versus without UDS in surgical outcome
Majumdar et al.,[17]RCT (Patient preference trial) Evaluation in the group without UDS included history, urine dipstick and three day bladder diary which was not comprehensive. Women showed a preference to undergo UDS. Women who had UDS were more motivated to comply with treatments and follow-ups. UDS did not improve incontinence episode frequency, it had utility in patient choice and subsequent treatment complianceResearch question was on women with LUTS and effect on treatment rather than women with SUI or stress predominant MUI and their outcome after surgery

Van Leijsen et al.,[30]

VUSIS2 study

A multicenter diagnostic cohort study with an embedded non-inferiority randomized controlled trial. The objective was to estimate whether a strategy of immediate surgery was non-inferior to a strategy of individually tailored therapy in women with stress urinary incontinence (SUI) but with discordant findings on UDS. Discordant findings on UDS included absence of SUI, DO, residual volume, small cystometric capacity, poor flow, low compliance and dysfunctional voiding. Conclusion: In women with uncomplicated SUI, an immediate mid-urethral sling operation is not inferior to individually tailored treatment based on urodynamic findingsThe research question was on the management of women with urodynamic diagnoses discordant to the clinical assessment rather than on the role of UDS in SUI management. All participants received UDS and were randomly allocated to receive either immediate surgery or individually tailored therapy based on UDS results

There were some differences in the extent of office evaluation, inclusion/exclusion criteria and duration of follow up, and some inequality in the urodynamics group compared with the office evaluation group for factors such as smoking, use of hormone replacement therapy, and the duration of symptoms across all the three studies.

Cure

Subjective cure was assessed using the Urogenital Distress Inventory Scale (UDI Scale) and/or the Patient Global Impression of Improvement (PGI Scale) studies.[9-11] Pooling the results from three studies for subjective outcome on the UDI Scale did not show a difference in the perception of cure of women randomised to the experimental group (office evaluation only) compared with the control group (office evaluation and urodynamics) (risk ratio [RR] 1.02, 95%CI 0.90–1.15, = 0.79). Objective cure was defined as a negative cough stress test. Pooled results of three studies found a similar response in both the experimental and control groups at 12–49 months' follow up (RR 1.01, 95%CI 0.93–1.11, = 0.77).[9-11]

Complications

Postoperative voiding dysfunction was reported in two studies.[9, 10] Only one study defined voiding dysfunction explicitly: patients who needed medical or surgical therapy to facilitate bladder emptying.[10] At the 1-year follow up, pooled results from these two studies did not show a statistically significant difference for this outcome (RR 1.54, 95%CI 0.61–3.89, P = 0.36). The incidence of new or continued treatment for urgency (RR 0.80, 95%CI 0.28–2.30, = 0.68) was not different in the two groups.

Discussion

Main findings

The collated evidence from the RCTs confirms that there is no difference in subjective or objective cure or postoperative complications (voiding dysfunction, new-onset urgency) in the office evaluation group compared with the group undergoing urodynamics and office evaluation.

Strengths

A systematic review of RCTs is known to reduce bias and random errors, and provides the highest quality of evidence. Our search was thorough and systematic without language restrictions. We attempted to contact the authors of published and unpublished studies to obtain further details but were unsuccessful with some authors. We have included only noninferiority RCTs in accordance with the recommendations on clinical trials assessing the utility of urodynamics.[18] We adhered to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) checklist for reporting the meta-analyses.[19] The degree of noninferiority in the study design was similar in the two RCTs (11% in the VALUE study and 8% in the VUSIS study).[20] Attrition bias and reporting bias were almost nonexistent in all three included RCTs. A fixed effects model is based on the assumption that the true effect is the same in all the studies. In reality, it is very unlikely that the included studies are identical and the true effect size is exactly the same. We used the random effects model for all four outcomes to accommodate one adequately powered and two underpowered studies, and due to the fact that there was some variation in the inclusion criteria, extent of office evaluation, etc.

We excluded nonrandomised retrospective studies analysing the outcome of surgery for SUI in women who had undergone urodynamics compared with the hypothetical omission of urodynamics because they might have unknown and uncontrolled confounding factors.[4, 21-24] Although there is a Cochrane review and other narrative reviews, in our systematic review we have focused our attention on the role of urodynamics in the surgical treatment of SUI.[6, 25] Three noninferiority RCTs carried out in different countries (of which two were multicentre RCTs) have similar findings, which increases our confidence in the results. They also enhance the generalisability of the findings, as the population in the studies included adult women of all age groups as well as those with coexisting urgency symptoms commonly seen in clinical practice.

Limitations

The quality of a meta-analysis depends on the quality of the included studies. We acknowledge several limitations of this review. Power calculation was reported in only two of the RCTs.[9, 10] One of these RCTs was stopped prematurely after randomisation of 59 patients due to slow recruitment, making it underpowered.[9] Blinding of participants was not possible due to the design of the RCTs and the nature of the intervention. Blinding of outcome assessors, though easily possible, was only carried out in one study.[10] There was some heterogeneity (I2 < 40%) among the included studies for objective cure, urgency and voiding dysfunction. Notwithstanding these limitations, all the studies indicated similar results.

The majority of women included in the review had a mid-urethral sling (MUS). A very small number of patients underwent colposuspension (= 2) or urethral bulking (= 4).[10] Such a small number of women undergoing different surgical procedures should not influence the outcomes in a RCT. The included studies may be criticised for the short duration of follow up of only 1 year.[9, 10] However, the mean duration of follow up was 49 months in the RCT by Maroto et al. and the results were still the same.[11]

Interpretation

A diagnostic test should only be performed if the results are going to influence management or the outcomes of treatment and direct the patient to an alternative pathway. The diagnostic accuracy of urodynamics in lower urinary tract conditions has never been studied due to the absence of an acceptable reference standard. The variation in urodynamic parameters may be up to 15% when the test is repeated in the same patient, challenging its reproducibility.[26]

Urodynamics may not differentiate the various pathologies of stress incontinence, for example bladder neck hypermobility or intrinsic sphincter deficiency. The diagnosis of intrinsic sphincter deficiency has not been standardised.[27] The aetiology or severity of SUI based on urodynamic parameters such as abdominal leak point pressure or maximum urethral closure pressure does not seem to alter the surgical outcome of a MUS.[23]

Uroflowmetry is an essential component of office evaluation. Uroflowmetry is known to have a high correlation with voiding cystometry. The negative predictive value (NPV) of normal uroflowmetry in ruling out abnormal voiding flowmetry is 97.3%.[28] Hence the likelihood of missing voiding dysfunction when women are screened with uroflowmetry and PVR on bladder scan alone should be negligible.

Coexisting DO is considered to be an adverse risk factor for the success of SUI surgery.[29] In a study by van Leijsen et al.,[30] women with a clinical history of SUI but with discordant urodynamic diagnoses (= 126) were randomised to immediate surgery (= 64) or to individually tailored therapy based on urodynamic diagnoses (= 62). Urinary incontinence had improved in 82% of women in the group with coexisting DO compared with 93% who improved in the group without DO (odds ratio [OR] 0.36, 95%CI 0.13–1.03). The outcome of cure in women undergoing immediate SUI surgery despite discordant urodynamic diagnoses was not inferior to the group undergoing tailored symptomatic treatment after 13 months followup.[30]

Urodynamics changed the diagnosis in 57% of women: it increased the diagnosis of voiding dysfunction (2.2–11.9%, < 0.001) and reduced the diagnosis of overactive bladder (41.6–25.2%, < 0.001). However, only 12.8% of those women had a urodynamics-driven change in their global treatment plan. These treatment changes were not associated with improved treatment success (OR 0.96, 95%CI 0.41–2.25, = 0.92) but were associated with increased postoperative treatment for urgency incontinence (OR 3.23, 95%CI 1.46–7.14, = 0.004).[31]

ICS defines complicated urinary incontinence (UI) as: incontinence with pain or haematuria, recurrent infections, suspected or proven voiding problems, significant pelvic organ prolapse or UI in those who have persistent incontinence or recurrent incontinence after pelvic irradiation, radical pelvic surgery, previous incontinence surgery or a suspected fistula.[32] This group of women with complicated UI will still need urodynamics.

Conclusions

Based on the above evidence, we conclude that urodynamics does not have any add-on value as long as detailed office evaluation is carried out prior to primary SUI surgery in women with isolated SUI or stress-predominant MUI who have a normal bladder capacity and PVR. Clinical models need to be constructed to universally standardise office evaluation. The diagnostic accuracy of a combination of tests used for office evaluation needs to be established in future studies. Methodologically robust RCTs with long-term follow up, powered to determine whether urodynamics-based alterations in management plan improve clinical outcomes, need to be carried out to ensure that results are consistent with short-term outcomes. The omission of preoperative urodynamics and the performance of comprehensive office evaluation in women with SUI or stress-predominant MUI (in the absence of voiding dysfunction and ≥Stage 2 prolapse) might lead to massive savings for healthcare systems and could also save women valuable time and embarrassment and from potential urinary tract infections. Evaluation of the cost-effectiveness of omission of urodynamics in the investigative pathway of women with SUI needs to be carried out.

Disclosure of interests

No conflict of interest is declared.

Contribution of authorship

S.R. was the first reviewer and wrote the manuscript. P.L. conceived the idea, acted as the second reviewer and contributed to the critical revision of the manuscript.

Details of ethical approval

Ethical approval was not required as this is a systematic review of published manuscripts.

Funding

S. Rachaneni is a Clinical Research Fellow on the NIHR-HTA funded BUS study.

Acknowledgements

We thank Mr Derick Yates, EBM tutor at the library of the Birmingham Women's Hospital, for performing the systematic search for this review.

Ancillary