European Urology

Clinical decision making in the field of pelvic organ prolapse (POP) and/or urinary incontinence (UI) surgery continues to be based on historical single-department experience or physician's perception. Randomised controlled trials (RCTs) represent occasionally useful information for surgical strategy [1]. Personal clinical and surgical experience alone may result in poorly reproducible predictions for applying a specific strategy at time of surgery as well as excessive variability. High-quality RCTs are infrequently reported in the context of urogynaecologic surgery, and even when these studies exist, they are not implemented over time or do not receive external validation [2]. Consequently, most of the surgical issues that pertain to indication and type of surgery for POP and UI cannot be solved using only RCT data.

Over the last decade, accurate preoperative evaluation of women with POP and/or UI and knowledge of patients’ expectations were developed as predictive tools with the aims of assisting in clinical decision making and providing more accurate, highly reproducible estimates of objective and subjective outcomes of interest after urogynaecologic surgery. Instruments include risk groupings, lookup tables, classification and regression tree analyses, prediction models that are often presented graphically in the form of nomograms, and artificial neural networks (ANNs).

In this issue of European Urology, Serati and colleagues used advanced computer-based ANN technology to investigate the correlation between urinary symptoms and urodynamic assessment in women with POP [3]. They wanted to know whether ANN could supplement, or even replace, urodynamic assessment in the preoperative work-up in women with POP. Of > 1758 patients with pelvic floor dysfunctions, 802 (45.6%) were eligible for inclusion in the advanced statistical model, which detected significant associations between baseline data, symptoms, anatomic findings, and urodynamic diagnosis. The authors worried that results might not be clinically useful: “[S]ymptoms such as SUI and OAB were commonly recorded as independent predictors of all the UDS [urodynamic] diagnoses investigated, and mostly in association with the diagnosis.” Serati and colleague concluded that although ANN is a powerful instrument for investigating complex biologic models, urodynamic evaluation is better. They expressed the view that clinical findings alone, even when supported by ANN, are insufficient for formulating an accurate diagnosis and that, to avoid unexpected outcomes, urodynamic evaluation should be “par for the course” when assessing urinary dysfunction before POP repair procedures.

Serati and colleagues deserve praise for having tried to provide a take-home message for clinical practice. Findings are strongly based on statistical analysis using a complex mathematical model. Outcomes are different if the model is not correctly applied, and these automatic building model procedures can indeed be erroneous because they do not consider the real impact of each predictor. According to the analysis by Serati et al, variables that achieve significance (p < 0.05) or association (p≤ 0.10) in the multiple logistic regressions in the full method were entered stepwise into further analysis. In our opinion, stepwise selection may not have been appropriate because when applied to explanatory variables, it serves to test different combinations of predictors/covariates systematically. Besides the obvious lack of any theoretical background, this approach fails to consider negative covariance, stepwise logistic regression tends to capitalise on chance, and results may not generalise to other similar samples. For all these reasons the method cannot be recommended [4].

We agree with Serati and colleagues that decisions about POP repair surgery cannot be based on symptoms and signs alone. Whether lower urinary tract symptoms (LUTS) are present or not, preoperative urodynamic assessment in women with POP aims at correlating observed or reported disease with a urodynamic finding or identifying subclinical disorders that could have an impact on the expected outcome. Consequently, invasive tests are not always indicated for healthy premenopausal women with low-stage POP and pure stress urinary incontinence (SUI), urethral hypermobility, and clinical signs of SUI at clinical examination who do not have storage, voiding, or postmicturitional symptoms and have not previously undergone urogynaecological surgery.

There are several challenges in managing women with POP and SUI: to identify intrinsic urethral sphincter deficiency in women with POP and SUI that requires a concomitant anti-incontinence procedure, to disclose detrusor overactivity or underactivity, and to know what is the bladder compliance. In all these cases urodynamic assessment might be considered mandatory. Transobturator slings are reported to be less effective in patients with intrinsic sphincter deficiency (ISD) [5] and [6]. Therefore if a transobturator sling is planned, identifying ISD might change the plan. Unfortunately, the literature does not solve the case [7]. In patients with or without concomitant incontinence, scheduled for POP repair, urodynamic assessment might allow a more selective use of incontinence surgery at the time of POP surgery [8]. In the era where patient safety is the most important concern, surgeons and their patients must weigh the risks and benefits of performing prophylactic procedures during POP surgery.

Other challenges are to detect detrusor over- or underactivity and determine bladder compliance. The incidental discovery of detrusor overactivity in a healthy woman with straightforward POP or SUI would not change the surgical plan. However, we believe that all surgeons who perform POP repair agree that urodynamic assessment should remain mandatory for all patients with LUTS. Detrusor underactivity increases the risk for postoperative voiding dysfunction, and urodynamic assessment serves to identify patients at increased risk. Currently we lack well-established thresholds for urodynamic values that would trigger a change in patient counselling when detrusor underactivity is diagnosed, but simple low-cost uroflowmetry might be used to discriminate candidates requiring invasive tests. Determining bladder compliance is crucial when planning an integral pelvic floor reconstruction because poor compliance could be worsened if outlet resistance is increased by, for example, a concomitant anti-incontinence procedure [9] and [10]. In such cases, identifying patients with low compliance is mandatory. However, in our experience the incidence of poor compliance is low in healthy women with straightforward POP or SUI, and thus urodynamic assessment could be avoided.

Par is a golfing term that denotes the predetermined number of strokes that a scratch golfer should require to complete a hole, a round (the sum of the pars of the played holes), or a tournament (the sum of the pars of each round). Par for the course is the “standard” for the golfer. So urodynamic assessment should be the “standard” for POP surgery. So far we have gained some clues, but our course remains in fieri.

Conflicts of interest

The authors have nothing to disclose.