It is known that straining to void is not productive for men with bladder outflow obstruction (BOO) [1] and that it has a higher likelihood of occurring in patients with detrusor underactivity (DU) [2]. We therefore investigated if analysis of the combination of abdominal strain and urine flow could be used as an easy diagnostic test for these conditions. 

A flow characteristic of straining has a fluctuating shape [3] so we hypothesised that features such as change in flow rate during change in abdominal pressure and rate of change of flow during straining were different between groups of obstructed, normal and underactive patients. Since the prostate / sphincter complex sits within the pelvic floor musculature, we also hypothesised that straining for an obstructed man would act to reduce flow by further compressing the urethra, whereas for an unobstructed man the effects of straining would be less. We also considered that to make a truly non-invasive diagnostic test, it would be preferable to use abdominal EMG rather than intraabdominal pressure measurements to assess straining.

This pilot study aimed to assess from retrospective urodynamic data the ability of combined straining and urine flow measurement to diagnose BOO in men who were observed to strain during their void, with a view to commissioning a full prospective study.

We gathered data from invasive urodynamics carried out in men who had happened to show straining of more than 10 cmH2O on the abdominal pressure (Pabd) line during their pressure flow study. We assessed maximum changes in Pabd and flow rate (Q), and the time of commencement of straining (before or during flow). For rates of change assessment, we considered the fastest rise in urine flow (dQ/dt) and measured the corresponding rate of change in abdominal pressure (dPabd/dt) at that time. In a follow up study, we assessed the ability of surface EMG of the rectus abdominis muscle to measure the presence of abdominal straining as confirmed by intrarectal pressure measurement by water-filled catheter balloon. Rate of change of Pabd was measured by exporting the urodynamic data to a spreadsheet, and plotting dPabd/dt against dQ/dt for the period of the fasted rise in Q.  The average value of dPabd/dt for that period was then plotted against the BOO index for each patient. The standard delay to the pressure signal of 0.6 seconds was present in the urodynamic software. Differences of groups compared to the obstructed group were assessed using Student’s t-test, and p<0.05 was considered significant. Ethical approval was granted for the retrospective use of anonymised patient data.

The urodynamic data for 28 men (10 BOO, 9 normal voiders, 9 DU) were found to include straining of more than 10 cmH2O during their voiding study.  The different parameters relating to Pabd and urine flow are summarised in Table 1.  The changes in Q, and the ratio of Pabd to Q change, between BOO and normal groups were statistically significantly different, but there was no difference between the DU and BOO groups in these parameters.  However, when considering the average gradient of Pabd during maximum Q change (Figure 1), the BOO group was statistically significantly different from both Normal and DU groups, suggesting it could be used as a diagnostic indicator.  Using dPabd/dt < -3 cmH2O/s as a cutoff, the test has sensitivity of 70% and specificity of 94%.

We assessed the differences in Pabd and Q traces during pressure flow studies in patients who strained of their own accord during voiding. We found that neither changes in pressure or flow, nor the ratio of the two, could be reliably used to differentiate men with BOO from both normal and underactive patients.  We also found that abdominal EMG could be used as a proxy for abdominal pressure measurement.
From our retrospective pilot data, it appears that the most sensitive indication of BOO using abdominal pressure and urine flow rate alone is the relation between the rate of change of abdominal pressure (dPabd/dt) and the rate of change of flow rate (dQ/dt), specifically when dQ/dt is at a maximum.  We measured the average dPabd/dt during flow rate rise, and found it appears to confirm our hypothesis that rising Pabd tends to increase Q in the unobstructed male, whereas falling Pabd tends to increase Q in the obstructed male. Apart from one patient, any dPabd/dt less than -3 cmH2O/s occurred in the presence of BOO. There was no difference in dPabd/dt between normal and underactive groups of unobstructed patients. 
In engineering terms, this relation between abdominal pressure (or abdominal EMG) and urine flow curves is known as the phase relationship. We suggest that the phase difference between Pabd and Q while a patient is asked to strain during voiding could be used as an indicator of the presence of BOO.  An analysis of phase could further incorporate rates of change of abdominal pressure during both rises in flow rates (as we have reported here) and also falls in flow rate, since anecdotally the flow rate often appears to rise as abdominal pressure falls in obstructed men. We now plan to commission a larger, prospective study to investigate whether straining on request in every patient can be used as BOO diagnostic. This is needed in order to eliminate the possibility that our pilot study was biased by using data from men who already strained during voiding. EMG will be pursued as a method of measuring the presence of abdominal strain. During this further study, we will also use machine learning (ML) techniques to analyse curve shapes, since there is a high likelihood that pattern recognition of abdominal pressure and urine flow traces could be a powerful tool in non-invasive patient assessment. As with any ML training stage, large amounts of patient data will be needed for this.

We present the first assessment of which we are aware of the analysis of abdominal pressure and urine flow trace gradients and phase during pressure flow studies.  We have shown in a pilot study that when taken during the maximum flow rate rise, an abdominal pressure gradient of less than -3 cmH2O/s will indicate BOO with a sensitivity of 70% and a specificity of 94%.  A larger study will now attempt to assess if this is a viable screening or diagnostic test for male patients with lower urinary tract symptoms, as it could be less invasive, cheaper and quicker than standard urodynamic pressure flow studies.

Reynard JM, Peters TJ, Lamond E, Abrams P. The significance of abdominal straining in men with lower urinary tract symptoms. Br J Urol 1995 Feb;75(2):148-53.
Gammie A, Kaper M, Dorrepaal C, Kos T, Abrams P. Signs and Symptoms of Detrusor Underactivity: An Analysis of Clinical Presentation and Urodynamic Tests From a Large Group of Patients Undergoing Pressure Flow Studies. Eur Urol 2016 Feb;69(2):361-9.
Tarcan T, Acar O, Agarwal MM, Rubilotta E, De Nunzio C, Rosier P.  ICS educational module: The practice of uroflowmetry in adults. Continence 2024 (9) doi.org/10.1016/j.cont.2023.101065. In press.

Continence 12S (2024) 101408
DOI: 10.1016/j.cont.2024.101408