The effect of urethral instrumentation on uroflowmetry


M.M. Issa, MD, Department of Urology, Emory University School of Medicine, 1365 Clifton Road, NE, Atlanta, Georgia 30322, USA.



To evaluate in a prospective study the effect of urethral instrumentation (flexible cystoscopy) on uroflowmetry, and in particular the peak urinary flow rate (Qmax).


Thirty-two consecutive patients (median age 61.8 years, range 24–80) undergoing flexible cystoscopy were included in the analysis. Patients with active urethral stricture disease or urinary infection were excluded. The indications for cystoscopy included haematuria (44%), voiding symptoms (66%), history of bladder cancer (19%), and history of perineal trauma (3%). Patients underwent uroflowmetry immediately before instrumentation. The postvoid residual volume (PVR) was measured by bladder catheterization. After cystoscopy the bladder was completely emptied and then filled with the same volume of sterile normal saline (bladder volume = voided volume + PVR), and the patient underwent a second uroflowmetry.


Patients with voiding symptoms (21, 66%) had a median (range) American Urological Association symptom score of 17 (4–34), a Bother score of 16 (1–23), and Quality of Life score of 3 (1–6). The mean Qmax was 16.9 (4.5–36.9) and 13.3 (4.5–39.4) mL/s before and after cystoscopy, respectively (P = 0.029). The mean percentage difference in Qmax was + 27 (− 23 to 139)% higher before than after cystoscopy. After cystoscopy, up to 25% (eight) and 21% (seven) patients had a lower Qmax, from > 15 to < 15 mL/s and from > 12 to < 12 mL/s, respectively. There were no significant differences in the bladder volume and PVR (P = 0.914 and 0.984, respectively).


Urethral instrumentation by flexible cystoscopy significantly alters Qmax. A ‘false’ mean change in Qmax (favouring improvement) of +27% would result if uroflowmetry data after instrumentation were used at baseline. Therefore, study protocols for benign prostatic obstruction should exclude uroflowmetry data obtained after urethral instrumentation; failure to exclude such data will lead to disproportionately greater improvements in Qmax that are independent of the therapy delivered.


Rehfisch first described the concept of uroflowmetry in 1897; since then, uroflowmetry has undergone numerous technological improvements to become a valuable tool for investigating LUTS [1]. While the test alone does not delineate the underlying causes of LUTS [2], the information obtained can be a valuable adjunct in the diagnosis, management and follow-up of patients. Furthermore, because it is simple, not invasive and safe, uroflowmetry has become an attractive tool for investigating BOO caused by BPH. Currently, the peak urinary flow rate (Qmax) obtained during the test is the most widely used objective variable in evaluating patients with symptomatic BPH. In 1994, the Agency for Health Care Policy and Research established guidelines for the diagnostic use of uroflowmetry [3].

Over the past two decades there has been a tremendous surge in the research and development of medical, thermal and surgical therapies for BPH. Virtually all clinical trials for BPH include uroflowmetry as part of the assessment, inclusion/exclusion criteria and follow-up. Indeed, the use of any new BPH therapy depends significantly on the Qmax values. In general, for a treatment to be considered successful, the Qmax must improve above the placebo range by 25–35%, and preferably by more than half. This ‘success’ criterion has flaws related to the technique of uroflowmetry, patient selection, timing of the test and interpretation; all these have been shown to influence Qmax. In addition, Qmax does not differentiate between various types of BOO or between BOO and detrusor hypocontractility [4,5]. Consequently, cystoscopy and urodynamic testing have been added to clinical study protocols in an attempt to clarify these issues. Often, urinary flow rate is measured immediately after cystoscopy, as the bladder is conveniently full. This practice adds yet another variable (urethral instrumentation) to the dilemma of uroflowmetry, as there is no current and acceptable information about the effect of urethral instrumentation on Qmax. Thus the objective of the present study was to evaluate the effect of urethral instrumentation (flexible cystoscopy) on uroflowmetry and in particular, Qmax.


Forty consecutive patients scheduled for diagnostic flexible cystoscopy underwent a routine evaluation that included a complete history and physical examination, AUA Symptom Score, Bother Score, Quality of Life Score, measurement of serum PSA and serum creatinine, urine analysis, and urine culture. Eight patients were excluded because of active urethral stricture disease, UTI or a history suggestive of neurogenic bladder. The data of the remaining 32 patients were used for the analysis (median age 61.8 years, range 24–80). The indications for diagnostic flexible cystoscopy included voiding symptoms in 21 (66%), haematuria in 14 (44%), a history of previously resected bladder cancer in six (19%), and a previous history of perineal trauma in one (3%).

All patients were instructed to drink sufficient fluid to have full bladders, without overdistension, at the time of their scheduled cystoscopy. On arrival at the cystoscopy suite they underwent standard uroflowmetry (Urodyn 1000, Dantec, Denmark). Furthermore, each uroflowmetry result was reviewed manually to ensure its accuracy. Data from patients with a voided volume of < 125 mL were excluded from analysis. Patients whose voided urine volume was > 125 mL were immediately placed supine on the cystoscopy table, had their external genitalia cleansed with povidone iodine solution, and 10 mL of 2% lidocaine gel instilled into their urethra in preparation for cystoscopy. A 15 F flexible cystoscope (CYF-3, Olympus, Tokyo, Japan; or model ACN-1, ACMI, USA) was introduced into the bladder with no fluid irrigation. The postvoid residual volume (PVR) was measured by emptying the bladder. Diagnostic cystoscopy was then performed in the standard fashion. At the end of the procedure the bladder was emptied completely and then filled to a capacity equal to that before cystoscopy (uroflowmetry voided volume + PVR). A second estimate of the uroflow was then obtained. The PVR was calculated by subtracting the voided volume from the bladder volume at the end of cystoscopy. Student's t-test was used to compare the uroflowmetry results before and after cystoscopy.


In patients who underwent cystoscopy to evaluate voiding symptoms, the median (range) AUA Symptom Score was 17 (4–34), the Problem Score Index (‘Bother Score’) 16 (1–23), and Quality of Life Score 3 (1–6). The median serum PSA was 0.9 (0.02–17.3) ng/mL, with 88% of patients having a normal serum PSA (< 4.0 ng/mL). In the remaining four patients with high PSA levels, three had negative prostate biopsies and one had prostatitis (PSA 17.3 ng/mL). The median serum creatinine was 11 (8–30) mg/L. Of the 24 patients who had BPH diagnosed clinically by symptoms and a DRE, 14 (58%) were on α-blockers (terazosin), one on androgen deprivation therapy (finasteride) and the remaining nine (38%) were receiving no medical therapy or phytotherapy. Three patients had had a previous TURP.

Cystoscopy was normal in five patients (16%) and abnormal in 27 (84%); the diagnosis included BPH in 24 (75%), bladder cancer in two (6%) and radiation cystitis in one (3%). Twenty-nine patients (91%) had bladder trabeculation, with grade I in six (19%), grade II in 14 (44%), grade III in six (19%) and grade IV in three (9%).

The mean volume of fluid in the bladder at the time of uroflowmetry testing was similar (322 and 326 mL) before and after cystoscopy, respectively (P = 0.91; Fig. 1) . The mean PVR at the end of the uroflowmetry tests was similar (50.5 and 50.2 mL) before and after cystoscopy, respectively (P = 0.98), but the mean Qmax was significantly different, at 16.9 and 13.3 mL/s, respectively (P = 0.029; Table 1) . This difference constitutes a 20% decrease in Qmax after cystoscopy, or a 27% greater Qmax before than after cystoscopy. Up to 25% (eight) and 21% (seven) patients had their Qmax decrease from > 15 to < 15 mL/s, or from > 12 to < 12 mL/s, respectively.

Figure 1.

Uroflowmetry graphs before cystoscopy (top) and afterward (bottom) in the same patient. The bladder volume at the time of the uroflowmetry tests was similar.

Table 1.  Results of uroflowmetry before and after cystoscopy
VariableBeforeAfterDifference (after vs before)P
Bladder volume, mL
Mean322326 +  4 (− 4) +  1.2 (− 1.2)0.914
Median257260 +  3 (− 3) +  1.1 (− 1.1)0.914
Qmax, mL/s
Mean  16.9  13.3 −  3.6 (+ 3.6) −  21 (+ 27)0.029
Median  14.7  12.2 −  2.5 (+ 2.5) −  17 (+ 20)0.029
Mean  50.5  50.2 −  0.3 (+ 0.3) −  0.5 (+ 0.5)0.984
Median  30.0  44.0 +  14 (− 14) +  46 (− 31)0.984


Uroflowmetry is a simple and noninvasive test used primarily to objectively determine the strength of the urinary stream in patients with voiding symptoms. The principle variable of uroflowmetry is Qmax; a weak urinary stream can be confirmed by a low Qmax, but such a finding does not differentiate between the various causes of the weak stream (BOO, detrusor hypocontractility or a combination of both). Of patients with low flow rates, 25–30% have detrusor hypocontractility as the main problem rather than BOO [4,5]. The validity of the test is influenced by the urine volume in the bladder at the time of urination. Ideally, the bladder should be sufficiently full (> 125 mL) but not over-distended. The upper limit of bladder capacity varies in each patient but it is generally accepted to have the patient urinate when he/she feels the urge to do so with no delay. Variability in interpreting the uroflow graphs and Qmax values has been discussed; Golomb et al.[6] reported variability in Qmax relating to circadian changes in a series of 32 patients, with a significantly greater Qmax (20%) in the evening (9.6 mL/s) than in the morning (8.0 mL/s). Grino et al.[7] reported marked variability in Qmax between machine-read and manually read values, and attributed this to artefacts not recognized by the machine. They concluded that manually reading the Qmax is more accurate. Van de Beek et al.[8] found considerable intra- and inter-observer variation in the interpretation of the uroflowmetry curves among 50 urologists and eight residents. More than half of the urologists interpreted the normal uroflow curves to be abnormal.

The effect of urethral instrumentation on Qmax has not been assessed in a controlled study; this is somewhat surprising, as pressure-flow studies, the ‘reference standard’ test for diagnosing obstruction, and during which urethral catheterization takes place, are completely dependent on the Qmax for the diagnosis of BOO. Furthermore, there is no consensus about patient selection for pressure-flow studies [9,10], making this test optional in evaluating patients with BPH [3,10]. This lack of consensus is partly because for most patients with BPH the test does not significantly alter management decisions or affect treatment outcome. In an attempt to be more selective, Jensen et al.[1] suggested the test should be used selectively, based on the Qmax. Using 15 mL/s as a threshold criterion, they showed that pressure-flow studies were useful in selecting patients for surgical therapy, and that treatment failure rates could be decreased to 8.3%. In general, a Qmax of < 15 mL/s is interpreted to be suggestive of BOO. Most BPH clinical trials use this Qmax as an inclusion/exclusion criterion. In the present study, up to 25% (eight) patients had a decrease in Qmax from > 15 to < 15 mL/s after instrumentation. Considering that Qmax is pivotal in clinical practice and trials, changes in its value after instrumentation may affect the diagnosis, patient selection, management decisions and results.

The variability in Qmax after instrumentation has been noted in two previous studies [11,12]. While in the present study the bladder volume was controlled by ensuring the same capacity at the time of the two uroflowmetry studies, this was not the case in the other studies. In one the Qmax was 3 mL/s lower when the bladder was filled transurethrally rather than suprapubically [12]; in the present study there was a similar mean decrease in Qmax of 3.6 mL/s. In the second study Qmax increased by 1.7 mL/s after urethral instrumentation at the time of the pressure-flow study [11], which is contrary to results of the present and other reports, and might be explained by variability such as the use of CO2 cystometry and changes in voided volumes.

The mechanism responsible for the changes in Qmax after urethral instrumentation is possibly related to an element of the associated tissue irritation and oedema. Another explanation would be a reactive increase in the tone of the bladder outlet, both the bladder neck and prostate. It is conceivable that the discomfort related to the instrumentation causes catecholamine release that acts on the adrenergic receptor-rich region of the bladder outlet.

In the present study every effort was made to avoid trauma and over-distension of the bladder during cystoscopy. We consider that the changes in Qmax after urethral instrumentation are more likely to be a consequence of changes in the bladder outlet rather than bladder contractility.

In conclusion, urethral instrumentation with flexible cystoscopy significantly changes Qmax; in about a quarter of patients it decreases from > 15 to < 15 mL/s after instrumentation. The present results indicate that a ‘false’ mean change in Qmax (favouring improvement) of + 27% (−23% to +139%) will result if investigators choose the uroflowmetry values after instrumentation when enrolling patients into studies of BPH therapies. Therefore, BPH study protocols should exclude uroflowmetry data obtained after urethral instrumentation; failure to exclude such data will lead to disproportionately better improvements in Qmax that are independent of the therapy.


maximum urinary flow rate


postvoid residual volume.