To investigate whether a mechanism of action of α-blockers on lower urinary tract symptoms (LUTS) involves improved perfusion of the LUT.
To investigate whether a mechanism of action of α-blockers on lower urinary tract symptoms (LUTS) involves improved perfusion of the LUT.
The accuracy of perfusion measurements using transrectal colour Doppler ultrasound (TRCDUS) and colour pixel density (CPD) was initially confirmed in a porcine model. Following this confirmation, measurements were taken from four healthy male volunteers and 19 patients with LUTS. The urinary bladder was filled slowly (50 mL/min) with 0.2 m KCl, which resembles the osmolarity of concentrated urine, and evaluated by cystometry. In parallel, TRCDUS and measurement of the CPD of the LUT were performed. The patients with LUTS were then treated with daily α-blocker (0.4 mg tamsulosin) for 5 weeks and urodynamic variables as well as perfusion were evaluated again.
In the healthy men, perfusion of the LUT increased considerably (157%) during filling of the bladder to a mean (sd) maximum cystometric capacity (Cmax) of 481 (28.9) mL. All the patients with LUTS had a reduced mean Cmax during filling with KCl at 322.4 (58.5) mL. The mean CPD in the urinary bladder and the prostate were only increased by 58.4% during filling with KCl. After α-blocker therapy the mean Cmax during filling with KCl rose to 382.5 (42.9) mL; furthermore, perfusion of the LUT measured by CPD was significantly increased (132.8%).
The present data strongly suggest that LUTS are associated with chronic ischaemia of the prostate and urinary bladder. α-blockers increase perfusion in the LUT and Cmax. These results might explain the therapeutic effects of α-blockers on LUTS.
transrectal colour Doppler ultrasound
colour pixel density
maximum cystometric capacity
laser Doppler flowmetry
region of interest
mean weighted difference
quality of life.
The exact aetiology of LUTS in humans is still poorly understood. In older men LUTS are commonly attributed to outlet obstruction from an enlarged prostate. It has long been assumed that LUTS originate from the prostate with obstruction causing obstructive symptoms, e.g. weak stream, hesitancy and intermittency. BOO in BPH is thought to be caused by two components: (i) a static component related to the increase in cellular mass reflecting the size of the prostate and (ii) a dynamic component related to variations in prostatic smooth muscle tone caused by stimulation of α1A-adrenoceptors . As the second step of the prevailing two-part theory of the development of LUTS, the bladder is thought to provoke storage symptoms, e.g. frequency, nocturia and urgency due to detrusor instability caused by bladder changes secondary to obstruction .
However, several findings have raised doubts about this theory. It has been shown that in more than a third of cases, LUTS are not associated with BOO [3,4]. Furthermore, symptom score questionnaires completed by women yielded identical scores compared with their age-matched male counterparts [5,6].
α1-blockers are widely used in the treatment of LUTS associated with BPH to reduce the dynamic component of BPH , different subtypes of α1-receptors have been identified in prostatic tissue, providing the basis for the development of subtype-selective ‘uroselective’α-blockers for treating LUTS [8–10].
But recent findings indicate that a reduction of the dynamic component of BPH by α1-receptor blockers cannot by itself explain successful relief from LUTS, particularly from the associated irritative symptoms [11,12]. α1-blockers are effective in patients with LUTS, irrespective of the presence of an enlarged prostate . It is well known that the maximum urinary flow rate as a standard urodynamic variable does not change dramatically after treatment with α-blockers . The site of action of these drugs might not therefore be exclusively the smooth muscle cells in the prostatic urethra or in the bladder, as has long been postulated in urological and pharmacological literature [9,14].
Furthermore, experimental data support the hypothesis that pelvic arterial insufficiency and chronic ischaemia are associated with marked changes in detrusor compliance and contractility of the bladder as well as changes of the prostate [15–17]. Recently, it was shown that bladder distension as well as the composition of urine, especially filling with 0.2 m KCl, which resembles highly concentrated urine, increased perfusion of the urinary bladder in humans [18,19]. Local regulation of human bladder blood flow and adaptive changes of perfusion are essential for normal function of the LUT .
To elucidate the aetiology of LUTS and the exact mechanism of action of α-blockers on the LUT, a sonographic-urodynamic study was performed. Besides urodynamic variables, vesical and prostatic blood flows were investigated using transrectal colour Doppler ultrasound (TRCDUS) and measurement of the colour pixel density (CPD). TRCDUS is an excellent and minimally invasive technique for measuring normal and pathological perfusion in the LUT [18,21,22].
Initially, the accuracy of TRCDUS and quantification of CPD to determine perfusion was tested in a porcine experimental model. The bladder of each of three female pigs was filled slowly three times with 0.2 m KCl, a concentration that resembles the osmolarity of concentrated urine. Laser Doppler flowmetry (LDF) as well as TRCDUS and measurement of the CPD were done simultaneously to enable comparison of the CPD measurements directly with LDF results .
LDF is generally accepted as the ‘gold standard’ for measuring perfusion in organs, including the urinary bladder [19,23]. LDF values are subject to very large tolerances because tissue site-to-tissue site and patient-to-patient variations are all additive variables. LDF therefore cannot provide absolute values of perfusion, although it can yield excellent relative measurements. Under proper monitoring techniques, changes detected by LDF are directly proportional to absolute volume flow changes in the sampled tissue.
This holds true for perfusion measurement using quantification of the CPD. To avoid variations from TRCDUS, all measurements were done in a standardized fashion by one experienced sonographer, using the same colour Doppler unit (Acuson Sequoia 512 unit) fitted with a high frequency transrectal US probe (7.5 MHz). All examinations were documented on magnetic optical disks. The digitally stored images were transferred to a computer workstation. With special computer software (Scion-Image-Analysis-Software, USA) the percentage of colour pixels detected sonographically in the region of interest (ROI) were calculated to determine the CPD during bladder filling. As the cross-sectional area of perfused vessels is measured, calculation of the CPD represents a highly reproducible and sensitive variable for determining perfusion. Repeated measurement of the CPD’s in selected ROI’s showed negligible variations within the same experimental settings in the pigs.
After establishing the accuracy of the CPD measurements in the porcine model and Review Board, TRCDUS and measurement of the CPD as well as cystometry were performed in four healthy volunteers and 19 patients with LUTS, validated by three experienced radiologists in reciprocal consensus. Written informed consent was obtained from all participants. In the present human study, again, 0.2 m KCl was used to fill the urinary bladder . The filling speed for all measurements was 50 mL/min. Parallel to the urodynamic variables recorded during each run, the CPD was measured in four arteries in the bladder neck and the prostate, at filling volumes of 0 mL, 100 mL and maximum cystometric capacity (Cmax) to measure and quantify the perfused vessels and, thus, total perfusion of the LUT (Fig. 1). Cmax was defined as the volume at which the patients experienced maximum urge and asked the filling to be stopped.
To avoid inter-individual variations, the CPD values were standardized to allow comparison of the many measurements in all patients; the mean CPD value measured at least three times in each different vessel in the bladder neck of the empty bladder and the prostate was defined as ‘starting point’ in each patient. The mean value was calculated from these measurements for further analysis. With this definition of a definite starting point, all inter-patient variations could be balanced, and subsequent increases in perfusion could be expressed as normalized proportional changes, described in mean-weighted differences (MWD) in percentages.
The mean (range) age of the patients was 62.3 (38–79) years and the mean total serum PSA concentration was 2.4 ng/mL. The mean (range) age of the four healthy volunteers was 61.2 (51–69) years. After initial evaluation, the 19 patients with uncomplicated LUTS were treated with tamsulosin (0.4 mg tamsulosin/day) for 5 weeks . The following conditions were considered as exclusion criteria for study enrolment: medical pre-treatment with an α-blocker, 5α-reductase inhibitor or anticholinergics; prostatic surgery or pelvic radiotherapy; uncontrolled diabetes; dementia or severe cardiac or respiratory failure; haematuria; recurrent UTIs or a residual urine volume of >150 mL. After α-blocker therapy, the perfusion measurements of the LUT as well as comparative cystometry were repeated. The mean IPSS, AUA-Quality of Life score (AUA-QoL) and 24-h frequency-volume chart values were evaluated before and after treatment.
For statistical analysis, the t-test was used to compare the perfusion of the LUT between the healthy young volunteers and the patients with LUTS, and to compare the changes of perfusion after treatment of the 19 patients with tamsulosin. A P ≤ 0.05 was considered to indicate statistical significance.
In the pig experiments, vessels in the bladder wall and bladder neck could be identified and blood flow measured easily. Blood flow in the LUT was increased 5-fold during filling with KCl, compared with that in the empty bladder. The CPD measurements of the LUT yielded similar results as those of LDF, considered as the ‘gold standard’ for perfusion measurements. It can therefore be safely assumed that the former method is accurate and reliable and has also the advantage of being a minimally invasive technique for quantification of changes of perfusion in the urinary bladder and prostate.
In the healthy volunteers the mean (sd, range) Cmax during filling with 0.2 m KCl was 481 (28.9, 453–512) mL. All cystograms showed normal compliance, no unstable contractions and normal intravesical pressure profiles. The mean (sd) increase of the CPD during filling with KCl was 157 (21%, Fig. 2).
In contrast to the healthy young men, all patients with LUTS had a strong pathological response to intravesical KCl. The mean Cmax during filling with KCl was 322.4 (58.5, 249–436) mL, which was statistically significantly reduced compared with the volunteers (P < 0.01).
The CPD values showed wide variations between patients. With the definition of the mean CPD value measured in the bladder neck of the empty bladder and the prostate as starting point, these inter-individual variations could be balanced. In general, perfusion of the LUT was considerably reduced in the patients with LUTS (Fig. 1).
Before α-blocker therapy, in the presence of KCl, the mean CPD increased by only 58.4% (Fig. 2) with distension reaching Cmax. After 5 weeks therapy with the α-blocker the only urodynamic variable that was changed by the medical treatment was the mean Cmax reaching a mean of 382.5 (42.9, 312–467) mL during filling with KCl. This variable was significantly increased after α-blocker therapy (P < 0.01; Fig. 3).
Furthermore, after α-blocker therapy the mean CPD values were markedly increased in the LUT (Fig. 1). In the presence of KCl the mean CPD value increased considerably by 132.8% at Cmax. The difference between the raw CPD values measured before and after α-blocker treatment was statistically significant (t-test two-tailed P < 0.01). The results of the comparative cystometry and measurement of the CPD after α-blocker therapy approached ‘normal’ values as measured in the healthy volunteers.
All the patients with LUTS responded very well to α-blocker treatment. Irritative symptoms were markedly improved after treatment with a mean IPSS and AUA-QoL value of 16.6 and 18.4 before treatment, and 8.8 and 10.2 after treatment, respectively. The 24-h-frequency and nocturnal-frequency decreased from 11.8 to 7.2 and 2.4 to 1.6, respectively, after α-blocker therapy.
This is the first study describing dynamic perfusion changes in men with LUTS and comparing these results with normal regulation of blood flow in healthy young men, using a minimally invasive sonographic technique that was initially evaluated in pigs. The present data strongly suggest that LUTS are associated with pathological perfusion in the urinary bladder and prostate.
Slow filling of the urinary bladder with KCl normally leads to a pronounced increase in perfusion in the LUT. By contrast, to measurements in healthy young volunteers and young pigs, increases of perfusion in the LUT from filling of the urinary bladder were much lower in patients with LUTS. This result is intriguing, given the fact that potassium is the classical ion of depolarization of cell membranes and that KCl is not used in routine urodynamic evaluation and classic pressure-flow studies. As human urine contains high concentrations of KCl, the urodynamic results and perfusion data during filling of the bladder with KCl can be directly compared with normal physiological regulation in healthy men and the pathological decrease of perfusion of the urinary bladder and prostate in patients with LUTS.
Age-dependent changes in the regulation of perfusion in the urinary bladder as well as changes of bladder urothelium might lead to higher excitability of afferent and efferent neurones as well as muscle cells in the LUT, as has been postulated previously [18,23]. This higher excitability would inevitably result in irritative bladder symptoms, the main component of LUTS.
Various conditions are associated with decreased perfusion of the LUT, including obstruction, ageing, menopause and interstitial cystitis [23–26]. Recently it was suggested that local hypoxia might be one of the triggers of stromal cell activation and subsequent hyperplastic growth in the prostate . The present clinical results strongly support the hypothesis that an age-related impairment of blood supply to the LUT plays a key role in the development of LUTS. This would also explain why the prevalence of LUTS increases with advancing age and that these symptoms develop with similar frequency in men and women [3–6]. Furthermore, chronic ischaemia of the LUT might be the cause of the significant increase in the volumetric amount of fibromuscular tissue in BPH [16,17].
During the last two decades, the therapeutic efficacy of α-blockers in the treatment of LUTS has been clearly shown in several clinical trials [14,28], and today α-blockers are the first-line medical treatment of LUTS. α-blockers are valued for rapid onset of action, effectiveness independent of prostate size and good therapeutic profile. The latter is supported by the development of selective α1-adrenoceptor antagonists and the excellent pharmacokinetics of these drugs .
The use of urodynamic studies at baseline and follow-up to evaluate changes in BOO in patients treated with α-blockers is still subject to controversy [30,31]. Most patients treated with α-blockers do not show substantial improvement of standard urodynamic variables [3,14], while on the other hand, irritative symptoms are greatly ameliorated.
Therefore, it has been postulated that other mechanisms of action of α-blockers must be involved in the successful treatment of LUTS in the elderly. Changes in α1-adrenoceptors expression in human blood vessels during ageing or extra-prostatic α1-adrenoceptors at the level of the spinal cord, ganglia and nerve terminals have been postulated to contribute to the pathogenesis of LUTS . Furthermore, a close relationship between ischaemia, obstruction, ageing and development of an overactive bladder has recently been thought to be involved in the pathogenesis of BPH and LUTS [16,17,33]. Ischaemia-initiated damages of intrinsic nerves, urothelium or smooth muscle cells, which are then even more susceptible to further damage by ischaemia, could eventually end in a vicious circle resulting in neuronal degeneration and bladder instability, which the patient then experiences as irritative bladder symptoms.
Clinical data supporting or disproving these theories are lacking. The second important finding of the present study, namely that α-blockers increase blood flow in the LUT and Cmax during filling with KCl, is of particular interest in this context. These results might, for the first time, explain the therapeutic effects of α-blockers on irritative LUTS, while standard urodynamic variables are not influenced. To date, the affect of treatment with α-blockers on perfusion of the LUT as well as Cmax during filling with 0.2 m KCl, which corresponds to highly concentrated urine, has not been addressed. But both parameters are strongly influenced by treatment with α-blockers. After α-blocker treatment, the values of Cmax during filling with KCl and CPD measurement during filling with KCl approached the ‘normal’ values measured in healthy volunteers. α-blockers cannot ‘rejuvenate’ the urinary bladder or the prostate, but as the present study shows, they can improve perfusion of the LUT, a mechanism that has been neglected to date. It must be emphasised that α-blockers are effective in treating LUTS although standard urodynamic variables are not influenced strongly.
Finally, in the Cochrane meta-analysis  4122 patients (aged 64 years) were systematically reviewed to assess the effects of tamsulosin in the treatment of LUTS. They stated that tamsulosin improved symptoms and peak urine flow relative to placebo. Indeed, regarding QoL nearly 40% of men receiving tamsulosin reported ‘moderate to marked improvement’ compared with 10% of men receiving placebo  and 90% of the treated men stated their symptoms were ‘much to slightly improved’ (vs 56% in placebo-arm; P = 0.03) . However, in marked contrast to the improved changes in QoL after α-blocker therapy the urinary flow measures were rather disappointing, as the MWDs were small with a 1.05 mL/s (9% improvement) for 0.4 mg tamsulosin.
Therefore, the present data emphasise the importance of the perfusion of the LUT for normal function, aetiology of LUTS and pharmacological treatment possibilities.
In the case of the pathophysiology of LUTS, a paradigm shift has already taken place in recent years . The present data strongly support the hypothesis that chronic ischaemia represents the main pathophysiological cause in the genesis of LUTS.
LUTS cannot be explained by simple mechanical models and evaluated by standard pressure-flow studies. In the future, exact evaluation of innervations, muscular function and perfusion of the LUT will be more important in the investigation of patients with LUTS. Further studies are needed to evaluate the exact mechanism of regulation of perfusion of the LUT and the exact site of action of α-blockers.
In conclusion, the present study provides new substantial insights into the pathophysiological cause of LUTS. In addition, it provides a plausible alternative explanation of how α-blockers act. Chronic ischaemia has been shown to play a key role in the multifactorial origin of the development of LUTS and in the treatment of LUTS with α-blockers. α-blockers increase reduced perfusion of the LUT and reduced bladder capacity, and these changes after treatment then lead to clinical improvement of LUTS.
The authors wish to thank Rajam Csordas-Iyer for valuable discussions and critical reading of the manuscript.