This review has five objectives: first, to identify systematically the salient literature on acute urinary retention (AUR); only one other review exists which was intended for a general rather than specialist urological readership , and addressed AUR only in adult men. Second, to classify the literature in a way that will be helpful to the reader; this is particularly important for the many case reports relating to AUR, which accounted for over half of the papers/abstracts identified. Third, using the published information, to try to understand the pathogenesis of AUR, which to date remains elusive. Fourth, to identify and appraise high-quality research studies which could influence the way that men with AUR are managed. Fifth, to identify information not so far published thus suggest a research agenda and order of priority.
AUR is a condition characterized by a sudden inability to micturate that is invariably but not always painful; painless AUR is rare and is usually associated with CNS pathology. Over 10% of men in their seventh decade will experience AUR over a 5-year period  and the risk increases to one in three over 10 years. AUR results in prostatectomy in 24–42% of men presenting in Britain and North America [3–8].
The search strategy
A Medline search (from 1966 to 1998) was conducted using the words ‘acute urinary retention’ or ‘acute retention of urine’; 446 records in English were identified. To include recent work, manual searches were also conducted of abstracts presented at the four leading urological congresses of the last 2 years (the AUA, Société International d’Urologie, European Association of Urology and the British Association of Urological Surgeons (BAUS)). By far the majority of papers were case reports describing unusual cases of AUR; these were then classified into broad groups.
AUR is both a significant public health issue and of consequence to the affected individual. In the past, AUR was considered an absolute indication for prostatectomy . Those patients who managed to avoid an operation by having a successful trial without catheter (TWOC) were at high risk of having prostatectomy within a year . Moreover, recent observational work has shown that a prostatectomy resulting from AUR is associated with increased morbidity and an increased risk of death during and after surgery . A sample of 1242 men who presented with AUR and underwent prostatectomy had an increased risk of perioperative complications and an excess risk of death at 30 days (relative risk, RR, 26.6; 95% CI 3.5–204.5) and at 90 days (RR 4.4, 95% CI 2.5–7.6) after surgery. These men were older, had larger prostate glands, higher American Society of Anesthesiologists grades and evidence of renal impairment, but these factors did not totally explain the excess risk. AUR was significantly associated with the requirement for a second operative procedure, usually for bleeding. A higher percentage of men with AUR failed to void after prostatectomy (9.2%), which became permanent (0.9%), compared with men who were operated for symptoms alone (2.3% and 0.1%, respectively).
Aetiology and pathogenesis
The presumed aetiologies of AUR occurring in 310 men that were admitted to a teaching hospital during a 2-year period  are listed in Table 1. Broadly, the causes of AUR can be classified into three categories ( Table 2). The first relates to any event or process which increases the resistance to the flow of urine. This can be either a simple mechanical obstruction, e.g. a urethral stricture, or a dynamic obstruction resulting from an increase in smooth or striated muscle tone, or both. In the second, AUR might result from an interruption of either the sensory innervation of the bladder wall or the motor supply of the detrusor muscle or secondary to the influence of drugs. For instance, a segmental spinal reflex of anovesical inhibition of the detrusor has been shown, with the pudendal nerve serving as the afferent limb and the parasympathetic as the efferent limb, but with no sympathetic involvement . This may be the mechanism that results in AUR after painful anorectal stimuli, e.g. anal fissures, after haemorrhoidectomy, prolapsed thrombosed haemorrhoids and anal intercourse. The third category relates to any situation which either permits or causes the bladder to over-distend. This situation is most commonly encountered after an operation performed under general anaesthesia [2, 14] in which the patient was not catheterized and the bladder allowed to distend. The condition is often complicated by any combination of the following: opiates or opioids that decrease the sensation of bladder fullness; anticholinergic medication inhibiting detrusor contractility; and the high α-adrenergic tone that invariably follows an operation may increase urinary sphincter tone.
Table 1. Aetiology of AUR in 310 men in a 2-year period
Carcinoma of the prostate
Table 2. Examples of case reports describing unusual presentations of AUR
Suggesting mechanical obstruction to bladder emptying
Severe urethral inflammation after exposure to a nonoxynol-9-based vaginal contraceptive pessary during unprotected intercourse
Published reports describe several processes that are thought by some authors to initiate a sequence of events that results in AUR. These include prostatic infarction [15–17], α-adrenergic activity [18–26] and neurotransmitter modulation [27–33].
The relationship between prostatic infarction and AUR was evaluated in two groups of 100 prostates for any histological evidence of infarction . One group consisted of consecutive patients with large prostates and who were in AUR, and the other of patients with BPH who underwent elective surgery. All patients underwent open enucleation of the prostate gland. In the group with AUR, 85% had prostatic infarcts while only 3% of the patients with BPH had. The authors postulated that prostatic infarction could result from distortion of the intraglandular vascular supply , trauma (including instrumentation), infection, stasis, thrombosis, atherosclerosis, embolism and BPH . The exact mechanism in the production of AUR was not previously determined. There is now a hypothesis whereby prostatic infarction leads to a neurogenic disturbance occurring at the prostatic urethra (a peri-ischaemic zone), which results in a failure of relaxation and subsequent AUR.
A recent study examined resected tissue from a group of ‘matched’ patients; 35 patients presented with AUR, 35 had symptoms alone and all underwent TURP . Prostatic infarction was noted histologically in 9% and 3% of the two groups, respectively but the differences were not statistically significant. In that study the proportion of cases reported to have prostatic infarction was small. In another study it was even less; in a community-based study  eight of 57 men with AUR underwent TURP within 6 months of the event (median 12 days). In these circumstances no foci of prostatic infarction were reported, but this might be because of the method of data collection, as it was a retrospective study and not specifically designed to assess prostatic infarction. An alternative explanation might lie with the risk factors leading to AUR; in that study, nearly half of the AUR events were associated with surgical procedures and up to 90% of these were performed under general anaesthesia. The authors concluded that in their chosen setting general anaesthesia had a precipitating role in a large proportion of cases of AUR. It is possible that prostatic infarction is not part of the pathogenesis of AUR after general anaesthesia.
The human prostate, and particularly the prostatic capsule, is known to be rich in α-adrenergic receptors , and the use of a rapid acting α-blocker has been shown to relieve AUR in some cases . Urodynamics performed in patients who presented in AUR but before bladder decompression showed some interesting features . The first was that the intravesical pressure was low, at 30–40 mmHg, and could be increased by abdominal straining. The second was that the bladder neck showed no point of high pressure, in contrast to a patient without AUR. The third was that the point of maximal urethral pressure was in the prostatic urethra and not in the external sphincter, which seemed to show no activity at all. Immediately after bladder decompression the intravesical pressure decreased to baseline, the pressures in the region of the bladder neck and prostate remained similar but the usual peak pressure at the level of the external sphincter reappeared. The administration of intravenous phentolamine to provide an α-adrenergic block resulted in a decrease in pressure along all segments of the posterior urethra, including the bladder neck, the prostate and the external sphincter.
It is generally agreed that bladder distension leads to an increase in sympathetic activity [21–23]. Experiments have confirmed that bladder distension results in a rise in intravesical pressure but beyond a certain point the pressure declines [24, 25]. The low pressure in the bladder neck during AUR may be caused by a mechanical stretching of the detrusor fibres pulling the bladder neck open or by a reciprocal relaxation of the bladder neck in the presence of the severe urge to void via spinal reflexes . As the bladder overfills, a reflex relaxation of the pelvic floor occurs in conjunction with a reduction in external sphincter and bladder neck pressures.
Taken together, these findings indicate that some cases of AUR are associated with a rise in the prostatic intraurethral pressure through an increase in α-adrenergic stimulation. Prostatic infarction and other processes such as acute prostatitis contribute to this process. Prolonged bladder distension in retention may directly affect detrusor function either by bladder wall ischaemia or axonal degeneration, resulting in the failure of detrusor muscle to initiate bladder contraction. Supplementary stimulation of the prostatic α-receptors caused by bladder distension, together with the secondary decrease in the intravesical pressure, perpetuates the retention.
The ratio of stromal to epithelial tissue
Several authors have been interested in the relative constituents of prostatic tissue and their effect on the natural history of BPH . In a retrospective study involving 10 aged-matched patients (five in retention and five not), the ratio of stromal to epithelial tissue, as determined by an investigator unaware of the source, was consistently decreased in patients with AUR . This may help to explain why finasteride, acting on epithelial tissue, can decrease the incidence of AUR. This finding was confirmed by a prospective controlled clinicopathological study comparing 35 patients who presented with AUR secondary to BPH and 35 with symptomatic BPH. All 70 patients underwent TURP and the prostatic chips were examined by one histopathologist unaware of the mode of presentation. The median age and age range, and prostatic weight, were comparable in the two groups. Those presenting with AUR had a significantly higher epithelial component (71% vs 60%) than had patients with symptomatic BPH and no AUR (P < 0.01) , suggesting that a differential rate of growth of the epithelial over the stromal component results in higher rates of AUR. The sequence of events could be as follows: differential epithelial/stromal growth predisposes the prostate to infarction through effects on the vascular supply; prostatic infarction leads to swelling and a rise in intraprostatic urethral pressure; this in turn stimulates efferent α-adrenergic stimulation, which results in a further rise in intraurethral pressure sufficient to exceed maximal detrusor pressure, augmented by a Valsalva manoeuvre.
Neurotransmitters and AUR
In addition to the classical neurotransmitters acetylcholine and noradrenaline, several other neuromodulators have been identified in the nerves supplying the bladder and urethra. Vasoactive polypeptide (VIP), neuropeptide Y (NYP), substance P (SP), enkephaline, somatostatin and nitric oxide have been implicated in AUR [27–29]. Reversible and irreversible changes were seen when AUR was induced experimentally in animals. Changes in nonadrenergic, noncholinergic neurotransmitters were reported in rats . During a forced diuresis against an obstructed urethra, there was a depletion of VIP, NYP and SP in the bladder wall of rats within 3 h; the depletion appeared to be transient and returned to normal soon after the distension was relieved. NYP and VIP may act as agonist and antagonist regulating local blood flow, and SP may influence the release of acetylcholine and noradrenaline; it may also modulate the release of other neurotransmitters . Not all the changes that have been reported were reversible. Experimental studies in guinea pigs have shown that if the cycle of AUR is not relieved, there was evidence of cell death in the intramural ganglia embedded in the bladder wall within 24 h, and this was established by 48 h . The existence of intramural ganglia in the human bladder has been documented  and are considered to constitute part of the pelvic ganglia, which integrate signals from both the sympathetic and parasympathetic bladder supply ( Fig. 1).
Risk factors for AUR
Reports suggest that the 10-year cumulative incidence rates of AUR are 4–73%; this broad range is a good example (if more were needed) that data drawn from small well-defined populations cannot readily be generalized. Many of these estimates come from single-centre case series, are not population-based, and therefore contain both known and unknown biases. To determine the incidence of AUR a longitudinal design is needed, preferably using a population-based approach; there is now such a study and the first results on the occurrence of AUR have been reported . In this study of 2115 community-dwelling men (in Minnesota, USA) not seeking consultation about their voiding function, the incidence of AUR varied by age, the severity of LUTS, peak urinary flow rate and prostate size. The cumulative incidence of AUR (the proportion of individuals who develop the condition over a certain period) increased with age. AUR in men who were < 60 years old was rare; men who were 70–79 years old had eight times the risk of AUR (95% CI 3.7–16.4) than had men in their fourth decade. At 5 years, the cumulative incidence increased from 1.6% among men 40–49 years old at baseline to 10% for men 70–79 years old at baseline. It can be calculated that a 60-year-old man would have a 23% probability of experiencing AUR if he were to reach the age of 80 .
In that study, men with moderate to severe urinary symptoms (IPSS > 7) had four times the risk of AUR than had men who were either asymptomatic or had mild symptoms (IPSS ≤ 7). The 5-year cumulative incidence increased from 1.5 to 6.8% in men with mild symptoms and from 2.1 to 13.8% for men with moderate to severe symptoms. For men younger than 60 years, repeat urination (within 10 min of urinating), and stopping and starting again during micturition, were significantly associated with a risk (RR, 95% CI) of experiencing AUR (3.7, 1.4–9.7, and 4.1, 1.7–10.1, respectively). By contrast, older men (≥ 61 years) appeared to be at twice the risk of experiencing AUR if they reported any symptoms at all. A weak urinary stream (smaller flow or weaker force of urinary stream) was the symptom most strongly associated with AUR (3.4, 1.7–7.0). An enlarged prostate (> 30 mL) was associated with AUR (3.0, 1.0–9.0) and a peak urinary flow rate of < 12 mL/s was also associated with AUR (3.4, 1.7–7.0) .
It is difficult to compare results from the above community study with those published previously, many of which comprised men with LUTS and BOO, and therefore of a different population group. The ease of access for consultation, the availability of health insurance, the varied methods of managing AUR, and the population studied could play a significant role in determining the outcome. The community study  was based essentially on white men, who were predominantly middle class, and almost half of AUR events were associated with surgical procedures; nearly 90% of these were performed under general anaesthesia. Extrapolation to another race or socio-economic group may be inappropriate. The interpretation and extrapolation of community data from patients not seeking medical management to men diagnosed with BPH must be cautious. In a large study of 2084 patients with an IPSS of ≥ 13 and a peak flow rate of ≤ 15 mL/s, the IPSS was not predictive of AUR . However, a higher serum PSA level at baseline, which was associated with a larger prostate volume at baseline, was associated with AUR . In a recent study examining the placebo arm of a large pharmaceutical study for men with enlarged prostates and LUTS, the PSA level was the strongest predictor of both AUR and the need for prostatic surgery . In a study of men with moderate to severe symptoms (IPSS > 7) in which 64 of 331 men had AUR, patients with a transition zone index (transition zone volume/total prostate volume, as measured by TRUS) of > 0.65 had a high risk of developing AUR (with sensitivity of 89% and a positive predictive value of 63%) .
AUR is uncommon in children, with most developing AUR secondary to cystitis, or brought on by infection or after catheterization ; other causes are listed in Table 3. In children, after decompression and treatment with antimicrobial agents in cases of cystitis, further investigations (renal tract ultrasonography, micturating cystography) are warranted to evaluate the renal tract anatomy and to exclude reflux or hydronephrosis.
Table 3. Causes of AUR in children and women
Congenital obstruction (anterior and posterior urethral valves, urethral polyp or atresia, ectopic ureterocele, hydrometrocolpos)
Acquired urethral obstruction (blood and/or tissue after surgery)
In a study of six Copenhagen hospitals serving a population of ≈700 000 over a 9-month period , the incidence of AUR in women was seven per 100 000 per year and the sex ratio (F:M) was 1:13. AUR in women may develop after surgery, or secondary to cystitis, infravesical obstruction, insufficient detrusor function, obstetrical and gynaecological conditions, or hysterical reactions  ( Table 3). Fowler et al. and Goodwin et al. described a syndrome in young women aged 15–30 years (usually in their 20s) who presented with painless AUR associated with residual volumes usually > 1000 mL. These young women commonly had a long history of voiding difficulty, the most commonly reported being infrequent voiding (< 2–3 times/24 h). All these women had neurological assessments that gave findings within normal limits; all underwent MRI of the pelvis, which again showed no abnormality. Some had the hirsutism/acne/menstrual irregularity complex associated with polycystic ovarian syndrome. Classically the urethral sphincter EMG reveals abnormal bursts of activity (decelerating bursts and complex repetitive discharges). This abnormal activity is typical of the repetitive, circuitous, self-excitatory activity that results from ephaptic excitation between muscle fibres. These bursts of depolarizing activity impair normal relaxation of the striated urethral sphincter, leading to incomplete bladder emptying and AUR. Because a muscle showing this abnormal EMG activity would be expected to increase in bulk through hypertrophy, TRUS was performed in 14 women with ‘Fowler’s syndrome’ and compared with that in 20 age-matched controls . The mean ( sd) volume of the sphincter in patients with Fowler’s syndrome of 3.05 (0.23) cm3 was significantly greater than that in the control group, of 1.30 (0.09) cm3 (P < 0.001), providing collateral evidence that the abnormal EMG activity is a significant factor in the pathophysiology of Fowler’s syndrome.
In most women, under-active detrusor function was the primary cause of AUR, at times preceded by a provocative event such as acute cystitis. Because of the underlying aetiology, a large proportion of women have been reported to have long-term sequelae . Half developed recurrent retention, whilst ≈39% had persistent voiding symptoms. The best long-term prognosis was in patients with correctable infravesical obstruction and in those with few urinary symptoms before AUR. All women require a pelvic examination, careful neurological assessment, urine culture and ultrasonography of the pelvis as the minimum investigations. The prognosis for the ability to void in the early stage after AUR was similar to that in men.
The management of AUR
AUR is treated by catheterization, in most cases urethrally mainly because it is easy and quick. Not all healthcare professionals (Accident & Emergency doctors, GPs or district nurses) are familiar with the insertion of suprapubic catheters, which are most often reserved for those cases where the urethral route has failed. Key points in the management of AUR in men are listed in Table 4.
Table 4. The management of AUR in men
Decreased risk of UTI and urethral stricture formation
permits a trial of micturition without removing the catheter
Patients catheterized and sent home
No adverse effect on outcome compared with patients who were hospitalized
May lead to avoidance of prostatectomy in 23% of patients with AUR
More likely in those aged >75 years
or those with drained volume >1000 mL or with detrusor contraction of <35 cmH2O
Associated with a higher percentage of successful TWOC
44% after a single catheterization 51% after 2 days 62% after 7 days
Three-fold higher success rate of a TWOC than in patients not on α-blockers
For AUR rather than for LUTS, is associated with a higher risk of complications during and after surgery blood transfusion hospital mortality
Suprapubic vs urethral catheterization
One of the most informative studies comparing these two methods of bladder drainage was undertaken by Horgan et al.. Using an unrandomized design they compared the outcome for 3 years after catheterization (suprapubic or urethral) in 86 consecutive patients. Of the 30 patients catheterized urethrally using a latex Foley catheter 12 (40%) had a UTI, whereas of the 56 who had suprapubic catheters placed, only 10 (18%) had a UTI (chi-squared, P < 0.05). Five (17%) of the patients catheterized urethrally subsequently developed urethral strictures, compared with none in the group catheterized suprapubically (P < 0.01). This may be partly explained by the use of latex catheters rather than the more inert biomaterials, e.g. silicone or hydrogel-coated catheters. A TWOC in seven of 11 patients who had been catheterized urethrally proved unsuccessful. The use of suprapubic catheterization may reduce the risk of UTI, urethral stricture formation and allows for a TWOC without the discomfort of catheter removal and the trauma of re-insertion. Suprapubic catheterization may be more comfortable for patients, easier to manage and more cost-effective .
Concerns about the safety of the suprapubic route seem misplaced if personnel with adequate training place the catheter . However, complications associated with its use have been reported, e.g. dislodgement , bowel perforation , acute peritonitis secondary to infected extravasated urine arising from the cystotomy site , late intraperitoneal posterior bladder wall perforation , migration of the suprapubic catheter to the anterior urethra , and implantation metastasis in a case of bladder cancer .
Hospitalise vs ‘catheterize and send home’
Once the patient is catheterized, local practice usually dictates whether the patient is hospitalized or sent home and later assessed in the outpatient department. In a large study of prostatectomy in the UK , 649 men (52.3% of the cohort) underwent prostatectomy after their admission to hospital with AUR. Most of the remaining men were sent home with a catheter (559, 45%) before a planned prostatectomy; half of these men underwent surgery within one month and most (88%) within 3 months. There were no differences in age, ASA grade, serum creatinine level or other preoperative variables between men admitted and those catheterized and sent home. The decision to admit or send home was based on local policy. Men hospitalized as a result of AUR stayed a mean of 5.0 days longer (95% CI 4.1–6.0) than men who were catheterized and sent home. The extra days comprised an extra 4.1 days (95% CI 3.4–4.8) before surgery and an extra day (95% CI 0.5–1.6) after surgery. Men who were admitted with AUR were more likely to require a second procedure for bleeding (4.6% vs 1.7%, P = 0.004, RR 2.8, 95% CI 1.3–6.0). Perioperative antimicrobial agents were given in 86% of patients in both groups. Uncomplicated urinary infection was more common after surgery in men who were catheterized and sent home (15.6% vs 9.5%, chi-squared P = 0.002) and consequently, more men in this group received antimicrobial agents after surgery (53.7% vs 45.9%, chi-squared P = 0.01). Prolonged catheterization leads to bacterial colonization of the urinary tract and might increase the risk of sepsis. However, no increased risk of major infective complications was detected. More men who were admitted with AUR were unable to pass urine postoperatively (10.9% vs 7.2%, P = 0.04); the symptomatic outcome and mortality were similar in both groups. The study concluded that it is safe for a man with AUR to be catheterized and sent home to await an elective prostatectomy in the next few weeks, provided he is not uraemic, septic, ill or dehydrated.
A trial without catheter
It is becoming increasingly accepted (for it has not always been) that a trial of micturition should be part of the management of a man presenting with AUR. In a study of 60 patients with AUR in whom the catheter was removed no later than 48 h after catheterization, 17 patients (28%) voided successfully . When these 17 patients were followed up for 6 months, six had required prostatectomy for severe LUTS while 11 had no or only minor symptoms. In that study, the authors showed that about a fifth of patients might avoid surgery if a TWOC was implemented as policy. Others have shown that up to 23% of patients in AUR may not require surgery . Possibly the main advantage of the suprapubic catheter is that a trial of micturition can be more easily undertaken.
Clearly, asking all men to undergo a TWOC might be wasteful of resources if only one in five men is likely to benefit. To assess which patients will void after a TWOC, 76 men admitted with AUR who underwent such a trial were assessed ; 18 (24%) were able to pass urine. A TWOC was found to be beneficial in three groups of patients, i.e. those with a UTI with no previous obstructive symptoms, those with gross constipation and those recently commenced on anticholinergic drugs.
If a TWOC was initially successful, 56% of men experienced recurrent AUR within a week, 62% after a month and 68% after a year. Patients with a retention volume of > 500 mL had a 3.6 times higher risk of recurrent retention and 2.7 times higher surgical treatment rate than patients with a volume of < 500 mL (P = 0.008 and P = 0.001, respectively). Patients with no provocative situation had a risk 1.9 times higher for recurrent retention and a 1.7 times higher surgical treatment rate than patients with a provocative situation (P = 0.02 and P = 0.009, respectively). Men who produced a maximum flow rate of > 5 mL/s were less likely to experience repeat AUR (7%). In contrast, men with poor flow rates (< 5 mL/s) were very likely to experience repeat AUR (75%). The volume of voided urine also appeared to be important. The ability to void at least 150 mL was associated with a significantly lower recurrent retention (8%) than in those with a voided volume of < 150 mL (60%). In addition, a residual volume of < 100 mL was associated with low risk of recurrent retention (P = 0.03). In that study, in which 173 of 223 patients with AUR were treated by simple transurethral catheter emptying and the remainder by an indwelling catheter, two-thirds of the patients developed recurrent retention; the risk was highest during the first few days after the retention period . Several factors were associated with failure, i.e. being older than 75 years, a drained volume of > 1000 mL and a detrusor contraction of < 35 cmH2O .
The duration of catheterization
Is there an optimal period of dependent drainage that should elapse before undertaking a TWOC? Djavan et al. addressed this question by randomizing men who presented in AUR into one of three groups: ‘in-and-out’ catheterization, and dependent catheter drainage for 2 or 7 days. On withdrawing the catheter, 44%, 51% and 62%, respectively, voided successfully. The authors found that differences in IPSS, total and transitional zone prostate volume and serum PSA level were not associated with the failure to void successfully. However, younger men (< 75 years old) with a retention volume of < 1000 mL and men who achieved maximum detrusor pressures after catheter removal of > 35 cmH2O were more likely to void after removal of the catheter. Patients who had retention volumes of > 1300 mL benefited most from prolonged catheterization.
α-Βlockers and a TWOC
A prospective, multicentre, randomized, placebo-controlled, double-blind study is currently recruiting patients to evaluate the effect of α-blockers on the success rate of a TWOC. A preliminary report was presented recently for the first 39 patients . Patients were randomized to receive either 5 mg of Alfuzosin (sustained-release) or placebo twice daily, the catheter was removed on the second morning of treatment and medication discontinued after four doses. The two groups were well matched for age, residual volume of urine and prostate size. Of the patients who received the α-blocker, 59% voided successfully in the TWOC, whilst only 18% of the placebo group voided successfully (P = 0.02).
In a multicentre randomized trial of men with moderate symptoms, 280 men were randomized to undergo surgery and 276 men to watchful waiting . Men managed by watchful waiting were eight times more likely to develop AUR than men treated with TURP (RR 0.12, 95% CI 0.02–0.98). Prostatectomy undertaken in a man with moderate symptoms substantially reduces the risk of AUR over the subsequent 3  to 5  years.
Men who fail their TWOC and who are fit for surgery will usually elect to undergo prostatectomy. Some physicians regard AUR and previous LUTS as an absolute indication for prostatectomy . However, prostatectomy performed as a result of AUR rather than for LUTS alone was associated with a higher risk (RR, 95% CI) of intraoperative complications (1.8, 1.3–2.5), blood transfusion (2.5, 1.8–3.3), postoperative complications (1.6, 1.2–2.0) and hospital mortality (3.3, 1.2–9.3) . The mean follow-up was 2.8 years and surgery was not associated with impotence or urinary incontinence.
Clean intermittent self-catheterization
For AUR initially treated by catheterization, CISC can offer a useful alternative to having a long-term indwelling catheter. CISC was shown to be effective in patients with multiple sclerosis in whom bladder involvement was up to 73–90%. Other conditions presenting with a neuropathic bladder, e.g. spina bifida, intravertebral disk lesions, spinal injuries or tumours, and patients with atonic bladders, could also benefit from CISC . CISC is a safe and simple technique, and patients with severe disability in wheelchairs, with paraplegia, anaesthetic perineum, spinal deformity, intention tremor, mental handicap, old age and blindness, were able to master CISC. In a group of 11 elderly women aged 70–86 years who developed AUR after operative fixation of a lower limb fracture, CISC was successful in 10 . These elderly women performed CISC 2–3 times daily for 9–40 days as inpatients and one patient continued CISC at home for 3 months. CISC was also useful in the treatment of women who developed AUR after vaginal delivery . In this group of patients, a period of catheterization usually resulted in resolution of the problem.
Future trends in AUR
There are four areas where advances are possible: prevention, secondary prevention, better systems of bladder drainage and improvements or alternatives to the standard treatment, i.e. prostatectomy.
AUR can be prevented; in a randomized, double-blind, placebo-controlled trial, men randomized to finasteride, a drug which reduces the size of the prostate by inhibiting the formation of DHT from testosterone, had a 57% lower risk (95% CI 40–69) of having AUR over 4 years than men receiving placebo . The finding of a consistent decrease in the ratio of stromal to epithelial tissue among patients with AUR [34, 36] may help to explain why finasteride, acting on epithelial tissue, is able to decrease the incidence of AUR. Because AUR was a relatively infrequent event in this group, the reduction in absolute risk of AUR occurring was 7% (99/1503) in the placebo group to 3% (42/1513) in the finasteride group. Interestingly, men who had spontaneous (no obvious cause) AUR whilst on finasteride were less likely to require prostatectomy than men receiving placebo (33% vs 72%). Because AUR is relatively infrequent and not life-threatening, few would at present argue that there should be a widespread prevention programme. To prevent a single episode of AUR or prostatectomy, 15 men with pre-existing urinary symptoms would have to be treated for 4 years to prevent one event. Selective use in men with known risk factors (moderate to severe LUTS, large prostates and poor urinary flow rates) is probably warranted. There may be a future role for secondary prevention. Further study is required to determine whether finasteride can prevent recurrent retention in a man who has had an episode of AUR and a successful TWOC.
As described, once AUR has occurred, men who received α-blockers were almost three times as likely to void successfully after catheter removal than men who received placebo during the period of catheter removal. In addition to the simple relaxation of muscle tone, α-blockers have been shown to alter contractile protein gene expression  and induce apoptosis in both epithelial and stromal cells in the prostate without affecting cell proliferation . Although this observation is unlikely to account for the therapeutic effect observed after a few doses, it may have a role in the long-term secondary prevention of AUR.
Another potential area for research lies with endothelin antagonists. Human prostate tissue produces endothelin and in the presence of smooth muscle, contractions have been observed which are not inhibited by α-blockers. Recently, a new orally active endothelin antagonist was found to inhibit endothelin-induced prostatic contractions in a dose-dependent way .
The implantation of metallic coils of wire into the prostatic urethra to relieve obstruction was first described by Fabian in 1980 . Since then, a variety of prostatic stents has been introduced, which are broadly classified as permanent and temporary. The permanent stents include the Titan, Urolume Wallstent, Memotherm and Ultraflex; the temporary stents include the Uro-spiral, Prostakath, intraurethral catheter, ProstaCoil, Memokath, Biofix, Barnes stent and Trestle . Permanent stents have wide spaces between their wires, allowing tissue ingrowth and permanent incorporation of the stent into the prostatic mucosa. Temporary stents remain in the urethral lumen, holding it open rather than becoming incorporated into the surrounding tissue. The loss of contact with urine in permanent stents reduces the risk of encrustation and infection, and the larger internal lumen permits subsequent endoscopic manoeuvres. However, the tissue response can result in luminal narrowing or even occlusion and the difficulty in removing the permanent stent is of concern to many urologists. Furthermore, the biocompatibility of the material, its potential carcinogenicity secondary to a chronic inflammatory reaction, tissue interaction and migration of corrosion particles are features yet to be resolved . The advantages of temporary stents include ease of insertion under local anaesthetic and relatively lower cost. Their disadvantage is that they are in contact with urine, leaving such a device prone to displacement, infection and encrustation. Except for the biodegradable stent, all the temporary stents should either be removed or replaced every 6–36 months .
Traditional catheterization might become unnecessary; the race is on amongst instrument makers to develop a device that will relieve AUR without the complications of encrustation, migration and infection, and simultaneously allow the bladder and external (voluntary) sphincter to function normally. These devices work by stenting open the bladder neck and prostate, thereby reducing the pressure required to initiate micturition. At present, such devices are slightly more difficult to place than a standard urethral catheter, but perhaps this will not be so for long. It is possible that such a device could be placed by a GP or in an accident and emergency department, allowing the patient to be discharged soon afterwards, relieved, continent and able to empty his bladder with no difficulty. We have not yet arrived at this point and will need to evaluate these devices carefully before their use becomes widespread; such trials are now being discussed.
Like catheters, temporary prostatic stents need to be removed at some time or they will become encrusted. At present they are extracted by pulling on a string lying in the urethra. However, biodegradable catheters that dissolve with time are also being considered, but presently are at an earlier stage of development. These could be used in a man with AUR and allowed to dissolve over the subsequent weeks, during which pharmacological agents could have more chance to work. Hospitalization would be avoided and the need for prostatectomy almost certainly reduced. A biodegradable self-reinforced poly DL-lactic acid spiral stent has been described to be safe and effective for the treatment of postoperative urinary retention after visual laser ablation of the prostate (VLAP) in a randomized study . Further studies are required to optimize the design, tailor the degradation sequences, incorporate properties such as radio-opacity and produce a range of biodegradable devices that degrade at different rates for different circumstances.
The interaction of urine with biomaterials has presented several challenges to overcome. Desirable characteristics of synthetic biomaterials used in the urinary tract include complete inertness, durability, partial flexibility, partial rigidity, resistance to encrustation, workability, sterilisability, permanence of physical characteristics and the ability to resist biofilm formation (protein adsorption and bacterial adhesion) . There is much to be done before a truly biocompatible stent becomes available. Indeed, more than a thousand years since Rhazes (a Persian scholar and physician in the ninth century CE) commented on the best type of material from which to construct a catheter, there remains disagreement about the form and substance of endourological devices. There is currently no completely inert biomaterial which is able to resist protein adsorption, bacterial adherence, encrustation and that causes no tissue reaction; moreover, the urinary system presents an unstable chemical environment, with supersaturation of uromucoids and crystalloids at the interphase between the material and urine, which creates a significant problem for long-term biocompatibility and biodurability of devices within this system. The design of stents can be improved and further systematic research needs to be conducted at the basic science level to device testing and clinical trial stages to minimize the complications associated with the use of a prostatic stent.
Alternative treatment modalities to prostatectomy
Clearly the role of prostatectomy as the best treatment for refractory AUR is being challenged. Interstitial treatments which involve the delivery of heat through a needle have been used with moderate success in AUR . Transurethral balloon laser thermotherapy (TUBAL-T) of the prostate was performed in 12 patients with BPH and urinary retention who were poor candidates for prostatectomy . Their mean (range) age was 78.9 (66–93) years and the mean duration of bladder catheterization before the procedure was 11 (2–48) weeks. Laser energy was applied with a Nd:YAG laser balloon under local anaesthetic. Spontaneous voiding was achieved in all patients at a mean of 2.8 (1–7) days after TUBAL-T and 75% of patients continued to void spontaneously in the long-term. The authors concluded that TUBAL-T was a safe and effective alternative to treating unfit patients with BPH in urinary retention. Holmium:YAG laser resection of the prostate (HoLRP) was performed in 36 men who presented in AUR . The mean (range) duration of postoperative catheterization was 1.5 (1–8) days and improved voiding was sustained through 6 months of follow-up. The long-term failure rate of urinary retention was 5.6%. The authors concluded that HoLRP was an effective surgical treatment for patients with BOO presenting in AUR. In another study, Nd:YAG laser coagulation prostatectomy was performed in 67 men (mean age 71 years) who presented with acute or chronic urinary retention . The median postoperative duration of catheterization was 10 days, the short-term failure rate was 6% and the long-term failure rate 9%. Nd:YAG laser transurethral evaporation of the prostate (TUEP) was performed in 22 consecutive patients presenting with urinary retention caused by BPH . All patients failed at least one voiding trial and the mean duration of catheterization before the procedure was > 30 days. Two patients were lost to follow-up and TUEP failed in two. The mean time to catheter removal was 3.5 days and all patients who successfully underwent TUEP were urinating spontaneously by 10 days. High-energy VLAP was performed in 17 men (mean age 69 years) in urinary retention ; the mean follow-up was 12 months and 82% were satisfied with their outcome. The authors concluded that high-energy VLAP was an effective procedure for the relief of BOO in men with AUR [80, 81]. Transurethral needle ablation of the prostate (TUNA) was performed in a pilot study of 10 patients in urinary retention . After treatment, nine patients voided at a median of 3 days but a further two required TURP. The authors concluded that their early experience was encouraging and that TUNA was an effective day-case procedure which could be undertaken with no anaesthesia.
However, currently prostatectomy has not been superseded; no doubt there will be further challenges to TURP but it remains the ‘gold standard’ with which all new procedures will need to compare favourably before they gain widespread acceptance.
There are several areas in AUR worthy of further research. The first relates to prevention by pharmacological means. The role of α-blockers in the primary prevention of men with moderate prostatic symptoms and in the secondary prevention of men who have successfully voided after a TWOC, with or without the administration of an α-blocker, needs to be defined. The effect of finasteride in the secondary prevention of AUR in men needs to be investigated. Other newer pharmaceutical agents, e.g. endothelin antagonists, require systematic studies into their mechanisms of action and potential usefulness in the prevention and treatment of AUR.
The second area is the association of AUR with general anaesthesia, a common precipitating event, which is preventable if the risk factors can be clearly shown and predicted. The third relates to the triggering events by prostatic infarction and α-adrenergic stimulation. Prostatic infarction and tissue swelling can be documented by TRUS soon after AUR. To study the role of α-adrenergic stimulation, patients admitted with AUR are catheterized suprapubically. The vesical, bladder neck, prostatic intraurethral and external sphincteric pressures are measured over 5 days. The optimal timing of a TWOC is determined when the detrusor pressure exceeds the intraurethral pressure. The effect of various pharmacological agents can also be assessed to determine their efficacy on several urodynamic variables.
This systematic review of the available literature should help to advance the understanding of the aetiology of AUR. Although prostatectomy remains the treatment of choice in most patients who present with AUR, new ways and alternative modalities are promising effective and less invasive means of treating the condition. Several exciting research avenues hold promise that prevention (primary or secondary) will play a greater role in the avoidance of AUR in susceptible patients, especially now that the aetiology of AUR is better understood.
S. Choong, Specialist Registrar in Urology.
M. Emberton, Senior Lecturer in Oncological Urology and Assistant Director, Clinical Effectiveness Unit, Royal Collegeof Surgeons of England.