Is the use of parasympathomimetics for treating an underactive urinary bladder evidence-based?
Martin C. Michel, Department Pharmacology & Pharmacotherapy, Academic Medical Center, Meibergdreef 15, 1105 AZ Amsterdam, the Netherlands. e-mail: firstname.lastname@example.org
An underactive urinary bladder (UUB), often occurring after surgery, can lead to urinary retention even in otherwise healthy people. We systematically reviewed published reports to determine whether the use of parasympathomimetic agents is warranted in patients with a UUB. Agents allegedly useful in treating UUB were identified from urology and pharmacology textbooks. A systematic search for randomized clinical trials in patients with UUB using these agents revealed 10 such studies. Controls typically received placebo or no treatment. While three studies reported statistically significant improvements relative to the control group, six did not and one even reported a significant worsening of symptoms. There was no evidence for differences between individual drugs, specific uses of such drugs, or in outcome measures. We conclude that the available studies do not support the use of parasympathomimetics for treating UUB, specifically when frequent and/or serious possible side-effects are taken into account.
underactive urinary bladder
randomized controlled trial
acute urinary retention
Underactivity of the urinary bladder (UUB), often leading to acute urinary retention (AUR), is a common complication after surgery, and can occur even after relatively small procedures in otherwise healthy people, and independent of surgery. While catheterization is the mainstay of treatment for AUR, medical treatment of the underlying detrusor underactivity is also possible and, in some cases, is even used as a preventive approach.
UUB could theoretically result from two causes, i.e. a lack of contractile stimulus and/or a lack of tissue responsiveness. The primary contractile stimulus for physiological voiding contractions of the detrusor is the parasympathetic transmitter acetylcholine, which acts on muscarinic receptors, mostly on those of the M3 subtype [1,2]. Thus, a lack of stimulus could be due to reduced transmitter release from parasympathetic nerve fibres innervating the detrusor, and/or from an increased degradation of such transmitter in the synaptic cleft, e.g. by the enzyme acetylcholinesterase. In turn, reduced transmitter release could result from a decreased input from the pontine micturition centre, as it can occur with neurogenic voiding dysfunction or as part of the relative detrusor denervation which can occur in the hypertrophied bladder , and/or from an impaired ability to release the transmitter, as it occurs after treatment with botulinum toxin . On the other hand, UUB could also result from impaired tissue responsiveness towards contractile stimuli. Such tissue ‘stunning’ has been reported, e.g. in the first day after pelvic surgery . Possible underlying mechanisms could include all steps involved in the signal transduction linking receptor activation to smooth muscle contraction 
Current approaches towards medical treatment of UUB aim at providing additional stimulation of muscarinic receptors on the detrusor cells. This can be done by a direct stimulation of muscarinic receptors by agonists such as bethanechol or carbachol. Alternatively, it can be done by inhibiting acetylcholinesterase, i.e. the enzyme inactivating the endogenous agonist acetylcholine, by agents such as distigmine, pyridostigmine or neostigmine. The latter approach depends on the presence of at least some endogenous acetylcholine, the effects of which can be amplified, and hence it is not expected to be effective if endogenous nervous input to the bladder is completely lacking, e.g. upon nerve injury. On the other hand, the use of direct muscarinic receptor agonists does not require the presence of endogenous acetylcholine to be effective. However, both approaches have little chance of being therapeutically effective if the detrusor underactivity results from impaired tissue responsiveness rather than from decreased availability of contractile stimulus.
Against this background recent editions of textbooks of both urology and pharmacology mention the use of parasympathomimetics in patients with UUB as a therapeutic option, but this does not correspond to our clinical experience of the poor efficacy of such medical treatment. As such drugs can have potentially harmful side-effects, we systematically sought and assessed reports to determine whether their use is evidence-based.
Agents allegedly useful in the treatment of UUB were identified from urology and pharmacology textbooks. We systematically searched Medline using the names of these compounds and the Medical Subject Heading terms ‘detrusor’, ‘bladder’, ‘underactivity’, ‘hypocontractility’, ‘areflexia’ and ‘atony’, and limited the search to randomized clinical trials (RCTs). The reference lists of identified articles were searched manually for additional studies, which revealed one study published before the existence of Medline.
In all we retrieved eight studies for the muscarinic receptor agonists bethanechol (seven) or carbachol (one) and three for the acetylcholinesterase inhibitor distigmine (Table 1; one study had mixed data with bethanechol and distigmine); no studies were found for pyridostigmine or neostigmine. Each study was independently evaluated by two investigators and scored for its technical quality and evidence level, according to the Oxford Centre for Evidence Based Medicine . As the judgement of the two investigators concurred in each case, no review by a third person was required.
Table 1. An overview of RCTs of parasympathomimetic agents in the treatment of UUB
|Bethanechol 50 mg × 3 oral from 3 days after surgery, vs no treatment|| I: prophylaxis of detrusor hypotonia after W-G op.|
E: hospital stay, catheter treatment, rate of cystitis and residual urine
| 40||Hospital stay 18.6 vs 15.5 days, catheter treatment 13.3 vs 9.6 days; rate of cystitis 25.0 vs 18.8%; residual urine <50 mL after 13 vs 8 days for no treatment vs bethanechol; all differences P < 0.01||2b|||
|Bethanechol 10 mg × 1 s.c. vs midazolam vs combination vs placebo|| I: treatment of AUR after anorectal surgery;|
E: incidence of catheterization
| 132||0 vs 69% responders for placebo and bethanechol (P = 0.05) irrespective of other treatment||2b|||
|Carbachol/diazepam 2 mg each vs alfuzosin 2.5 mg vs placebo, all ×1 oral|| I: treatment of AUR after general surgery|
E: voiding within 2 h after medication
| 249||No significant difference between groups||2b|||
|Distigmine 0.5 mg i.m. × 1 for 4 days vs placebo|| I: treatment of AUR after prostatectomy;|
E: flow rates and re-catheterization rate
| 93||No significant difference between groups||2b|||
|Bethanechol 25, 50 or 100 mg × 1 oral vs placebo 60 min before urodynamic investigation|| I: women with persistent high residual urine but no sign of neurological disease or BOO|
E: urodynamic changes
| 48||No significant difference between groups for voided volume, residual volume, % residual volume, mean flow rate and intravesical pressure||2b|||
|Bethanechol 4 × 50 mg daily oral + intravesical PGE2×1/week vs placebo for 6 weeks || I: Treatment of UUB;|
E: residual urine
| 19 ||Relative to baseline statistically significant reduction with active treatment but not with placebo, but effect size judged as ‘limited therapeutic effect’ by investigator||1b|||
|Distigmine 5 mg × 1 oral vs phenoxybenzamine 10 mg × 2 oral vs intravesical PGF2α 7.5 mg vs placebo from 1 day after surgery|| I: prevention of AUR after vaginal surgery for genital prolapse.|
E: residual volume after surgery
| 100||Statistically significant increase of residual urine for distigmine vs placebo||2b|||
|Bethanechol 15 mg every 4 h (6 doses) vs no treatment|| I: prevention of AUR postpartum;|
E: catheterization and residual volume
|1796||No significant difference between both groups||2b|||
|Bethanechol 20 mg × 3 or distigmine 5 mg × 3 oral vs urapidil 30 mg × 2 vs combined for 4 weeks|| I: treatment of UUB|
E: mean and max flow rate, postvoid residual volume IPSS
| 119||No significant effect of cholinergic agonists vs baseline||2b|||
|Bethanechol 25 mg × 1 oral vs placebo for 2 weeks in cross-over design|| I: treatment of UUB|
E: residual urine, max detrusor pressure and urinary flow
| 16||Significant reduction of residual urine and increase in max urinary flow vs placebo (P < 0.02 and <0.03), detrusor pressure tended to increase||1b|||
RCTs on the use of parasympathomimetic agents related to UUB have been reported for both prevention [8–10] and treatment settings [11–17], using a large variety of conditions. Thus, studies were reported for postpartum women , in patients after general surgery , prostatectomy , Wertheim–Meigs operation , vaginal prolapse surgery , or anorectal surgery , in patients with urodynamically confirmed detrusor underactivity [15–17] and in women with excessive residual urine but with no neurological disease . Some studies involved only men or only women, whereas others have investigated a mixed population. These studies have used placebo [9,11–15,17], no treatment [8,10] or other active treatments  as control, and involved 19–1796 patients. There was considerable heterogeneity in treatment methods (Table 1). Thus, in some cases the parasympathomimetic drug was administered as part of a combined treatment [12,15,16], whereas others have used monotherapy. Furthermore, the dose, route of administration, dosing interval, and duration of treatment also differed among studies. The efficacy of treatment was also assessed heterogeneously, using both clinical and urodynamic endpoints.
Based on this marked heterogeneity of underlying conditions, interventions and measured endpoints, it was not possible to identify a set of core elements common to all studies, i.e. to apply the PICO rules of meta-analysis, i.e. patient (or disease), intervention (such as a drug or test), comparison (another drug, placebo or test) and outcome. There were too few studies for each drug other than bethanechol to draw reliable drug-specific conclusions, but it is to be expected on mechanistic grounds that all direct and indirect parasympathomimetics should share the same effects on the bladder as long as they are adequately dosed. Specific endpoints and results of the various studies, with a grading of the corresponding evidence levels, are given in Table 1.
A statistically significant benefit relative to the control group was reported in three studies, of which one was in a preventive and two in a treatment setting, and all involved the use of bethanechol [8,11,17]. However, six studies did not detect a benefit [10,12–16] and one even found a significant worsening  relative to the control group. Thus, no consistent picture has emerged, although the combined studies involved 2612 patients, i.e. the equivalent of a full pharmaceutical development programme. While the heterogeneity of patient populations, treatments and assessments of treatment efficacy makes a formal meta-analysis difficult, the overall pattern of the data suggests that discrepancies among studies were unrelated to treatment vs preventive strategies, gender or study size, i.e. smaller studies failing to detect an effect due to lack of statistical power. While all three positive studies involved bethanechol [8,11,17], even for this agent no consistent picture emerges, as there are more negative studies [10,14–16].
Taken together the currently available data show little if any beneficial effect of parasympathomimetic agents in preventing or treating UUB. That some of these agents are officially registered in many countries for treating UUB does not necessarily contradict this conclusion, as such registration was typically obtained many years ago, i.e. when the drug approval process was not driven by adequately powered RCTs.
Given the key role of muscarinic receptors in the regulation of detrusor tone [1,2], we consider two explanations for these rather poor treatment responses. First, it is possible that UUB represents an unresponsiveness of the detrusor towards contractile stimuli, as has been noted within a day after pelvic surgery in experimental animals . This condition should be refractory to cholinergic treatment on mechanistic grounds. In such circumstances only agents directly acting on detrusor smooth muscle contractility have promise, but such drugs are not yet on the clinical horizon. Second, it is possible that the currently used parasympathomimetic agents are used in too low a dose to be effective on the detrusor. The reason for such underdosing could be a fear of side-effects, e.g. flushing, nausea, vomiting, diarrhoea, gastrointestinal cramps, bronchospasms, headache, salivation, sweating, and difficulty with visual accommodation; moreover, rare but important side-effects include acute and severe cardiovascular depression, which can result in an acute circulatory failure and cardiac arrest and hence be potentially lethal . Unless bladder-specific parasympathomimetic agents are identified, this appears to be an inherent limitation of their use.
Regardless of the above mechanistic considerations, our analysis clearly shows little evidence to support the use of muscarinic receptor agonists and/or acetylcholinesterase inhibitors in the treatment of UUB. Even if parasympathomimetic drugs have a minor beneficial effect, this would need to be weighed against the risk of frequent mild and rare serious side-effects. Therefore, the unfavourable balance between minor, if any, beneficial effects on the one hand and considerable side-effects on the other suggests that parasympathomimetics should not be used routinely in preventing or treating UUB.
CONFLICT OF INTEREST