Chapter 27. Neurophysiology & Pharmacology of the Lower Urinary Tract

The bladder, in concert with the urethra and the pelvic floor, is responsible for storage and periodic expulsion of urine. The integrated function of these components of the lower urinary tract (LUT) is dependent on a complex control system in the brain, spinal cord and peripheral ganglia, and on local regulatory factors (de Groat, 2006). Dysfunction of the central nervous control systems or of the components of the LUT can produce insufficient voiding and retention of urine, or different types of urinary incontinence (mainly urgency and stress incontinence), or the symptom complex of the “overactive bladder” (OAB), characterized by urgency, frequency with or without urgency incontinence, often with nocturia (Abrams et al, 2002).

Pharmacologic treatment of urinary incontinence and LUT symptoms (LUTSs) including OAB is the main option, and several drugs with different modes and sites of action have been tried (Andersson, 2007; Andersson and Wein, 2004; Andersson et al, 2009a, 2009b; Ouslander, 2004; Zinner et al, 2004). However, to be able to optimize treatment, knowledge about the mechanisms of micturition and of the targets for treatment is necessary. Theoretically, failure to store urine can be improved by agents that decrease detrusor activity and increase bladder capacity, and/or increase outlet resistance.

In this chapter, a brief review is given of the normal nervous control of the LUT and of some therapeutic principles used in the treatment of urinary incontinence.

Normal micturition occurs in response to afferent signals from the LUT (Birder et al, 2010; de Groat, 2006; Fowler et al, 2008). Both bladder filling and voiding are controlled by neural circuits in the brain, spinal cord, and peripheral ganglia. These circuits coordinate the activity of the smooth muscle in the detrusor and urethra with that of the striated muscles in the urethral sphincter and pelvic floor. Suprapontine influences are believed to act as on–off switches to shift the LUT between the two modes of operation: storage and elimination. In adults, urine storage and voiding are under voluntary control and depend upon learned behavior. In infants, however, these switching mechanisms function in a reflex manner to produce involuntary voiding. In adults, injuries or diseases of the central nervous system (CNS) can disrupt the voluntary control of micturition and cause the reemergence of reflex micturition, resulting in OAB and detrusor overactivity (DO). Because of the complexity of the CNS control of the LUT, OAB and DO can occur as a result of a variety of neurological disorders as well as changes in the peripheral innervation and smooth and skeletal muscle components (Andersson and Arner 2004; Andersson and Wein, 2004).

Filling of the bladder and voiding involve a complex pattern of afferent and efferent signaling in parasympathetic (pelvic nerves), sympathetic (hypogastric nerves), and somatic (pudendal nerves) pathways. These pathways constitute reflexes, which either keep the bladder in a relaxed state, enabling urine storage at low intravesical pressure, or which initiate bladder emptying by relaxing the outflow region and contracting detrusor. Integration of the autonomic and somatic efferents result in that contraction of the detrusor muscle is preceded by a relaxation of the outlet region, thereby facilitating bladder emptying. On the contrary, during the storage phase, the detrusor muscle is relaxed and the outlet region is contracted to maintain continence.

The sacral parasympathetic pathways mediate contraction of the detrusor smooth muscle and relaxation of the outflow region. The preganglionic parasympathetic neurons are located to the sacral parasympathetic nucleus (SPN) in the spinal cord at the level of S2–S4. The axons pass through the pelvic nerves and synapse with the postganglionic nerves either in the pelvic plexus, in ganglia on the surface of the bladder (vesical ganglia), or within the walls of the bladder and urethra (intramural ganglia). The ganglionic neurotransmission is predominantly mediated by acetylcholine acting on nicotinic receptors, although the transmission can be modulated by adrenergic, muscarinic, purinergic, and peptidergic presynaptic receptors. The postganglionic neurons in the pelvic nerve mediate the excitatory input to the normal human detrusor smooth muscle by releasing acetylcholine acting on muscarinic receptors (see later). However, an atropine-resistant (nonadrenergic, noncholinergic [NANC]) contractile component is regularly found in the bladders of most animal species. Such a component can also be demonstrated in functionally and morphologically altered human bladder tissue (O'Reilly et al, 2002), but contributes only to a few percent to normal detrusor contraction (Andersson and Wein, 2004). Adenosine triphosphate (ATP) is the most important mediator of the NANC contraction, although the involvement of other transmitters cannot be ruled out (Andersson and Wein, 2004). The pelvic nerve also conveys parasympathetic nerves to the outflow region and the urethra. These nerves exert an inhibitory effect on the smooth muscle, by releasing nitric oxide and other transmitters (Andersson and Wein, 2004).

The sympathetic innervation of the bladder and urethra originates from the intermediolateral nuclei in the thoracolumbar region (T10–L2) of the spinal cord. The axons leave the spinal cord via the splanchnic nerves and travel either through the inferior mesenteric ganglia (IMF) and the hypogastric nerve or pass through the paravertebral chain to the lumbosacral sympathetic chain ganglia and enter the pelvic nerve. Thus, sympathetic signals are conveyed in both the hypogastric nerve and the pelvic nerve. The ganglionic sympathetic transmission is, like the parasympathetic preganglionic transmission, predominantly mediated by acetylcholine acting on nicotinic receptors. Some preganglionic terminals synapse with the postganglionic cells in the paravertebral ganglia or in the IMF, while other synapse closer to the pelvic organs and short postganglionic neurons innervate the target organs. Thus, the hypogastric and pelvic nerves contain both pre- and postganglionic fiber. The predominant effect of the sympathetic innervation is to contract the bladder base and the urethra. In addition, the sympathetic innervation inhibits the parasympathetic pathways at spinal and ganglionic levels. In the human bladder, electrical field stimulation in vitro causes nerve release of noradrenaline, which in the normal detrusor causes relaxation. However, the importance of the sympathetic innervation for relaxation of the human detrusor has never been established. In contrast, in several animal species, the adrenergic innervation has been demonstrated to mediate relaxation of the detrusor during filling (Andersson and Arner, 2004).

The somatic innervation of the urethral rhabdosphincter and of some perineal muscles (eg, compressor urethrae and urethrovaginal sphincter) is provided by the pudendal nerve. These fibers originate from sphincter motor neurons located in the ventral horn of the sacral spinal cord (levels S2–S4) in a region called Onuf's (Onufrowicz's) nucleus (Thor and de Groat, 2010; Thor and Donatucci, 2004).

The afferent nerves to the bladder and urethra originate in the dorsal root ganglia at the lumbosacral level of the spinal cord and travel via the pelvic nerve to the periphery (de Groat and Yoshimura, 2010; Kanai and Andersson, 2010). Some afferents originate in dorsal root ganglia at the thoracolumbar level and travel peripherally in the hypogastric nerve. The afferent nerves to the striated muscle of the external urethral sphincter travel in the pudendal nerve to the sacral region of the spinal cord. The most important afferents for the micturition process are myelinated Aδ-fibers and unmyelinated C-fibers traveling in the pelvic nerve to the sacral spinal cord, conveying information from receptors in the bladder wall. The Aδ-fibers respond to passive distension and active contraction, thus conveying information about bladder filling. The activation threshold for Aδ-fibers is 5–15 mm H2O. This is the intravesical pressure at which humans report the first sensation of bladder filling. C-fibers have a high mechanical threshold and respond primarily to chemical irritation of the bladder urothelium/suburothelium or to cold. Following chemical irritation, the C-fiber afferents exhibit spontaneous firing when the bladder is empty and increased firing during bladder distension. These fibers are normally inactive and are therefore termed “silent fibers.” Afferent information about the amount of urine in the bladder is continuously conveyed to the mesencephalic periaqueductal gray (PAG), and from there to the pontine micturition center (PMC), also called Barrington's nucleus (Holstege, 2005; Kuipers et al, 2006).

Recent evidence suggests that the urothelium/suburothelium may serve not only as a passive barrier but also as a specialized sensory and signaling unit, which, by producing nitric oxide, ATP, and other mediators, can control the activity in afferent nerves, and thereby the initiation of the micturition reflex (Andersson, 2002; Birder, 2010; Birder and de Groat, 2007; de Groat, 2004). The urothelium has been shown to express, for example, nicotinic, muscarinic, tachykinin, adrenergic, bradykinin, and transient receptor potential (TRP) receptors (Birder, 2010; Birder and de Groat, 2007). Low pH, high K+, increased osmolality, and low temperatures can all influence afferent nerves, possibly via effects on the vanilloid receptor (capsaicin- [CAP] gated ion channel, TRPV1), which is expressed both in afferent nerve terminals and in the urothelial cells (Birder et al, 2001; Birder et al, 2002). A network of interstitial cells, extensively linked by Cx43-containing gap junctions, was found to be located beneath the urothelium in the human bladder (Brading and McCloskey, 2005; McCloskey, 2010; Sui et al, 2002; Sui et al, 2004). This interstitial cellular network was suggested to operate as a functional syncytium, integrating signals and responses in the bladder wall. The firing of suburothelial afferent nerves, conveying sensations and regulating the threshold for bladder activation, may be modified by both inhibitory (eg, nitric oxide) and stimulatory (eg, ATP, tachykinins, prostanoids) mediators. ATP, generated by the urothelium, has been suggested as an important mediator of urothelial signaling (Andersson, 2002). Supporting such a view, intravesical ATP induces DO in conscious rats (Pandita and Andersson, 2002). Furthermore, mice lacking the P2X3 receptor were shown to have hypoactive bladders (Cockayne et al, 2000; Vlaskovska et al, 2001). Interstitial cells can also be demonstrated within the detrusor muscle (Brading and McCloskey, 2005; McCloskey, 2010). They may be involved in impulse transmission, but their role has not been clarified.

There seem to be other, thus far unidentified, factors in the urothelium that could influence bladder function (Andersson and Wein, 2004). Even if these mechanisms can be involved in, for example, the pathophysiology of OAB, their functional importance remains to be established.

During the storage phase, the bladder has to relax in order to maintain a low intravesical pressure. Urine storage is regulated by two separate storage reflexes, of which one is sympathetic (autonomic) and the other is somatic (Thor and Donatucci, 2004). The sympathetic storage reflex (pelvic-to-hypogastric reflex) is initiated as the bladder distends (myelinated Aδ-fibers) and the generated afferent activity travels in the pelvic nerves to the spinal cord. Within the spinal cord, sympathetic firing from the lumbar region (L1–L3) is initiated, which, by effects at the ganglionic level, decreases excitatory parasympathetic inputs to the bladder. Postganglionic neurons release noradrenaline, which facilitates urine storage by stimulating β3-adrenoceptors (ARs) in the detrusor smooth muscle (see later). As mentioned previously, there is little evidence for a functionally important sympathetic innervation of the human detrusor, which is in contrast to what has been found in several animal species. The sympathetic innervation of the human bladder is found mainly in the outlet region, where it mediates contraction. During micturition, this sympathetic reflex pathway is markedly inhibited via supraspinal mechanisms to allow the bladder to contract and the urethra to relax. Thus, the Aδ afferents and the sympathetic efferent fibers constitute a vesico-spinal-vesical storage reflex, which maintains the bladder in a relaxed mode while the proximal urethra and bladder neck are contracted.

In response to a sudden increase in intra-abdominal pressure, such as during a cough, laugh, or sneeze, a more rapid somatic storage reflex (pelvic-to-pudendal reflex), also called the guarding or continence reflex, is initiated. The evoked afferent activity travels along myelinated Aδ afferent nerve fibers in the pelvic nerve to the sacral spinal cord, where efferent somatic urethral motor neurons, located in the nucleus of Onuf, are activated. Afferent information is also conveyed to the PAG and from there to the PMC (the L region). From this center, impulses are conveyed to the motor neurons in the nucleus of Onuf. Axons from these neurons travel in the pudendal nerve and release acetylcholine, which activates nicotinic cholinergic receptors on the rhabdosphincter, which contracts. This pathway is tonically active during urine storage. During sudden abdominal pressure increases, however, it becomes dynamically active to contract the rhabdosphincter. During micturition, this reflex is strongly inhibited via spinal and supraspinal mechanisms to allow the rhabdosphincter to relax and permit urine passage through the urethra. In addition to this spinal somatic storage reflex, there is also supraspinal input from the pons, which projects directly to the nucleus of Onuf and is of importance for voluntary control of the rhabdosphincter (Blok et al, 1997; Holstege, 2005; Sugaya et al, 2005).

Vesico-Bulbo-Vesical Micturition Reflex

Electrophysiological experiments in cats and rats provide evidence for a voiding reflex mediated through a vesico-bulbo-vesical pathway involving neural circuits in the pons, which constitute the PMC. Other regions in the brain, important for micturition, include the hypothalamus and cerebral cortex (Fowler et al, 2008; Griffiths, 2004; Griffiths et al, 2005; Holstege, 2005). Bladder filling leads to increased activation of tension receptors within the bladder wall and thus to increased afferent activity in Aδ-fibers. These fibers project on spinal tract neurons mediating increased sympathetic firing to maintain continence as discussed earlier (storage reflex). In addition, the spinal tract neurons convey the afferent activity to more rostral areas of the spinal cord and the brain. As mentioned previously, one important receiver of the afferent information from the bladder is the PAG in the rostral brain stem (Fowler et al, 2008; Holstege, 2005; Kuipers et al, 2006). The PAG receives information both from afferent neurons in the bladder and from more rostral areas in the brain, that is, cerebral cortex and hypothalamus. This information is integrated in the PAG and the medial part of the PMC (the M region), which also control the descending pathways in the micturition reflex. Thus, PMC can be seen as a switch in the micturition reflex, inhibiting parasympathetic activity in the descending pathways when there is low activity in the afferent fibers and activating the parasympathetic pathways when the afferent activity reaches a certain threshold. The threshold is believed to be set by the inputs from more rostral regions in the brain. In cats, lesioning of regions above the inferior colliculus usually facilitates micturition by elimination of inhibitory inputs from more rostral areas of the brain. On the other hand, transections at a lower level inhibit micturition. Thus, the PMC seems to be under a tonic inhibitory control. A variation of the inhibitory input to PMC results in a variation of bladder capacity. Experiments on rats have shown that the micturition threshold is regulated by, for example, gamma-aminobutyric acid (GABA)-ergic inhibitory mechanisms in the PMC neurons.

Vesico-Spinal-Vesical Micturition Reflex

Spinal lesions rostral to the lumbosacral level interrupt the vesico-bulbo-vesical pathway and abolish the supraspinal and voluntary control of micturition (Anderson and Wein, 2004). This results initially in an areflexic bladder accompanied by urinary retention. An automatic vesico-spinal-vesical micturition reflex develops slowly, although voiding is generally insufficient due to bladder-sphincter dyssynergia, that is, simultaneous contraction of bladder and urethra. It has been demonstrated in chronic spinal cats that the afferent limb of this reflex is conveyed through unmyelinated C-fibers, which usually do not respond to bladder distension, suggesting changed properties of the afferent receptors in the bladder. Accordingly, the micturition reflex in chronic spinal cats is blocked by CAP, which blocks C-fiber–mediated neurotransmission.

Central Nervous System Targets

Anatomically, several CNS regions may be involved in micturition control: supraspinal structures, such as the cortex and diencephalon, midbrain, and medulla, and also spinal structures (Fowler et al, 2008; Fowler and Griffiths, 2010; Griffiths, 2004; Griffiths et al, 2005; Holstege, 2005; Sugaya et al, 2005). Several transmitters are involved in the micturition reflex pathways described earlier and may be targets for drugs aimed for control of micturition (de Groat and Yoshimura, 2001). However, few drugs with a CNS site of action have been developed (Andersson and Pehrson, 2003).

Opioid Receptors

Endogenous opioid peptides and corresponding receptors are widely distributed in many regions in the CNS of importance for micturition control (de Groat and Yoshimura, 2001). It has been well established that morphine, given by various routes of administration to animals and humans, can increase bladder capacity or block bladder contractions. Furthermore, given intrathecally to anesthetized rats and intravenously to humans, the mu-opioid receptor antagonist, naloxone, has been shown to stimulate micturition, suggesting that a tonic activation of mu-opioid receptors has a depressant effect on the micturition reflex. However, intrathecal naloxone was not effective in stimulating micturition in conscious rats at doses blocking the effects of intrathecal morphine (Andersson and Wein, 2004).

Morphine given intrathecally was effective in patients with DO due to spinal cord lesions, but it was associated with side effects, such as nausea and pruritus. Further side effects of opioid receptor agonists comprise respiratory depression, constipation, and abuse (Andersson and Wein, 2004). Attempts have been made to reduce these side effects by increasing selectivity toward one of the different opioid receptor types. At least three different opioid receptors—μ, δ, and κ—bind stereospecifically with morphine and have been shown to interfere with voiding mechanisms. Theoretically, selective receptor actions, or modifications of effects mediated by specific opioid receptors, may have useful therapeutic effects for micturition control.

Tramadol is a well-known analgesic drug. By itself, it is a weak mu-receptor agonist, but it is metabolized to several different compounds, some of them almost as effective as morphine at the mu-receptor. However, the drug also inhibits serotonin (5-HT) and noradrenaline reuptake (Raffa and Friderichs, 1996). This profile is of particular interest, since both mu-receptor agonism and amine reuptake inhibition may be useful principles for treatment of DO/OAB.

When tramadol is given to a normal, awake rat, the most conspicuous changes in the cystometrogram are increases in threshold pressure and bladder capacity. Naloxone can more or less completely inhibit these effects (Pandita et al, 2003). However, there are differences between the effects of tramadol and morphine. Morphine has a very narrow range between the doses causing inhibition of micturition and those increasing bladder capacity and evoking urinary retention. Tramadol has effects over a much wider range of doses, which means that it could be therapeutically more useful for micturition control. It may be speculated that the difference is dependent on the simultaneous influence of the 5-HT and noradrenaline uptake inhibition (Pandita et al, 2003).

In rats, tramadol abolished experimentally induced DO caused by cerebral infarction (Pehrson et al, 2003). Tramadol also inhibited DO induced by apomorphine in rats (Pehrson and Andersson, 2003)—a model of bladder dysfunction in Parkinson's disease. Whether or not tramadol may have a clinically useful effect on DO/OAB remains to be studied in randomized controlled clinical trials (RCTs).

Safarinejad and Hosseini (2006) evaluated in a double-blind, placebo-controlled, randomized study the efficacy and safety of tramadol in patients with idiopathic DO. A total of 76 patients 18 years or older were given 100 mg tramadol sustained release every 12 hours for 12 weeks. Clinical evaluation was performed at baseline and every 2 weeks during treatment. Tramadol significantly reduced the number of incontinence periods and induced significant improvements in urodynamic parameters. The main adverse event was nausea. It was concluded that in patients with nonneurogenic DO, tramadol provided beneficial clinical and urodynamic effects. Even if tramadol may not be the best suitable drug for treatment of LUTS/OAB, the study proofs the principle of modulating micturition via the mu-receptor.

Serotonin (5-HT) Mechanisms

Lumbosacral autonomic, as well as somatic, motor nuclei (Onuf's nuclei) receive a dense serotonergic input from the raphe nuclei, and multiple 5-HT receptors have been found at sites where afferent and efferent impulses from and to the LUT are processed (Ramage, 2006). The main receptors shown to be implicated in the control of micturition are the 5-HT1A, 5-HT2, and 5-HT7 receptors (Ramage, 2006). There is some evidence in the rats for serotonergic facilitation of voiding; however, the descending pathway is essentially an inhibitory circuit, with 5-HT as a key neurotransmitter.

It has been speculated that selective serotonin reuptake inhibitors (SSRIs) may be useful for treatment of DO/OAB. On the other hand, there are reports suggesting that the SSRIs in patients without incontinence actually can cause incontinence, particularly in the elderly, and one of the drugs (sertraline) seemed to be more prone to produce urinary incontinence than the others (Movig et al, 2002). Patients exposed to serotonin uptake inhibitors had an increased risk (15 out of 1000 patients) for developing urinary incontinence. So far, there are no RCTs demonstrating the value of SSRIs in the treatment of DO/OAB.

Duloxetine is a combined noradrenaline and serotonin reuptake inhibitor, which has been shown to significantly increase sphincteric muscle activity during the filling/storage phase of micturition in the cat acetic acid model of irritated bladder function (Katofiasc et al, 2002; Thor et al, 1995). Bladder capacity was also increased in this model, both effects mediated centrally through both motor efferent and sensory afferent modulation. The effects of duloxetine was studied in a placebo-controlled study comprising women with OAB (Steers et al, 2007) and was, compared with placebo, shown to cause significant improvements or decreases in voiding and incontinence episodes, for increases in the daytime voiding intervals, and for improvements in quality-of-life (I-QoL) scores. Urodynamic studies showed no significant increases in maximum cystometric capacity or in the volume threshold for DO.

GABA Mechanisms

Both in the brain and the spinal cord, GABA has been identified as a main inhibitory transmitter (de Groat and Yoshimura, 2001). GABA functions appear to be triggered by binding of GABA to its inotropic receptors, GABAA and GABAC, which are ligand-gated chloride channels, and its metabotropic receptor, GABAB (Chebib and Johnston, 1999). Since blockade of GABAA and GABAB receptors in the spinal cord and brain (Pehrson and Andersson, 2002) stimulated rat micturition, an endogenous activation of GABAA+B receptors may be responsible for continuous inhibition of the micturition reflex within the CNS. In the spinal cord, GABAA receptors are more numerous than GABAB receptors, except for the dorsal horn where GABAB receptors predominate.

Experiments using conscious and anesthetized rats demonstrated that exogenous GABA, muscimol (GABAA receptor agonist), and baclofen (GABAB receptor agonist) given intravenously, intrathecally, or intracerebroventricularly inhibit micturition (Pehrson et al, 2002). Baclofen given intrathecally attenuated oxyhemoglobin-induced DO, suggesting that the inhibitory actions of GABAB receptor agonists in the spinal cord may be useful for controlling micturition disorders caused by C-fiber activation in the urothelium and/or suburothelium (Pehrson et al, 2002).

Stimulation of the PMC results in an immediate relaxation of the external striated sphincter and a contraction of the detrusor muscle of the bladder demonstrated in cats a direct pathway from the PMC to the dorsal gray commissure of the sacral cord (Blok et al, 1997). It was suggested that the pathway produced relaxation of the external striated sphincter during micturition via inhibitory modulation by GABA neurons of the motoneurons in the sphincter of Onuf (Blok et al, 1997). In rats, intrathecal baclofen and muscimol ultimately produced dribbling urinary incontinence (Pehrson et al, 2002).

Thus, normal relaxation of the striated urethral sphincter is probably mediated via GABAA receptors (Pehrson et al, 2002; Pehrson and Andersson, 2002), GABAB receptors having a minor influence on motoneuron excitability (Rekling et al, 2000).

Gabapentin was originally designed as an anticonvulsant GABA mimetic capable of crossing the blood–brain barrier (Maneuf et al, 2003). The effects of gabapentin, however, do not appear to be mediated through interaction with GABA receptors, and its mechanism of action remains controversial (Maneuf et al, 2003), even if it has been suggested that it acts by binding to a subunit of the α2δ unit of voltage-dependent calcium channels. Gabapentin is also widely used not only for seizures and neuropathic pain but also for many other indications, such as anxiety and sleep disorders, because of its apparent lack of toxicity.

In a pilot study, Carbone et al (2003) reported on the effect of gabapentin on neurogenic DO. These investigators found a positive effect on symptoms and significant improvement in urodynamic parameters after treatment with gabapentin, and suggested that the effects of the drug should be explored in further controlled studies in both neurogenic and nonneurogenic DO. Kim et al (2004) studied the effects of gabapentin in patients with OAB and nocturia not responding to antimuscarinics. They found that 14 out of 31 patients improved with oral gabapentin. The drug was generally well tolerated, and the authors suggested that it can be considered in selective patients when conventional modalities have failed. It is possible that gabapentin and other α2δ ligands (eg, pregabalin and analogs) will offer new therapeutic alternatives.

Noradrenaline Mechanisms

Noradrenergic neurons in the brainstem project to the sympathetic, parasympathetic, and somatic nuclei in the lumbosacral spinal. Bladder activation through these bulbospinal noradrenergic pathways may involve excitatory α1-ARs, which can be blocked by α1-AR antagonists (Yoshiyama et al, 2000). In rats undergoing continuous cystometry, doxazosin, given intrathecally, decreased micturition pressure, both in normal rats and in animals with postobstruction bladder hypertrophy. The effect was much more pronounced in the animals with hypertrophied OABs. Doxazosin given intrathecally, but not intra-arterially, to spontaneously hypertensive rats exhibiting bladder overactivity, normalized bladder activity (Persson et al, 1998). It was suggested that doxazosin has a site of action at the level of the spinal cord and ganglia.

A central site of action for α1-AR antagonists has been discussed as an explanation for the beneficial effects of these drugs in LUTS (especially storage symptoms) associated with benign prostatic hyperplasia (BPH) (Andersson and Gratzke, 2007; Andersson and Wein, 2004).

Dopamine Mechanisms

Patients with Parkinson's disease may have neurogenic DO, possibly as a consequence of nigrostriatal dopamine depletion and failure to activate inhibitory D1 receptors (Andersson, 2004). However, other dopaminergic systems may activate D2 receptors, facilitating the micturition reflex. Apomorphine, which activates both D1 and D2 receptors, induced bladder overactivity in anesthetized rats via stimulation of central dopaminergic receptors. The effects were abolished by infracollicular transection of the brain and by prior intraperitoneal administration of the centrally acting dopamine receptor blocker, spiroperidol. It has been shown that the DO induced by apomorphine in anesthetized rats resulted from synchronous stimulation of the micturition centers in the brainstem and spinal cord, and that the response was elicited by stimulation of both dopamine D1 and D2 receptors. Blockade of central dopamine receptors may be expected to influence voiding; however, the therapeutic potential of drugs having this action has not been established (Andersson and Wein, 2004).

NK-1 Receptor Mechanisms

The main endogenous tachykinins, substance P, neurokinin A (NKA), and neurokinin B (NKB), and their preferred receptors, NK1, NK2, and NK3, respectively, have been demonstrated in various CNS regions, including those involved in micturition control (Covenas et al, 2003; Lecci and Maggi, 2001; Saffroy et al, 2003).

Aprepitant, an NK-1 receptor antagonist used for treatment of chemotherapy-induced nausea and vomiting (Massaro and Lenz, 2005), significantly improved symptoms of OAB in postmenopausal women with a history of urgency incontinence or mixed incontinence, as shown in a well-designed pilot RCT (Green et al, 2006). Aprepitant was generally well tolerated and the incidence of side effects, including dry mouth, was low. Another NK-1 receptor antagonist, serlopitant, significantly decreased daily micturitions but did not offer advantages in efficacy compared with tolterodine (Frenkl et al, 2010). The results of these studies suggest that NK-1 receptor antagonism holds promise as a potential treatment approach for OAB, but so far, the drugs available have not been very effective.

Peripheral Targets

There are many possible peripheral targets for pharmacologic control of bladder function (Andersson and Arner, 2004). Although many effective drugs are available targeting these systems, most of them are less useful in the clinical situation due to the lack of selectivity for LUT, which may result in intolerable side effects.

Muscarinic Receptors

Muscarinic receptors comprise five subtypes, M1–M5, encoded by five distinct genes, and in both animal and human bladders, the mRNAs for all muscarinic receptor subtypes have been demonstrated, with a predominance of mRNAs encoding M2 and M3 receptors. These receptors are also functionally coupled to G proteins, but the signal transduction systems vary (Andersson and Arner, 2004; Giglio and Tobin, 2009).

Detrusor smooth muscle contains muscarinic receptors mainly of the M2 and M3 subtypes. The M3 receptors in the human bladder are the most important for detrusor contraction (Andersson and Wein, 2004). In the human detrusor, Schneider et al (2004) confirmed that the muscarinic receptor subtype mediating carbachol-induced contraction was the M3 receptor, and they also demonstrated that the L-type calcium channel blocker, nifedipine, almost completely inhibited carbachol-induced detrusor contraction, whereas an inhibitor of store-operated Ca2+ channels caused little inhibition. The Rho-kinase inhibitor, Y-27632, produced a concentration-dependent attenuation of the carbachol-induced contractile responses. Schneider et al (2004) concluded that carbachol-induced contraction of human detrusor is mediated via M3 receptors, and furthermore, largely depends on transmembrane Ca2+-flux through nifedipine-sensitive calcium channels as well as activation of the Rho-kinase pathway. These conclusions were supported by Takahashi et al (2004) who found that in human detrusor muscle, carbachol induces contraction, not only by increasing [Ca2+] but also by increasing the Ca2+ sensitivity of the contractile apparatus in a Rho-kinase and protein kinase C-dependent manner.

It has been suggested that M2 receptors may oppose sympathetically mediated smooth muscle relaxation, mediated by β-ARs (Hegde, 1997). M2 receptor stimulation may also activate nonspecific cation channels and inhibit KATP channels through activation of protein kinase C. However, the functional role for the M2 receptors in the normal bladder has not been clarified, but in certain disease states, M2 receptors may contribute to contraction of the bladder. Thus, in the denervated rat bladder, M2 receptors, or a combination of M2 and M3 receptors mediate contractile responses. Both types of receptor seemed to act in a facilitatory manner to mediate contraction (Braverman et al, 2002). In obstructed, hypertrophied rat bladders, there was an increase in total M2 receptor density but a reduction in M3 receptor density (Braverman and Ruggieri, 2003). The functional significance of this change for voiding function has not been established. Pontari et al (2004) analyzed bladder muscle specimens from patients with neurogenic bladder dysfunction to determine whether the muscarinic receptor subtype mediating contraction shifts from M3 to the M2 receptor subtype, as found in the denervated, hypertrophied rat bladder. They concluded that although normal detrusor contractions are mediated by the M3 receptor subtype, in patients with neurogenic bladder dysfunction, contractions can be mediated by the M2 receptors.

Muscarinic receptors may also be located on the presynaptic nerve terminals and participate in the regulation of transmitter release. The inhibitory prejunctional muscarinic receptors have been classified as M2 in the rabbit and rat, and M4 in the guinea pig, rat, and human bladder. Prejunctional facilitatory muscarinic receptors appear to be of the M1 subtype in the rat and rabbit urinary bladder (Andersson and Arner, 2004). Prejunctional muscarinic facilitation has also been detected in human bladders. The muscarinic facilitatory mechanism seems to be upregulated in OABs from chronic spinal cord–transected rats. The facilitation in these preparations is primarily mediated by M3 muscarinic receptors (Somogyi et al, 2003).

Muscarinic receptors have also been demonstrated in the urothelium and in the suburothelium (Bschleipfer et al, 2007; Chess-Williams, 2002; Mansfield et al, 2005), but their functional importance has not been clarified. It has been suggested that they may be involved in the release of an unknown inhibitory factor (Chess-Williams, 2002), or they may be directly involved in afferent signaling, and thus a target for antimuscarinic agents, explaining part of the efficacy of these drugs in DO/OAB (Andersson, 2004; Andersson and Yoshida, 2003; Kim et al, 2005; Yokoyama et al, 2005).

Antimuscarinics

In general, antimuscarinics can be divided into tertiary and quaternary amines (Abrams and Andersson, 2007; Guay, 2003). They differ with regard to lipophilicity, molecular charge, and even molecular size, tertiary compounds generally having higher lipophilicity and molecular charge than quaternary agents. Atropine, darifenacin, fesoterodine (and its active metabolite 5-hydroxymethyl-tolterodine), oxybutynin, propiverine, solifenacin, and tolterodine are tertiary amines. They are generally well absorbed from the gastrointestinal tract and should theoretically be able to pass into the CNS, dependent on their individual physicochemical properties. High lipophilicity, small molecular size, and less charge will increase the possibilities to pass the blood–brain barrier, but for some of the drugs, this counteracted by active transport out of the CNS. Quaternary ammonium compounds, like propantheline and trospium, are not well absorbed, pass into the CNS to a limited extent, and have a low incidence of CNS side effects (Guay, 2003). They still produce well-known peripheral antimuscarinic side effects, such as accommodation paralysis, constipation, tachycardia, and dryness of mouth.

Many antimuscarinics are metabolized by the P450 enzyme system to active and/or inactive metabolites (Guay, 2003). The most commonly involved P450 enzymes are CYP2D6 and CYP3A4. The metabolic conversion creates a risk for drug–drug interactions, resulting in either reduced (enzyme induction) or increased (enzyme inhibition, substrate competition) plasma concentration/effect of the antimuscarinic and/or interacting drug. Antimuscarinics secreted by the renal tubules (eg, trospium) may theoretically be able to interfere with the elimination of other drugs using this mechanism.

Antimuscarinics are still the most widely used treatment for urgency and urgency incontinence (Andersson et al, 2009a, 2009b). However, currently used drugs lack selectivity for the bladder, and effects on other organ systems may result in side effects, which limit their usefulness. For example, all antimuscarinic drugs are contraindicated in untreated narrow angle glaucoma.

Theoretically, drugs with selectivity for the bladder could be obtained, if the subtype(s) mediating bladder contraction, and those producing the main side effects of antimuscarinic drugs, were different. Unfortunately, this does not seem to be the case. One way of avoiding many of the antimuscarinic side effects is to administer the drugs intravesically. However, this is practical only in a limited number of patients.

Clinical efficacy. The clinical relevance of efficacy of antimuscarinic drugs relative to placebo has been questioned (Herbison et al, 2003). However, large meta-analyses of studies performed with the currently most widely used drugs (Chapple et al, 2005, 2008; Novara et al, 2008) clearly show that antimuscarinics are of significant clinical benefit.

None of the antimuscarinic drugs in common clinical use (darifenacin, fesoterodine, oxybutynin, propiverine, solifenacin, tolterodine, or trospium) is ideal as a first-line treatment for all OAB/DO patients. Optimal treatment should be individualized, implying that the patient's comorbidities and concomitant medications, and the pharmacological profiles of the different drugs, should be taken into consideration (Chapple et al, 2008).

Tolerability and safety. An extensive literature supports that antimuscarinics for the treatment of OAB symptoms are generally well tolerated. The adverse effect profiles of the different drugs are determined by their organ and muscarinic receptor subtype selectivities and pharmacokinetic parameters. The most commonly reported adverse effects are dry mouth, constipation, headache, and blurred vision.

Among the more serious concerns related to antimuscarinic use is the risk of cardiac adverse effects, particularly increases in heart rate and QT prolongation and induction of polymorphic ventricular tachycardia (torsade de pointes). It should be emphasized that QT prolongation and its consequences are not related to blockade of muscarinic receptors, but rather linked to inhibition of the hERG potassium channel in the heart (Roden, 2004). Thus, QT prolongation is not a class effect of antimuscarinics. In general, the cardiovascular safety for antimuscarinic drugs seems to be good. However, the potential of the different agents to increase heart rate or to prolong the QT time has not been extensively explored. Differences between the drugs cannot be excluded, but risk assessments based on available evidence are not possible.

Another concern is that antimuscarinic drugs commonly used to treat OAB can be associated with CNS side effects including cognitive dysfunction, memory impairment, dizziness, fatigue, and headache. With the exception for oxybutynin, CNS-related side effects are not commonly found when investigated. The potential to cause CNS-related adverse effects may differ between drugs, but in the absence of comparative trials, relative risk assessments are not possible.

For detailed discussion of the clinical efficacy, tolerability, and safety of the individual antimuscarinics, see Andersson et al (2009b).

Adrenergic Receptors

Alpha-ARs

Most investigators agree that there is a low expression of α-ARs in the human detrusor (Michel, 2006). Malloy et al (1998) found that two-thirds of the α-AR mRNA expressed was α1D, and one-third was α1A (there was no α1B). It has been suggested that a change of subtype distribution may be produced by outflow obstruction. Nomiya and Yamaguchi (2003) confirmed the low expression of α-AR mRNA in normal human detrusor, and further demonstrated, in contrast to data from animal experiments (Hampet et al, 2000), that there was no upregulation of any of the adrenergic receptors with obstruction. In addition, in functional experiments, they found a small response to phenylephrine at high drug concentrations with no difference between normal and obstructed bladders. Thus, in the obstructed human bladder, there seems to be no evidence for α-AR upregulation or change in subtype, although this finding was challenged by Bouchelouche et al (2005), who found an increased response to α1-AR stimulation in obstructed bladders. Whether or not this would mean that the α1D-ARs in the detrusor muscle are responsible for DO or OAB is unclear.

Sugaya et al (2002) investigated the effects of intrathecal tamsulosin (blocking α1A/D-ARs) and naftopidil (blocking preferably on α1D-ARs) on isovolumetric bladder contractions in rats. Intrathecal injection of tamsulosin or naftopidil transiently abolished these contractions. The amplitude of contraction was decreased by naftopidil but not by tamsulosin. It was speculated that in addition to the antagonistic action of these agents on the α1A-ARs of prostatic smooth muscle, both agents (especially naftopidil) may also act on the lumbosacral cord (α1D-ARs). This observation is of particular interest considering the findings that in the human spinal cord, α1D-AR mRNA predominated overall (Smith et al, 1999). Ikemoto et al (2003) gave tamsulosin and naftopidil to 96 patients with BPH for 8 weeks in a crossover study. Although naftopidil monotherapy decreased the I-PSS for storage symptoms, tamsulosin monotherapy decreased the I-PSS for voiding symptoms. However, this difference (which was suggested to depend on differences in affinity for α1-AR subtypes between the drugs) could not be reproduced in a randomized head-to-head comparison between the drugs (Gotoh et al, 2005).

Beta-ARs

It has been known for a long time that isoprenaline, a non–subtype selective β-AR agonist, can relax bladder smooth muscle (Andersson, 1993). Even if the importance of β-ARs for human bladder function still remains to be established (Andersson and Arner, 2004), this does not exclude that they can be useful therapeutic targets. All three subtypes of β-ARs (β1, β2, and β3) can be found in the detrusor muscle of most species, including humans (Michel and Vrydag, 2006), and also in the human urothelium (Otsuka et al, 2008). However, the expression of β3-AR mRNA (Nomiya and Yamaguchi, 2003; Michel and Vrydag, 2006) and functional evidence indicate a predominant role for this receptor in both normal and neurogenic bladders (Michel and Vrydag, 2006). The human detrusor also contains β2-ARs, and most probably both receptors are involved in the physiological effects (relaxation) of noradrenaline in the bladder (Andersson and Arner, 2004; Michel and Vrydag, 2006). β3-AR agonists have a pronounced effect on spontaneous contractions of isolated detrusor muscle (Biers et al, 2006), which may be the basis for their therapeutic effects in OAB/DO.

It is generally accepted that β-AR-induced detrusor relaxation is mediated by activation of adenylyl cyclase with the subsequent formation of cAMP (Andersson, 1999). However, there is evidence suggesting that in the bladder, β-AR agonists can mediate relaxation via K+ channels (particularly BKca channels), independent of cAMP (Frazier et al, 2008; Hristov et al, 2009; Takemoto et al, 2008; Uchida et al, 2005).

The in vivo effects of β3-AR agonists on bladder function have been studied in several animal models. It has been shown that β3-AR agonists increase bladder capacity with no change in micturition pressure and residual volume. For example, Hicks et al (2007) studied the effects of the selective β3-AR agonist, GW427353, in the anesthetized dog and found that the drug evoked an increase in bladder capacity under conditions of acid-evoked bladder hyperactivity, without affecting voiding.

β3-AR selective agonists are currently being evaluated as potential treatment for OAB/DO in humans (Colli et al, 2007). One of these, mirabegron (YM187), was given to patients with OAB in a controlled clinical trial (Chapple et al, 2008). The primary efficacy analysis showed a statistically significant reduction in mean micturition frequency, compared with placebo, and with respect to secondary variables, mirabegron was significantly superior to placebo concerning mean volume voided per micturition, mean number of incontinence episodes, nocturia episodes, urgency incontinence episodes, and urgency episodes per 24 hours. The drug was well tolerated, and the most commonly reported side effects were headache and gastrointestinal adverse effects. The results of this proof of concept study showed that the principle of β3-AR agonism may be useful for treatment of patients with OAB/DO.

Ion Channels

Calcium Channels

There is no doubt that an increase in [Ca2+]i is a key process required for the activation of contraction in the detrusor myocyte. However, it is still uncertain whether this increase is due to influx from the extracellular space and/or release from intracellular stores. Furthermore, the importance of each mechanism in different species, and also with respect to the particular transmitter studied, has not been firmly established (Kajioka et al, 2002).

Theoretically, inhibition of calcium influx by means of calcium antagonists would be an attractive way of inhibiting DO/OAB. However, there have been few clinical studies of the effects of calcium antagonists in patients with DO. Naglie et al (2002) evaluated the efficacy of nimodipine for geriatric urgency incontinence in a randomized, double-blind, placebo-controlled crossover trial, and concluded that this treatment was unsuccessful.

Thus, available information does not suggest that systemic therapy with calcium antagonists is an effective way to treat DO/OAB (Andersson et al, 2009a, 2009b; Andersson and Wein, 2004).

Potassium Channels

Potassium channels represent another mechanism to modulate the excitability of the smooth muscle cells. There are several different types of K+-channels and at least two subtypes have been found in the human detrusor: ATP-sensitive K+-channels (KATP) and large conductance calcium-activated K+-channels (BKCa). Studies on isolated human detrusor muscle and on bladder tissue from several animal species have demonstrated that K+-channel openers reduce spontaneous contractions as well as contractions induced by carbachol and electrical stimulate. However, the lack of selectivity of presently available K+-channel blockers for the bladder versus the vasculature has thus far limited the use of these drugs. The first generation of K-channel openers, such as cromakalim and pinacidil, were found to be more potent as inhibitors of vascular smooth muscle than of detrusor muscle (Andersson and Arner, 2004). No effects of cromakalim or pinacidil on the bladder were found in studies on patients with spinal cord lesions or detrusor instability secondary to outflow obstruction. Also with more recently developed KATP-channel openers, claimed to have selectivity toward the bladder, negative results have been obtained in an RTC on patients with idiopathic DO (Chapple et al, 2006).

Thus, at present there is no clinical evidence to suggest that K+-channel openers represent a treatment alternative for DO/OAB (Andersson et al, 2009a, 2009b; Andersson and Wein, 2004).

Vanilloid Receptors

The TRP channel superfamily has been demonstrated to be involved in nociception and mechanosensory transduction in various organ systems, and studies of the LUT have indicated that several TRP channels, including TRPV1, TRPV2, TRPV4, TRPM8, and TRPA1, are expressed in the bladder, and may act as sensors of stretch and/or chemical irritation. However, the roles of these individual receptors for normal LUT function and in LUTS/DO/OAB have not been established. TRPV1 is the channel best investigated. By means of CAP, a subpopulation of primary afferent neurons innervating the bladder and urethra, the “CAP-sensitive nerves,” has been identified. It is believed that CAP exerts its effects by acting on specific “vanilloid” receptors (TPVR1), on these nerves. CAP exerts a biphasic effect: initial excitation is followed by a long-lasting blockade, which renders sensitive primary afferents (C-fibers) resistant to activation by natural stimuli. In sufficiently high concentrations, CAP is believed to cause “desensitization” initially by releasing and emptying the stores of neuropeptides, and then by blocking further release. Resiniferatoxin (RTX) is an analogue of CAP, approximately 1000 times more potent for desensitization than CAP, but only a few hundred times more potent for excitation. Possibly, both CAP and RTX can have effects on Aδ-fibers. It is also possible that CAP at high concentrations (mM) has additional nonspecific effects.

The rationale for intravesical instillations of vanilloids is based on the involvement of C-fibers in the pathophysiology of conditions such as bladder hypersensitivity and neurogenic DO. In the healthy human bladder, C-fibers carry the response to noxious stimuli, but they are not implicated in the normal voiding reflex. After spinal cord injury, major neuroplasticity appears within bladder afferents in several mammalian species, including man (de Groat and Yoshimura, 2006). C-fiber bladder afferents proliferate within the suburothelium and become sensitive to bladder distention. Those changes lead to the emergence of a new C-fiber–mediated voiding reflex, which is strongly involved in spinal neurogenic DO. Improvement of this condition by defunctionalization of C-fiber bladder afferents with intravesical vanilloids has been widely demonstrated in humans and animals.

Despite available information (including data from randomized controlled trials) suggests that both capsaicin and RTX may have useful effects in the treatment of neurogenic DO, and that they may have beneficial effects also in nonneurogenic DO in selected cases refractory to antimuscarinic treatment (Andersson et al, 2009a, 2009b), they are no longer widely used.

Botulinum Toxin-Sensitive Mechanisms

Seven immunologically distinct antigenic subtypes of botulinum toxin (BTX) have been identified: A, B, C1, D, E, F, and G. Types A and B are in clinical use in urology, but most studies have been performed with BTX A type. BTX is believed to act mainly by inhibiting acetylcholine release from cholinergic nerve terminals interacting with the protein complex necessary for docking acetylcholine vesicles, but the mechanism of action may be more complex (Apostolidis et al, 2006; Simpson, 2004; Smith et al, 2003; Yokoyama et al, 2002). Apostolidis et al (2006) proposed that a primary peripheral effect of BTX is “the inhibition of release of acetylcholine, ATP, substance P, and reduction in the axonal expression of the CAP and purinergic receptors. This may be followed by central desensitization through a decrease in central uptake of substance P and neurotrophic factors.”

The BTX-produced chemical denervation is a reversible process and axons are regenerated in about 3–6 months. Given in adequate amounts BTX inhibits release not only of acetylcholine but also of several other transmitters. The BTX molecule cannot cross the blood–brain barrier and therefore has no CNS effects.

BTX injected into the external urethral sphincter was initially used to treat spinal cord injured patients with detrusor-external sphincter dyssynergia (Smith et al, 2002; Yokoyama et al, 2002). The use of BTX has increased rapidly, and successful treatment of neurogenic DO by intravesical BTX injections has now been reported by several groups (Cruz and Silva, 2004; Leippold et al, 2003; Sahai et al, 2005). BTX may also be an alternative to surgery in children with intractable OAB (Schurch and Corcos, 2005). However, toxin injections may also be effective in refractory idiopathic DO (Anger et al, 2010; Rapp et al, 2004). Intravesical injection of BTX resulted in improvement in medication refractory OAB symptoms. However, the risk of increased postvoid residual and symptomatic urinary retention was significant. Several questions remain concerning the optimal administration of BTX-A for the patient with OAB.

Adverse effects, for example, generalized muscle weakness, have been reported (De Laet and Wyndaele, 2005), but seem to be rare.

Summary and Future Aspects

To effectively control bladder activity, and to treat urinary incontinence, identification of suitable targets for pharmacological intervention is necessary. Such targets may be found within or outside the CNS. LUTSs, including OAB/DO, are all conditions that can have major effects on quality of life and social functioning. Antimuscarinic drugs are still first-line treatment—they have often good initial response rates, but adverse effects and decreasing efficacy cause long-term compliance problems. A new target is TRP channel superfamily, which has been demonstrated to be involved in nociception and mechanosensory transduction in various organ systems. Studies of the LUT have indicated that several TRP channels, including TRPV1, TRPV2, TRPV4, TRPM8, and TRPA1, are expressed in the bladder and may act as sensors of stretch and/or chemical irritation. However, the roles of these individual receptors for normal LUT function and in LUTS/DO/OAB have not been established. There may be several other new possibilities to treat LUTS/OAB/DO. For example, β3-AR agonists (eg, mirabegron) are in phase 3 trials after promising initial results, and the principle of β3-AR agonism seems clinically useful.

There is currently increasing interest in drugs modulating the micturition reflex by a central action. However, central nervous mechanisms have so far not been preferred targets for drugs aimed to treat OAB, since selective actions may be difficult to obtain. Drugs with a central mode of action such as NK-1 receptor antagonists, tramadol, and gabapentin have positive proof of concept documented in RCTs. Even if neither of these drugs can be recommended for general use in the treatment of LUTS/OAB/DO, they illustrate that agents with a target in the CNS can be therapeutically useful. Thus, even if antimuscarinic drugs remain the first-line treatment of the OAB, and their favorable efficacy/tolerability–safety ratio has been confirmed, new drugs are needed.

Bibliography