Urinary Incontinence in the Bitch: An Update



Urinary incontinence (UI), defined as the involuntary loss of urine during the filling phase of the bladder (Abrams et al. 2002), is a commonly seen problem in veterinary practice. Urinary sphincter mechanism incompetence (USMI) after spaying is the most common micturition disorder, and its medical treatment is normally successful, even though the underlying pathophysiological mechanism is not fully understood. Hormonal changes inducing structural and functional alterations in the bladder, as well as in the urethra composition, are discussed. To manage incontinent patients successfully, possible underlying abnormalities besides USMI should be ruled out. In the majority of cases, history, physical examination and simple tests including urinalysis and urine bacterial culture lead to a presumed aetiology. If USMI is the most likely cause, then the advantage of further diagnostic tests should be discussed with the owner before starting a trial therapy with alpha-adrenergic drugs. Potential side effects of this therapy have to be mentioned even though they rarely occur. It is important to thoroughly evaluate the success of the initial treatment. Its failure should lead to further diagnostic testing. Specialized clinical assessments may provide an aetiological diagnosis, and this could serve as a basis for discussing further treatment options. Surgical procedures, which may in rare cases cause irreversible side effects, may be instituted. If incontinence reoccurs after initial treatment was successfully performed, the diagnostic work-up including urinalysis should always be repeated. As results of urinalysis did not correlate well with results of bacterial culture, a urine culture is recommended (Comer and Ling 1981). Cystocentesis is the preferred method of urine collection (Bartges 2004). Equivocal results of quantitative cultures of urine samples obtained during midstream voiding or by catheterization require repeat collection by cystocentesis (Comer and Ling 1981).


According to the International Continence Society (Anonymus, 1977), urinary incontinence is an objectively demonstrable involuntary loss of urine that is a social or hygienic problem. The symptom UI occurs when the processes of storing and passing urine and therefore the ability to retain urine voluntarily is disrupted. To be continent, the following requirements have to be met (Arnold et al. 2006):

  • 1.The ureters open solely into the bladder
  • 2.During the normal filling or the storage phase, the competence of the urethra and accommodation of the bladder is a prerequisite to store urine at a low and stable pressure. This requires that:
    1. a)The urinary bladder has to be extendible without an increase in the intravesical pressure. This relationship between bladder volume and bladder pressure, which is referred to as bladder compliance, is most probably dependent on various processes including a spinal parasympathetic reflex, intrinsic reflex activity in the vesical plexus, inhibition from the spinal cord and the intrinsic properties of the bladder muscle (Wyndaele et al. ).
    2. b)The intra-urethral pressure has to be higher than the intravesical pressure to prevent urine outflow. During the extension of the bladder, low-level vesical afferent firing occurs from tension receptors. This activates sympathetic reflexes, which mediate closure of the bladder neck and cranial urethra as well as inhibition of the detrusor muscle. This, in turn, contributes to sufficient urethral closure.
  • 3.The complex regulation of micturition has to be coordinated; the sensations of bladder filling, which are recorded by signals from stretch and volume receptors, have to be transmitted to centres of the spinal cord. At certain bladder fullness, afferent signals have to be conveyed to the cortical centres to consciously perceive the bladder filling stage. Increasing afferent neuronal activity of the bladder due to bladder fullness has to activate the micturition centre in the pons. Over a certain range of bladder filling, the micturition centre should be under voluntary control to prevent micturition at an inappropriate time. Activation of the micturition centre has to inhibit the spinal reflexes to allow relaxation of the smooth muscle sphincter that, in turn, opens the urinary outflow tract. Via parasympathetic innervation, the detrusor muscle has to be activated to support bladder emptying.

Work-up of UI

Looking at this complex regulatory mechanism, the multifarious causes of UI are not surprising. In dogs referred for the investigation of UI, USMI is the most common diagnosis made in adults followed by ureteral ectopia, detrusor instability, neurogenic incompetence, intersexuality, bladder neoplasia, cystitis, ureterovaginal fistula, vaginal tumours and vesiculo-vaginal fistula. However, in young animals, ectopic ureters and intersexuality are at the top of the list (Krawiec 1989; Holt 1999). The work-up of each individual case should intend to eliminate other possible diagnoses including pelvic, neurological and lower urinary tract abnormalities leading to UI.

To facilitate clinical work-up, it is useful to differentiate between UI associated with other signs of central nervous system dysfunction and non-neurogenic UI (Stone and Barsanti 1992a). Dogs with non-neurogenic UI can further be subdivided into dogs whose bladders remain distended, and dogs that can empty their bladders, although urine dribbles between micturitions. Most incontinent dogs belong to the last group and the physical conditions for their incontinence are:

  1. a) Decreased resting urethral pressure, which is exceeded by a normal intravesical pressure resulting in an insufficient urethral closure pressure and therefore resulting in urine dribbling. Examples are congenital USMI, urethral hypoplasia or acquired USMI after gonadectomy.
  2. b) Involuntary bladder contraction or bladder hypoplasia, where the excessive intravesical pressure exceeds the normal resting urethral pressure resulting in urine loss. Detrusor instability may occur with any inflammatory or neoplastic changes of the urogenital tract, but also in dogs after gonadectomy without any evidence of bladder wall irritations.
  3. c) Congenital or acquired malformations bypassing the urethral closure mechanism (i.e. ectopic ureters or ureterovaginal fistula).

If a patient is presented by the owner with presumed UI, first of all the ‘involuntary’ loss of urine must be distinguished from behavioural problems, dysuria or polyuria by asking the owner about the location and position of the dog during episodes of UI. Urgency or increased urine production as well as psychological problems may be responsible for urinating in the ‘wrong’ place. However, affected dogs still take a micturition position or try to reach the front door, which certainly is not the case when leakage occurs. Patient data such as gender, age, breed and reproductive status already reduce the number of possibilities. UI already occurring during puppyhood is indicative of a congenital malformation such as ectopic ureters, congenital USMI, persistent urachus, bladder diverticulum, bladder or urethral hypoplasia. If the first occurrence of UI is noted after neutering, the most likely diagnoses are USMI and/or detrusor instability, but ureterovaginal fistula and ectopic ureters should still be considered. Taking a detailed history as well as performing a thorough physical examination, observing micturition and the measurement of residual bladder volume may reduce the number of differentials and helps decisions on further laboratory and imaging techniques (Fig 1).

Figure 1.

Decision tree for the work-up of urinary incontinence (UI) in the female dog

UI after gonadectomy

In intact bitches and in male dogs, the risk for UI is low (0–1%), but in spayed bitches, the prevalence varies between 5% and 20%, and for certain breeds, it may be up to 60% (Arnold 1997; Thrusfield et al. 1998; Angioletti et al. 2004). The first episode of UI is usually observed 2–5 years after spaying, but may occur immediately or up to 10 years after surgery (Holt 1987; Arnold 1997; Thrusfield et al. 1998; de Bleser et al. 2011). Therefore, for a long time, the relationship between spaying and the development of UI was not recognized. The first time UI was described as a side effect of spaying was about 50 years ago (Joshua 1965), but it took another 20 years until the causal relationship between removal of the ovaries and UI was demonstrated (Thrusfield 1985). However, the influence of spaying on UI was again questioned last year: a systematic review of peer-reviewed original English analytic journal articles was performed to evaluate the strength of evidence for an association between neutering, age at neutering and UI in bitches. Of 1853 records screened, only three studies were identified that examined the effect of neutering or age at neutering on the risk of UI, and were judged to be at moderate risk of bias and included in the study. The conclusion was that overall, the evidence is neither consistent nor strong enough to make firm recommendations on the effect of neutering or age at neutering on the risk of UI (Beauvais et al. 2012). Based on their own results (Arnold 1997; Stocklin-Gautschi et al. 2001; Reichler et al. 2005), the authors of the present article are convinced that spaying is the most important cause for UI in the bitch. In one of their studies which focuses on the levels of gonadotropins in bitches with differences in risk for UI, of 195 intact and 310 spayed female dogs, 20% of the spayed dogs showed involuntary loss of urine occurring the first time after spaying, whereas only 2% of the intact dogs were incontinent (Reichler et al. 2005). In addition to spaying itself, the following risk factors for UI are described: breed, body weight, adiposity, timing of neutering relative to onset of puberty, the length of the urethra, the position of the bladder neck and tail docking (Thrusfield 1985; Holt and Thrusfield 1993; Blendinger et al. 1995b; Arnold 1997; Thrusfield et al. 1998; de Bleser et al. 2011). The general rule of increasing risk of UI with increasing body weight (Holt and Thrusfield 1993; Arnold 1997; de Bleser et al. 2011) seems to apply in particular for certain larger breeds with a breed disposition like the Boxer, Rottweiler, Doberman, Irish Setter, Weimaraner, Springer Spaniel, Bobtail and Giant Schnauzer (Holt and Thrusfield 1993; Blendinger et al. 1995b; Arnold 1997; Angioletti et al. 2004; de Bleser et al. 2011), while dogs of other breeds with a similar body weight such as the German Shepherd and the Labrador Retriever are rarely affected (Arnold 1997; Reichler et al. 2005; de Bleser et al. 2011). The surgical method of spaying, ovariectomy vs ovariohysterectomy, seems to have no effect on the risk for UI (van Goethem et al. 2006), whereas the risk factor ‘timing of neutering’ is still controversially discussed. According to a population study, the spaying of puppies at the age of <12 weeks should be avoided as the risk of UI seems to be increased (Spain et al. 2004). The comparison of two studies from Switzerland indicates that there is a lower risk of UI in bitches spayed shortly before than after puberty (Arnold 1997; Stocklin-Gautschi et al. 2001). The almost twice as high incidence of UI in bitches spayed after the first heat compared with bitches spayed immediately before the first heat was confirmed by another study of the same group (Reichler et al. 2005). In contrast to these results, data of a large case–control study in England (de Bleser et al. 2011) showed no difference in the risk of UI with regards to the timing of neutering relative to onset of puberty, or the age of the dog at the time of surgery. The latest retrospective study on the prevalence of UI in spayed bitches came to the conclusion that of all variables considered, such as age at time of surgery, body weight, number of litters prior to spaying and severity of incontinence, the only significant association was found between body weight and incontinence. Dogs weighing more than 15 kg were about seven times more likely to become incontinent, but the overall prevalence of UI in this study of 566 dogs was only 5.12% (Forsee et al. 2013). Unfortunately, timing of neutering relative to onset of puberty was not part of the statistical analysis.

The timing of gonadectomy relative to the onset of puberty may have an influence on the degree of UI. Sixty percentage of early spayed incontinent bitches show an uncontrolled loss of urine not only while sleeping but also when awake, during resting, and occasionally in a sitting position or on a walk (Arnold 1997). In contrast, bitches spayed after puberty are mainly incontinent during sleep, and even though more dogs spayed after the first heat become incontinent, the frequency of incontinent episodes is considerably lower (Stocklin-Gautschi et al. 2001).

Pathophysiology of UI after spaying

In spite of intensive research, the underlying pathophysiological mechanism for UI after spaying is still not fully elucidated. It is well known that within one year of spaying, the urethral closure pressure is significantly reduced. If the closure pressure drops below a critical level, the bitch becomes incontinent (Arnold 1997). Apart from this urethral sphincter incompetence (Rosin and Barsanti 1981; Richter and Ling 1985; Arnold 1997; Nickel 1998), a hyperactive bladder (Nickel 1998) may also contribute to UI after spaying. As a possible pathophysiological mechanism of these physical changes, besides the earlier postulated oestrogen deficiency, a change in the secretion of GnRH, FSH and LH (by the hypothalamic and pituitary gland) is discussed (Reichler et al. 2003, 2004, 2005, 2006; Burgherr et al. 2007; Ponglowhapan et al. 2007). Their receptors are expressed in the lower urinary tract, but to what extent the expression of these receptors changes after spaying is controversial. The reduced number of smooth muscle fibres and the higher collagen content in the lower urinary tract may also be responsible for the onset of UI after gonadectomy (Coit et al. 2008; Ponglowhapan et al. 2008). These changes in collagen could explain the reduced muscarinergic excitability and contractility of the smooth muscle fibres in spayed bitches (Augsburger and Cruz-Orive 1995; Coit et al. 2008; Byron et al. 2010). In contrast to the differences found between pre- and post-menopausal women, the ratio of collagen type II to collagen type I in periurethral tissues was similar in sexually intact and neutered female dogs (Byron et al. 2010). Recently, it was shown that there is a reduced amount of glycosaminoglycan in the tissues of the lower urinary tract which may favour a hyperactive bladder and this, in turn, leads to UI (Ponglowhapan et al. 2011). Prostaglandins may possibly play a role in modulating the reflex of micturition. Not only their receptors, but also cyclooxygenase, the enzyme responsible for their synthesis is considerably less expressed in the lower urinary tract of spayed bitches (Ponglowhapan et al. 2009, 2010).

Therapy of UI after spaying

Medical therapy is successful in most cases independent of the age at spaying and the degree of incontinence. It is however very important that any underlying disease causing polyuria and/or a concomitant cystitis are diagnosed and correctly treated. The first line of therapy is alpha-adrenergic agonists, which are commonly used to stimulate the alpha-adrenergic receptors expressed in the internal urethral sphincter, leading to an increase in the urethral closure pressure (Richter and Ling 1985). The success rate varies from 86% to 97% for phenylpropanolamine (1.5 mg/kg BW 1–3x/day) and 74–93% for ephedrine (1–2 mg/kg BW 2x/day) (Richter and Ling 1985; Blendinger et al. 1995a; Arnold 1997; Claeys et al. 2011). According to a study comparing the effects of 1.5 mg/kg tid phenylpropanolamine with the same dosage of pseudoephedrine, an alpha-agonist and stereoisomer of ephedrine, phenylpropanolamine seems to be superior (Byron et al. 2007). It produced a higher urethral closure pressure, and the side effects such as gastrointestinal irritations, anorexia, nervousness, aggressiveness or apathy were less often observed (Byron et al. 2007). Interestingly, in male dogs, where neutering, especially of large breed dogs, may also favour the acquired form of USMI, conservative management is frequently disappointing, as only 44% respond satisfactorily to treatment with phenylpropanolamine (Aaron et al. 1996). The use of alpha-adrenergic agonists is contraindicated in diseases where an increase in blood pressure should be avoided, as in most kidney diseases, heart problems or glaucoma (Burgherr et al. 2007).

Oestrogens sensitize the alpha-receptors and thus are suitable for a combined therapy with alpha-adrenergic agonists to potentiate their effects in females that are not fully responsive to phenylpropanolamine alone. Furthermore, studies in humans and rats indicate that oestrogens increase bladder capacity and stimulate the growth and proliferation of cells (Versi et al. 1988; Fleischmann et al. 2002; Teng et al. 2002; Yu et al. 2009). Due to the rare but severe complication of bone marrow suppression after the use of long-acting synthetic oestrogen preparations (De Schepper et al. 1977), only short-acting oestrogens like oestriol should be used. A study on the pharmacokinetics of oestriol in canine plasma after repeated oral administration of 2 mg oestriol once daily for seven consecutive days indicates that no accumulation occurs (Hoeijmakers et al. 2003). Treatment with oestriol (0.5–2 mg/dog/day PO for 5–7 days, then every other day and/or reduce dose every week to establish minimal effective dose for UI after spaying) improved continence in 65–83% of treated dogs (Arnold 1997; Mandigers and Nell 2001). Possible side effects are oestrous-like symptoms such as swelling of the vulva and attractiveness to male dogs (Mandigers and Nell 2001). In bitches that are not hysterectomized, the use of oestrogens should be carefully evaluated (Mandigers and Nell 2001), as they may cause uterine diapedesis leading to a bloody vaginal discharge and possibly uterine disease (Schotanus et al. 2008).

In some cases, substances used in humans that have an effect on the urinary bladder can also be applied for the dog. Possible medications are anticholinergic drugs such as duloxetine (first trials with 30- or 60-mg enteric-coated pellet in a capsule (Teng et al. 2002; Yu et al. 2009)), oxybutynin (0.2 mg/kg bid) (Noel et al. 2013) or the tricyclic antidepressant imipramine (5–20 mg per os bid) (Stone and Barsanti 1992b; Noël et al. 2010). Both imipramine and the anticholinergic spasmolytic flavoxate (10 mg/kg per os bid) stabilize the urinary bladder and also increase urethral closure pressure (Benson et al. 1977). Flavoxate can easily be combined with alpha-adrenergic agonists and/or oestrogens if, after ruling out cystitis, a detrusor instability is suspected to contribute to UI.

The subcutaneous application of GnRH depot analogues (Deslorelinacetate 4.7 mg/dog) increases bladder compliance and is successful as a single therapy in approximately 50% of bitches with UI (Reichler et al. 2003, 2006). GnRH analogues are especially suitable for patients showing serious side effects after therapy with alpha-adrenergic agonists, or if alpha-adrenergic agonists are contraindicated. So far, there have been no reports on side effects after the use of GnRH analogues in ovariectomized bitches. The use of GnRH depot analogues also restored continence in a neutered male dog (Greer 2012) and in a spayed cat (Pisu and Veronesi 2013). It was hypothesized that the observed effect of GnRH depot analogues is due to a direct effect of GnRH on the urinary bladder, as neither gonadotropin levels of responsive and unresponsive treated dogs nor gonadotropin receptor expression in the bladder differed between intact and spayed dogs (Reichler et al. 2006, 2007). Furthermore, urethral closure pressure did not reflect the treatment effect (Reichler et al. 2007). It is therefore very interesting that not only GnRH depot analogues, but also GnRH immunization restored continence in 4 of 9 female dogs (Donovan et al. 2013). In patients not regaining continence after the use of alpha-adrenergic agonists or GnRH analogues, the two medications can be combined as an additional treatment option.

Possible surgical treatments are urethropexy, colposuspension, different sling techniques, a hydraulic urethral sphincter and the submucosal endoscopic/laparoscopic/cystoscopic injection of urethral bovine cross-linked collagen, polyacrylamide hydrogel (BulkamidTM, ArthramidTM, both Contura A/S, Soborg, Denmark) or hyaluronic acid (DefluxTM; Qmed, Uppsala, Sweden) (Rawlings 2002; Barth et al. 2005; Claeys et al. 2010a,b; Byron et al. 2011; Delisser et al. 2012; Reeves et al. 2013). Compared with the other surgical techniques, the endoscopic treatment has the advantage of offering visual control of the bladder neck, urethra and the ureteral openings. The injection can also be repeated when needed (Barth et al. 2005). If urinary continence is not fully achieved, medical therapy can be given in addition. Thus, the overall success rate is 75% (Barth et al. 2005). Success rates of 80% are reported for the transobturator vaginal tape inside-out and the percutaneously controlled urethral hydraulic occluder (Claeys et al. 2010a; Currao et al. 2013; Reeves et al. 2013).

Due to the great impact of UI in humans, intensive research for other treatments is ongoing. Cell therapy including bone marrow mesenchymal cells (de la Garza-Rodea et al. 2011), adipose (Lin et al. 2010) and amniotic fluid-derived stem cells (Kim et al. 2012) as well as autologous skeletal muscle precursor cells (Eberli et al. 2012) is currently proposed to restore functional muscle cells and aid in sphincter closure in women with sphincter-associated incontinence. In a non-human primate model of stable urinary sphincter deficiency, intra-urinary sphincter injection of skeletal muscle precursor cells produced stable restoration of sphincter structure and function (Badra et al. 2013).

Conflicts of interest

The authors declare that they have no conflicts of interest.

Author contributions

IM Reichler wrote the manuscript, M Hubler revised it critically. The flow chart was discussed and designed by both authors together.