Objective To determine the effect of repeated and prolonged vaginal distension on the leak-point pressure (LPP) and urethral anatomy in the female rat, as prolonged vaginal distension has been clinically correlated with signs of stress urinary incontinence (SUI).
Materials and methods Sixty female rats were placed into one of five groups; four groups underwent one of four vaginal distension protocols using a modified 10 F Foley catheter, i.e. prolonged (1 h), brief (0.5 h), intermittent (cycling inflated/deflated for 0.5 h) or sham distension. All animals had a suprapubic bladder catheter implanted 2 days after and were assessed urodynamically 4 days after vaginal distension. The fifth group of rats acted as controls and did not undergo vaginal distension, but did have a suprapubic bladder catheter placed and urodynamics assessed. To measure LPP the rats were anaesthetized with urethane, placed supine and the bladder filled with saline (5 mL/h) while bladder pressure was measured via the bladder catheter. LPPs were measured three times in each animal by manually increasing the abdominal pressure until leakage at the urethral meatus, when the external abdominal pressure was rapidly released. Peak bladder pressure was taken as the LPP and a mean value calculated for each animal. Immediately after measuring LPP the urethra was removed and processed routinely for histology (5 µm sections, stained with haematoxylin/eosin and trichrome). The means ( sem ) were compared using a Kruskal–Wallis one-way anova on ranks, followed by a Dunn's test, with P < 0.05 indicating a significant difference.
Results Both LPP and the external increase in abdominal pressure were significantly lower after prolonged distension, at 31.4 (1.7) and 19.8 (1.2) cmH 2 O, than in the sham group, at 41.1 (3.2) and 32.0 (4.7) cmH 2 O, respectively. There were no significant differences in LPP or in the increase in abdominal pressure between the brief, intermittent and sham groups. Qualitative histology showed that prolonged distension resulted in extensive disruption and marked thinning of urethral skeletal muscle fibres. Brief and intermittent distension showed mild and focal disruptions, respectively.
Conclusions As observed clinically, prolonged vaginal distension results in a lower LPP, greater anatomical injury and increased severity of SUI. These results suggest that ischaemia is important in the development of SUI after prolonged vaginal distension.
Stress urinary incontinence (SUI) is a significant medical problem affecting ≈ 25 million American women ; 37% of non-institutionalized women aged > 60 years and almost 11% of women aged 30–59 years have urinary incontinence . Thus, incontinence is currently a major problem and is projected to increase with the ageing of the population in the USA. Understanding the causes of urinary incontinence is necessary to develop more effective treatments.
The primary causal factor associated with SUI in women is vaginal parity [3–7], presumably through combined neuromuscular and connective tissue injury [8–12]. The development of an animal model that reproduces the injuries associated with human vaginal parity is important for understanding the pathophysiology of SUI. This knowledge can aid in developing preventive strategies and treatment methods for SUI. The animal model of simulated birth trauma allows an assessment of the effects of prolonged dilation and compression of pelvic floor tissues on both the leak-point pressure (LPP) and the neuromuscular and connective tissue properties of the pelvic floor.
The purpose of the present study was to determine if prolonged vaginal distension (PVD) produced greater anatomical injury and lower LPPs than brief or intermittent distension. We tested various methods of vaginal distension in female rats and evaluated their effects via functional (LPP) and anatomical outcomes (urethral histology).
Materials and methods
Sixty virgin female Sprague-Dawley rats (175–200 g) were used and placed into either a control group or into one of four groups with differing protocols of vaginal distension. The rats undergoing distension were anaesthetized with a mixture of intraperitoneal ketamine (100 mg/kg) and xylazine (15 mg/kg) and placed supine. All of these rats had the vagina pre-dilated with incrementally increasing dilators. A modified 10 F Foley catheter was inserted into the vagina and sutured into place with one 3/0 silk suture. The sham group had a catheter inserted but not inflated (nine rats) and the brief group had a sustained 0.5 h inflation (10 rats). The prolonged group (PVD) had a sustained 1 h inflation (16 rats), and the intermittent group had five cycles of inflation for 5 min and deflation for 1 min (11 rats). The control group (14 rats) had no catheter inserted or vaginal pre-dilatation.
Two days after vaginal distension all five groups of rats had a transperitoneal suprapubic bladder catheter implanted, as described by Malmgren et al.. Briefly, the rats were anaesthetized as before and a catheter (PE-50 tubing with a flared tip) implanted in the bladder dome and tunnelled subcutaneously to the back of the neck, where it exited the skin. The catheter was capped and the incision closed in two layers.
Two days after implanting the bladder catheter (4 days after vaginal distension) all rats were anaesthetized with intraperitoneal urethane (1.2 g/kg) to assess the LPP, as described previously . The rats were placed supine at the level of zero pressure and the bladder emptied manually. The bladder was then filled with room-temperature saline at 5 mL/h through the bladder catheter, while the bladder pressure was recorded. The bladder catheter was connected to both a syringe pump and a pressure transducer. All bladder pressures were referenced to air pressure at the level of the bladder. Pressure and force transducer signals were amplified, recorded on a chart recorder and digitized for computer data collection at 10 samples/s. LPPs were measured at half bladder capacity (≈ 0.4 mL) by slowly and manually increasing the abdominal pressure until a leak occurred, when the external abdominal pressure was rapidly released (Fig. 1). The peak bladder pressure was taken as the LPP. The LPP at half-bladder capacity was tested 3–5 times on each rat. The bladders were manually emptied using Crede's manoeuvre and refilled between LPP measurements.
The animals were killed immediately after completing the measurements of LPP, and the vagina, bladder and urethra harvested and immersion-fixed in formalin. The tissues were embedded in paraffin, sectioned (5 µm) and stained with haematoxylin and eosin, and trichrome. Mid-urethral cross-sections including the external urethral sphincter of five rat urethras in each group were qualitatively analysed by two uropathologists.
Quantitative data are presented as the mean (sem); the mean value of all LPPs for each rat, and subsequently a mean for all rats within each group, was calculated. Student's t-test was used to compare the sham group with the undistended controls. A Kruskal–Wallis one-way anova on ranks, followed by Dunn's test, was used to compare the experimental groups, with P < 0.05 taken to indicate significant differences.
There were no significant differences in LPP between the sham and control groups, at 41.7 (5.7) and 40.7 (4.0) cmH2O, respectively. Nor was there a significant difference in the external increase in bladder pressure between them, at 30.0 (5.2) and 36.5 (3.0) cmH2O, respectively. Therefore, these two groups were combined into a single sham/control group for comparison with the experimental groups.
The mean LPP from the PVD group was significantly lower than in the sham/control group (Table 1). There were no significant differences in LPP between the brief or intermittent vaginal distension and the sham/control groups. The external increase in bladder pressure necessary to reach the LPP was also significantly lower after PVD than in the sham/control, brief or intermittent distension groups.
Table 1. The mean ( sem ) values of LPP and increase in abdominal pressure in the four groups
Urethral cross-sections from the sham group showed undisrupted skeletal and smooth muscle fibres circumferentially around the rat urethra (Fig. 2a). There were no signs of ischaemia, as indicated by wavy or thinning skeletal muscle fibres. After brief distension there was some focal disruption of the skeletal muscle layer with thinning of the skeletal muscle fibres, but the skeletal muscle layer was essentially complete (Fig. 2b). After intermittent distension the urethra showed focal disruption of the skeletal muscle layer and mild thinning of the muscle fibres (Fig. 2c). The most prominent damage was in the PVD group, with extensive disruption of the skeletal muscle layer and marked thinning of muscle fibres (Fig. 2d); there was also a marked paucity of muscle fibres around the urethra.
Clinically it has been suggested that a prolonged second-stage of labour in humans is associated with neuromuscular pelvic floor injury [9,11]. Studies show that some women incur a pudendal nerve denervation injury during vaginal delivery [9,11,15,16]. In some patients this may be severe enough to cause weakness of the pelvic floor muscles and result in poor transmission of abdominal pressure changes to the proximal urethra, resulting in SUI. Also, vaginal delivery alone may adversely affect urethral sphincter function, because of pressure-induced ischaemic injury . In addition, multiparity may cause repeated injury and greater urethral dysfunction with each vaginal birth .
An animal model that reproduces the pelvic floor, urethral sphincter and neurological damage sustained during vaginal delivery could be useful for studying the pathophysiology of SUI and for developing new treatments. Heidekamp et al. investigated a pudendal-nerve transected rat as a potential model of SUI. Alterations in the behavioural pattern of voiding after pudendal nerve transection showed the importance of pudendal innervation in the female rat continence mechanism. Although Heidekamp et al. did not present functional evidence of incontinence by measuring the LPP, they showed urethral striated muscle atrophy after pudendal nerve transection . Sakamoto et al. found similar alterations in rat voiding behaviour after a pudendal nerve crush injury, but did not undertake functional testing.
Lin et al. developed a vaginal-distension model in rats and tested the level of dysfunction produced by this injury. Their method of urethral functional testing involved either tickling the rat nostril with a whisker or using chilli powder to induce sneezing and urine leakage. With this method of assessing LPP, less than half of the rats showed stress incontinence 4 weeks after a 4 h vaginal distension . A new method of testing urethral function using a tilt table may be able to differentiate between the relative contributions of the external and internal sphincter to continence .
We developed a new method of measuring LPP in rats that allows for a graded measure of function and dysfunction . We showed that this method is reliable and reproducible within the same rat and between rats in the same group [14,23]; also, the method is not volume-dependent . Using this method of assessing LPP, we have shown that PVD results in a lower LPP than in the sham/control group. Brief and intermittent distension did not cause a significantly lower LPP than in the sham group. Also, the external increase in abdominal pressure was significantly lower only after PVD, further corroborating the differences in LPP and indicating that these differences are not a result of differences in baseline bladder pressure.
There are limitations to these models, as rats are quadrupeds and have a lax abdominal wall. Therefore, increased intra-abdominal pressure caused by a cough, sneeze or external pressure in a rat does not result in the same pressure forces seen in the human female pelvis [22,23]. Further developments are underway to generate a more even force transmission to the abdomen during LPP testing.
One of the more interesting results of the present study was the histological injury to the urethra as a result of vaginal distension. The PVD group had the greatest histological evidence of urethral damage, especially to the skeletal muscle layer. The evidence suggested that one mechanism of this injury may be ischaemia. Similarly, women with a prolonged second-stage of labour also have pressure-induced ischaemia to the urethral skeletal muscle . Pudendal nerve damage may be an additional mechanism underlying the association between prolonged second-stage labour and incontinence . During human vaginal delivery the pudendal nerves sustain direct injury in the pelvis and traction injury during elongation of the birth canal. Snooks et al. presented direct evidence for pudendal neuropathy after vaginal delivery in a 5-year follow-up study.
In summary, the mechanism of pelvic injury is probably multifactorial; as in humans, the present rat model shows evidence of pressure-induced ischaemia, pelvic floor injury and dysfunction of the urethral continence mechanism. With further understanding of the pathophysiology of the animal model, it may be possible to prevent and/or treat these injuries in the acute phase, to prevent resultant SUI in humans. Many new treatments for a damaged external urethral sphincter are currently being developed. In particular, recent studies show the ability to generate muscle stem cells that can form and organize into muscle structures which are genetically the same as the host. Preliminary studies indicate that muscle stem cells can be injected in the periurethral area and improve the urethral continence mechanism [25–28]. This model will certainly be useful for investigational studies.
This research was supported in part by Veterans Administration, Rehabilitation Research and Development Service, NIH R01HD38679, and a grant from the Society of Women in Urology.