- Email: [email protected]
Pudendal Nerve Stretch Reduces External Urethral Sphincter Activity in Rats
Pudendal Nerve Stretch Reduces External Urethral Sphincter Activity in Rats Kamran P. Sajadi,* Dan L. Lin, James E. Steward, Brian Balog, Charuspong Dissaranan, Paul Zaszczurynski, Bradley C. Gill, Hai-Hong Jiang, James M. Kerns and Margot S. Damaser From the Division of Urology, Oregon Health and Science University (KPS), Portland, Oregon, Advanced Platform Technology Center, Louis Stokes Cleveland Veterans Affairs Medical Center (DLL, PZ), Department of Biomedical Engineering, Lerner Research Institute (JES, BB) and Glickman Urologic and Kidney Institute, Cleveland Clinic (CD, BCG, HHJ), Cleveland (MSD), Ohio, and Department of Anatomy and Cell Biology, Rush University Medical Center (JMK), Chicago, Illinois
Purpose: Most animal models of stress urinary incontinence simulate maternal injuries of childbirth since delivery is a major risk factor but they do not reproduce the nerve stretch known to occur during human childbirth. We hypothesized that pudendal nerve stretch produces reversible dysfunction of the external urethral sphincter. Materials and Methods: Female virgin Sprague-Dawley® rats were anesthetized with urethane. Bilateral pudendal nerve stretch or sham injury was performed for 5 minutes. External urethral sphincter electromyography and leak point pressure were recorded immediately before and after, and 10, 30, 60 and 120 minutes after pudendal nerve stretch. Post-pudendal nerve stretch results were compared to prestretch values and to values in sham injured animals. The pudendal nerves underwent qualitative histological assessment. The nucleus of Onuf was evaluated by immunohistochemistry and polymerase chain reaction for -APP and c-Fos expression as markers of neuronal activity and injury. Results: A total of 14 rats underwent bilateral pudendal nerve stretch (9) or sham injury (5). Each nerve was stretched a mean ⫾ SEM of 74% ⫾ 18% on the left side and 63% ⫾ 13% on the right side. Electromyography amplitude decreased significantly immediately after stretch compared to before stretch and after sham injury (p ⫽ 0.003) but it recovered by 30 minutes after stretch. There was no significant change in leak point pressure at any time. Two hours after injury histology showed occasional neuronal degeneration. -APP and c-Fos expression was similar in the 2 groups. Conclusions: Acute pudendal nerve stretch produces reversible electrophysiological dysfunction but without leak point pressure impairment. Pudendal nerve stretch shows promise in modeling injury. It should be tested as part of a multi-injury, chronic, physiological model of human childbirth injury.
Abbreviations and Acronyms
-APP ⫽ -amyloid precursor protein EMG ⫽ electromyography EUS ⫽ external urethral sphincter LPP ⫽ leak point pressure PBS ⫽ phosphate buffered saline PCR ⫽ polymerase chain reaction PNC ⫽ pudendal nerve crush PNS ⫽ pudendal nerve stretch SUI ⫽ stress urinary incontinence VD ⫽ vaginal distension Submitted for publication January 13, 2012. Supported by the Research Project Committee, and Glickman Urologic and Kidney Institute, Cleveland Clinic, Cleveland, Ohio. Study received institutional animal care and use committee approval. * Correspondence: Division of Urology, Oregon Health and Science University, CH10U, 3303 Southwest Bond Ave., Portland, Oregon 97232 (telephone: 503-346-1500; e-mail: kpsajadi@ gmail.com).
Key Words: urethra; urinary incontinence, stress; pudendal nerve; electromyography; disease models, animal STRESS urinary incontinence, the involuntary leakage of urine associated with increased abdominal pressure such as exercise, laughing or coughing, is a prevalent, bothersome, costly condition
affecting millions of women.1–3 SUI is strongly associated with age and vaginal childbirth, the latter due to ischemic and direct damage of the pelvic floor musculature and nerves.4,5 In
0022-5347/12/1884-1389/0 THE JOURNAL OF UROLOGY® © 2012 by AMERICAN UROLOGICAL ASSOCIATION EDUCATION
http://dx.doi.org/10.1016/j.juro.2012.06.006 Vol. 188, 1389-1395, October 2012 RESEARCH, INC. Printed in U.S.A.
AND
www.jurology.com
1389
1390
PUDENDAL NERVE STRETCH REDUCES EXTERNAL URETHRAL SPHINCTER ACTIVITY IN RATS
addition, EUS innervation via the pudendal nerve is prone to injury during vaginal childbirth by stretch and crush.6,7 Animal models of childbirth injury simulate direct pelvic floor damage via VD and pudendal nerve injury via nerve crush or avulsion.8 Three-dimensional models of human childbirth suggest that significant stretch of the pudendal nerve occurs during vaginal childbirth with 13% strain to the branch that innervates the EUS.7 Animal models of other peripheral nerve injuries indicate that this degree of stretch results in significant nerve impairment.9,10 However, to our knowledge no animal model of vaginal childbirth injury to date has specifically incorporated PNS injury. VD and/or PNC create models of SUI that are recoverable but up to 30% of women with postpartum SUI do not recover.11 Durable animal models of SUI have been described, such as pudendal nerve transection and urethrolysis, but they involve drastic injuries that do not have physiological correlates in human childbirth.8 An integrated model involving physiological mechanisms of injury, such as VD with PNC and PNS, if nonrecoverable, would more accurately replicate human vaginal childbirth injury. With the aim of designing a multiple injury model we hypothesized that PNS produces significant, reversible EUS denervation and dysfunction, as measured by anatomical and functional outcomes.12,13
MATERIALS AND METHODS Dissection and Measurement Techniques After approval from the institutional animal care and use committee age matched, virgin female Sprague-Dawley rats were anesthetized with intraperitoneal urethane (1.2 mg/kg). An anterior transpubic dissection was made to expose the bladder, urethra, pelvic floor and pudendal nerves bilaterally, as previously described.12 An inverted Y-shaped incision was made over the pelvis, the inferior epigastric vessels were ligated and divided, the pubic symphysis was divided sharply and the pubic bones were separated with blunt instruments. Bipolar parallel platinum electrodes (30 gauge needles 2 mm apart) were placed on the anterior surface of the mid urethra over the EUS to record EMG. The electrodes were connected to a Model ISO-80 amplifier (World Precision Instruments, Sarasota, Florida) with 3 Hz to 3 kHz bandpass frequencies and a Chart™ 5 electrophysiological recording system with a 10 kHz sampling rate. A pursestring 6-zero silk suture was placed in the bladder dome. Through a small cystotomy a polyethylene-50 catheter with a flared tip was introduced and secured in place as a suprapubic tube. The catheter was connected to a Model BLPR2 pressure transducer and a SYS-TBM4M Transbridge transducer amplifier (World Precision Instruments) and a Model 200 syringe pump (KD Scientific, New Hope, Pennsylvania). Vesical pressure was zeroed at the bladder level. Vesical pressure and EUS EMG were recorded simultane-
ously as the bladder was filled with saline at 5 ml per hour with rats free to void via the urethra. Baseline LPP was assessed before injury in all rats, as previously described.13 When half bladder capacity was attained, gentle pressure was applied to the rat bladder and slowly increased until leakage was seen via the urethra, when applied abdominal pressure was discontinued abruptly. The process was repeated 3 times per rat and the 3 values were used to calculate the mean. Peak pressure minus baseline pressure was recorded as LPP.
Pudendal Nerve Stretch To grossly quantify stretch 2 marks were made 5 mm apart on each nerve in the middle of the exposed segment using a sterile Kendall™ Devon surgical marker. DeBakey forceps or Castroviejo surgical calipers with the tips covered by polyethylene tubing were placed in the ischiorectal fossa beneath the nerve on each side. For rats undergoing PNS the instrument was opened to push apart the vagina and pelvic side wall, allowing the pudendal nerve to passively stretch between them for 5 minutes. For sham injured rats the instruments were left in situ for 5 minutes without opening them. Before and during stretch calibrated digital photographs were taken of the marked nerves with an ophthalmological surgical microscope at 5⫻ magnification. The distance between the marks, measured from center to center, was compared before and during stretch to quantify the degree of stretch.
Post-Stretch Assessment EUS EMG and LPP were repeated immediately after, and 10, 30, 60 and 120 minutes after bilateral stretch or sham injury. Recording began immediately after intervention since prior studies demonstrated an immediate effect with bilateral injury.14
Pudendal Neuroanatomy Two hours after injury or sham injury the rats were sacrificed via intracardiac perfusion with cold saline. The pelvic floor and pudendal nerves were excised and fixed in glutaraldehyde (2.5%) and paraformaldehyde (0.5%) fixative in cacodylate buffer, refrigerated overnight and changed to buffer after 24 hours. The pudendal nerves were microdissected from the pelvis, as previously described.15 Transverse semithin nerve sections were embedded with Durcupan™ and stained with toluidine blue. Nerve sections were evaluated qualitatively by a blinded reviewer (JMK).
Spinal Cord Evaluation A dorsal midline incision was made with the rat prone to expose the spine. Thoracolumbar laminectomy was performed to expose the lumbosacral spinal cord, which was covered with liquid nitrogen and frozen in situ. The lumbosacral cord was excised and stored in liquid nitrogen until sectioning. The area containing the nucleus of Onuf (L6-S1) was sectioned on a cryostat into 14 m sections. Alternating sections were placed on glass slides for immunohistochemistry and on membrane slides for laser microdissection and PCR.
Immunohistochemistry Alternating immunohistochemistry slides were stained for -APP, a marker of peripheral nerve traumatic stretch
PUDENDAL NERVE STRETCH REDUCES EXTERNAL URETHRAL SPHINCTER ACTIVITY IN RATS
injury, or for c-Fos, a proto-oncogene that is an indirect marker of neuronal activity.10,16 After storage in a freezer at ⫺20C slides were fixed using 10% neutral buffered formalin for 10 minutes at room temperature, rinsed thoroughly with PBS and incubated in 0.6% hydrogen peroxide in PBS for 30 minutes. Application of 3% normal goat serum (S-1000) in PBS was followed by application of Avidin D blocking solution (SP-2001) and biotin blocking solution (SP-2001, Vector Laboratories, Burlingame, California). Slides were incubated overnight at 4C with 1:1,000 rabbit monoclonal anti--APP or 1:1,000 rabbit polyclonal anti-c-Fos antibody (ab32136 and ab7963, respectively, Abcam®). Sections were exposed to secondary 1:200 biotinylated goat anti-rabbit IgG (BA-1000) and incubated with ImmPACT™ DAB Substrate (SK-4105). Slides were counterstained with hematoxylin, rinsed, dipped in bluing solution (7301, Richard Allan Scientific, Kalamazoo, Michigan), dehydrated with ethanol and xylene, and fixed with Permount™. Omission of primary antibodies served as a negative control. Positive controls consisted of postnatal day 3 rat brain for -APP17 and rat spinal cord for c-Fos. All slides were evaluated with the reviewer (KPS) blinded to the experimental group using a semiquantitative grading scale of 0 to 3 according to the intensity of staining for -APP and c-Fos.
Laser Microscope Dissection and PCR Membrane slides were removed from ⫺80C storage, thawed for 15 seconds, rehydrated with ultrapure nuclease-free water, stained with thionin, dehydrated with ethanol and fixed with xylene. Using a laser dissection microscope pudendal motoneurons in the nucleus of Onuf were collected from each specimen. Dissected cells were captured in a microcentrifuge tube containing lysis solution. RNA isolation was performed using an RNeasy® Mini Kit for RNA isolation, followed by reverse transcription with a high capacity cDNA reverse transcription kit (Applied Biosystems®). Due to the small sample size cDNA was preamplified using TaqMan® PreAmp Master Mix with amplification targeted to c-Fos (Rn02396759) or -APP (Rn00570673, Applied Biosystems). A 40 cycle PCR reaction was run for each primer and c-Fos or -APP expression was compared between the 2 groups. Expression was quantified relative to a single sham injured control using the ⌬⌬cycle threshold method with 18S rRNA as the endogenous control.
Outcomes and Statistical Analysis EUS EMG was quantitatively assessed by determining the mean rectified amplitude of the potential, as we have done previously.18 Power supply interference (60 and 120 Hz) was filtered with a digital band pass filter (59 to 61 and 119 to 121 Hz using Chart 5). LPP and EMG mean amplitude at the baseline and peak of LPP testing were calculated at each time point. A guarding reflex was defined as an increase in mean EMG amplitude during LPP testing compared to baseline amplitude. Data were analyzed with 2-way repeated measures ANOVA. Post hoc pairwise comparisons to pre-injury measurements were performed using the Dunnett test. Results were tested for correlation with the percent of stretch and averaged between the left and right sides using the Spearman corre-
1391
lation. Semiquantitative immunohistochemistry and realtime PCR results were compared between the sham injured and PNS groups using the Student t test. For all statistical tests p ⬍0.05 was considered statistically significant. SigmaPlot™ v11 was used. Quantitative data are shown as the mean ⫾ SEM.
RESULTS EMG and LPP An initial series of PNS procedures performed to assess technique feasibility showed a dramatic change in EUS EMG after 5 minutes of stretch. Thus, we stretched the nerve for 5 minutes in subsequent rats. A total of 24 rats underwent PNS or sham injury. Rats were excluded from study due to excessive noise during EMG that prohibited adequate analysis. Rats were also excluded when pre-injury mean EMG amplitude was less than 5 V (less than the 5th percentile) since they likely experienced inadvertent nerve injury during dissection before injury was applied. Nine PNS and 5 sham injured rats weighing a mean of 266 ⫾ 5 gm were included in analysis. In rats that underwent PNS the pudendal nerve was stretched a mean of 74% ⫾ 18% on the left side and 63% ⫾ 13% on the right side, which was not significantly different. EUS EMG amplitude was significantly decreased immediately after PNS but recovered quickly thereafter (p ⬍0.001, figs. 1, A and 2). Sham nerve stretch did not affect EUS EMG. A guarding reflex was present before PNS and in sham injured animals at all time points. In contrast, it was significantly decreased immediately after PNS (p ⫽ 0.003, fig. 3). It began to return 30 minutes later (fig. 1, B). EMG amplitude did not vary significantly with time in sham injured animals (fig. 2). This suggests that extended anesthesia as well as dissection and exposure of the nerve did not detrimentally affect function. As a fraction of the pre-injury value, average EMG amplitude immediately after injury was 48% in the PNS group and 102% in the sham injured group. EMG amplitude during LPP immediately after PNS was lower than after sham injury but not significantly so (mean 26 ⫾ 5 vs 38 ⫾ 3 V, p ⫽ 0.09). EMG amplitude remained significantly decreased 10 minutes after PNS but on post hoc analysis this was not significantly different from sham injury or prestretch values. The decrease in EMG amplitude did not significantly correlate with the percent of nerve stretch. There was minimal change in LPP from before to immediately after stretch (mean 43 ⫾ 6 to 40 ⫾ 15 cm H2O) (fig. 3, A). Changes in LPP with time were not significant (p ⫽ 0.421). There was a slight decrease in LPP 120 minutes after sham stretch, similar to that after PNS, but this was not significantly
1392
PUDENDAL NERVE STRETCH REDUCES EXTERNAL URETHRAL SPHINCTER ACTIVITY IN RATS
Figure 1. A, characteristic baseline EMG during filling before and after PNS and sham injury. EMG immediately after injury in PNS group shows mostly background noise (sine wave) but motor units began returning 10 minutes after injury. Sham injured EMG is largely unchanged. B, combined EMG and vesical pressure show guarding reflex, defined as increased EMG activity during LPP testing, before PNS. Response was blunted after PNS but recovered 30 minutes after injury.
different from baseline (mean 21 ⫾ 3.4 vs 33 ⫾ 6.9 cm H2O, p ⫽ 0.077). Histology Small motor fascicles of the nerve were normal in appearance in PNS and sham injured specimens. Isolated axons in the larger fascicles from the PNS group rarely showed degeneration (fig. 4). Nonetheless, there was no apparent structural difference between pudendal nerves subjected to PNS or sham injury. Cytoplasmic staining for -APP and nuclear staining for c-Fos were observed in the nucleus of Onuf in all PNS and sham injured animals (fig. 5). On a scale of 0 to 3 there was no significant difference in the semiquantitative score between the PNS and sham injured groups for -APP (mean 1.4 ⫾ 0.7, range 0 to 3 vs 1.6 ⫾ 0.2, range 1 to 2) or c-Fos (mean 2.2 ⫾ 0.4, range 1 to 3 vs 2 ⫾ 0, range 2 to 2). Quantitative Real-time PCR Expression of c-Fos mRNA was not significantly different after PNS than after sham injury (1.13 ⫾ 0.11 vs 0.93 ⫾ 0.05). Similarly -APP mRNA expression
was not significantly different after PNS compared to that in sham injured rats (mean 1.24 ⫾ 0.36 vs 0.84 ⫾ 0.09).
DISCUSSION We developed a method of PNS that produces a rapid, demonstrable and reversible decrease in EUS EMG activity. EUS EMG is dysfunctional in women with urodynamic stress incontinence19 and it has been used to measure urethral function in animal models.12,20,21 The primary outcome of this study was EUS EMG since it reflects changes immediately after injury.14,15 PNS rats demonstrated an immediate decrease in EUS EMG activity but this recovered relatively quickly (by 10 minutes) with subsequent return of the guarding reflex. In contrast, EUS EMG remained weak with little response to LPP 1 week after VD and/or PNC.12 Although our study was an acute one and no direct comparisons to other models were made, it seems likely that PNS is not as severe an injury as VD or PNC.
PUDENDAL NERVE STRETCH REDUCES EXTERNAL URETHRAL SPHINCTER ACTIVITY IN RATS
1393
Figure 2. EUS EMG amplitude before and after PNS or sham injury. Amplitude in PNS group was markedly decreased after injury but began to recover 10 minutes after injury. Each group showed small but not statistically significant decrease during 120 minutes. Asterisk indicates statistically significantly different vs baseline and sham injury.
We applied PNS passively by pushing apart the pelvic side wall and vagina, causing the portion of the nerve between them to stretch, without applying pressure directly to the nerve or urethra. In this way we stretched the nerve without crushing it at a site of contact. Since this method was identical in each rat, the vector of nerve stretch was constant between rats. Different PNS methods may provide different results, suggesting an area for future investigation. For the first 2 hours after PNS LPP was not significantly decreased compared to that in sham injured rats and it was comparable to that in controls in previous studies.12,13 There was a gradual, consistent decrease in LPP in each group from 30 minutes after injury, although this was not statistically significant. The EMG change without a change in LPP points to the complex dependence of LPP on multiple contributors. Periurethral smooth muscle, urethral support and mucosa, central nervous system innervation, and possibly pelvic floor muscles and fascia contribute to LPP.21,22 Since none were injured in this model, they may have maintained LPP even with a deficit in EMG. When considering the effect of an injury on the end organ, we must consider the type of injury, its severity and duration, and the time point at which the effect is measured. We used PNS 5 minutes in duration since dramatic electrophysiological differences were seen immediately in the initial pilot
procedures. In humans a prolonged second stage of labor of greater than an hour is a significant risk factor for SUI.23 Given the short life span and different physiology of rats, such periods are not necessarily comparable. LPP attains a nadir 4 days after PNC or VD, of which the latter injures the pudendal nerve distally.13,24 –26 Thus, LPP may also be decreased if studied at longer time points after PNS. LPP is reduced similarly after a VD of 1 or 4 hours, although recovery was slower with longer duration distention, suggesting that longer stretch could lead to slower LPP and/or EUS EMG recovery.20 A model that allows for survival and followup is necessary to determine whether there is a change in LPP after this injury. Our method of quantifying nerve stretch did not correlate with the severity of injury, as measured by EUS EMG. Singh et al used a similar method of marking spinal nerve roots and calibrating digital images before and after stretch but in that study the proximal end of the nerve was cut and secured to prevent slippage.10 In another series the rat tibial nerve was severed and the degree of stretch was measured directly with the proximal end fixed.26 We tested EUS function and did not sever the nerve to provide more accurate length measurements. Thus, we cannot conclude whether our measurement method was insufficiently precise and/or the degree of stretch was unimportant.
1394
PUDENDAL NERVE STRETCH REDUCES EXTERNAL URETHRAL SPHINCTER ACTIVITY IN RATS
Figure 5. Cytoplasmic staining in nucleus of Onuf, as rated by blinded observer on intensity scale of 0 to 3. A and C, -APP at 1:1,000. B and D, c-Fos at 1:1,000. A and B, intensity score of 1. C and D, intensity score of 3. Scale bar indicates 50 m.
Figure 3. Functional outcomes after PNS. A, LPP measured from baseline to peak pressure did not significantly differ between PNS and sham injured rats or significantly with time in either group. B, EUS EMG response to LPP (guarding reflex) was significantly decreased after PNS but not after sham injury. EMG amplitude during LPP began to recover 10 minutes after injury. Post-injury guarding reflex was decreased but did not attain statistical significance vs sham injury. Asterisk indicates statistically significantly different vs baseline and sham injury.
We also noted no significant change in c-Fos or -APP in the PNS group 2 hours after injury. c-Fos is an early reactive proto-oncogene that is a nonspecific marker of neuronal activity and increases after VD.16 After sciatic nerve transection c-Fos protein
Figure 4. A, pudendal nerve motor fascicles appear normal in sham injured rat. B, note rare degenerating axons in PNS rat. Scale bars indicate 10 m.
expression peaked 2 hours after injury.27 The timing of spinal cord harvest at 2 hours after PNS was based on the data described. However, prior research in rat models of traumatic brain injury showed increased c-Fos mRNA expression 5 minutes after injury, which peaked 30 minutes after injury and remained elevated 5 hours after injury.28 This suggests that 2 hours after injury we would have captured a difference if there had been one. Therefore, the extent and/or duration of stretch used was likely insufficient to measurably increase c-Fos expression. It is surprising that a 63% to 74% stretch did not induce a more obvious nerve lesion since the reported threshold is 8% to 15% and connective tissue covering in the rodent is less robust than in larger nerves and species.29 Perhaps the duration of stretch in our study was insufficient. The length of the isolated nerve and its anchor points might also be considered for improvement. Finally, although technically difficult, multiple stretch injuries with time may better simulate the clinical setting and induce more pronounced pathological conditions and physiological changes. -APP is transported by fast antegrade axoplasmic transport and it pools where transport is impaired, making it ideal for studies of traumatic axonal injury and nerve stretch.10,30 The -APP accumulation is seen as early as 30 minutes after injury in the brain.30 Previously -APP was evaluated in the central axons and peripheral nerves. To our knowledge no group has previously evaluated its accumulation in the nucleus of Onuf. Singh et al found increased -APP accumulation after spinal nerve root stretch in rats.10 However, the rat pudendal nerve is much smaller than the spinal nerve
PUDENDAL NERVE STRETCH REDUCES EXTERNAL URETHRAL SPHINCTER ACTIVITY IN RATS
root, making immunohistochemical examination of the nerve unfeasible.
CONCLUSIONS This technique serves as a promising model of PNS in the rat with an immediate, significant decrease in EUS EMG amplitude at baseline and during the guarding reflex. LPP was not affected in the acute post-injury period. Although EUS EMG activity re-
1395
covers quickly, PNS produces a measurable effect that should be studied in combination with other models of simulated childbirth injury to produce a more physiological correlate to human childbirth injury.
ACKNOWLEDGMENTS This study was done at the Lerner Research Institute, Cleveland Clinic.
REFERENCES 1. Hu TW, Wagner TH, Bentkover JD et al: Cost of urinary incontinence and overactive bladder in the United States: a comparative study. Urology 2004; 63: 461.
11. Viktrup L, Rortveit G and Lose G: Risk of stress urinary incontinence twelve years after the first pregnancy and delivery. Obstet Gynecol 2006; 108: 248.
2. Thom DH, Nygaard IE and Calhoun EA: Urologic diseases in America project: urinary incontinence in women—national trends in hospitalizations, office visits, treatment and economic impact. J Urol 2005; 173: 1295.
12. Jiang HH, Pan HQ, Gustilo-Ashby MA et al: Dual simulated childbirth injuries result in slowed recovery of pudendal nerve and urethral function. Neurourol Urodyn 2009; 28: 229.
3. Haylen BT, de Ridder D, Freeman RM et al: An international urogynecological association/international continence society joint report on the terminology for female pelvic floor dysfunction. Neurourol Urodyn 2010; 49: 4. 4. Richter HE, Burgio KL, Brubaker L et al: Factors associated with incontinence frequency in a surgical cohort of stress incontinent women. Am J Obstet Gynecol 2005; 193: 2088. 5. Hannestad YS, Rortveit G, Sandvik H et al: A community-based epidemiological survey of female urinary incontinence: the Norwegian EPINCONT study. Epidemiology of Incontinence in the County of NordTrøndelag. J Clin Epidemiol 2000; 53: 1150. 6. Snooks SJ, Swash M, Henry MM et al: Risk factors in childbirth causing damage to the pelvic floor innervation. Int J Colorectal Dis 1986; 1: 20. 7. Lien K, Morgan DMM, Delancey JOL et al: Pudendal nerve stretch during vaginal birth: A 3D computer simulation. Am J Obstet Gynecol 2005; 192: 1669. 8. Sajadi KP, Gill BC and Damaser MS: Neurogenic aspects of stress urinary incontinence. Curr Opin Obstet Gynecol 2010; 22: 425. 9. Diao E, Andrews A and Diao J: Animal models of peripheral nerve injury. Oper Tech Orthop 2004; 14: 153. 10. Singh A, Kallakuri S, Chen C et al: Structural and functional changes in nerve roots due to tension at various strains and strain rates: an in-vivo study. J Neurotrauma 2009; 26: 627.
13. Damaser MS, Broxton-King C, Ferguson C et al: Functional and neuroanatomical effects of vaginal distention and pudendal nerve crush in the female rat. J Urol 2003; 170: 1027. 14. Zutshi M, Salcedo LB, Zaszczurynski PJ et al: Effects of sphincterotomy and pudendal nerve transection on the anal sphincter in a rat model. Dis Colon Rectum 2009; 52: 1321. 15. Kerns JM, Damaser MS, Kane JM et al: Effects of pudendal nerve injury in the female rat. Neurourol Urodyn 2000; 19: 53. 16. Lin AS, Carrier S, Morgan DM et al: Effect of simulated birth trauma on the urinary continence mechanism in the rat. Urology 1998; 52: 143. 17. Ohta M, Kitamoto T, Iwaki T et al: Immunohistochemical distribution of amyloid precursor protein during normal rat development. Brain Res Dev Brain Res 1993; 75: 151. 18. Steward JE, Clemons JD, Zaszczurynski PJ et al: Quantitative evaluation of electrodes for urethral sphincter electromyography during bladder-tourethral guarding reflex. World J Urol 2010; 28: 365. 19. Allen RE, Hosker GL, Smith ARB et al: Pelvic floor damage and childbirth: a neurophysiological study. Br J Obstet Gynaecol 1990; 97: 770. 20. Yang Z, Dolber PC and Fraser MO: Differential vulnerabilities of urethral afferents in diabetes and discovery of a novel urethra-to-urethra reflex. Am J Physiol Renal Physiol 2010; 2981: F118. 21. Jiang HH, Salcedo LB, Song B et al: Pelvic floor muscles and the external urethral sphincter have
different response to applied bladder pressure during continence. Urology 2010; 75: 1515.e1. 22. Jiang HH, Salcedo LB and Damaser MS: Quantification of neurological and other contributors to continence in female rats. Brain Res 2011; 1382: 198. 23. Dolan LM, Hosker GL, Mallett VT et al: Stress incontinence and pelvic floor neurophysiology 15 years after the first delivery. BJOG 2003; 110: 1107. 24. Pan HQ, Kerns JM, Lin DL et al: Increased duration of simulated childbirth injuries results in increased time to recovery. Am J Physiol Regul Integr Comp Physiol 2007; 292: R1738. 25. Damaser MS, Samplaski MK, Parikh M et al: Time course of neuroanatomical and functional recovery after bilateral pudendal nerve injury in female rats. Am J Physiol 2007; 293: F1614. 26. Lundborg G and Rydevik B: Effects of stretching the tibial nerve of the rabbit. A preliminary study of the intraneural circulation and the barrier function of the perineurium. J Bone Joint Surg Br 1973; 55: 390. 27. Chi SI, Levine JD and Basbaum AI: Peripheral and central contributions to the persistent expression of spinal cord fos-like immunoreactivity produced by sciatic nerve transection in the rat. Brain Res 1993; 617: 225. 28. Yang K, Mu KS, Xue JJ et al: Regional and temporal profiles of c-fos and nerve growth factor mRNA expression in rat brain after lateral cortical impact injury. J Neurosci Res 1995; 42: 571. 29. Lundborg G: The nerve trunk. In: Nerve Injury and Repair: Regeneration, Reconstruction, and Cortical Remodeling, 2nd ed. Philadelphia: Elsevier Science 2004; pp 38-38. 30. Stone JR, Singleton RH and Povlishock JT: Antibodies to the C-terminus of the beta-amyloid precursor protein (APP): a site specific marker for the detection of traumatic axonal injury. Brain Res 2000; 871: 288.
Purpose: Most animal models of stress urinary incontinence simulate maternal injuries of childbirth since delivery is a major risk factor but they do not reproduce the nerve stretch known to occur during human childbirth. We hypothesized that pudendal nerve stretch produces reversible dysfunction of the external urethral sphincter. Materials and Methods: Female virgin Sprague-Dawley® rats were anesthetized with urethane. Bilateral pudendal nerve stretch or sham injury was performed for 5 minutes. External urethral sphincter electromyography and leak point pressure were recorded immediately before and after, and 10, 30, 60 and 120 minutes after pudendal nerve stretch. Post-pudendal nerve stretch results were compared to prestretch values and to values in sham injured animals. The pudendal nerves underwent qualitative histological assessment. The nucleus of Onuf was evaluated by immunohistochemistry and polymerase chain reaction for -APP and c-Fos expression as markers of neuronal activity and injury. Results: A total of 14 rats underwent bilateral pudendal nerve stretch (9) or sham injury (5). Each nerve was stretched a mean ⫾ SEM of 74% ⫾ 18% on the left side and 63% ⫾ 13% on the right side. Electromyography amplitude decreased significantly immediately after stretch compared to before stretch and after sham injury (p ⫽ 0.003) but it recovered by 30 minutes after stretch. There was no significant change in leak point pressure at any time. Two hours after injury histology showed occasional neuronal degeneration. -APP and c-Fos expression was similar in the 2 groups. Conclusions: Acute pudendal nerve stretch produces reversible electrophysiological dysfunction but without leak point pressure impairment. Pudendal nerve stretch shows promise in modeling injury. It should be tested as part of a multi-injury, chronic, physiological model of human childbirth injury.
Abbreviations and Acronyms
-APP ⫽ -amyloid precursor protein EMG ⫽ electromyography EUS ⫽ external urethral sphincter LPP ⫽ leak point pressure PBS ⫽ phosphate buffered saline PCR ⫽ polymerase chain reaction PNC ⫽ pudendal nerve crush PNS ⫽ pudendal nerve stretch SUI ⫽ stress urinary incontinence VD ⫽ vaginal distension Submitted for publication January 13, 2012. Supported by the Research Project Committee, and Glickman Urologic and Kidney Institute, Cleveland Clinic, Cleveland, Ohio. Study received institutional animal care and use committee approval. * Correspondence: Division of Urology, Oregon Health and Science University, CH10U, 3303 Southwest Bond Ave., Portland, Oregon 97232 (telephone: 503-346-1500; e-mail: kpsajadi@ gmail.com).
Key Words: urethra; urinary incontinence, stress; pudendal nerve; electromyography; disease models, animal STRESS urinary incontinence, the involuntary leakage of urine associated with increased abdominal pressure such as exercise, laughing or coughing, is a prevalent, bothersome, costly condition
affecting millions of women.1–3 SUI is strongly associated with age and vaginal childbirth, the latter due to ischemic and direct damage of the pelvic floor musculature and nerves.4,5 In
0022-5347/12/1884-1389/0 THE JOURNAL OF UROLOGY® © 2012 by AMERICAN UROLOGICAL ASSOCIATION EDUCATION
http://dx.doi.org/10.1016/j.juro.2012.06.006 Vol. 188, 1389-1395, October 2012 RESEARCH, INC. Printed in U.S.A.
AND
www.jurology.com
1389
1390
PUDENDAL NERVE STRETCH REDUCES EXTERNAL URETHRAL SPHINCTER ACTIVITY IN RATS
addition, EUS innervation via the pudendal nerve is prone to injury during vaginal childbirth by stretch and crush.6,7 Animal models of childbirth injury simulate direct pelvic floor damage via VD and pudendal nerve injury via nerve crush or avulsion.8 Three-dimensional models of human childbirth suggest that significant stretch of the pudendal nerve occurs during vaginal childbirth with 13% strain to the branch that innervates the EUS.7 Animal models of other peripheral nerve injuries indicate that this degree of stretch results in significant nerve impairment.9,10 However, to our knowledge no animal model of vaginal childbirth injury to date has specifically incorporated PNS injury. VD and/or PNC create models of SUI that are recoverable but up to 30% of women with postpartum SUI do not recover.11 Durable animal models of SUI have been described, such as pudendal nerve transection and urethrolysis, but they involve drastic injuries that do not have physiological correlates in human childbirth.8 An integrated model involving physiological mechanisms of injury, such as VD with PNC and PNS, if nonrecoverable, would more accurately replicate human vaginal childbirth injury. With the aim of designing a multiple injury model we hypothesized that PNS produces significant, reversible EUS denervation and dysfunction, as measured by anatomical and functional outcomes.12,13
MATERIALS AND METHODS Dissection and Measurement Techniques After approval from the institutional animal care and use committee age matched, virgin female Sprague-Dawley rats were anesthetized with intraperitoneal urethane (1.2 mg/kg). An anterior transpubic dissection was made to expose the bladder, urethra, pelvic floor and pudendal nerves bilaterally, as previously described.12 An inverted Y-shaped incision was made over the pelvis, the inferior epigastric vessels were ligated and divided, the pubic symphysis was divided sharply and the pubic bones were separated with blunt instruments. Bipolar parallel platinum electrodes (30 gauge needles 2 mm apart) were placed on the anterior surface of the mid urethra over the EUS to record EMG. The electrodes were connected to a Model ISO-80 amplifier (World Precision Instruments, Sarasota, Florida) with 3 Hz to 3 kHz bandpass frequencies and a Chart™ 5 electrophysiological recording system with a 10 kHz sampling rate. A pursestring 6-zero silk suture was placed in the bladder dome. Through a small cystotomy a polyethylene-50 catheter with a flared tip was introduced and secured in place as a suprapubic tube. The catheter was connected to a Model BLPR2 pressure transducer and a SYS-TBM4M Transbridge transducer amplifier (World Precision Instruments) and a Model 200 syringe pump (KD Scientific, New Hope, Pennsylvania). Vesical pressure was zeroed at the bladder level. Vesical pressure and EUS EMG were recorded simultane-
ously as the bladder was filled with saline at 5 ml per hour with rats free to void via the urethra. Baseline LPP was assessed before injury in all rats, as previously described.13 When half bladder capacity was attained, gentle pressure was applied to the rat bladder and slowly increased until leakage was seen via the urethra, when applied abdominal pressure was discontinued abruptly. The process was repeated 3 times per rat and the 3 values were used to calculate the mean. Peak pressure minus baseline pressure was recorded as LPP.
Pudendal Nerve Stretch To grossly quantify stretch 2 marks were made 5 mm apart on each nerve in the middle of the exposed segment using a sterile Kendall™ Devon surgical marker. DeBakey forceps or Castroviejo surgical calipers with the tips covered by polyethylene tubing were placed in the ischiorectal fossa beneath the nerve on each side. For rats undergoing PNS the instrument was opened to push apart the vagina and pelvic side wall, allowing the pudendal nerve to passively stretch between them for 5 minutes. For sham injured rats the instruments were left in situ for 5 minutes without opening them. Before and during stretch calibrated digital photographs were taken of the marked nerves with an ophthalmological surgical microscope at 5⫻ magnification. The distance between the marks, measured from center to center, was compared before and during stretch to quantify the degree of stretch.
Post-Stretch Assessment EUS EMG and LPP were repeated immediately after, and 10, 30, 60 and 120 minutes after bilateral stretch or sham injury. Recording began immediately after intervention since prior studies demonstrated an immediate effect with bilateral injury.14
Pudendal Neuroanatomy Two hours after injury or sham injury the rats were sacrificed via intracardiac perfusion with cold saline. The pelvic floor and pudendal nerves were excised and fixed in glutaraldehyde (2.5%) and paraformaldehyde (0.5%) fixative in cacodylate buffer, refrigerated overnight and changed to buffer after 24 hours. The pudendal nerves were microdissected from the pelvis, as previously described.15 Transverse semithin nerve sections were embedded with Durcupan™ and stained with toluidine blue. Nerve sections were evaluated qualitatively by a blinded reviewer (JMK).
Spinal Cord Evaluation A dorsal midline incision was made with the rat prone to expose the spine. Thoracolumbar laminectomy was performed to expose the lumbosacral spinal cord, which was covered with liquid nitrogen and frozen in situ. The lumbosacral cord was excised and stored in liquid nitrogen until sectioning. The area containing the nucleus of Onuf (L6-S1) was sectioned on a cryostat into 14 m sections. Alternating sections were placed on glass slides for immunohistochemistry and on membrane slides for laser microdissection and PCR.
Immunohistochemistry Alternating immunohistochemistry slides were stained for -APP, a marker of peripheral nerve traumatic stretch
PUDENDAL NERVE STRETCH REDUCES EXTERNAL URETHRAL SPHINCTER ACTIVITY IN RATS
injury, or for c-Fos, a proto-oncogene that is an indirect marker of neuronal activity.10,16 After storage in a freezer at ⫺20C slides were fixed using 10% neutral buffered formalin for 10 minutes at room temperature, rinsed thoroughly with PBS and incubated in 0.6% hydrogen peroxide in PBS for 30 minutes. Application of 3% normal goat serum (S-1000) in PBS was followed by application of Avidin D blocking solution (SP-2001) and biotin blocking solution (SP-2001, Vector Laboratories, Burlingame, California). Slides were incubated overnight at 4C with 1:1,000 rabbit monoclonal anti--APP or 1:1,000 rabbit polyclonal anti-c-Fos antibody (ab32136 and ab7963, respectively, Abcam®). Sections were exposed to secondary 1:200 biotinylated goat anti-rabbit IgG (BA-1000) and incubated with ImmPACT™ DAB Substrate (SK-4105). Slides were counterstained with hematoxylin, rinsed, dipped in bluing solution (7301, Richard Allan Scientific, Kalamazoo, Michigan), dehydrated with ethanol and xylene, and fixed with Permount™. Omission of primary antibodies served as a negative control. Positive controls consisted of postnatal day 3 rat brain for -APP17 and rat spinal cord for c-Fos. All slides were evaluated with the reviewer (KPS) blinded to the experimental group using a semiquantitative grading scale of 0 to 3 according to the intensity of staining for -APP and c-Fos.
Laser Microscope Dissection and PCR Membrane slides were removed from ⫺80C storage, thawed for 15 seconds, rehydrated with ultrapure nuclease-free water, stained with thionin, dehydrated with ethanol and fixed with xylene. Using a laser dissection microscope pudendal motoneurons in the nucleus of Onuf were collected from each specimen. Dissected cells were captured in a microcentrifuge tube containing lysis solution. RNA isolation was performed using an RNeasy® Mini Kit for RNA isolation, followed by reverse transcription with a high capacity cDNA reverse transcription kit (Applied Biosystems®). Due to the small sample size cDNA was preamplified using TaqMan® PreAmp Master Mix with amplification targeted to c-Fos (Rn02396759) or -APP (Rn00570673, Applied Biosystems). A 40 cycle PCR reaction was run for each primer and c-Fos or -APP expression was compared between the 2 groups. Expression was quantified relative to a single sham injured control using the ⌬⌬cycle threshold method with 18S rRNA as the endogenous control.
Outcomes and Statistical Analysis EUS EMG was quantitatively assessed by determining the mean rectified amplitude of the potential, as we have done previously.18 Power supply interference (60 and 120 Hz) was filtered with a digital band pass filter (59 to 61 and 119 to 121 Hz using Chart 5). LPP and EMG mean amplitude at the baseline and peak of LPP testing were calculated at each time point. A guarding reflex was defined as an increase in mean EMG amplitude during LPP testing compared to baseline amplitude. Data were analyzed with 2-way repeated measures ANOVA. Post hoc pairwise comparisons to pre-injury measurements were performed using the Dunnett test. Results were tested for correlation with the percent of stretch and averaged between the left and right sides using the Spearman corre-
1391
lation. Semiquantitative immunohistochemistry and realtime PCR results were compared between the sham injured and PNS groups using the Student t test. For all statistical tests p ⬍0.05 was considered statistically significant. SigmaPlot™ v11 was used. Quantitative data are shown as the mean ⫾ SEM.
RESULTS EMG and LPP An initial series of PNS procedures performed to assess technique feasibility showed a dramatic change in EUS EMG after 5 minutes of stretch. Thus, we stretched the nerve for 5 minutes in subsequent rats. A total of 24 rats underwent PNS or sham injury. Rats were excluded from study due to excessive noise during EMG that prohibited adequate analysis. Rats were also excluded when pre-injury mean EMG amplitude was less than 5 V (less than the 5th percentile) since they likely experienced inadvertent nerve injury during dissection before injury was applied. Nine PNS and 5 sham injured rats weighing a mean of 266 ⫾ 5 gm were included in analysis. In rats that underwent PNS the pudendal nerve was stretched a mean of 74% ⫾ 18% on the left side and 63% ⫾ 13% on the right side, which was not significantly different. EUS EMG amplitude was significantly decreased immediately after PNS but recovered quickly thereafter (p ⬍0.001, figs. 1, A and 2). Sham nerve stretch did not affect EUS EMG. A guarding reflex was present before PNS and in sham injured animals at all time points. In contrast, it was significantly decreased immediately after PNS (p ⫽ 0.003, fig. 3). It began to return 30 minutes later (fig. 1, B). EMG amplitude did not vary significantly with time in sham injured animals (fig. 2). This suggests that extended anesthesia as well as dissection and exposure of the nerve did not detrimentally affect function. As a fraction of the pre-injury value, average EMG amplitude immediately after injury was 48% in the PNS group and 102% in the sham injured group. EMG amplitude during LPP immediately after PNS was lower than after sham injury but not significantly so (mean 26 ⫾ 5 vs 38 ⫾ 3 V, p ⫽ 0.09). EMG amplitude remained significantly decreased 10 minutes after PNS but on post hoc analysis this was not significantly different from sham injury or prestretch values. The decrease in EMG amplitude did not significantly correlate with the percent of nerve stretch. There was minimal change in LPP from before to immediately after stretch (mean 43 ⫾ 6 to 40 ⫾ 15 cm H2O) (fig. 3, A). Changes in LPP with time were not significant (p ⫽ 0.421). There was a slight decrease in LPP 120 minutes after sham stretch, similar to that after PNS, but this was not significantly
1392
PUDENDAL NERVE STRETCH REDUCES EXTERNAL URETHRAL SPHINCTER ACTIVITY IN RATS
Figure 1. A, characteristic baseline EMG during filling before and after PNS and sham injury. EMG immediately after injury in PNS group shows mostly background noise (sine wave) but motor units began returning 10 minutes after injury. Sham injured EMG is largely unchanged. B, combined EMG and vesical pressure show guarding reflex, defined as increased EMG activity during LPP testing, before PNS. Response was blunted after PNS but recovered 30 minutes after injury.
different from baseline (mean 21 ⫾ 3.4 vs 33 ⫾ 6.9 cm H2O, p ⫽ 0.077). Histology Small motor fascicles of the nerve were normal in appearance in PNS and sham injured specimens. Isolated axons in the larger fascicles from the PNS group rarely showed degeneration (fig. 4). Nonetheless, there was no apparent structural difference between pudendal nerves subjected to PNS or sham injury. Cytoplasmic staining for -APP and nuclear staining for c-Fos were observed in the nucleus of Onuf in all PNS and sham injured animals (fig. 5). On a scale of 0 to 3 there was no significant difference in the semiquantitative score between the PNS and sham injured groups for -APP (mean 1.4 ⫾ 0.7, range 0 to 3 vs 1.6 ⫾ 0.2, range 1 to 2) or c-Fos (mean 2.2 ⫾ 0.4, range 1 to 3 vs 2 ⫾ 0, range 2 to 2). Quantitative Real-time PCR Expression of c-Fos mRNA was not significantly different after PNS than after sham injury (1.13 ⫾ 0.11 vs 0.93 ⫾ 0.05). Similarly -APP mRNA expression
was not significantly different after PNS compared to that in sham injured rats (mean 1.24 ⫾ 0.36 vs 0.84 ⫾ 0.09).
DISCUSSION We developed a method of PNS that produces a rapid, demonstrable and reversible decrease in EUS EMG activity. EUS EMG is dysfunctional in women with urodynamic stress incontinence19 and it has been used to measure urethral function in animal models.12,20,21 The primary outcome of this study was EUS EMG since it reflects changes immediately after injury.14,15 PNS rats demonstrated an immediate decrease in EUS EMG activity but this recovered relatively quickly (by 10 minutes) with subsequent return of the guarding reflex. In contrast, EUS EMG remained weak with little response to LPP 1 week after VD and/or PNC.12 Although our study was an acute one and no direct comparisons to other models were made, it seems likely that PNS is not as severe an injury as VD or PNC.
PUDENDAL NERVE STRETCH REDUCES EXTERNAL URETHRAL SPHINCTER ACTIVITY IN RATS
1393
Figure 2. EUS EMG amplitude before and after PNS or sham injury. Amplitude in PNS group was markedly decreased after injury but began to recover 10 minutes after injury. Each group showed small but not statistically significant decrease during 120 minutes. Asterisk indicates statistically significantly different vs baseline and sham injury.
We applied PNS passively by pushing apart the pelvic side wall and vagina, causing the portion of the nerve between them to stretch, without applying pressure directly to the nerve or urethra. In this way we stretched the nerve without crushing it at a site of contact. Since this method was identical in each rat, the vector of nerve stretch was constant between rats. Different PNS methods may provide different results, suggesting an area for future investigation. For the first 2 hours after PNS LPP was not significantly decreased compared to that in sham injured rats and it was comparable to that in controls in previous studies.12,13 There was a gradual, consistent decrease in LPP in each group from 30 minutes after injury, although this was not statistically significant. The EMG change without a change in LPP points to the complex dependence of LPP on multiple contributors. Periurethral smooth muscle, urethral support and mucosa, central nervous system innervation, and possibly pelvic floor muscles and fascia contribute to LPP.21,22 Since none were injured in this model, they may have maintained LPP even with a deficit in EMG. When considering the effect of an injury on the end organ, we must consider the type of injury, its severity and duration, and the time point at which the effect is measured. We used PNS 5 minutes in duration since dramatic electrophysiological differences were seen immediately in the initial pilot
procedures. In humans a prolonged second stage of labor of greater than an hour is a significant risk factor for SUI.23 Given the short life span and different physiology of rats, such periods are not necessarily comparable. LPP attains a nadir 4 days after PNC or VD, of which the latter injures the pudendal nerve distally.13,24 –26 Thus, LPP may also be decreased if studied at longer time points after PNS. LPP is reduced similarly after a VD of 1 or 4 hours, although recovery was slower with longer duration distention, suggesting that longer stretch could lead to slower LPP and/or EUS EMG recovery.20 A model that allows for survival and followup is necessary to determine whether there is a change in LPP after this injury. Our method of quantifying nerve stretch did not correlate with the severity of injury, as measured by EUS EMG. Singh et al used a similar method of marking spinal nerve roots and calibrating digital images before and after stretch but in that study the proximal end of the nerve was cut and secured to prevent slippage.10 In another series the rat tibial nerve was severed and the degree of stretch was measured directly with the proximal end fixed.26 We tested EUS function and did not sever the nerve to provide more accurate length measurements. Thus, we cannot conclude whether our measurement method was insufficiently precise and/or the degree of stretch was unimportant.
1394
PUDENDAL NERVE STRETCH REDUCES EXTERNAL URETHRAL SPHINCTER ACTIVITY IN RATS
Figure 5. Cytoplasmic staining in nucleus of Onuf, as rated by blinded observer on intensity scale of 0 to 3. A and C, -APP at 1:1,000. B and D, c-Fos at 1:1,000. A and B, intensity score of 1. C and D, intensity score of 3. Scale bar indicates 50 m.
Figure 3. Functional outcomes after PNS. A, LPP measured from baseline to peak pressure did not significantly differ between PNS and sham injured rats or significantly with time in either group. B, EUS EMG response to LPP (guarding reflex) was significantly decreased after PNS but not after sham injury. EMG amplitude during LPP began to recover 10 minutes after injury. Post-injury guarding reflex was decreased but did not attain statistical significance vs sham injury. Asterisk indicates statistically significantly different vs baseline and sham injury.
We also noted no significant change in c-Fos or -APP in the PNS group 2 hours after injury. c-Fos is an early reactive proto-oncogene that is a nonspecific marker of neuronal activity and increases after VD.16 After sciatic nerve transection c-Fos protein
Figure 4. A, pudendal nerve motor fascicles appear normal in sham injured rat. B, note rare degenerating axons in PNS rat. Scale bars indicate 10 m.
expression peaked 2 hours after injury.27 The timing of spinal cord harvest at 2 hours after PNS was based on the data described. However, prior research in rat models of traumatic brain injury showed increased c-Fos mRNA expression 5 minutes after injury, which peaked 30 minutes after injury and remained elevated 5 hours after injury.28 This suggests that 2 hours after injury we would have captured a difference if there had been one. Therefore, the extent and/or duration of stretch used was likely insufficient to measurably increase c-Fos expression. It is surprising that a 63% to 74% stretch did not induce a more obvious nerve lesion since the reported threshold is 8% to 15% and connective tissue covering in the rodent is less robust than in larger nerves and species.29 Perhaps the duration of stretch in our study was insufficient. The length of the isolated nerve and its anchor points might also be considered for improvement. Finally, although technically difficult, multiple stretch injuries with time may better simulate the clinical setting and induce more pronounced pathological conditions and physiological changes. -APP is transported by fast antegrade axoplasmic transport and it pools where transport is impaired, making it ideal for studies of traumatic axonal injury and nerve stretch.10,30 The -APP accumulation is seen as early as 30 minutes after injury in the brain.30 Previously -APP was evaluated in the central axons and peripheral nerves. To our knowledge no group has previously evaluated its accumulation in the nucleus of Onuf. Singh et al found increased -APP accumulation after spinal nerve root stretch in rats.10 However, the rat pudendal nerve is much smaller than the spinal nerve
PUDENDAL NERVE STRETCH REDUCES EXTERNAL URETHRAL SPHINCTER ACTIVITY IN RATS
root, making immunohistochemical examination of the nerve unfeasible.
CONCLUSIONS This technique serves as a promising model of PNS in the rat with an immediate, significant decrease in EUS EMG amplitude at baseline and during the guarding reflex. LPP was not affected in the acute post-injury period. Although EUS EMG activity re-
1395
covers quickly, PNS produces a measurable effect that should be studied in combination with other models of simulated childbirth injury to produce a more physiological correlate to human childbirth injury.
ACKNOWLEDGMENTS This study was done at the Lerner Research Institute, Cleveland Clinic.
REFERENCES 1. Hu TW, Wagner TH, Bentkover JD et al: Cost of urinary incontinence and overactive bladder in the United States: a comparative study. Urology 2004; 63: 461.
11. Viktrup L, Rortveit G and Lose G: Risk of stress urinary incontinence twelve years after the first pregnancy and delivery. Obstet Gynecol 2006; 108: 248.
2. Thom DH, Nygaard IE and Calhoun EA: Urologic diseases in America project: urinary incontinence in women—national trends in hospitalizations, office visits, treatment and economic impact. J Urol 2005; 173: 1295.
12. Jiang HH, Pan HQ, Gustilo-Ashby MA et al: Dual simulated childbirth injuries result in slowed recovery of pudendal nerve and urethral function. Neurourol Urodyn 2009; 28: 229.
3. Haylen BT, de Ridder D, Freeman RM et al: An international urogynecological association/international continence society joint report on the terminology for female pelvic floor dysfunction. Neurourol Urodyn 2010; 49: 4. 4. Richter HE, Burgio KL, Brubaker L et al: Factors associated with incontinence frequency in a surgical cohort of stress incontinent women. Am J Obstet Gynecol 2005; 193: 2088. 5. Hannestad YS, Rortveit G, Sandvik H et al: A community-based epidemiological survey of female urinary incontinence: the Norwegian EPINCONT study. Epidemiology of Incontinence in the County of NordTrøndelag. J Clin Epidemiol 2000; 53: 1150. 6. Snooks SJ, Swash M, Henry MM et al: Risk factors in childbirth causing damage to the pelvic floor innervation. Int J Colorectal Dis 1986; 1: 20. 7. Lien K, Morgan DMM, Delancey JOL et al: Pudendal nerve stretch during vaginal birth: A 3D computer simulation. Am J Obstet Gynecol 2005; 192: 1669. 8. Sajadi KP, Gill BC and Damaser MS: Neurogenic aspects of stress urinary incontinence. Curr Opin Obstet Gynecol 2010; 22: 425. 9. Diao E, Andrews A and Diao J: Animal models of peripheral nerve injury. Oper Tech Orthop 2004; 14: 153. 10. Singh A, Kallakuri S, Chen C et al: Structural and functional changes in nerve roots due to tension at various strains and strain rates: an in-vivo study. J Neurotrauma 2009; 26: 627.
13. Damaser MS, Broxton-King C, Ferguson C et al: Functional and neuroanatomical effects of vaginal distention and pudendal nerve crush in the female rat. J Urol 2003; 170: 1027. 14. Zutshi M, Salcedo LB, Zaszczurynski PJ et al: Effects of sphincterotomy and pudendal nerve transection on the anal sphincter in a rat model. Dis Colon Rectum 2009; 52: 1321. 15. Kerns JM, Damaser MS, Kane JM et al: Effects of pudendal nerve injury in the female rat. Neurourol Urodyn 2000; 19: 53. 16. Lin AS, Carrier S, Morgan DM et al: Effect of simulated birth trauma on the urinary continence mechanism in the rat. Urology 1998; 52: 143. 17. Ohta M, Kitamoto T, Iwaki T et al: Immunohistochemical distribution of amyloid precursor protein during normal rat development. Brain Res Dev Brain Res 1993; 75: 151. 18. Steward JE, Clemons JD, Zaszczurynski PJ et al: Quantitative evaluation of electrodes for urethral sphincter electromyography during bladder-tourethral guarding reflex. World J Urol 2010; 28: 365. 19. Allen RE, Hosker GL, Smith ARB et al: Pelvic floor damage and childbirth: a neurophysiological study. Br J Obstet Gynaecol 1990; 97: 770. 20. Yang Z, Dolber PC and Fraser MO: Differential vulnerabilities of urethral afferents in diabetes and discovery of a novel urethra-to-urethra reflex. Am J Physiol Renal Physiol 2010; 2981: F118. 21. Jiang HH, Salcedo LB, Song B et al: Pelvic floor muscles and the external urethral sphincter have
different response to applied bladder pressure during continence. Urology 2010; 75: 1515.e1. 22. Jiang HH, Salcedo LB and Damaser MS: Quantification of neurological and other contributors to continence in female rats. Brain Res 2011; 1382: 198. 23. Dolan LM, Hosker GL, Mallett VT et al: Stress incontinence and pelvic floor neurophysiology 15 years after the first delivery. BJOG 2003; 110: 1107. 24. Pan HQ, Kerns JM, Lin DL et al: Increased duration of simulated childbirth injuries results in increased time to recovery. Am J Physiol Regul Integr Comp Physiol 2007; 292: R1738. 25. Damaser MS, Samplaski MK, Parikh M et al: Time course of neuroanatomical and functional recovery after bilateral pudendal nerve injury in female rats. Am J Physiol 2007; 293: F1614. 26. Lundborg G and Rydevik B: Effects of stretching the tibial nerve of the rabbit. A preliminary study of the intraneural circulation and the barrier function of the perineurium. J Bone Joint Surg Br 1973; 55: 390. 27. Chi SI, Levine JD and Basbaum AI: Peripheral and central contributions to the persistent expression of spinal cord fos-like immunoreactivity produced by sciatic nerve transection in the rat. Brain Res 1993; 617: 225. 28. Yang K, Mu KS, Xue JJ et al: Regional and temporal profiles of c-fos and nerve growth factor mRNA expression in rat brain after lateral cortical impact injury. J Neurosci Res 1995; 42: 571. 29. Lundborg G: The nerve trunk. In: Nerve Injury and Repair: Regeneration, Reconstruction, and Cortical Remodeling, 2nd ed. Philadelphia: Elsevier Science 2004; pp 38-38. 30. Stone JR, Singleton RH and Povlishock JT: Antibodies to the C-terminus of the beta-amyloid precursor protein (APP): a site specific marker for the detection of traumatic axonal injury. Brain Res 2000; 871: 288.