STRUCTURAL ASSESSMENT OF THE URETHRAL SPHINCTER IN WOMEN WITH URINARY RETENTION

0022-5347/05/1734-1246/0 THE JOURNAL OF UROLOGY® Copyright © 2005 by AMERICAN UROLOGICAL ASSOCIATION

Vol. 173, 1246 –1251, April 2005 Printed in U.S.A.

DOI: 10.1097/01.ju.0000156833.84907.3d

STRUCTURAL ASSESSMENT OF THE URETHRAL SPHINCTER IN WOMEN WITH URINARY RETENTION D. E. ANDRICH,* D. RICKARDS, D. N. LANDON, C. J. FOWLER

AND

A. R. MUNDY

From the Institute of Urology, the Institute of Neurology (DNL, CJF), and The National Hospital for Neurology and Neurosurgery (CJF), Queen Square, London, United Kingdom

ABSTRACT

Purpose: The pathophysiology of urinary retention in women is generally unknown but a subgroup of women with urinary retention have been diagnosed as having so-called primary disorder of sphincter relaxation on the basis of an abnormal urethral sphincter electromyogram. It was suggested this sphincter overactivity could lead to work hypertrophy of the urethral rhabdosphincter and in this study we looked for any evidence of such muscle fiber hypertrophy. Materials and Methods: In 9 women 18 to 45 years old (mean age 31.6) with urinary retention and overactive urethral sphincter electromyogram, light and electron microscopy were used to examine core needle biopsies of the urethral rhabdosphincter taken under transvaginal ultrasound control. Of the 9 patients only 5 biopsies processed for light microscopy and 4 processed for electron microscopy contained striated urethral muscle fibers. The results of these biopsies were compared to the morphology of a control specimen from a postmenopausal woman without a history of urinary retention. Results: On light microscopy the urethral rhabdosphincter fiber diameter did not differ among patients (mean average 7.6 ␮m), was less than that reported in the literature (15 to 20), but did not differ from that of the control (mean 9.9). In all patients electron microscopy showed excessive peripheral sarcoplasm with lipid and glycogen deposition, and sarcoplasmic accumulation of normal mitochondria. These ultrastructural abnormalities were not seen in the control. Conclusions: To our knowledge this is the first morphological description of the urethral rhabdosphincter in a subgroup of women with urinary retention. Mean rhabdosphincter fiber diameter was approximately the same in patients and controls. This study does not support the previous theory that urethral sphincter overactivity in a subgroup of women with urinary retention leads to work hyperplasia of urethral rhabdosphincter fibers. An alternative hypothesis is suggested. KEY WORDS: urinary retention, histology; microscopy, electron; muscle fibers, slow-twitch; urethra

Obstructed voiding in women can present clinically in a variety of ways from mainly irritative lower urinary tract symptoms with partial retention to complete retention of urine. Intermittent self-catheterization is the mainstay of treatment, although recently sacral neuromodulation has been introduced in this group of patients.1 In the last 3 decades anatomical, neurophysiological and clinical urodynamic based studies have demonstrated the structural components of the urethral sphincter and their function, but the pathogenesis of urinary retention in women remains elusive. In the 1970s it was suggested that external sphincter spasticity was the cause of functional urinary outflow obstruction in women with idiopathic retention of urine.2, 3 Detrusor areflexia with failure of the external urethral sphincter to relax was demonstrated in many urodynamic studies4 and it was speculated that local reflex mechanisms at the spinal cord level and suprasacral psychosomatic influences are important in the pathophysiology.5 Although emotional and psychosomatic influences were identified in the pathogenesis of idiopathic retention and

widely believed to be important,6 – 8 treating these women with behavioral therapy never gained popularity.9 In the mid 1980s the psychogenic concept of the failure of the external sphincter to relax was challenged on the basis of finding abnormal myotonic-like electromyogram (EMG) activity in the urethral rhabdosphincter of women with urinary retention.10, 11 Fowler’s syndrome was formulated to consist of urinary retention, abnormal electromyographic activity of the urethral rhabdosphincter and polycystic ovaries.12 It was further speculated that this overactivity would lead to changes of work hypertrophy of the urethral rhabdosphincter.13 MATERIALS AND METHODS

There were 9 women between 18 and 45 years old (mean age 31.6) with urinary retention of at least 2 years who showed electromyographic signs of urethral sphincter overactivity and were recruited to the study, which was approved by the hospital trust research ethics committee. Of the 9 patients 3 had partial retention, 4 had complete retention, and the other 2 started with partial retention which gradually deteriorated into complete retention. Precipitating factors included urinary tract infection (UTI), postoperative retention and bed-wetting (table 1). Of the 4 patients with retention after urinary tract infection 3 had partial retention. Vaginal delivery in our patients ranged from 0 to 3 (average 1.2). One woman had polycystic ovaries, and another woman was postmenopausal and taking hormone replacement ther-

Submitted for publication March 1, 2004. Study received hospital trust research ethics committee approval. Nothing to disclose. * Correspondence: Institute of Urology, 48 Riding House St., London W1P 7PN (e-mail: [email protected]). Editor’s Note: This article is the fourth of 5 published in this issue for which category 1 CME credits can be earned. Instructions for obtaining credits are given with the questions on pages 1438 and 1439. 1246

CISC

⫹

⫺

⫺

⫺

Ab

CISC

Medication

Bladder emptying EMG assisted biopsy Striated muscle content: Light microscopy biopsy score Electron microscopy biopsy score

Polycystic ovaries, dysphagia, childhood UTI, ileostomy for constipation

Hysterectomy

Lower back pain, asthma, irritable bowel syndrome Inhaler

Medical history

Acute ⫹ persistent

Acute ⫹ persistent

Acute ⫹ persistent

⫹

⫺

CISC

Complete

Complete

Complete

28 0 Postop

3

Retention degree Retention development

39 II Postop

2

26 II Single enuresis episode

1

Age Para Retention onset

Pt No.

4

⫺

⫹⫹

CISC

Sodium valproate, carbamazepine

Partial retention after hysterectomy, complete retention after appendectomy Epilepsy

Complete

37 II Postop

⫺

⫺

Cetirizine hydrochloride, inhaler, Ab

Codeine ⫹ acetaminophen, dihydrocodeine, diclofenac sodium, tramadol, sulfasol, Ab Void ⫹ CISC

⫺

⫹⫹

Void Yes

Asthma

6

Spinal syrinx

Partial

45 III UTI

Gradual

5

Gradual

Partial

18 0 UTI

TABLE 1. Patient characteristics 7

8

⫹

⫹

⫹⫹

CISC Yes

Void ⫹ CISC Yes

⫹⫹⫹

Fluoxetine hydrochloride, inhaler

Partial retention after appendectomy, with time complete retention Asthma

Complete

23 0 Postop

Hormone replacement therapy

Hysterectomy (not followed by postop retention!)

Gradual

Partial

44 II UTI

9

⫹

⫺

CISC Yes

Carbamazepine, Ab

Epilepsy, endometriosis, childhood UTI, sex therapy

Acute ⫹ persistent

22 0 Childhood UTI, acute urinary retention at 14 yrs Complete

ASSESSMENT OF URETHRAL SPHINCTER IN WOMEN WITH URINARY RETENTION

1247

1248

ASSESSMENT OF URETHRAL SPHINCTER IN WOMEN WITH URINARY RETENTION

apy during the investigation. Of the study participants 8 performed clean intermittent self-catheterization (CISC). Control tissue was obtained from a 60-year-old multiparous postmenopausal woman who underwent cystourethrectomy for histologically proven chronic active cystitis. She received chemotherapy for breast cancer 2 years before cystectomy, which was initially fluorouracil, epirubicin and cyclophosphamide followed by mitomycin C, methotrexate and paclitaxel. The oncologist deliberately omitted methotrexate in view of bladder dysfunction and considered that the other chemotherapeutic agents were unlikely to cause any bladder symptoms. She had no previous history of obstructed voiding. We are aware of the limitations of having only 1 control subject without EMG evaluation, but for this challenge logistical it was extremely difficult to obtain age matched fresh urethral tissue. Transvaginal ultrasound guided urethral biopsy. In study patients urethral core needle biopsies were performed under transvaginal ultrasound control. The patient was positioned in left lateral position and prophylactic antibiotics (Ab 120 mg gentamicin) were administered intravenously. Urethral tissue was anesthetized with 10 ml of 1% lidocaine and the patient was catheterized with a urethral catheter (16Fr Foley) to facilitate visualization of the urethral lumen. Urethral cores (3 to 4) were taken with a Tru-Cut biopsy needle (Pro-Mag 2.2 automated biopsy gun, 18 gauge ⫻ 20 mm ACN biopsy needle, Manan Medical Products, Wheeling, Illinois, PBN Medical, Stenlose, Denmark) running through a guide of the transvaginal ultrasound probe (9 MHz transrectal probe). The biopsy needle was aimed at the most lateral aspect of the hypoechoic zone anteriorly where the bulk of the omega-shaped urethral rhabdosphincter fibers were expected to be (fig. 1). The biopsy cores were immediately put into fixative solution for light microscopy (1% glutaraldehyde/ 1.5% paraformaldehyde) and electron microscopy (3% glutaraldehyde buffered at pH 7.4 in 0.1 molar sodium cacodylate and 5 mM calcium chloride). Local pressure was applied to the biopsy site until bleeding stopped (usually after 5 to 10 minutes). EMG assisted transvaginal ultrasound guided urethral biopsy. The biopsies of the last 4 patients were taken with additional electromyographic guidance to improve rhabdosphincter localization accuracy. The biopsy needle was di-

FIG. 1. Longitudinal transvaginal ultrasound section of urethra in woman with urinary retention. Catheter balloon is seen in left corner. Catheter delineates urethral lumen with hyperechoic zone, and more peripheral hypoechoic zone representing different muscle layers of urethra. Only in most peripheral part of hypoechoic zone was there electromyographic activity, indicating location of urethral rhabdosphincter. Tip of EMG needle is seen in right corner.

rected at the tip of the concentric EMG needle where action potentials of the urethral rhabdosphincter were detected. The necessary advancement in depth was estimated by knowing the forward thrust of the biopsy needle from the point of entry into the tissue. After the biopsy procedure all but 1 patient remained catheterized until pain-free and selfcatheterization could be resumed. Patients were contacted by telephone to monitor adverse events. In terms of control tissue retrieval, the urethra of the control patient was transected from the bladder neck perioperatively, and immediately put into fixatives for light microscopy and electron microscopy as previously described. Urethral tissue processing: light and electron microscopy. Light microscopy sections of patient and control tissue were routinely stained with hematoxylin and eosin, and photographs taken at low power (fig. 1). For electron microscopy patient and control tissues were cut into 5 mm pieces in sequential order. The tissue was post-fixed en bloc with 1% aqueous osmium tetroxide for 2 hours, dehydrated and embedded in agar 100 resin and kept overnight at 60C for polymerization. Semi-thin sections were cut at 1 ␮m, stained with 1% toluidine blue/1% borax for light microscopic examination and photographically documented with a Nikon FXA (Nikon Inc., Garden City, New York). All blocks with striated muscle content were ultrathin sectioned and sections were collected on naked 3 mm 300 mesh copper grids. They were then stained with 25% uranyl acetate in 50% methyl alcohol for 20 minutes, followed by Reynold’s lead citrate for another 20 minutes and then examined in a Jeol 1200 EX (Jeol Ltd., Tokyo, Japan) electron microscope at 80 kV. Muscle fiber types and ultrastructural characteristics were identified by DNL14 and micrographs were made of areas of interest. Morphometric fiber diameter analysis. For the control and each of the 5 patients with striated urethral muscle fibers the entire light microscopy biopsy core was photographed and the striated muscle fibers counted (JVC Video 3-CCD camera, Leica monitor and Adobe and QWin Colour program software). Muscle fiber diameter was calculated from the cross-sectional area in which 450 pixels equaled 400 ␮m2. We pragmatically excluded obliquely cut myofibers from fiber diameter analysis by choosing a width-to-length ratio of greater than or equal to 2:1 to define oblique myofibers. The thereby obtained corrected fiber diameter distribution was statistically analyzed (fig. 2 and table 2).

FIG. 2. Box plot diagram comparing mean fiber diameter distribution of control and women with urinary retention. Median fiber diameter is similar between patients (2– 6) and control (1). Box represents interquartile range (IQR), bar median value and whiskers represent data points within IQR. Dots represent data outside IQR (outliers).

ASSESSMENT OF URETHRAL SPHINCTER IN WOMEN WITH URINARY RETENTION

1249

TABLE 2. Fiber diameter analysis Observations

Control Pt

Total myofibers/core Analyzed fibers/core Min fiber diameter (␮m) Max fiber diameter (␮m) Mean (␮m) Standard deviation

85 56 3.9 19.4 9.85 3.85

Pt No. 1 203 150 2.8 13.2 6.12 1.78

RESULTS

Our patients tolerated the biopsy procedure well without serious complications. Of 9 patients 4 had mild to moderate self-limiting bleeding for 1 or 2 days. Discomfort at the biopsy site occurred in 2 patients until the day after the procedure and was controlled with oral analgesia. In 1 of the 2 patients discomfort was probably more related to bladder spasm since catheter removal resolved the problem. A clinical urinary tract infection only developed in the patient who refused urethral catheterization following the biopsy, which resolved with oral antibiotics. There was no evidence in any patient that sphincter function was affected adversely by the biopsy procedure. Additional electromyographic localization of the urethral rhabdosphincter seemed to increase the number of positive biopsy cores for striated muscle content but this difference could not be analyzed statistically due to the small number of patients. The sphincter EMG added 10 to 15 minutes to the biopsy procedure but did not lead to increased morbidity. Of the 9 patient biopsy cores 5 contained a variable number of urethral rhabdosphincter fibers. As seen in figure 3 the muscle fibers were cut at different degrees of obliquity and at variable locations of each given myofiber. All striated muscle fibers of each 5 cores were counted but only about half were used for fiber diameter analysis after oblique myofibers were disregarded. The number of analyzed striated muscle fibers per core varied between 150 and 541 (median 244) and the 5 patients had an average striated urethral myofiber diameter of 7.6 ␮m (range 6.1 to 8.6), which was slightly less than the mean fiber diameter of our control patient (9.9 ␮m) (table 2). Mean fiber diameter distribution comparing control with pa-

FIG. 3. Light micrograph of urethral rhabdosphincter biopsy in woman with urinary retention. Bundles of transverse, oblique and longitudinal striated muscle fibers are seen with connective tissue separating bundles and each individual muscle fiber. Small muscle fibers differ in diameter. It is clearly seen in right corner that increase in fiber diameter is due to fact that muscle fibers are cut obliquely. At edge in right corner 1 fiber is even cut longitudinally. Toward left side striated muscle bulk tapers out, and striated muscle fibers intermingle with smooth muscle cells and connective tissue. There is muscle fiber bundle with quite small and circular cut surface which could represent end of spindle-shaped muscle fibers. H & E, reduced from ⫻500.

2 1,135 541 3.2 23.3 8.06 3.05

3 473 244 2.6 19 7.87 3.51

4 350 179 2.8 25.9 7.55 3.95

5 923 407 3.2 18.2 8.59 2.6

tients is illustrated in figure 2. Median striated myofiber diameter is similar between patients and control. Although the control core contained fewer myofibers for analysis, statistical expert opinion was that this number was sufficient for analysis overall. There was no obvious correlation between fiber size and age of the patient, parity or menopausal status, or history of polycystic ovaries. Only 4 of 9 patient cores for electron microscopic examination contained urethral rhabdosphincter fibers. As expected with any muscle biopsy in which length is not controlled, the myofibers were hypercontracted. Many of the examined urethral rhabdosphincter fibers of all 4 patients showed abnormal amounts of sarcoplasm with increased glycogen deposition and accumulation of ultrastructurally normal mitochondria (figs. 4 and 5). These findings were not seen in the control sample (fig. 6). The majority of urethral rhabdosphincter fibers showed ultrastructural characteristics of type I slow twitch fibers14 but formal biochemical fiber type analysis was not undertaken in this study.

FIG. 4. Excess glycogen. Relaxed fiber is shown in which myofibrils are cut obliquely but are widely separated by excessive amounts of interstitial glycogen. Most myofibers were hypercontracted whereby I-bands disappear and Z-line is in close apposition to A-band.

1250

ASSESSMENT OF URETHRAL SPHINCTER IN WOMEN WITH URINARY RETENTION

FIG. 5. Sarcoplasmic accumulation of mitochondria. Sarcoplasmic hyperplasia is seen in partial longitudinal section of hypercontracted fiber showing nucleus at edge of myofiber surrounded by mass of sarcoplasm rich in mitochondria.

DISCUSSION

The prevalence of nonurological, nonneurogenic urinary retention in women is unknown but it is a heterogenous group of patients with a diverse clinical presentation. In this study we investigated a selected group in whom the only common variable was the electromyographic urethral rhabdosphincter overactivity. To obtain samples of urethral rhabdosphincter we used the established ultrasound guided core needle biopsy technique similar to the transrectal ultrasound guided biopsy of the prostate gland. This method proved to be well tolerated and did not cause major complications. Urethral rhabdosphincter orientation. It became clear that methodological difficulties had to be taken into account when urethral rhabdosphincter biopsy specimens are the material to be studied. The loose bundles of the urethral rhabdosphincter run in slightly differing orientations but are mainly oblique around the female urethra forming the bulk at mid urethral level.15 It is impossible to orient the biopsy needle at an angle that would hit all fibers in cross-section. Unlike with standard skeletal muscle biopsy taken by surgical excision where the muscle is fixed in stretched longitudinal orientation, the muscle fibers of a needle biopsy will coil due to lack of stretch and subsequent tissue fixation will increase total contraction even more.16 The control patient tissue, which was taken by excision biopsy, was also not fixed in stretched orientation. Fiber obliquity. There are few reports in the literature on female urethral rhabdosphincter fiber diameter measurements but it seems that the problem of fiber obliquity has not been addressed methodologically.17, 18 Schroder and Reske-Nielsen measured the lesser diameter of a fiber19 and other investigators even measured the maximum oblique

FIG. 6. Ultrastructure of control striated muscle fibers. Oblique section through control fiber showing close packing of myofibrils without excess glycogen or mitochondria. Narrow rim of sarcoplasm is normal.

diameter of muscle fibers.20 The greater the degree of obliquity the greater the error of measurement. A variety of approaches to this problem have been explored in applications to skeletal muscle but to our knowledge there is no report in the literature addressing this issue with particular reference to the urethral rhabdosphincter. Although striated urethral muscle fibers are grossly oriented in bundles, nevertheless they originate at different levels. Muscle fibers are spindleshaped and orientation is oblique. Therefore, mixtures of different fiber diameters of varying obliquity are to be expected. We dealt with this problem by excluding every muscle fiber with a cut surface width-to-length ratio greater than or equal to 2:1 so that major error could be avoided. Other ultrastructural methods of identifying true cross-section as used in skeletal muscle morphometry proved impracticable due to the limited number of fibers within a core in true cross-section. With our chosen width-to-length ratio we worked with a practical approximation of the true transverse fiber diameter. We disregarded almost half of the fibers in a core due to obliquity. We suspect that a similar effort to deal with fiber obliquity was not undertaken by other investigators who obtained fiber diameters approximately twice the size of those in this study. Fiber diameter. In this study the mean urethral rhabdosphincter fiber diameter in women with urinary retention was 7.6 ␮m (range 6.1 to 8.6), which was slightly less than the mean fiber diameter of the control patient (9.9 ␮m). Gosling et al quoted an average urethral rhabdosphincter fiber size of 15 to 20 ␮m.15 Ho et al assessed the relationship of muscle fiber diameter and sarcolemmal neuronal nitric oxide synthase (snNOS) immunoreactivity.17 They reported a fiber diameter of 10.1 to 15 ␮m for snNOS negative and 15.1 to 20 ␮m for snNOS positive urethral striated muscle fibers

ASSESSMENT OF URETHRAL SPHINCTER IN WOMEN WITH URINARY RETENTION

obtained from postmortem cystourethrectomy specimens of 4 female cadaveric organ donors age 40 to 63 years.17 With differences in tissue fixation techniques (Ho et al used snap frozen tissue) and the previously mentioned general limitation of a muscle needle biopsy, it is difficult to compare our data directly with that of the literature. It is possible that our smaller rhabdosphincter diameter is due to fixation artefact. Accepting the limitation of only 1 control, nevertheless our findings seem to argue against the hypothesis that neurophysiologically overactive urethral rhabdosphincter (detected by EMG) leads to muscle fiber hypertrophy of the urethral rhabdosphincter in women with urinary retention.13 Ultrastructural findings. Electron microscopy revealed that all patient myofibers had excessive amounts of sarcoplasmic glycogen and mitochondria, suggesting increased metabolic activity of these cells, which would be in keeping with a neurophysiologically overactive urethral sphincter.14 We cannot comment on the electrophysiological behavior of the control patient but similar ultrastructural features were not seen. Whether the sarcoplasmic accumulation of mitochondria and glycogen should be considered a pathological finding is uncertain but needs to be confirmed with more patients from a wider clinical spectrum. We did not analyze fiber type histochemically primarily because we wanted to keep the number of biopsies taken per patient in this study to a minimum. However, the majority of the striated muscle fibers showed ultrastructural characteristics of type I slow twitch fibers. Detailed fiber analysis might be valuable in future studies. CONCLUSIONS

It appears that the size of urethral rhabdosphincter fibers is not abnormal in women with urinary retention and urethral rhabdosphincter overactivity, which seems to argue against the hypothesis that the neurophysiological urethral sphincter overactivity leads to muscle fiber hypertrophy of the urethral rhabdosphincter. However, the observed ultrastructural changes (the increased amount of sarcoplasmic glycogen and mitochondria) reflecting the increased energy requirement of an overactive urethral rhabdosphincter would make physiological sense in this subgroup of women with urinary retention. Study patients donated urethral tissue for this project. Prof. F. Scaravilli’s Department of Neuropathology at the Institute of Neurology, London contributed to the study. In particular Mr. A. Beckett prepared the light microscopy slides, Mr. M. Groves helped with morphometric fiber diameter analysis and Mr. B. Young prepared the material for electron microscopy. In addition, Prof. K. D. MacRae provided statistical expert opinion. REFERENCES

1. Aboseif, S., Tamaddon, K., Chalfin, S., Freedman, S., Mourad, M. S., Chang, J. H. et al: Sacral neuromodulation in functional

2. 3. 4. 5. 6. 7. 8.

9. 10. 11.

12.

13. 14.

15.

16. 17.

18. 19. 20.

1251

urinary retention: an effective way to restore voiding. BJU Int, 90: 662, 2002 Tanagho, E. A., Miller, E. R., Lyon, R. P. and Fisher, R.: Spastic striated external sphincter and urinary tract infection in girls. Br J Urol, 43: 69, 1971 Allen, T. D.: Psychogenic urinary retention. South Med J, 65: 302, 1972 Blaivas, J. and Labib, K. B.: Acute urinary retention in women: complete urodynamic evaluation. Urology, 10: 383, 1977 Raz, S. and Smith, R.: External sphincter spasticity syndrome. J Urol, 115: 443, 1976 Williams, G. E. and Johnson, A. M.: Recurrent urinary retention due to emotional factors; report of a case. Psychosom Med, 18: 77, 1956 Chapman, A. H.: Psychogenic urinary retention in women; report of a case. Psychosom Med, 21: 119, 1959 Krane, R. J. and Siroky, M. B.: Functional voiding disorders in women. In: Clinical Neuro-Urology, 2nd ed. Edited by R. J. Krane and M. B. Siroky. Boston: Little, Brown & Co., pp. 445– 457, 1991 Hafner, R. J.: Biofeedback treatment of intermittent urinary retention. Br J Urol, 53: 125, 1981 Fowler, C. J. and Kirby, R. S.: Electromyography of urethral sphincter in women with urinary retention. Lancet, 1: 1455, 1986 Fowler, C. J. and Kirby, R. S.: Abnormal electromyographic activity (decelerating burst and complex repetitive discharges) in the striated muscle of the urethral sphincter in 5 women with persisting urinary retention. Br J Urol, 57: 67, 1985 Fowler, C. J., Christmas, T. J., Chapple, C. R., Parkhouse, H. F., Kirby, R. S. and Jacobs, H. S.: Abnormal electromyographic activity of the urethral sphincter, voiding dysfunction, and polycystic ovaries: a new syndrome? BMJ, 297: 1436, 1988 Wiseman, O. J., Swinn, M. J., Brady, C. M. and Fowler, C. J.: Maximum urethral closure pressure and sphincter volume in women with urinary retention. J Urol, 167: 1348, 2002 Landon, D. N.: Skeletal muscle-normal morphology, development and innervation. In: Skeletal Muscle Pathology. Edited by F. Mastaglia and J. Walton. Edinburgh: Churchill Livingstone, pp. 1– 87, 1982 Gosling, J., Dixon, J. and Humpherson, J.: Gross and microscopic anatomy of the urethra II. In: Functional Anatomy of the Urinary Tract. An Integrated Text and Colour Atlas. Edinburgh: Churchill Livingstone, pp. 5.1–5.20, 1982 Eisenberg, B. R. and Mobley, B. A.: Size changes in single muscle fibers during fixation and embedding. Tissue Cell, 7: 383, 1975 Ho, K. M. T., Borja, M. C., Persson, K., Brading, A. F. and Andersson, K.-E.: Expression of nitric oxide synthase immunoreactivity in the human female intramural striated urethral sphincter. J Urol, 169: 2407, 2003 Hayek, H.: Die weibliche Harnro¨hre. In: Handbuch der Urologie. II. Anatomie und Embryologie. Berlin: Springer Verlag, pp. 315–323, 1969 Schroder, H. D. and Reske-Nielsen, E.: Fiber types in the striated urethral and anal sphincters. Acta Neuropathol, 60: 278, 1983 Hale, D. S., Benson, J. T., Brubaker, L., Heidkamp, M. C. and Russell, B.: Histologic analysis of needle biopsy of urethral sphincter from women with normal and stress incontinence with comparison of electromyographic findings. Am J Obstet Gynecol, 180: 342, 1999