In elderly women, the urethral striated muscle sphincter or rhabdosphincter (RS) does not show a completely circular arrangement, but is located on the pubic and/or lateral side of the urethra [1, 2]. However, Perucchini et al. [3] have quantitatively demonstrated that a thin striated muscle layer remains along the vaginal side of the urethra, especially in young women. A recent detailed study using human female fetuses [4] has shown there is no striated muscle, but a connective tissue "raphe" along the vaginal side of the urethra. This embryology is absolutely consistent with Strasser et al. [1], but seems to rule out the possibility of a vaginal-sided striated muscle sphincter. However, Masumoto et al. [5] have reported that the urethral striated muscle is likely to extend to the vaginal side of the urethra when the vagina merges with the urethra at the higher position: at this stage, the vagina is in the process of descent to the future vestibulum. Thus, the embryological development of the striated sphincter in females is still unclear.

In addition, with a few exceptions, the fetal development of the urethrovaginal sphincter (which is a well developed striated muscle in children and adults [2, 6, 7]) has not been well described [8], and there remains a further possibility that the urethrovaginal sphincter extends posteriorly to surround the thin fetal vagina. In both the fetal urethral and urethrovaginal striated sphincters, we have considered that the topographical anatomy during vaginal descent makes it difficult to understand this aspect of the developmental process. Does the descending vagina disturb the development of a striated sphincter, thus preventing a completely circular arrangement, as stated by Sebe et al. [4]? To address this issue, we examined the fetal topographical anatomy of the urethral RS with the aim of reappraising the existing concept of the difference in its development between the sexes.

The present study was performed in accordance with the provisions of the Declaration of Helsinki 1995 (as revised in Edinburgh 2000). We examined the paraffin-embedded histology of 22 mid-term fetuses (11 males and 11 females). The specimens varied in size and stage: the crown-rump lengths (CRL) of the male specimens were 27, 29, 29, 30, 39, 46, 48, 52, 84, 95, and 103 mm, while those of female specimens were 36, 48, 48, 62, 76, 78, 84, 94, 92, 100, and 103 mm. These sizes corresponded to ovulational ages of approximately 8-13 weeks [9].

All the specimens were part of the large collection kept at the Embryology Institute of the Universidad Complutense, Madrid, and were the products of miscarriages and ectopic pregnancies managed at the Department of Obstetrics of the University. Approval for the study was granted by the university ethics committee (approval number, B-08/374). Because of the nature of the specimens, we were unable to rule out the possibility that they contained pathology. However, no pathology was found in the developing umbilical vessels, liver, intestine, adrenal, and kidney in the specimens examined. Identification of gender was based on observations of the urogenital fold, including the primitive uterus.

After routine procedures for paraffin-embedded histology, most of the specimens were cut almost horizontally (anterior side, tilted inferiorly) at a thickness of 5 µm and at intervals of 50 (100) µm. Sagittal sections were limited to those from male and female fetuses with a CRL of 103 mm, and were cut at a thickness of more than 10 µm because of their large size. Depending on the size of the specimen, 30-200 sections were needed for complete observation, including the entire urethra. Most sections were stained with hematoxylin and eosin, but some were stained with azan (red and blue) or azocarmine (orange).

The present study demonstrated that, in both genders, the urethral striated sphincter developed on the inferior or distal side of the site where the genital tracts merged with the urethra. The male fetal urethra changed its direction immediately inferior to the future prostatic colliculus. Thus, the usual model in which the striated muscle surrounds the urethra in the horizontal plane (Fig. 4A) does not correspond to the fetal morphology earlier than 13 weeks, or a CRL of 95 mm. Therefore, at these stages, the future membranous urethra seems to run horizontally along the antero-posterior axis, and the early male striated sphincter extends in the frontal plane (Fig. 4B). In females, by contrast, there was no distal urethra, but the future vestibule opened inferiorly on the inferior side of the vaginal inferior end. Therefore, it appeared impossible for the female urethral striated muscle to "surround" this fissure-like structure corresponding to the future vestibule. Likewise, the urethrovaginal sphincter did not appear to reach the posterior midline area because of the inferiorly opened fetal vestibule. In fact, in adults, the vagina is tightly attached to the posterior wall of the urethra [7, 10, 11]. However, we are unable to rule out the possibility that, before descent of the vagina, the urethral striated muscle extends posteriorly into the space between the vagina and the urethra.

Experiments using an avian chimera system have shown that the perineal muscles originate from myotomal cells in common with the lower extremities [12]. The urethral striated muscle likely originates from a precursor shared with the perineal muscles. In fact, the fetal urethral striated sphincter was located adjacent to the bulbospongiosus muscle, whereas in females, the vestibular bulb later separated the two. Sasaki et al. [13] and Yamaguchi et al. [14], studying mouse embryos, demonstrated an intimate topographical relationship between the bulbospongiosus and the external anal sphincter. The myogenic cells require an attachment or insertion. Mesenchymal tissues around the urethra inferior or distal to the vaginal end, as well as around the anorectum, seem to be favorable for myogenic cell migration and attachment. Conversely, the striated muscle does not provide a specific interfacial tissue for insertion to the vagina, but simply attaches to the latter [15].

In large specimens of both genders, a membranous structure was seen connecting the crus penis or clitoris to the striated sphincter. A similar membranous structure was described by Kato et al. [8] in fetuses at a much later stage than the present specimens. This might represent the earliest form of the urogenital diaphragm or the perineal membrane. Notably, the membranous structure separated the male striated sphincter from the bulbospongiosus, whereas the vestibular bulb played the same role in females. This topographical anatomy is maintained in adults in the urogenital diaphragm [16] and the perineal membrane [8].

A change in topographical anatomy is necessary for transition from the fetal morphology shown in Fig. 4B, C to the adult form: the horizontal male urethra needs to change into the membranous urethra along the supero-inferior axis, while the urethrovaginal sphincter does so along the anteroposterior axis. The developing prostate is likely to contribute significantly to this change in males. In fact, in a CRL 94-mm male fetus with a definite prostate, Oelrich [17] observed an adult-like course of the urethra as well as the striated muscle extending superiorly along the prostate. Thus, the change may occur at a stage slightly later than the present specimens. In contrast, a well-known diagram of the female perineum by Oelrich [6] was based on a 27-year-old woman, and not on observations of fetuses. However, Kato et al. [8] described the urethrovaginal sphincter at 20 weeks (CRL, >180 mm). During the hypothetical topographical change at the late stage, mechanical stress is likely to be responsible for differentiation of the membranous structure between the crus and the striated sphincter (see above) into the male urogenital diaphragm or the female perineal membrane. Subsequently, the female urethral striated muscle seems to extend posterolaterally along the superficial surface of the developed perineal membrane.