OBJECTIVE: To describe analytically the anatomic variety and laparoscopic findings observed in patients with Rokitansky syndrome throughout an 11-year span.
METHODS: We analyzed the laparoscopic and chart records of 106 consecutive patients who underwent surgery for the creation of a neovagina, according to the modified laparoscopic Vecchietti procedure.
RESULTS: A hypoplastic vagina was observed in 61 women. Müllerian remnants, laterally displaced in the pelvis, were identified in 92 cases; 42 (25.9%, 95% confidence interval [CI] 19.2–33.3) of 162 müllerian remnants were cavitated and contained endometrial mucosa. Ovaries were extrapelvic in 17 (16.0%, 95% CI 9.6–24.4) cases. Anomalies of the urinary tract were identified in 32 (30.2%, 95% CI 21.7–39.9) patients; unilateral renal agenesis was the most frequent finding (18 cases; 18%, 95%CI 10.4–25.5).
CONCLUSION: Rokitansky syndrome has a wide variability of anatomic presentations, and, as the do other congenital anomalies of the female genital tract, it represents a continuum of embryonic malformations, which occur at different stages of development.
LEVEL OF EVIDENCE: III
The Mayer-Rokitansky-Küster-Hauser syndrome is defined as the agenesis of the uterus and vagina (class Ie of the American Fertility Society),1 with normal ovarian function and development of secondary sexual characteristics. Although this malformation has been partially described by Mayer in 1829,2 by Rokitansky in 1838,3 and subsequently by Küster in the first review in 1910,4 it was only in 1961 that it obtained its current definition uterus bipartitus solidus rudimentarius cum vagina solida.5 However, such definition does not specify the extreme variability of the anatomic characteristics of this syndrome. Indeed, there may be complete absence of the vagina, although a retrohymenal fovea of 1–2 cm is frequently present. The variability of the anatomic structures of the upper genital tract is even greater. In 1910 Küster4 described as typical findings of this syndrome the presence, instead of a uterus, of a small bipartite uterus and normal ovaries. In fact, as suggested by Rock,1 the anatomic setting of the upper genital tract is much broader, in which the müllerian remnants can be represented by bilateral solid strands of muscular tissue located on the lateral pelvic walls, or conversely, by bilateral rudimentary uterine bulbs containing a cavity lined with functioning endometrial tissue. The tubes are described as hypoplastic, whereas the ovaries are reported to be normal in the majority of cases.
Although we have performed elective treatment of the Rokitansky syndrome since 1993 by means of the modified Vecchietti laparoscopic procedure,6 we have recorded all anatomic details observed during laparoscopy of all operated patients. To better define the anatomic variability, especially of the upper genital tract, in the Rokitansky syndrome, we have analyzed the records of 106 patients who had undergone a laparoscopic exploration of the pelvis during the surgical creation of a neovagina according to the Vecchietti modified approach.6
MATERIALS AND METHODS
The patients assessed in this study were 106 women with Rokitansky syndrome operated on for the creation of a neovagina by the first author (L.F.), from 1993 to 2004. Patients were referred to us by other Italian hospitals, because our Institute is a primary care center for Rokitansky syndrome. The study was exempt from approval from the local institutional review board according to the ethics guidelines at our institution.
All women had primary amenorrhea. The clinical criteria for the diagnosis of Rokitansky syndrome were normal external genitalia, normal axillary and pubic hair, absence of at least the superior two thirds of the vagina and of the median uterine structures, in a woman with a normal female karyotype. All patients underwent a pelvic transabdominal and transrectal ultrasound scan, karyotyping, and a urinary tract ultrasonogram. In all cases we also performed a magnetic resonance imaging scan to evaluate the volume and possible cavitation of the uterine remnants.
All patients underwent creation of a neovagina according to the modified laparoscopic Vecchietti procedure.6 Such approach systematically involved an intraumbilical entrance as well as the insertion of two suprapubic trocars, through which the traction threads of the Vecchietti device were passed. The endoscopic examination of the pelvis included assessment of the possible presence and characteristics of uterine remnants, and the presence or absence of fibromuscular tissue, rather than a simple peritoneal fold, was reported between the bladder and rectum. When present, the uterine remnants were evaluated with regard to morphology, size by means of a graduated endoscopic probe, location, and relation to the pelvic inlet and internal orifice of the inguinal canal. In all uterine remnants we identified the uteroovarian and round ligament and the tubal insertion point. In addition, the tubes were assessed for location, relationship to the uterine remnant, as well as the length, presence, and characteristics of the infundibulum. The ovaries were evaluated for location, size, and functional characteristics (ie, the presence of folliculogenic structures). Both internal inguinal orifices were observed for the presence of a hernial defect. Finally, retroperitoneal masses representing pelvic displacement of the kidneys were investigated. Confidence intervals for estimated proportions were calculated.
The subjects were all white, and the mean age in this case series was 17.2 years (range 15–33 years). In all cases the laparoscopic inspection was adequate and easily achieved, allowing the assessment of all the characteristics under study. In four cases in which an endometrial cavity was present in the rudimentary uterine horns and the patient complained of cyclic pelvic pain, the horns were removed.
A vaginal fovea 1 cm or more depth was observed in 61 patients (57.5%). The maximal length reported was 2.5 cm in four cases (Table 1). When present, the vagina was represented in most cases by no more than a shallow invagination of the vestibular mucosa at the normal site of the vaginal orifice. In a minority of cases it was constituted by an introitus with a hymen followed by a blind pouch in which the mucosa had folds and was lined by an iodine-positive epithelium using the Schiller's test.7
In all cases the uterus was absent. A müllerian remnant of at least 1 cm diameter was seen in 92 patients (86.8%). The remnant was unilateral in 22 cases and bilateral in 70. The maximal size observed was 4.5×2.5 cm. Only 12 remnants in seven patients had a maximal diameter of 4 cm or more. The smaller müllerian remnants were club-shaped, whereas the larger ones appeared uterus-like. The round and uteroovarian ligaments as well as the tubes were always present, with a size that appeared inversely proportional to the uterine rudiment's volume. The former structures were always inserted into the lateral side of the rudiment, whereas its medial side tapered in a medial direction. From the combined assessment of ultrasound and magnetic resonance imaging scans, the smaller uterine horns were always noncavitated whereas 42 (25.9%, 95% confidence interval [CI] 19.2–33.3) of the 162 müllerian remnants were cavitated and contained endometrial mucosa. Of the 12 horns of larger size, 10 in six patients clearly appeared cavitated with the presence inside of hypoechogenic material (ie, liquid mass or hematometra). In close contiguity to the uterine rudiments, and of variable size and length, could be clearly identified the round and uteroovarian ligaments (Fig. 1).
A fibromuscular streak located at the vesicorectal fold was observed in 28% of cases, whereas a simple peritoneal fold was found in 78% (Table 1). The fibromuscular streak was never thicker than 1 cm. The connecting strand is generally thought to represent the anlage of a rudimentary cervix.8
A tubal structure (ie, at least a rudimentary infundibulum) was always observed when a müllerian remnant was present. In cases where a müllerian remnant was absent, a homolateral tubal rudiment was never observed. The tubes of the larger uterine horns showed fimbriae, an infundibulum, and part of the ampulla with its lumen. In five cavitated uterine horns, the cavity communicated with the tubal lumen and peritoneum as demonstrated by the transmural injection of indigo carmine.
The ovaries were absent unilaterally in three patients. In all other cases the ovaries were bilaterally present, although in 6 cases there was a marked unilateral hypoplasia (Table 1). In 17 patients the ovaries had an extrapelvic location (bilateral in six cases, unilateral in 11). All stages of follicular development could be observed on the ovarian surface. These findings confirm that the ovaries seem to be comparable to the gonads of normal women of reproductive age. In one case was identified a dermoid cyst of the right ovary that had undergone torsion, with the partial involvement of the rudimentary uterine horn.
An evident hiatus of the internal inguinal ring was observed unilaterally or bilaterally in 22 (20.7%, 95% CI 13.5–29.7) patients. In no cases were herniated organs identified. In 18 cases the magnetic resonance imaging showed in the retrovesicorectal space tiny fibrous cavitated structures with a diameter of a few millimeters that were interpreted as a possible remnant of Gartner's duct.
Overall, 32 cases presenting with anomalies of the urinary tract were identified (30.2%, 95% CI 21.7–39.9) (Table 1). In cases of unilateral renal agenesis, the residual kidney was usually located in the pelvis (16 of 18). We observed that the presence of a single pelvic kidney was clearly associated with an extrapelvic dislocation of the ovaries (12 of 16 cases). In nine cases both ovaries were extrapelvic.
Our study presents a detailed description of the genital anatomic variants in Rokitansky syndrome, and their frequency, in a large and uniform series. This case series represents a group of patients observed in a relatively short 11-year span, from the same research group and assessed with the same diagnostic approaches (laparoscopy, ultrasound scan, and magnetic resonance imaging) for the whole study period. The past literature has plenty of contributions, including original ones, which have tried to identify the anatomic characteristics of the Rokitansky syndrome. The results of such studies are inadequate due to the poor number and clear limitation of the diagnostic examinations. In 1977 Schmid-Tannwald and Hauser,9 in 10 patients with Rokitansky syndrome described the presence of anomalies of the internal genital organs differing from the typical form of the syndrome (absence of one of the uterine rudimentary horns, anomalies of the tubes and ovaries). These cases were thus defined as “atypical” presentations of the Rokitansky syndrome. Rock1 first had described the presence of cavitated uterine remnants with potentially functioning endometrium, but reported an anecdotal literature and no systematic study. More recent literature is instead substantially focused on new therapeutic ideas. Creatsas in 200110 describes the results of his technique, providing a description of the anatomic aspects of the operated patients, which is superficial and nonanalytical. Recently, only two studies assessed analytically the anatomic characteristics of Rokitansky syndrome. Strubbe and colleagues11 make a distinction between the two types of the syndrome. Type A is characterized by a perfect symmetry of the uterine buds and the absence of tubal anomalies. Type B instead is characterized by an asymmetry of the uterine buds along with tubal anomalies. These authors regard urinary tract anomalies (found in 37% of cases in this series) as well as ovarian abnormalities (found in 15% of cases) to be present only in type B of the syndrome.
Oppelt et al12 instead presented a study that attempted to analyze the anatomic problems of Rokitansky syndrome. However, the study has several important drawbacks. The case series is heterogeneous because it results from the experience of multiple centers and lacks a uniform diagnostic approach. In fact it is a retrospective study that collected clinical information by correspondence with physicians regarding patients evaluated over a long time period at different institutes. In addition, such a case series is not particularly large, and several data are missing for some patients.
Among the various observations and in agreement with Rock,1 our study shows that in patients with Rokitansky syndrome, the rudimentary horns are frequently cavitated. In the majority of cases there is no evidence of endometrial functioning (hematometra), and in fact only a small number of cases experienced cyclic pain symptoms. Such finding warrants a systematic search of this anatomic characteristic to provide, before corrective surgery, an adequate counseling on the therapeutic possibilities, ie, the removal of cavitated and symptomatic uterine remnants or the attempt to anastomose these to the vaginal vestibulum. The latter procedure has been described by our research group in a recent report.13
The presence of such a large number of cavitated uterine remnants is in contradistinction to the classical anatomic definition of the Rokitansky syndrome, by which the uterus is absent or bipartitus solidus.4 The anatomic elements that represent the common denominator of the syndrome seem rather to be agenesis of the cervix and vagina and the absence of functioning median uterine structures. As a consequence, cases of cervicovaginal agenesis with a cavitated hemiuterus, which now have been difficult to classify, would be included in this syndrome.
Our study confirms prior findings10 concerning the ovaries, which are absent or hypoplastic in extremely rare cases, while often displaced to an extrapelvic localization. As formerly suggested by Strubbe et al,11 no strict relationship was seen between the presence of symmetrical uterine rudimentary horns and the absence of ovarian and urinary tract anomalies. Our results instead show a distinct association between the extrapelvic displacement of the ovaries and the pelvic localization of a single kidney. Even in the absence of a clear median fibrotic thickening, the posterior limit of the bladder is unequivocally defined by the presence of a thin peritoneal fold that easily identifies the vesicorectal border. The peritoneal fold arises in the frontal plane from the dorsal wall of the bladder. Behind this fold lie the pouch of Douglas and the rectum; it is considered as the precursor of the broad ligaments.3 The extremely frequent finding of defects in the internal inguinal rings suggests that the possibility of an endoscopic correction concomitant to the laparoscopic neovagina creation should be discussed with the patient in advance. In conclusion, the Rokitansky syndrome has a wide variability of anatomic presentations, and it represents a continuum of embryonic malformations that occur at various stages of embryologic development.
1. Rock JA, Breech LL. Surgery for anomalies of the Mullerian ducts. In: Rock JA, Jones HW, editors. Te Linde's operative gynecology. Philadelphia (PA): Lippincott Williams & Wilkins; 2003. p. 705–52.
2. Mayer CA. Uber verdoppelungen des uterus and ihre arten, nebst bemerkungen uber hasenscharte und wolfsrachen. J Chir Augen 1829;13:525–64.
3. Rokitansky KF. Uber die sogenannten Verdoppelungen des Uterus. Med Jahrb Osterr Staat 1838;26:39–77.
4. Kuester H. Uterus bipartitus solidus rudimentarius cum vagina solida. Z Geburtshilfe Gynakol 1910;67:692–718.
5. Hauser GA, Schreiner WE. Mayer-Rokitansky-Kuester syndrome. Rudimentary solid bipartite uterus with solid vagina [in German]. Schweiz Med Wochenschr 1961;91:381–4.
6. Fedele L, Bianchi S, Zanconato G, Raffaelli R. Laparoscopic creation of a neovagina in patients with Rokitansky syndrome: analysis of 52 cases. Fertil Steril 2000;74:384–9.
7. Fedele L, Bianchi S, Berlanda N, Fontana E, Raffaelli R, Bulfoni A, et al. Neovaginal mucosa after Vecchietti's laparoscopic operation for Rokitansky syndrome: structural and ultrastructural study. Am J Obstet Gynecol 2006;195:56–61.
8. Ludwig KS. The Mayer-Rokitansky-Küster syndrome. An analysis of its morphology and embryology. Part I: Morphology. Arch Gynecol Obstet 1998;262:1–26.
9. Schmid-Tannwald I, Hauser GA. Atypical forms of the Mayer-Rokitansky-Kuster-syndrome (author's transl) [in German]. Geburtshilfe Frauenheilkd 1977;37:386–92.
10. Creatsas G, Deligeoroglou E, Makrakis E, Kontoravdis A, Papadimitriou L. Creation of a neovagina following Williams vaginoplasty and the Creatsas modification in 111 patients with Mayer-Rokitansky-Kuster-Hauser syndrome. Fertil Steril 2001;76:1036–40.
11. Strubbe E, Willemsen W, Lemmens JA, Thinjin CJP, Rolland R. Mayer-Rokitansky-Küster-Hauser syndrome: distinction between two forms based on excretory urographic, sonographic, and laparoscopic findings. Am J Roentgenol 1993;160:331–4.
12. Oppelt P, Renner SP, Kellermann A, Brucker S, Hauser GA, Ludwig KS, et al. Clinical aspects of Mayer-Rokitansky-Küster-Hauser syndrome: recommendations for clinical diagnosis and staging. Hum Reprod 2006;21:792–7.
13. Fedele L, Bianchi S, Berlanda N, Bulfoni A, Fontana E. Laparoscopic creation of a neovagina and recovery of menstrual function in a patient with Rokitansky syndrome: a case report. Hum Reprod 2006;21:3287–9.