ARTICLE |
a Departments of Pediatric Surgery
b Pediatric Endocrinology, Necker Enfants-Malades Hospital, Paris V University, Paris, France
| ABSTRACT |
|---|
|
|
|---|
METHODS. From 1984 to 2007, we treated 32 patients for obstructive uterovaginal duplication in our institution. We separated them in 2 groups (prepubertal [n = 8] and pubertal [n = 24]) and analyzed their clinical and radiologic presentations and their treatments. Patients >18 years of age (n = 22) were recontacted.
RESULTS. For the prepubertal group, the median age at diagnosis was 6 months. Postnatal ultrasound showed an absent ipsilateral kidney in 6 case subjects, although 4 patients had a prenatal diagnosis of ipsilateral multicystic dysplastic kidney. This renal anomaly was associated with a pelvic sonolucent mass in 3 case subjects, allowing us to prenatally suspect the diagnosis. All of the patients were cured by vaginal approach. For the pubertal group, the median delay of diagnosis after menarche was 9 months. Among patients managed in an emergency setting (n = 11), there were 9 misdiagnoses with inappropriate abdominal surgeries, including 3 hysterectomies of the obstructed hemiuterus. Concerning long-term results, 5 patients were attempting to have children, with successful pregnancies for 4 of them. One patient suffered from infertility attributable to contralateral isthmic stenosis after a hysterectomy of the obstructed hemiuterus.
CONCLUSIONS. Obstructive uterovaginal duplication is a benign pathology when properly managed. Therefore, management of abdominal pain in peripubertal girls should include systematic abdominal and gynecologic examinations completed by radiologic pelvic and renal evaluation. Surgical treatment should be performed by vaginal approach to avoid infertility. Concerning the origin of the malformation, the high association of prenatal dysplastic kidneys and postnatal absent kidneys suggests anomalies of both wolffian and müllerian duct development, contrasting with the classic definition of this malformation.
Key Words: uterovaginal duplication renal agenesis blind hemivagina hematocolpos hematometra
Female genital tract development anomalies represent a broad spectrum of pathologies that are frequently misdiagnosed and inappropriately managed during the pediatric period. Uterovaginal duplications are described as the consequence of a failure in the fusion process of the 2 müllerian ducts around 9 weeks of gestation.1–3 The incidence of this malformation is estimated within the range of 0.1% to 3.8%.4,5 Obstructive uterovaginal duplication, also called blind hemivagina malformation, is rare and generally diagnosed after the onset of menses or, in some cases, around birth, corresponding with the 2 periods of steroid hormonal impregnation in childhood. It combines a duplex uterine cavity, 2 cervices and 2 hemivaginas, 1 of which ends blindly. This induces a retentive situation of the obstructive hemigenital tract. Obstructive uterovaginal duplications are always associated with anomalies of the ipsilateral wolffian derivates (kidney and ureter).3,4,6–8 Concerning the ureter, the spectrum generally includes absent hemitrigone at cystoscopy with or without ectopic ureter termination in the wall of the blind hemivagina.3,9
Management of this malformation is frequently done in an emergency setting. Misdiagnoses are frequent and end up in inappropriate surgical procedures with irreversible injuries on the genital tract or residual urinary incontinence if an ectopic ureter with vaginal ending is neglected. The optimum management of this benign disease should fulfill 2 major goals: suppression of the obstruction and ability of normal carriage.
The aim of this study was to review all of the patients managed for obstructive uterovaginal duplication in our institution over a 20-year period, to list and analyze the clinical presentations depending on the age at diagnosis, and to evaluate their long-term outcomes.
| PATIENTS AND METHODS |
|---|
|
|
|---|
Radiologic Examinations and Classification
Radiologic evaluation consisted mainly on abdominal-pelvic ultrasound but also pelvic computed tomography scans or MRI depending on the cases. When possible, we retrospectively classified the uterine malformation according to the radiologic evaluation. Didelphic uterus was defined as 2 completely separate hemiuteruses with 2 cervices. "Septate uterus" was defined as a uterus with a smooth, normal, external shape but divided into 2 cavities by an internal longitudinal septum and 2 cervices, in contrast with the standard definition of a septate uterus.10 Retrospective study permitted us, in some patients, to assess the size of the hematocolpos (retention of menstrual blood in the blind hemivagina) or hydrocolpos (mucous retention), depending on the age of the patient at diagnosis. If present, the size of the hematometra and hematosalpinx was also assessed.
In a few former patients of the prepubertal group with antenatal ultrasound diagnosis of cystic dysplastic kidney, we performed an intravenous pyelogram to assess the remaining function of the ipsilateral kidney. In recent case subjects, this was assessed by scintigraphy.
Surgical Treatment
Surgical management consisted mainly of a complete resection of the vaginal septum through an intravaginal approach. Nowadays we use a harmonic scalpel for septum division. Sometimes septum division has been preceded by a vaginal drainage as an emergency procedure. For a few patients, the first procedure was performed by using an abdominal approach (unilateral supracervical hysterectomy or hysterovaginectomy) completed by using a vaginal approach for the septum division. Preoperative complementary evaluation has frequently included a vaginoscopy, which provided an evaluation of both cervices and vaginal walls.
Concerning the management of renal anomalies, few cystoscopies (n = 4) have been performed to explore the trigone area and to confirm the lack of ipsilateral hemitrigone. Nephrectomies, when needed, were performed through a suprapubic transversal incision.
| RESULTS |
|---|
|
|
|---|
|
A 1-stage vaginal approach with septum resection was performed for 7 patients. One has still not been operated on, because the vulvar mass disappeared rapidly after birth. Retention was a hydrocolpos in all 7 of the patients, with 1 hydropyocolpos (case 7). We performed 4 ipsilateral nephrectomies in patients for whom we discovered an ectopic ureter ending in the blind hemivagina. Among these patients, 3 had an antenatal diagnosis of ipsilateral cystic dysplastic kidney, and it was always a nonfunctioning kidney (at intravenous pyelogram or scintigraphy). Nephrectomy was performed during the same anesthesia in 3 patients and 13 months later with concomitant contralateral antireflux procedure in 1 case (case 2). The median delay between assessment of diagnosis and surgical treatment was 20 days (range: 0–287 days). The longest one corresponds with a girl (case 1) who had an antenatal diagnosis of a small pelvic cystic mass with unilateral cystic dysplastic kidney, but at birth there was no vulvar mass; during follow-up, the suspected hydrocolpos increased until becoming a vulvar mass.
Pubertal Group
Age at diagnosis in the pubertal group (n = 24) spread from 11 to 17 years, with a median age of 14 years. All of these girls had menses, and the median delay of diagnosis after menarche was 9 months (range: 1–48 months). Clinical presentations are summarized in Table 2.
|
|
Overall, we have to deplore 5 hysterectomies (20.8%) in this pubertal group (1 of them, who had the operation elsewhere, unfortunately concerned the contralateral normal uterus). The median number of surgical procedures per patient was 2 for the group managed in an emergency situation compared with 1 for the remaining group.
Blind hemivagina was right and left sided, respectively, in 13 (54%) and 11 (46%) patients. The uterus was didelphic in 19 patients (79%), septate in 3 patients, and undefined in 2 patients. In all of the patients we found a hematocolpos. Hematometra was present in 13 (54%) patients and hematosalpinx in 3 (12.5%) patients. In these 3 patients, median time between menarche and diagnosis was significantly higher than in the rest of the group (24 months), explaining the tubal, uterine, and vaginal retention. The median size of the blind hemivagina was 10.0 cm (range: 6.0–16.0 cm). Ipsilateral and contralateral uterus sizes were, respectively, 8.0 cm (range: 6.0–8.0 cm) and 4.5 cm (range: 2.8–6.9 cm).
Concerning urologic anomalies, the results are summarized on Table 4. In 2 patients we found, by vaginal approach, an ipsilateral ureter ending in the blind hemivagina. In 1 case, the patient had a prenatal history of cystic dysplastic kidney; we performed, in this case, a complementary nephroureterectomy. For the other patient, because the ureter was blind, we did no further procedure. Two patients had both of their kidneys noted by standard ultrasound; one of them had a functioning ipsilateral renal dysplasia with normal contralateral kidney, and the other one had an ectodermal dysplasia syndrome with bilateral renal dysplasia that ultimately required transplantation. In these 2 patients there were no anomalies on ureter ending.
|
Long-term Outcomes
The current median age of the patients was 23 years old (range: 3–36 years) at the time of this study, and the median time since surgery was 11 years. Among the 22 patients >18 years of age recontacted by a questionnaire (2 in the infant group and 20 in the pubertal group), only 11 answered.
There was no recurrence of obstruction, and, concerning sexuality, the results were good, with painless intercourses in all of the patients. Regarding fertility, 6 successful pregnancies were reported by 4 patients, with 3 within the former obstructed hemiuterus. These patients have had a vagina septum resection associated, in 1 case, with an ipsilateral supracervical hysterectomy. Delivery was achieved by cesarean section in 4 patients. One patient, who had a supracervical hysterectomy, complained of infertility. Gynecologic investigations concluded to primary sterility because of an isthmic stenosis. The remaining patients had not had a child project yet.
| DISCUSSION |
|---|
|
|
|---|
Neonatal or even prenatal cases are rare in the literature19 but represent
25% of our series. Prenatal diagnosis can be made as early as the second trimester, but, because of the small retention at this term, it is generally performed during the third trimester. After birth, diagnosis is evoked in front of a soft translucent vulvar mass. The differential diagnosis that could be discussed with these clinical features is an imperforate hymen with hydrocolpos. Because it is difficult to confirm the vaginal duplication by routine physical examination at this age, attentive vulvar examination should focus on the hymen membrane, which is pushed forward but laterally by the blind hemivagina. Moreover, an ultrasound scan confirming an absent or dysplastic kidney would be a further argument for the diagnosis. In the other hand, antenatal or postnatal findings of a dysplastic cystic or absent kidney with or without sonolucent pelvic mass should alert on the risk of further genital anomaly in female newborns.
Retention in the blind hemivagina at this age can be explained by the estrogenic impregnation of the fetus and the newborn. If the diagnosis is not evident during this period (small retention), the hydrocolpos will decrease as hormonal impregnation on the uterine cervix decreases, and the diagnosis will be delayed until puberty, when menses retention shall appear. However, diagnosis can sometimes be made during childhood, before puberty, like for the 6-year-old girl in our series. For this patient, whose complaints were chronic abdominal pains, we found an ectopic ureter ending in the blind hemivagina with a renal remnant. One can wonder whether the hydrocolpos was not because of a small but persistent urinary secretion (urocolpos). This configuration could explain the chronic retention out of any hormonal impregnation. Therefore, careful examination of the vaginal wall should be performed, searching for ectopic ureter ending. The discovery of this ectopic insertion is not easy during vaginal examination (an injection of a contrast agent before resection could be helpful). If the ureter is permeable, we recommend performing an associated nephroureterectomy to avoid postoperative urinary incontinence after resection of the septum by vaginal approach.
We have found, in this prepubertal group, 4 patients with an ectopic ureter ending. In all of the patients we have found a small, mostly pelvic, renal remnant, although preoperative radiologic examinations had failed to show it in 3 patients. These findings suggest that the classical definition of this malformation, including an agenesis of the ipsilateral kidney,7,20,21 is arguable. This "agenesis" has always been defined radiologically and mostly by ultrasound or intravenous pyelogram in the literature. With regard to the small size of these remnants and their frequent ectopic location in our series, we could wonder whether we do not reach the limit of imaging diagnosis, as suggested by Prada Arias et al.22 These 4 early cases more likely suggest a mechanism of dysplasia followed by nonfunctional hypoplasia than a pure agenesis. This dysplasia could be explained by the obstruction caused by the ureteral anomaly (either fibrotic ureter or ectopic situation). Additional questions concern the other patients without any ectopic ureter discovered during vaginal surgery. Cystoscopies, when performed, revealed an absent ipsilateral hemitrigone, and imaging examinations found an absent ipsilateral kidney. This absent hemitrigone associated with an absent ipsilateral kidney does not permit us to conclude whether it is a real agenesis or a hypoplasia. Nevertheless, regarding those results, one could imagine an early antenatal ureteral atresia because of major obstruction, inducing a dysplasia of the metanephric blastema and early renal hypoplasia. Classic descriptions of blind hemivagina malformation suggest that the primum movens in this malformation involves an anomaly of fusion of the 2 lateral müllerian ducts. If this malformation was only the result of a müllerian misdevelopment, it is difficult to understand the frequent ectopic ureteral situation encountered in our patients. These data rather suggest a sequence including both a unilateral wolffian and müllerian misdevelopment, which means a misdevelopment from earlier than the ninth week of gestation. Regarding the chronology of the different embryologic events, it could be argued that the primum movens in the embryogenesis of this malformation is a unilateral anomaly of incorporation of the distal part of the wolffian duct within the presumptive bladder. Because of close relations between wolffian and müllerian ducts in this area, this may lead to anomalies of the unilateral müllerian duct, leading to anomalies in fusion processes of the both müllerian ducts. This hypothesis has been suggested previously by Gilsanz et al,3 and we think that our results support these data. Moreover, we think that this wolffian anomaly of incorporation also induces an anomaly in the incorporation of the ureteric bud within the presumptive bladder (ectopic vaginal ureteral ending). This situation leads to a renal misdevelopment (but not agenesis, because the metanephric blastema is induced by the ureteric bud before it incorporates in the posterior wall of the presumptive bladder) and dysplasia, for which severity depends on the obstruction at the ureter insertion.
Concerning hemivaginal obstruction, the failure of fusion of the both müllerian ducts could induce that the malformative duct would not reach the sinusal tubercle of the urogenital sinus. Thus, the resorption of the vaginal plate would not take place during the second trimester, and the blind hemivagina would be of müllerian origin without the normally distal part coming from the urogenital sinus, whereas the normal contralateral hemivagina would be complete. This mechanism could also explain the clinical fact that the blind hemivagina, when moderately dilated, is found as a bulging mass high in the normal hemivaginal sidewall. Another hypothesis to explain the malformation has been proposed by Acién et al23,24 in their work on the embryology of the female genital tract. They hypothesize that, contrary to the classical description, the entire vagina originates from the fused wolffian ducts. Thus, in obstructive uterovaginal duplications, the blind hemivagina is attributable to the failure of 1 wolffian duct to reach the urogenital sinus. This interesting hypothesis probably needs to be put to the test of the different pediatric situations encountered, particularly in the cases of disorders of sexual differentiation.
Among the 11 patients admitted to the emergency department, we deplore 9 inappropriate laparotomies for erroneous diagnoses, followed by 7 incomplete cures of blind hemivagina and 2 unworthy hysterectomies, with 1 concerning the normal contralateral hemiuterus. When managing acute abdomen in an emergency department at this age, appendicitis and adnexal torsion are, indeed, the main causes evoked and diagnosed. Regarding our results, it seems that systematic abdominal and gynecologic examinations completed by radiologic evaluation with a special focus on renal region should be a part of the management of acute pains in peripubertal girls. Concerning radiologic evaluation, even if statistics in our series are not in favor of an ultrasound scan in an emergency, it allows, if it is specifically asked for by the radiologist, a simple examination of both the lumbar renal region and pelvic region, for arguments for the diagnosis even if hematocolpos of a small size are not easy to display by simple pelvic ultrasound scan. In these cases, and, moreover, if ultrasound reveals an absent kidney, MRI is the accurate examination for diagnosis,25 and we must avoid unnecessary exposure to radiation.
Concerning the other clinical signs seen after menarche, 2-step menstruations of red and black blood, which suggest a communicating blind hemivagina, seem to be very specific; indeed, uterine adenomyosis or endometrial tumor are rare at this age. For 5 patients in this series, the only reason for referral was this sign. When the opening is large enough, patients can feel no pain and just complain of this particular sign. These openings can sit at a uterine or vaginal level. Because the flow is not completely interrupted in these patients, the median delay for diagnosis is longer. Another 2 patients had a small communication between the 2 hemivaginas discovered during surgery. They presented for one with an acute abdomen with fever and for the other with a chronic vaginal suppuration resisting local treatment. Vaginal communications in these 2 patients were too thin to permit efficient blood flow but permeable enough for bacterial colonization to develop.
Optimum surgical management of this malformation is through a vaginal approach with complete resection of the septum.6,20,21,26 There is no need to specifically drain the hematometra or hematosalpinx if present, because after a wide opening of the obstructed vagina, they will naturally drain. An ultrasound scan performed 1 month after surgery will show the progressive decrease in the size of the uterine cavity and recovery to a normal size after 3 months. Except in extremely rare presentations with fever and peritonitis by retention infection, the surgical cure is not an emergency. Efficient analgesic medication permits us generally to alleviate symptoms. In patients where surgery has to be delayed, it is also possible to stop menstruation by continuous estroprogestative treatment. The hematocolpos will progressively decrease by blood lysis. Because surgery of this malformation is easier when the obstructed vagina is dilated, estroprogestative pills should be stopped
3 months before surgery. Multiple procedures increase the risk of ascending infection with the well-known risks on fertility; therefore, 1-stage septum resection by vaginal approach is all that is warranted. Laparotomy is not required in the management of these patients,6,21 even when severe endometritis or salpingitis is present. The number of infertilities after ipsilateral hysterectomies in our series and in others21,26 supports this fact. Moreover, the ability to achieve a normal carriage in the former obstructed hemitract21 confirms that obstructive uterovaginal duplication is a benign malformation, and when these patients are properly managed, long-term results are good.
| FOOTNOTES |
|---|
Address correspondence to Claire Nihoul-Fékété, MD, PhD, Hôpital Necker Enfants-Malades, Service de Chirurgie Pédiatrique, 149 Rue de Sèvres, 75743 Paris Cedex 15, France. E-mail: claire.fekete{at}nck.aphp.fr
The authors have indicated they have no financial relationships relevant to this article to disclose.
| What's Known on This Subject Obstructive uterovaginal duplication is a very rare malformation consisting of a blind hemigenital tract associated with ipsilateral renal anomaly.
|
| What This Study Adds We discuss a large series of 32 cases with long-term follow-up along with the embryologic origin of the malformation.
|
| REFERENCES |
|---|
|
|
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||