Vaginal Pull Through Procedure To Treat Isolated Lower One Third Vaginal Agenesis

Author Information

Saxena A*, Pardeshi S**, Gupta AS***.
(* Junior Resident, ** Assistant Professor, *** Professor, Department of Obstetrics and Gynecology, Seth GS Medical College and KEM Hospital, Mumbai, India.)


Vaginal agenesis is a rare condition. Mayer-Rokitansky-Küster-Hauser (MRKH) syndrome or Mullerian agenesis is the most common etiology which is described as congenital absence of uterus and vagina with female genotype, normal secondary sexual characteristics and ovaries. Absence or agenesis of various segments is usually uncommon. We are presenting one case of an adolescent girl who developed massive hematocolpos due to absence of the lower one third of the vagina.


Vaginal agenesis has an incidence of 1 in 4,000 to 1 in 10,000 females.[1] It may occur rarely as an isolated finding.[2, 3]. Here we are reporting one such case of vaginal agenesis, who presented with abdominal pain and an abdominal mass in our emergency receiving room and was managed surgically to drain the hematometra and create an outflow tract, while keeping her menstrual and reproductive function intact.

Case Report

A 15 years old, unmarried, student of grade VII, who had not attained menarche, came to our receiving room at 1:00 am with complaints of pain in abdomen. She was apparently alright till 3 days ago, when she had sudden pain in abdomen and her mother felt an abdominal mass. She was immediately taken to a private hospital and investigations were carried out. Amongst routine investigations, hemoglobin was 7.8 gm%, total leucocytes were 9,480/mm3, platelets were 1,93,000/mm3 and creatinine was 2.9 mg%. She was given symptomatic treatment and referred to our hospital.
On examination, per abdominally, a huge cystic mass arising from the pelvis and going up to the right hypochondrium with mild tenderness was felt. Infraumbilical, the mass was in the midline and in the upper abdomen the mass was felt more on the right side of the midline. It was fusiform to cylindrical in shape. On local examination, labia majora, labia minora and urethral and anal opening were normal but the vaginal opening was blind and flat. On per rectal examination, cystic collection was felt approximately 6 cms above the anal opening. Fecal matter was felt further. Abdominal movements of the mass were transmitted to its lower end. An impression of cervical atresia, with a functioning uterus with hematometra and hematosalpinx with renal compromise was made.
Ultrasonography of the abdomen and pelvis was suggestive of a very large cystic lesion with homogeneous internal echoes. The mass measured 21x10 cm and was seen extending from pelvis to the epigastric region. Uterine shadow was not seen separately. MRI was done which was suggestive of hematometra, hematocolpos and bilateral hydrosalpinx with blind ending cervical canal, which gave the impression of cervical atresia. Bilateral hydronephrosis and hydroureter were also probably due to extrinsic compression. Right kidney was found to be small, measuring 4.9x3.3 cms. Left kidney was of normal dimension.
Opinions of urologists and nephrologists were taken. She was catheterized and after fitness from the urologist, nephrologist and anesthetist an exploratory laparotomy with consent for possible hysterectomy and excision of the hematosalpinx was done.
Exploratory laparotomy was done under spinal and epidural anesthesia, through an infra umbilical midline incision. In vitro, large hematocolpos and hematometra were seen extending up to the right hypochondrium, just under the liver surface. Aspiration of the hematocolpos was done. After draining 100 cc of old blood, a 2-inch vertical incision was made on the anterior surface of the mass.  Almost 2 liters of blood from the hematocolpos and hematometra were drained after which the uterus was exteriorized. On inspection ovaries were normal, both tubes were stretched and had a retort shape.They were distended to  due to hematosalpinx. Uterus was about 10 weeks in size even after drainage of the old blood. An over distended vagina was seen. A bimanual digital palpation through the anterior vaginal wall incision towards the blind introitus, revealed a distance of 3 cms from the vaginal dimple.
In view of the grossly elongated vagina (empty vagina was just above the umbilicus) a decision to pull through the vagina and creation of an introitus by fixing the lower end of the vaginal mucosa to the perineal skin was made. Lithotomy position was then given and under all aseptic precautions, a cruciate incision was made below the urethral opening at the site of the vaginal dimple. Careful dissection between the bladder anteriorly and the rectum posteriorly was guided by palpating the digit of the surgeon who was at the abdominal end of the procedure. Vagina was distended abdominally with 100 cc of methylene blue to detect opening of the blind upper vagina into the dissected space. A long Kelly’s forceps from above was passed till it indented the roof of the dissected space and was felt from below. Area checked to confirm it was away from the bladder anteriorly and rectum posteriorly. Incision was taken on the tip of the Kelly’s forceps and the incision was then extended laterally. The vaginal edges were held all around and then undermined to prevent tension on the vagina. Fascia in the dissected space was incised laterally on both sides to circumvent possibility of vaginal stricture. Hemostasis in the bed was confirmed. The vaginal mucosal edges were then sutured with simple, interrupted sutures to the fascia radially with 0 no. polyglactin sutures. Fascia was then reinforced and sutured to the perineal skin from lower end of labia minora to the other labia minora. Anteriorly this was sutured to the tissue below the urethral meatus. One finger could easily be passed through the neovaginal opening. Abdominal anterior vaginotomy was sutured with simple, interrupted sutures 0 no. polyglactin sutures. Hematosalpinx was left untouched to allow natural healing. She tolerated the procedure well and got her menses after 4 days.

Figure 1. MRI. Yellow arrow shows hematocolpos which was mistaken for the atretic cervix. Yellow and green arrow is the hematometra.

Figure 2. Yellow and green arrow mark the distended, drained vagina. Blue and yellow arrow is the site of anterior colpotomy.

Figure 3. Blind Vagina

Figure 4. Dissection in the blind vagina roof to reach the upper cannalized part of the vagina.

Figure 5. Tip of the Kelly’s forceps (Yellow Arrow)

Figure 6. Anastomosed upper patent vagina to the introitus.

Daily vaginal digital examinations were done to check and maintain the patency of the vagina. After 2 weeks of the procedure, she and her mother were taught the correct usage of glass dilators and she was discharged with instructions to use the vaginal dilator daily. Weekly follow ups were advised but she was lost to follow up for almost a month.
When she finally followed up in OPD, a vaginal stricture had formed even though her mother gave history of regular use of glass dilators. On local examination, a vaginal stricture at about 3 cm from the introitus was seen. No epithelialization was seen in the 3 cm area. Opening was identified with a uterine sound and an artery forceps could be introduced. Clear fluid flowed out but tip of finger couldn’t be admitted.
She was posted for a second procedure where dilation of the stricture with radial incision and with placement of amnion graft in the lower one third of the vagina with Styrofoam vaginal mold was planned. Stricture was dilated with Hegar dilators upto no.20 till a opening of 2 cm in diameter was created. Radial incisions at 10’o clock, 2’o clock, 4’o clock and 8’o clock were taken to release the stricture and the flaps thus created were sutured anteriorly to the vestibule. Posteriorly a flap could not be raised due to lack of epithelization. An amnion graft was placed over the Styrofoam mold into the vagina so as to ensure the raw area had contact with amnion graft. Introitus was closed. Foley’s catheter was inserted in the bladder and left in situ. After a week, the mold was changed under all aseptic precautions. The soft mold was changed again after a week and replaced by a prosthetic mold which she or her mother could change as and when required. She was discharged with instructions to wear the mold every night till she becomes sexually active and to follow up as instructed. Her monthly followup shows a patent vagina. She is menstruating at regular intervals and wearing the mold at night regularly.

Figure 7. Placing vaginal mold with amnion graft (arrows).

Figure 8. Vagina closed after inserting mold.


Vaginal agenesis is the result of complete failure of the vaginal plate to canalize. Mullerian agenesis may be of varying degrees, distal to proximal vagina, cervix or uterus or near total absence of all mullerain structures with associated degrees of Mullerian aplasia.[4]
The diagnosis may be unsettling for the patient and her family. Therefore, attention has to be paid not only to the correction of the anatomical abnormality but also to the psychosocial issues.[5] Counseling about future fertility and sexual function is also extremely important. In patients with isolated vaginal agenesis with a functioning uterus and cervix, normal fertility can be expected in future. However, infertility issues may occur due to hematosalpinx, endometriosis and the surgery itself.

The timing for both nonsurgical and surgical correction of this anomaly is purely elective unless the patient is symptomatic.[5] As such in patients with intact uterus experiencing cyclic pelvic pain as a result of obstructed menstrual fluid flow, surgical correction at the earliest is mandatory at the time of diagnosis. In the literature, although more than 10 surgical procedures for neovaginal reconstruction have been described so far, an ideal approach has not yet been identified. Emran et al reconstructed the vagina using a sigmoid colon pedicle flap successfully. The proximal part of the created vagina was connected to the cervix and a Foley’s catheter was left in situ for 7 days to prevent obstruction.[6] Beksak described a procedure in a 14 year old girl with a normal female genotype and secondary sexual characteristics with complains of primary amenorrhea and cyclic pain in the pelvis. A vaginal dimple without a normal vaginal orifice was seen on local examination and an enlarged uterus was felt on bimanual recto-abdominal examination. A transverse incision at the site of vaginal dimple was taken. Blunt dissection was done to create a potential space between the bladder and urethra in front and rectum behind. The surgeon palpated the catheter in front and assistant’s finger in rectum to avoid injury while dissecting. A neovagina was obtained which was 9-10 cms long and 3-4 cms wide. in spite of this, the collection couldn’t be reached and an abdominal approach was required to drain the hematometra. An 8-french Foleys catheter was placed in its central lumen with its tip located in the uterine cavity. Its balloon was insufflated with 3 mL saline. Endometrium and myometrium were closed separately. To maintain the mold in position, it was sutured to the labia majora. Hyalobarrier gel was applied between the mold and the neovaginal walls.[7] Many a times it is not possible to drain a hematometra or hematocolpos through a vaginal approach and an abdomino-perineal approach is often required.

We primarily decided an abdominal procedure as the hematometra reported on MRI was upto the right hypochondrium and differential diagnosis was of cervical atresia or hematocolpos. Since it was an over distended vagina containing more than 2 liters of blood which we drained we realized that we would be able to pull through this stretched vagina and fix it to the introitus thus resulting in a vaginal reconstruction with normal vaginal epithelium. We also used an abdomino-perineal approach to align the upper vagina into the dissected space and its fixation to the introitus. This was successfully performed and the daily digital vaginal dilatation maintained it patent. However, though the vaginal dilatation was taught to the girl’s mother and also confirmed, the upper vagina retracted from the introital sutures. Hence in the second sitting we had to excise the constriction ring of about 1cm in width between junction of upper vagina and lower vagina and place a graft on it.
Many complications are associated with the procedure. Bowel and bladder injury are serious complications that have to be explained to the family before surgery. Vaginal molds which are used to prevent restenosis may be associated with many problems. An unsatisfactory reconstruction due to poor drainage, graft maceration, sloughing and graft detachment may occur due to non ideal vaginal molds. Using an appropriate vaginal mold is the key to achieving good results.
Although multiple options are available for allografts, amniotic membrane was used by us as an allograft in vaginal reconstruction as it is inexpensive, easily available and has low antigenicity.


Various grafts have been used to line the neovagina. Pull through of the upper vagina upto the introitus should theoretically prevent stricture formation provided the upper vagina does not retract back. However, daily wearing of the vaginal mold will still have to be continued till the girl maintains an actice sexual life.


We would like to thank Dr. S. V. Parulekar for his valuable intraoperative inputs.

  1. Evans TN, Poland ML, Boving RL. Vaginal malformations. American Journal of Obstetrics and Gynecology. 1981;141(8):910-920.
  2. Current evaluation of amenorrhea. Practice Committee of the American Society for Reproductive Medicine. Fertility and Sterility. 2006;86(5 Suppl 1):148-155.
  3. Müllerian agenesis: diagnosis, management, and treatment. ACOG Committee Opinion No. 728. American College of Obstetricians and Gynecologists. Obstet Gynecol 2018;131:e35–42.
  4. Miller RJ, Breech LL. Surgical correction of vaginal anomalies. Clin Obstet Gynecol. 2008;51(2):223-36.
  5. Laufer MR. Congenital absence of the vagina: in search of the perfect solution. When, and by what technique, should a vagina be created? Curr Opin Obstet Gynecol. 2002;14(5):441-4.
  6. Erman-Akar M, Ozkan O, Ozkan O, Yucel S, Dolay K, Ertugrul F, et al. Uterine preservation and vaginal reconstruction in a patient with congenital vaginal agenesis presenting with cyclic menouria. J Minim Invasive Gynecol. 2011;18(5):682-5.
  7. Beksac MS, Salman MC, Dogan NU. A new technique for surgical treatment of vaginal agenesis using combined abdominal-perineal approach. Case Reports in Medicine. 2011;2011:1-6.

Saxena A, Pardeshi S, Gupta AS. Vaginal Pull Through Procedure To Treat Isolated Lower One Third Vaginal Agenesis. JPGO 2019. Volume 6 No.4. Available from: