Müllerian agenesis is also known as Müllerian aplasia, vaginal agenesis, or Mayer-Rokitansky-Küster-Hauser syndrome. It is a congenital anomaly manifested by failure of Müllerian duct development. More specifically, it is characterized by the absence of reproductive system tissues derived from the Müllerian (paramesonephric) ducts, which comprise the fallopian tubes, uterus, and upper one-third of the vagina. Notably, the lower two-thirds of the vagina develop from the urogenital sinus and contribute to the development of the urethra and urinary bladder. Patients with Müllerian agenesis typically exhibit normal secondary sexual features because the ovaries are normal as they develop from the intermediate mesoderm [1,2].

Genetically, Müllerian agenesis is proposed to take place because of autosomal dominant inheritance with partial penetrance and variable expressivity, which in turn plays a key role in the intricacy implicated in pinpointing the exact underlying pathogenesis [1-3]. Clinically, Müllerian agenesis can be categorized into two types. Type 1 is characterized by the absence of structures derived from the Müllerian ducts, including the upper vagina, cervix, uterus, and fallopian tubes. In contrast, type 2 is similar to type 1, in addition to further anomalies of the body, including the skeletal and renal systems. Müllerian duct aplasia, renal agenesis, and cervicothoracic somite dysplasia is a type 2 Müllerian agenesis [4].

Vaginal agenesis is associated with several concerns for both partners, including sexual dissatisfaction, displeasurable orgasms, and distorted body image [5]. Management of vaginal agenesis can be surgical via vaginoplasty or nonsurgical dilation [2,6,7]. The goal of surgical management is to create a neovagina that enhances satisfaction with sexual intercourse [5]. Neovagina creation can be performed using various surgical methods, most commonly the McIndoe, Vecchietti, Baldwin, Williams, and Davydov techniques [7-9]. Several graft options are available for constructing a neovagina. These include grafts from the skin, small intestine, large colon, peritoneum, and the amnion [7-9]. Furthermore, heterologous biological grafts, such as tilapia fish skin, can also be used in neovaginoplasty [10,11].

Only a limited number of studies have documented experiences with vaginoplasty using amniotic membrane grafts. All of these investigations were mostly single case reports [12,13] and small-sized retrospective series [14-18]. Hence, additional research is important to bridge the literature gap, enrich the existing body of evidence, and open valuable doors for prospective, comprehensive systematic reviews and meta-analyses.

Therefore, the present study aimed to report a single-center experience of vaginoplasty using amniotic membrane grafts for the management of 10 patients with vaginal agenesis. Clinical, surgical, and functional data are presented.

The present study was a single-arm retrospective case series. This study was performed at the Department of Obstetrics and Gynecology, King Faisal Specialist Hospital and Research Center, Jeddah, Saudi Arabia. Due to the rarity of congenital conditions, no sample size calculation was performed. Instead, the study included all patients who met the following inclusion criteria: 1) patients who failed non-surgical treatment with a vaginal dilator; 2) patients who underwent McIndoe vaginoplasty using amniotic membrane graft; 3) the McIndoe vaginoplasty procedures were conducted at the Department of Obstetrics and Gynecology; 4) the procedures had been taken place from January 1, 2019 to March 31, 2023; 5) patients had confirmed clinical and radiographic diagnosis of vaginal agenesis/hypoplasia; 6) patients had a follow-up of at least 6 months postoperatively; and 7) patients had complete clinical, surgical, and functional data.

All patients underwent vaginoplasty using the McIndoe technique [19] with fresh membrane graft. Briefly, the patients were admitted the night before the procedure and administered a relevant fleet enema. Intraoperatively, the patients received epidural anesthesia combined with spinal and/or epidural anesthesia for postoperative pain control. A foley catheter was inserted into the lithotomy position. Next, a plane was dissected between the bladders, and a space was created using sharp and blunt dissections beginning from the fourchette. This step was carefully executed to avoid injury to the bladder and rectum by inserting a finger in the rectum, with careful attention to be anterior. Amniotic membrane grafts were retrieved as sterilized as possible from consenting parturient women who delivered within the last 6 hours before vaginoplasty. The grafts were repeatedly rinsed and cleaned in normal saline solution and stored in this solution until further use (Fig. 1A). The rubber mold (vaginal dilator) was covered with the prepared amniotic membrane, which was sutured around the mold after the application of a condom between the sutured amniotic membrane graft and the mold itself (Fig. 1B, 1C). A 3-0 Vicryl (Ethicon, Somerville, NJ, USA) was used as an interrupted suture before insertion of the mold in the newly generated vagina. Meticulous hemostasis was ensured, the mold was inserted, and a rectal examination was performed before inserting the mold. At this point, the mold was introduced and the vaginal introitus was sutured (obliterated using Ethibond [Ethicon] 0 or 1). Abdulrahim Gari, et al. Vaginoplasty using fresh human amnion All patients were administered an intraoperative course of intravenous antibiotics (metronidazole and ciprofloxacin), followed by an oral course of the same medications for 2 weeks postoperatively. Patients were often discharged within 5 days. On postoperative day 14, the patients were examined again under anesthesia to assess the healing process and to remove the vaginal mold for irrigation, cleaning, and reinsertion. The patients were kept overnight for education on how to self-insert the vaginal mold and were discharged the following day. The follow-up evaluation of the patients included clinical examination at 2 and 4 weeks, 2, 3, and 6 months, and yearly thereafter. The clinical assessment involved both vaginal and rectal examinations in addition to vaginoscopy. Patients were advised to refrain from sexual intercourse for 8 weeks.

Three types of data were collected: preoperative, intraoperative, and postoperative. The preoperative data included patient age, body mass index, karyotype, comorbidities, association with renal or skeletal malformations, and vaginal length before surgery using magnetic resonance imaging and clinical assessment. Intraoperative data included operative time, estimated intraoperative blood loss, and occurrence of intraoperative complications (blood transfusion, ureteric injury, blood vessel injury, urinary bladder injury, and bowel injury). Intraoperative blood loss was assessed gravimetrically. Blood was measured in suction canisters, and surgical towels were weighed before and after use to track absorbed blood loss. Postoperative data included postoperative complications (leukocytosis, fever, pelvic abscess, open wound, chronic vaginal discharge, chronic pain, wound infection, and reoperation for complications), length of the vagina after surgery measured via clinical assessment, rate of self-reported sexual satisfaction (yes/no format), and mean score for the validated Arabic version [20] of the female sexual function index (FSFI) [21]. The FSFI is a validated tool for evaluating sexual function in women across six key areas: desire, arousal, lubrication, orgasm, satisfaction, and pain. It comprises 19 questions, with responses typically ranging from 0 to 5. Higher scores within each domain indicate better sexual function, as calculated by summing the scores for the respective questions. The total FSFI score, which was derived by combining all domain scores, reached a maximum of 36, and indicating optimal sexual function across all measured aspects. It is widely used in clinical and research settings to assess the impact of various factors on female sexual health and outcomes [21]. Anatomical success was attained if the postoperative vaginal length measured clinically ≥5 cm, in addition to a width sufficient to comfortably accommodate the insertion of two fingers. Functional success was achieved when an FSFI score of >26.5 was reached FSFI [22] 6 months after the surgery. This cutoff was selected based on a validation study indicating that an FSFI total score of 26.55 (rounded to 26.6) effectively distinguished between women with and without sexual dysfunction [22].

Data were analyzed using the Social Package for Social Sciences (SPSS) version 27.0 for Windows (IBM, Armonk, NY, USA). The descriptive data were reported as numbers and percentages or means±standard deviations (SD) and ranges (minimum-maximum).

Overall, 10 patients underwent McIndoe vaginoplasty using fresh human amniotic membrane grafts. Table 1 summarizes the preoperative data are summarized in Table 1. The mean±SD age and body mass index were 30.1±7.8 years (range, 21-43) and 26.95±7.33 kg/m² (range, 18-42.4), respectively. All the patients had a normal female karyotype (46XX). One patient (10%) had hypertension. While none of the patients had associated skeletal anomalies, four patients (40%) had associated renal anomalies. The preoperative length of vagina using magnetic resonance imaging was done in eight patients only and the mean±SD was 2.63±1.36 cm (range, 1-5.5). On the other hand, the preoperative length of vagina measured clinically was done in all patients and the mean±SD was 2.15±0.85 cm (range, 1.5-4.5).

Table 2 summarizes the intraoperative data are summarized in Table 2. The mean±SD operative time and estimated blood loss were 83.5±50.31 minutes (range, 42-210) and 122.0±75.69 mL (range, 20-250), respectively. None of the patients required intraoperative blood transfusion or experienced injury to vital organs such as the ureters, blood vessels, bowels, or urinary bladder.

Table 3 summarizes the postoperative data are summarized in Table 3. Four patients (40%) developed postoperative leukocytosis within 48 hours, and one patient (10%) experienced chronic pelvic pain that lasted more than 6 months postoperatively. No major postoperative complications, such as pelvic abscess, open wound, wound infection, chronic vaginal discharge, thrombosis, or reoperation, were reported. All patients attained anatomical success and the mean±SD postoperative vaginal length measured clinically was 5.5±0.53 cm (range, 5-6).

Table 4 summarizes the results of the validated Arabic version of the FSFI values of amnion vaginoplasty patients compared to those of a historical cohort of healthy individuals [21]. The mean±SD total score of the 10 patients was 29.2±2.4, which was closely similar to an archival cohort of 131 healthy patients (30.2±6.1) [21]. In our cohort, considering the clinical cut-off of <26.5 for defining female sexual dysfunction [22], eight patients were found to be sexually functional women, whereas two patients were found to be sexually dysfunctional women. Self-reported sexual satisfaction at 7 months postoperatively was reported by eight patients (80%).

This study reports the efficacy and safety of McIndoe vaginoplasty with fresh amniotic membrane grafts in patients with vaginal agenesis. Overall, 10 patients were analyzed between January 2019 and March 2023 with a follow-up period of at least 6 months. Collectively, 100% and 80% of the patients achieved anatomical and functional success, respectively. None of the patients experienced major intra-or postoperative complications. Overall, our results highlight the rarity of vaginal agenesis and emphasize the practical viability, safety, and efficacy (defined by anatomical and functional success rates) of McIndoe vaginoplasty using fresh membrane membrane graft in patients with vaginal agenesis.

Several investigations have assessed both the functional and anatomical aspects of various surgical and nonsurgical methods of vaginoplasty in patients with vaginal agenesis. However, there is not a universally accepted “ideal” method established yet [23]. The selection of vaginoplasty can be governed by a battery of considerations, including patient readiness, inclinations, beliefs, complications, cost-effectiveness, and the surgeon’s expertise [16]. Irrespective of the surgeon’s personal choice, it is crucial to provide the patient with comprehensive information about treatment options, probable aftermath, long-term durability, and the impact on long-term sexual activity.

The complex nature of vaginal agenesis and the limited availability of therapeutic modalities present challenges in conducting well-designed prospective, randomized, double-blind controlled trials (RCTs), both for ethical and practical considerations. To date, only one RCT conducted by Cao et al. [24] has compared the laparoscopic Davydov procedure (n=21) with the Baldwin procedure (n=11). The authors favored the laparoscopic Davydov procedure based on the anatomical and functional findings, patient contentment, reduced intraoperative bleeding, and shorter operative duration [24]. During the 5-year study period, only 10 cases of vaginal agenesis were identified. Indeed, the scarcity of vaginal agenesis means that healthcare institutes often specialize in only one vaginoplasty technique, leading to non-comparative investigations that carry a high likelihood of reporting and publication bias. Moreover, the existing body of evidence is highly diverse and lacks standardization, comprehensive details, sufficient follow-up, and robust statistical analysis [9]. While no evidence-based management algorithm has been determined thus far, the Committee Opinion on Adolescent Health Care from the American College of Obstetricians and Gynecologists stated that conservative management, primarily involving the use of vaginal dilators, could be satisfactory and should be considered as the first-line option because of the comparatively high success rate and few perioperative aftermaths [25-27]. This earlier recommendation was echoed in two comprehensive review [8,9]. Nonetheless, when considering vaginoplasty as a surgical option, several approaches are available [7-9], with the McIndoe procedure being among the most frequently performed techniques [19]. Nevertheless, there is a lack of consensus regarding the material used for lining the neovaginal canal wall. Numerous studies have documented the use of various synthetic and biological materials to coat the neovaginal cavity and promote epithelialization [7-9]. Among such graft materials is the amniotic membrane graft used during vaginoplasty [12-18].

In our study, we chose a human amniotic membrane graft because of its several advantages. Such advantages comprise the lower cost and easy accessibility which can be obtained aseptically and stored in normal saline at 4°C for 2 to 3 days [13]. Moreover, it has been successfully used as a biological dressing for the treatment of various burns and wounds [28,29]. In addition, they are immunologically safe with a lower occurrence of immune rejection when implanted subcutaneously because they do not express human leukocyte antigens [30]. In addition, it exhibits antimicrobial functions by secreting lysozymes, which lower the likelihood of postoperative wound infection [31]. In addition, when implanted subcutaneously, it tends to adhere securely, guard granulation tissue, and prompt epithelialization through its intrinsic capability of undergoing metaplasia to transform into squamous epithelium [13]. In addition, it eliminates the need for a split-thickness skin graft, which can cause scarring at donor sites in some patients undergoing McIndoe vaginoplasty. Overall, these characteristics make the amniotic membrane graft stand out as an excellent graft material for constructing neovaginas.

The analysis of our series showed that all patients attained anatomical success defined by a postoperative vaginal length of ≥5 cm in addition to comfortable insertion of two fingers into the neovagina. The literature states that the cutoff for defining anatomical success is not solidly determined and ranges from 5 to 10 cm [12-18]. In contrast, our results showed that only 80% of the patients self-reported sexual satisfaction (yes/no format). In our study, functional success was determined according to the total FSFI score 6 months after surgery. Traditionally, the functional success of vaginoplasty for sexual activity can be evaluated using subjective (patient self-reported yes/no format) and/or objective (FSFI) tools [21,22]. Apart from four cases of leukocytosis and one case of chronic pelvic pain lasting more than 6 months, none of the patients in our series experienced major complications, such as injuries to internal organs, including the ureter, bowel, bladder, or vital vessels. Unfortunately, the cause of chronic pelvic pain could not be identified and may be attributed to various factors, including structural causes such as nerve damage or entrapment, inflammation, infection, adhesions, and pelvic floor dysfunction, and psychosocial factors such as stress. Our satisfactory findings for anatomical and Abdulrahim Gari, et al. Vaginoplasty using fresh human amnion functional success rates as well as perioperative safety for patients undergoing vaginoplasty using membrane graft are in agreement with relevant published literature (case reports and case series) that have been summarized elsewhere [16,32].

Rectal and local infections are the primary risk factors for rectovaginal fistulas. During dissection of the vesicorectal space, it is crucial for the surgeon to prevent contamination of the amniotic membrane graft by bacteria from the rectal flora. To facilitate this, the surgeon may insert an index finger into the rectum during rectovesical dissection, which helps to identify the anatomical boundaries more clearly and reduces the risk of inadvertent damage to the rectal mucosa [16].

The present research had several shortcomings that ought to be outlined. The small sample size represents a major shortcoming that could not be avoided owing to the rarity of the congenital condition. Additionally, the small sample size limited our ability to conduct robust comparative analyses. Due to the retrospective study design, selection bias could not be avoided. Additional shortcomings include the lack of evaluation of functional sexuality at long-term intervals (e.g., 12 months) and examination of the epithelialization of the vaginal mucosa.

In view of the limitations of our results and the existing body of literature, additional case series and RCTs, if possible, with larger sample sizes are required. An interesting investigation will examine the anatomical, functional, and complication rates of vaginoplasty in patients receiving fresh versus frozen membrane grafts. An additional prospective inquiry may include exploring the long-term outcomes of vaginoplasty patients at intervals of to 3-5 years postoperatively. Lastly, worthwhile research may be geared toward identifying a cohort of patients with vaginal agenesis (stratified according to certain clinical and radiological parameters) who are most likely to benefit from vaginoplasty using amniotic membrane grafts.

In the absence of a universally acknowledged “ideal” treatment approach and comprehensive RCTs, individuals with vaginal agenesis should be provided with unbiased and expert information regarding both surgical and nonsurgical choices. This approach enables the selection of a personalized ideal method according to patient preferences.

Overall, surgical dissection of the vesicorectal space and utilization of a fresh amniotic membrane graft to create a neovagina over a vaginal mold are technically feasible, safe, and linked to favorable anatomical and functional outcomes in patients with vaginal agenesis.