Robot-assisted laparoscopic colpectomy in female-to-male transgender patients; technique and outcomes of a prospective cohort study

We performed a single-center prospective cohort study in 36 female-to-male transgender patients. Patients underwent the procedure as described below after informed consent was obtained. The study protocol (2016.224) has been approved by the Medical Ethics Review Committee of VU University Medical Centre (OHRP number IRB00002991). The FWA number assigned to VU University Medical Center is FWA00017598.

Patients were included between July 2011 and June 2015. One gynecologist performed all procedures with extensive experience in robotic surgery including robot-assisted hysterectomies and oncological surgery (JT).

With the introduction of the uterine mobilizer (V-Care^® Uterine Manipulator, ConMed, Utica, USA) vaginally, a suture is placed in the vaginal epithelium 10 mm proximal of the urethra as a bordering landmark for the colpectomy (see Fig. 1A, B). Vaginal epithelium 15 mm around the ostium urethrae is not removed as this tissue will be used during urethra elongation to create a second protective layer covering the urethral anastomosis. A rectum cannula is placed in order to identify the rectum during the colpectomy.

A small incision was made approximately 20 cm above the pubic bone for insertion of a Veress needle to create a pneumoperitoneum up to 20 mmHg. A total of 5 trocars were used (4 for the robot arms and 1 for the assistant). The da Vinci^® surgical system (Intuitive Surgical, Inc., Sunnyvale, CA, USA) was positioned to the patient’s right side, with a scrub nurse or assistant to the left side of the patient. The robotic instruments used were the monopolar scissors, bipolar fenestrated forceps and the grasping forceps. During surgery, pneumoperitoneum was kept at 8–10 mmHg. The patient was now placed in Trendelenburg position for optimal visualization of the operating area.

A robot-assisted TLH and BSO are performed according to a standardized protocol [17, 18]. The ureters are lateralized and dissected up to the vesicoureteric junction in preparation of the colpectomy. The uterus and adnexa are removed through the vagina. Maintenance of pneumoperitoneum is accomplished by introducing a glove-covered gauze in the distal part of the vagina at the rim of the introitus. Hereafter, the robot-assisted colpectomy is performed.

Anteriorly, the epithelium is dissected carefully with monopolar scissors to ensure the coagulation of the well-defined vasculature of the vaginal wall. Dissection is completed to approximately one centimeter proximal of the urethra (marked by the suture, see Fig. 1B) and posteriorly up to the level of the posterior commissure. The epithelium removed is as thin as possible to prevent nerve injury to adjacent structures, to prevent fistula to bladder, urethra or rectum, and to prevent bleeding from the perivaginal plexus. It is necessary to dissect the ureters up to the vesicoureteric junction in order to visualize the ureters during removal of proximal vaginal epithelium. After removal of the entire vaginal wall, the vaginal apex is sutured laparoscopically by suturing the remnants of the rectovaginal septum and endopelvic fascia of the vesicovaginal space together. During this step, it is crucial to carefully pay attention to the route of both ureters to avoid lesions or kinking. After hemostasis, the instruments are removed and the abdominal incisions closed. Finally, any residual epithelium at the level of the introitus is removed vaginally and the bulbocavernosus muscles are approximated with a few sutures to narrow the introitus. As a standard procedure, an absorbable hemostat is applied to the surgical field to prevent diffuse bleeding. The introitus is not closed, as access for the urethral anastomosis at a later stage is required.

Postoperative care was congruent with our standard vaginal colpectomy protocol consisting of bed rest during the first day after surgery and bladder training at postoperative day 2. Bladder training consisted of clamping the transurethral catheter twice for 3–4 h. When the urge to urinate is present both times, the catheter was removed. Spontaneous micturition must occur within 3–4 h. Postvoid residual urinary volume (PVR) was tested ultrasonographically after spontaneous micturition, and PVR must be <100 mL repeatedly (twice). Patients with a PVR of more than 100 mL were treated with an indwelling urinary catheter for 24 h. After successful bladder training and voiding, patients were discharged.

Herein, we have established that RaLC is a feasible procedure in FtM transgender patients; a clear learning curve could be observed concerning surgery time and blood loss. In this series of patients, we have encountered one major postoperative complication (1 patient with postoperative bleeding) and 7 minor postoperative complications (5 patients with urine voiding problems, 1 patient with an UTI and 1 patient with urine voiding problems and an UTI). All 8 complications were successfully treated or resolved with conservative treatment.

Although this study has its limitations, such as a small number of patients and a relatively short follow-up period, the study we present shows the safety, feasibility and efficacy of the robot-assisted colpectomy combined with robot TLH/BSO as a single-step surgical procedure.

Consistent with the IDEAL criteria (Idea, Development, Exploration, Assessment and Long-term study) for the evaluation and implementation of an innovational surgical procedure, this is a prospective development study with an initial small group of patients, where a clear description of the surgery is outlined, without omissions and including learning curves as well. As these characteristics represent the stage 2a ‘Development’ based on these IDEAL criteria, this stage may be assigned to this study [20].

In this study, we did not encounter severe complications such as fistula. However, the number of patients included in this pilot study is too low to conclude (yet) that RaLC is a safer procedure than vaginal colpectomy. Also there is very little literature available showing complication rates for vaginal colpectomy in patients.

Vaginal (partial) colpectomy is mainly described in patients with deep infiltrating endometriosis and in patients with cancer and may therefore have higher complication risk than our patient category. Complications described are mostly urological: 3.5–14.3% persistent urinary retention [12, 13, 21], and one paper describes 12.5% bladder perforation (1/8 patients) [22]. Complication rates for fistula (vesicovaginal or rectovaginal) are not reported in the literature, although there is a reported 0.7% incidence on rectal lacerations during surgery that involved the posterior vaginal wall [23]. In this series of patients, published by Hoffman et al. [23], the majority of procedures to the posterior vaginal wall included was for prolapse; other procedures were for neoplasia and transgender. In the 6 patients included, there was one serious adverse event: a rectal laceration of 2 cm (1 of 6 patients; 16.6% in this particular subgroup) [23]. None of these patients had further postoperative sequelae.

Sehnal et al. [24] published in 2008 a prospective randomized controlled trial in which they compared 3 methods of hysterectomy (through median laparotomy, Pfannenstiel incision or laparoscopy) in 61 FtM patients. The authors conclude that TLH is the best treatment because of better wound healing, low complication rate, short admission time and no damage to the rectus abdominis muscle and epigastric vessels (important for phalloplasty using a rectus abdominis flap). In none of the cases, a direct (vaginal) colpectomy was performed. Furthermore, they did not include vaginal hysterectomy as they deemed this procedure too difficult in this patient group. Kaiser et al. [25] published in 2011 a series of 106 FtM patients undergoing bilateral subcutaneous mastectomy, vaginal hysterectomy with BSO and complete vaginal colpectomy in the same surgical session. Although a series of more than 100 colpectomies in a transgender population is remarkable, Kaiser et al. did not describe any further detail regarding technique and results of colpectomy specifically. The conversion risk to laparotomy in this study was 2.8% of cases, and the major complication rate was 5.4% (bowel perforation 0.9%, bladder perforation 0.9%, compartment syndrome 1.8%, intra-abdominal bleeding 1.8%) [25.] If the vaginal approach is preferred and the narrow vagina limits appropriate access, one may consider the Schuchardt incision. This incision has been used for radical vaginal hysterectomy. Briefly, the skin is incised over 3–4 cm from the lower and middle third of the left labium majus. The vaginal wall, which is now under tension, is incised down to the levator ani muscle, and the rectum is displaced. The layers of the perineum are divided to expose the levator, which is then divided almost completely [26]. Schuchardt technique is known to cause considerable morbidity, such as an increased risk of bleeding, wound dehiscence, nerve damage, vaginal vault prolapse and in rare cases bowel evisceration. The preceding clearly shows that laparoscopic hysterectomy in patients is not inferior to vaginal hysterectomy.

Benefits of our technique over the vaginal route include an excellent exposure of the operating area with a complete overview of the vaginal epithelium, blood loss is minimal, postoperative recovery is fast with rapid postoperative mobilization and discharge, and there is almost immediate regain of bladder function after surgery. Admission to the hospital is longer after RaLC than after other laparoscopic procedures. This might be due to prudence with a new technique and unknown postoperative sequelae. This is underlined by the adjustment of the protocol concerning admission period during our cohort study. Further decrease in admission period is feasible. This is in line with the fact that more and more laparoscopic and robotic-assisted hysterectomies are executed in daycare [27‐32].

Since patient recovery in our study was fast, we considered removing the catheter immediately after surgery instead of day 2. Joshi et al. [33] found no difference in UTI between patients who had their catheter removed immediately after an abdominal hysterectomy compared to patients who had their catheter removed after 24 h, while recatheterization was slightly more frequent in the early removal group. Although these patients are not completely comparable to colpectomy patients, it seems that early removal of urinary catheters in colpectomy patients (day 1) should be feasible.

The clinical implications are not yet clear but may include a faster return to work, less complications and less blood loss. Although these results are promising, this is a complex procedure even for the experienced robot surgeon and a clear learning curve is seen. Implementation of this technique on a larger scale and by more surgeons might not show the same positive results. The technique is still in the developmental phase according to the IDEAL criteria. Therefore, we advise to utilize this technique solely in a highly specialized center in a research setting in order to evaluate the technique on a larger scale.

To our knowledge, this is the first study to report on RaLC combined with robot TLH and BSO in FtM patients. Future research should be directed toward quality of life assessment and long-term follow-up. Furthermore, RaLC in combination with robot TLH/BSO needs to be compared to the standard two-step procedure of TLH/BSO and consecutive vaginal colpectomy. Also analysis should include complication rates of subsequent phalloplasty procedures.