Average blood loss for male-to-female SRS is minimal. However, pelvic blood flow is impressive and blood loss can quickly exceed 600 cc even among experienced surgeons. Risk of hemorrhage increases with BMI >30, in younger patient age, and in patients who experience tumescence (erection) at the time of surgery. Less blood loss is seen in older patients, in those with prior orchiectomy, and in patients with medical problems that compromise microvascular blood flow (e.g., diabetes and hypertension) [2]. A 2012 German study involving 332 patients who underwent single-stage vaginoplasty reported an average of 150 mL blood loss with very low need for blood transfusion (2 %) [3]. For most surgeons, modest blood loss normally does not result in transfusion. The author’s experience was that transfusion was necessary in just 7 of 1,135 (0.6 %) GRS procedures (Bowers ML, 2014, MTF vaginoplasty blood loss, unpublished data). Surgeons typically no longer have blood on site, favoring a type and screen as opposed to type and cross. Threshold criteria for blood transfusion have also changed and become increasingly selective. In general, a hematocrit of greater than 22 % (hemoglobin 7.0 g/dL) immediately postoperatively, in the absence of symptoms and significant cardiac risk factors, is well tolerated by patients and does not require transfusion.

Infection of tissue, whether grafted, a flap, or adjacent to the surgical site, is unusual but may arise because of compromised blood supply and tissue stretching. One study reported a 5 % rate of SSI (superficial abscesses, largely labial) that was successfully managed with incision and drainage and cephalosporin-based antibiotics [3]. A Detroit center reported on 43 patients who underwent genital reconstructive operations between years 1984 and 2008 and found a 21 % rate of SSI in those who underwent male-to-female GRS. Importantly, SSIs were independently associated with increased frequency of total operative procedures [4]. Other authors have suggested that development of SSI, particularly severe infection, contributed to complications such as loss of vaginal depth or stricture requiring additional surgical revision [3]. In general, the incidence of SSI is lower with increased surgeon experience and reduced operative time [5].

The etiology of this complication relates to devascularization of tissues secondary to ligation of arterial supply, compression of arterial structures, or poor host blood supply. Medical history (presence or absence of comorbidity) and nutritional status can also play a significant role in determining outcome and successful healing. Diabetes, while not a contraindication to GRS per se, is associated with greater risk for tissue necrosis. Selective debridement of tissues, wet-to-dry dressings, and “a tincture of time” are the best allies in treatment of these issues. Reopening of a portion of the incision and drainage is indicated for abscess but is rarely necessary. Pelvic and perineal circulation tends to be excellent and resilient, allowing many of these losses to ultimately resolve with such conservative management.

Necrosis of the neoclitorisis, fortunately, is uncommon. Most surgeons report an incidence of 1–3 % [3, 6–9], but rates as high as 10 % were noted in one review series [10]. Even in these cases, the outlook remains reasonably positive, as most patients can achieve orgasm with or without clitoral sensation. The best evidence for this comes from earlier techniques for GRS (Drs. Biber and Schrang) where clitoral preservation using homologous tissues was not done. In those procedures, amputation of the dorsal nerves and arteries at the crus meant that postoperative sensation arising from the neoclitoris was not possible. Despite a lack of sensory innervation from the dorsal nerves, cavernosa, and glans, most of these patients – via innervations from residual spongiosum tissue, retained in these surgeons’ patients as the neoclitoris – were able to orgasm. When clitoral necrosis does occur, the necrosis is most often partial with neovascularization and some residual sensation possible. Finally, for the rare patient whose clitoris loses both blood supply and innervation, the other areas retaining erectile or erogenous sensation (spongiosum, G-spot, prostate, etc.) can provide enough stimuli to allow orgasm. If patients remain anorgasmic after 1 year, topical testosterone cream (1 % compounded) can be helpful.

Labial necrosis is uncommon, with wound dehiscence being a more frequently reported problem [3]. Dehiscence is generally focal and localized and, in such cases, is best managed with local wound care and healing by secondary intention. Cosmetic complaints or scarring can be addressed at subsequent labiaplasty. Labiaplasty is normally not offered until at least 3 months following primary GRS to permit optimal healing and restoration of blood flow in flaps. For surgeons who perform a so-called two-stage procedure, this secondary cosmetic procedure is routine. In patients who have undergone vaginoplasty in a single stage, labiaplasty and/or scar revision could be offered as an additional surgical procedure but is rarely necessary due to the remarkable healing capacity of this area of the body.

Major surgery predisposes patients to a risk of deep venous thrombosis or pulmonary embolism. The pathophysiology relates to acute inflammatory reaction caused by tissue trauma, activation of clotting cascade, and venous slowing or stasis created by prolonged immobility [10]. Risk is further compounded by certain comorbidities such as age, obesity, tobacco smoking, malignancy, and certain medications. While long-term use of cross-gender hormones appears to be safe [11–13], these patients, theoretically, may be at greater risk of VTE due to use of exogenous hormones during the perioperative period. Rare reports of nonfatal VTE associated with GRS are found in the literature [14].

To reduce the risk of thromboembolic complications, most clinicians have advocated discontinuation of exogenous hormones several weeks prior to plastic surgery procedures [12, 15]. In fact, half of British plastic surgeons who responded to a survey advocated discontinuation of hormone replacement therapy prior to surgery, with most recommending a hormone-free interval of 5–6 weeks prior to reassignment [13]. Although a dose-related risk assessment of HRT does not exist, these recommendations have been accepted as fact, although route of administration and type of hormone rather than presence or absence of HRT may be more important in terms of risk assessment for VTE. Transdermal application of estrogens appears to hold a lower risk for VTE than oral administration [16]. Additionally, ethinyl estradiol is known to increase VTE risk by 20-fold in MTF transsexuals on cross-sex HRT and is no longer recommended [11, 13]. Recent data suggests that reduction of estrogens prior to surgery, rather than outright discontinuation, may improve perioperative feelings of well-being without increasing the risk of VTE. Accordingly, not all surgeons who perform GRS consistently recommend discontinuation of exogenous hormones prior to surgery [17]. Early postoperative mobilization, sequential compression devices, and prophylactic anticoagulant therapy may be sufficient prophylaxis for the majority of low-risk clientele.

Inadvertent entry into the bowel, bladder, or urethra is an inherent risk of MTF vaginoplasty. Although rare and not disastrous in experienced hands, these complications need immediate recognition and repair if major fistulous sequelae are to be avoided. A fistula represents the gravest of potential complications and can occur even with expert repair. A mechanical bowel prep with or without antibiotic is recommended preoperatively by most surgeons when embarking upon perineal dissection to create the neovaginal cavity. The key anatomic landmark followed is the central tendon of the perineum immediately beneath the urethra, a route similarly taken in perineal prostatectomies. Urologic structures are normally well protected with the use of a handheld retractor (Ferreira Breast Retractor 95 × 25 mm, smooth end with fiber-optic light) held squarely in the midline. For rectal protection, aside from assiduously following the central tendon, keeping a portion of it posterior to the dissection field, surgeons variably use other methods. This can include the use of a Lowsley retractor placed transurethrally to retract the urologic structures anteriorly to allow the rectum to drop as dissection advances cephalad. Others employ an assistant’s finger or dilator in the rectum, retracting posteriorly in addition to the anterior retraction. This allows the surgeon an additional palpable posterior backstop as dissection proceeds. Thermal injury is a potential problem but does not appear to be a significant factor in viscous injury [18]. Some have also advocated blunt dissection as opposed to sharp or cautery dissection, although this surgical philosophy remains debatable. Fistula rates after viscous injury are rarely reported but should not occur if a tension-free repair of the viscous injury has been performed [19].

Rectovaginal fistula presents with passage of stool or rectum through the neovagina. Presentation later than 6 weeks post-op is unusual, although reports of late RV fistula in patients with inflammatory bowel disease have been reported. Aside from RV fistula arising de novo secondary to inflammatory bowel disease, the etiology of most other RV fistulas is likely surgical. Dilator related injuries causing RV fistula have been reported. Cautery injuries do not appear to be causative. Most injuries occur within 3 cm of the vaginal introitus, although a higher injury is possible. Diagnosis is best confirmed by Gastrografin enema. Referral to a colorectal surgeon is indicated for repair.