Temporal Trends and Outcomes of Gender-Affirming Gynecologic Surgery

Performance of gender-affirming gynecologic surgery is increasing in the United States.

Abstract

We performed a population-based analysis to examine the trends in utilization and outcomes of gynecologic surgical procedures performed for gender-affirming care using the National Inpatient and Ambulatory Surgery Samples. Overall, 3,344,186 (95% CI, 3,281,163–3,407,209) unique individuals who underwent gynecologic surgery were identified. Transgender and gender-diverse (TGD) individuals accounted for 19.7% (345/1,748) of the labiaplasty–clitoroplasty procedures, 15.0% (3,355/22,433) of vaginoplasties, 2.0% (212/10,679) of vaginectomies, 0.2% (5,369/2,346,748) of hysterectomies, and 0.1% (515/962,578) of oophorectomies. The percentage of cases performed in TGD individuals increased substantially from 2016 to 2020 for vaginoplasty (6.2–24.1%) and clitoroplasty–labiaplasty (8.0–35.1%). Among patients who underwent vaginoplasty, the complication rate was significantly higher among TGD individuals (16.5% vs 11.1%) (adjusted odds ratio 1.85, 95% CI, 1.22–2.81). For the other procedures, the complication rate was not higher.

Utilization of gender-affirming surgery has increased over the past decade.^1,2 A variety of gynecologic procedures, including lower genital tract operations (vaginectomy, vaginoplasty and labiaplasty) as well as oophorectomy and hysterectomy, may be performed as part of gender-affirming care. However, data are largely lacking on utilization and outcomes of gynecologic surgical procedures performed for gender-affirming care. We examined trends in use of gender-affirming gynecologic surgery using population-based data and describe rates of major complications and length of stay for individuals undergoing these procedures for gender-affirming care compared with general gynecologic care.

METHODS

Data from the 2016–2020 National Inpatient Sample and the National Ambulatory Surgery Sample were used in this population-based analysis to identify the sample cases and weighted to yield national estimates while accounting for sampling design. The National Inpatient Sample captures inpatient medical care and procedures, and the National Ambulatory Surgery Sample collects data on selected ambulatory surgical interventions from hospital-affiliated facilities in the United States.^3,4 Data for the study were deidentified and deemed exempt by the Columbia University IRB.

Patients aged 18 years or older were included if they underwent one of the five gynecologic surgical procedures that may be performed as part of gender-affirming care: hysterectomy, oophorectomy, vaginectomy, vaginoplasty, and labiaplasty–clitoroplasty.^2,5,6 Patients with gynecologic cancers were excluded. These procedures were considered as being performed for gender-affirming care if the patient had International Classification of Diseases, Tenth Revision, Clinical Modification diagnostic codes related to gender dysphoria (F64.0-64.9, Z87.890) (referred to as transgender and gender-diverse [TGD] individuals hereafter). To include each person only once, a hierarchy of procedures was applied (Appendix 1, available online at http://links.lww.com/AOG/D741). Performance of concomitant procedures including colporrhaphy, colpopexy, and incontinence repair was recorded. The outcome measures included major perioperative complications (measured using a previously described coding schema⁷) and length of stay.

Trends of gender-affirming surgery were assessed from 2016 to 2020 using Rao-Scott χ² test with Bonferroni adjustment. Major perioperative complications and length of stay were described for patients undergoing surgery for gender-affirming and general gynecologic care. Multivariable logistic regression analysis accounting for sample design was used to estimate the association between gender-affirming surgery and major perioperative complications. All analyses were conducted using SAS 9.4.

RESULTS

Overall, 2,015,062 unique individuals who underwent the gynecologic procedures of interest were identified, and the sample size was weighted to 3,344,186 (95% CI, 3,281,163–3,407,209). The weighted number of individual procedures ranged from 1,748 for labiaplasty–clitoroplasty to 2,346,748 for hysterectomy (Appendices 1 and 2, available online at http://links.lww.com/AOG/D741).

Gender-affirming surgery accounted for 19.7% (345/1,748) of the labiaplasty–clitoroplasty procedures, 15.0% (3,355/22,433) of vaginoplasties, 2.0% (212/10,679) of vaginectomies, 0.2% (5,369/2,346,748) of hysterectomies, and 0.1% (515/962,578) of oophorectomies. The percentage of gender-affirming surgery cases increased between 2016 and 2020 for vaginoplasty (6.2–24.1%) and clitoroplasty–labiaplasty (8.0–35.1%). In contrast, the percentage of oophorectomies performed in TGD individuals declined (0.2–0.04%) and the percentage of hysterectomies and vaginectomies was relatively stable over time (Fig. 1).

Fig. 1. Trends of gender-affirming gynecologic procedures from 2016 to 2020. Vaginectomy (P=.49), vaginoplasty (P=.03), and clitoroplasty–labiaplasty (P=.002) (A); hysterectomy (P=.03) and oophorectomy (P=.001) (B). Error bars show 95% CIs. TGD, transgender and gender-diverse.

Chen. Outcomes of Gender-Affirming Gynecologic Surgery. O&G Open 2024.

Among patients who underwent vaginoplasty, the complication rate was significantly higher in TGD individuals (16.5% vs 11.1%) (adjusted odds ratio [aOR] 1.85, 95% CI, 1.22–2.81) (Table 1). The complication rate was higher among TGD individuals who underwent vaginectomy (37.1% vs 20.3%), but it was no longer significant after accounting for demographic factors due to limited sample size (aOR 0.59, 95% CI, 0.21–1.67). In contrast, TGD individuals who underwent clitoroplasty–labiaplasty (21.9% vs 30.3%, aOR 0.52, 95% CI, 0.31–0.88) and hysterectomy (5.2% vs 8.3%, aOR 0.64, 95% CI, 0.52–0.79) experienced complications less commonly than cisgender women. The complication rates for oophorectomy were similar for TGD individuals and cisgender women (6.2% vs 4.2%, aOR 1.30, 95% CI, 0.68–2.48). The higher complication rate for vaginectomy and vaginoplasty was largely due to bladder injury; the higher complication rate for cisgender women after clitoroplasty–labiaplasty was driven by wound infections (Appendix 3, available online at http://links.lww.com/AOG/D741). The median length of hospital stay was greater for TGD individuals for both vaginectomy (4.2 vs 0 days) and vaginoplasty (2.5 vs 0 days). The median length of stay was similar between TGD individuals and cisgender women for the other procedures.

Table 1.

Outcomes in Individuals Undergoing Gynecologic Procedures^*,^†

Procedure	Any Complication			Complications Other Than Transfusion		Transfusion		Length of Stay [Median (IQR)]
	Cisgender	TGD	TGD vs Cisgender	Cisgender	TGD	Cisgender	TGD	Cisgender	TGD
Hysterectomy	194,296 (8.3)	281 (5.2)	0.64 (0.52–0.79)	156,797 (6.7)	264 (4.9)	56,660 (2.4)	52 (1.0)	0 (0–0.4)	0 (0–0)
Oophorectomy	40,013 (4.2)	32 (6.2)	1.30 (0.68–2.48)	36,741 (3.8)	32 (6.2)	5,336 (0.6)	—‡	0 (0–0)	0 (0–0)
Vaginectomy	2,124 (20.3)	78 (37.1)	0.59 (0.21–1.67)	2,098 (20.0)	73 (34.7)	149 (1.4)	16 (7.6)	0 (0–0)	4.2 (0.7–5.0)
Vaginoplasty	2,112 (11.1)	554 (16.5)	1.85 (1.22–2.81)	2,095 (11.0)	489 (14.6)	26 (0.1)	95 (2.8)	0 (0–0)	2.5 (2.0–3.8)
Clitoroplasty–labiaplasty	425 (30.3)	75 (21.9)	0.52 (0.31–0.88)	424 (30.2)	75 (21.9)	—‡	—‡	0 (0–0)	0 (0–0)

IQR, interquartile range; TGD, transgender and gender-diverse.

Data are n (%) or adjusted odds ratio (95% CI) unless otherwise specified.

^*For patients undergoing multiple procedures, the hierarchy of hysterectomy then oophorectomy then vaginectomy then vaginoplasty then clitoroplasty–labiaplasty was applied.

^†Logistic regression models were fit to examine the association between TGD individuals and any complication separately for each gynecologic procedure while accounting for weight and sampling design. Covariates included TGD vs cisgender; age; year; race and ethnicity; insurance status; income status; hospital bed size, location, and teaching status; region; and comorbidity score. For hysterectomy, the model was further adjusted for route of hysterectomy and concomitant oophorectomy, vaginectomy, vaginoplasty, clitoroplasty–labiaplasty, anterior or posterior repair, incontinence repair, and colpopexy. For vaginectomy, the model did not include insurance status due to small cell size, and it was further adjusted for concomitant vaginoplasty. For vaginoplasty, the model was further adjusted for concomitant clitoroplasty–labiaplasty.

^‡Data suppressed due to small cell size (fewer than 10).

DISCUSSION

These data suggest that performance of gender-affirming gynecologic surgery is increasing in the United States. Overall, gender-affirming surgery was more common among lower genital tract reconstructive operations than among hysterectomy and oophorectomy. Operative morbidity for vaginectomy and vaginoplasty was higher in TGD individuals than in cisgender women.

We recognize a number of important limitations. There may be undercapture or misclassification of gender-affirming surgery, particularly given the difficulties of distinguishing sex assigned at birth and a patient's gender identity and the fact that not all TGD individuals receive a diagnosis indicating gender dysphoria. We recognize that performance of a gynecologic procedure in TGD individuals may be for reasons other than gender affirmation. We lack granularity on the specific surgical procedure performed for each grouping of procedures. Undoubtedly there is variability in the specific operative interventions performed for each procedure class, particularly for the lower genital tract operations. A priori, our goal was to report on broad trends and not distinguish particular operative factors associated with complications. Although we selected only major complications likely to generate a billing code, some complications may have been underreported or were not captured.

In sum, these findings suggest that use of gender-affirming gynecologic surgery is increasing, particularly of the lower genital tract. Given the high morbidity of these procedures, further research is needed to assess both short-term and long-term outcomes.