Vaginal Reconstruction in the Pediatric Population: An Analysis of a National Database

Abstract

Objective:

In the pediatric population, vaginoplasties can be performed in patients with either congenital malformations or acquired conditions. To our knowledge, there has been no study to date investigating the outcomes of vaginoplasty in the pediatric population using a nationwide database. Here, we present a national cohort study of perioperative characteristics and 30-day complications of vaginoplasty in pediatric patients.

Methods:

A level II retrospective, prognosis cohort study was performed using the Pediatric National Surgical Quality Improvement Program (NSQIP-P) database from 2012 to 2020. Data from patients age 0 to 18 who underwent vaginoplasty was queried using CPT code 57,335. Descriptive analysis was performed to elucidate patterns in patient demographics, perioperative characteristics, and 30-day post-operative outcomes.

Results:

A total of 183 patients were identified. Median age was 2.41 years (IQR 0.9 to 12.1). In this population, 58.5% of patients had ASA class 2, and 33.3% ASA class 3. Congenital malformation was present in 75.9%. Average total length of stay was 2.7 days (SD = 3.8) and readmission rate was 7.86%. Complications included urinary tract infection (3.3%), bleeding/transfusions (2.2%), organ/space surgical site infection (1.1%), and superficial incisional surgical site infection (0.6%). The most common procedures performed simultaneously with vaginoplasty included cystourethroscopy (n = 66), clitoroplasty for intersex state (n = 58), and plastic repair of introitus (n = 22).

Conclusion:

Vaginoplasties in the pediatric population were found to have low rates of 30-day readmission and low incidence of 30-day postoperative complications. Further studies focusing on prospective clinical data related to pediatric vaginoplasty can help identify factors to improve long-term outcomes in this population.

1. Introduction

In the pediatric population, vaginoplasties can be performed to restore functioning female anatomy in patients with congenital disorders, such as congenital adrenal hyperplasia (CAH), who are born with partial or complete vaginal agenesis [1–3]. The proposed motivation for performing feminizing genitoplasty procedures earlier in life is to produce the best psychological outcome for the child by helping them to avoid stigma, optimizing future self-esteem and sexual satisfaction, and reducing psychosocial risk [4–7]. However, there has been significant debate over whether or not vaginoplasties should be performed at an early stage in this pediatric population, as the procedure is not considered medically necessary, and complications of feminizing genitoplasties can have potential adverse effects on patients’ future adult sex-lives [6–9]. Much of this debate focuses on weighing the possibility of complications from the procedure, against the potential significant positive psychological impact of performing vaginoplasty at a young age [4,6,7].

Currently, most of the literature examining the safety and immediate post-operative complications of vaginoplasty in the pediatric population consist of case studies and small cohort studies examining specific complications or procedures [1,5,7,8,10]. To our knowledge, there has been no study to date that examines current demographic trends in pediatric vaginoplasty or the post-operative outcomes of vaginoplasty in the pediatric population using a nationwide database. In this study we examined data extracted from the American College of Surgeons (ACS) National Surgical Quality Improvement Program - Pediatric database (NISQIP-P), in an effort to investigate vaginoplasties in the pediatric populations by examining demographic data, as well as perioperative factors and post-operative outcomes in this population.

2. Methods

For this retrospective cohort study, data was pulled from the American College of Surgeons Pediatric National Surgical Quality Improvement Program (NSQIP-P) database. The NSQIP-P database reports on a variety of factors including demographic data, perioperative variables, and post operative complications, based on 30-day outcomes. This study was exempt from Institutional Board Review, as data was extracted from the NSQIP-P database and contains deidentified data. A Retrospective review was performed from January 2012 to December 2020 of all patients from age 0–18 years old, who underwent vaginoplasty procedure (CPT 57335). Variables analyzed included age, ethnicity, surgical specialty, presence of congenital malformation, type of congenital malformation, ASA class, readmission, reoperation, length of stay, operation time, complications, and concurrent and/or other procedures. SAS software was used via SAS Studio for analysis of this data. Descriptive analysis was performed, in which categorical variables were calculated as percentages, and continuous variables were analyzed as means and standard deviations and/or medians and interquartile ranges.

3. Results

A total of 183 patients were identified. The median age at the time of surgery was 2.41 years (IQR 0.9–12.1 years) (Table 1). Of these patients, 15.3% (n = 28) had been born prematurely (25–36 completed weeks of gestation), 79.2% (n = 145) were born at term, and 5.46% (n = 10) were unknown. ASA class in patients was shown to be ASA 1 in 8.2% (n = 15) (normal/healthy), ASA 2 in 58.5% (n = 107) (mild systemic disease), and ASA 3 in 33.3% (n = 61) (severe systemic disease). Congenital malformation was reported in 76% (n = 149) of patients (Table 2). The most common malformations included “Congenital Adrenal Hyperplasia (non-classical) or androgenital disorder” (n = 74, 40.4%) and “Indeterminate sex and pseudohermaphroditism” (n = 52, 28.4%) (Supplemental Table 1).

Table 1.

Demographics.

	Number (%) or Median (IQR)
Age	2.41 (IQR 0.9–12.1)
Ethnicity
White	110 (60.11%)
Black or African American	32 (17.49%)
Asian	10 (5.46%)
Native Hawaiian or Other Pacific Islanders	2 (1.09%)
American Indian or Alaska Native	1 (0.55%)
Unknown/Not Reported	28 (15.3%)
Hispanic ethnicity
Yes	26 (14.21%)
No	148 (80.87%)
Unknown	9 (4.92%)

Table 2.

Patient and surgical characteristics.

	Number (%) or Median (IQR)
ASA Classification
Class 1	15 (8.2%)
Class 2	107 (58.5%)
Class 3	61 (33.3%)
Laparoscopic/MIS Procedure/Open a
Laparoscopic/MIS Only	13 (8.02%)
Laparoscopic/MIS and Open	9 (5.56%)
Open Only or N/A	140 (86.42%)
Transfer Status
Admitted from home/clinic/doctor’s office	179 (97.81%)
Admitted through ER	2 (1.0%)
Transferred from outside hospital	2 (1.09%)
Surgical Specialty
General Surgery	1 (0.55%)
Gynecology	8 (4.37%)
Pediatric Surgery	25 (13.66%)
Pediatric Urology	137 (74.86%)
Plastics	1 (0.55%)
Urology	11 (4%)
Total Operation Time	202.5 (126–270)
Total Length of Hospital Stay	2 (1–3)
Any Other Procedure
Yes	145 (79.2%)
No	38 (20.8%)
Readmission b	11 (7.86)
Urinary tract infection	2 (18.2%)
Disorders of adrenal glands	1 (9.09%)
Intestinal obstruction without mention of hernia	1 (9.09%)
Injury other and unspecified	1 (9.09%)
No reason specified	6 (54.5%)
Reoperation c	6 (4.32%)
Debridement	1 (16.7%)
Exploratory laparotomy	1 (16.7%)
Placement of a nephrostomy catheter	1 (16.7%)
Placement of a urinary bladder catheter	1 (16.7%)
Incision and drainage of vaginal hematoma	1 (16.7%)
No reason specified	1 (16.7%)
Complications	11 (6.01%)
Superficial Incisional Surgical Site Infection	1 (0.55%)
Organ/Space Surgical Site Infection	2 (1.09%)
Urinary Tract Infection	6 (3.28%)
Bleeding/Transfusions	4 (2.19%)
Cardiac Risk Factor
No cardiac risk factors	162 (88.52%)
Minor cardiac risk factors	11 (6.01%)
Major cardiac risk factors	9 (4.92%)
Severe cardiac risk factor	1 (0.55%)
Congenital Malformation	139 (75.96%)
Prematurity
25–26 weeks gestation	2 (1.09%)
27–28 weeks gestation	1 (0.55%)
29–30 weeks gestation	2 (1.09%)
31–32 weeks gestation	7 (3.83%)
33–34 weeks gestation	8 (4.37%)
35–36 weeks gestation	8 (4.37%)
No	145 (79.23%)
Unknown	10 (5.46%)

^aData missing for 21 of 183 patients (11.5%).

^bData missing for 43 of 183 patients (23.5%).

^cData missing for 44 of 183 patients (24%).

Average operation time was 205.9 min (standard deviation 130.9 min), and total length of hospital stay was 2.72 days (standard deviation 3.8 days, median 2 days, IQR 1–3 days). Of the vaginoplasty surgeries performed in this population, majority were performed by pediatric urology 74.9% (n = 137), followed by pediatric surgery 13.7% (n = 25) and adult urology 6% (n = 11) specialties. Other procedures were performed simultaneously with vaginoplasty in 79.2% (n = 145) of patients, including cystourethroscopy (n = 66, 36.1%), clitoroplasty for intersex state (n = 58, 31.7%) and plastic repair of introitus (n = 22, 12.0%) (Supplemental Table 2). Postoperative complications reported included superficial incisional surgical site infection 0.55% (n = 1), organ/space surgical site infection 1.09% (n = 2), urinary tract infection 3.28% (n = 6), bleeding/transfusions 2.19% (n = 4). Readmission rate was 7.86% (n = 11), of which 70% (n = 7) were related to the initial operation, with the most common reason being UTI (Table 2). Reoperation rate was 4.32% (n = 6), listed in Table 2.

4. Discussion

In this study, data from 183 female pediatric patients extracted from the ACS NSQIP-P database was analyzed in order to evaluate post-operative outcomes of vaginoplasty in the pediatric population. The median age of patients in this population was found to be 2.41 years, and 58.5% were ASA class 2 (mild systemic disease) and 33.3% were ASA class 3 (severe systemic disease). The current study demonstrated low rates of 30-day postoperative complication (6%), readmission (7.9%), and reoperation (4%) for vaginoplasty in a population in which median age was less than 2.5 years, and more than 90% of patients were ASA class 2 or 3 (mild to severe systemic disease).

In the present study, 76% of patients were reported to have congenital malformations, with the most common being “congenital adrenal hyperplasia or androgenital disorder” affecting 40% of patients, and “indeterminate sex and pseudohermaphroditism” affecting 28% of patients (the term “pseudohermaphroditism” is taken directly from the NSQIP-P dataset, but is outdated, and refers to an individual with a disorder of sexual development). Due to the nature of the ICD9 coding system used to categorize/classify congenital malformations in NSQIP-P, there are diagnoses that can be written/recorded in multiple ways, thus making it harder to determine which diagnoses are truly the most common. For example, Mayer Rokitansky Kuster Hauser syndrome (MRKH) is one of the most common forms of DSD, but there is no ICD9 code specific to MRKH, rather there are multiple ICD9 codes that could be associated with MRKH (such as 752.49 - “Other congenital anomalies of cervix vagina and external female genitalia”), so it is impossible to tell from the data recorded in NSQIP-P how common MRKH truly is in the current study population. Furthermore, it is similarly difficult to determine from the available data what proportion of patients undergoing vaginoplasty for cloacal repair were included. There are several patients who are noted to have congenital malformations consistent with cloacal malformation, and may be represented by ICD9 codes 751.2 – “Atresia and stenosis of large intestine, rectum, and anal canal” (10 patients), 752.49-“Other congenital anomalies of cervix vagina and external female genitalia” (10 patients), comprising about 11% of the patients included in this study. However, given that these codes could also describe issues other than cloacal malformation, it is possible that this dataset does not include patients undergoing vaginoplasty for cloacal/anorectal malformation.

The current study demonstrated low 30-day postoperative complication, readmission, and reoperation rates (6%, 7.9% and 4%, respectively) after vaginoplasty in the pediatric population. Previous studies have also shown low immediate complication and reoperation rates after vaginoplasty [9–12]. However, these studies have had limited sample sizes, with many being cases studies, and the largest only including around 50 patients [9–16]. In a cohort of 50 pediatric patients under the age of 2 years who underwent vaginoplasty, Baskin et al. found a complication rate of 18%, with 14% of complications requiring further surgery. In comparison, the current study had a sample size of 183 patients and showed even lower rates of 30-day post-operative complication and reoperation, with an overall complication rate of 6% and a reoperation rate of 4%.

Previous studies have examined perioperative variables and postoperative outcomes in gender-affirming vaginoplasty in adult patients [17–19]. Compared to earlier investigation by Mishra et al. of the NSQIP database for adults undergoing vaginoplasty (n = 488) [19], the median operative time (202.5 min in pediatric, 271 min in adult) and total length of stay (2 days in pediatric, 4 days in adult) were lower in the pediatric population. Furthermore, when comparing to their study, in pediatric population the rates of wound dehiscence, superficial surgical site infection, deep surgical site infection (1.64% in pediatric, 9% in adult) were lower. However, the rate of UTI in the pediatric population (3.28% in pediatric, 2% in adult) and readmission rate (7.86% in pediatric, 4.3% in adult) were slightly higher. Reoperation rate was similar between the two populations (4.32% in pediatric, 4.7% in adults) [19].

Given that more than half of the patients included in the current study were noted to have “congenital adrenal hyperplasia or androgenital disorder,” or other forms DSD, in discussing the appropriateness of performing vaginoplasty in this population, it is important to consider the optimal age for surgery, as every DSD patient is unique [20]. Although there has been concern for higher rate of complication when vaginoplasties are performed in young children, in the present study, rates of immediate 30-day post-operative complications were found to be lower than previously thought and are seemingly comparable to adult rates at the same post-operative follow-up time. Previous studies have explored the surgery-timing preferences of differences of sexual development patients [21–23]. By administering questionnaires to individuals with a DSD diagnosis, Bennecke et al. demonstrated that nearly two-thirds of individuals with CAH think that genital surgeries are most appropriate in infancy (1 month–3 years) or childhood (4 years–12 years) [21]. Although patients’ satisfaction and preference for the timing of vaginoplasty could not be concluded from the current study, our data does show that the age at which the vaginoplasties were performed tended to be relatively young. Indeed, the median age of the patients was 2.41, and almost 75% were performed in patients under age 12.

As previously mentioned, the argument for performing vaginoplasty at a young age includes the theoretical potential for maximizing self-esteem and psychological benefit [4–7]. However, surgically assigning anatomical gender before a patient has capacity to make informed decisions regarding their own gender identity can be detrimental to psychological well-being as well [20]. Furthermore, although there are a variety of medical arguments for performing vaginoplasty at a young age (such as preventing conditions resulting from obstruction of menstruation and optimizing future functionality [24,25]) there is much debate over whether or not these medical arguments are actually valid [24]. Due to the scope of the data examined in this study, we cannot comment on what the appropriate age is for vaginoplasty, but we can say that if it is done (whether electively, or by necessity), in the immediate 30-day postoperative period, the procedure has low rates of postoperative complication, readmission, and reoperation.

It is important to note that, especially in the DSD population, appropriateness of surgery is also related to patient autonomy and decision making [26–29]. In the pediatric population, especially for those in infancy or early childhood, decisions regarding DSD-related surgery are made by the parents, and the patients themselves are often not directly involved in decision making [21,24,26]. This study shows that if the decision is made to perform surgery early in life, it can be done with acceptable 30-day postoperative complication rates, comparable to the same surgery in adults. However, there are many psychosocial aspects of this surgery that cannot be adequately assessed through NSQIP-P data alone. This decision-making process is ethically complex, and while this study demonstrates that complication rates for pediatric vaginoplasty are acceptably low, it does not mean that vaginoplasty—or surgery for DSD in general—is the right decision for every patient and every family. As previous research has suggested, one of the most important aspects of this decision-making process is that the parents feel supported and able to leave an open or non-restrictive sexual future for their children [7,27].

The most significant limitations of this paper include the fact that our measurement of outcomes is limited to 30 days, and our inability to examine rates of complications important to future sexual health such as vaginal stenosis, scarring, or issues with mucous production that may arise months to years later. Similarly, NSQIP-P does not include data that would allow us to examine the effects of vaginoplasty at a young age on future functional considerations such as menstruation, sexual intercourse, or obstetrics/fertility. Thus, while we are able to establish the immediate, 30-day postoperative outcomes of vaginoplasty in the pediatric population, we cannot draw any conclusions about long-term safety of vaginoplasty in the pediatric population. It is also important to note that the NSQIP-P database does not provide any information regarding the type of vaginoplasty performed, so we cannot draw any conclusions about the safety and outcomes of specific techniques, only of “vaginoplasty for intersex state” in general. Type of vaginoplasty is an important consideration in examining long-term outcomes, as after care can vary greatly between methods. For example, when tissue from the bowel is used to construct a neovagina, long-term care involves monitoring for long-term complications such as colitis or cancer development [30]. Due to the nature of the NSQIP-P data, we were also unable to investigate whether patients received perioperative antibiotics or had foley catheters—potentially informative measures to examine for this study given that UTI was the most common complication and most common reason for readmission. In this study we extracted data from the NSQIP-P database using the CPT code for “vaginoplasty for intersex state,” but we cannot guarantee that in doing so we did not miss some number of cases wherein patients underwent vaginoplasty for vaginal agenesis (DSD-related or otherwise), that were simply marked with a different CPT code. Furthermore, we were not able to compare vaginoplasty outcomes in pre-pubertal and post-pubertal populations, as we have no way of determining when puberty occurred for each patient from the NSQIP-P data. The ability to more directly compare these groups would be useful in the future to determine the extent to which factors present only in the post-pubertal population, such as fully estrogenized skin and the use of vaginal dilators as adjunctive treatment, might influence healing after vaginoplasty. Finally, one of the most significant arguments in favor of performing vaginoplasty at a younger age is the potential for improved psychological outcome and maximizing self-esteem and sexual function. Unfortunately, because of the nature of the data recorded in NSQIP, we are unable to determine long-term satisfaction after vaginoplasty, or the psychological impacts of these procedures.

5. Conclusion

In this study we were able to use NSQIP-P to examine demographic trends in pediatric vaginoplasty, as well as gain insight into the frequency and types of post-operative complications in these procedures. In this pediatric population, vaginoplasties were shown to have low rates of 30-day readmission and reoperation, as well as low incidence of immediate complications. Further studies focusing on larger prospective clinical data related to pediatric vaginoplasty should be done to help elucidate factors that can improve long-term outcomes in this population. Such studies should include examination of more long-term complication such as scarring, stenosis, mucous production, as well as outcomes that include menstruation, sexual function, and obstetric considerations. Indeed, there may be great benefit in longitudinal study of not only long-term surgical outcomes, but also psychological development in vaginoplasty patients.