Incidence, Prevalence, Diagnostic Delay, and Clinical Presentation of Female 46,XY Disorders of Sex Development

Context:

The prevalence of phenotypic females with a 46,XY karyotype is low, thus current knowledge about age and clinical presentation at diagnosis is sparse even for the most frequent conditions, androgen insensitivity syndrome (AIS), and gonadal dysgenesis.

Objective:

To estimate incidence, prevalence, age at diagnosis, and clinical presentation at diagnosis in 46,XY females.

Design and Setting:

A nationwide study covering all known females with a 46,XY karyotype in Denmark since 1960. The diagnosis of 46,XY disorder of sex development (DSD) was determined by medical record evaluation, data from the Danish National Patient Registry, and genetic testing, if available.

Patients:

A total of 166 females registered as 46,XY females in the Danish Cytogenetic Central Registry were identified.

Results:

A total of 124 females were classified as having 46,XY DSD, 78 with AIS and 25 with gonadal dysgenesis, whereas the remaining subjects had a variety of different diagnoses. The prevalence of 46,XY females was 6.4 per 100 000 live born females, and for AIS and gonadal dysgenesis, it was 4.1 and 1.5 per 100 000, respectively. Median age at diagnosis was 7.5 years (95% confidence interval, 4.0–13.5; range, 0–34 y) in AIS and 17.0 years (95% confidence interval, 15.5–19.0; range, 0–28 y) in gonadal dysgenesis (P = .001). Clinical presentation was dependent on cause of DSD.

Conclusions:

The first estimate on prevalence of 46,XY females is 6.4 per 100 000 live born females. The presentation of AIS and gonadal dysgenesis is distinctly different, with AIS being diagnosed during childhood and gonadal dysgenesis during pubertal years. The presenting phenotype is dependent on the cause of 46,XY DSD.

Disorders of sex development (DSDs) are congenital conditions in which the development of the chromosomal, gonadal, or anatomical sex is atypical. The term 46,XY DSD females refers to phenotypic females with a male genotype, previously also termed sex reversal, intersex, or pseudohermaphroditism (1, 2). In this article, the term 46,XY females is used.

The most frequent causes of 46,XY females are androgen insensitivity (androgen insensitivity syndrome [AIS]) and gonadal dysgenesis. AIS is caused by X-linked androgen receptor mutations, which impairs androgen-dependent male sexual differentiation to various degrees (3–5). Gonadal dysgenesis can be caused by mutations in testes determining factors on the Y chromosome (6). In 10%–20% of cases, a mutation is found in the sex-determining region Y (SRY) gene (7, 8). Other rare causes of female 46,XY DSD are enzymatic defects in the androgen biosynthesis cascade (eg, 5a-reductase deficiency, 17β-hydroxysteroid dehydrogenase [or 17β-ketosteroid reductase] deficiency, and Leydig cell dysfunction) (4, 9). However, in a considerable proportion of 46,XY females, no specific gene mutation can yet be determined.

Clinically, individuals with complete AIS present with female external genitalia. The gonads have differentiated to testes, and the typical location of testes is intraabdominal, in the inguinal canals or in the labia. Due to normal production of anti-Müllerian hormone in the fetus, the Müllerian structures regress (ie, the upper part of the vagina, uterus, and the Fallopian tubes). During puberty, some degree of breast development is seen due to aromatization of excessive amounts of androgens into estrogens, but pubic and axillary hair are scanty or absent (3). Individuals with partial AIS show varying degrees of virilization at birth. Thus, these patients may be raised as either females or males (4, 6, 9).

Individuals with pure 46,XY gonadal dysgenesis present with normal external female genitalia. The gonads are streak gonads without hormone production, so Müllerian structures are present, but no development of secondary female characteristics during puberty is seen (6, 7, 10).

46,XY females are rare. Data on the incidence are sparse and estimates vary widely. The incidences of AIS and gonadal dysgenesis are reported to be 1–5 per 100 000 births (11–13) and 1 per 80 000 births (7), respectively.

The aims of this nationwide cohort study were 1) to estimate the incidence and prevalence of 46,XY females; 2) to describe the complaints and clinical findings leading to referral and subsequent diagnosis; 3) to determine the age at diagnosis; 4) to assess the degree of virilization at diagnosis and the localization of gonads; 5) to assess the frequency of gonadectomy and/or genital surgery; and 6) to assess the frequency of hormone replacement therapy (HRT).

Patients and Methods

Since 1968, all Danish residents have been registered in the Civil Registration System with an assigned unique 10-digit civil registration number, from which date and year of birth as well as sex can be identified. The civil registration number enables accurate matching of data from different registries. The health care system in Denmark is a public tax funded system ensuring all citizens equal access to health care.

Identification of 46,XY females

The Danish Cytogenetic Central Registry (DCCR) was founded in 1968, and all individuals karyotyped in Denmark since 1960 are recorded in the registry, excluding stillbirths. From DCCR, all individuals with a male karyotype and officially registered as females in the Civil Registration System were identified. Data on civil registration number, karyotype, and date of karyotyping were retrieved. In total, 165 individuals karyotyped postnatally during November 1966 until December 2014 were identified. One additional patient karyotyped in 2015 was known at one department of clinical genetics, but due to some delay from karyotyping to registration in DCCR, the patient was not included in the dataset retrieved from DCCR.

All in-patient contacts in Denmark have been registered in the Danish National Patient Registry (DNPR) since 1977 and, likewise, all out-patient contacts since 1996. From DNPR, International Classification of Diseases (ICD)-8 codes, ICD-10 codes, and information on the relevant departments were retrieved for all patients identified in DCCR.

In order to confirm the diagnosis of female 46,XY DSD, evaluation of medical records was essential. If medical records were insufficient to confirm the diagnosis or if the medical record was nonexisting, senior specialists in DSD evaluated the available data (ICD-8 and ICD-10 codes) retrieved from the DNPR and determined the status of the patient by consensus (Figure 1). The etiology of the condition was determined on the basis of karyotype, tests for specific gene mutations (if available), or clinical phenotype retrieved from clinical medical records.

Figure 1

Illustration of the cohort of 46,XY females and the evaluation process. *, 1 patient was only identified from a department of clinical genetics due to delay in the update of the DCCR; **, 7 transsexual males, 2 registration errors; ***, ICD-8 and ICD-10 codes retrieved from the DNPR. Abbreviation: StAR, steroid acute regulatory gene.

Patients that had a specific gene mutation (n = 26) were categorized according to the mutation. Patients that were not genetically tested, or in whom a specific gene mutation was not identified by testing, were categorized retrospectively according to their clinical presentation based on findings in medical records. We categorized patients according to the recently published criteria into two broad groups consisting of 1) disorders of gonadal development (gonadal dysgenesis) and 2) disorders of androgens synthesis or action (AIS) (2). Patients were categorized as gonadal dysgenesis if they presented with remnants of gonads (streak gonads). No subcategorization into pure gonadal dysgenesis or mixed gonadal dysgenesis was possible. Patients without a verified AR mutation were categorized as AIS if they fulfilled one or more of the following criteria: gonads described as testes, history of AIS in the family, absence of uterus and other Müllerian derivatives, or extraabdominal localization of gonads. No subcategorization into Complete Androgen Insensitivity Syndrome or Partial Androgen Insensitivity Syndrome was possible. If no diagnosis could be ascertained, patients were categorized as 46,XY females without known etiology (unclassified 46,XY females).

Diagnosis

Age at diagnosis was defined as age at diagnosis evaluated from the medical record review (clinically or genetically), or as age at karyotyping retrieved from DCCR, whichever came first.

From the medical record review, following data were accessed: 1) complaints and physical findings leading to referral and subsequent diagnosis; 2) degree of genital virilization, quantified retrospectively, according to the Prader scale (14) from descriptions of the genitalia in the medical records; 3) localization of the gonads; 4) presence of uterus; 5) whether or not gonadectomy had been performed; 6) genital surgery; and 7) prescription of HRT.

Statistics

The incidence was estimated as the average number of annually diagnosed 46,XY females per million females in the background population. Incidence was estimated for the combined cohort of 46,XY females, for AIS, and for gonadal dysgenesis. Observation periods started in 1934 (combined cohort of 46,XY females), 1945 (AIS), and 1972 (gonadal dysgenesis) according to the year for the first relevant diagnosis. The prevalence was estimated as the average number of born 46,XY females per 100 000 live born females in the background population. Prevalence was estimated for the combined cohort of 46,XY females, for AIS, and for gonadal dysgenesis. Observation periods started in 1911 (combined cohort of 46,XY females), 1917 (AIS), and 1945 (gonadal dysgenesis) according to the first birth year for patients in the relevant cohort. All observation periods ended in 2015. Cases were clustered according to 5-year calendar time periods, and the number of diagnosed patients per 5 years was divided by the sum of their 5 respective birth cohorts.

Data concerning the background population was retrieved from Statistics Denmark (www.dst.dk).

During assessment of clinical presentation (complaints leading to referral and subsequent diagnosis, degree of virilization [Prader stage], localization of gonads, gonadectomy, presentation of uterus, genital surgery, and prescribed HRT), cases with missing data on the relevant topic for evaluation were excluded.

Time trend in incidence and prevalence was analyzed using Poisson regression. Time trend in age at diagnosis was analyzed using linear regression. Regression was made using every calendar year for which at least one 46,XY female was diagnosed. Median age for patients with AIS and patients with gonadal dysgenesis was compared using Kruskal-Wallis test. Patients classified as AIS and patients verified with AIS due to an identified AR gene mutation were compared on phenotypic presentation using χ² test. P < .05 was considered significant. Statistical analyses were made in StataCorp 13.1 for Windows. The study was approved by the Danish Data Protection Agency (journal number 2012–41-0047) and the Danish Health Authorities (journal number 3–3013-472/1/).

Results

From the DCCR, 165 patients, officially registered as females, were identified with a male karyotype, and additionally, one patient from a department of clinical genetics. The evaluation of diagnoses is shown in Figure 1. As shown, 10 patients were from Greenland and Faroe Islands and thus excluded. Medical records were available for 120 patients (77%). Nine patients were excluded as non-46,XY females (transsexuality, n = 7; registration error, n = 2), leaving a verified diagnosis of 46,XY DSD in 111 patients; hereof 26 patients with an identified specific gene mutation (Supplemental Table 1). Evaluation of DNPR data was made on further 36 patients, whereof 13 patients were verified as 46,XY females. Thus, in combination, 124 patients were confirmed as 46,XY females. The patients were categorized as follows: AIS (n = 78), gonadal dysgenesis (n = 25), 17β-hydroxysteroid dehydrogenase deficiency (n = 1), steroidogenic acute regulatory protein deficiency (n = 3), Frasier syndrome (Wilms tumor 1 gene mutation) (n = 1), 17α-hydroxylase deficiency (n = 1), and unclassified (n = 15). Karyotypes for the cohort are shown in Supplemental Table 2.

Incidence

During 1934–2015, the average annual number of background population females at risk of being diagnosed was 2 400 000, hence the average minimum incidence of 46,XY females was 0.6 per million females. For AIS and gonadal dysgenesis the average minimum incidence was 0.4 per million females and 0.2 per million females, respectively (Figure 2A).

Figure 2

A, The incidence of 46,XY females per million females in Denmark that was at risk of being diagnosed each year (all 46,XY females, dark blue; AIS, cyan; gonadal dysgenesis, blue). B, Prevalence of 46,XY females per 100 000 live born females during 1911–2015 (all 46,XY females, dark blue; AIS, cyan; gonadal dysgenesis, blue). Dashed lines indicate the average prevalence of all 46,XY females, AIS, and gonadal dysgenesis during 1951–2000.

There was a significant increase in the overall incidence of female 46,XY DSD throughout the entire observation period 1934–2015 for all 46,XY females combined (P = .0003) but not in the incidences of AIS (P = .17) and gonadal dysgenesis (P = .70).

Prevalence

The average prevalence of 46,XY females combined was 3.5 cases per 100 000 live born females. The average prevalence of AIS and gonadal dysgenesis was 2.3 and 1.2 per 100 000 live born females, respectively. During the observation period for the combined cohort of 46,XY females (1911–2015), there was an overall significant increase in prevalence (P < .0001).

A marked increase in prevalence was observed from 1951–2000 with an average prevalence of 6.4 46,XY females per 100 000 live born females. During the same period, the prevalence of AIS and gonadal dysgenesis increased to 4.1 and 1.5 per 100 000 live born females, respectively (Figure 2B).

From 2001 to 2015, the prevalence of 46,XY females decreased to 1.5 46,XY females per 100 000 live born females, and the prevalence of AIS and gonadal dysgenesis decreased to 1.1 and 0.0 per 100 000 live born females, respectively (Figure 2B).

Age at diagnosis

The median age at diagnosis for all 46,XY females was 14.0 years (95% confidence interval [CI], 10.3–16.0) (Figure 3A). Patients with AIS had a median age at diagnosis of 7.5 years (95% CI, 4.0–13.5; range, 0–34 y), whereas patients with gonadal dysgenesis had a median age at diagnosis of 17.0 years (95% CI, 15.5–19.0; range, 0–28 y) (Figure 3B); this difference in age at diagnosis was significant (P = .001).

Figure 3

A, Age at diagnosis in 46,XY females during 1934–2015. Dashed lines indicate the age of which 25%, 50%, and 75% of the patients are diagnosed. B, Number of patients by age at diagnosis (AIS, cyan; gonadal dysgenesis, blue). Dashed lines indicating age of which 50% of the patients with were diagnosed.

Studying the entire cohort, 25% were diagnosed by the age of 3.0 years, and 75% were diagnosed by the age of 18.0 years, whereas diagnosis was seen as late as 67 years (Figure 3A).

During the observation period, no significant change in age at diagnosis was observed for the entire cohort of 46,XY females (P = .47), for AIS (P = .10), or for gonadal dysgenesis (P = .57).

Complaints of 46,XY females leading to referral

46,XY females presented with a wide variety of complaints and physical findings leading to referral and subsequent diagnosis (Table 1). More than one complaint was present at referral in 16% (15/96) of all patients. The main complaint in the entire cohort was primary amenorrhea (42/96, 44%), whereas in AIS, it was inguinal hernia (23/67, 34%), a family history of AIS (11/67, 16%), or clitoral hypertrophy (10/67, 15%). Among patients with gonadal dysgenesis, the most frequent complaint was primary amenorrhea (17/25, 68%).

Table 1

Complaints Leading to Diagnosis of 46,XY Females

	AIS n = 78	Gonadal Dysgenesis n = 25	Total n = 124
Amenorrhea	22	17	42
Inguinal hernia	23	0	23
History of AIS in the family	11	0	11
Clitoral hypertrophy	10	1	11
Inability to have intercourse	5	1	6
Ambiguous genitalia	3	0	3
Virilization during puberty	2	0	3
Examination in relation to abdominal pain	1	2	3
Examination for other reason	1	0	2
Dysmorphic signs	0	0	2
Mismatch between prenatal genetic gender and gender at birth	1	0	1
Infertility	0	1	1
Unknown	11	3	28

Complaints leading to diagnosis in 46,XY females with AIS, gonadal dysgenesis, and other forms of female 46,XY DSD in a nationwide Danish cohort. Patients are classified according to information from diagnostics for specific gene mutations or information from medical records. Total refers to all females with 46,XY DSD, including persons with other diagnoses than AIS and gonadal dysgenesis.

Degree of virilization

Most patients (73/99, 74%) had a normal external female phenotype according to Prader staging (stage 0). Prader stages greater than zero was mainly observed in patients with AIS (21/71), most commonly being Prader stage 1 (12/20, 60%) (Table 2), whereas all patients with gonadal dysgenesis were classified with Prader stage 0.

Table 2

Clinical Presentation of 46,XY Females

	AIS n = 78	Gonadal Dysgenesis n = 25	Total n = 124
Prader stage
Prader stage 0	50	17	73
Prader stage 1	12	2	16
Prader stage 2	6	1	7
Prader stage 3	3	0	3
Unknown	7	5	25
Localization of gonads
Abdominal	20	20	41
Inguinal	29		31
Labial	10		10
Abdominal and labial	3		3
Abdominal and inguinal	3		4
Subcutaneous	1		1
Unknown	12	5	34
Gonadectomy
Yes	66	23	95
No	6		7
Unknown	6	2	22
Median age at gonadectomy	8.3 (0.6–62.4)	18.0 (6.1–28.5)	12.2 (0.6–62.4)
Uterus
Yes	0	24	26
No	73	0	82
Unknown	5	1	16
Genital surgery
Yes	36	4	41
No	33	19	60
Unknown	9	2	23
Median age at first genital surgery	19.2 (2.7–35.9)	11.3 (6.1–16.4)	18.8 (2.7–35.9)
Prescription of HRT
Yes	64	23	94
No	8		9
Unknown	6	2	21
Median age at first prescription of HRT	14 (10.0–42.0)	17 (10.0–29.0)	15.5 (10.0–42.0)

Degree of virilization, localization of gonads, gonadectomy, presence of uterus, genital surgery, and prescribed HRT for the entire cohort of 46,XY females. Total refers to all 46,XY females including persons with other diagnoses than AIS and gonadal dysgenesis.

Localization of gonads, gonadectomy, and uterus

Among patients categorized as AIS, one or both gonads were localized inguinally or labially in 68% (45/66), whereas the remaining patients with AIS and all patients with gonadal dysgenesis had both gonads localized abdominally (Table 2).

Gonadectomy was performed in 93% (95/102) of the patients, and it was performed at a significantly younger age in patients with AIS than in patients with gonadal dysgenesis (P = .002) (Table 2). In patients with AIS and a Prader stage of 0, 54% (19/35) were gonadectomized before puberty. Median time from diagnosis to gonadectomy was 1.3 years (range, 0.1–34.4 y) for AIS and 0.8 years (range, 0.01–6.5 y) for gonadal dysgenesis.

A uterus was present in all patients with gonadal dysgenesis and absent in all patients with AIS (Table 2).

Genital surgery

Genital surgery was primarily performed in patients with AIS (n = 36/41) (Table 2). Details regarding type of genital surgery are shown in Table 3.

Table 3

Genital Surgery in 46,XY Females

	AIS n = 36	Gonadal Dysgenesis n = 4	Total n = 41
Vaginoplasty	26	2	29
Clitoroplasty	9	2	11
Vaginal introitus revision	3	0	3
Vaginoplasty and clitoroplasty	6	1	7
Unknown	5	1	6

Genital surgery in 46,XY females divided on type of surgery.

Of the patients who underwent genital surgery 46% (19/41) were classified as Prader stage 0, and 18 of these patients had AIS, whereas one patient had gonadal dysgenesis. All AIS patients had either vaginoplasty or vaginal introitus revision. No data were available on the indication or type of genital surgery in the one patient with gonadal dysgenesis and Prader stage 0.

Age at first genital surgery was known in 26 patients, and 34% (9/26) of these patients underwent surgery before the age of 15. Patients exclusively treated with vaginal dilatation are not included.

Hormone replacement therapy

HRT was prescribed for 91% of the patients (94/103) (Table 2). The few patients (n = 9), in whom HRT was not prescribed, were mainly patients with AIS (n = 8), and three of these patients had not yet reached pubertal age. There was no difference in age at first prescription of HRT comparing the AIS and gonadal dysgenesis groups (P = .11).

AIS patients with an identified AR mutation vs AIS patients classified on clinical presentation

There were no differences between AIS patients with an identified AR mutation and those classified as AIS based on the available information for localization of gonads (abdominal vs extraabdominal [inguinal or labial]) (P = .17) and age at diagnosis (P = .08). Prader stage was significantly lower in patients with identified AR mutations than in patients with clinically classified AIS (P = .02).

Discussion

In this first national epidemiologic study of 46,XY females, we report a prevalence of 6.4 per 100 000 live born females. The diagnosis of 46,XY DSD in females is made throughout life. However, 75% of the patients are diagnosed before the age of 18 years. The diagnostic delay was longer in patients with gonadal dysgenesis than in AIS.

In Denmark, karyotyping became available in 1960. Thus, females with DSD deceased before that time had no possibility of being karyotyped and included in the study, leading to an underestimation of the prevalence in the early birth cohorts. A substantial increase in prevalence is observed from 1951, which implies that cases from these birth cohorts either were born or reached puberty at a time when karyotyping as the primary laboratory examination had become an established diagnostic procedure. The decrease in prevalence observed in the birth cohorts from 2001 and onwards suggests that a number of cases still are awaiting diagnosis. Thus, we consider the prevalence of 46,XY females to be best reflected by the estimated prevalence in the birth cohort during 1951–2000. Hence, the prevalence of 46,XY females is estimated to be 6.4 per 100 000 live born females, and the prevalence of AIS and gonadal dysgenesis is 4.1 and 1.5 per 100 000 live born phenotypic females, respectively. Although there may be differences in prevalence between ethnic groups, we assume that our estimates may be extrapolated to other populations as well.

Previously, the prevalence of AIS has been reported to be 1–5 per 100 000 births (11–13) and 1 per 80 000 births in gonadal dysgenesis (7). To our knowledge, there are no reports on prevalence of the total group of 46,XY females. We pooled data from 11 large cytogenetic survey studies (15–25) and calculated a prevalence of 46,XY females of 4 per 100 000 female births (ie, three 46,XY females observed among 75 628 female births). In comparison, our estimate of a total prevalence of 6.4 46,XY females per 100 000 live born females was about 50% higher, which is most likely due to the centralized registration of all karyotypes in Denmark. It appears reasonable to assume that the condition of female 46,XY DSD has a high detection rate during the lifespan due to its distinct clinical phenotypes, although the fact that some were diagnosed late indicates that some patients may still not have received a diagnosis. This is contrary to findings in similar studies on other DSD groups like Klinefelter and Turner syndromes, in whom the proportion of diagnosed patients, due to frequent nondiagnosis, is lower than the true prevalence expected from systematic cytogenetic surveys (26, 27).

It is important to emphasize that our estimates of prevalence are conservative estimates, because 23 patients registered as 46,XY females in DCCR were excluded from the study, due to nonexisting medical records or because the diagnosis was impossible to verify from DNPR data. The decrease in prevalence in recent times could theoretically be due to elective abortions of cases diagnosed prenatally. However, less than 10% of Danish women have prenatal karyotyping performed (26). This number is currently decreasing with the introduction of the combined double/triple test (measurement of maternal serum levels of α-fetoprotein, estriol, β-human chorionic gonadotropin), malformation ultrasound scan (28), and noninvasive testing of cell-free fetal DNA in maternal serum (29). In addition, we have only confirmation of abortion of two female fetuses with 46,XY, both in pregnant women from Greenland (personal communication DCCR). Thus, abortions are expected to only marginally influence the estimates. The decrease in recent years is therefore most likely due to cases that have not yet been diagnosed.

Besides a presentation of primary amenorrhea and incomplete pubertal development, females with 46,XY DSD can be referred due to discrepancy between prenatal karyotype and sex at birth, clinical signs of ambiguous genitalia, inguinal hernia (AIS), virilization during puberty, and a family history with affected siblings (9). However, a common belief has been that most 46,XY females primarily receive their diagnosis during puberty (3, 6). In this study, this was predominantly the case in patients with gonadal dysgenesis, in whom primary amenorrhea was the main complaint (68%). Patients with AIS, however, were diagnosed at a median age of 7.5 years, mostly due to inguinal hernia, and in some cases immediately after birth.

The risk of malignancy in the gonads of patients with gonadal dysgenesis is increased up to 30%–50% (30, 31). Therefore, bilateral gonadectomy has been internationally recommended (15). In this study, all patients with gonadal dysgenesis were gonadectomized at a median time of 0.8 years after diagnosis. Because there seems to be only a minor risk of malignancy in the testes in patients with complete AIS (31), recent recommendations suggest that gonads of patients with complete AIS may be left in situ throughout puberty, and maybe beyond (32). A wide range of AR mutations and their biological activity has been reported (5), but the risk of malignancy has not been systematically studied. Thus, it can be difficult on an individual basis to determine with certainty whether there is a degree of androgen action as in partial AIS, which increases the risk of malignancy, and a Prader stage of 0 at birth does not exclude androgen effects during puberty. The gonadectomies performed in our patients with AIS reflects previous clinical recommendations.

A diagnosis of 46,XY DSD may carry a stigma, and traditionally, information about the diagnosis was in some cases concealed from the patient and family. Today, most patients and health care providers recommend full disclosure of the diagnosis to the patient, if possible already during childhood. An early diagnosis may allow better adjustment to the diagnosis, a better understanding of the condition and a better compliance to treatment (2, 6, 9). However, there are no data proving that early disclosure is beneficial for the patient in the long term. Thus, studies on the topic are needed. Minimizing diagnostic delay may be beneficial when considering the increased malignancy risk in patients with gonadal dysgenesis. We found the diagnostic delay to be constant during the study period, although a shortening over time was expected, due to increased vigilance among clinicians and due to better diagnostic tools. This indicates that the diagnosis of 46,XY DSD is difficult in many cases, and a substantial decrease in diagnostic delay will only be possible with neonatal screening, which is not realistic from a cost-benefit perspective.

HRT of 46,XY females is recommended from puberty and onwards, especially after gonadectomy, in order to induce pubertal development and growth, and to prevent osteoporosis and cardiovascular disease. In individuals with a uterus, HRT also ensures normal uterine growth and regular menstruations to facilitate later egg donation and pregnancy (33–35). In this study, HRT was prescribed for 94% of patients, who had reached an age where hormone therapy is recommended. We have no data on compliance with hormone treatment.

The strength of the present study is the nationwide approach, covering all Danish female patients verified with 46,XY DSD, thus providing valid estimates of the prevalence of this condition. Further, to our knowledge, this is the largest study of 46,XY females to date.

Limitations of the study are the possible exclusion of some female patients who may in fact be 46,XY females, but in whom lack of data made a verification of 46,XY DSD impossible. Likewise, due to lack of data, some patients were included in the study as “unclassified 46,XY females.” Thus, they only contributed to the estimates of the entire cohort. Further, as in all epidemiological studies, a risk of misclassification is present, particularly for patients without accessible genetic data. Hence, in order to diminish the risk of misclassification, patients were only classified as AIS or gonadal dysgenesis without further subcategorization. In conclusion, the national estimate of the prevalence of 46,XY females was 6.4 per 100 000 live born females. The prevalence of AIS and gonadal dysgenesis was 4.1 and 1.5 per 100 000 live born females, respectively. The age at diagnosis was dependent on the specific etiology.

Acknowledgments

We thank Jan Hansen (data manager) from the Danish Cytogenetic Central Registry for identifying 46,XY females in the registry. We also thank the following MDs throughout the country for enabling access to the medical records. Thomas Larsen, Department of Gynecology and Obstetrics, Aalborg University Hospital; Erik Østergaard, Department of Pediatrics, Aalborg University Hospital; Hans Ole Daugaard, Department of Gynecology and Obstetrics, Sydvestjysk Sygehus; Helle Ejdrup Bredkjær, Department of Gynecology and Obstetrics, Herlev Hospital; Carsten Byrjalsen, Department of Gynecology and Obstetrics, Region Hospital Herning; Elke Longin, Department of Pediatrics, Hospital Vendsyssel; Peter Hornnes, Department of Gynecology and Obstetrics, Hospital Nordsjælland, Hillerød; Charlotte Wilken-Jensen, Department of Gynecology and Obstetrics, Hvidovre Hospital; Annemette Wildfang Lykkebo, Department of Gynecology and Obstetrics, Kolding Sygehus, Marianne Jakobsen, Department of Pediatrics, Kolding Sygehus; Lene Lundvall, Department of Gynecology and Obstetrics, Rigshospitalet; Per Bagi, Department of Urology, Rigshospitalet; Niels Uldbjerg, Department of Gynecology and Obstetrics, Aarhus University Hospital; Rune Weiss Næraa, Department of Pediatrics, Aarhus University Hospital; Jesper Schou, Department of Surgery, Sydvestjysk Sygehus; Anders Bojesen, Department of Clinical Genetics, Vejle Sygehus; Ida Vogel and Anne Skakkebæk, Department of Clinical Genetics, Aarhus University Hospital; Sonja Kindt, Department of Gynecology and Obstetrics, Region Hospital Viborg; Helle Klyver, Clinic of Plastic Surgery, Rigshospitalet; Eva Kleberg Andersen, Department of Gynecology and Obstetrics, Region Hospital Randers; Steen Axel Hertel, Neonatal Clinic, Rigshospitalet; Jonas Manthorpe, Department of Pediatrics, Roskilde Sygehus. Secretaries at all these departments are thanked for their invaluable help identifying and attaining the medical records.

This work was supported by research grants from Health Research Fund of Central Denmark Region, the A.P. Møller Foundation “Fonden til Laegevidenskabens Fremme,” and the Novo Nordisk Foundation.

Disclosure Summary: A.B., T.H.J., K.S., J.F., K.M.M., and M.H.V. have nothing to disclose. C.H.G. has received honoraria from Pfizer and Novo Nordisk for scientific talks.

Abbreviations

 
• AIS

  androgen insensitivity syndrome

 
• CI

  confidence interval

 
• DCCR

  Danish Cytogenetic Central Registry

 
• DNPR

  Danish National Patient Registry

 
• DSD

  disorder of sex development

 
• HRT

  hormone replacement therapy

 
• ICD

  International Classification of Diseases.