Abstract
Clinical practice guidelines for individuals with Turner syndrome (TS) recommend screening for neuropsychological concerns (NC) and mental health concerns (MHC). However, current provider screening and referral patterns for NC and MHC are not well characterized. Additionally, prevalence of and risk factors for NC and MHC vary across studies. This multicenter chart review study examined the prevalence, risk factors for, and management of NC and MHC in a cohort of 631 patients with TS from three pediatric academic medical centers. NC and/or MHC were documented for 48.2% of patients. Neuropsychological evaluation recommendations were documented for 33.9% of patients; 65.4% of the sample subsequently completed these evaluations. Mental health care recommendations were documented in 35.0% of records; subsequent documentation indicated that 69.7% of these patients received such services. Most notably, rates of documented MHC, NC, and related referrals differed significantly by site, suggesting the need for standardized screening and referral practices. TS diagnosis in early childhood was associated with an increased risk of NC. Spontaneous menarche was associated with increased risk of MHC. Younger age at growth hormone initiation was associated with both increased risk of isolated NC and co-occurring NC and MHC. Mosaic karyotype was associated with decreased risk of NC and MHC.
Keywords: Turner syndrome, neuropsychology, mental health, monosomy X, multidisciplinary clinics, sex chromosome aneuploidy, differences of sex development
INTRODUCTION
Turner syndrome (TS) is diagnosed in people with a karyotype containing one X chromosome and rearrangement, total, or partial absence of the second sex chromosome. TS occurs in an estimated 1/2,000 female-assigned infants and may affect many aspects of physical and neurocognitive development. Phenotypic manifestations vary significantly across affected individuals, but commonly include short stature and ovarian insufficiency (Gravholt et al., 2017). While most individuals with TS have a relatively normal full-scale IQ (Hong, Kent, & Kesler, 2009), weaknesses in aspects of cognitive functioning including visual-spatial reasoning, working memory, processing speed, and social cognition are common (Delooz, Van den Berghe, Swillen, Kleczkowska, & Fryns, 1993; el Abd, Turk, & Hill, 1995; Wolstencroft & Skuse, 2019). Additionally, comorbid neuropsychological concerns (NC) such as specific learning disabilities in mathematics, attention deficit hyperactivity disorder (ADHD), executive function deficits, and autism spectrum disorder (ASD) are more common in individuals with TS (Björlin Avdic et al., 2021). One study reported that up to 25% of youth with TS met ADHD criteria relative to 1.3% of the general female population (Russell et al., 2006). However, associations between these neuropsychological concerns with other aspects of the TS phenotype, such as growth and pubertal timing, are less well-defined (Messina et al., 2007).
The incidence of mental health concerns (MHC) such as anxiety, depression, schizophrenia, and eating disorders among individuals with TS has received little empirical study (Morris, Tishelman, Kremen, & Ross, 2020). There has been some suggestion of an increased risk of anxiety and depression (Cardoso et al., 2004) as well as psychotic disorders and body dysmorphia in TS (Björlin Avdic et al., 2021), but these findings have not been consistently observed across TS patient cohorts.
Specific phenotypic manifestations of TS, such as social challenges, may also be associated with MHC, by contributing to isolation and low self-esteem (Wolstencroft & Skuse, 2019). Medical concerns associated with TS, such as orthopedic complaints (scoliosis, leg length discrepancy or Madelung’s deformity) and diabetes, have been associated with reduced health-related quality of life, which may in turn be associated with poor self-perceptions and a heightened risk for mood disorders (van den Hoven et al., 2020). Primary ovarian insufficiency and subfertility, which occur in a majority of girls and women with TS, are associated with lifetime risk of major depression, including in adolescence (Wasserman et al., 2022), and can cause significant distress among women with TS (Sutton et al., 2006). Although desire for childbearing is common among women with TS (van Hagen et al., 2017), only 2–7% achieve spontaneous pregnancy, and a high rate of medical complications such as hypertensive disorders and pre-eclampsia has been reported (Hagman et al., 2013). Additionally, medical providers may share information about infertility in ways that do not prioritize patients’ needs for psychosocial support, or delay sharing this information until adolescence or adulthood, increasing patients’ experiences of isolation and mistrust of medical providers (Sutton et al., 2006).
Individuals with TS may experience increased body image-related distress, though the relationship between self-reported body image and classic physical manifestations of TS is complex. In particular, evidence is conflicting regarding the impact of short stature on self-image in adolescents with TS (Jez et al., 2018; Vijayakanthi, Marcus, Fritz, Xiang, & Fadoju, 2022). Short stature due to other endocrine disorders has been associated with increased risk of depression (Kostev, Teichgräber, Konrad, & Jacob, 2019), and teasing from peers about appearance has been established a risk factor for depression in TS (Rickert, Hassed, Hendon, & Cunniff, 1996). However, neither short stature nor treatment with growth hormone have been definitively linked to increased risk of depression in TS (Bannink, Raat, Mulder, & De Muinck Keizer-Schrama, 2006).
Both NC and MHC can negatively impact quality of life by creating challenges around academic, vocational, and activities of daily living. Such difficulties with navigating both day-to-day tasks and social/interpersonal relationships may complicate the ability of young people with TS to successfully adapt to the transitions of emerging young adulthood (Hutaff-Lee, Bennett, Howell, & Tartaglia, 2019). Additionally, the clinical course of MHC is not well-described in adolescents and young adults with TS.
Although clinical practice guidelines recommend routine screening for NC and MHC in patients with TS (Gravholt et al., 2017), it is unknown whether this is occurring in clinical practice, even within high-volume centers with multidisciplinary TS clinics that have dedicated psychologists. Our group has demonstrated previously that the majority of pediatric endocrinologists do not routinely screen for NC or MHC in their patients with TS (Davis et al., 2021). It is unclear if other clinicians on an interdisciplinary team are screening for these conditions, or perhaps referring patients for additional evaluation with specialized neuropsychology and/or mental health providers, even in the absence of formal or informal screening for NC or MHC. Additionally, how often patients with TS access neuropsychological and/or mental health evalutions or treatment when referrals are placed has not been previously reported. Previous work suggests that only a minority of patients with TS who experience clinically significant mental health concerns establish care with mental health providers (Alexandrou et al., 2022).
An understanding of the clinical correlates of NC and MHC in TS may help inform counseling and screening practices for providers caring for patients with TS by identifying high-risk groups. Based on general population risk factors, as well as evidence from prior research (Krantz, Landin-Wilhelmsen, Trimpou, Bryman, & Wide, 2019), we hypothesized that TS-specific features such as genotype, timing of diagnosis, height, growth hormone treatment, and pubertal onset would be associated with NC and MHC among persons with TS. In addition to identifying who is at highest risk, we hypothesized there would be variability in clinical approaches to screening, referring, and evaluating for NC and MHC between sites despite the clinical practice guideline recommendations for universal screening and assessment. In addition to elucidating who is most likely to be referred for these services, an understanding of current referral patterns could help to identify gaps in current clinical care. This information is necessary in order to improve adherence to screening guidelines and ultimately, to reduce the burden of NC and MHC for people with TS.
The aims of this three-site study were (1) to identify the prevalence of documented NC and MHC in pediatric patients with TS; (2) to quantify the association of NC and MHC with genetic, phenotypic, and clinical management-related risk factors; and (3) to describe documented referral practices for neuropsychological and mental health services among pediatric patients with TS, as well as documentation of receipt of these services.
METHODS
Approach
A retrospective medical record review of individuals with TS cared for at three large pediatric academic medical centers was performed (June 2018-April 2019). At Sites A (Northeast) and B (Midwest), clinical care is provided to individuals with TS in stand-alone sub-specialty clinics or multidisciplinary clinics which were not dedicated to the care of patients with TS. Subspecialists seeing patients with TS included providers from endocrinology, cardiology, nephrology, genetics, gastroenterology, otolaryngology, developmental pediatrics, psychology, social work, neurology, and neuropsychology. At Site C (West), a multidisciplinary clinic specific to the care of patients with TS was established in 2015. This clinic includes specialists in neuropsychology and psychology in addition to other medical specialties. In this practice model, visits with subspecialist providers from multiple disciplines occur on the same day within the same clinical encounter.
At all sites, electronic medical records were queried to identify patients with a TS diagnosis who were seen by a subspecialist for a TS-related concern between March 2013 and March 2018. Patients who had not been seen for TS-related care and those not meeting criteria for TS per the 2017 Clinical Practice Guidelines (Gravholt et al., 2017), including those with atypical genital development/genital variation or male phenotype, were excluded. All three hospitals’ institutional review boards approved this protocol prior to commencing data collection.
Data Collection
Three pediatric endocrinologists and two pediatric psychologists collaborated to develop a standardized chart abstraction form designed to characterize TS patients and medical, mental health, and neuropsychological care practices of providers across medical centers. Chart abstraction data from each site was stored in a single Research Electronic Data Capture (REDCap) database housed at Site B. Chart reviews were performed by clinical research assistants who were specifically trained on this protocol or by the clinician researchers themselves. Documentation of NC or MHC in the past medical history or problem list, or clinician documentation of known or suspected diagnoses of NC and/or MHC were abstracted. NC were defined as attention deficit hyperactivity disorder (ADHD), autism spectrum disorder (ASD), global developmental delay, and learning disorders. MHC were defined as anxiety, depression, bipolar disorder, obsessive-compulsive disorder (OCD), eating disorder, body dysmorphic disorder, and gender dysphoria. If NC or MHC were not documented in the medical chart, they were coded as not being present.
Additionally, the following covariates were also abstracted: patient demographics; karyotype; medical comorbidities (e.g., congenital heart disease); age of TS diagnosis (divided into age brackets representing prenatal, early postnatal, and late postnatal diagnosis); age of thelarche and/or menarche and whether thelarche/menarche occurred spontaneously; age of hormone treatment initiation including estrogen and growth hormone; and adult height (if attained).
Analysis
Primary outcomes included: (1) percentage of patients with documented NC; (2) percentage of patients with documented MHC; (3) percentage of patients with documented NC and/or MHC concerns; (4) percentage of patients with documentation of referral for evaluation of NC and/or MHC; and (5) percentage of patients documented to have received services (neuropsychological evaluation, mental health counseling/therapy) among those who were referred.
Descriptive statistics were used to summarize patient characteristics and clinical factors overall and by site, presented as mean (standard deviation) for continuous variables and percentage (frequency) for categorical variables. Analyses of variance were used to compare continuous variables across the three sites, and Chi-square tests were used for categorical variables.
Odds ratios and 95% confidence intervals estimated from logistic regression were used to assess associations between medical and demographic features and primary outcomes. For age-related variables (e.g., menarche, thelarche, age of sex hormone replacement initiation, and age of growth hormone initiation), we ran additional separate analyses for each of dependent variable to adjust for age at last visit. Variables with probable clinical significance and P≤0.1 on the univariate analysis were entered into a multivariate model. Pearson’s Chi-squared test assessed differences between sites with respect to NC, MHC, and referral for neuropsychological and/or mental health services. For this exploratory study, all statistical tests were 2-sided with P<0.05 considered statistically significant. All data were analyzed using SAS version 9.4 (SAS Institute, Cary, NC, USA).
RESULTS
Population Characteristics
The cohort included 631 patients: 146 at Site A, 193 at Site B, and 292 at Site C. The demographics significantly differed between sites in race, ethnicity, insurance type and age at most recent clinic visit (Table 1), overall representing a diverse study sample. In addition, the clinical characteristics of the patients differed by site, including karyotype, as a higher prevalence of patients with non-mosaic 45,X were seen at Site C, age of TS diagnosis with an overall older age at Site B, and lower prevalence of menarche (particularly spontaneous menarche) at Site C. Overall, 21% of the cohort had documented NC, 11% had MHC, and 10% had documentation of both NC and MHC (Figure 1).
Table 1.
Demographic and Clinical Characteristics
| Characteristic | Overall (N=631) | Site A (Northeast) (n=146) |
Site B (Midwest) (n=193) |
Site C (West) (n=292) |
P value |
|---|---|---|---|---|---|
|
Age at most recent clinic visit, years
Mean ±SD [range] |
13.4 ± 7.8 [0,56.4] |
13.9 ± 6.5 [0.3,26.3] |
14.6 ± 10.2 [0.0,56.4] |
12.3 ± 6.2 [0.2,38.4] |
0.004 |
| Race (n, %) | <0.0001 | ||||
| American Indian/Alaskan Native | 2 (0.32%) | 0 | 0 | 2 (0.68%) | |
| Asian | 26 (4.1%) | 13 (8.9%) | 7 (3.6%) | 6 (2.1%) | |
| Black/African American | 50 (7.9%) | 11 (7.5%) | 30 (15.5%) | 9 (3.1%) | |
| Native Hawaiian or Other Pacific Islander | 1 (0.2%) | 1 (0.7%) | 0 | 0 | |
| White | 465 (73.7%) | 92 (63.0%) | 148 (76.7%) | 225 (77.1%) | |
| More than one race | 33 (5.2%) | 2 (1.4%) | 8 (4.2%) | 23 (7.9%) | |
| Unknown/unreported | 54 (8.6%) | 27 (18.5%) | 0 | 27 (9.3%) | |
| Ethnicity (n, %) | <0.0001 | ||||
| Hispanic/Latino | 107 (17.0%) | 17 (11.6%) | 6 (3.1%) | 84(28.8%) | |
| Not Hispanic/Latino | 463 (73.4%) | 87 (59.6%) | 182 (94.3%) | 194(66.4%) | |
| Refused/not reported | 61 (9.7%) | 42 (28.8%) | 5 (2.6%) | 14(4.8%) | |
| Insurance type (n, %) | <0.0001 | ||||
| Private/commercial | 338 (53.6%) | 99 (67.8%) | 93 (48.1%) | 146 (50.0%) | |
| Medicaid | 233 (36.9%) | 42 (28.8%) | 77 (39.9%) | 114 (39.0%) | |
| Self-pay | 9 (1.4%) | 3 (2.1%) | 3 (1.6%) | 3 (1.0%) | |
| Military | 19 (3.0%) | 0 | 0 | 19 (6.5%) | |
| Other | 24 (3.8%) | 2 (1.4%) | 20 (10.4%) | 2 (0.7%) | |
| Unknown/unreported | 8 (1.3%) | 0 | 0 | 8 (2.7%) | |
| Karyotype (n, %) | 0.006 | ||||
| 45,X | 238 (37.7%) | 44 (30.1%) | 69 (35.8%) | 125 (42.8%) | |
| 45,X/46,XX | 93 (14.7%) | 19 (13.0%) | 25 (13.0%) | 49 (16.8%) | |
| 45,X/46,XY | 33 (5.2%) | 12 (8.2%) | 9 (4.7%) | 12 (4.1%) | |
| 45,X/47,XXX | 25 (4.0%) | 9 (6.2%) | 4 (2.1%) | 12 (4.1%) | |
| Other | 6 (1.0%) | 4 (2.7%) | 1 (0.5%) | 1 (0.3%) | |
| Ring X | 54 (8.6%) | 15 (10.3%) | 22 (11.4%) | 17 (5.8%) | |
| Isochromosome X | 101 (16.0%) | 28 (19.2%) | 32 (16.6%) | 41 (14.0%) | |
| Autosomal Translocation | 9 (1.4%) | 4 (2.7%) | 2 (1.0%) | 3 (1.0%) | |
| Deletion on the X chromosome | 35 (5.6%) | 7 (4.8%) | 10 (5.2%) | 18 (6.2%) | |
| Unknown/unreported | 37 (5.9%) | 4 (2.7%) | 19 (9.8%) | 14 (4.8%) | |
| Congenital heart disease (n, %) | 0.046 | ||||
| Yes | 140 (22.2%) | 43 (29.5%) | 36 (18.7%) | 61 (20.9%) | |
| No | 491 (77.8%) | 103 (70.6%) | 157 (81.4%) | 231 (79.1%) | |
| TS Diagnosis age, years (n, %) | 0.002 | ||||
| Prenatal | 130 (20.6%) | 34 (23.3%) | 22 (11.4%) | 74 (25.3%) | |
| Postnatal | 482 (76.4%) | 107 (73.3%) | 162 (83.9%) | 213 (73.0%) | |
| Postnatal, Ages 0–5 yrs | 242 (38.4%) | 50 (34.3%) | 72 (37.3%) | 120 (41.1%) | |
| Postnatal, Ages 6+ yrs | 222 (35.2%) | 55 (37.7%) | 78 (40.4%) | 89 (30.5%) | |
| Postnatal, Unknown age | 18 (2.9%) | 2 (1.4%) | 12 (6.2%) | 4 (1.4%) | |
| Unknown/unreported | 19 (3.0%) | 5 (3.4%) | 9 (4.7%) | 5 (1.7%) | |
| Menarche (n, %) | <0.0001 | ||||
| No | 333 (52.8%) | 68 (46.6%) | 95 (49.2%) | 170 (58.2%) | |
| Yes | 257 (40.7%) | 66 (45.2%) | 69 (35.8%) | 122 (41.8%) | |
| Unknown/unreported | 41 (6.5%) | 12 (8.2%) | 29 (15.0%) | 0 | |
| Spontaneous menarche (n = 257); (n, %) | 0.14 | ||||
| Yes | 93 (36.2%) | 29 (43.9%) | 27 (19.1%) | 37 (30.3%) | |
| No | 134 (52.1%) | 34 (51.5%) | 33 (47.8%) | 67 (54.9%) | |
| Unknown/unreported | 30 (11.7%) | 3 (4.6%) | 9 (13.0%) | 18 (14.8%) | |
| Age sex hormone replacement therapy initiated (n, %) | <0.001 | ||||
| Not ever on sex hormone replacement therapy | 320 (50.8%) | 68(46.6%) | 88(45.6%) | 164(56.2%) | |
| <13 years | 96(15.2%) | 24(16.4%) | 18(9.3%) | 54(18.5%) | |
| 13–15 years | 113(17.9%) | 30(20.6%) | 29(15.0%) | 54(18.5%) | |
| 16+ years | 72(11.4%) | 18(12.3%) | 37(19.2%) | 17(5.8%) | |
| Unreported | 30(4.8%) | 6(4.1%) | 21(10.9%) | 3(1.0%) | |
| Age growth hormone initiated (n, %) | 0.007 | ||||
| ≥ 9 years | 154(24.4%) | 42(28.8%) | 46(23.8%) | 66(22.6%) | |
| 6–8 years | 89(14.1%) | 26(17.8%) | 17(%8.8) | 46(15.8%) | |
| 0–5 years | 175(27.7%) | 30(20.6%) | 58(30.1%) | 87(29.8%) | |
| Not ever started | 192(30.4%) | 44(30.1%) | 59(30.6%) | 89(30.5%) | |
| Unreported | 21(3.3%) | 4(2.7%) | 13(6.7%) | 4(1.4%) | |
| Final height, cm (n=227) Mean ±SD | 151.2 ± 8.2 | 150.7 ± 6.7 | 151.2 ± 9.7 | 151.6 ± 8.0 | 0.77 |
| Ever seen in TS multidisciplinary clinic (n, % yes) | 110 (17%) | 0 | 0 | 110 (36.4%) | |
Figure 1.
Neuropsychological (NC) and/or mental health concerns (MHC) and associated medical and treatment factors in N=631 pediatric patients with TS from three U.S. sites*
TS= Turner Syndrome
GH= Growth hormone
HRT= Sex hormone replacement therapy
*Site is associated with all NC and MHC outcomes (see Tables 3–5).
Neuropsychological Concerns, Referrals, and Assessments
NC concerns were documented in 30.3% of the cohort overall, with ADHD being the most common concern (14.3%), followed by global developmental delay, learning disorders, and autism spectrum disorders (Table 2). Developmental concerns with or without mention of global developmental delay were present in 40.9% (n = 258) of the overall sample; learning concerns with or without a documented learning disorder were also common (48.5%; n = 305). Site of care was associated with odds of documented NC, with patients with TS seen at Site B and Site C with ~2.5 times greater odds of having documented NC than those seen in the northeast (Site A) (Table 3). Patients receiving care within the TS multidisciplinary clinic at Site C were more likely to have a documented recommendation for a neuropsychological evaluation (56%), compared to those seen at Site A, B, or outside of the multidisciplinary clinic at Site C (28–30%). However, only a minority of patients had any documentation of completing neuropsychological evaluations (20–22% across all sites).
Table 2.
Neuropsychological and Mental Health Concerns by Site and Receipt of Multidisciplinary Care†
| Overall (N=631) | Site | Multidisciplinary TS clinic (n=292) | |||||||
|---|---|---|---|---|---|---|---|---|---|
| Characteristic | A(Northeast) (n=146) |
B (Midwest) (n=193) | C (West) (n=292) | P value | No (n=182) |
Yes (n=110) |
P value | ||
| Neuropsychological evaluation recommended | 0.04 | <0.0001 | |||||||
| No | 417 (66.1%) | 105 (71.9%) | 134 (69.4%) | 178 (61.0%) | 130 (71.4%) | 48 (43.6%) | |||
| Yes | 214 (33.9%) | 41 (28.1%) | 59 (30.6%) | 114 (39.0%) | 52 (28.6%) | 62 (56.4%) | |||
| Neuropsychological evaluation completed | 0.02 | 0.08 | |||||||
| No | 74 (34.6%) | 11 (26.8%) | 14 (23.7%) | 49 (43.0%) | 27 (51.9%) | 22 (35.5%) | |||
| Yes | 140 (65.4%) | 30 (73.2%) | 45 (76.3%) | 65 (57.0%) | 25 (48.1%) | 40 (64.5%) | |||
| Mental health service recommended | <0.0001 | < 0.0001 | |||||||
| No | 410 (65.0%) | 113 (77.4%) | 124 (64.3%) | 173 (59.2%) | 133 (73.1%) | 40 (36.4%) | |||
| Yes | 221 (35.0%) | 33 (22.6%) | 69 (35.8%) | 119 (40.8%) | 49 (26.9%) | 70 (63.6%) | |||
| Mental health service received | 0.38 | 0.63 | |||||||
| No | 67 (30.3%) | 7 (21.2%) | 24 (34.8%) | 36 (30.2%) | 16 (32.7%) | 20 (28.6%) | |||
| Yes | 154 (69.7%) | 26 (78.8%) | 45 (65.2%) | 83 (69.8%) | 33 (67.4%) | 50 (71.4%) | |||
| Neuropsychological or mental health diagnoses/concerns | 0.0002 | <0.0001 | |||||||
| No | 326 (51.8%) | 96 (65.8%) | 84 (43.5%) | 146 (50.3%) | 109 (60.6%) | 37 (33.6%) | |||
| Yes | 303 (48.2%) | 50 (34.3%) | 109 (56.5%) | 144 (49.7%) | 71 (39.4%) | 73 (66.4%) | |||
| Neuropsychological Concerns | |||||||||
| ADHD | 90 (14.3%) | 18 (12.3%) | 31 (16.1%) | 41 (14.0%) | 22 (12.1%) | 19 (17.3%) | |||
| Autism | 25 (4.0%) | 4 (2.7%) | 10 (5.2%) | 11 (3.8%) | 6 (3.3%) | 5 (4.6%) | |||
| Global developmental delay | 87 (13.8%) | 5 (3.4%) | 29 (15.0%) | 53 (18.2%) | 25 (13.7%) | 28 (25.5%) | |||
| Learning disorder | 56 (8.9%) | 7 (4.8%) | 16 (8.3%) | 33 (11.3%) | 17 (9.3%) | 16 (14.6%) | |||
| Mental Health Concerns | |||||||||
| Anxiety | 110 (17.4%) | 17 (11.6%) | 36 (18.7%) | 57 (19.5%) | 23 (12.6%) | 34 (30.9%) | |||
| Bipolar disorder | 1 (0.2%) | 0 | 1 (0.5%) | 0 | 0 | 0 | |||
| Depression | 42 (6.7%) | 8 (5.5%) | 19 (9.8%) | 15 (5.1%) | 6 (3.3%) | 9 (8.2%) | |||
| OCD | 9 (1.4%) | 1 (0.7%) | 3 (1.6%) | 5 (1.7%) | 3 (1.7%) | 2 (1.8%) | |||
| Psychosis | 2 (0.3%) | 0 | 2 (1.0%) | 0 | 0 | 0 | |||
| Eating disorder | 9 (1.4%) | 1 (0.7%) | 2 (1.0%) | 6 (2.1%) | 4 (2.2%) | 2 (1.8%) | |||
| Body dysmorphic disorder | 3 (0.5%) | 0 | 0 | 3 (1.0%) | 2 (1.1%) | 1 (0.9%) | |||
| Gender dysphoria | 3 (0.5%) | 0 | 1 (0.5%) | 2 (0.7%) | 2 (1.1%) | 0 | |||
| Other diagnoses | 117 (18.5%) | 13 (8.9%) | 76 (39.4%) | 28 (9.6%) | 12 (6.6%) | 16 (14.6%) | |||
For some variables in this retrospective chart review, ns do not add up to 100% due to missing data
Table 3.
Association of Neuropsychological Concerns (NC) with Patient Characteristics†
| NC (Yes) | NC (No) | Univariate | Multivariate | |||||
|---|---|---|---|---|---|---|---|---|
| Variables | n=192 | n=326 | OR (95%CI) | P value | OR (95%CI) | P value | ||
| Age of diagnosis‡ (n, %) | ||||||||
| ≥6 years | 46 (24%) | 125 (38.3%) | 1.00 | 1.00 | ||||
| 0–5 years | 98 (51%) | 111 (34%) | 2.40 (1.56, 3.70) | <0.001 | 2.24 (1.36, 3.67) | <0.001 | ||
| Prenatal | 38 (19.8%) | 71 (21.8%) | 1.45 (0.87, 2.44) | 0.16 | 1.50 (0.83, 2.74) | 0.18 | ||
| Age at last visit (n, %) | ||||||||
| 0–10 years | 77 (23.9%) | 104 (31.9%) | 1.00 | 1.00 | ||||
| 11–16 years | 55 (28.6%) | 97 (29.7%) | 0.77 (0.49, 1.19) | 0.24 | 0.91 (0.56, 1.46) | 0.68 | ||
| ≥17 years | 60 (31.2%) | 125 (38.2%) | 0.65 (0.42, 0.99) | 0.046 | 0.92 (0.55, 1.53) | 0.75 | ||
| Karyotype (n, %) | ||||||||
| 45,X | 78 (40.6%) | 116 (35.6%) | 1.00 | 1.00 | ||||
| 45,X/46,XX; 45,X/47,XXX | 26 (13.5%) | 75 (23.0%) | 0.52 (0.30, 0.87) | 0.01 | 0.65 (0.37, 1.14) | 0.14 | ||
| 45,X/46,XY | 7 (3.6%) | 22 (6.7%) | 0.47 (0.19, 1.16) | 0.10 | 0.62 (0.24, 1.60) | 0.33 | ||
| Ring X; Isochromosome X; Autosomal Translocation; other | 59 (30.7%) | 83 (25.5%) | 1.06 (0.68, 1.64) | 0.80 | 1.38 (0.85, 2.23) | 0.18 | ||
| Deletion on the X chromosome | 13 (6.8%) | 11 (3.4%) | 1.76 (0.75, 4.12) | 0.19 | 1.88 (0.77, 4.57) | 0.16 | ||
| Site (n, %) | ||||||||
| A (Northeast) | 27 (14.1%) | 96 (29.4%) | 1.00 | 1.00 | ||||
| B (Midwest) | 62 (32.3%) | 84 (25.8%) | 2.62 (1.53, 4.50) | <0.001 | 2.77 (1.58, 4.86) | <0.001 | ||
| C (West) | 103 (53.6%) | 146 (44.8%) | 2.51 (1.53, 4.12) | <0.001 | 2.49 (1.48, 4.19) | <0.001 | ||
| Race (n, %) | ||||||||
| Non-White | 42 (21.9%) | 50 (15.3%) | 1.00 | |||||
| White | 138 (71.9%) | 241 (73.9%) | 1.47 (0.93, 2.32) | 0.14 | ||||
| Congenital heart disease (n, %) | ||||||||
| No | 139 (72.4%) | 256 (78.5%) | 1.00 | |||||
| Yes | 53 (27.6%) | 70 (21.5%) | 1.39 (0.92, 2.11) | 0.11 | ||||
| Insurance type (n, %) | ||||||||
| Private/commercial | 109 (56.8%) | 215 (66.0%) | 1.00 | 1.00 | ||||
| Medicaid | 81 (42.2%) | 107 (32.8%) | 1.49 (1.03, 2.16) | <0.001 | 1.30 (0.88,1.93) | 0.19 | ||
| Final height (n, %) | ||||||||
| Not reached yet | 134 (69.8%) | 210 (64.4%) | 1.35 (0.85, 2.13) | 0.20 | ||||
| <150 cm | 23 (12.0%) | 42 (12.9%) | 1.16 (0.60, 2.21) | 0.66 | ||||
| ≥ 150 cm | 35 (18.2%) | 74 (22.7%) | 1.00 | |||||
| Menarche (n, %) * | ||||||||
| No menarche | 114 (59.4%) | 163 (50.0%) | 1.00 | |||||
| Yes, spontaneous | 27 (14.1%) | 47 (14.4%) | 0.95 (0.51, 1.79) | 0.88 | ||||
| Yes, not spontaneous | 35 (18.2%) | 76 (23.3%) | 0.77 (0.42, 1.41) | 0.39 | ||||
| Thelarche (n, %) * | ||||||||
| No thelarche | 84 (43.8%) | 123 (37.7%) | 1.00 | |||||
| Yes, spontaneous | 43 (22.4%) | 82 (25.2%) | 1.12 (0.61, 2.06) | 0.71 | ||||
| Yes, not spontaneous | 50 (26.0%) | 81 (24.8%) | 1.36 (0.73, 2.53) | 0.33 | ||||
| Age growth hormone started‡ (n, %) * | ||||||||
| ≥9 years | 25 (13.0%) | 90 (27.6%) | 1.00 | 1.00 | ||||
| 6–8 years | 28 (14.6%) | 43 (13.2%) | 2.26 (1.17, 4.36) | 0.02 | 2.48 (1.25, 4.89) | 0.009 | ||
| 0–5 years | 77 (40.1%) | 75 (23.0%) | 3.45 (1.97, 6.03) | <0.001 | 3.18 (1.78, 5.65) | <0.001 | ||
| Not ever started | 58 (30.2%) | 106 (32.5%) | 1.75 (0.98, 3.14) | 0.06 | 1.81 (0.99, 3.33) | 0.054 | ||
| Age sex hormone replacement therapy initiated (n, %) * | ||||||||
| Not ever on sex hormone replacement therapy | 107 (55.7%) | 164 (50.3%) | 1.00 | |||||
| <13 years | 32 (16.7%) | 48 (14.7%) | 1.33 (0.74, 2.38) | 0.34 | ||||
| 13–15 years | 26 (13.5%) | 61 (18.7%) | 0.88 (0.48, 1.63) | 0.68 | ||||
| 16+ years | 21 (10.9%) | 35 (10.7%) | 1.32(0.63, 2.73) | 0.46 | ||||
For some variables in this retrospective chart review, ns do not add up to 100% due to missing data
In multivariate model, age of diagnosis and age of growth hormone started were entered in the model separately because the two variables are highly correlated (r=0.82, P<0.001)
Univariate analyses for menarche, thelarche, age of growth hormone and age of sex steroid replacement therapy initiation were adjusted for age at last visit.
Independent factors associated with a higher risk for NC included TS diagnosis between 0–5 years of age, Medicaid insurance, and younger age at growth hormone initiation (after controlling for age at last visit, Table 3). Mosaic karyotypes were associated with lower odds of documented NC compared to individuals with non-mosaic 45,X and complex karyotypes. In the multivariate analysis, site, age at diagnosis, and age at growth hormone initiation remained significant.
Mental Health Concerns
MHC concerns were documented for 20.4% of the cohort (Table 4). Anxiety was the most common mental health concern (17.4%), followed by depression (6.7%) (Table 2). Site of care was associated with odds of documented MHC; patients at Site B had 3.4 times greater odds, and those at Site C had twice greater odds of having documented concerns as compared to those seen at Site A in the northeast (Table 4).
Table 4.
Association of Mental Health Concerns (MHC) with Patient Characteristics†
| MHC (Yes) | MHC (No) | Univariate | Multivariate | |||||
|---|---|---|---|---|---|---|---|---|
| Variables | n=129 | n=326 | OR (95%CI) | P value | OR (95%CI) | P value | ||
| Age of diagnosis‡ (n, %) | ||||||||
| ≥6 years | 55 (42.6%) | 125 (38.3%) | 1.00 | |||||
| 0–5 years | 40 (31.0%) | 111 (34.0%) | 0.82 (0.51, 1.33) | 0.42 | ||||
| Prenatal | 23 (17.8%) | 71 (21.8%) | 0.74 (0.42, 1.30) | 0.29 | ||||
| Age at last visit (n, %) | ||||||||
| 0–10 years | 25 (19/4%) | 104 (31.9%) | 1.00 | 1.00 | ||||
| 11–16 years | 47 (36.4%) | 97 (29.8%) | 2.02 (1.15, 3.52) | 0.01 | 1.78 (0.89, 3.56) | 0.10 | ||
| ≥17 years | 57 (44.2%) | 125 (38.3%) | 1.90 (1.11, 3.25) | 0.02 | 1.33 (0.61, 2.90) | 0.48 | ||
| Karyotype (n, %) | ||||||||
| 45,X | 51 (39.5%) | 116 (35.6%) | 1.00 | 1.00 | ||||
| 45,X/46,XX; 45,X/47,XXX | 24 (18.6%) | 75 (23.0%) | 0.73 (0.41, 1.28) | 0.27 | 0.46 (0.22, 0.99) | 0.046 | ||
| 45,X/46,XY | 4 (3.1%) | 22 (6.7%) | 0.41 (0.14, 1.26) | 0.12 | 0.40 (0.13, 1.28) | 0.12 | ||
| Ring X; Isochromosome X; Autosomal Translocation; other | 32 (24.8%) | 83 (25.5%) | 0.88 (0.52, 1.48) | 0.62 | 0.75 (0.42, 1.35) | 0.34 | ||
| Deletion on the X chromosome | 12 (9.3%) | 11 (3.4%) | 2.48 (1.03, 5.99) | 0.04 | 1.97 (0.76, 5.12) | 0.16 | ||
| Site (n, %) | ||||||||
| A (Northeast) | 20 (15.5%) | 96 (29.4%) | 1.00 | 1.00 | ||||
| B (Midwest) | 49 (38.0%) | 84 (25.8%) | 2.80 (1.54, 5.09) | 0.001 | 3.37 (1.77, 6.41) | <0.001 | ||
| C (West) | 60 (46.5%) | 146 (44.8%) | 1.97 (1.12, 3.48) | 0.02 | 2.03 (1.12, 3.70) | 0.02 | ||
| Race (n, %) | ||||||||
| Non-White | 21(16.3%) | 50 (15.3%) | 1.00 | |||||
| White | 101 (78.3%) | 241 (73.9%) | 1.00 (0.57, 1.76) | 0.99 | ||||
| Congenital heart disease (n, %) | ||||||||
| No | 108 (83.7%) | 256 (78.5%) | 1.00 | |||||
| Yes | 21 (16.3%) | 70 (21.5%) | 0.71 (0.42, 1.22) | 0.21 | ||||
| Insurance type (n, %) | ||||||||
| Non-Medicaid | 85 (65.9%) | 215 (66.0%) | 1.00 | |||||
| Medicaid | 42 (32.5%) | 107 (32.8%) | 1.27 (0.23, 7.03) | 0.97 | ||||
| Final height (n, %) | ||||||||
| Not reached yet | 68 (52.7%) | 210 (64.4%) | 0.75 (0.46, 1.23) | 0.25 | ||||
| <150cm | 29 (22.5%) | 42 (12.9%) | 1.60 (0.85, 3.00) | 0.14 | ||||
| ≥ 150cm | 32 (24.8%) | 74 (22.7%) | 1.00 | |||||
| Menarche (n, %) * | ||||||||
| No menarche | 52 (40.3%) | 163 (50.0%) | 1.00 | 1.00 | ||||
| Yes, spontaneous | 30 (23.2%) | 47 (14.4%) | 1.70 (0.89, 3.23) | 0.11 | 2.44 (1.00, 5.91) | 0.049 | ||
| Yes, not spontaneous | 31 (24.0%) | 76 (23.3%) | 1.10 (0.58, 2.06) | 0.77 | 0.75 (0.36, 1.58) | 0.45 | ||
| Thelarche (n, %) * | ||||||||
| No thelarche | 31 (24.0%) | 123 (37.7%) | 1.00 | 1.00 | ||||
| Yes, spontaneous | 41 (31.8%) | 82 (25.2%) | 1.76 (0.89, 3.49) | 0.10 | 1.15 (0.47, 2.80) | 0.76 | ||
| Yes, not spontaneous | 43 (33.3%) | 81 (24.8%) | 1.88 (0.93, 3.80) | 0.08 | 1.64 (0.65, 4.12) | 0.29 | ||
| Age of growth hormone started (n, %) * | ||||||||
| ≥9 years | 37 (28.7%) | 90 (27.6%) | 1.00 | |||||
| 6–8 years | 23 (17.8%) | 43 (13.2%) | 1.38 (0.73, 2.62) | 0.33 | ||||
| 0–5 years | 35 (27.1%) | 75 (23.0%) | 1.28 (0.73, 2.25) | 0.39 | ||||
| Not ever started | 28 (21.7%) | 106 (32.5%) | 0.80 (0.44, 1.46) | 0.47 | ||||
| Age sex hormone replacement therapy initiated (n, %) * | ||||||||
| Not ever on sex hormone replacement therapy | 46 (35.7%) | 164 (50.3%) | 1.00 | 1.00 | ||||
| <13 years | 22 (17.1%) | 48 (14.7%) | 1.38 (0.71, 2.66) | 0.34 | 1.35 (0.59, 3.13) | 0.48 | ||
| 13–15 years | 29 (22.5%) | 61 (18.7%) | 1.51 (0.81, 2.82) | 0.20 | 1.27 (0.58, 2.80) | 0.55 | ||
| 16+ years | 22 (17.1%) | 35 (10.7%) | 2.22 (1.06, 4.66) | 0.03 | 1.52 (0.63, 3.66) | 0.35 | ||
For some variables in this retrospective chart review, ns do not add up to 100% due to missing data
In multivariate model, age of diagnosis and age of growth hormone started were entered in the model separately because the two variables are highly correlated (r=0.82, P<0.001)
Univariate analyses for menarche, thelarche, age of growth hormone and age of sex steroid replacement initiation were adjusted for age at last visit.
Several variables were significantly associated with increased risk of MHC in multivariate analysis, including older age at last visit; spontaneous menarche; and older age at hormone replacement therapy initiation. Because many of these variables were independently associated with older patient age in our cohort, an analysis controlling for age at last visit was performed. Only sex steroid replacement at age >16 years remained significant, associated with a twofold greater risk of having documented MHC (Table 4).
Co-occurring Neuropsychological and Mental Health Concerns
As shown in Table 5, 60 patients were documented to have both NC and MHC. Site of care was again associated with differences in documentation of both NC and MHC; the odds of patients at Site B (Midwest) and Site C (West) having both documented MHC and NC were 7.0 and 5.5 times those of patients at Site A (Northeast). In analysis controlling for age at last visit, growth hormone initiation at younger ages was associated with greater odds of having both NC and MHC. Those patients in whom growth hormone was initiated at 0–5 years had 3.45 times the odds of documented NC and MHC as compared to youth who started growth hormone older than 9 years (Table 5).
Table 5.
Patient Characteristics Associated with Both Neuropsychological and Mental Health Concerns†
| MHC and NC (Yes) | MHC and NC (No) | Univariate | Multivariate | |||||
|---|---|---|---|---|---|---|---|---|
| Variables | n=60 | n=326 | OR (95%CI) | P value | OR (95%CI) | P value | ||
| Age of diagnosis‡ (n, %) | ||||||||
| ≥6 years | 20 (33.3%) | 125 (38.4%) | 1.00 | |||||
| 0–5 years | 25 (41.7%) | 111 (34.0%) | 1.41 (0.74, 2.67) | 0.30 | ||||
| Prenatal | 10 (16/7%) | 71 (21.8%) | 1.65 (0.55, 4.90) | 0.37 | ||||
| Age at last visit (n, %) | ||||||||
| 0–10 years | 12 (20.0%) | 104 (31.9%) | 1.00 | |||||
| 11–16 years | 22 (36.7%) | 97 (29.8%) | 1.97 (0.92, 4.19) | 0.08 | ||||
| ≥17 years | 26 (43.3%) | 125 (38.3%) | 1.80 (0.87, 3.75) | 0.11 | ||||
| Karyotype (n, %) | ||||||||
| 45,X | 28 (46.7%) | 116 (35.6%) | 1.00 | |||||
| 45,X/46,XX; 45,X/47,XXX | 11 (18.3%) | 75 (23.0%) | 0.61 (0.29, 1.29) | 0.20 | ||||
| 45,X/46,XY | 1 (1.7%) | 22 (6.7%) | 0.19 (0.02, 1.46) | 0.11 | ||||
| Ring X; Isochromosome X; Autosomal Translocation; other | 12 (20.0%) | 83 (25.5%) | 0.60 (0.29, 1.25) | 0.17 | ||||
| Deletion on the X chromosome | 6 (10.0%) | 11 (3.4%) | 2.26 (0.77, 6.63) | 0.14 | ||||
| Site (n, %) | ||||||||
| Site A (Northeast) | 4 (6.7%) | 96 (29.4%) | 1.00 | 1.00 | ||||
| Site B (Midwest) | 26 (43.3%) | 84 (25.8%) | 7.43 (2.49, 22.15) | <0.001 | 6.98 (2.28, 21.32) | 0.001 | ||
| Site C (West) | 30 (50.0%) | 146 (44.8%) | 4.93 (1.68, 14.44) | 0.004 | 5.47 (1.82, 16.44) | 0.003 | ||
| Race (n, %) | ||||||||
| Non-White | 12 (20.0%) | 50 (15.3%) | 1.00 | |||||
| White | 46 (76.7%) | 241 (73.9%) | 1.26 (0.62, 2.54) | 0.52 | ||||
| Congenital heart disease (n, %) | ||||||||
| No | 49 (81.7%) | 256 (78.5%) | 1.00 | |||||
| Yes | 11 (18.3%) | 70 (21.5%) | 0.81 (0.41, 1.66) | 0.58 | ||||
| Insurance type (n, %) | ||||||||
| Non-Medicaid | 42 (70.0%) | 215 (66.0%) | 1.00 | |||||
| Medicaid | 18 (30.0%) | 107 (32.8%) | 0.86 (0.47, 1.57) | 0.62 | ||||
| Final height (n, %) | ||||||||
| Not reached yet | 35 (58.3%) | 210 (64.4%) | 0.77 (0.40, 1.47) | 0.43 | ||||
| <150 cm | 9 (15.0%) | 42 (12.9%) | 1.99 (0.40, 2.44) | 0.98 | ||||
| ≥150cm | 16 (26.7%) | 74 (22.7%) | 1.00 | |||||
| Menarche (n, %) * | ||||||||
| No menarche | 26 (43.3%) | 163 (50.0%) | 1.00 | |||||
| Yes, spontaneous | 13 (21.7%) | 47 (14.4%) | 1.44 (0.60, 3.48) | 0.42 | ||||
| Yes, not spontaneous | 16 (26.7%) | 76 (23.3%) | 1.12 (0.48, 2.61) | 0.80 | ||||
| Thelarche (n, %) * | ||||||||
| No thelarche | 18 (30.0%) | 123 (37.7%) | 1.00 | |||||
| Yes, spontaneous | 17 (28.3%) | 82 (25.2%) | 1.01 (0.89, 3.49) | 0.98 | ||||
| Yes, not spontaneous | 19 (31.7%) | 81 (24.8%) | 1.15 (0.44, 2.98) | 0.78 | ||||
| Age of growth hormone started (n, %) *, ‡ | ||||||||
| ≥9 years | 9 (15.0%) | 90 (27.6%) | 1.00 | |||||
| 6–8 years | 12 (20.0%) | 43 (13.2%) | 2.98 (1.16, 7.66) | 0.02 | 3.72 (1.37, 10.07) | 0.01 | ||
| 0–5 years | 22 (36.7%) | 75 (23.0%) | 3.45 (1.47, 8.06) | 0.004 | 3.20 (1.32, 7.74) | 0.01 | ||
| Not ever started | 14 (23.3%) | 106 (32.5%) | 1.76 (0.70, 4.33) | 0.23 | 1.60 (0.62, 4.15) | 0.33 | ||
| Age sex hormone replacement therapy initiated (n, %) * | ||||||||
| Not ever on sex hormone replacement therapy | 26 (75.0%) | 164 (50.3%) | 1.00 | 1.00 | ||||
| <13 years | 11 (18.3%) | 48 (14.7%) | 1.39 (0.58, 3.34) | 0.46 | 1.44 (0.62, 3.38) | 0.40 | ||
| 13–15 years | 10 (16.7%) | 61 (18.7%) | 1.06 (0.43, 2.60) | 0.90 | 1.31 (0.56, 3.10) | 0.53 | ||
| 16+ years | 12 (20.0%) | 35 (10.7%) | 2.45 (0.92, 6.51) | 0.07 | 3.08 (1.29, 7.36) | 0.01 | ||
For some variables in this retrospective chart review, ns do not add up to 100% due to missing data
In multivariate model, age of diagnosis and age of growth hormone started were entered in the model separately because the two variables are highly correlated (r=0.82, P<0.001)
Univariate analyses for menarche, thelarche, age of growth hormone and age of sex steroid replacement initiation were adjusted for age at last visit.
Referral and Receipt of Neuropsychological and/or Mental Health Services
Neuropsychological evaluation was recommended for 33.9% of patients of whom 65.4% had documentation indicating a completed neuropsychological evaluation. Similar rates of referral (35%) and utilization (69.7%) of mental health services were documented for our cohort. There were notable differences between clinical sites in referral practices for neuropsychological evaluation and mental health evaluation (Table 2). Individuals receiving care at Site C (West) were more likely overall to have recommendations for neuropsychological evaluations and mental health services. Furthermore, those who were seen in the multidisciplinary clinic within Site C were the most likely to have neuropsychological evaluation and mental health services recommended. Patients with TS seen in the Site C multidisciplinary clinic were more likely to have NC or MHC documented in their medical records when compared to those seen at Site C who were not seen in the multidisciplinary clinic at the time charts were abstracted. Individuals receiving care at Site A had a greater likelihood of mental health services being received when recommended and individuals receiving care at Site B had a greater likelihood of completing recommended neuropsychological evaluation. There was no significant difference in the likelihood of accessing mental health services based on having been seen in a multidisciplinary clinic or not.
DISCUSSION
In this large, retrospective study across three geographically distinct regions in the U.S., we observed that NC were documented in 30% and MHC in 20% of pediatric patients with TS. This represents a higher prevalence of NC among patients with TS compared to general population estimates (Bitsko et al., 2022; Zablotsky & Alford, 2020), but a similar prevalence of MHC (Bitsko et al., 2022). Nearly 10% of youth with TS were documented to have both NC and MHC. Discrepancies between population estimates of NC and MHC and the prevalence in our cohort may be due in part to methodological differences; NC and MHC in this study were ascertained via retrospective chart review, reliant on provider documentation, whereas general population studies utilized prospective symptom assessments. While several variables were associated with NC and/or MHC, the most notable difference was in rates of documented NC and MHC by site. In addition, differences in referral patterns and completion of evaluations were noted between sites, reflecting substantial variability in clinical care practices despite international consensus guidelines (Gravholt et al., 2017).
Neuropsychological Concerns in Turner Syndrome
The most common NC documented in our population was ADHD, which has previously been reported to be more common among individuals with TS (Björlin Avdic et al., 2021; H. F. Russell et al., 2006). The prevalence of ADHD was lower in our TS sample than rates reported in prospective studies (24–51%). However, these studies often capture individuals with ADHD symptoms such as impaired attention who may not have been formally diagnosed with ADHD (Green et al., 2015; Russell et al., 2006). Despite these variable estimates, our results support that ADHD is more prevalent among individuals with TS than expected in the general pediatric population, and more research is needed to understand the pathophysiology, clinical features, and optimal management of ADHD symptomatology among youth with TS.
Risk factors for NC included diagnosis of TS occurring in early childhood, early growth hormone initiation and non-mosaic/complex karyotype. These associations likely reflect the increased severity of clinical presentation in those diagnosed with TS in early childhood (vs. incidental identification prenatally or more mild clinical presentations not diagnosed until later childhood) which may result in earlier initiation of growth hormone treatment, as well as the increased severity of phenotypic manifestations in individuals with non-mosaic/complex karyotype. Individuals with TS caused by ring chromosome have been described to have a higher risk of developmental delay and learning disability (Mauger, 2018). Increased risk of neurocognitive deficits has been ascribed to deletions of Xp, due in part to PAR1 haploinsufficiency. The presence of a cell line with two intact sex chromosomes (e.g., 45,X/46,XX, 45,X/47,XXX, or 45,X/46,XY) may be protective, given the presence of intact PAR1 on the X chromosome, or the homologous region on the Y chromosome (Ross et al., 2000). Additional research is needed on the specific genotype contributions to neuropsychological phenotype in TS to further elucidate the biological underpinnings of these deficits that could potentially be targeted with interventions.
Lower odds of NC in individuals with private insurance could reflect the effect of higher socio-economic status. Low socio-economic status is thought to influence risk of some neuropsychological diagnoses, such as ADHD, in part through increased rates of stress and adversity (Goodman et al., 2003; Russell et al., 2016).
Mental Health Concerns in Turner Syndrome
Anxiety and depression were the most common mental health diagnoses in our population, findings that are consistent with previous work (Alexandrou et al., 2022; Lesniak-karpiak, Mazzocco, & Ross, 2003). However, we observed a lower documented prevalence of MHC in this retrospective pediatric study relative to prospective studies which have documented rates of depression in up to 62.5% of adult patients with TS (Cardoso et al., 2004; Schmidt, 2006). Nevertheless, research on depression in TS youth is limited, and robust studies remain a high priority for future research.
There are several potential explanations for the low rates of MHC documented in our population of patients with TS. First, the chart review design limits the ability to capture those with symptoms that are not documented in the medical record. Second, lower prevalence of MHC in our sample may be related to the young age of our cohort (mean age at last visit: 13.4 years), particularly because most existing estimates of mood disorder prevalence in patients with TS have focused on adults (Morris et al., 2020). In patients with TS, as in the population as a whole, MHC often become clinically apparent in adolescence and young adulthood (Bitsko et al.,2022). Additionally, challenges related to neurocognitive and physical differences and associated impacts on daily functioning may become more pronounced with maturity as academic, social, and independence expectations increase. Our cohort may have been young enough that clinically significant MHC had not yet become evident to either patients, caregivers, or care providers. A third possible explanation for lower rates of MHC in our population may be insufficient recognition of the mental health needs of patients with TS. Patients experiencing internalizing symptoms such as anxiety and/or depressive symptoms may not be recognized due to inadequate screening, recognition of symptoms during brief medical appointments, and/or referral to mental health services. We previously demonstrated that most pediatric endocrinologists do not routinely screen for MHC, with less than half of their patients routinely receiving MHC screening, primarily due to perceived barriers such as time constraints, lack of comfort with screening, and systemic problems in mental health care (Davis et al., 2021).
Our finding that MHC were twofold more likely in patients with spontaneous menarche and sex hormone replacement at older ages (16+ years) may suggest that, in contrast to NC, MHC do not seem to be associated with more severe TS phenotype. In fact, a milder TS phenotype may contribute to anxiety and depression symptomatology. It is possible that patients with milder TS phenotypes, and in particular, those with fewer neurocognitive deficits, may be more aware of their TS-related physical differences relative to peers, which in turn may contribute to negative impacts on social and emotional functioning and accompanying mental health symptoms.
In contrast to our findings, prior work has suggested a contribution of ovarian insufficiency and/or estrogen deficiency to increased depressive symptoms (Schmidt et al., 2011), which may be partially remediable with initiation of estrogen therapy (Ross et al., 1996). Clearly, further empirical research on the genetic and environmental underpinnings of MHC in TS is greatly needed.
Co-occurring Neuropsychological and Mental Health Concerns in Turner Syndrome
Medical care differences (e.g., initiation of growth hormone) for individuals diagnosed with both NC and MHC could be a product of differences in care practices by site. As discussed above, documentation of referral for neuropsychological testing and mental health services varied by site, as did clinical care approaches for TS. It is possible that psychosocial complexity driven by co-existing NC and MHC could lead to delays in accessing care or difficulty adhering to medical recommendations. For example, for individuals with significant developmental delays or mental health concerns (i.e., anxiety), caregivers may struggle to routinely administer some medical treatments, such as daily injections of growth hormone for short stature). As a result, such treatments may be consistently provided later than is recommended by clinical practice guidelines for these patients.
Lack of clear associations between demographic and medical characteristics and co-occurring NC and MHC could be due in part to the relatively small number of patients who had documentation of both NCH and MHC. This may also reflect delays in diagnosis of TS for patients who are more psychosocially complex. However, further investigation is needed to better understand risk factors for both NC and MHC.
Practice Patterns
Though neuropsychological evaluations are recommended for all individuals with TS in clinical practice guidelines, providers caring for patients with TS are not universally documenting referrals for these services or the outcomes of existing referrals. Barriers to referral for neuropsychological evaluation and mental health services may include discomfort with screening for NC and/or MHC, lack of time to devote to screening for these concerns during the clinic visit, and limited resources for these services in some communities (Davis et al., 2021). The increased frequency of referral for neuropsychological evaluations and mental health services at Site C, the only site with a multidisciplinary clinic specifically for TS including a psychologist and neuropsychologist, suggests that embedding personnel qualified to administer appropriate screening measures may increase referral rates for neuropsychological evaluations. It is also possible that the collaboration across providers from multiple specialties in the context of a multidisciplinary team increases awareness of MHC and NC and establishes systems for screening and referral. Low rates of mental health service recommendations at all sites indicate an area for improvement of clinical care for patients with TS.
Our finding that, in contrast to NC, MHC are not associated with more severe phenotypes, suggests that these patients are an at-risk population who may benefit from targeted screening and relatedly, referrals for mental health interventions. Specifically, clinicians may wish to utilize therapeutic models that address some of the social factors contributing to MHC in youth with TS, including focused support for navigating school and peer relationships.
This study did not assess what is driving low rates of medical provider-recommended mental health and neuropsychological services. Possible contributing factors include deficits in screening practices or non-standardized screening practices; differences in provider discipline and/or education about NC and MHC in TS; a perception that screening is out of the scope of practice of a medical provider; limits on clinical time enabling adequate screening; differences in the approach to referral; and/or lack of availability of resources in the practice environment. Our previous work has suggested a potential approach to addressing some of these barriers by developing and validating a TS-specific tool to screen for NC and MHC, and some of this work is underway (Davis et al., 2021; Wolfe, Hutaff-Lee, & Wilkening, 2021).
Based on our findings, there may be a significant advantage for youth with TS to be seen in a dedicated multidisciplinary clinic, in line with current recommendations (Gravholt, et al., 2017). However, even though patients seen in the multidisciplinary clinic at Site C were more likely to have neuropsychological diagnoses recommended, there were not significant differences in completion of neuropsychological evaluations. Across all sites in our large, multisite cohort, only approximately 20% of patients had completed neuropsychological evaluations, a finding suggesting that screening for NC and recommendations for care are not the only limitations to accessing these services for patients with TS. Future studies are needed to better understand how other barriers— including resource limitations such as long wait times, limited numbers of specialists who perform evaluations, and lack of interest in/perceived benefits of neuropsychological evaluations among patients and guardians— affect completion rates.
These findings and differences among sites in documentation and referral for NC and MHC point to the need for standardization of routine neuropsychological and mental health screening/referral practices for patients with TS. Clinical practice guidelines (Gravholt, et al., 2017) recommend routine neuropsychological screening and assessment for all girls with TS but make limited reference to the optimal age and approach to conducting screenings and assessments. The need for practical guidance around implementing these recommendations is particularly marked considering the limited numbers of pediatric neuropsychologists and mental health specialists in the U.S., of whom few have and training or expertise in TS. Even in the context of routine neuropsychological evaluation, little is known about the frequency with which patients’ access recommended services, a question addressed by the current study. Still, further work is needed to identify the specific barriers to pursuing neuropsychological evaluation and/or mental health services in this population, and to understand the feasibility of implementing comprehensive neuropsychological evaluations in all patients with TS across a variety of clinical environments.
Limitations
This study was a retrospective chart review, with data extracted based on provider documentation in the medical record. As such, there are likely errors of omission, particularly given that providers may not document NC and MHC. Furthermore, there are likely variations in documentation and referral practices across providers and institutions. As some patients may have accessed neuropsychology or mental health care outside of the hospital systems in which medical records were reviewed, we may not have accurately captured our patients’ referrals for MHC or NC by others (e.g., primary care physicians; school counselors), or the utilization of neuropsychology and mental health care outside of the hospital setting.
Conclusions
NC and MHC are frequently observed in patients with TS, but despite this few have completed neuropsychological evaluations and/or therapy. In this study, markers of a more severe TS phenotype, such as diagnosis in early childhood, were associated with increased risk for documented NC. In contrast, MHC were associated with later age of estrogen replacement, which may suggest that social factors, such as perceived difference from peers, play a role in anxiety and depression. Pediatric specialists who care for patients with TS—including geneticists and endocrinologists— should consider screening for NC and MHC, guided by disease-associated risk factors. While screening efforts may be limited by provider comfort with screening and logistical constraints (e.g., time), specialists can incorporate questions about mood, behavior, and developmental and/or school functioning into their visits and refer to mental health clinicians and/or neuropsychologists as indicated. Similarly, brief, free, and easily administered instruments, such as the Patient-Reported Outcomes Measurement Information System (PROMIS®) have been piloted with some success in TS populations (Alexandrou et al., 2022) and could be integrated into practice. In our study, referral practices for neuropsychological and mental health services varied substantially across sites, further highlighting the need for standardized screening and referral practices for this at-risk population.
Acknowledgements:
We wish to thank Maddy McClinchie for her help with manuscript formatting and editing.
Funding:
This study was supported in part by NICHD K23HD092588 (awarded to SD).
Footnotes
Disclosures: The authors declare no conflicts of interest.
DATA SHARING:
Data is stored in a RedCap Database which can be made available on request.
REFERENCES
- Alexandrou E, Corathers S, Gutmark-Little I, Casnellie L, Gerstle M, Tatum J, Khoury J, & Backeljauw P (2022). Improving anxiety screening in patients with Turner syndrome. Hormone Research in Pediatrics, 95(1), 68–75. 10.1159/000524169 [DOI] [PubMed] [Google Scholar]
- Bannink EM, Raat H, Mulder PG, & de Muinck Keizer-Schrama SM (2006). Quality of life after growth hormone therapy and induced puberty in women with Turner syndrome. The Journal of Pediatrics, 148(1), 95–101. 10.1016/j.jpeds.2005.08.043 [DOI] [PubMed] [Google Scholar]
- Bitsko JH, Claussen AH, Lichstein J, Black LI, Jones SE, Danielson ML, Hoenig JM, Davis Jack SP, Brody DJ, Gyawaki SG, Maenner MJ, Warner M, Holland KM, Perou R, Crosby AE, Blumberg SJ, Avenevoli S, Kaminski JW, & Ghandour RM (2022). Mental health surveillance among children — United States, 2013–2019 (72(2), 1–42). MMWR Supplement. 10.15585/mmwr.su7102a1 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Björlin Avdic H, Butwicka A, Nordenström A, Almqvist C, Nordenskjöld A, Engberg H, & Frisén L (2021). Neurodevelopmental and psychiatric disorders in females with Turner syndrome: A population-based study. Journal of Neurodevelopmental Disorders, 13(1). 10.1186/s11689-021-09399-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Cardoso G, Daly RJ, Haq NA, Hanton L, Rubinow DR, Bondy CA, & Schmidt PJ (2004). Current and lifetime psychiatric illness in women with Turner syndrome. Gynecological Endocrinology, 19(6), 313–319. 10.1080/09513590400021227 [DOI] [PubMed] [Google Scholar]
- Davis S, Crerand C, Hutaff-Lee C, Thompson T, Tishelman A, Samara O, Umbaugh H, Nahata L, & Kremen J (2020). Neurodevelopmental and mental health screening for patients with Turner syndrome in pediatric endocrine clinics: Results of a pediatric endocrine society survey. Hormone Research in Pediatrics, 93(11–12), 643–650. 10.1159/000516126 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Delooz J, van den Berghe H, Swillen A, Kleczkowska A, & Fryns JP (1993). Turner syndrome patients as adults: A study of their cognitive profile, psychosocial functioning and psychopathological findings. Journal of Genetic Counseling, 4(3), 169–179. [PubMed] [Google Scholar]
- el Abd S, Turk J, & Hill P (1995). Psychological characteristics of Turner syndrome. Journal of Child Psychology and Psychiatry, 36, 1109–1125. [DOI] [PubMed] [Google Scholar]
- Goodman E, Huang B, Wade TJ, & Kahn RS (2003). A multilevel analysis of the relation of socioeconomic status to adolescent depressive symptoms: Does school context matter? The Journal of Pediatrics, 143(4), 451–456. 10.1067/s0022-3476(03)00456-6 [DOI] [PubMed] [Google Scholar]
- Gravholt CH, Andersen NH, Conway GS, Dekkers OM, Geffner ME, Klein KO, Lin AE, Mauras N, Quigley CA, Rubin K, Sandberg DE, Sas TCJ, Silberbach M, Söderström-Anttila V, Stochholm K, van Alfen-van derVelden JA, Woelfle J, Backeljauw PF, & International Turner Syndrome Consensus Group. (2017). Clinical practice guidelines for the care of girls and women with Turner syndrome: Proceedings from the 2016 Cincinnati International Turner syndrome meeting. European Journal of Endocrinology, 177(3), G1–G70. 10.1530/eje-17-0430 [DOI] [PubMed] [Google Scholar]
- Green T, Bade Shrestha S, Chromik LC, Rutledge K, Pennington BF, Hong DS, & Reiss AL (2015). Elucidating X chromosome influences on attention deficit hyperactivity disorder and executive function. Journal of Psychiatric Research, 68, 217–225. 10.1016/j.jpsychires.2015.06.021 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Hagman A, Loft A, Wennerholm UB, Pinborg A, Bergh C, Aittomäki K, Nygren KG, Bente Romundstad L, Hazekamp J, & Söderström-Anttila V (2013). Obstetric and neonatal outcome after oocyte donation in 106 women with Turner syndrome: a Nordic cohort study. Human reproduction (Oxford, England), 28(6), 1598–1609. 10.1093/humrep/det082 [DOI] [PubMed] [Google Scholar]
- Hong D, Scaletta Kent J, & Kesler S (2009). Cognitive profile of Turner syndrome. Developmental Disabilities Research Reviews, 15(4), 270–278. 10.1002/ddrr.79 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Hutaff-Lee C, Bennett E, Howell S, & Tartaglia N (2019). Clinical developmental, neuropsychological, and social–emotional features of Turner syndrome. Child: Care, Health and Development, 181(1), 126–134. 10.1002/ajmg.c.31687 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Jeż W, Tobiasz – Adamczyk B, Brzyski P, Majkowicz M, Pankiewicz P, & Irzyniec T (2018). Social and medical determinants of quality of life and life satisfaction in women with Turner syndrome. Advances in Clinical and Experimental Medicine, 27(2), 229–236. 10.17219/acem/66986 [DOI] [PubMed] [Google Scholar]
- Kostev K, Teichgräber F, Konrad M, & Jacob L (2019). Association between chronic somatic conditions and depression in children and adolescents: A retrospective study of 13,326 patients. Journal of Affective Disorders, 245, 697–701. 10.1016/j.jad.2018.11.014 [DOI] [PubMed] [Google Scholar]
- Krantz E, Landin-Wilhelmsen K, Trimpou P, Bryman I, & Wide U (2019). Health-related quality of life in Turner syndrome and the influence of growth hormone therapy: A 20-Year Follow-Up. The Journal of Clinical Endocrinology &Metabolism, 104(11), 5073–5083. 10.1210/jc.2019-00340 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Lesniak-Karpiak K, Mazzocco MMM, & Ross JL (2003). Behavioral assessment of social anxiety in females with Turner or fragile X syndrome. Journal of Autism and Developmental Disorders, 33(1), 55–67. 10.1023/a:1022230504787 [DOI] [PubMed] [Google Scholar]
- Maenner MJ, Blumberg SJ, Kogan MD, Christensen D, Yeargin-Allsopp M, & Schieve LA (2016). Prevalence of cerebral palsy and intellectual disability among children identified in two U.S. national surveys, 2011–2013. Annals of Epidemiology, 176(3), 139–148. 10.1016/j.annepidem.2016.01.001 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Mauger C, Lancelot C, Roy A, Coutant R, Cantisano N, & le Gall D (2018). Executive functions in children and adolescents with Turner syndrome: A systematic review and Meta-Analysis. Neuropsychology Review, 28(2), 188–215. 10.1007/s11065-018-9372-x [DOI] [PubMed] [Google Scholar]
- Messina M, Zirilli G, Civa R, Rulli I, Salzano G, Versa TA, & Valenzise M (2007). Neurocognitive profile in Turner’s syndrome is not affected by growth impairment. Journal of Pediatric Endocrinology and Metabolism, 20(6), 677–684. 10.1515/jpem.2007.20.6.677 [DOI] [PubMed] [Google Scholar]
- Morris LA, Tishelman AC, Kremen J, & Ross RA (2019). Depression in Turner syndrome: A systematic review. Archives of Sexual Behavior, 49(2), 769–786. 10.1007/s10508-019-01549-1 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Rickert VI, Hassed SJ, Hendon AE, & Cunniff C (1996). The effects of peer ridicule on depression and self image among adolescent females with Turner syndrome. Journal of Adolescent Health, 19(1), 34–38. 10.1016/1054-139x(95)00225-h [DOI] [PubMed] [Google Scholar]
- Ross JL, McCauley E, Roeltgen D, Long L, Kushner H, Feuillan P, & Cutler GB Jr (1996). Self-concept and behavior in adolescent girls with Turner syndrome: Potential estrogen effects. The Journal of Clinical Endocrinology and Metabolism, 81(3), 926–931. 10.1210/jc.81.3.926 [DOI] [PubMed] [Google Scholar]
- Russell AE, Ford T, Williams R, & Russell G (2015). The association between socioeconomic disadvantage and attention deficit/hyperactivity disorder (ADHD): A systematic review. Child Psychiatry & Human Development, 47(3), 440–458. 10.1007/s10578-015-0578-3 [DOI] [PubMed] [Google Scholar]
- Russell HF, Wallis D, Mazzocco MMM, Moshang T, Zackai E, Zinn AR, Ross JL, & Muenke M (2006). Increased prevalence of ADHD in Turner syndrome with no evidence of imprinting effects. Journal of Pediatric Psychology, 31(9), 945–955. 10.1093/jpepsy/jsj106 [DOI] [PubMed] [Google Scholar]
- Schmidt PJ, Cardoso GMP, Ross JL, Haq N, Rubinow DR, & Bondy CA (2006). Shyness, social anxiety, and impaired self-esteem in Turner syndrome and premature ovarian failure. JAMA, 295(12), 1374–1376. 10.1001/jama.295.12.1374 [DOI] [PubMed] [Google Scholar]
- Schmidt PJ, Luff JA, Haq NA, Vanderhoof VH, Koziol DE, Calis KA, Rubinow DR, & Nelson LM (2011). Depression in women with spontaneous 46, XX primary ovarian insufficiency. The Journal of Clinical Endocrinology and Metabolism, 96(2), E278–E287. 10.1210/jc.2010-0613 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Sutton EJ, McInerney-Leo A, Bondy CA, Gollust SE, King D, & Biesecker B (2005). Turner syndrome: four challenges across the lifespan. American Journal of Medical Genetics. Part A, 139A(2), 57–66. 10.1002/ajmg.a.30911 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Sutton EJ, Young J, McInerney-Leo A, Bondy CA, Gollust SE, & Biesecker BB (2006). Truth-telling and Turner Syndrome: the importance of diagnostic disclosure. The Journal of Pediatrics, 148(1), 102–107. 10.1016/j.jpeds.2005.08.022 [DOI] [PubMed] [Google Scholar]
- van den Hoven AT, Bons LR, Dykgraaf R, Dessens AB, Pastoor H, de Graaff L, Metselaar MR, Kneppers-Swets A, Kardys I, Mijnarends H, Zweerus F, Hazelzet JA, Utens E, van den Bosch AE, & Roos-Hesselink JW (2020). A value-based healthcare approach: Health-related quality of life and psychosocial functioning in women with Turner syndrome. Clinical Endocrinology, 92(5), 434–442. 10.1111/cen.14166 [DOI] [PMC free article] [PubMed] [Google Scholar]
- van Hagen IM, Duijnhouwer AL, Ten Kate-Booij MJ, Dykgraaf RH, Duvekot JJ, Utens EM, & Roos-Hesselink JW (2017). Wish to conceive and concerns to develop cardiovascular complications during pregnancy in patients with Turner syndrome. Journal of Psychosomatic Obstetrics and Gynaecology, 38(1), 45–52. 10.1080/0167482X.2016.1216961 [DOI] [PubMed] [Google Scholar]
- Vijayakanthi N, Marcus DJ, Fritz SP, Xiang Y, & Fadoju D (2022). Body Image, Self-perception, and Satisfaction Among Girls With Turner Syndrome: A Prospective Cross-sectional Study. The Journal of Clinical Endocrinology and Metabolism, 107(4), e1382–e1389. 10.1210/clinem/dgab889 [DOI] [PubMed] [Google Scholar]
- Wasserman HM, Kalkwarf HJ, Altaye M, Yolton K, Kanj RV, & Gordon CM (2022). Characterizing the Adolescent Premature Ovarian Insufficiency Phenotype: A Case Control Study. Journal of Pediatric and Adolescent Gynecology, S1083–3188(22)00298–4. Advance online publication. 10.1016/j.jpag.2022.10.003 [DOI] [PubMed] [Google Scholar]
- Wolfe KR, Hutaff-Lee C, & Wilkening G (2021). Neuropsychological screening in pediatric multidisciplinary clinics: Group characteristics and predictive utility. Archives of Clinical Neuropsychology, 37(4), 789–797. 10.1093/arclin/acab090 [DOI] [PubMed] [Google Scholar]
- Wolstencroft J, Mandy W, & Skuse D (2019). Experiences of social interaction in young women with Turner syndrome: A qualitative study. Child: Care, Health and Development, 46(1), 46–55. 10.1111/cch.12710 [DOI] [PubMed] [Google Scholar]
- Zablotsky B, & Alford JM (2020). Racial and ethnic differences in the prevalence of attention-deficit/hyperactivity disorder and learning disabilities among children aged 3–17 years. NCHS Data Brief, 358, 1–8. [PubMed] [Google Scholar]