Turner syndrome: skin, liver, eyes, dental and ENT evaluation should be improved

Introduction Turner syndrome association with multi-organ system comorbidities highlights the need for effective implementation of follow-up guidelines. We aimed to assess the adequacy of care with international guidelines published in 2007 and 2017 and to describe the phenotype of patients. Methods In this multicenter retrospective descriptive cohort study, we collected growth and pubertal parameters, associated comorbidities, treatment, and karyotype in patients diagnosed at age <18 years between 1993 and 2022. We assessed age-appropriate recommendation follow-up (children, adolescents and adults) according to the 2007 guidelines if the last visit was before 2017 (18 recommendations) and the 2017 guidelines if the last visit was after 2017 (19 recommendations). Results We included 68 patients followed at Lausanne University Hospital (n=64) and at Neuchatel Regional Hospital (RHNe) (n=4). 2.9% of patients underwent all recommended investigations. Overall, 68.9 ± 22.5% and 78.5 ± 20.6% of the recommendations were followed, before and after 2017 respectively. High implementation rates were found for height, weight and BMI (100%), cardiac (80 to 100%) and renal (90 to 100%) imaging. Low implementation rates were found for Ear, Nose and Throat (ENT) (56.5%), skin (38.5%), dental (23.1%), ophthalmological (10%) and cholestasis (0 to 29%) assessments, depending on age and time of visit. In adults (n=33), the mean proportion of followed recommendations was lower before than after 2017: 63.5 ± 25.8% vs. 78.7 ± 23.4%, p=0.039. Conclusion Growth parameters, cardiac and renal imaging are well followed. However, efforts should be made for dental, ENT, ophthalmological, skin and cholestasis assessments. Adequacy of follow-up improved with the quality of transition to adult care.


Introduction
Turner syndrome (TS), caused by the complete or partial absence of one of the two X chromosomes, is the most common sex chromosome disorder in females, affecting approximately 1 in 2,000 live-born females. Comorbidities of TS may involve the endocrine system (the main clinical features are short stature, hypogonadism due to ovarian dysgenesis, and thyroid disease), the cardiovascular system (congenital heart disease, aortopathy, vasculopathy, arterial hypertension) and neuropsychocognitive development (e.g. learning difficulties). Other possible manifestations include hearing loss, orthopedic disorders (hip dysplasia, scoliosis, osteoporosis), renal and urinary tract disorders, (metabolic syndrome (hypertension, insulin-resistance, diabetes, overweight) or autoimmune disorders (hypothyroidism, celiac disease). This explains the need for screening and lifelong multidisciplinary follow-up of these patients (1). As growth and pubertal disorders are the main complaints of patients with Turner syndrome, the pediatric endocrinologist is usually the first to start the workup. The transition of care to adult specialists will most often occur at the end of puberty.
International clinical practice guidelines for TS were first published in 2007 (2) and updated in 2017, with recommendations for care across the lifespan and covering all health issues and comorbidities (3). Based on these recommendations, the pediatric and adult endocrine units of the Lausanne University Hospital developed an in-house clinical care guideline in 2011. This document was upgraded in 2019 into a mobile, electronic health (m-health) tool called the TS health transition passport (Supplementary Data Sheet 1). The aim is to support patients' understanding of their condition and improve effective transition to adult-oriented care.
The aims of our study were, first, to assess the adequacy of care according to published international recommendations in a tertiary center with pediatric and adult endocrine units (Lausanne University Hospital) and in a general hospital pediatric service (Neuchatel Regional Hospital) and, second, to describe the clinical, biological and radiological phenotype of patients.

Methods
We conducted a retrospective study on patients affected with Turner syndrome.

Patients' selection
We included patients with karyotype-confirmed TS, diagnosed before the age of 18 years and followed between 1993 and 2022 at the pediatric and adult endocrinology units of Lausanne University Hospital (CHUV) and the pediatric endocrinology unit of Neuchatel Regional Hospital (RHNe). No adult patients were followed at the RHNe. We excluded patients with less than 5% mosaic cells or with a written refusal.

Data collection
We collected clinical, radiological and biological data obtained during follow-up from medical records including karyotype, growth parameters, growth laboratory tests (IGF-1 and IGFBP-3 concentrations), growth treatment information and pubertal parameters.
Height was expressed as standard deviation (SD) using healthy female growth charts. Serum IGF-1 and IGFBP-3 were routinely assayed using IGF-1 and IGFBP-3 immunoassay kits: Nichols Institute Diagnostics from 1995 to 2005; Immulite by Siemens thereafter. Reference values were from Le Bouc Y for the Nichols Institute Diagnostics kit (4) and from Elminger et al. (5) for the Immulite kit.

Adequacy of follow-up
We defined the last visit as the last endocrinology consultation and if there was none, the last specialist consultation for patients who were no longer followed up in the pediatric unit. The last visit was the date for assessing adequacy of follow-up. We defined good adequacy of recommendation if ≥65% of patients followed it.
To assess the adequacy of follow-up, patients were divided into 2 groups: those whose last visit was before 2017 and those whose last visit was after 2017. Their follow-up was compared with the appropriate guidelines at the time, i.e. the 2007 guidelines and the 2017 guidelines, according to their age and pubertal status. The results were as follows: children, adolescents (girls with a Tanner stage S2 stage or higher -spontaneous or with estrogen therapy-and <18 years old), and adults.

Statistical analyses
Qualitative data were expressed as absolute number (percentage) and quantitative data as median (interquartile range -IQR) or as mean ± Standard Deviation (SD). We compared groups with Kruskall-Wallis, Chi 2 or Fisher's exact tests, using R statistical software. Values of p smaller than 0.05 were considered statistically significant.
This study was approved by the Local Ethics Committee (N°2 021-00229).

Results
We included 68 patients ( Figure 1). Mean age at diagnosis was 6.3 ± 5.3 (range: 0 to 16.7) years. At last visit, there were 13 children, 22 adolescents and 33 adults. Mean age at last visit was 17.6 ± 7.5 (range: 1.2 to 40.5) years. We identified 17 patients (25%) who had their last visit before 2017 and 51 (75%) patients who had their last visit after 2017 (Table 1). All children had their last visit after 2017.
In our clinic, we implemented 3 additional recommendations: IGFBP-3, thyroid US and bone age assessment, that are usually performed in children affected with growth retardation or dysthyroidism (Supplementary Table 1).

Follow-up adequacy
Overall, 2.9% (n=2/68, one child and one adult last seen after 2017) of the patients underwent all recommended investigations. The overall mean proportion of recommendations followed was 76.1 ± 21.4%: 68.9 ± 22.5% of the 2007 recommendations for patients last seen before 2017 and 78.5 ± 20.6% of the 2017 recommendations for patients last seen after 2017 (Table 1 and  Supplementary Table 1).

Adequacy of follow-up according to recommendations
The recommendations with the highest implementation rate were height, weight and BMI (100%), and cardiac (range: 80 to 100%) and renal (range: 90 to 100%) imaging. The recommendations with the lowest implementation rate were bone mineral density (in adults last seen before 2017: 30%), skin examination (in children: 38.5%), ENT (in adolescents last seen before 2017: 57% and in adults last seen after 2017: 56.5%), ophthalmological (in adolescents and adults last seen before 2017: respectively: 0% and 10%, and children: 61.5%) and dental consultations for the whole cohort (Table 1). Liver function biomarkers were often not assayed, especially ASAT and ALAT for adolescents last seen before 2017 (43%), and gGT and ALP for the whole cohort (Table 1).
We found a difference for HbA1C between adolescents and adults last seen before 2017 (100% vs. 40% p=0.035) and for fertility counseling among all patients last seen after 2017 (38.5% of children, 93.3% of adolescents and 91.3% of adults, p<0.001).

Adequacy of follow-up according to age
In children (n=13), the overall mean proportion of recommendations followed was 75.5 ± 19.1% (Table 1). We found good adequacy (i.e., ≥65% of recommendations followed) for 11/16 (69%) of the recommendations in children. In adolescents (n=22), we found no difference between overall followed recommendations for patients last seen before 2017 compared to patients last seen after 2017 (76.6 ± 15.1% vs. 80.9 ± 18.3%, p=0.306). We found good adequacy for 9/15 (60%) of the recommendations for adolescents last seen before 2017 and 14/17 (82%) of the recommendations for adolescents last seen after 2017. In contrast, in adults (n=33), the mean proportion of overall followed recommendations was lower before than after 2017: 63.5 ± 25.8% vs. 78.7 ± 23.4%, respectively, p=0.039. We found good adequacy for 8/16 (50%) of the recommendations for adults last seen before 2017 and for 13/17 (76%) of the recommendations for adults last seen after 2017.
All children and adolescents had cardiac imaging, whereas 20% and 13% of adults last seen before and after 2017 respectively did not have any cardiac imaging; the difference was not significant. Patient flow chart. *Age at last visit: £ Girls with a Tanner S2 stage or above (spontaneously or with estrogen therapy) and aged below 18 years old.
Height and height velocity at diagnosis were significantly different across the groups (p=0.031 and p=0.019, respectively) ( Table 2). Nevertheless, final height was similar across the groups (p=0.106).
Among the 86.8% of patients who had GH treatment, height and height velocity at the start of GH were significantly different across the groups (p=0.017 and 0.006 respectively) ( Table 3). One year after the start of GH treatment, height velocity was no longer different across the groups (p=0.971) while the height remained significantly different (p=0.021). Bone age delay of more than 1 year was found in 23/46 (50%) of patients at the start of GH treatment and in 12/46 (26.1%) of patients at the end of GH. In these 12 patients, GH therapy was discontinued despite the bone delay, because of patients' willingness.

Discussion
In this cohort of children, adolescents and adults patients with TS, a small minority of patients had a complete follow-up according to the international guidelines. However, the most important and potentially serious comorbidities were well followed-up, especially growth parameters, cardiac assessment and renal ultrasound. Adequacy of follow-up improved with quality of transition to adult care.
Four studies evaluated the adequacy of care according to international guidelines: three compared the adequacy of followup with the 2007 guidelines, and one with the 2017 guidelines (6)(7)(8)(9). Two studies from France and Poland found that less than 5% of adult patients received all the medical investigations recommended  3 The p-values were calculated across groups: "Children", "Adolescents" and "Adults" with the Fisher's Exact Test; 4 1 patient > 10 Y; 5 10 children and 11 adolescents with GH treatment, 6 To do when suggestive symptoms 7 2 adolescents and 10 adults with dysthyroidsm, 8 Considered to be present in all adults in our cohort, 9 if normal anatomy, otherwise according to the opinion of the cardiologist.
in the 2007 guidelines. They showed that liver enzymes were often not assayed (6,7). The prevalence of liver disease is higher in adults with TS, especially with elevated gGT rather than transaminases (10). Reported complications include non-alcoholic steatohepatitis, hepatic architectural changes such as cirrhosis, and biliary lesions such as sclerosing cholangitis (1,11). An American study evaluated the medical care of girls with TS compared to the 2007 guidelines. In our study, adherence to recommendations was higher for followup of lipid levels, liver enzymes, blood glucose, thyroid function, ENT assessments, fertility counselling, celiac screening and bone mineral density and cardiac imaging, depending on the study (6-8). However, celiac screening, ENT and ophthalmological assessments were lower than in the study published by Hoag and colleagues (9). This suggests an improvement in the management of patients with TS. There is still room for improvement in care coordination. The implementation of new models of care coordination could help (12). We focused on more recommendations from the 2007 and 2017 guidelines than these studies, especially dental, eye and skin examinations. Lack of compliance with these follow-up visits can lead to reduced quality of life. This highlights the importance of awareness among clinicians. Dermatological screening aims to detect lymphedema, dermatitis, eczema, psoriasis and multiple pigmented nevi (1, 13). Eye disorders include high rates of strabismus, visual impairment such as myopia, or sightthreatening abnormalities such as papilledema (14). The time interval recommended by the 2007 guidelines for seeing an ophthalmologist was very restrictive. This might explain the poor follow-up adequacy of ophthalmic consultations for patients with last visit before 2017. Dental disorders include a wide range of manifestations from micrognathia to abnormal dental development (1). The fact that dental consultations are not covered by the Swiss National Health Insurance may explain the low number of dental consultations in our cohort. The distribution of karyotypes in our cohort showed slightly higher proportion of X structural rearrangement than previously published (15). The comorbidities and their distribution between the different karyotypes in our cohort were globally consistent with the literature (1,(16)(17)(18). In our study, we found lower proportions of heart and liver diseases but similar proportions of thyroid and celiac diseases (15). Nevertheless, the number of comorbidities should be correlated with the adherence to recommended follow-up.
All recommendations, except HbA1c for patients last seen before 2017 and fertility counselling for patients last seen after 2017, were monitored equally between children, adolescents and adults. This may reflect the structured transition clinic between pediatric and adult endocrinological care that we have developed at the CHUV. This transition endocrine clinic has improved, as suggested by a better follow-up of adults last seen before 2017 compared to those last seen after 2017. As all RHNe patients are children or adolescents, they did not have transition. Our group has shown the usefulness of integrating transition passports as a usable, understandable health tool for patients and physicians, to reduce gaps in transition from pediatric to adult-oriented care (19). Our study suggests that the CHUV pediatric and adult TS transition passport could provide patients with a better understanding of their follow-up, of treatment, fertility care and comorbidities.
GH treatment followed the 2017 guidelines in terms of starting age, doses and IGF-1 monitoring. Estrogen treatment was started at low doses and increased over 2 to 3 years as recommended. However, the majority of patients started later than recommended. This discrepancy may be explained by the fact that only five patients started treatment after the publication of the 2017 guidelines (3). Late initiation of estrogen therapy can be detrimental to bone and uterine health (20). Data are consistent with no change in adult height when low-doses estrogen is started before the age of 12, as recommended.
The strengths of our study include the comparison of long-term medical follow-up of girls and women with TS between the two published international guidelines. Studies on the adequacy of medical follow-up in girls and women with TS are limited (6)(7)(8)(9), and most have compared the adequacy of follow-up with the 2007 guideline. We divided our cohort into 3 different age groups, as in the study by Hoag and colleagues (9). We also included additional important recommendations such as dental and ophthalmological Quantitative data are expressed as median and interquartile range (IQR). SD, Standard Deviation. *The p-values were calculated across 3 groups: "Complete Monosomy X", "45,X Mosaicism" and "X Structural Rearrangement" with the Kruskal-Wallis test.
consultations and skin examinations, which are rarely investigated, and we could perform a karyotype in all patients. This study looked at patients with TS followed at a large university center (CHUV) with many experienced specialists, including cardiologists, radiologists, ear, nose and throat specialists, adult endocrinologists, and others. As a result, follow-up of these patients may not be the same throughout Switzerland. Moreover, in other studies, most adult patients with TS were followed up by general practitioners, who were sometimes unaware of the TS diagnosis. For this reason, we developed a patient oriented electronic health (m-health) TS health transition passport, to avoid loss of medical information (Supplementary Data Sheet 1). A limitation of our study is its retrospective design, which is justified by the rarity of the disease. However, we minimized missing data through in-depth analysis of medical records.
Our study highlights the importance of improving awareness among patients themselves and primary care physicians of the broad spectrum and variability of TS presentation at different ages. We should aim to reduce health inequalities by making multidisciplinary clinics and comprehensive care available and accessible. It is also important to ensure adequate medical and social support for transition of young adults and care of adults with TS. We also should involve the patient, who gains autonomy and responsibility for her health care during adolescence and young adulthood. Our TS education program, launched in 2011 and improved in 2019, aims to address these challenges. We aim to serve as a regional resource for the community and for physicians in our community.
In conclusion, complete guideline adherence in TS patient care and follow-up should be improved, especially in bone mineral density, liver, ophthalmic, ENT, dermatological and dental assessment. Our results open a field for possible future research on patient education and healthcare organizations: how to understand the lack of awareness in TS, how to improve structural problems, how to implement a complete work-up or how to spread these guidelines among non-endocrinologists, especially pediatricians and general practitioners. A better followup has already been observed compared to the studies before 2017, which makes us optimistic for the future.  One year after start of GH  Quantitative data are expressed as median and interquartile range (IQR). AMH, Anti-Müllerian Hormone; E2, Oestrogen; FSH, Follicle Stimulating Hormone; LH, Luteinizing hormone. *These p-values were calculated across 3 groups "Complete Monosomy X", "45,X Mosaicism" and "X Structural Rearrangement" with the Kruskal-Wallis test. **These p-values were calculated across 2 groups "45,X Mosaicism" and "X Structural Rearrangement". AMH was detectable in one patient in the "Complete Monosomy X" group. Liver disease** n (%) 3 (4.4%) 1 (4.2%) 0 (0%) 2 (8.3%) 0 (0%) 0.771 *Bone disease referred to osteopenia, Leŕi-Weill dyschondrosteosis and cartilaginous protrusion of the ribs. **Liver diseases were steatohepatitis with hepatomegaly for a 16 years and a 20 years old patient and inflammatory liver disease for a 30 years old patient. ***The p-values were calculated across 3 groups "Complete Monosomy X", "45,X Mosaicism" and "X Structural Rearrangement" with the Chi 2 test or with Fisher's exact test when Chi 2 test was not applicable.

Data availability statement
The original contributions presented in the study are included in the article/Supplementary Material, further inquiries can be directed to the corresponding author/s.

Ethics statement
The studies involving human participants were reviewed and approved by Commission cantonale (VD) d'ethique de la recherche sur l'etre humain (CER-VD). Written informed consent from the participants' legal guardian/next of kin was not required to participate in this study in accordance with the national legislation and the institutional requirements.

Author contributions
JL collected the data, performed the data analysis and interpretation, performed the statistical analysis, wrote, and critically revised the manuscript. KB designed the study, was responsible for the data analysis and interpretation, and for the statistical analysis, wrote, and critically revised the manuscript. NV and MH interpreted the data and critically revised the manuscript. SS-V, MA, TB, IR, NS, TR, SF, NP, MH, and KB followed the patients, PB was responsible for hormonal assay, FB was responsible for genetic investigation. AG provided feedback on the study design. All authors contributed to the article and approved the submitted version.

Funding
Open access funding by University of Lausanne.