Aims: To investigate whether karyotype, mid-childhood (6–10 years) follicle-stimulating hormone (FSH) and luteinizing hormone (LH) levels, and ultrasound ovary visualization results can be used as indicators of spontaneous puberty in Turner syndrome (TS). Methods: The analysis was based on clinical and biochemical data from 110 TS girls aged >13 years at the end of the study (1,140 visits between 1996 and 2015). The study population was divided according to karyotype: 45,X and non-45,X. Results: The mean age ± standard deviation at diagnosis was 10.7 ± 4.0 years, and the follow-up duration was 5.9 ± 3.3 years. Spontaneous puberty was confirmed in 48% and menarche in 20% of the subjects, less frequently in 45,X girls. The mean age at Tanner stage B2 was 13.7 ± 2.4 years and that at menarche 14.2 ± 1.7 years, regardless of the karyotype. The median FSH level at 6–10 years was 8.16 IU/L, which was significantly lower than <6 years and >10 years. The median LH level at 6–10 years was 0.35 IU/L, which was lower than >10 years. The chance of spontaneous menarche was decreased in girls with FSH ≥6.7 IU/L between 6 and 10 years. Conclusions: Although spontaneous puberty and menarche occur more frequently in non-45,X girls, the karyotype cannot be used to predict them. However, the chance of spontaneous menarche can be predicted based on gonadotropin cut-off values. There was no correlation between ultrasound ovary visualization results and spontaneous puberty.

Turner syndrome (TS) is one of the most common genetic disorders, with an incidence of 1 per 2,500 live-born girls [1]. It is caused by complete or partial absence of one of the X chromosomes. The clinical presentation can range from mild growth deficiency with subtle phenotypic features to severe growth impairment with marked dysmorphic traits, pubertal failure, and cardiac defects. According to the literature, 20% of TS patients show some degree of pubertal development, with menarche in approximately 16% and regular menstrual cycles in 6% of cases [2]. Only 2–5% of TS patients achieve spontaneous pregnancy [3]. Thus approximately 90% of TS girls and women require or will require oestrogen replacement therapy to initiate, progress, or maintain pubertal development. It is recommended that women with TS should receive oestrogen and progestin replacement, known to have long-term effects on puberty, fertility, metabolism, and psychological functioning [4, 5]. However, patient compliance with treatment is often variable, especially during late adolescence, when transition to adult care is being planned [6]. The international consensus is to initiate oestrogen therapy at the age of 12 years in the absence of spontaneous puberty and if follicle-stimulating hormone (FSH) levels are elevated. However, recent data have also shown some positive effects of ultralow-dose oestrogen therapy in early childhood [7, 8].

Prediction of spontaneous puberty and cyclical menstruation at a prepubertal or early pubertal age would be useful for determining the potential need for oestrogen replacement and the timing of its initiation. To date, no such indicator has been established. An argument can be made for examining the karyotype, since ovarian function in some patients with TS is influenced by this. For example, the prevalence of spontaneous puberty is 54% among subjects with the 45,X/46,XX karyotype compared with 6% among those with the 45,X karyotype [9], with the former girls being 9 times more likely to achieve secondary sexual characteristics and to show preservation of ovarian function in the long term. This chance is even higher in very rarely occurring mosaicism with the 47,XXX cell line (45,X/47,XXX or 45,X/46,XX/47,XXX) [10].

It has been demonstrated that ovaries can be detected by transabdominal ultrasound in almost 40% of younger TS patients. However, it is not easy to make a precise distinction between a true ovary and a streak gonad [11].

The aim of this study was to investigate the frequency of spontaneous puberty symptoms, and to determine whether the karyotype, pelvic ultrasound data (visualization of ovaries), and in particular the gonadotropin profile at 6–10 years can be predictors of spontaneous puberty in TS patients.

The study encompassed 139 consecutive patients with TS who between 1996 and 2015 reported for follow-up visits every 3–6 months at the Department of Paediatrics, Paediatric Endocrinology and Diabetes, Medical University of Silesia, Katowice, Poland. Twenty-nine patients were excluded from the study: 20 girls were <13 years at the end of observation, 6 girls had already been treated with oestrogen at the beginning of the study, and 3 girls with the 45,X/46,XY karyotype had undergone gonadectomy due to the risk of malignant transformation. The final analysis encompassed 110 patients. In all cases, TS was diagnosed based on a cytogenetic analysis using peripheral lymphocytes, and confirmed by karyotyping with routine G-banding, including counting of at least 10 metaphases, 3 of which were fully analysed.

Throughout the study, all the patients underwent routine clinical visits during which thorough clinical examinations including pubertal staging according to the method of Tanner [12] and anthropometric measurements were performed by a single paediatric endocrinologist (A.M.G.).

The study group was divided into subsets according to karyotype and spontaneous puberty. The karyotype subset was divided into a 45,X and a non-45,X subgroup. The pubertal subset was divided into patients who remained at Tanner stage B1 and those at Tanner stage B2 or higher (subgroup B ≥2). Additionally, patients with spontaneous menarche (M1) and those without (M0) were recorded.

All blood samples were analysed in the same laboratory. FSH and luteinizing hormone (LH) were measured using a chemiluminescence immunometric assay (IMMULITE Systems). Oestradiol was measured using the electrochemiluminescence method (electrochemiluminescence immunoassay). Due to the low sensitivity of the measurement (the detection limit was 18.4 pmol/L), only the detection criterion (detected or not) was considered. Transabdominal pelvic ultrasound examinations were performed with a 5-MHz convex transducer (Siemens Acuson Antares 5.0, Acuson Sequoia, and Acuson 128 XP). The presence of ovaries was determined by two gynaecologists (A.D.-C. and K.W.). A detectable and measurable ovary with or without follicles was considered a true ovary. A detectable hypoechogenic but unmeasurable structure without follicles could not be accurately distinguished from streak gonads, and was not considered a true ovary [11].

Statistics and Data Analysis

Statistical analyses were performed with PQStat version 1.4.8 and SPSS version 20 (IBM, Armonk, NY, USA). Data are presented as means and standard deviations, medians and ranges, and percentages; unless stated otherwise, they are given in the text as mean (SD)/median (range). Comparisons between two groups were performed with two-sided Student t tests or Fisher exact tests, as appropriate. The Cochran-Cox or Mann-Whitney U test was used in the case of unequal variances or if the assumption of normal distribution was not met. For multiple comparisons, p was adjusted with the Benjamini-Hochberg correction. Dunn’s post hoc test with the Sidak correction, following Kruskal-Wallis analysis of variance, was applied to compare three groups. Cut-offs were determined by receiver operating characteristic curve analysis. p values <0.05 were considered significant.

The study was conducted in accordance with the Declaration of Helsinki. Informed consent was obtained from each patient over the age of 16 years, or a parent or a legal custodian.

Clinical Presentation

The information on the 110 patients, grouped according to karyotype, is given in Table 1. Of all these 110 patients, 43 were followed up between the age of 6 and 10 years. The mean age (SD)/median age (range) at TS diagnosis and/or at the first available blood sample was 10.7 (4.0)/11.1 (0.4–18.4) years. The duration of observation was 5.9 (3.3)/5.3 (0.5–15.0) years, with a mean number of 10 visits.

Table 1.

Karyotype distribution in the 110 girls with Turner syndrome

 Karyotype distribution in the 110 girls with Turner syndrome
 Karyotype distribution in the 110 girls with Turner syndrome

No signs of spontaneous puberty were observed in 57/110 (52%) of the girls (subgroup B1). Spontaneous puberty was observed in 53/110 (48%) of the patients (subgroup B ≥2) and was significantly less frequent in 45,X than in non-45,X girls (31 vs. 69%; p = 0.02). The age at breast development in subgroup B ≥2 was 13.7 (2.4)/13.9 (9.4–18.2) years. Breast development in the 45,X girls, if it did occur, was not significantly later than in the non-45,X girls (14.6 [9.5–17.7] vs. 13.2 [9.4–16.7] years; p = 0.67).

Spontaneous menarche, at the age of 14.20 (1.7)/14.0 (12.4–19.1) years, occurred in 22/110 (20%) of the patients. There was no significant difference in median age (range) at menarche between the 45,X and the non-45,X subgroup (12.9 [12.2–14.6] vs. 14.3 [12.6–16.9] years; p = 0.105), but menarche was more frequent among the non-45,X patients (32.6 vs. 9.8%; p = 0.02).

Spontaneous menarche occurred in 22/53 (41.5%) of the patients with spontaneous breast development (subgroup B ≥2). The median age (range) at breast development in this subgroup was 13.9 (11.6–15.2) years for the 45,X patients and 13.6 (10.4–16.0) years for the non-45,X patients.

FSH and LH Values

Between 1996 and 2015, a total of 1,140 clinical visits were performed, and samples for serum FSH, LH, and oestradiol assessment being drawn on 372, 370, and 311 occasions.

The median FSH value in the 43 patients observed between the age of 6 and 10 years was significantly lower than in the younger or the older age group (p = 0.02 and p < 0.0001, respectively). The LH values at 6–10 years were lower than >10 years (p < 0.0001) but no different from those <6 years (p = 0.88) (Table 2).

Table 2.

Median values (range) of FSH, LH, and E2 in the age subgroups of girls with Turner syndrome

 Median values (range) of FSH, LH, and E2 in the age subgroups of girls with Turner syndrome
 Median values (range) of FSH, LH, and E2 in the age subgroups of girls with Turner syndrome

The phasic pattern of gonadotropin concentration was similar in subgroups B1, B ≥2, M0, M1, 45,X, and non-45,X (Fig. 1, 2). However, after the age of 10 years, the median FSH values (range) were higher in subgroup B1 than in subgroup B ≥2 (99.0 [0–279.5] vs. 70.6 [0.36–265.0] IU/L; p < 0.001), in subgroup M0 than in subgroup M1 (98.5 [0–279.5] vs. 14.6 [0.36–143.5] IU/L; p = 0.0001), and in subgroup 45,X than in subgroup non-45,X (101.0 [0–279.5] vs. 80.5 [0.36–265.0] IU/L; p = 0.033).

Fig. 1.

FSH levels depending on karyotype (a), spontaneous breast development (b), and occurrence of menarche (c). FSH, follicle-stimulating hormone; 45,X, patients with monosomic karyotype; non-45,X, patients with miscellaneous karyotypes; B ≥2, patients with spontaneous breast development (Tanner stage 2 or higher); B1, patients without spontaneous breast development (Tanner stage 1); M0, patients without spontaneous menarche; M1, patients with spontaneous menarche; Poly., polynomial trend line order 6.

Fig. 1.

FSH levels depending on karyotype (a), spontaneous breast development (b), and occurrence of menarche (c). FSH, follicle-stimulating hormone; 45,X, patients with monosomic karyotype; non-45,X, patients with miscellaneous karyotypes; B ≥2, patients with spontaneous breast development (Tanner stage 2 or higher); B1, patients without spontaneous breast development (Tanner stage 1); M0, patients without spontaneous menarche; M1, patients with spontaneous menarche; Poly., polynomial trend line order 6.

Close modal
Fig. 2.

LH levels depending on karyotype (a), spontaneous breast development (b), and occurrence of menarche (c). LH, luteinizing hormone; 45,X, patients with monosomic karyotype; non-45,X, patients with miscellaneous karyotypes; B ≥2, patients with spontaneous breast development (Tanner stage 2 or higher); B1, patients without spontaneous breast development (Tanner stage 1); M0, patients without spontaneous menarche; M1, patients with spontaneous menarche; Poly., polynomial trend line order 6.

Fig. 2.

LH levels depending on karyotype (a), spontaneous breast development (b), and occurrence of menarche (c). LH, luteinizing hormone; 45,X, patients with monosomic karyotype; non-45,X, patients with miscellaneous karyotypes; B ≥2, patients with spontaneous breast development (Tanner stage 2 or higher); B1, patients without spontaneous breast development (Tanner stage 1); M0, patients without spontaneous menarche; M1, patients with spontaneous menarche; Poly., polynomial trend line order 6.

Close modal

Similarly, after the age of 10 years, the median LH values (range) were higher in subgroup B1 than in subgroup B ≥2 (17.2 [0–64.0] vs. 13.8 [0–137.0] IU/L; p = 0.023), in subgroup M0 than in subgroup M1 (16.9 [0–13.0] vs. 6.9 [0–82.2] IU/L; p = 0.039), and in subgroup 45,X than in subgroup non-45,X (17.3 [0–137.0] vs. 12.8 [0–46.7] IU/L; p = 0.002).

Between the ages of 6 and 10 years, the only difference that reached significance after correction was that in FSH levels, which were lower in subgroup M1 (3.6 [0.23–49.9] IU/L) than in subgroup M0 (8.9 [1.0–157.0] IU/L) (p = 0.047).

In order to determine how FSH could be of use in predicting spontaneous menarche in the future, we calculated sensitivity and specificity values for different FSH thresholds in girls aged 6–10 years (Fig. 3). The optimal threshold was 6.7 IU/L (sensitivity 62%, specificity 66%). All patients with FSH <0.76 IU/L had spontaneous menarche (100% specificity, but only 23% sensitivity). No patient with FSH >49.9 IU/L had spontaneous menarche (sensitivity 100%, but only 14% specificity).

Fig. 3.

Sensitivity (-■-■-■-■-) and specificity (-●-●-●-●-) of FSH at the age of 6–10 years for prediction of menarche. Based on this plot, 6.7 IU/L (sensitivity 62%, specificity 66%) was accepted as a reasonable cut-off, since it is the point where the curves meet and the two measures are the most balanced. The higher the value of FSH, the higher its sensitivity as a predictor of menarche but the lower its specificity; thus a cut-off of 49.9 IU/L has 100% sensitivity but 14% specificity. By contrast low FSH levels are specific but not sensitive, so that a cut-off of 0.76 IU/L has 100% specificity but only 23% sensitivity. FSH, follicle-stimulating hormone.

Fig. 3.

Sensitivity (-■-■-■-■-) and specificity (-●-●-●-●-) of FSH at the age of 6–10 years for prediction of menarche. Based on this plot, 6.7 IU/L (sensitivity 62%, specificity 66%) was accepted as a reasonable cut-off, since it is the point where the curves meet and the two measures are the most balanced. The higher the value of FSH, the higher its sensitivity as a predictor of menarche but the lower its specificity; thus a cut-off of 49.9 IU/L has 100% sensitivity but 14% specificity. By contrast low FSH levels are specific but not sensitive, so that a cut-off of 0.76 IU/L has 100% specificity but only 23% sensitivity. FSH, follicle-stimulating hormone.

Close modal

Oestradiol

Due to the low sensitivity of the oestradiol assay, the results were quantified as nominal variables above or below the detection threshold. In 43/110 (39.1%) of the study population, oestradiol was above the lower limit of quantification, and this occurred more frequently in subgroup B ≥2 (27/53 [50.9%]) than in subgroup B1 (15/57 [26.3%]; p = 0.005). It was measurable in 19/61 (31.1%) of the 45,X subgroup and in 22/49 (44.9%) of the non-45,X subgroup (p = 0.166). Detailed data on the measurable oestradiol values for the age subgroups are shown in Table 2.

Pelvic Ultrasound

Transabdominal pelvic ultrasound with assessment of ovaries was performed an average of 3 times on each patient, a total of 324 times in all the groups. In 4 cases, no results were available. The median age (range) at the first examination was 11.1 (0.4–18.3) years and at the last examination 15.3 (10.8–20.0) years. The ovary/ovaries was/were described for 63/106 (59.4%) of the study population, being seen in 34/59 (57.5%) 45,X girls versus 29/47 (61.7%) non-45,X girls (p = 0.87), and in 29/56 (51.7%) B1 girls versus 34/50 (68%) B ≥2 girls (p = 0.43) (Table 3). There was no correlation between the presence of ovaries and spontaneous puberty (Yule’s Q 0.13; p = 0.39).

Table 3.

Ultrasound visualization of ovaries in the 106 girls with Turner syndrome, grouped by the presence of pubertal symptoms and by karyotype

 Ultrasound visualization of ovaries in the 106 girls with Turner syndrome, grouped by the presence of pubertal symptoms and by karyotype
 Ultrasound visualization of ovaries in the 106 girls with Turner syndrome, grouped by the presence of pubertal symptoms and by karyotype

In the girls with ovaries detected on ultrasound examination, the median FSH and LH levels (range) at the age of 6–10 years were not significantly lower than in the girls without visualization of ovary tissues (6.81 [0.23–89.0] vs. 8.18 [2.04–157.0] IU/L, p = 0.36; 0.36 [0.10–32.0] vs. 0.28 [0.42–27.4] IU/L, p = 1).

In this longitudinal study many of the 110 girls were observed for up to 10 years. Nearly half showed spontaneous pubertal symptoms and 20% had spontaneous menarche. This is at variance with previous findings, in which spontaneous puberty occurred less often [2, 13, 14]. There is no clear explanation for our observation. The difference in results may be partly accounted for by the fact that karyotype analysis is performed more commonly than previously, even in girls with a mild phenotype. At present almost all patients with short stature have karyotyping performed, facilitating an accurate diagnosis at an earlier age.

In our study, the monosomic 45,X TS patients were twice less likely to develop spontaneous puberty and had a 3 times lower chance of spontaneous menarche than the non-45,X subgroup. Hagen et al. [9] proved that the miscellaneous TS karyotype is highly predictive of remaining ovarian function. However, cases of 45,X patients with preserved fertility do exist [15]. Therefore, a more reliable factor for prediction is needed, particularly if ultralow-dose oestrogen therapy for younger girls is being contemplated.

Our study showed that the age at onset of spontaneous puberty and menarche cannot be predicted based on the karyotype, since there was no difference between the 45,X and the non-45,X group. Therefore there is no reason to wait longer for the first symptoms of puberty in non-45,X patients than in 45,X patients.

Moreover, the period between the mean age at spontaneous breast development and the mean age at menarche was surprisingly short in our study (0.8 years). Thus, even if symptoms of spontaneous puberty occur, they are delayed and may be followed by a rapid puberty, associated with a worse final height prognosis.

Our results confirm previous reports on the biphasic pattern of gonadotropin secretion in TS, being highest in childhood, declining between 6 and 10 years, and then rising again in adolescence [16, 17]. Hagen et al. [9] claimed that the biphasic age pattern of gonadotropins is preserved in all patients, irrespective of the karyotype. Although some researchers reported normal or significantly lower prepubertal gonadotropin levels in patients with spontaneous menarche than in patients with induced menarche [2, 18], a concrete cut-off level has as yet to be established. Aso et al. [19] suggested an FSH level <10 mIU/mL at 12 years as an index of spontaneous and cyclical menstruation in TS. We found a cut-off level for serum FSH of 6.7 mIU/mL at 6–10 years to be most predictive of spontaneous menarche, with a sensitivity and specificity of 62 and 66%, respectively. This cut-off level applies to an age range hitherto regarded as being of limited use in interpreting gonadotropin levels. Our study shows that gonadotropin measurement in mid-childhood is in fact of value for predicting spontaneous puberty, which is useful in determining the potential need for oestrogen replacement and the timing of its initiation. Therefore, gonadotropin measurements should not be underestimated in any age group.

With regard to oestradiol measurement, it is impossible to conclude whether this is a reliable factor in the absence of assays with sufficient sensitivity. Although we observed some correlation between detectable concentrations of oestradiol and spontaneous puberty, more targeted research is warranted.

Since we found no correlation between spontaneous puberty and the visualization of ovarian structures, we cannot recommend using this method as a puberty indicator.

The limitation of our longitudinal study, designed 20 years ago, is the absence of measurements of inhibin B and anti-Müllerian hormone (AMH). Hagen et al. [9] found undetectable inhibin B levels in all TS patients with premature ovarian failure measured prior to pubertal onset. TS patients with primary amenorrhoea were shown to have significantly lower levels of inhibin B than TS girls with spontaneous puberty and healthy controls [20]. Inhibin B is therefore considered a screening test for assessing ovarian reserve and a longitudinal marker of the possible decline of ovarian function in TS. However, its specificity is limited, as 37% of healthy girls have undetectable inhibin B levels [21]. AMH appears to be a reliable marker in this field, and is associated with karyotype as well as with ovarian function, the lowest AMH levels being found in 45,X monosomy patients and in those with no signs of spontaneous puberty [22, 23]. AMH has been found to predict disturbance at puberty onset in young TS girls and premature ovarian failure in adolescent and adult TS patients [22‒24]. Nevertheless, none of the cited studies proved the superiority of AMH over FSH in predicting spontaneous puberty. Based on an analysis of 15 girls, Lunding et al. [24] confirmed that an AMH level below –2 SD (4 pmol/L) predicted failure to enter puberty. However, a strong negative correlation between AMH and FSH was found in a multicentre study, but due to its cross-sectional design the authors stressed the need for a longitudinal study to determine if AMH at a younger age is predictive and whether multiple measurements are needed over time [23].

Although new markers of gonadal function are being tested, it seems that the application of easily available indicators such as gonadotropins should be the prime procedure. The results of our study confirm that the assessment of gonadotropins, in particular FSH, in girls aged between 6 and 10 years is also useful for predicting spontaneous puberty. This is especially important in the context of earlier induction of puberty or the use of ultralow-dose oestrogens in younger girls with TS.

The authors wish to thank Sandra Lindon for the proofreading of this manuscript.

The authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of the study.

Statutory work at the Medical University of Silesia.

A.M. Gawlik and M. Hankus designed the study, analysed the database, and wrote the manuscript. K. Soltysik and K. Szeliga prepared and analysed the patient database. A. Drosdzol-Cop and K. Wilk participated as gynaecological consultants. E. Malecka-Tendera, A. Antosz, and A. Zachurzok collaborated in preparing the manuscript.

1.
Stochholm K, Juul S, Juel K, Naeraa RW, Gravholt CH: Prevalence, incidence, diagnostic delay, and mortality in Turner syndrome. J Clin Endocrinol Metab 2006; 91: 3897–3902.
2.
Pasquino AM, Passeri F, Pucarelli I, Segni M, Municchi G: Spontaneous pubertal development in Turner’s syndrome. Italian Study Group for Turner’s Syndrome. J Clin Endocrinol Metab 1997; 82: 1810–1813.
3.
Hovatta O: Pregnancies in women with Turner’s syndrome. Ann Med 1999; 31: 106–110.
4.
Gawlik A, Hankus M, Such K, Drosdzol-Cop A, Madej P, Borkowska M, Zachurzok A, Malecka-Tendera E: Hypogonadism and sex steroid replacement therapy in girls with Turner syndrome. J Pediatr Adolesc Gynecol 2016; 29: 542–550.
5.
Gawlik A, Malecka-Tendera E: Transitions in endocrinology: treatment of Turner’s syndrome during transition. Eur J Endocrinol 2013; 170:R57–R74.
6.
Gawlik A, Kaczor B, Kaminska H, Zachurzok-Buczynska A, Gawlik T, Malecka-Tendera E: Quality of medical follow-up of young women with Turner syndrome treated in one clinical center. Horm Res Paediatr 2012; 77: 222–228.
7.
Ross JL, Quigley CA, Cao D, Feuillan P, Kowal K, Chipman JJ, Cutler GB Jr: Growth hormone plus childhood low-dose estrogen in Turner’s syndrome. N Engl J Med 2011; 364: 1230–1242.
8.
Quigley CA, Wan X, Garg S, Kowal K, Cutler GB Jr, Ross JL: Effects of low-dose estrogen replacement during childhood on pubertal development and gonadotropin concentrations in patients with Turner syndrome: results of a randomized, double-blind, placebo-controlled clinical trial. J Clin Endocrinol Metab 2014; 99:E1754–E1764.
9.
Hagen CP, Main KM, Kjaergaard S, Juul A: FSH, LH, inhibin B and estradiol levels in Turner syndrome depend on age and karyotype: longitudinal study of 70 Turner girls with or without spontaneous puberty. Hum Reprod 2010; 25: 3134–3141.
10.
Lim HH, Kil HR, Koo SH: Incidence, puberty, and fertility in 45,X/47,XXX mosaicism: report of a patient and a literature review. Am J Med Genet A 2017, Epub ahead of print.
11.
Cleemann L, Holm K, Fallentin E, Skouby SO, Smedegaard H, Møller N, Borch-Christensen H, Jeppesen EM, Wieslander SB, Andersson AM, Cohen A, Højbjerg Gravholt C: Uterus and ovaries in girls and young women with Turner syndrome evaluated by ultrasound and magnetic resonance imaging. Clin Endocrinol (Oxf) 2011; 74: 756–761.
12.
Marshall WA, Tanner JM: Variations in pattern of pubertal changes in girls. Arch Dis Child 1969; 44: 291–303.
13.
Negreiros LP, Bolina ER, Guimarães MM: Pubertal development profile in patients with Turner syndrome. J Pediatr Endocrinol Metab 2014; 27: 845–849.
14.
Tanaka T, Igarashi Y, Ozono K, Ohyama K, Ogawa M, Osada H, Onigata K, Kanzaki S, Kohno H, Seino Y, Takahashi H, Tajima T, Tachibana K, Tanaka H, Nishi Y, Hasegawa T, Fujita K, Yorifuji T, Horikawa R, Yokoya S: Frequencies of spontaneous breast development and spontaneous menarche in Turner syndrome in Japan. Clin Pediatr Endocrinol 2015; 24: 167–173.
15.
Mortensen KH, Rohde MD, Uldbjerg N, Gravholt CH: Repeated spontaneous pregnancies in 45,X Turner syndrome. Obstet Gynecol 2010; 115(pt 2):446–449.
16.
Chrysis D, Spiliotis BE, Stene M, Cacciari E, Davenport ML: Gonadotropin secretion in girls with Turner syndrome measured by an ultrasensitive immunochemiluminometric assay. Horm Res 2006; 65: 261–266.
17.
Conte FA, Grumbach MM, Kaplan SL: A diphasic pattern of gonadotropin secretion in patients with the syndrome of gonadal dysgenesis. J Clin Endocrinol Metab 1975; 40: 670–674.
18.
Alves ST, Gallicchio CT, Guimarães MM, Santos M: Gonadotropin levels in Turner’s syndrome: correlation with breast development and hormone replacement therapy. Gynecol Endocrinol 2003; 17: 295–301.
19.
Aso K, Koto S, Higuchi A, Ariyasu D, Izawa M, MiyamotoIgaki J, Hasegawa Y: Serum FSH level below 10 mIU/mL at twelve years old is an index of spontaneous and cyclical menstruation in Turner syndrome. Endocr J 2010; 57: 909–913.
20.
Messina MF, Aversa T, Salzano G, Costanzo D, Sferlazzas C, Mirabelli S, Zirilli G, Lombardo F: Inhibin B in adolescents and young adults with Turner syndrome. J Pediatr Endocrinol Metab 2015; 28: 1209–1214.
21.
Gravholt CH, Naeraa RW, Andersson AM, Christiansen JS, Skakkebaek NE: Inhibin A and B in adolescents and young adults with Turner’s syndrome and no sign of spontaneous puberty. Hum Reprod 2002; 17: 2049–2053.
22.
Hagen CP, Aksglaede L, Sørensen K, Main KM, Boas M, Cleemann L, Holm K, Gravholt CH, Andersson AM, Pedersen AT, Petersen JH, Linneberg A, Kjaergaard S, Juul A: Serum levels of anti-Müllerian hormone as a marker of ovarian function in 926 healthy females from birth to adulthood and in 172 Turner syndrome patients. J Clin Endocrinol Metab 2010; 95: 5003–5010.
23.
Visser JA, Hokken-Koelega AC, Zandwijken GR, Limacher A, Ranke MB, Flück CE: Anti-Müllerian hormone levels in girls and adolescents with Turner syndrome are related to karyotype, pubertal development and growth hormone treatment. Hum Reprod 2013; 28: 1899–1907.
24.
Lunding SA, Aksglaede L, Anderson RA, Main KM, Juul A, Hagen CP, Pedersen AT: AMH as predictor of premature ovarian insufficiency: a longitudinal study of 120 Turner syndrome patients. J Clin Endocrinol Metab 2015; 100:E1030–E1038.