Abstract
Context
Congenital hypogonadotropic hypogonadism/Kallmann syndrome (CHH/KS) is a rare condition characterized by gonadotropin deficiency and pubertal failure. Adult height (AH) in patients with CHH/KS has not been well studied.
Objective
To assess AH in a large cohort of patients with CHH/KS.
Patients
A total of 219 patients (165 males, 54 females). Parents and siblings were included.
Methods
AH was assessed in patients and family members. AH was compared to the general French population, mid parental target height (TH) and between patients and same-sex siblings. Delta height (∆H) was considered as the difference between AH and parental TH. ∆H was compared between patients and siblings, normosmic CHH and KS (CHH with anosmia/hyposmia), and according to underlying genetic defect. We examined the correlations between ∆H and age at diagnosis and therapeutically induced individual statural gain.
Results
Mean AH in men and women with CHH/KS was greater than that in the French general population. Patients of both sexes had AH > TH. Males with CHH/KS were significantly, albeit moderately, taller than their brothers. ∆H was higher in CHH/KS compared to unaffected siblings (+6.2 ± 7.2 cm vs +3.4 ± 5.2 cm, P < 0.0001). ∆H was positively correlated with age at diagnosis. Neither olfactory function (normosmic CHH vs KS) nor specific genetic cause impacted ∆H. Individual growth during replacement therapy inversely correlated with the age at initiation of hormonal treatment (P < 0.0001).
Conclusions
CHH/KS is associated with higher AH compared to the general population and mid-parental TH. Greater height in CHH/KS than siblings indicates that those differences are in part independent of an intergenerational effect.
Introduction
Congenital hypogonadotropic hypogonadism (CHH) with normal olfactory function (normosmic CHH) or with olfactory impairment (Kallmann syndrome, KS) is a rare cause of disrupted puberty and infertility caused by deficient secretion (or action) of gonadotropin-releasing hormone (GnRH) (1, 2). The estimated prevalence is approximately 1/4000 and 1/10 000 respectively (1, 3). Clinically, CHH/KS is characterized by absent (or incomplete) pubertal development. In males, testosterone deficiency is evidenced by undervirilization while females typically present with primary amenorrhea and variable breast underdevelopment in the setting of estradiol deficiency (1). Accordingly, the vast majority of cases are not diagnosed until late adolescence or early adulthood (1, 4, 5). Diagnostic delay and thus initiation of replacement therapy have negative psychological consequences in both genders (1, 2, 4, 5).
Puberty in both sexes is associated with accelerated skeletal growth and bone maturation. Long bone growth ends when the hyaline cartilage comprising the epiphyseal plates (so-called ‘growth plates’) fuse (6). There is a paucity of data on pubertal growth and adult height of men and women born with CHH/KS (7, 8, 9). A study of 36 patients suggests that the sex steroid deficiency associated with CHH/KS neither accelerates nor delays growth (9). Only one study has examined the effect of sex steroid treatment on height parameters in this population (7). Moreover, intergenerational effects (i.e. how height changes in successive generations) is yet to be taken into account in this patient population (10, 11). To date, much of the literature on growth and height in endocrine disorders of puberty has focused on growth hormone deficiency/multiple pituitary hormone deficiency and Turner syndrome (12, 13).
This study aims to assess adult height in a sizable cohort of patients with CHH/KS and compare height with the general population, mid-parental target height and same sex siblings. Secondarily, we aimed to examine adult height according to clinical presentation (i.e. normosmic CHH vs KS) and identified rare genetic variants (mutations) underlying CHH/KS. In addition, we examine the effects of treatment (i.e. sex steroid replacement) on stature.
Patients and methods
This study was conducted at a single large tertiary academic medical center (Department of Endocrinology and Reproductive Diseases, Bicêtre University Hospital, France). The National Agency of Medical Safety reviewed the study protocol and provided ethics approval (RCB 2017-A01584-49). All participants provided written informed consent in accordance with French Bioethics Law and the principles of the Declaration of Helsinki.
Patients
Patients with CHH/KS were enrolled in the study. Period of recruitment of the study population ranged from 1975 to 2018. A definitive diagnosis of isolated GnRH deficiency (CHH/KS) was based on the following (1, 2): (1) absent or incomplete puberty; (2) low serum sex steroid levels (males: testosterone <1.5 ng/mL, females: estradiol (E2), <35 pg/mL) in the setting of low or inappropriately normal (<5.5 IU/L) serum gonadotropin levels (luteinizing hormone (LH, <5.6 IU/L), follicle-stimulating hormone (FSH, <7.5 IU/L)); (3) otherwise normal anterior pituitary function evidenced by normal basal (>80 ng/mL)/stimulated (>180 ng/mL) cortisol, normal basal prolactin (>8.0 and <20 ng/mL) and normal thyroid function (free T4: 12.0–22 pmol/L, TSH 0.8–4.3 mIU/L) and normal stimulated growth hormone (GH peak >10 ng/mL) either before or after 2005 via insulin tolerance test (14); (4) normal serum DHEAS (males: >640 ng/mL, females: >412 ng/mL), iron/ferritin concentrations and age-adjusted insulin-like growth factor-1 (IGF-1) levels (before 1995: >172 ng/mL after 2005 >198 ng/mL) (15, 16, 17); (5) normal MRI of the hypothalamic–pituitary region; and (6) no functional causes of hypogonadotropic hypogonadism (i.e. eating disorders, low BMI, malabsorption, chronic diseases). In cases when patients were diagnosed with CHH/KS between 13 and 17 years of age, the diagnosis was confirmed with repeat hormone profiling following a treatment washout after age 18 (total testosterone <1.5 ng/mL, LH <5.6 IU/L and FSH <7.6 IU/L in males, and persistent amenorrhea associated with estradiol <35 pg/mL, LH <6.5 IU/L and FSH <8.5 IU/L in females after a 3-month withdrawal of previous sex steroid replacement therapy). Patients were considered normosmic CHH or having Kallmann syndrome (CHH + anosmia or hyposmia) based on olfactometry testing and cranial imaging (olfactory bulb hypoplasia/aplasia) as previously described (18, 19). In patients in whom CHH or KS was diagnosed between 13 and 17 years, the definitive diagnosis was made at 18 years in order to exclude a constitutional delay of growth and puberty, and after a 3-month withdrawal of a sex steroid replacement therapy. After washout, all male CHH/KS patients had total serum testosterone levels below 1.5 ng/mL, and all CHH/KS females had serum estradiol levels below 35 pg/mL associated with persistent amenorrhea.
CHH/KS patients without available parental height measures were excluded from the study. We also excluded affected parents (i.e., with nCHH or KS) from the study.
Anthropometric measures
Height (in cm) was measured using a wall-mounted stadiometer at first visit and before introduction of hormonal replacement therapy. Patients were defined as reaching adult height (AH) by two or more stable height measurements (<1.5 cm difference between two measurements performed at least 6 months apart). Whenever available, parents and siblings were measured using a wall-mounted stadiometer. When not present in hospital consultation, in a subgroup we asked for parent and siblings’ heights by telephone survey. As a reference population, we used publicly available data on the general French population (1980–2010) - French National Institute of Statistics (http://www.insee.fr/) and French National Institute of Health (http://invs.santepubliquefrance.fr/). We used the French National Nutritional and Health Study (ENNS, CNIL no. 905481; Paris Cochin CPP no. 2264), funded by the French National Agency of Public Health (Agence nationale Santé publique France, formerly Institut de veille sanitaire (InVS)). The survey encompassed a sample from the French general population by using a three-stage sampling: (1) municipalities or groups of municipalities (n = 190) stratified by urban unit size and eight major regions; (2) households from phone lists (landlines, red lists, mobile phones and unbundled); (3) one individual per household, adult or child, using the birthday method, see also https://epidemiologie-france.aviesan.fr). We used in our study the latest database published in July 7, 2011. This study included 3115 healthy adults from French general population aged 18–74. Data are consistent with less recent figures from other French National Health public databases (https://cress-umr1153.fr/index.php/courbes-carnet-de-sante/; https://www.epsilon.insee.fr/jspui/bitstream/1/19240/1/estat_1981_132_3.pdf).
Mid parental target height (TH) for each patient was determined (in cm) using Tanner’s formulas (male TH = (father’s height + mother’s height + 13 cm) × 0.5, female TH = (father’s height + mother’s height − 13 cm) × 0.5) (20). Delta height (∆H) was calculated as the difference (in cm) between adult and mid parental target height (AH − TH = ∆H) (21). A ∆H equal to zero indicates adult and target height are equivalent. A positive ∆H value indicates that AH is superior to TH, while negative ∆H values indicate that AH is inferior to TH. Because delta height (∆H = AH − TH) were not statistically different in male and female (see Results section) we were able to compare ∆H data of male CHH/KS from siblings of both genders. The individual growth during replacement therapy was calculated as the difference between the AH and the basal height measured prior to initiating hormonal therapy (i.e. sex steroid replacement).
Genetic analyses
Following informed consent, genomic DNA was extracted from peripheral blood leukocytes using standard protocols, and direct sequencing was performed using traditional Sanger sequencing. From a total of 215 patients with available genetic analysis we found mutations in the coding regions of the following genes underlying CHH/KS: ANOS1 (formerly KAL1, n = 20/215, 9.3%), FGFR1 (n = 36/215, 16.8%), PROK2 (n = 5/215, 2.3%), PROKR2 (n = 13/215, 6.1%), CHD7 (n = 8/215, 3.7%), SEMA3A (n = 11/215, 5.1%), GNRH1 (n = 2/215, 0.9%), GNRHR (n = 4/215, 1.9%), KISS1 (n = 0/215, 0%), KISS1R (n = 7/215, 3.3%), TAC3 (n = 2/215, 0.9%), TACR3 (n = 2/215, 0.9%) and WDR11 (n = 0/215, 0%). These genes were sequenced as previously reported (22, 23, 24, 25, 26, 27, 28, 29). We did not find any deleterious genetic mutation in 103 cases out of 215.
Statistical analyses
Patient and sibling statural parameters were compared using Student t-tests. Chi-square or Fisher’s exact tests were employed for categorical comparisons between groups as appropriate. We used linear regression to examine the relationship between ∆H and age at diagnosis/start of hormonal substitution and therapeutically induced growth respectively. Sample size for paired tests was calculated with Sample Size Calculator (https://clincalc.com/stats/samplesize.aspx) and Statulator Beta (http://statulator.com/SampleSize/ss2PM.html). By analyzing previous pertinent literature, a number of 34 pairs are needed to detect a statistic significant difference (alpha 0.05, power 0.8). For comparison of height measures according to genetic variants, we performed Kruskal–Wallis and ANOVA tests. Graphics and statistical analyses were conducted using GraphPad Prism, version 5.0f (GraphPad Software, Inc.). Data are reported as individual values or using mean and s.d. unless otherwise noted. A P value less than 0.05 was considered statistically significant.
Results
A total of 219 patients (165 males, 54 females) were included in the study (Table 1). Of these patients, 56 males and 16 females were initially diagnosed between 13 and 17 years of age – all of whom had CHH/KS confirmed after age 18 years. Cases were equally divided between normosmic CHH (n = 109) and Kallmann syndrome (n = 110). A total of 172/219 (79%) of patients had siblings.
Clinical and biochemical characteristics of patients with CHH/KS at diagnosis.
| Male CHH/KS (n = 165) | Normal range in males | Female CHH/KS (n = 54) | Normal range in females | |
|---|---|---|---|---|
| Age at diagnosis (years) | 17.5 ± 6.0 | – | 16 ± 3.3 | – |
| Age at treatment initiation (years) | 17.8 ± 6.2 | – | 16.3 ± 3.2 | – |
| Weight (kg) | 76 ± 23 | – | 63 ± 12 | – |
| BMI (kg/m2) | 24.8 ± 6 | 20–25 | 23 ± 4 | 21–25 |
| Testicular volume (mL) | 3.3 ± 2.4 | >12 | – | – |
| Ovarian volume (mL) | – | – | 1.8 ± 1.5 | >2.6 to <10 |
| Normosmic CHH/KS (n) | 76/91 | – | 33/19 | – |
| Hormonal profile | ||||
| Total testosterone (ng/mL) | 0.48 ± 0.7 | 3.4–8.5 | 0.27 ± 0.1 | 0.17–0.59 |
| Total estradiol (pg/mL) | 9.1 ± 8.9 | 10–38 | 14 ± 11 | 13–97 |
| LH (IU/L) | 0.8 ± 1.1 | 1.6–5.6 | 1.6 ± 1.9 | 2.2–8.2 |
| FSH (IU/L) | 1.0 ± 1.2 | 2.2–7.4 | 2.5 ± 2.1 | 2.4–7.9 |
| IGF-1 (ng/mL) | 251 ± 105 | 198–397* | 252 ± 112 | 170–367* |
| Prolactin (ng/mL) | 7.2 ± 4.3 | 9–20 | 9.0 ± 6.1 | 9.0–21 |
| TSH (mIU/L) | 2.2 ± 1.7 | 0.8–4.3 | 1.9 ± 1.1 | 1.1–3.9 |
| FT4 (pmol/L) | 15 ± 3.4 | 12–22 | 14.6 ± 2.4 | 12.5–22.4 |
| Cortisol (8:00 h) (ng/mL) | 192 ± 76 | 94–214 | 173 ± 71 | 89–225 |
| DHEAS (ng/mL) | 1766 ± 929 | 643–3731 | 1323 ± 725 | 416–3685 |
Data are reported as means ± s.d.
*Normative values for circulating IGF1 are from previous published work (15, 16, 17).
BMI, body mass index; DHEAS, dehydroepiandrosterone sulfate; FSH, follicle-stimulating hormone; FT4, free thyroxine; IGF-1, insulin-like growth factor 1; LH, luteinizing hormone; n, number of patients; TSH, thyroid-stimulating hormone.
Median adult height (AH) in male and female CHH/KS patients was above the median height reported in the French reference population (Figs 1A and 2A). Parental AH of the patients did not differ from the reference population – neither 1980 nor 2010 (Supplementary Fig. 1, see section on supplementary materials given at the end of this article).
Adult and mid-parental target height in CHH/KS males and their brothers. (Panel A) Individual values (n = 165) and mean ± s.e. of male CHH/KS adult height versus mid-parental target height. Dark grey zone shows the 2.5 to the 97.5 centile of the French male general population. Light grey indicates the min and max limits of the French general population. Dotted line indicates median. General French population data are from the public database of the French National Institute of Health, http://invs.santepubliquefrance.fr/. Mid-parental target height has been calculated by means of the parental height according to Tanner formula (20). ****P < 0.0001. (Panel B) Comparison between the adult height in a subgroup of CHH/KS males and that of their unaffected brothers (n = 77). Individual values and mean ± s.e. are shown. *P < 0.05.
Citation: European Journal of Endocrinology 182, 2; 10.1530/EJE-19-0537
Adult and mid-parental target height in CHH/KS males and their brothers. (Panel A) Individual values (n = 165) and mean ± s.e. of male CHH/KS adult height versus mid-parental target height. Dark grey zone shows the 2.5 to the 97.5 centile of the French male general population. Light grey indicates the min and max limits of the French general population. Dotted line indicates median. General French population data are from the public database of the French National Institute of Health, http://invs.santepubliquefrance.fr/. Mid-parental target height has been calculated by means of the parental height according to Tanner formula (20). ****P < 0.0001. (Panel B) Comparison between the adult height in a subgroup of CHH/KS males and that of their unaffected brothers (n = 77). Individual values and mean ± s.e. are shown. *P < 0.05.
Citation: European Journal of Endocrinology 182, 2; 10.1530/EJE-19-0537
Adult and mid-parental target height in CHH/KS males and their brothers. (Panel A) Individual values (n = 165) and mean ± s.e. of male CHH/KS adult height versus mid-parental target height. Dark grey zone shows the 2.5 to the 97.5 centile of the French male general population. Light grey indicates the min and max limits of the French general population. Dotted line indicates median. General French population data are from the public database of the French National Institute of Health, http://invs.santepubliquefrance.fr/. Mid-parental target height has been calculated by means of the parental height according to Tanner formula (20). ****P < 0.0001. (Panel B) Comparison between the adult height in a subgroup of CHH/KS males and that of their unaffected brothers (n = 77). Individual values and mean ± s.e. are shown. *P < 0.05.
Citation: European Journal of Endocrinology 182, 2; 10.1530/EJE-19-0537
Adult and mid-parental target height in CHH/KS females and their sisters. (Panel A) Individual values and mean ± s.e. of female CHH/KS adult height versus mid-parental target height (n = 54). Dark grey zone shows the 2.5 to the 97.5 centile of the French male general population. Light grey indicates the min and max limits of the French general population. Dotted line indicates median. General French population data are from the public database of the French National Institute of Health, http://invs.santepubliquefrance.fr/. Mid-parental target height has been calculated by means of the parental height according to Tanner formula (20). ****P < 0.0001. (Panel B) Comparison between the adult height in a subgroup of CHH/KS females and that of their unaffected sisters. Individual values (n = 16) and mean ± s.e. are shown. ns, non-significant.
Citation: European Journal of Endocrinology 182, 2; 10.1530/EJE-19-0537
Adult and mid-parental target height in CHH/KS females and their sisters. (Panel A) Individual values and mean ± s.e. of female CHH/KS adult height versus mid-parental target height (n = 54). Dark grey zone shows the 2.5 to the 97.5 centile of the French male general population. Light grey indicates the min and max limits of the French general population. Dotted line indicates median. General French population data are from the public database of the French National Institute of Health, http://invs.santepubliquefrance.fr/. Mid-parental target height has been calculated by means of the parental height according to Tanner formula (20). ****P < 0.0001. (Panel B) Comparison between the adult height in a subgroup of CHH/KS females and that of their unaffected sisters. Individual values (n = 16) and mean ± s.e. are shown. ns, non-significant.
Citation: European Journal of Endocrinology 182, 2; 10.1530/EJE-19-0537
Adult and mid-parental target height in CHH/KS females and their sisters. (Panel A) Individual values and mean ± s.e. of female CHH/KS adult height versus mid-parental target height (n = 54). Dark grey zone shows the 2.5 to the 97.5 centile of the French male general population. Light grey indicates the min and max limits of the French general population. Dotted line indicates median. General French population data are from the public database of the French National Institute of Health, http://invs.santepubliquefrance.fr/. Mid-parental target height has been calculated by means of the parental height according to Tanner formula (20). ****P < 0.0001. (Panel B) Comparison between the adult height in a subgroup of CHH/KS females and that of their unaffected sisters. Individual values (n = 16) and mean ± s.e. are shown. ns, non-significant.
Citation: European Journal of Endocrinology 182, 2; 10.1530/EJE-19-0537
In male CHH/KS patients, mean AH was 179.0 ± 8.2 cm (median: 179 cm, interquartile range (IQR): 174–185 cm, Fig. 1A). Comparing AH to mid-parental target height (TH) revealed that on average, male patients with CHH/KS surpassed their mid-parental TH (179.0 ± 8.2 vs 175.0 ± 5.6 cm, P < 0.0001, Fig. 1A). Interestingly, male CHH/KS patients were significantly taller than their brothers (181.3 ± 7.9 vs 179.0 ± 6.9, P = 0.045) (Fig. 1B).
In female patients mean AH was 166.0 ± 6.7 cm (median: 166 cm, IQR: 162–170.5 cm, Fig. 2A). Similarly, female patients with CHH/KS surpassed their mid-parental TH (166 ± 6.2 vs 162 ± 4.6 cm, P < 0.0001, Fig. 2A). Female patients with CHH/KS (n = 16) did not differ from their sisters (n = 16) (166.1 ± 6.2 vs 165.0 ± 5.8 cm, P = 0.26) (Fig. 2B).
Examining delta height (∆H = AH − TH) showed that values were similar in male and female patients (4.0 ± 7.2 vs 4.3 ± 6.1 cm, P = 0.8). Among unaffected siblings, mean ∆H was also positive (3.4 ± 5.2 cm) – consistent with progressive generational statural gain (10, 11). However, patient ∆H (n = 172) was significantly greater than unaffected siblings (6.2 ± 7.2 vs 3.4 ± 5.2 cm, P < 0.0001, Fig. 3A). We examined the ∆H distribution and found the vast majority of CHH/KS patients 161/219 (73.5%) had a positive ∆H (Fig. 3B). Further, patients were more likely to have positive ∆H compared to their unaffected siblings (n = 172, 88.4% vs 73.8%, P = 0.0009, see Supplementary Fig. 2).
Difference between adult and mid-parental target height (delta) in pooled male and female CHH/KS patients. (Panel A) Delta height values between pooled male and female CHH/KS patients and their male and female siblings (n = 172) (see text, Methods and Results section). Data are shown as box and whiskers (min to max). Red line indicates zero (no difference between adult and mid-parental target height). ****P < 0.0001. Note that delta height in siblings is above zero (generational effect, see Results section). (Panel B) Distribution of positive (AH > TH, >0 cm) and negative (AH < TH, <0 cm) ∆H in CHH/KS population. Each vertical bar represents the number of patients. On x-axis numbers represent the delta (adult – mid parental target height, in cm). Vertical red line indicates zero (no difference between adult and mid parental target height). (Panel C) Individual values of delta height (∆H) in male and female CHH/KS patients according to the age at first treatment (in years), indicating correlation between the delta height (in cm) and the age at first replacement therapy. Best fit and 95% confidence interval is shown. A full colour version of this figure is available at https://doi.org/10.1530/EJE-19-0537.
Citation: European Journal of Endocrinology 182, 2; 10.1530/EJE-19-0537
Difference between adult and mid-parental target height (delta) in pooled male and female CHH/KS patients. (Panel A) Delta height values between pooled male and female CHH/KS patients and their male and female siblings (n = 172) (see text, Methods and Results section). Data are shown as box and whiskers (min to max). Red line indicates zero (no difference between adult and mid-parental target height). ****P < 0.0001. Note that delta height in siblings is above zero (generational effect, see Results section). (Panel B) Distribution of positive (AH > TH, >0 cm) and negative (AH < TH, <0 cm) ∆H in CHH/KS population. Each vertical bar represents the number of patients. On x-axis numbers represent the delta (adult – mid parental target height, in cm). Vertical red line indicates zero (no difference between adult and mid parental target height). (Panel C) Individual values of delta height (∆H) in male and female CHH/KS patients according to the age at first treatment (in years), indicating correlation between the delta height (in cm) and the age at first replacement therapy. Best fit and 95% confidence interval is shown. A full colour version of this figure is available at https://doi.org/10.1530/EJE-19-0537.
Citation: European Journal of Endocrinology 182, 2; 10.1530/EJE-19-0537
Difference between adult and mid-parental target height (delta) in pooled male and female CHH/KS patients. (Panel A) Delta height values between pooled male and female CHH/KS patients and their male and female siblings (n = 172) (see text, Methods and Results section). Data are shown as box and whiskers (min to max). Red line indicates zero (no difference between adult and mid-parental target height). ****P < 0.0001. Note that delta height in siblings is above zero (generational effect, see Results section). (Panel B) Distribution of positive (AH > TH, >0 cm) and negative (AH < TH, <0 cm) ∆H in CHH/KS population. Each vertical bar represents the number of patients. On x-axis numbers represent the delta (adult – mid parental target height, in cm). Vertical red line indicates zero (no difference between adult and mid parental target height). (Panel C) Individual values of delta height (∆H) in male and female CHH/KS patients according to the age at first treatment (in years), indicating correlation between the delta height (in cm) and the age at first replacement therapy. Best fit and 95% confidence interval is shown. A full colour version of this figure is available at https://doi.org/10.1530/EJE-19-0537.
Citation: European Journal of Endocrinology 182, 2; 10.1530/EJE-19-0537
Patient’s age at diagnosis was correlated with ∆H (Fig. 3C).
∆H did not differ between patients with normosmic CHH and those with KS (4.4 ± 6.4 vs 4.4 ± 7.3 cm, P = 0.9, Fig. 4A). Similarly, no significant differences in ∆H values were identified based on genetic causes in the 214 patients who had genetic screening (Fig. 4B).
Delta height (∆H) (difference between adult height and mid-parental target height) in pooled male and female CHH/KS patients according to olfactory status (KS versus nCHH) and genetic form. (Panel A) Individual delta height values in Kallmann syndrome versus normosmic CHH patients (nCHH). Dotted line indicates zero (no difference between adult height and mid-parental target height). (Panel B) Individual delta height (∆H) values in CHH/KS patients with and in those without identified genetic abnormalities. Dotted line indicates zero (no difference between adult and mid-parental target height). Data are reported as individual ∆H values in cm; mean ± s.e. are also indicated in each group.
Citation: European Journal of Endocrinology 182, 2; 10.1530/EJE-19-0537
Delta height (∆H) (difference between adult height and mid-parental target height) in pooled male and female CHH/KS patients according to olfactory status (KS versus nCHH) and genetic form. (Panel A) Individual delta height values in Kallmann syndrome versus normosmic CHH patients (nCHH). Dotted line indicates zero (no difference between adult height and mid-parental target height). (Panel B) Individual delta height (∆H) values in CHH/KS patients with and in those without identified genetic abnormalities. Dotted line indicates zero (no difference between adult and mid-parental target height). Data are reported as individual ∆H values in cm; mean ± s.e. are also indicated in each group.
Citation: European Journal of Endocrinology 182, 2; 10.1530/EJE-19-0537
Delta height (∆H) (difference between adult height and mid-parental target height) in pooled male and female CHH/KS patients according to olfactory status (KS versus nCHH) and genetic form. (Panel A) Individual delta height values in Kallmann syndrome versus normosmic CHH patients (nCHH). Dotted line indicates zero (no difference between adult height and mid-parental target height). (Panel B) Individual delta height (∆H) values in CHH/KS patients with and in those without identified genetic abnormalities. Dotted line indicates zero (no difference between adult and mid-parental target height). Data are reported as individual ∆H values in cm; mean ± s.e. are also indicated in each group.
Citation: European Journal of Endocrinology 182, 2; 10.1530/EJE-19-0537
Anthropometric measures prior to hormonal replacement initiation were available for 118 patients. As shown in Fig. 5, we observed an inverse relationship between the age of treatment initiation and individual statural gain during replacement therapy (r = −0.57, P < 0.0001). In other words, patients had lower statural growth potential when the treatment was initiated later. It should be noted that when analyzing data about period recruitment, we found a negative correlation between the year of birth and the age at treatment initiation, indicating that age at treatment initiation was earlier in recent decades (negative correlation: r = 0.35; P < 0.0001).
Individual post-therapeutic statural growth according to the age at beginning of hormonal treatment in CHH/KS patients. In this figure, each individual point indicates the difference between the adult height and the height before the hormonal substitution. On y-axis, this growth gain is expressed in cm. On x-axis, the age at the beginning of treatment is expressed in years. Blue line shows the nonlinear fit. Inset: individual post-replacement therapy growth gain in CHH/KS patients in whom hormonal substitution started before or after the age of 16 years solely for indicating more growth potential in younger than older CHH/KS patients; ****P < 0.0001. A full colour version of this figure is available at https://doi.org/10.1530/EJE-19-0537.
Citation: European Journal of Endocrinology 182, 2; 10.1530/EJE-19-0537
Individual post-therapeutic statural growth according to the age at beginning of hormonal treatment in CHH/KS patients. In this figure, each individual point indicates the difference between the adult height and the height before the hormonal substitution. On y-axis, this growth gain is expressed in cm. On x-axis, the age at the beginning of treatment is expressed in years. Blue line shows the nonlinear fit. Inset: individual post-replacement therapy growth gain in CHH/KS patients in whom hormonal substitution started before or after the age of 16 years solely for indicating more growth potential in younger than older CHH/KS patients; ****P < 0.0001. A full colour version of this figure is available at https://doi.org/10.1530/EJE-19-0537.
Citation: European Journal of Endocrinology 182, 2; 10.1530/EJE-19-0537
Individual post-therapeutic statural growth according to the age at beginning of hormonal treatment in CHH/KS patients. In this figure, each individual point indicates the difference between the adult height and the height before the hormonal substitution. On y-axis, this growth gain is expressed in cm. On x-axis, the age at the beginning of treatment is expressed in years. Blue line shows the nonlinear fit. Inset: individual post-replacement therapy growth gain in CHH/KS patients in whom hormonal substitution started before or after the age of 16 years solely for indicating more growth potential in younger than older CHH/KS patients; ****P < 0.0001. A full colour version of this figure is available at https://doi.org/10.1530/EJE-19-0537.
Citation: European Journal of Endocrinology 182, 2; 10.1530/EJE-19-0537
We dichotomized the data and observed that patients starting treatment before the age of 16 years exhibited significantly greater individual statural growth potential compared to their counterparts who started treatment later (16.6 ± 5.9 vs 4.1 ± 5.2 cm, P < 0.0001 Fig. 5, inset). Important and significant individual statural growth differences were also observed when dichotomizing patients starting hormonal treatment before or after the age of 14 (18.1 ± 8.1 vs 6.8 ± 7.1 cm, P < 0.0001), and before or after the age of 18 (13.1 ± 7.3 vs 2.5 ± 3.9 cm, P < 0.0001).
Post-therapeutic growth was greater in male (+9.2 ± 8.8 cm, n = 98) than in female CHH/KS (+4.4 ± 6.1 cm, n = 33, P = 0.0051).
Bone age was assessed in a subgroup of 59 CHH/KS patients (15 females, 44 males) before hormonal treatment. Bone age was significantly lower than chronological age (females: 16.7 ± 1.8 vs 13.2 ± 1.0, P < 0.0001; males: 16.3 ± 2.9 vs 13.3 ± 1.2, P < 0.0001, Supplemental Fig. 3A). A positive correlation was found between chronological and bone age (R = 0.73, P < 0.0001, Supplementary Fig. 3B).
Discussion
Herein we report adult height (AH) in a single-center cohort of men and women with CHH/KS who were identified using rigorous, well-defined criteria. We found that mean AH of patients exceeded the general population, mid-parental target height (TH) and sibling adult height. Few studies have examined height in patients with CHH/KS. These studies are quite dated, limited by small sample sizes and are confounded by the fact that some studies include patients with other causes of pubertal delay (7, 8, 12, 30). Our findings of increased height among CHH/KS patients is consistent with the previous study by Uriarte and colleagues (8). These investigators analyzed AH in 41 males with CHH/KS and found that compared to a reference group of young American males, male CHH/KS patients were 3 cm taller on average. To our knowledge, the present work is the first to compare AH in women with CHH/KS with a same-sex reference population.
Distinct from prior studies, we accounted for genetic background by comparing patient AH to mid-parental target height (i.e. target height (TH)) as well as to AH in sex-matched siblings. This is relevant, as familial genetic component is known to have a major effect on AH (31). In the present study, both male and female patient AH significantly exceeded TH. These observations indicate CHH/KS does not have a deleterious genetic effect on height potential – as has been previously suggested (32). Rather, patients with CHH/KS are likely to have higher growth potential compared to their unaffected relatives. The present study is the first to use healthy, sex-matched siblings as genetic controls to examine AH in patients with CHH/KS. This is particularly relevant due to the observed generational effect on AH in which offspring are progressively taller than previous generations. Thus, even in healthy populations, the AH in offspring of both sexes exceeds mid-parental target height (10). Indeed, we observed a positive delta height (∆H) in CHH/KS unaffected siblings, consistent with this generational effect. We further controlled for this in the present study by including same-sex siblings. We revealed that male CHH/KS patients were significantly taller than their healthy brothers. We failed to detect a significant difference between female CHH/KS patients and their sisters, probably because of an insufficient sample size. However, by analyzing ∆H differences and thus by pooling males and females together, CHH/KS patients were significantly taller than their siblings. Our findings suggest that part of the difference between the patients’ adult height and their mid-parental target height could be explained by a secular trend (10). However when compared to their siblings, a modest but significant increase in height was also observed. Thus, the increased statural growth potential in CHH/KS is not solely due to a generational effect, but could be explained by the slow and delayed maturation and welding of growth cartilage secondary to low endogenous sex steroids levels associated with this condition.
It is worthwhile to note that a minority of patients had a negative ∆H (AH below mid-parental target height (TH)). However, this was not unique to patients. Indeed, brothers/sisters were more likely than their affected siblings (CHH/KS) to have negative ∆H. The relative low stature observed in a minority of patients thus seems unrelated to the disease and its treatment. Growth and AH are multifactorial and contributors other than genetic potential play a part. It is widely acknowledged that environmental and nutritional factors play a role in adult height (33).
We observed a positive relationship between age at CHH/KS diagnosis and ∆H. Patients who are diagnosed earlier have relatively lower ∆H values. A possible explanation for this finding is that, before diagnosis and prior to any hormonal treatment, because of the very low concentrations of sex steroids (34, 35) linear growth could continue without cartilage epiphyseal fusion until sex steroid replacement is initiated.
Neither form (i.e. normosmic CHH vs KS) nor genetic cause had any specific observed effect on AH. Thus, the increased AH appears to be intrinsic to congenital hypogonadotropic hypogonadism – i.e. absence of increased sex steroids typical during puberty. It is also interesting that patients harboring mutations in FGFR1 did not have impaired AH. This gene has known effects on bone development and is associated with skeletal phenotypes in CHH/KS (1, 2, 36, 37). However, the present findings are in agreement with those from a prior study that included five KS patients (38) suggesting that FGFR1 mutations do not alter the development/maturation of the hyaline cartilage comprising the epiphyseal plates.
Clinically, height is a major consideration for patients. Abnormal growth, disrupted puberty and short stature are among the most common causes for endocrine consultation among adolescents (4, 39). Moreover, looking physically different from peers is a major concern at presentation (1, 2, 4, 39). Importantly, the AH prognosis specific to each disease influences clinical decision-making regarding the timing for initiating treatment and selected sex steroid doses. In the present study, as expected we show that individual statural growth during treatment was greatest when treatment was initiated earlier. An explanation is that younger CHH/KS patients just have more growth potential than older CHH patients. We must however emphasize that in CHH/KS patients even when replacement treatment is started after age 18 years, further growth can occur – albeit much less. In the handful of patients who were started on sex steroid treatment around the age of 30 years had no observable change in height. Such observation suggests that the growth plates had fused (6) despite a severe (yet not complete (34)) sex steroid deficiency. Low circulating sex steroids levels, mostly derived from adrenal precursors, are present in CHH/KS patients (34) and may contribute to slower bone maturation in these patients.
Conclusions
We demonstrate that CHH/KS does not negatively affect stature in adulthood regardless of sex, clinical form (normosmic CHH vs KS) or genetic etiology. Men and women with CHH/KS are, on average, taller than the general population, exceed their mid-parental target height and male CHH/KS have greater adult stature than unaffected brothers.
Our data also suggest that an earlier treatment does not seem to negatively impact adult stature but solely make it closer to mid-parental target height. On the other hand, late treatment could produce a taller than predicted adult stature, but further studies are needed to definitively answer to this hypothesis. Finally, we cannot exclude that the potential modest increase in adult stature associated with a late treatment could be obtained at the expense of a detriment of quality of life, lack of masculinization or feminization of CHH/KS adolescents, and suboptimal bone mass peak acquisition.
Supplementary materials
This is linked to the online version of the paper at https://doi.org/10.1530/EJE-19-0537.
Declaration of interest
The authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of this study.
Funding
This work was supported in part by the Programme Hospitalier de Recherche Clinique ‘Hypo-Proteo,’ Assistance Publique Hôpitaux de Paris, and Ministry of Health (to J Y); Agence Française de Lutte contre le Dopage (the French antidoping agency) (to J Y); Fondation pour la Recherche Médicale Grant FRM-2009 (to J Y); University Paris-Sud (Bonus Qualite Recherche) (to J Y); Agence Nationale de la Recherche Grant ANR-09-GENO-017-01 (to J Y); Programme Hospitalier de Recherche Clinique, French Ministry of Health Grant P081216/IDRCB 2009-A00892-55 (Variété), Assistance Publique Hôpitaux de Paris (to P C).
References
- 1↑
Young J, Xu C, Papadakis GE, Acierno JS, Maione L, Hietamaki J, Raivio T, Pitteloud N. Clinical management of congenital hypogonadotropic hypogonadism. Endocrine Reviews 2019 669–710. (https://doi.org/10.1210/er.2018-00116)
- 2↑
Boehm U, Bouloux PM, Dattani MT, de Roux N, Dode C, Dunkel L, Dwyer AA, Giacobini P, Hardelin JP & Juul A et al. Expert consensus document: European Consensus Statement on congenital hypogonadotropic hypogonadism – pathogenesis, diagnosis and treatment. Nature Reviews: Endocrinology 2015 547–564. (https://doi.org/10.1038/nrendo.2015.112)
- 3↑
Fromantin M, Gineste J, Didier A, Rouvier J. Impuberism and hypogonadism at induction into military service. Statistical study. Problemes Actuels d’Endocrinologie and de Nutrition 1973 179–199.
- 4↑
Dwyer AA, Quinton R, Pitteloud N, Morin D. Psychosexual development in men with congenital hypogonadotropic hypogonadism on long-term treatment: a mixed methods study. Sexual Medicine 2015 32–41. (https://doi.org/10.1002/sm2.50)
- 5↑
Dzemaili S, Tiemensma J, Quinton R, Pitteloud N, Morin D, Dwyer AA. Beyond hormone replacement: quality of life in women with congenital hypogonadotropic hypogonadism. Endocrine Connections 2017 404–412. (https://doi.org/10.1530/EC-17-0095)
- 6↑
Bourguignon JP. Linear growth as a function of age at onset of puberty and sex steroid dosage: therapeutic implications. Endocrine Reviews 1988 467–488. (https://doi.org/10.1210/edrv-9-4-467)
- 7↑
Dickerman Z, Cohen A, Laron Z. Growth in patients with isolated gonadotrophin deficiency. Archives of Disease in Childhood 1992 513–516. (https://doi.org/10.1136/adc.67.4.513)
- 8↑
Uriarte MM, Baron J, Garcia HB, Barnes KM, Loriaux DL, Cutler GB Jr. The effect of pubertal delay on adult height in men with isolated hypogonadotropic hypogonadism. Journal of Clinical Endocrinology and Metabolism 1992 436–440. (https://doi.org/10.1210/jcem.74.2.1449545)
- 9↑
Varimo T, Hero M, Laitinen EM, Miettinen PJ, Tommiska J, Kansakoski J, Juul A, Raivio T. Childhood growth in boys with congenital hypogonadotropic hypogonadism. Pediatric Research 2016 705–709. (https://doi.org/10.1038/pr.2015.278)
- 10↑
Cole TJ. The secular trend in human physical growth: a biological view. Economics and Human Biology 2003 161–168. (https://doi.org/10.1016/S1570-677X(02)00033-3)
- 11↑
Ong KK, Ahmed ML, Dunger DB. Lessons from large population studies on timing and tempo of puberty (secular trends and relation to body size): the European trend. Molecular and Cellular Endocrinology 2006 8–12. (https://doi.org/10.1016/j.mce.2006.04.018)
- 12↑
Bourguignon JP, Vandeweghe M, Vanderschueren-Lodeweyckx M, Malvaux P, Wolter R, Du Caju M, Ernould C. Pubertal growth and final height in hypopituitary boys: a minor role of bone age at onset of puberty. Journal of Clinical Endocrinology and Metabolism 1986 376–382. (https://doi.org/10.1210/jcem-63-2-376)
- 13↑
Klein KO, Rosenfield RL, Santen RJ, Gawlik AM, Backeljauw PF, Gravholt CH, Sas TCJ, Mauras N. Estrogen replacement in Turner syndrome: literature review and practical considerations. Journal of Clinical Endocrinology and Metabolism 2018 1790–1803. (https://doi.org/10.1210/jc.2017-02183)
- 14↑
Chanson P, Cailleux-Bounacer A, Kuhn JM, Weryha G, Chabre O, Borson-Chazot F, Dubois S, Vincent-Dejean C, Brue T & Fedou C et al. Comparative validation of the growth hormone-releasing hormone and arginine test for the diagnosis of adult growth hormone deficiency using a growth hormone assay conforming to recent international recommendations. Journal of Clinical Endocrinology and Metabolism 2010 3684–3692. (https://doi.org/10.1210/jc.2010-0295)
- 15↑
Ducos B, Cabrol S, Houang M, Perin L, Holzenberger M, Le Bouc Y. IGF type 1 receptor ligand binding characteristics are altered in a subgroup of children with intrauterine growth retardation. Journal of Clinical Endocrinology and Metabolism 2001 5516–5524. (https://doi.org/10.1210/jcem.86.11.7985)
- 16↑
Galland F, Kamenicky P, Affres H, Reznik Y, Pontvert D, Le Bouc Y, Young J, Chanson P. McCune-Albright syndrome and acromegaly: effects of hypothalamopituitary radiotherapy and/or pegvisomant in somatostatin analog-resistant patients. Journal of Clinical Endocrinology and Metabolism 2006 4957–4961. (https://doi.org/10.1210/jc.2006-0561)
- 17↑
Chanson P, Arnoux A, Mavromati M, Brailly-Tabard S, Massart C, Young J, Piketty ML, Souberbielle JC & VARIETE Investigators. Reference values for IGF-I serum concentrations: comparison of six immunoassays. Journal of Clinical Endocrinology and Metabolism 2016 3450–3458. (https://doi.org/10.1210/jc.2016-1257)
- 18↑
Maione L, Benadjaoud S, Eloit C, Sinisi AA, Colao A, Chanson P, Ducreux D, Benoudiba F, Young J. Computed tomography of the anterior skull base in Kallmann syndrome reveals specific ethmoid bone abnormalities associated with olfactory bulb defects. Journal of Clinical Endocrinology and Metabolism 2013 E537–E546. (https://doi.org/10.1210/jc.2012-3553)
- 19↑
Maione L, Cantone E, Nettore IC, Cerbone G, De Brasi D, Maione N, Young J, Di Somma C, Sinisi AA & Lengo M et al. Flavor perception test: evaluation in patients with Kallmann syndrome. Endocrine 2016 236–243. (https://doi.org/10.1007/s12020-015-0690-y)
- 20↑
Tanner JM, Goldstein H, Whitehouse RH. Standards for children’s height at ages 2–9 years allowing for heights of parents. Archives of Disease in Childhood 1970 755–762. (https://doi.org/10.1136/adc.45.244.755)
- 21↑
Straetemans S, De Schepper J, Thomas M, Verlinde F, Rooman R & BESPEED. Validation of prediction models for near adult height in children with idiopathic growth hormone deficiency treated with growth hormone: a Belgian Registry study. Hormone Research in Paediatrics 2016 161–168. (https://doi.org/10.1159/000448553)
- 22↑
Dode C, Levilliers J, Dupont JM, De Paepe A, Le Du N, Soussi-Yanicostas N, Coimbra RS, Delmaghani S, Compain-Nouaille S & Baverel F et al. Loss-of-function mutations in FGFR1 cause autosomal dominant Kallmann syndrome. Nature Genetics 2003 463–465. (https://doi.org/10.1038/ng1122)
- 23↑
Dode C, Teixeira L, Levilliers J, Fouveaut C, Bouchard P, Kottler ML, Lespinasse J, Lienhardt-Roussie A, Mathieu M & Moerman A et al. Kallmann syndrome: mutations in the genes encoding prokineticin-2 and prokineticin receptor-2. PLoS Genetics 2006 e175. (https://doi.org/10.1371/journal.pgen.0020175)
- 24↑
Francou B, Paul C, Amazit L, Cartes A, Bouvattier C, Albarel F, Maiter D, Chanson P, Trabado S & Brailly-Tabard S et al. Prevalence of KISS1 receptor mutations in a series of 603 patients with normosmic congenital hypogonadotrophic hypogonadism and characterization of novel mutations: a single-centre study. Human Reproduction 2016 1363–1374. (https://doi.org/10.1093/humrep/dew073)
- 25↑
Hanchate NK, Giacobini P, Lhuillier P, Parkash J, Espy C, Fouveaut C, Leroy C, Baron S, Campagne C & Vanacker C et al. SEMA3A, a gene involved in axonal pathfinding, is mutated in patients with Kallmann syndrome. PLoS Genetics 2012 e1002896. (https://doi.org/10.1371/journal.pgen.1002896)
- 26↑
Kim HG, Ahn JW, Kurth I, Ullmann R, Kim HT, Kulharya A, Ha KS, Itokawa Y, Meliciani I & Wenzel W et al. WDR11, a WD protein that interacts with transcription factor EMX1, is mutated in idiopathic hypogonadotropic hypogonadism and Kallmann syndrome. American Journal of Human Genetics 2010 465–479. (https://doi.org/10.1016/j.ajhg.2010.08.018)
- 27↑
Pingault V, Bodereau V, Baral V, Marcos S, Watanabe Y, Chaoui A, Fouveaut C, Leroy C, Verier-Mine O & Francannet C et al. Loss-of-function mutations in SOX10 cause Kallmann syndrome with deafness. American Journal of Human Genetics 2013 707–724. (https://doi.org/10.1016/j.ajhg.2013.03.024)
- 28↑
Salenave S, Chanson P, Bry H, Pugeat M, Cabrol S, Carel JC, Murat A, Lecomte P, Brailly S & Hardelin JP et al. Kallmann’s syndrome: a comparison of the reproductive phenotypes in men carrying KAL1 and FGFR1/KAL2 mutations. Journal of Clinical Endocrinology and Metabolism 2008 758–763. (https://doi.org/10.1210/jc.2007-1168)
- 29↑
Marcos S, Sarfati J, Leroy C, Fouveaut C, Parent P, Metz C, Wolczynski S, Gerard M, Bieth E & Kurtz F et al. The prevalence of CHD7 missense versus truncating mutations is higher in patients with Kallmann syndrome than in typical CHARGE patients. Journal of Clinical Endocrinology and Metabolism 2014 E2138–E2143. (https://doi.org/10.1210/jc.2014-2110)
- 30↑
Raboch J, Reisenauer R. Analysis of body height in 829 patients with different forms of testicular pathology. Andrologia 1976 265–268. (https://doi.org/10.1111/j.1439-0272.1976.tb02148.x)
- 31↑
Jelenkovic A, Hur YM, Sund R, Yokoyama Y, Siribaddana SH, Hotopf M, Sumathipala A, Rijsdijk F, Tan Q & Zhang D et al. Genetic and environmental influences on adult human height across birth cohorts from 1886 to 1994. eLife 2016 e20320. (https://doi.org/10.7554/eLife.20320)
- 32↑
Kaushanski A, Laron Z. Growth pattern of boys with isolated gonadotropin deficiency. Israel Journal of Medical Sciences 1979 518–521.
- 33↑
Perkins JM, Subramanian SV, Davey Smith G, Ozaltin E. Adult height, nutrition, and population health. Nutrition Reviews 2016 149–165. (https://doi.org/10.1093/nutrit/nuv105)
- 34↑
Giton F, Trabado S, Maione L, Sarfati J, Le Bouc Y, Brailly-Tabard S, Fiet J, Young J. Sex steroids, precursors, and metabolite deficiencies in men with isolated hypogonadotropic hypogonadism and panhypopituitarism: a GCMS-based comparative study. Journal of Clinical Endocrinology and Metabolism 2015 E292–E296. (https://doi.org/10.1210/jc.2014-2658)
- 35↑
Trabado S, Maione L, Salenave S, Baron S, Galland F, Bry-Gauillard H, Guiochon-Mantel A, Chanson P, Pitteloud N & Sinisi AA et al. Estradiol levels in men with congenital hypogonadotropic hypogonadism and the effects of different modalities of hormonal treatment. Fertility and Sterility 2011 2324–2329, 2329.e1–3. (https://doi.org/10.1016/j.fertnstert.2011.03.091)
- 36↑
Maione L, Dwyer AA, Francou B, Guiochon-Mantel A, Binart N, Bouligand J, Young J. GENETICS IN ENDOCRINOLOGY: Genetic counseling for congenital hypogonadotropic hypogonadism and Kallmann syndrome: new challenges in the era of oligogenism and next-generation sequencing. European Journal of Endocrinology 2018 R55–R80. (https://doi.org/10.1530/EJE-17-0749)
- 37↑
Villanueva C, Jacobson-Dickman E, Xu C, Manouvrier S, Dwyer AA, Sykiotis GP, Beenken A, Liu Y, Tommiska J & Hu Y et al. Congenital hypogonadotropic hypogonadism with split hand/foot malformation: a clinical entity with a high frequency of FGFR1 mutations. Genetics in Medicine 2015 651–659. (https://doi.org/10.1038/gim.2014.166)
- 38↑
Hero M, Laitinen EM, Varimo T, Vaaralahti K, Tommiska J, Raivio T. Childhood growth of females with Kallmann syndrome and FGFR1 mutations. Clinical Endocrinology 2015 122–126. (https://doi.org/10.1111/cen.12504)
- 39↑
Dwyer AA, Phan-Hug F, Hauschild M, Elowe-Gruau E, Pitteloud N. TRANSITION IN ENDOCRINOLOGY: Hypogonadism in adolescence. European Journal of Endocrinology 2015 R15–R24. (https://doi.org/10.1530/EJE-14-0947)
