Testosterone concentrations in female athletes and

This article was downloaded by: [Uniwersytet Przyrodniczy W Poznaniu] On: 08 May 2015, At: 03:28 Publisher: Taylor & Francis Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK

European Journal of Sport Science Publication details, including instructions for authors and subscription information: http://www.tandfonline.com/loi/tejs20

Testosterone concentrations in female athletes and ballet dancers with menstrual disorders a

Karolina Łagowska & Karina Kapczuk

b

a

Department of Human Nutrition and Hygiene, Dietetic Division, Poznan University of Life Sciences, Poznań, Poland b

Department of Perinatology and Gynecology, Division of Gynecology, Karol Marcinkowski University of Medical Sciences, Poznań, Poland Published online: 08 May 2015.

Click for updates To cite this article: Karolina Łagowska & Karina Kapczuk (2015): Testosterone concentrations in female athletes and ballet dancers with menstrual disorders, European Journal of Sport Science, DOI: 10.1080/17461391.2015.1034786 To link to this article: http://dx.doi.org/10.1080/17461391.2015.1034786

PLEASE SCROLL DOWN FOR ARTICLE Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and should be independently verified with primary sources of information. Taylor and Francis shall not be liable for any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of the Content. This article may be used for research, teaching, and private study purposes. Any substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http:// www.tandfonline.com/page/terms-and-conditions

European Journal of Sport Science, 2015 http://dx.doi.org/10.1080/17461391.2015.1034786

ORIGINAL ARTICLE

Testosterone concentrations in female athletes and ballet dancers with menstrual disorders KAROLINA ŁAGOWSKA1 & KARINA KAPCZUK2 Department of Human Nutrition and Hygiene, Dietetic Division, Poznan University of Life Sciences, Poznan´, Poland, Department of Perinatology and Gynecology, Division of Gynecology, Karol Marcinkowski University of Medical Sciences, Poznan´, Poland

1

Downloaded by [Uniwersytet Przyrodniczy W Poznaniu] at 03:28 08 May 2015

2

Abstract Menstrual disorders are common among female athletes and ballet dancers. Endocrine changes, such as high testosterone (HT) levels and high luteinizing hormone (LH)/follicle-stimulating hormone (FSH) ratios, may suggest functional ovarian hyperandrogenism which may induce such dysfunction. The aim of this study was therefore to evaluate endocrine status in female athletes and ballet dancers with menstrual disorders. Their nutritional status and dietary habits were analysed in relation to the testosterone levels. In a cross-sectional approach, 31 female athletes (18.1 ± 2.6 years) and 21 ballerinas (17.1 ± 0.9) with menstrual disorders participated in the study. The levels of serum LH, FSH, progesterone (P), estradiol (E2), prolactin (PRL), thyroid-stimulating hormone, testosterone (T) and sex hormone-binding globulinwere measured to assess hormonal status. In addition, the free androgen index (FAI) was calculated. Nutritional status, total daily energy expenditure and nutritional habits were evaluated. Girls were assigned to one of the following groups: low testosterone (LT) level, normal testosterone level or HT level. There were significant differences between ballerinas and other female athletes in terms of testosterone levels, FAI, age at the beginning of training, length of training period and age at menarche. The PRL level was lowest in the LT group while the FAI index was highest in the HT group. Daily energy and carbohydrate intakes were significantly lower in the HT group. T levels in the study subjects were found to be associated with nutritional factors, energy availability, age at the beginning of training and frequency of training. This is the first report of HT levels being associated with the status of a female ballet dancer, the age of menarche and the length of the training history. Further research is necessary to confirm the results in a larger study group. Keywords: Body composition, health, metabolism, nutrition

Introduction Menstrual disorders are common among female athletes, particularly in esthetic and endurance disciplines. About 16– 61 percent of female athletes are estimated to suffer from hypothalamic–pituitary menstrual disorders, which are mostly due to low energy availability, negative energy balance, inadequate nutritional status [low body weight (BW) and fat mass in particular], emotional stress and excessive physical exercise that frequently begins before puberty (Klentrou & Plyley, 2003; Mudd, Fornetti, & Pivarnik, 2007; Nattiv et al., 2007). For the same reasons, menstrual disorders are also very common among ballet dancers. In a study conducted by Doyle-Lucas, Akers, and Davy (2010), 40% of ballet dancers suffered from amenorrhea and 6.7% from oligomenorrhea,

while in a study by Stokić, Srdić, and Barak (2005), amenorrhea was found in 20% of ballerinas and oligomenorrhea in 10%. In female athletes and ballet dancers who suffer from menstrual dysfunction, disturbances in the secretion of gonadoliberin (GnRH pulsatility) are observed, resulting in reduced levels of luteinizing hormone (LH) and follicle-stimulating hormone (FSH) hormones, as well as of oestrogen (hypoestrogenism). Such women frequently exhibit hypoglycemia or hypoinsulinemia and reduced basal metabolism, in addition to low leptin levels.Laughlin and Yen (1996) reported increased secretion of GH, cortisol and IGF binding protein-1 (IGFBP-1) in amenorrheic athletes compared to regularly menstruating athletes and sedentary controls. The authors suggested that these endocrine changes are adaptations

Correspondence: K. Łagowska, Department of Human Nutrition and Hygiene, Dietetic Division, Poznan University of Life Sciences, Wojska Polskiego Str., Poznan´, Poland. E-mail: [email protected] © 2015 European College of Sport Science


2

K. Łagowska & K. Kapczuk

to energy deficiency. Other endocrine changes, such as high testosterone (HT) levels and high LH/FSH ratios, may suggest functional ovarian hyperandrogenism (Maimoun et al., 2013). In studies by Coste et al. (2011) and Hagmar, Berglund, Brismar, and Hirschberg (2009), no functional hypogonadism was reported in swimmers with menstrual disorders; instead, hyperandrogenism was found in this population. In a study conducted by Rickenlund et al. (2003), athletes with menstrual disorders were found to have abnormally high testosterone/sex hormone-binding globulin (SHBG) ratios, increased levels of free testosterone, increased LH/FSH ratios and significantly reduced SHBG levels, as compared to controls and to other athletes, both with and without menstrual disorders. The purpose of the present cross-sectional study was to investigate the potential effects of different types of intensive training on the hormonal changes in adolescent ballet dancers and female athletes and to determine whether dietary habits and nutritional status also contribute to hyperandrogenism.

Materials and methods Subjects Forty-five well-trained female athletes (rowers, synchronised swimmers and triathlonists) from various sports clubs in Poznań and 27 female ballet dancers from the Poznań Ballet School were recruited to the study. Thirty-one female athletes and 21 female ballet dancers completed the study. One of inclusion criteria for athletes was to have taken part in national-level sport competitions during the last 12 months. The dancers were all ballet students. All women with training volumes of over 10 hours per week and with total training periods of more than three years were eligible. Other inclusion criteria were as follows: menstrual disorders within the last 12 months and non-use for the last 12 months of hormonal contraception or other medications that might interfere with the hypothalamic–pituitary–gonadal axis. Only subjects with gynecologic age >18 months were included. Written informed consent was obtained from all participants and their parents. The study was approved by the Poznań Medical Ethics Committee (no. 334/09). Menstrual status Each subject completed a two-part medical questionnaire. The questions in the first part concerned menstruation: age at menarche, length of menstrual cycles and history of amenorrhea. Primary amenorrhea was diagnosed where there was no onset

of menses by 15 years of age, while secondary amenorrhea was diagnosed where menstruation was absent for 6 months or for more than three times the previous cycle length. Menstrual periods that occurred more than 35 days apart were described as oligomenorrhea (3–6 menstruation bleedings per year; Nattiv et al., 2007). Each participant underwent a gynaecological evaluation performed by the co-author of this study; the evaluation included pelvic ultrasound and tests of the levels of LH, FSH, progesterone (P), estradiol (E2), prolactin (PRL), thyroid-stimulating hormone (TSH), testosterone (T) and SHBG, in order to exclude independent causes of amenorrhea or oligomenorrhea, such as pregnancy, primary ovarian failure, hyperprolactinemia, thyroid dysfunction or polycystic ovary syndrome. Hirsutism was graded according to the standard Ferriman–Gallwey score, by which the density of terminal hairs is scored at nine different body sites (Ferriman & Gallwey, 1961; Maimoun et al., 2013). In each of these areas, a score of 0–4 was assigned, with a total score of >8 defining hirsutism. Acne was graded according to the Pillsbury method, which counts acne lesions in order to rank severity as grade I (absent or minor, 1– 9 comedones), grade II (mild, 10–19 comedones), grade III (moderate, >20 comedones, inflammation) and grade IV (severe; Pillsbury, Kligman, & Shelley, 1961; Maimoun et al., 2013). None of the subjects presented hirsutism, acne, alopecia or voice deepening. Thirty-one female athletes and 21 female ballet dancers were assigned to three subgroups according to their T level: 27 (athletes = 9, ballet dancers = 18) composed the high T level (HT; T > 50 ng/dl), 21 (athletes = 18, ballet dancers = 3) composed the normal T level (NT; 10 ≤ T ≤ 50 ng/dl) and 4 (all athletes) composed the low T level (T < 10 ng/dl) (Ankarberg & Norjavaara, 1999; Maimoun et al., 2013; Sultan & Paris, 2006).

Body weight and body composition measurements In order to evaluate nutritional status, anthropometrical indices of height and weight were measured using an anthropometer coupled with a WPT 200 OC calibrated medical scale (Rad Wag). Body mass index (BMI, kg/m2) was calculated as BW divided by squared body height. Participants wore minimal clothing during measurements, which were rounded to the nearest 0.5 kg and 0.5 cm. Analysis of body fat mass (FM) and fat-free mass (FFM) was performed in the morning after an overnight fast, with subjects lying in a supine position, using a Bodystat 1500 apparatus, as described by Heyward and Wagner (2003). All athletes had empty bladders at the time of measurements and had been instructed

Testosterone concentrations in female athletes and ballet dancers to abstain from caffeine and alcohol for 24 hours prior to the measurement, as well as to refrain from performing strenuous exercise on the day prior to measurement. The measurement was not performed during menstruation. Body composition measurement was conducted, taking into account individual and environmental factors (Heyward & Wagner, 2003). The bioelectrical impedance method potentially raises some controversies. However, the DEXA method was not used, due to young age of study participants and the potential adverse effects (UV).


Blood sampling and biochemical analyses Blood samples were obtained between days 2 and 5 of the menstrual cycle (in the early follicular phase) in menstruating subjects and randomly in amenorrheic subjects. The samples were taken between 6.00 am and 9.00 am following overnight fasting and rest. Subjects were instructed to abstain from taking caffeine and alcohol for 24 hours prior to blood sampling and from performing strenuous exercise on the day of sampling. LH, FSH, E2, P, PRL, TSH, T and SHBG concentrations were measured by immunochemical methods using the Chemiluminescent Microparticle Immunoassay (CMIA) and Microparticle Chemiflex Flexible assay protocols, making use of diagnostic sets and an Architect automatic analyzer (Abbott, Germany). All hormone levels were determined in duplicate and, to reduce inter-assay variation, all of plasma samples were analysed in a single session. The ratio between the total serum T and SHBG (the T/SHBG ratio or “free androgen index”) is considered to be a useful indicator of T activity in women (Rotterdam ESHRE/ASRMSposnsored PCOS Consensus Workshop Group, 2004) and was thus also calculated here. The range 1.9–5.4 was considered to be normal. Nutritional habits Seven consecutive days of dietary records were obtained under the supervision of dieticians. All meals (including their recipes and item masses), non-meal foods, beverages and fluids were recorded in diary form with the assistance of a Photographic Album of Dishes (Szponar, Wolnicka, & Rychlik, 2000). Participants met regularly with a registered dietician to receive training in accurately recording dietary intake and to review the completed energy and nutrient intake logs. The participants received written guidelines regarding proper measurement and reporting of food portions and preparation. The daily diets were analysed for energy and nutrient levels (fat, protein and carbohydrate) using the Dietician computer software package, based on Polish food composition

3

tables (Kunachwicz, Nadolna, Przygoda, & Ivanow, 2005). Physical activity determination Part two of the questionnaire referred to sport activities: type of activity, age at the beginning of training in the activity, type of training and number of training hours per day and per week. Total daily energy expenditure and energy availability For three days, each subject wore a heart-rate monitor (Polar Sport Tester, RS 400, Finland) to estimate total daily energy expenditure (TDEE). The relationship between heart rate (HR) and oxygen consumption (VO2) was established for each subject. A K4b2 respirometer by Cosmed Italy, coupled with an ergometer, was used to measure oxygen consumption, according to the procedure presented by Pillsbury et al. (1961) and Spurr et al. (1988), as modified by Ceesay et al. (1989) and Tanhoffer, Tanhoffer, Raymond, Hills, and Davis (2012). The measurements were carried out two or more hours after meals, and after the subject had rested for 30 minutes after arriving at the laboratory. The results were obtained by simultaneous measurement of HR and VO2 for the following activities carried out sequentially: lying in a supine position, sitting quietly, standing quietly and continuous graded exercise on a cycle ergometer. The data were pre-edited to remove spurious HR data, and the total energy expenditure was then calculated using the Flex-HR method. In this method, a FlexHR value is defined for each subject such that good and poor correlations between HR and VO2 fall respectively above and below it. The Flex-HR was therefore calculated as the mean of the highest HR for resting activities (supine, sitting and standing) and the lowest HR for exercise activities. At the end of the measurement session, the researchers transferred the minute-by-minute records of the last 24 hours from the instrument to a database. The 24hour energy balance was calculated as the difference between the means of seven consecutive days of 24hour energy intake and the TDEE as a three-day mean. Energy availability was calculated by subtracting exercise energy expenditure from the total daily energy intake, adjusted for fat-free mass (FFM kg; Nattiv et al., 2007). Data analysis Means and standard deviations for quantitative variables were calculated. The normality of the distribution was tested using the Shapiro–Wilk statistic. The comparison of the means between the

4

K. Łagowska & K. Kapczuk

testosterone-level subgroups was performed through analysis of variance when the data distribution was normal and the Kruskal–Wallis test where the continuous variables were skewed. The connection between testosterone and other variables was tested using Spearman’s or Pearson’s rank correlation tests. Statistical analysis was performed using Statistica 8.0 (StatSoft, 2008). P values of less than 0.05 were considered statistically significant.

Results


Study subjects Ballet dancers began physical training at a significantly earlier age than the athletes. Their training periods were longer, and they devoted more time per week to training. It is interesting that the dancers were older at the onset of menarche than the athletes were. Significant differences between anthropometric parameters and body composition in the groups were also observed. BW, BMI and FM (kg) were significantly lower in ballet dancers, while FFM (kg) was significantly higher. Among the hormonal parameters, serum testosterone levels were higher in the dancers, while the free androgen index (FAI) index was lower. Daily energy, fat, protein and carbohydrate intakes were significantly higher in the athletes. In this group, significantly higher TDEE, exercise energy expenditure and energy availability were also observed. In both the athletes and the ballet dancers, the energy balance was negative and energy availability was lower than the critical value of 30 kcal/kg FFM/day (Table I).

Differences in parameter between testosterone-level groups There were significant differences between the study groups in terms of age at the beginning of training and the length of the training period. Girls with high levels of testosterone began training the earliest, and their training period was the longest. There were no significant differences in BW or body composition between the study groups, but there were differences in hormone levels. There were also significant differences in PRL levels, FAI, and age at menarche. The athletes from the low testosterone (LT) group had the lowest levels of PRL while those from the HT group had the lowest FAIs; they were also the youngest at onset of menarche. Statistically significant differences were also observed in energy expenditure and nutrient intake. Daily energy and carbohydrate intakes were significantly lower in subjects with increased testosterone. These results, combined with daily energy expenditure, revealed a negative energy balance in all groups. TDEE and exercise energy

Table I. Characteristics of study participants

Parameters

Athletes (n = 31)

Ballet dancers (n = 21)

Baseline characteristics Age (years) 18.1 ± 2.6 17.1 ± 0.9 Height (cm) 169.7 ± 6.7 167.1± 4.5 Age at the beginning 11.2 ± 3.5 6.7 ± 0.5 of training (years) Training period 6.8 ± 3.3 10.7 ± 1.2 (years) Number of training 5.2 ± 1.1 5.2 ± 0.5 session per week (n/week) Training hours per 4.0 ± 1.8 4.7 ± 1.1 day (hours/day) Training hours per 19.5 ± 7.2 24.9 ± 6.9 week (hours/week) Anthropometric characteristics BW (kg) 59.3 ± 5.3 52.9 ± 5.8 BMI (kg/m2) 20.6 ± 1.4 18.9 ± 1.7 FM (%) 20.6 ± 3.7 18.7 ± 3.2 FM (kg) 12.2 ± 2.4 9.9 ± 2.2 FFM (%) 79.4 ± 3.7 81.3 ± 3.2 FFM (kg) 47.1 ± 4.9 43.0 ± 4.8 Hormonal and menstrual cycles’ characteristics T (ng/dl) 37.28 ± 21.85 72.95 ± 25.92 TSH (µIU/ml) 1.74 ± 0.80 1.96 ± 1.33 PRL (ng/ml) 13.0 ± 9.33 16.38 ± 8.88 SHBG (nmol/l) 62.79 ± 41.91 71.62 ± 22.03 FAI 2.89 ± 2.51 3.85 ± 1.71 LH (mlU/ml) 3.04 ± 1.63 3.00 ± 1.51 FSH (mlU/ml) 5.01 ± 2.37 5.23 ± 2.32 LH/FSH 0.84 ± 0.56 0.72 ± 0.54 E2 (pg/ml) 36.5 ± 19.4 39.9 ± 21.7 P (ng/ml) 0.54 ± 0.99 0.53 ± 0.50 Age at menarche 13.0 ± 1.2 13.9 ± 1.3 (years) Secondary 6 (19.4) 6 (28.6) amenorrhea (n, %) Oligomenorrhea 25 (80.6) 15 (71.4) (n, %) Energy and nutrients intake Energy (kcal) 2354 ± 539 1640 ± 412 Fat (g) 92.2 ± 27.5 57.0 ± 18.9 Protein (g) 75.6 ± 14.8 58.8 ± 13.3 Carbohydrate (g) 305.4 ± 78.0 222.8 ± 55.7 TDEE (kcal/day) 2642 ± 348 2038 ± 279 EB (kcal/day) −288 ± 477 −384 ± 267 EEE (kcal/day) 959 ± 174 731 ± 207 EA (kcal/kg 28.3 ± 9.2 21.7 ± 7.2 FFM/day)

P value

NS NS