CLINICAL SCIENCES

Growth and Maturation of Adolescent Female Gymnasts, Swimmers, and Tennis Players
MARTA C. ERLANDSON1, LAUREN B. SHERAR1, ROBERT L. MIRWALD1, NICOLA MAFFULLI2, and ADAM D. G. BAXTER-JONES1
1

College of Kinesiology, University of Saskatchewan, Saskatoon, CANADA; and 2Department of Trauma and Orthopaedic Surgery, Keele University School of Medicine, North Staffordshire Hospital, Staffordshire, UNITED KINGDOM

ABSTRACT ERLANDSON, M. C., L. B. SHERAR, R. L. MIRWALD, N. MAFFULLI, and A. D. G. BAXTER-JONES. Growth and Maturation of Adolescent Female Gymnasts, Swimmers, and Tennis Players. Med. Sci. Sports Exerc., Vol. 40, No. 1, pp. 34–42, 2008. Intensive training at a young age may adversely affect the growth and sexual maturation of female athletes, resulting in compromised adult stature. Purpose: To compare the somatic growth, sexual maturation, and final adult height of elite adolescent female athletes. Methods: Serial measures of height, sitting height, and breast and pubic hair development were taken on 81 gymnasts, 60 swimmers, and 81 tennis players between 8 and 19 yr of age. Menarcheal age, parental heights, maternal menarcheal age, and number of training hours were also recorded. Final adult heights were obtained from a subsample of the athletes (N = 110). Results: Gymnasts were significantly shorter than tennis players and swimmers at all chronological ages during adolescence, and they attained menarche at an older age (P G 0.05). No significant differences were found in adult heights. During adolescence, no difference were found in standing height to sitting height ratios, leg length to standing height ratios, or sitting height to leg length ratios between sports (P > 0.05). Conclusion: The results from this study suggest that regular training did not affect final adult stature and that, when aligned by biological age, the tempo of sexual maturation was similar in these young athletes. Key Words: TRAINING, MENARCHE, PEAK HEIGHT VELOCITY, ADULT STATURE

T

o be successful at the international level of sport competition, it is believed that intensive training must begin before puberty (27). For example, in elite female artistic gymnastics, athletes normally peak in performance around 16 yr of age and are considered extraordinary if still competing internationally in their 20s. Thus, elite female gymnasts undertake training of progressive intensity and volume from an early age. Some gymnasts start training as young as 5 or 6 yr of age and can train 20–30 hIwkj1 year-round throughout their childhood and adolescence (9). It has been suggested that this type of intensive training at such a young age could adversely affect the growth and biological maturation of young female athletes (1,9,10,14,31,33). However, this is a view disputed by others (6,32), because a cause-and-effect relationship between intensive training at a young age and reduced growth has not yet been demonstrated. To answer this question, the effects of environment must be isolated

Address for correspondence: Marta Erlandson, B.Sc., College of Kinesiology, University of Saskatchewan, 87 Campus Drive, Saskatoon, SK, S7N 5B2, Canada; E-mail: marta.erlandson@usask.ca. Submitted for publication November 2006. Accepted for publication August 2007. 0195-9131/08/4001-0034/0 MEDICINE & SCIENCE IN SPORTS & EXERCISEÒ Copyright Ó 2007 by the American College of Sports Medicine DOI: 10.1249/mss.0b013e3181596678

from those of genetic predisposition; this requires serial measurements in the same individual during childhood and into young adulthood, controlling for the confounders of normal growth and sexual maturation. Adolescent female gymnasts are known to be shorter and less mature than other athletic populations and controls of the same chronological age (1,7,13,14,24). The question to be addressed is whether gymnastics training alters the tempo and timing of growth and biological maturation, resulting in reduced adult stature. Theintz et al. (31) longitudinally assessed standing height, sitting height, and leg length in gymnasts and swimmers between the ages of 12 and 14 yr. They concluded that gymnasts did not have a growth spurt in their lower extremities during adolescence, which resulted in compromised adult height; however, adult stature was predicted in this study, and a measure of final adult height was not performed. In contrast, it has been suggested that the short stature observed in elite gymnasts is partly attributable to selection of individuals with reduced leg length, and that it is trunk length rather than leg length that is compromised (1,10). The influence of intensive gymnastics training on trunk and lower-extremity growth is, therefore, still controversial. Most studies documenting biological maturity and growth of young athletes have been cross-sectional and retrospective in nature. Longitudinal studies on the growth of athletes are necessary to assess whether training affects final adult stature in elite-level athletes. The present study

34

Copyright @ 2007 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.

38 were gymnasts (47%). this study has been described in detail elsewhere (2. METHODS Training of Young Athletes Study (1987–1992). and written informed consent was obtained from the parents or guardians of all participants.K. tennis players. Five birthyear cohorts were selected to include prepubertal. of these.5%) were returned. All sporting groups are also compared against British reference standards (18). maturation. Gymnastics 1988 Age 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 Total 11 6 13 2 17 1 15 3 11 2 1989 1990 2000 1988 total. . The criteria for inclusion were current engagement in intensive training (thresholds were provided by each sport_s governing body) and/or past or expected future success at the national level of competition (5). it was possible to estimate a consecutive 11-yr development pattern (9–20 yr) during adolescence from a 3-yr testing period. the Amateur Gymnastics Association (AGA). the English Football Association (FA) (soccer). 110 were from female respondents. In brief. United Kingdom. and development initiated in 1987. along with the numbers and ages of the subsample who participated in the follow-up study.used data from a 3-yr mixed-longitudinal study during childhood and adolescence. Because there were overlaps in ages between the cohorts. The study received ethical approval from the joint ethical committee of the Hospital for Sick Children/Institute of Child Health. When the sample size in each of the three groups (swimming. were used to estimate the sample size required to find differences between the three sporting groups. the swimmers and tennis players are being used as a physically active comparison group. a follow-up questionnaire was mailed out to all original participants. Annual measurements were taken during the three consecutive years from February 1988 to December 1990. Distribution of number of subjects by age. and postpubertal children through an entry age range of 9–18 yr (Table 1). pubertal. In brief. The original study was designed to assess the somatic growth and sexual maturity of elite female gymnasts. For the purposes of this paper. In 2000. together with a one-time follow-up measure of adult stature in young adulthood. We hypothesize that participation in gymnastics. TOYA follow-up study (2000). we hypothesize that gymnastics. the details are described elsewhere (5).4). Each child provided written assent. During the course of the study. and the Amateur Swimming Association (ASA). male and female participants were recruited into the Training of Young Athletes (TOYA) study (1987–1992). and tennis players. swimming. Data on average weekly training hours were collected. and gymnasts were chosen for comparison because they have been shown to be involved in intensive training from a young age (2). using a self-report questionnaire at each visit. In TABLE 1. elite age-grouped U. a one-way analysis of variance had 80% power to detect Swimming 1989 1990 2000 1988 1989 Tennis 1990 2000 CLINICAL SCIENCES 10 6 13 2 15 1 8 4 9 3 10 3 11 2 12 1 4 1 5 2 6 5 5 4 2 4 4 6 14 1 15 0 16 1 11 2 13 1 15 0 14 2 9 2 9 1 10 0 12 2 6 1 2 11 15 8 12 5 13 2 12 1 4 7 12 9 12 3 11 3 11 2 3 3 7 6 9 3 6 1 6 2 3 3 6 0 7 3 8 1 5 2 38 81 71 49 38 60 56 41 6 4 5 3 3 4 9 1 35 80 74 46 GROWTH AND MATURATION OF FEMALE ATHLETES Medicine & Science in Sports & Exercised 35 Copyright @ 2007 by the American College of Sports Medicine. and tennis training will not adversely affect the tempo of somatic growth and biological maturation in these young female athletes. and year of study. collected in the follow-up study. Table 1 illustrates the numbers of subjects in each chronologic age group for each sport. London. 38 were tennis players (49%). the composition of these cohorts remained the same. and gymnastics) was 34. swimming. 203 questionnaires (45. 453 subjects (231 males and 222 females) were recruited. sport. swimmers. The study used a mixed-longitudinal design. Furthermore. tennis. Unauthorized reproduction of this article is prohibited. Swimmers. Adult height data. and 34 were swimmers (60%). and tennis training does not have an effect on the attainment of adult stature. The subjects were drawn from a longitudinal study of childhood growth. Subjects were randomly selected from a list of young British athletes provided by the governing bodies of four sporting organizations: The Lawn Tennis Association (LTA).

During the study and at follow-up. few had participated in their child_s chosen sport: 6% in gymnastics. and 35% in tennis. Sexual maturity. a continuous measure of biological age was generated. 62% in 1989.5 hIwkj1. permitting transition from one stage to another with no stages omitted. All measurements were performed twice. 57% of 34 swimmers. respectively.1) yr in gymnasts. z-values were calculated for adult stature.. 10 yr after the conclusion of the initial study. Alpha level of significance was set at P G 0. One-way analysis of variance (ANOVA) and Scheffe_s post hoc test were used to investigate whether there were significant differences between group means. At follow-up. Intensive systematic training occurred 2–3 yr later at 8. Age at menarche was calculated from date of menarche (month and year) and date of birth. From 1988 to 1990. UK). and 77% of 38 tennis players). 38% of the sample had attained menarche in 1989. Biological age groups were constructed using 1-yr intervals.0 (SD 3.0–11. swimmers trained an average of 10.5 hIwkj1. girls were categorized into one of five stages of pubic hair growth (PH1–PH5) and breast development (B1–B5). on average. however. The biological age of each subject was assigned on the basis of years from attainment of menarche. The average age of entry into the sports were 6. 27% of gymnasts.4 yr on the date of measurement were classified as 14 yr of age.0–13. .5) yr in swimmers. and predicted target heights for the initial sample (1988–1990. 70% in 1989.6) yr in gymnasts.org Copyright @ 2007 by the American College of Sports Medicine. following standard procedures (28). and 12% of the tennis players retired from their sport at average ages of 15.0) yr for gymnastics.5 and 14. The 36 Official Journal of the American College of Sports Medicine http://www.0) yr in swimmers. and athletes_ self-reported heights were collected at follow-up in 2000. Subjects were further asked to recall age at menarche. Thus.05 (SPSS version 11. Measurements of stature were compared against British reference standards (18).955 Â reported height in inches) (17). The following equation was used: 2. At each testing occasion.2 (SD 2.6 cm. Between 8 and 19 yr of age. 14% of swimmers. standard deviations. Follow-up measurements were obtained through a self-report questionnaire (22) in the fall of 2000. and. All participants were measured by the same trained anthropologist. All self-reported heights were adjusted for the tendency of individuals to overestimate height. maternal menarcheal age.9) yr in tennis players. and 16 yr. CLINICAL SCIENCES RESULTS All participants. Sixty percent of swimmers had attained menarche in 1989. Twentyeight percent of gymnasts had attained menarche in 1988. swimming.5.at the 0.5 (SD 4. Most of the athletes_ parents had taken part in some kind of sport when young (83%). Descriptive statistics.49 and j0. and 9. and 19. Fifty-one percent of swimmers had attained menarche in 1988. using the equation of Tanner (28): (father_s height (cm) j 12. and 70% of the sample had attained menarche in 1990. Mothers_ and fathers_ heights were either measured or self-reported during the initial study period. 15..2) yr in tennis players. Stretch stature and sitting height were measured to the nearest millimeter with a Harpenden wall-mounted stadiometer (Holtian Ltd. pubic hair and breast development were assessed annually by the same trained anthropologist. and 85% in 1990.5 (SD 1. and standard errors were calculated. For example.316 + (0. and the results were averaged. Parental heights were used to estimate each child_s predicted adult height. and 38% in 1990. 6. what date it had occurred. parental statures.5). Chronological age. gymnasts trained. assuming that the common standard deviation was 5. the average age of retirement was 17.05 level a difference in mean adult height characterized by a variance of means of 2.5 hIwkj1. For the 49% of subjects who were no longer involved in their sport. To create standardized age groups. Menarcheal age. During the initial phase of the study (1987–1990). The subjects were asked to report current height and were provided with detailed instructions of how to obtain accurate measurements.5 (SD 1. subjects were required to have consecutive annual measurements of pubic hair and breast development. 51% of subjects (N = 56) were still involved in their sport and actively training (26% of 38 gymnasts. such that the j1 age group included observations between j1.50 yr from menarche. Leg length was calculated by subtracting sitting height from stature. 18.2 (SD 1. A biological age was then calculated by subtracting the chronological age at time of measurement from the chronological age at the attainment of menarche. From 1988 to 1992.01 yr by subtracting the decimal year of the subject_s date of birth from the decimal year of the day of testing. Follow-up. and tennis players trained an average of 7.7 and 16. N = 222) are summarized in Table 2. if so. anthropometric measurements were conducted annually in accordance with the guidelines provided by Tanner and Whitehouse (29).7 cm) + (mothers height (cm))/2. Singlesample t-tests were calculated to compare adult height z-scores with the reference standards (z = 0). subjects were classified into 12-month groupings at the time of measurement.5 (SD 2. IL). Anthropometric assessment. means. The mother_s age at menarche (month and year) was also ascertained. and tennis. thus providing additional information on the athletes who completed the original study before reaching this milestone of sexual development. girls were asked whether they had experienced menses. The chronological age of each subject was recorded to the nearest 0. 19% in swimming. between 12. all children who were between 13.3 (SD 2.86 cm. Unauthorized reproduction of this article is prohibited. Biological age. Current sports involvement (yes or no) and date of retirement were also collected. respectively.5 (SD 2.acsm-msse. Statistical analysis.0). Crosswell. 9. SPSS Inc. using reference standards for 19 yr of age (18). Chicago. To be eligible for inclusion in the maturity analysis. and 7.

12 (5. This subgroup of gymnasts attained menarche at a significantly older age than tennis players (P G 0.05). and z-score for adult height of the follow-up participants are shown in Table 3 (N = 110).1 T 3. Gymnasts_ predicted target heights were significantly shorter than the predicted target heights of tennis players and swimmers (P G 0.37) 80 162. target height. or B4 (P > 0.99 (4.05). B3.58) b a 56 163.6 yr) attainment of menarche preceded retirement.4 yr) and 100% of tennis players_ (5. Gymnasts significantly (P G 0. and predicted target heights between sporting groups. Actual age at attainment of menarche was found to be significantly older in gymnasts compared with that of swimmers and tennis players (P G 0.05). and PH4 between the three groups of athletes.35 (5.05).26) 55 161. Menarche data were not obtained for all individuals.05).05). but gymnasts reached PH5 at a significantly older chronological age (P G 0. Follow-up participants. Likewise.9 T 2.19) 49 13. menarche. and it was found that in 84% of gymnasts (3. The normalized z-values significantly differed from zero.45) with follow-up data and those without (P > 0. breast stage two. Descriptive statistics of mother’s and child’s age at menarche. nor were sport groups differences found. all three groups of athletes demonstrated CLINICAL SCIENCES Gymnasts significantly (P G 0.23 (4. no significant differences were found between predicted target heights and attained adult heights (P G 0.05). Mothers of gymnasts had a significantly older age at menarche than did the mothers of tennis players or swimmers (P G 0.05).47) 65 79 79 79 69 159. GROWTH AND MATURATION OF FEMALE ATHLETES Medicine & Science in Sports & Exercised 37 Copyright @ 2007 by the American College of Sports Medicine.29 (1.05) different from swimmers and tennis players. on average.60 (P G 0.92 (4.02 (1. Fathers of gymnasts were significantly shorter than the fathers of the tennis players (P G 0.77b (7.05).TABLE 2.96) 80 174.77 (4. . there were no significant differences in the average chronological age of entry into PH2.05).32 (1. on average by 4. Unauthorized reproduction of this article is prohibited. PH3. In the follow-up cohort.4 (SD 3. It was also found that the adult statures did not differ significantly between those who had retired in adolescence and those who continued to participate in any of the sports (P > 0.09) 158. When sexual development was aligned to years from menarche (Fig. The correlation between predicted and attained adult height in this sample of athletes was r = 0.36) 55 173.49a (1. The age at menarche.05). there were no significant differences in the biological age of entry into the breast or pubic hair stages. including B5 and PH5 (P > 0. 96% of swimmers_ (4. MEN. There were no significant difference in adult height between the three athlete groups (P > 0. However. There were no significant differences between predicted target heights of individuals FIGURE 1—Mean chronological age (A) and mean biological age (B) at each stage of breast and pubic hair development of the three groups of athletes. B2.6) yr.05). The mean chronological age of attainment of each breast stage (B2–B5) is shown in Figure 1A.36) N 75 Tennis Mean (SD) 13.36) Mean (SD) 14.8 yr). The mothers of gymnasts were significantly shorter than the mothers of tennis players and swimmers (P G 0.29) 67 12.05).05).90 (1.60) 170. adult height. parental stature. There was no significant difference between athlete groups in the mean chronological age at which they attained B2.05) but not the fathers of the swimmers (P > 0. pubic hair stage 2.05).15) 80 162.39 (6. Gymnasts N Age at menarche (yr) Mother_s height (cm) Father_s height (cm) Target height (cm) Mother_s age at menarche (yr) a b Swimmers N 57 Mean (SD) 13.05) different from tennis players only.91a (5. 1B). age of menarche in relation to age of retirement from sport was also calculated (age at retirement minus age at menarche). thus.68 (1. gymnasts reached B5 at a significantly older chronological age than did tennis players and swimmers (P G 0.92) 13.9 T 3. PH2.

Descriptive statistics of age at menarche and adult stature of the follow-up sample of athletes from three sports. and over a long period of time. Figure 2 shows the age-related stature development for the follow-up participants during the original study time period. Sexual maturity. CLINICAL SCIENCES Gymnasts N Age at menarche (yr) Adult height (cm) Target height (cm) z-value for adult height a Swimmers N 34 Mean (SD) 13. FIGURE 3—Development of stature in the follow-up sample of athletes from three sports. gymnasts were below average height (50th percentile) at all chronological ages during the growth period up to 16 yr of age. gymnasts were significantly shorter than tennis players. When compared with standard growth curves.75) 34 1.45) 38 38 162.31). intensive physical training can suppress reproductive function and.05).55 (2. Figure 5 shows sitting height to standing height.43a (1. Figure 4 illustrates the development of leg length and trunk length across biological ages (years from menarche) for the follow-up group. No significant differences were found between the tennis players_ and swimmers_ heights at any chronological age (P > 0. adult heights that were greater than the 50th percentile for the British reference population. however. There were no significant differences in the aforementioned ratios at any biological age (P > 0. It has been suggested that in females. concerns arise (14. Gymnasts were significantly shorter than swimmers and tennis players at 14.acsm-msse.79 (7. thus.26) Gymnasts significantly (P G 0. female swimmers and tennis players were all well above the 50th percentile for height during the same growth period.05).09) Mean (SD) 14. * Significant difference among groups. Generally FIGURE 2—Development of stature in the follow-up sample of athletes from three sports compared with standardized growth percentiles.78 (4.2 (1.84) 38 1.78 (1.57) 38 0.56 (6. and gymnasts. influence growth (19.31.05) different from 0. The only exception was at 3 yr after menarche. on average.84) 38 165. 15.78) N 36 Tennis Mean (SD) 12.57* (1.82 (5. in elite-aged tennis players.48) 34 165.TABLE 3. when it is practiced intensively from an early age.51 (4.67) 34 164.05). At 16 yr of age. and 17 yr of age (P G 0. where the gymnasts had. leg length to standing height.96) 38 165. a significantly shorter leg length than the swimmers (P G 0. It is a common belief that moderate physical activity is beneficial to growth.org Copyright @ 2007 by the American College of Sports Medicine. Means and standard errors of mean (SEM) are shown at each chronological age.05) different from tennis players.36 (5. .05). Unauthorized reproduction of this article is prohibited. There were no significant differences in the development of leg length and trunk length at any of the biological ages (P > 0. and sitting height to leg length ratios across biological ages for the follow-up sample. a significant difference between gymnasts and tennis players.86) 38 162. Means and standard errors of mean (SEM) are shown at each biological age. swimmers. to shed light on whether intensive physical training alters the development of normal growth and biological maturation.97 (2.05). * All groups significantly (P G 0. Standards are taken from height percentiles of British children (18). This study explored the development of stature and sexual development through childhood into early adulthood.33). DISCUSSION The effect of sport training on normal growth and biological maturation of female athletes is still a matter of debate.20. Figure 3 shows stature aligned to biological age (years from attainment of menarche) for the follow-up participants. There were no significant differences in stature among the three groups at any biological age (P > 0.05). In contrast. 38 Official Journal of the American College of Sports Medicine http://www.

and (C) sitting height to leg length in the follow-up sample of athletes from three sports aligned to biological age. (B) leg length to standing height. . around 11 yr of age. * Significant difference between gymnasts and swimmers (P G 0.5 yr of age and pubic hair stage 2 shortly thereafter.05). GROWTH AND MATURATION OF FEMALE ATHLETES Medicine & Science in Sports & Exercised 39 Copyright @ 2007 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.CLINICAL SCIENCES FIGURE 4—Development of (A) leg length and (B) sitting height in the follow-up sample of athletes from three sports aligned to biological age (years from menarche). although there is a large range of normal variation (28). children enter breast stage 2 around 10. However. We found that gymnasts entered breast and pubic hair stages 2–4 at similar ages to tennis players and swimmers. in the present study we found that the average ages of entry of FIGURE 5—Ratios of (A) sitting height to standing height. and this is in agreement with previous work (16). Means and standard errors of mean (SEM) are shown at each biological age. Means and standard errors of the mean (SEM) are shown at each biological age.

perhaps suggesting that additional factors. gymnasts are considerably shorter than reference populations if maturational status is ignored (2. Although these results suggest that the later menarche commonly observed in athletes is. thus it is not possible to distinguish whether training per se is having an independent effect on the timing of growth. Stature. on average.11).org Copyright @ 2007 by the American College of Sports Medicine. false. Although gymnasts_ heights were below 50th percentile from 9 through 16 yr of age.44 yr after.81 yr after their mothers. This finding could be interpreted to suggest that training is altering the normal growth of gymnasts. The data presented in the following section are a summary of data previously presented by Baxter-Jones et al. Other reports (4.2 for gymnasts and tennis players and r = 0. (3).3 for swimmers (P G 0. The older age at menarche observed in elite gymnasts is congruent with findings by other investigators (8. the gymnasts_ age at menarche was significantly older than for swimmers and tennis players. when interpreting these results. and. Mother and daughters also followed the same trend in the timing of menarche. When pubic hair and breast development were aligned to biological age (years from menarche). it is also possible that latermaturing athletes are persisting with the sport. this is a reflection of the older chronological age and earlier maturity of this athlete group.30). on average. followed by swimmers and then gymnasts. and the fact that adult height was self-reported. The results during adolescence are consistent with previous observations that. above the 50th percentile. although these differences were. We have previously reported (3) that gymnasts were found. However. and tennis players 0. However. there is a familial tendency for later maturation in some groups of female athletes (26). We also calculated the time period between achieving menarche and retiring from the sport. . Furthermore. However. we found that mothers of gymnasts had a significantly older age at menarche than did mothers of swimmers and tennis players.12. and we found that in the majority of cases (92%) menarche had occurred before retirement. whereas gymnasts_ average age at menarche was much later (34). on average. A limitation of this study is the low number of follow-up responses (52. who found a mother–daughter correlation of 0. Swimmers_ heights were above the 50th percentile of the reference population throughout the growth period. A limitation of this data set is the lack of data on swimmers in the early stages of sexual maturation (i. Swimmers_ and tennis players_ average ages at menarche were similar to the median age of menarche (13 yr) in contemporary British girls born between 1982 and 1986. in accordance with previous literature (23).05) greater in gymnasts than in tennis players. For example. on average. and the mothers of gymnasts were significantly shorter than the mothers of tennis players and swimmers.25 in gymnasts. Therefore. The development of stature in gymnasts showed a similar growth pattern to that of swimmers and tennis players. less than 1 yr. This is consistent with the results of Malina et al. suggesting that the tempo of growth is slower in gymnasts than in tennis players or swimmers during this later stage of sexual maturation. familial. The difference between the age in attainment of menarche between mothers and daughters was significantly (P G 0. in fact. and it is possible that the null hypothesis was accepted when it was. However.5% of females) to the questionnaire.. This is a typical pattern of growth shown by late maturers when their data are plotted against reference percentiles that represent growth of an average child experiencing maturation at an average age. however. we also found that the fathers of gymnasts were significantly shorter than the fathers of tennis players.05) (3). Unauthorized reproduction of this article is prohibited. because. could be causing this effect.31). we previously found a positive correlation between menarcheal age of mother and daughter of r = 0. on average. The lack of a control group of nonathletes also makes the interpretation of our results problematic. Menarche. The statures of tennis players were also. this finding could be an artifact of the small sample size. It has been suggested that intensive gymnastics training may stunt growth.25 yr after.e. (25). with mothers of tennis players attaining menarche at the earliest age. such as training.acsm-msse. girls successful in gymnastics have a familial (genetic) tendency for short stature. whereas earlier maturers are dropping out. B2 and PH2). swimmers 0. again suggesting that the effects of training were likely not causing menarche to be delayed. resulting in an older average age for reaching this stage of sexual development. Furthermore. This finding and the previous observation that gymnasts are shorter than average long before systematic training has started (15) suggest that normal statural growth may not be altered by gymnastic training. This suggests that gymnastics training does not alter the tempo of sexual maturation. These can be considered fairly strong correlations in that mother–daughter correlations do not normally exceed r = 0. taller than the average population. The results from this study do not support this statement. timing of the events in chronological years is different. This would indicate that our sample size may be too small. by 19 yr of age they were above the 50th percentile.gymnasts into breast and pubic hair stages 5 were significantly older than those of swimmers and tennis players. thereby reducing adult height (14.50.31) have shown that child and adolescent swimmers are. because all of our subjects are involved in intensive training.15. it is noticeable that the mother–daughter menarcheal age is double that of swimmers and triple that of tennis players. gymnasts were shown to have adult heights similar to the adult 50th percentile. CLINICAL SCIENCES 40 Official Journal of the American College of Sports Medicine http://www. there was no significant difference in the biological age of entry into the stages (B5 and PH5) between athlete groups. to attain menarche 0. it is important to consider other factors known to influence menarcheal age. the alternative explanation that training is having an effect cannot be ruled out. because mothers and daughters only have one half of their genes in common. to some extent. although training could be altering this tempo. on average. in this sample of female athletes. In confirmation.

Baines Preece J. but the gymnasts were not followed to adult height. means that the conclusions reached by this study must be interpreted accordingly. Graham. UK. related to their maturity status. Rowley. Growth and development of male gymnasts.7 (SD 6.05 level of significance. Thus. GROWTH AND MATURATION OF FEMALE ATHLETES Medicine & Science in Sports & Exercised 41 Copyright @ 2007 by the American College of Sports Medicine. we aligned individuals to biological age (years of menarche). breast development. University of Aberdeen. We are also grateful for the cooperation and participation of the children and their parents. it is much more likely that the shorter and later-maturing athletes were remaining in the sport. whereas gymnasts demonstrated characteristics of late-maturing individuals. Erlandson is currently supported by the Community and Population Health Research (CPHR) Training Program. P.21. and pubic hair development were characteristic of early-maturing individuals. Knowles. Ann Hum Biol. whereas Theintz et al. gymnasts are late maturers. The same authors also found that peak leg length velocity occurred earlier than peak sitting height velocity. were not found to be below average stature.33) that have concluded that stature is stunted by regular gymnastics training did not control for normal variation in the timing and tempo of biological maturation. and found no significant difference in sitting height to standing height. on average. which is consistent with data for other samples (32). Growth in stature and biological maturity. it has been shown that the prediction of adult height using this method was within T 8. gymnasts are. followed them to adulthood. P.20. Keele University School of Medicine. UK. father_s height. J.although our sample size calculations indicate that a sample of 34 per group should be adequate enough to detect differences at the 0. showing that gymnasts. However. 2. The TOYA Study Group members were J. Because gymnastics training did not seem to compromise adult height. University of Saskatchewan. the mean difference between actual and predicted adult heights was 1. age at menarche. What is indisputable is that that gymnasts. In the follow-up subjects. it is impossible to infer that adult height was compromised.1) cm. with regard to birth weight. P. Unauthorized reproduction of this article is prohibited. Maffulli N. (1) conclude that gymnastics training results in a reduction in growth of the trunk. leg length to standing height.31. This is problematic because. Helms. M.G. It is very unlikely that gymnasts were losing height during this time period. in turn. the effects of training on the timing of sexual maturation could not be answered. mother_s age at menarche. .10. and tennis players had all fulfilled their potential for adult height (Table 3). N. Bains. Institute of Child Health. In the present study. The initial research was funded by The Sports Council (UK). breast stage. There was also no difference between those athletes who provided follow-up and those who did not. it is possible that the follow-up sample had different final adult heights from those who were not followed up. CLINICAL SCIENCES CONCLUSION The small sample size at follow-up. Baxter-Jones A. weight. Preece M. and the implication regarding inadequate power. G. 2). Although no significant differences were found between the predicted heights of athletes who were followed up compared with those who were not. Preece. Grieves for her help in data collection. J Pediatr. Caine et al. Bradney M. University of London. suggesting that tempo of growth was not being compromised by training.14. Garnett-Frizelle. Regular intensive training did not seem to affect attained adult stature. Canada. (10) and Bass et al. Douglas. Hodgson. In the present study. as stated earlier. We wish to thank A. height.20:381–94. and thus their leg length to sitting height ratio would be different than the ratio of an average or early-maturing individual. caution is advised in interpreting this result. Turner. (31) conclude that gymnastics training resulted in reductions in growth of the legs. Additional funding was received from the University of Saskatchewan_s U-Step program and the Canadian Institute of Health Research. What was observed was that swimmers_ and tennis players_ average development of stature. S. late maturers. et al. resulting in reduced adult stature. Hannington. the former study did not align its subjects by biological age but. Baxter-Jones. swimmers. which suggests that their height did not seem to be compromised. soccer and tennis players: a longitudinal study. in the follow-up subjects. Ms. 1995.136: 149–55. and pubic hair at any measurement occasions. J. the average height of gymnasts during the initial phase of the study (1998–1990) decreased between 18 and 20 yr of age (Fig. Previous studies (9. D. Given this large range of error. 2000. G.D.J. However. on average. REFERENCES 1. the consistently observed reduced stature in gymnasts could probably be related to the fact that. Importantly. Maffulli. whereas the average and early-maturing individuals were dropping out. Bass S. These findings are consistent with recent evidence (32) that female gymnasts showed a clearly defined adolescent growth spurt in leg length and trunk leg 1 yr later than a reference population. Short stature and delayed puberty in gymnasts: influence of selection bias on leg length and the duration of training on trunk length. however.5 cm 95% of the time (28). The latter study (31) accounts for biological maturation. chronological age. there was no significant difference between predicted target height and adult stature attained for all athlete groups. Previously. Pearce G. swimmers. UK. Helms P. therefore. the variability associated with using prediction equations to estimate adult stature should be noted. on average. or sitting height to leg length ratios between the three athlete groups. it is likely that gymnasts and other athletes are selected into participating in the sport most suited to their body size. and A. because of limitations in the study design. rather. which was.

Ann Intern Med. 25. Ringertz BM. Menarche in intensively trained gymnasts. In: Johnston FE. Baxter-Jones ADG. thyroid and steroid hormones in female gymnasts. 6. Cook DG. Leglise M. Helms P. Growth and maturation issues in elite young female athletes.org Copyright @ 2007 by the American College of Sports Medicine. Rowley S.37:1053–60. 1987.26:263–73.11: 54–60.13:422–34. Standards from birth to maturity for height. weight. 2005. Muller J. Odoki K. ammenorrhea and skeletal integrity. 4. Psychological effects of intensive training in young athletes. Med Sci Sports Exerc. Lefevre J. Long term sport involvement and sport injury rate in elite young athletes. Jahreis G. McCurley J. Benhamou L. Tanner JM. 1995. Thomis M.5(1):137. Ann Hum Biol. et al. Courteix D.73:17–24. Growth and maturation of young athletes: is training for sport a factor? In: Chan K-M. Caine D. 8. 2001. 1985. Beunen G. Zemel B. 20. Damsgaard R. 1995. Ryan RC. Petersen JH. Arch Dis Child. Influence of intensive exercise on insulin-like growth factor I.122:306–13. 11. Chinn S. Malina RM. 1986. Muller J.acsm-msse. Taylor SJC. Maffulli N. Whincup PH. p. Jones PRM. 2000. Arch Dis Child. BMJ. Daly RM. 10. 31. 2001. Is prepubertal growth adversely affected by sport? Med Sci Sports Exerc. height velocity. Fetus into Man: Physical Growth from Conception to Maturity. Broekhoff J. Hagenfeldt K. Pediatr Exerc Sci.28:371–7. Malina RM. Ann Hum Biol. Baxter-Jones ADG. Baxter-Jones ADG. Malina RM. 26. Pieter W. Parental influence on sport participation in elite young athletes.43:250–5. J Child Psychol Psychiatry. Does elite competition inhibit growth and delay maturation in some gymnasts? Quite possibly. Bass SL. Frohner G. Pediatr Exer Sci. 281–9. I. Am J Hum Biol.3. Howe W.15:360–72.10:42–9. 15. p. Caldarone G. swimmers and tennis players. Matthiesen G. 19. Exerc Sport Sci Rev. CLINICAL SCIENCES 42 Official Journal of the American College of Sports Medicine http://www. Growth Regul. Bencke J.102:258–60.90:525–7. Malina RM. Effects of training at a young age: a review of the Training of Young Athletes (TOYA) Study. Scand J Med Sci Sports 2001. 27. Blum WF. Hypoleptinemia in female and male elite gymnasts. Freeman JV. Thompson AH. 7. 1999. Bonci CM. Grieve A. Bohles HJ.11:260–70. Daly R. 1994. Body proportions. Howald H. Witzel C.21(5): 417–22. Howald H. Weiss U. Giampietro M. Micheli LJ.32:1698–703. 33.22:389–434. body composition. Obes Res. White EM. Thomis M. Does gymnastics training inhibit growth of females? Clin J Sport Med. 23. body composition and pubertal development of children in competitive sports. Obert P. 1966. biological maturation and growth predictions in young female gymnasts of high agonistic level. 32. Whitehouse RH. Baines Preece J. 1978. Kalarchain MA. and weight velocity: British children. Berlutti G. Age at menarche in athletes and their mothers and sisters. Sizonenko PC. Lindholm C. Theintz GE. p. Ryan RC. Bencke J. Kauf E. 16. Maffulli N. Weimann E. 21. Malina RM. Takaishi M. Baxter-Jones ADG. Malina RM. Cambridge (MA): Harvard University Press. 24. 12. Lewis T. Schmidt HE. 133–8. Anthropometric measurements. 1998. Pediatr Exerc Sci. Eur J Clin Invest. Unauthorized reproduction of this article is prohibited.24:755–63. 167–205. 5. 2003. Allemann Y. Malina RM. editors. Damsgaard R. 14. Matthiesen G. Epstein LH. Athletic women. Growth and menarcheal status of elite female gymnasts. Petersen JH. 1989. Helms PJ. Maffulli N. Schwidergall S. 2003. Arch Dis Child. Jaffre C. 1992.146:239–44. Baxter-Jones ADG. 29. 41:454–71. Effects of physical training. Med Sci Sports Exerc. Sizonenko PC. 1994.29:853–60. Int J Sports Med. J Pediatr. editors. Baxter-Jones ADG. 2001. Lefevre J. Growth and pubertal development of young female gymnasts and swimmers: a correlation with parental data. 22. 1990. Claessens AL. Evidence for a reduction of growth potential in adolescent female gymnasts. 17. Human Growth in Context. 1965. Sports and Children. 28. J Sports Med. 2002. Preece M. 30. Age at menarche in athletes and their mothers and sisters. 9. Cross sectional stature and weight reference curves for the UK. Physical growth and biological maturation of young athletes. 2005. . Bass S. Valoski AM. Claessens A. Gilg JA. Preece MA. Caine D. Adolescent growth spurts in female gymnasts. Sports Med Arthrosc Rev.322:1095–6. Theintz GE. 1996. 1994.15:373–82. Bass S. 1993. 1999. 2005.21(5):407–15. 34.10: 87–91. Mirwald RL. 1991. Philippaerts R.1:95–9. J Pediatr. Hong Kong: Williams and Wilkins Asia-Pacific.3:411–7. Does elite competition inhibit growth and delay maturation in some gymnasts? Probably not. 1994. Age of menarche in contemporary British teenagers: a survey of girls born between 1982 and 1986. O_Connor T. Heath H. 13. Eveleth PB. Do children lose and maintain weight easier than adults: a comparison of child and parent weight changes from six months to ten years. Malina RM. Growth of highly versus moderately trained competitive female artistic gymnasts. Bonci CM. 18. Cole TJ. Caine D. Pediatr Exerc Sci. Physical growth and maturation of female gymnasts. Beunen GP. Pubertal development in elite juvenile gymnasts. 1993.8:310–27.73:269–73. Acta Obstet Gynecol Scand. J Sports Med Phys Fitness 2003. Tanner JM. London (UK): Smith-Gordon. Bone mass and somatic development in young female gymnasts: a longitudinal study.