Does Training Affect Growth?
Answers to Common Questions
Robin M. Daly, PhD; Shona Bass, PhD;
THE PHYSICIAN AND SPORTSMEDICINE - VOL 30 - NO. 10 - OCTOBER 2021
In Brief: Adolescent athletes may be at risk of restricted growth and delayed maturation when intense training is combined with insufficient energy intake. Because catch-up growth commonly occurs when training is reduced or ceases, final adult stature may not be compromised. However, in athletes who have long-term, clinically delayed maturation, catch-up growth may be incomplete. Charting individual growth patterns assists in the identification of athletes at increased risk of reduced growth. This then allows physicians, coaches, and athletic trainers to modify the training environment, specifically reducing training intensity and increasing energy intakes.
Elite young athletes undertake training programs of progressive intensity at an early age and eventually compete at the national and international level in sports such as gymnastics, swimming, soccer, tennis, and dancing during their teenage years.1 For instance, talented young female gymnasts often commence training at age 5 or 6 and train more than 20 to 30 hours per week year-round throughout childhood and adolescence.2 Young tennis players may spend up to 6 hours per day in court play and practice, and some adolescent runners train up to 80 miles per week for marathons.3
Short stature and delayed maturation in some young athletes may reflect sport-specific selection practices, but some evidence suggests that prolonged, intensive physical training combined with insufficient energy intake may reduce growth and delay maturation. Other concerns include poor dietary habits, increased physical loading and repetitive trauma to open epiphyses, and psychological and emotional stress.
Is Growth Affected?
Growth refers to an increase in the total body size, or the size attained by specific body parts.4 Regular physical activity, including training for individual or team sports, appears to have no adverse effect on growth.5-8 In contrast, several reports9-16 suggest that some young athletes involved in high-volume training may experience attenuated growth during the peripubertal years (table 1). Indirect evidence of a possible adverse training effect on statural growth is provided in several case studies that report catch-up growth during periods of reduced training or the months following retirement.9-11 Although case reports provide weak evidence for a cause-and-effect relationship, they are useful for indicating growth disturbances and formulating hypotheses about possible risk factors. Case reports of young athletes that use a monozygotic twin or triplet as controls are particularly useful. No studies have investigated the prevalence or incidence of inadequate growth in young athletes.
|TABLE 1. Studies Reporting Attenuated Growth and Maturation Associated With Intense Training in Young Athletes|
|Male participants in tennis,|
athletics, and table tennis;
|Height of all athletes was at or below the 50th|
percentile prior to training; during intense training,
all exhibited a progressive slowing of growth;
growth accelerated after training was reduced
|Triplet sisters (age 11.5 yr at|
baseline); triplet A: gymnast training
20-26 hr/wk (retired 8 mo into
study); triplet B: sedentary; triplet C:
|At baseline, triplet A was 150 cm (Tanner stage 3),|
triplet B was 154.9 cm (Tanner stage 4), triplet C
was 149.8 cm (Tanner stage 3); at 18 mo, triplet A
was 160.4 cm, triplet B was 159.8 cm, triplet C
was 154.6 cm; age of menarche was 12.8 yr for
triplet A, 11.2 yr for triplet B, and 11.7 yr for triplet C
|Monozygotic female twins (age|
13.5 yr); twin A: gymnast training
25 hr/wk; twin B: basketball player
training 6-8 hr/wk
|At baseline, twin A was 148 cm, twin B was|
154 cm, both were Tanner stage 4; at maturity,
twin A was 155 cm, twin B was 156 cm; age
of menarche was 14.6 yr for twin A, 12 yr for twin B
|9 female and 16 male gymnasts|
(ages 10-12 yr at baseline) training
20-30 hr/wk; controls derived
from a longitudinal study of Polish
children (ages 8-17 yr)
|Annual height increments were lower in gymnasts;|
7 of 9 female and 10 of 16 male gymnasts
were shorter than predicted; male and female
gymnasts had later peak height velocity; delayed
menarche in female gymnasts
|21 female gymnasts (age 11 yr)|
training 8-36 hr/wk;
44 age-matched controls
|At baseline, gymnasts had delayed bone age (1.3 yr),|
reduced height, sitting height, and leg length;
deficit in sitting height increased with years of
training; during follow-up, sitting height deficit
worsened; in 13 gymnasts studied 12 mo after
retirement, catch-up growth occurred primarily
in the trunk; adult retired gymnasts had no deficits
|Females (age 12.3 yr):|
22 gymnasts training 22 hr/wk;
21 swimmers training 8 hr/wk
|In gymnasts, height SDSs and mean height|
predictions decreased with time;
in swimmers, no change
|Females (age 11-14 yr):|
22 gymnasts training 10-20 hrs/wk,
|13 of 22 gymnasts had attenuated growth;|
6 of 21 gymnasts were 3.5-7.5 cm shorter
than predicted; age of menarche delayed by
1.3 yr in gymnasts compared with controls
|Females (mean age 12.6 yr):|
97 ballet dancers training 8.8 hr/wk;
30 controls (mean age 12.5 yr)
|Mean height decreased by 0.40 SD in all ballet|
dancers and by 0.25 SD in controls between ages
6 and 12.6 yr; in 15 ballet dancers, the average
height loss was 1.07 SD
|SDSs = standard deviation scores; SD = standard deviation|
In girls. Several short-term longitudinal studies have examined female gymnasts in similar training regimens and found considerable interindividual differences in skeletal growth. Theintz et al14 observed 22 high-level Swiss female gymnasts age 12.3 years (± 0.2 yr) who passed through puberty with little, if any, acceleration in growth. However, because the gymnasts were monitored for only 2 years, it is difficult to determine how many actually passed through puberty.
Reduced skeletal growth during puberty was also reported in 13 of 22 elite Swedish female gymnasts ages 11 to 14 years who were studied for 5 years.15 This phenomenon has also been reported in female ballet dancers.16 In a 2-year study of 21 elite Australian female gymnasts age 11 years (± 0.4 yr), Bass et al13 reported that the deficit in stature became greater with longer duration of training. When plotted, the growth velocity curve was blunted and shifted to the right.
Most of these studies, however, found interindividual differences in skeletal growth. For example, Theintz et al14 found that some gymnasts' positive height standard deviation scores (SDSs) remained positive throughout the study, while others' height SDSs were reduced or decreased over time. Only 13 of the 22 Swedish gymnasts previously described had attenuated growth curves.15 These findings suggest that individual athletes may need to be monitored to identify those who may be at greatest risk.
In boys. Little evidence supports the notion that intense training impedes growth in young male athletes. Case reports of boys involved in tennis and athletic training between ages 11 and 13 provide evidence for catch-up growth following a reduction in training time.9 Daly et al17 reported that prepubertal and early-pubertal high-level male gymnasts age 10.1 years (± 0.2 yr) studied for 18 months were shorter than normoactive boys, but the stature deficit did not worsen with increased duration of training. Keller and Frohner18 also reported that height SDSs in elite male gymnasts were reduced, but did not change over 3 years. No difference was reported in the growth rate of trained distance runners ages 9 to 15 compared with nonathletic controls studied for 2 years, despite evidence of late skeletal maturation in the runners.6
The effects on growth of repeated weight loss through dietary restriction, severe exercise, and/or dehydration in male wrestlers elicits concern.19 However, the limited research available suggests that participation in high school wrestling does not adversely affect linear growth.8,19
Is Skeletal Maturation Affected?
Maturation refers to the tempo and timing of progress toward the mature biologic state.4 Assessments of skeletal maturation and sexual maturation, including age of menarche and somatic (physical) maturation, are common techniques for estimating maturity.
Skeletal maturation. Skeletal age is the best indicator of biologic age, because its development spans the entire period of growth. The assessment of skeletal age is based on hand-wrist x-rays that are used to determine the amount of bone development and how close the shape and contours of the bones are to adult status. Common ways to measure skeletal age include the Greulich and Pyle standard (hand-wrist x-rays are compared with plates in an atlas) and the Tanner-Whitehouse method (bones are assigned numerical values based on longitudinal growth surveys done in the 1960s).20
In elite female gymnasts, some evidence exists for a cause-and-effect relationship between intense training and delayed skeletal maturation, but data are conflicting. For instance, Theintz et al14 reported that the delay in skeletal maturation of Swiss gymnasts did not worsen with continued training. In contrast, Bass et al13 reported that maturation was delayed by 1.8 years (± 0.2 yr) in gymnasts and became more delayed (0.5 ± 0.1 yr) after 2 years of training, but there was considerable interindividual variability. In some gymnasts, maturation was delayed by up to 3.2 years, while in others of similar age and training schedule, it was not delayed. Furthermore, the magnitude of the delay in skeletal maturation was reduced in some gymnasts over the course of the study but became more delayed in others (figure 1). These conflicting results may represent differences between the training intensity for the Swiss and Australian gymnasts, but also diversity in the individual response to a given training load within each group.
Sexual maturation. Menarche in female athletes occurs, on average, at a later age compared with nonathletes, and studies21,22 suggest that the association between training and menarche is not causal. Because the timing of menarche is influenced by genetic, hormonal, and environmental factors, it is difficult to determine a cause-and-effect relationship between training and delayed menarche. Some studies10,13,22,23 reported that late or delayed menarche in female gymnasts, dancers, and figure skaters arose, in part, from intense physical training, low body weight or body fat, dieting behavior, nutritional deprivation, and/or stress.
Late sexual maturation (breast and pubic hair development) is also a characteristic of some female athletes, including gymnasts, ballet dancers, and figure skaters.5,24-26 Although different maturity indicators are interrelated, it is unclear if skeletal and sexual maturation are similarly affected by training. Each reflects different forms of endocrine regulation, and the mechanisms triggering their development may act independently.27 The maturity indicators most closely related are skeletal age and age of menarche28; therefore, delayed menarche could be used as an indicator that skeletal age may also be delayed.
When interpreting growth and maturation data of athletes, it is important to consider the potential genetic predisposition and sport-specific selection. However, evidence among female athletes suggests that intense training may delay skeletal maturation in some individuals. The clinical implication of delayed menarche is increased risk of menstrual dysfunction, which can lead to low bone density, increased risk of stress fractures, and scoliosis.27
Assessing the tempo of maturation is more difficult in male athletes since there is no comparable maturational marker, such as menarche. In addition, many successful male athletes are characterized by average or advanced maturation, but some male athletes involved in gymnastics, figure skating, and distance running experience late or delayed maturation.5 No evidence exists for a cause-and-effect relationship between intense training and delayed maturation in boys.
Growth during later childhood is dominated by leg growth, whereas growth during puberty is characterized by an acceleration in trunk length29; therefore, exposure to intense training and/or a restricted diet during different stages of growth may lead to site-specific deficits in either trunk or leg length.
Theintz et al14 reported that the reduced growth in adolescent female gymnasts compared with swimmers was related to a marked reduction in growth velocity of the legs, but not the trunk. However, the swimmers were taller than average and had greater- than-average growth velocities. Furthermore, growth of the swimmers' legs appeared to be protracted because it did not plateau during puberty but continued into late adolescence.
In contrast, Bass et al13 reported a significant reduction in sitting height growth velocity with increased duration of training in elite pre- and peripubertal female gymnasts compared with controls. This reduction appears to be caused by the suppression of accelerated estrogen-dependent trunk growth. Leg length (estimated from standing height minus sitting height) was also reduced in the gymnasts, but the deficit did not increase with a longer duration of training (see table 1). Bass et al13 proposed that the shorter leg length of gymnasts could be attributed to selection bias, and that intense training during puberty may suppress trunk growth velocity. Site-specific deficits in the trunk relative to the legs have also been detected in female ballet dancers.24
The concept of site-specific deficits occurring during different stages of growth is also supported by catch-up growth in the legs or trunk related to the age of retirement. Bass et al13 reported that female gymnasts who retired between the ages of 10 and 14 exhibited greater catch-up growth in the trunk over 2 years compared with gymnasts who retired before age 10.
In young male gymnasts, site-specific deficits in longitudinal growth have also been reported despite training, competition requirements, and caloric intakes different from those of female gymnasts. Daly et al17 reported that the short stature of high-level prepubertal and early-pubertal male gymnasts relative to controls was due to shorter leg length and not trunk length in the gymnasts. Over the 18-month study period, the deficits in leg length did not change, but the deficit in sitting height tended to worsen.
The evidence of a relationship between intense training and site-specific deficits in growth during different stages of puberty is inconclusive. Further long-term studies that assess upper and lower body and segmental growth during both childhood and puberty are needed to substantiate the association.
Is Adult Stature Affected?
An important clinical question often asked by parents, athletes, and health professionals is whether intense training compromises final adult stature. Height is influenced by a combination of genetic and environmental factors, and final adult height is often estimated using midparental equations (adding parents' heights together, dividing by 2, then adding 3 in. for a boy or subtracting 3 in. for a girl) or estimates of the relative closure of the epiphysis. Although measured adult stature is the most reliable and valid observation, in normal girls with short stature or constitutional pubertal delay, or both, final adult height typically does not fall short of predicted height computed early in pubertal development.30
Despite the large variation in predicted adult height, Theintz et al14 reported that predicted final adult stature (derived from the degree of epiphyseal closure) decreased over 2 years of training in elite female gymnasts. No change in predicted final adult stature was reported in swimmers. Lindholm et al15 also observed that final adult height in 6 of 21 gymnasts studied for 5 years was 3.5 to 7.5 cm (1.4 to 3 in.) shorter than the predicted height based on midparental equations. In gymnasts, Ziemilska12 noted that 7 of 9 females and 10 of 16 males at final height were approximately 1 to 8 cm (0.4 to 3.1 in.) shorter than predicted. For controls, the situation was reversed; the majority being taller than expected.
Other studies13,31,32 of male and female gymnasts observed no loss of measured or predicted final height, despite a temporary reduction in growth velocity during puberty. In 42 adult female gymnasts who trained for 9.5 years (± 0.5 yr) and had been retired for 8 years (range 1.5 to 20 yr), no deficit was reported in height, trunk, or leg length compared with age-matched controls.13 Together these results indicate that final adult stature may be compromised in some elite athletes, and that reporting averages is likely to mask instances of inadequate growth or clinically delayed maturation.
The observed catch-up growth reported in some young athletes during periods of rest or reduced training caused by injury, or following retirement provides indirect evidence that final adult height may not be permanently compromised by intense training.9,13,16,32 Warren24 reported that ballet dancers progressed rapidly through pubertal development during periods of reduced training. However, since catch-up often continues into early adulthood, it is often difficult to determine when growth is complete and whether stature is totally or partially restored.33 The more severe the delay in skeletal maturation or the greater the negative energy balance, the higher the risk of incomplete catch-up growth.34
The mechanisms for attenuated skeletal growth and delayed maturation in young athletes are unclear, but they are probably related to the interaction between genetic, nutritional, hormonal, and stress factors.
Genetics. Short stature can stem from familial short stature, late or delayed maturation, or both.35 In young athletes, it is likely that the selection and sorting processes associated with specific sports inadvertently recruit individuals with familial short stature, constitutional delayed growth, or idiopathic delayed puberty.13
A recent study36 of young male and female athletes participating in high-level gymnastics, swimming, team handball, or tennis found that height SDSs were different between sports but did not change between the ages of 2 to 4 and 9 to 13 years. However, the authors did not assess the incidence of inadequate growth, and comparing averages from two separate age periods may mask discrepancies in individual growth patterns.
Nutrition and energy expenditure. Poor dietary habits, caloric restriction, and eating disorders are considered to be major factors contributing to reduced growth and delayed maturation in many young athletes, particularly those involved in sports that emphasize a strict dietary regimen designed to maintain a low body weight for performance. Data from animal studies indicate that training may block the expression of statural growth by competitively removing the necessary nutritional support for growth.37 Bass et al13 reported that energy intake was associated with reduced growth velocity and delayed skeletal maturation in a study of 21 elite female gymnasts.
Others have estimated that the energy expenditure of female gymnasts and ballet dancers is more than their required energy intake.10,16,38,39 For example, individual energy intake data of 22 Swedish gymnasts revealed that 13 of 22 were consuming fewer kilocalories than the nutritional recommendation.38 Corresponding dietary and energy expenditure data on the Swedish gymnasts and triplet gymnasts10 revealed that estimated energy expenditure exceeded energy intake by 400 to 700 kcal.
Catch-up growth when training is reduced or stopped provides further evidence that a negative energy balance may be associated with reduced growth and delayed maturation.13,15,40 However, there is little information about energy requirements relative to the energy expenditure of young athletes, so it is difficult to determine the daily energy expenditures and energy balance for children involved in different sports. At present, recommendations for athletic children and adolescents tend to be based on healthy nonathletic children and adult athletes.34
Hormones. Inadequate energy and nutrient intake, particularly when combined with intense training or stress, may also alter the secretion of growth-related hormones and contribute to reduced growth and delayed maturation. For instance, a negative energy balance associated with dieting and/or exercise is known to reduce the level of insulin-like growth factor-1 (IGF-1), which provides a useful measure of daily integrated growth hormone secretion.41,42
Low IGF-1 levels have been reported in young female gymnasts when compared with controls and swimmers40,43,44 and were associated with negative energy balances and reduced growth rates in pre- and peripubertal female gymnasts.13 In contrast, no differences were detected in IGF-1 levels between prepubertal and early-pubertal male gymnasts and normoactive controls studied for 18 months.17 Because energy or protein restriction can lead to reduced IGF-1 levels, it is likely these differences between male and female gymnasts are due to the poor dietary practices of female gymnasts.
Because rising sex steroid levels during puberty augment the secretion of growth hormone and IGF-1,45 it is possible that low sex steroid concentrations related to intense training, inappropriate nutrition, or chronic stress may affect growth by altering the secretion of growth hormone or IGF-1. However, it is difficult to assess the influence of exercise on many growth and reproductive hormones because of their pulsatile and circadian pattern of release. Nevertheless, some authors have proposed that exercise-related growth and maturation alterations may disrupt the hypothalamic-pituitary-gonadal axis.14,27
Intense training may affect the signals driving the gonadotropin-releasing hormone pulse generator that stimulates the release of luteinizing hormone (LH) and follicle-stimulating hormone (FSH).46 Low levels of estrogen, testosterone, LH, FSH, and progesterone have been reported in female gymnasts, runners, and ballet dancers, presumably in response to training, inappropriate nutrition, and/or delayed maturation.24,25,47,48
In young male athletes, the hypothalamic-pituitary-gonadal axis seems to be less sensitive to chronic exercise.40 Limited information exists regarding hormonal alterations in response to intense training in boys. However, there are reports of reduced levels of growth hormone, sex hormone-binding globulin, or testosterone in male gymnasts and wrestlers.49,50 Furthermore, it is unclear whether reduced growth and delayed maturation in young male and female athletes represent a hormonal adaptation to training, or whether they are mediated via stress in combination with poor dietary practices and are casually related to reduced growth and delayed maturation.27
Physeal injuries. Linear growth deficits may also arise from stress injuries of the lower-extremity physes in young athletes.51,52 Although incidence data are lacking, there are several case reports and case series53-57 of young athletes, representing a variety of sport activities, with stress-injured lower-extremity physes. While these lower-extremity injuries all healed with rest and without complication to growth, there is some recent evidence of premature, partial, or complete closure of the stress-injured distal radial growth plate in skeletally immature female gymnasts58-62. It seems unlikely, however, that lower-extremity physeal injury would result in a reduction in adult stature, although leg-length discrepancy resulting from unilateral injury might be an outcome.
Psychological stress. Additional factors that may interact with intense training and negative energy balance to alter growth potential include the psychological and emotional stress associated with maintaining body weight, frequent competitions, year-long training, altered social relationships with peers, and demanding parents or coaches.63
Jahreis et al44 suggested that intense exercise may increase stress, thereby increasing cortisol levels to a level that could suppress growth hormone and IGF-1. In this study, cortisol levels did not change after 3 days of intensive gymnastics training but were very high at the outset, indicating long-term stress. Daly et al64 reported that the anabolic to catabolic balance represented by the IGF-1-to-cortisol ratio was reduced in male gymnasts during periods of intense training relative to normoactive controls. Although growth was not affected, these results reflect a catabolic state, perhaps resulting from overtraining, inadequate recovery, or inadequate caloric intake.
Does a Training Threshold Exist?
If a training threshold exists, the growth potential of a young athlete may be compromised when he or she exceeds the threshold. Theintz et al14 reported that training more than 15 to 18 hours per week before and during puberty may represent chronic stress capable of attenuating growth and final adult height. However, little scientific evidence supports a training threshold or recommendations of a safe and effective amount of exercise for young athletes.
In most studies of elite athletes, training intensity (ie, elements per minute, biomechanical loads, or skill difficulty) is typically not described; rather, training is routinely reported in terms of hours per week. Daly et al65 reported that young male gymnasts training between 10 and 29 hours per week sustained, on average, 102 impacts per session on the upper extremities and 217 on the lower extremities. The magnitude of such impacts peaked at 3.6 and 10.4 times body weight respectively. Video and heart rate data also illustrated that gymnastics training was characterized by intermittent bursts of activity, interspersed with equivalent or longer periods of rest and recovery time that accounted for about 63% of the total training time.
If a training threshold does exist, it must be specific to the physiologic, musculoskeletal, nutritional, and psychological demands of different sporting activities. It would be inappropriate to make generalizations from one sport to another, from boys to girls, or even from child to child. Training 15 hours per week may be excessive for one child but not another; therefore, the establishment of a training threshold is difficult considering the varying individual responses to the stress of elite sporting competition. It is possible that some children training below the threshold may still be at risk.
Compelling data indicate that some young athletes experience attenuated growth during training, followed by catch-up growth during periods of reduced training or retirement. Although a cause-and-effect relationship between training and growth has not been demonstrated, it is possible that growth is retarded by inadequate nutrition for a given level of activity, particularly during the periadolescent period. Guidelines for healthcare professionals caring for young athletes involved in intense training regimens include:
Training level. Understand and record the current details of the athletic activity (eg, type, intensity, frequency of participation, hours per session) at each office visit and during routine preparticipation exams.
Charting. Plot the athlete's growth quarterly, preferably beginning before puberty, on a standard chart (eg, National Center for Health Statistics) that displays percentile curves. Any observed tendency for an athlete to miss his or her predicted curve should trigger concern and prompt evaluation for potentially correctable factors.
Coaches can record the data, thereby helping to confirm their place on the healthcare team. Coaches must be alert to the potential negative consequences of appearing to favor particular growth patterns over others, and to the importance of avoiding injudicious comments about a particular athlete's body composition, either approving or disapproving. Data should be collected objectively to avoid comparing one athlete to another.
By noting the onset of rapid growth periods, coaches may be able to institute reduced training loads and limit skill progressions during times when the athlete may be especially vulnerable to injury. Adolescent girls should also be encouraged to keep accurate records of menstruation. Any cessation of menses for more than 2 or 3 months should trigger concern, prompting evaluation for correctable causes at an early stage.
Unique attributes. Remember that even though a particular athlete may remain within "average" parameters for sex and age, the actual situation may be far from normal for that individual and must be evaluated individually. Athletes perceived as having an eating disorder, or showing any tendency toward such a disorder, should be referred promptly for intervention.
Coach the coaches. Maintain good communication to help the coach understand sound nutrition and recognize signs of possible disordered eating and overtraining syndromes. Coaches should be made aware of the potentially harmful significance athletes may place on seemingly trivial comments that coaches might make.
Teamwork. Foster excellent relationships with the athlete's parents and coaches as well as counselors, nutritionists, and other community resources. This helps identify and treat problems. All referrals should be channeled through the physician to facilitate communication between all involved parties.
The prevailing opinion has been that intense training has no apparent effect on growth and maturation, perhaps because most studies report only average changes. However, there is evidence of reduced growth in some athletes when individual growth and maturation patterns are investigated. Any sign of reduced growth or delayed maturation should be investigated. Paying attention to the many interrelated factors that influence final adult height will help young elite athletes reach their full potentials.
Dr Daly is a National Health and Medical Research Council public research fellow and Dr Bass is a senior lecturer at the School of Health Sciences at Deakin University in Melbourne, Australia. Dr Caine is a professor in the department of physical education, health, and recreation, and Dr Howe is a team physician in the department of athletics at Western Washington University in Bellingham, Washington. Address correspondence to Robin M. Daly, PhD, School of Health Sciences, 221 Burwood Hwy, Deakin University, Victoria, Australia 3125; e-mail to [email protected].
Disclosure information: Drs Daly, Bass, Caine, and Howe disclose no significant relationship with any manufacturer of any commercial product mentioned in this article. No drug is mentioned in this article for an unlabeled use.