SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods and Procedures
  5. Results
  6. Discussion
  7. Acknowledgments
  8. Disclosure
  9. REFERENCES

Few large studies have evaluated the emergence of sexual dimorphism in fat distribution with appropriate adjustment for total body composition. The objective of this study was to determine the timing and magnitude of sex differences in regional adiposity from early childhood to young adulthood. Regional fat distribution was measured using dual-energy X-ray absorptiometry (trunk and extremity fat using automatic default regions and waist and hip fat using manual analysis) in 1,009 predominantly white participants aged 5–29 years. Subjects were divided into pre (Tanner stage 1), early (Tanner stages 2–3), late (Tanner stages 4–5), and post (males ≥20 years and females ≥18 years) pubertal groups. Sexual dimorphism in trunk fat (adjusted for extremity fat) was not apparent until late puberty, when females exhibited 17% less (P < 0.001) trunk fat than males. By contrast, sex differences in waist fat (adjusted for hip fat) were apparent at each stage of puberty, the effect being magnified with age, with prepubertal girls having 5% less (P = 0.027) and adult women having 48% less (P < 0.0001) waist fat than males. Girls had considerably more peripheral fat whether measured as extremity or hip fat at each stage. Sex differences in regional adiposity were significantly greater in young adults than in late adolescence. Exclusion of overweight participants did not materially affect the estimates. Sexual dimorphism in fat patterning is apparent even prepubertally with girls having less waist and more hip fat than boys. The magnitude of the sex difference is amplified with maturation, and particularly from late puberty to early adulthood.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods and Procedures
  5. Results
  6. Discussion
  7. Acknowledgments
  8. Disclosure
  9. REFERENCES

Marked sexual dimorphism in total and regional body composition is apparent in adults, with men having greater lean tissue mass and a more central fat pattern compared with the more peripheral fat distribution typically observed in adult women (1). It is now well accepted that significant gender differences in total body composition are apparent even in prepubertal children (2,3,4). However, the emergence of sex differences in fat patterning appears to be more controversial. Although anthropometric techniques suggested that sex-specific fat patterning emerges during puberty (5,6), recent work utilizing direct measures of body fat distribution including dual-energy X-ray absorptiometry (DXA) and magnetic resonance imaging has suggested that variation in fat patterning may also exist prepubertally (3,4,7,8,9,10). However, the literature is not consistent, with several groups reporting no gender differences in regional adiposity in young children (11,12,13). Much of this discrepancy can probably be attributed to variations in the regional indexes of choice, the particular ages being studied, differences in relative fatness of the groups, and use of different adjustments for covariates (14).

Few previous groups have studied large samples of participants over the full span of growth from prepuberty to early adulthood to examine the timing of the appearance of sexual dimorphism in fat distribution (8,11). Elucidating such information is important because of the role of body composition during growth in predicting adult disease risk (15). The aim of this study was to examine when differences in fat distribution measured using anthropometry and DXA become apparent in a large cross-sectional cohort of children and young adults aged 5–29 years.

Methods and Procedures

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods and Procedures
  5. Results
  6. Discussion
  7. Acknowledgments
  8. Disclosure
  9. REFERENCES

Subjects

Cross-sectional data on total and regional body composition were available for 1,009 predominantly white children and young adults aged 5–29 years who had participated in various studies investigating body composition and health in our laboratory from 1996 to 2007 (16,17,18,19,20,21,22). All studies were approved by the Lower South Ethics Committee and written informed consent was obtained from each participant or parent/guardian. A brief medical history was obtained by questionnaire and no participant was taking medication that would affect his or her body composition and no subjects had a history of constitutional delay in growth or maturation. Tanner stage of pubertal development was assessed in all subjects <19 years of age by the parent/guardian of the younger children or by the adolescent themselves using standard descriptions and pictures (23). Participants were divided into pre (Tanner stage 1), early (Tanner stages 2 and 3), late (Tanner stages 4 and 5) (8), and post (males aged ≥20 years and females ≥18 years) pubertal groups for analysis.

Anthropometric measures

Duplicate measures of height (wall-mounted stadiometer) and weight (electronic scales) were obtained with the participants wearing light clothing and no shoes. Duplicate measures of waist (minimum circumference between the rib cage and the iliac crest using a nonstretchable tape measure) and hip (maximum protuberance of the buttocks) circumferences were obtained in 908 (90%) and 846 (84%) of participants, respectively. Measurements were obtained by several trained examiners following the same protocols. In our laboratory, coefficients of variation (CVs%) for height, weight, and waist circumference in young children are <2% (17). BMI was calculated as weight in kilograms/height in meters squared. Overweight and obesity were defined using international reference norms (24) for those aged 5–18 years and BMI 25–29.9 and BMI ≥30 for overweight and obesity, respectively, for those ≥19 years of age.

DXA scanning

All DXA measurements were performed and analyzed by one experienced operator with a Lunar DPX-L scanner (software package 4.7; Lunar, Madison, WI) using standard procedures. The scanner determines total fat mass (kg) and the fat content (kg) of specific anatomical regions including trunk and extremity fat (automatic default regions) and central and peripheral fat (manual regions of interest) as shown in Figure 1. The trunk region consists of the area bordered by a horizontal line below the chin, vertical borders lateral to the ribs, and oblique lines passing through the femoral necks. The arm region consists of all tissues outside these lateral borders and the leg region of all tissue below the oblique lines. Extremity fat was defined as the sum of arm and leg fat (7). Central and peripheral regions of interest were also determined using manual analysis. The central region of interest (waist fat) was a box 9.6 cm high with the lower border positioned superior to the iliac crest. The peripheral region (hip fat) consists of the same sized box, positioned so that the center of the box was at the level of the greater trochanters (25). In our laboratory, the CVs for repeated in vivo scans on 10 adults were 2.6% for fat mass, 2.5% for fat percentage, and <3.5% for all regional measurements.

image

Figure 1. Automatic (a: trunk, arm, and leg) and manual (b: waist and hip) regions of interest.

Download figure to PowerPoint

Statistics

All data were analyzed using STATA (Stata Statistical Software, Release 8.0, 2003; StataCorp, College Station, TX). Analysis of covariance, which adjusted for age and included a sex × group interaction to test for a different pattern of results across the four different stages of puberty for males and females, was used to analyze the data. Logistic regression was used to analyze the data for overweight status. Regression models were used to examine the sex differences, presented as ratios of female relative to male values, for trunk fat adjusted for extremity fat, and for extremity fat adjusted for trunk fat, for the four pubertal stages. Sex differences for waist fat and hip fat, and waist girth and hip girth were also estimated. All variables not normally distributed (weight, BMI, fat masses) were log transformed before analysis.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods and Procedures
  5. Results
  6. Discussion
  7. Acknowledgments
  8. Disclosure
  9. REFERENCES

Table 1 shows means (s.d.) in total body composition for each pubertal group by sex. Sex differences were observed in all indexes with the exception of the prevalence of overweight/obesity. Furthermore, the interaction between sex and stage indicated that the pattern of differences between males and females was generally higher in each successive stage of puberty. In general, higher values are apparent across puberty for most indexes and for both sexes, with the exception of body fat in males. Although absolute fat mass is approximately twofold higher in adolescent and older males compared with prepubertal males, relative fatness (fat mass index, % fat) is highest during early puberty. By contrast, females display steady gains in absolute and relative fatness with maturation. Interestingly, the significantly higher body weight of young adults compared with adolescents in late puberty, was not matched by higher stature, and was explained by a higher deposition of lean than fat mass.

Table 1.  Total body composition by sex and puberty
inline image

Significant differences in regional fat distribution are also apparent with both sex and pubertal development (Table 2). The greater total body fat of females is also reflected by greater deposition of absolute regional fat at each site with most stages of maturation. In males, higher values are observed for trunk fat mass with age whereas extremity fat mass remains similar from early to postpuberty. Significant interactions between pubertal stage and sex were also observed for most measures showing that the pattern of differences between males and females depended on the level of maturation. By contrast, the nonsignificant interaction effect for waist circumference shows that the sex difference in waist circumference did not differ according to stage of puberty.

Table 2.  Regional body composition by sex and puberty
inline image

Table 3 demonstrates regional fat distribution at each stage of pubertal maturation as ratios (female values relative to males). Interestingly, sexual dimorphism in trunk fat deposition was not apparent until late puberty, and became more marked in young adults, where women had 34% less trunk fat than men, once adjusted for extremity fat. By contrast, differences in waist fat were apparent at each stage of puberty, although this effect magnified with increasing age, ranging from a difference of 5% (P = 0.027) in prepubertal children to 48% (P < 0.001) in young adults. Considerable differences in peripheral fat distribution were also apparent even in the youngest children. Thus prepubertal girls had 2% (P = 0.283) more extremity fat and 6% (P < 0.001) more hip fat than prepubertal boys and these differences became more marked at each successive stage of puberty. Although substantial sex differences in regional adiposity were apparent by late puberty, Table 3 demonstrates that this dimorphism becomes considerably more marked in young adults. For example, adjusted trunk fat was 17% (P < 0.001) and 34% (P < 0.001) lower in late and postpubertal females, respectively, relative to males, with corresponding figures of 35% (P < 0.001) and 48% (P < 0.001) in these groups for adjusted waist fat. Similar patterns were observed for peripheral fat distribution. Waist circumference was significantly lower and hip circumference significantly higher in girls from early puberty onward.

Table 3.  Sex differences (as ratios) in various measures of regional fat distribution across puberty (females relative to males)
inline image

Exclusion of overweight participants did not materially affect these estimates. The adjusted ratios at each pubertal stage were similar in the total sample (N = 1,009) and when analyses were restricted to the normal-weight participants (N = 846) for trunk or extremity fat (Figure 2) and waist or hip fat (data not shown).

image

Figure 2. Ratios (females relative to males) for adjusted trunk fat (dark lines) and adjusted extremity fat (light lines) for the total group (solid lines) and lean subjects only (dashed lines).

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods and Procedures
  5. Results
  6. Discussion
  7. Acknowledgments
  8. Disclosure
  9. REFERENCES

Our data demonstrate that sexual dimorphism in regional fat distribution becomes apparent at different stages of puberty, depending on the index chosen. Significant sex differences in trunk fat were not evident until late puberty, whereas even in prepubertal children, girls had significantly less waist fat than boys. Differences in peripheral fat were also evident at younger ages, whether measured using extremity fat (sum of leg and arm fat) or hip fat (fat directly over the greater trochanters). Sex differences in waist and hip circumferences were also observed from early puberty. The magnitude of the sex differences in regional adiposity increased significantly with maturation, so that striking differences were evident between adolescents in late puberty and young adults in their 20s.

The precise timing at which sex differences in regional adiposity is said to become apparent during growth is conflicting in the literature, due in part to the variety of methods used to assess centrality and to choice of statistical adjustments for other body components. In general, studies utilizing skinfold measurements have demonstrated that although absolute thicknesses are often higher in females at least from 8 years of age onward (6,26,27), demarcation in relative skinfold thickness occurs during puberty, primarily due to increased deposition of peripheral fat in females (5,6). Sexual dimorphism in circumference measures is also apparent from a young age with elevated waist circumference and/or waist-to-hip ratio or lower hip circumferences being reported in boys as young as 5–7 years (4,26,28,29). Our data support this earlier work, demonstrating that females have lower waist and greater hip circumferences from early puberty. However, there are reservations regarding the use of anthropometric measures to provide indexes of regional adiposity, because of variation in body proportions with growth (15). The advent of DXA, magnetic resonance imaging, and computed tomography provide more accurate and sensitive evaluation techniques to assess changes in body composition, even in young children (15).

Central fat distribution assessed by DXA has typically been reported using trunk fat as a percentage of total fat (% trunk fat) or the trunk-to-leg fat ratio. Although two groups found that differences in trunk-to-leg fat ratio were not apparent until after puberty (11,30), others (4) reported higher trunk-to-leg fat ratio values in 7–8-year-old girls compared with boys (4). Conversely, Wang et al. (10) in a large sample of Chinese children, reported that 6–11-year-old boys have a higher % trunk fat than girls. Regardless of these discrepant findings, alternative methods of analyzing DXA data are warranted given the problems of using ratios in statistical analyses (31) and the difficulties in interpreting measures of adiposity which include the measure of interest (e.g., trunk fat) in both the numerator and the denominator (% trunk fat) (32). Comparing regional fat masses without making adjustments for body size is clearly inappropriate (15); girls will have more body fat in every region as they develop, given their considerably higher total body fat levels and the strong correlations which are present between regional fat and total fat. Thus as recommended recently (15), we adjusted specific central regions for their corresponding peripheral region, as well as adjusting for height, but not for weight. When we included weight as a covariate as other groups have done (7,8), the variance inflation factors indicated that multicollinearity between variables was an issue.

Our results support and extend earlier work (7,8) demonstrating that sexual dimorphism is apparent not only in peripheral fat but also in central fat during puberty. Although expected differences in trunk fat were not apparent until late puberty, we observed that even young prepubertal girls stored less fat in the waist region and more fat peripherally (over the hips) than boys. Use of the DXA manual “waist” region of interest, which focuses narrowly on fat located in the mid-abdominal regional of the body, may provide a more appropriate measure of “central” fat than is provided by the larger default trunk fat region (33), given that DXA cannot differentiate between visceral and subcutaneous fat. However, others have demonstrated that trunk fat measured by DXA is strongly correlated with visceral fat in young children (34). Furthermore, central fat (DXA) is similarly associated with adverse plasma lipid profiles in children as is visceral adipose tissue (35,36). DXA is also a more available and less invasive method of assessing body composition in children, compared with computed tomography or magnetic resonance imaging, which are more costly, take longer to complete and have a larger radiation dose (computed tomography only). Visceral fat accumulation is thought to be similar in prepubertal boys and girls (3,12), particularly after appropriate adjustment for subcutaneous fat (13), although not all studies are consistent (9,37). On the other hand, in prepubertal girls of this age, more of this central fat may be subcutaneous (3).

Interestingly, we observed greater sexual dimorphism in regional adiposity in young adults aged 18–29 years when compared with adolescents in late puberty. Much of this response was probably driven by the differences we observed between adolescent and young adult males; the latter had 2 kg more total fat and significantly more trunk fat, but similar levels of extremity fat relative to adolescent males. Similar results have been reported recently in a longitudinal study of 17-year-old males followed for 8 years; here both active and inactive men became fatter and stored more of that fat abdominally by study end (38).

Our study is cross-sectional and therefore cannot determine actual changes in body composition within individuals. Pubertal status was self-reported rather than examined, and others (39) have reported only moderate concordance between self- and physician-based assessment. However, Duke et al. (23) show that 86–91% of adolescents correctly estimate their developmental age. We also do not have measures of blood hormone status, such as insulin-like growth factor I, sex steroids, and leptin which might explain variation in body composition over time and between the sexes (40,41). Others (4) have shown that variation in insulin-like growth factor I and sex steroids explain <18% of the variance in body composition in prepubertal children. However, this may underestimate the contribution of sex hormones, given that blood sampling in Garnett et al. (4) occurred in the early morning, and sex steroids peak during the night. Although recent research has provided greater understanding of the role of sex hormones on fat partitioning, considerable work is also required to determine the role of chromosomal sex (XX, XY) and prenatal hormones in determining fat patterning later in life (42).

Strengths of our study include measurements on a large database of children and young adults varying markedly in shape and size. Measurements were made on the same DXA scanner, and analysis of all DXA data was completed by the same person. In addition, we excluded participants with any known endocrine disorders or with a history of constitutional delay in growth or maturation from our study.

In conclusion, our study shows significant sexual dimorphism in fat patterning is apparent even prepubertally, with girls having less waist and more hip fat than boys. The magnitude of the sex difference is amplified with maturation, with young adults displaying considerably higher central fat deposition compared with those in late adolescence. Progression to young adulthood may be a time of considerable fat deposition, particularly fat that is more centrally distributed, at least in men (38). Although the predominant predictor of fat gain in men at this time was age, appropriate participation in physical activity may modulate the overall effect (38). Thus late adolescence/young adulthood could be an appropriate time for targeted lifestyle interventions to limit fat deposition.

Acknowledgments

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods and Procedures
  5. Results
  6. Discussion
  7. Acknowledgments
  8. Disclosure
  9. REFERENCES

This research was supported by funding from Health Research Council, the University of Otago, the Child Health Research Foundation, Otago Medical Research Foundation, and the Dunedin School of Medicine Dean's Bequest Fund.

REFERENCES

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods and Procedures
  5. Results
  6. Discussion
  7. Acknowledgments
  8. Disclosure
  9. REFERENCES