help button home button Endocrine Society JCEM JCEM Call for Nominations for EIC
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Journal of Clinical Endocrinology & Metabolism, doi:10.1210/jc.2004-2041
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
90/5/2580    most recent
Author Manuscript (PDF)
Right arrow Submit a related Letter to the Editor
Right arrow Purchase Article
Right arrow View Shopping Cart
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Copyright Permission
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Misra, M.
Right arrow Articles by Klibanski, A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Misra, M.
Right arrow Articles by Klibanski, A.
Right arrowPubmed/NCBI databases
*Compound via MeSH
*Substance via MeSH
Medline Plus Health Information
*Eating Disorders
Hazardous Substances DB
*ESTRADIOL
*HYDROCORTISONE
Related Collections
Right arrow Female Endocrinology
Right arrow Metabolism
Right arrow Neuroendocrinology and Pituitary
Right arrow Pediatric Endocrinology
The Journal of Clinical Endocrinology & Metabolism Vol. 90, No. 5 2580-2587
Copyright © 2005 by The Endocrine Society

Hormonal Determinants of Regional Body Composition in Adolescent Girls with Anorexia Nervosa and Controls

Madhusmita Misra, Karen K. Miller, Cecilia Almazan, Megan Worley, David B. Herzog and Anne Klibanski

Neuroendocrine Unit (M.M., K.K.M., C.A., M.W., A.K.) and Eating Disorders Unit (D.B.H.), Massachusetts General Hospital and Harvard Medical School, and Pediatric Endocrine Unit (M.M.), Massachusetts General Hospital for Children and Harvard Medical School, Boston, Massachusetts 02114

Address all correspondence and requests for reprints to: Anne Klibanski, M.D., Chief, Neuroendocrine Unit, Massachusetts General Hospital, BUL 457, 55 Fruit Street, Boston, Massachusetts 02114. E-mail: aklibanski{at}partners.org.


   Abstract
Top
 Abstract
Introduction
 Subjects and Methods
 Results
 Discussion
 References
 
We have previously demonstrated that girls with anorexia nervosa (AN) have higher levels of GH and cortisol and lower levels of estradiol than healthy adolescents. The effects of endocrine alterations on regional body composition in AN, however, have not been reported. We, therefore, enrolled 23 adolescent girls with AN and 20 healthy girls of comparable maturity in a study examining regional body composition. Levels of estradiol and IGF-I, as well as measures of GH and cortisol concentration (using cluster analysis of data obtained from frequent sampling q30' for 12 h overnight) were examined to determine hormonal determinants of regional body composition in adolescent girls with AN and controls. Girls with AN were followed for 1 yr and examined again at weight recovery (10% increase in body mass index) (n = 11).

Percent trunk fat and trunk to extremity fat ratio (T/E fat) were significantly reduced in girls with AN compared with healthy adolescents (P = 0.001 and 0.01, respectively). Percent trunk lean mass and trunk to extremity lean mass ratio (T/E lean) were higher in AN than in controls (P = 0.01 and 0.009); percent extremity lean mass was lower in AN (P = 0.009). In healthy controls, total area under the curve (AUC) for GH correlated inversely with percent trunk fat and T/E fat (r = –0.66, P = 0.002 and r = –0.62, P = 0.003). Similar correlations were observed between other measures of GH concentration (mean and nadir) and percent trunk fat and T/E fat. No relationship was observed between GH concentration and regional lean mass or between cortisol concentration and regional body composition. In contrast, GH concentration did not predict regional body composition in adolescents with AN on regression analysis. However, nadir cortisol concentration correlated inversely with percent extremity lean mass (r = –0.49; P = 0.02) and positively with percent trunk lean mass and T/E lean (r = 0.48, P = 0.03; and r = 0.49, P = 0.02) in girls with AN. A similar trend was observed between other measures of cortisol concentration (mean cortisol and AUC) and percent trunk lean mass and T/E lean in AN. Trunk fat was lowest in girls with AN who had high GH but low cortisol levels (based on median values), whereas some preservation of trunk fat was observed in girls with low GH and high cortisol levels. Weight recovery occurred in seven of 11 girls with low GH and high cortisol values; however, only two of the nine girls with high GH and low cortisol recovered weight. High GH with lower cortisol levels may thus be a marker of greater severity of AN.

Our results suggest that in healthy controls, GH concentration predicts regional body composition and favors a redistribution of body fat such that T/E fat ratio decreases. In AN, however, high levels of GH and cortisol have contrasting associations with fat mass. High cortisol levels in AN predict a redistribution of lean body mass such that extremity lean mass decreases. Further studies are necessary to better understand the implications of these data.


   Introduction
Top
 Abstract
 Introduction
Subjects and Methods
 Results
 Discussion
 References
 
WE HAVE PREVIOUSLY demonstrated marked alterations in levels of GH, cortisol, and estradiol (E2) (1, 2), hormones that have effects on body composition, in adolescent girls with anorexia nervosa (AN) compared with healthy adolescents of comparable maturity. However, the effects of alterations in these hormones on regional body composition have not been examined in AN. Because marked changes also occur in regional body composition in AN (3), this disorder is an excellent model in which to study the effects of the aforementioned hormones on regional body composition. In our studies, girls with AN had significantly higher GH concentrations than healthy adolescents, despite lower IGF-I levels, suggestive of a nutritionally acquired resistance to GH (1). Of note, high levels of GH in girls with AN did not correlate with markers of bone turnover, although in healthy adolescents, strong positive correlations were observed between GH concentration and bone turnover markers, suggestive of resistance to GH effects at the level of bone. It is not known whether a similar resistance to GH exists at the level of fat or muscle. We have also observed high serum cortisol concentrations and low levels of E2 in adolescents with AN (1, 2); however, associations of high cortisol and low E2 levels with regional fat mass or lean mass have not been reported.

GH is lipolytic (4), and children and adults with GH deficiency have higher fat mass and lower lean mass than healthy controls, with a truncal fat distribution (5, 6, 7, 8, 9). Obese individuals with abdominal fat accumulation also have low GH levels (10). Administration of recombinant human GH (rhGH) reverses these effects both in conditions of GH deficiency (5, 6, 7, 8, 9, 11) and abdominal obesity (12, 13) and increases lean mass. AN represents the opposite end of the spectrum to obesity, and high levels of GH occur in this disorder (1, 14, 15, 16, 17, 18).

Cortisol excess has antilipolytic effects (4, 19) causing a redistribution of fat mass and favoring a truncal accumulation of fat (20, 21), which resolves with normalization of cortisol levels (22). A loss of muscle mass with hypercortisolemia has also been reported (23, 24). E2 causes a gynecoid distribution of body fat (25, 26). Estrogen deficiency, as in menopause, leads to an android redistribution of body fat (27), with increased trunk fat at the expense of extremity fat, and estrogen administration decreases android fat in postmenopausal women (28).

In this paper, we have examined regional body composition in adolescent girls with AN and in healthy adolescents of comparable maturity in relation to levels of GH, cortisol, and E2 to determine whether changes in regional body composition are commensurate with the expected effects of high GH, high cortisol, and low E2 levels on body composition.


   Subjects and Methods
Top
 Abstract
 Introduction
 Subjects and Methods
Results
 Discussion
 References
 
Subject selection

We enrolled 23 adolescent girls with AN (meeting the Diagnostic and Statistical Manual of Mental Disorders IV criteria) and 20 healthy adolescents for this study. Girls with AN did not differ from controls in chronological age (16.3 ± 1.6 vs. 15.5 ± 1.8 yr; P not significant) or bone age (15.8 ± 1.5 vs. 15.8 ± 2.0 yr; P not significant). Methods of recruitment for our subjects, clinical characteristics, hormonal data, but not regional body composition data, have been reported earlier (1). Five girls with AN and four healthy adolescents had not attained menarche at the time of the study. Other girls with AN had been amenorrheic for at least 3 months at study initiation. Healthy controls did not have a past or present history of eating disorders. The Institutional Review Board of Partners HealthCare approved the study, and informed consent and assent were obtained from all subjects and their parents.

Experimental protocol

All subjects were screened with a history, physical examination, and a blood sample to rule out thyroid dysfunction, hypergonadotropic hypogonadism, and hyperprolactinemia. Eligible subjects had frequent blood sampling every 30 min overnight (2000 h on the night of admission to 0800 h the next morning) for GH and cortisol at the General Clinical Research Center of Massachusetts General Hospital (Boston, MA). Data obtained from frequent sampling were analyzed by the computerized algorithm Cluster (1X2) using described methods (29). Total area under the curve (AUC), nadir, and mean GH and cortisol concentrations are reported in this paper. Details of data from Cluster analysis of GH and cortisol concentration have been previously reported (1, 30). Fasting serum was obtained for IGF-I and E2. All subjects had body composition assessed by dual energy x-ray absorptiometry (DXA). Subjects were followed over 1 yr, and frequent sampling for GH and cortisol was repeated at weight recovery [defined as a 10% increase in body mass index (BMI)] (2). Eleven subjects recovered weight over the study period, and frequent sampling data were available for 10 of these weight-recovered adolescents.

Anthropometric measurements

Height for all subjects was measured using a single stadiometer at the General Clinical Research Center. Measurements were obtained in triplicate and averaged. Weight was measured on an electronic scale in a hospital gown. BMI was calculated as the ratio of weight (in kilograms) to height (in meters)2. Standards of Greulich and Pyle (31) were used to determine bone age of our subjects by a single investigator, a pediatric endocrinologist.

Biochemical assessment

Serum GH concentrations and IGF-I were determined by immunoradiometric assays (Nichols Institute Diagnostics, San Juan Capistrano, CA), detection limit of 0.05 ng/ml and an intraassay coefficient of variation of 2.4–9.4% for GH, and detection limit of 30 µg/liter and coefficient of variation 3.1–4.6% for IGF-I. We measured serum cortisol with an RIA (Diagnostic Products Corp., Los Angeles, CA; limit of detection 1 µg/dl, sensitivity 0.21 µg/dl, coefficient of variation 2.5–4.1%). Although this assay measures steroid metabolites other than cortisol, it remains the standard assay for this steroid hormone. Urine free cortisol was measured by the hospital laboratory using the GammaCoat 125I RIA (Diasorin Inc., Stillwater, MN; detection limit 1 µg/dl; coefficient of variation 7%). E2 levels were assayed by ultrasensitive RIA (Diagnostic Systems Laboratories, Inc., Webster, TX; detection limit, 8.1 pmol/liter; coefficient of variation, 6.5–8.9%).

Body composition

Body composition, including validated measures of fat mass and lean body mass, was determined by whole-body DXA (QDR 4500, Hologic Inc., Waltham, MA) (32, 33). The precision error (SD) of DXA has been reported to be 425 g for whole-body fat and fat-free mass (32), with a correlation of 0.99 with a four-compartment model body composition method for measuring fat-free mass and 0.93–0.97 with multislice computed tomography for measuring regional fat-free mass (33). Whole-body DXA is recommended for determining body composition in children (34, 35), and the coefficients of variation for lean body mass and fat mass by DXA in the pediatric weight range have been reported as 1.0 and 4.1%, respectively (36, 37). Calculations for regional body composition have been previously described (3) and are summarized here: percent trunk fat, (trunk fat/total fat) x 100; percent extremity fat, (total extremity fat/total fat) x 100; trunk to extremity fat (T/E fat), percent trunk fat/percent extremity fat; percent trunk lean, (trunk lean mass/total lean mass) x 100; percent extremity lean, (total extremity lean mass/total lean mass) x 100; and trunk to extremity lean (T/E lean), percent trunk lean/percent extremity lean.

Statistical methods

All data are described as mean ± SD. The data were analyzed using the JMP program (version 4; SAS Institute, Inc., Cary, NC). Student’s t test was used to calculate differences between means. When data were not normally distributed, the Wilcoxon rank sum test was used. When comparisons involved more than two groups, ANOVA was used followed by the Tukey Kramer’s test for intergroup comparisons. We used paired analysis to compare endpoints at weight recovery vs. baseline. Correlational analyses were used to determine predictors of regional body composition.


   Results
Top
 Abstract
 Introduction
 Subjects and Methods
 Results
Discussion
 References
 
Clinical characteristics

Clinical characteristics of our subjects and hormonal data have been reported in a previous study (1) and are summarized here. BMI of girls with AN was significantly lower compared with controls (16.7 ± 1.2 vs. 21.9 ± 3.7 kg/m2, P < 0.0001). Total and percent fat mass were also significantly lower in girls with AN compared with healthy adolescents (fat mass, 8.8 ± 2.7 vs. 17.4 ± 5.7 kg, P < 0.0001; percent fat mass, 18.3 ± 3.9 vs. 29.4 ± 5.4%; P < 0.0001). Lean body mass did not differ between the groups (37.0 ± 3.9 kg in AN vs. 38.8 ± 6.6 kg in controls, P not significant). Girls with AN had significantly lower IGF-I levels (315 ± 141 vs. 528 ± 157 ng/ml, P < 0.0001) and significantly higher GH concentrations compared with controls on Cluster analysis (mean GH, 5.3 ± 2.7 vs. 3.4 ± 1.4 ng/ml, P = 0.006; nadir GH, 2.3 ± 1.4 vs. 0.9 ± 0.8 ng/ml, P = 0.000; and total AUC for GH, 3879 ± 2007 vs. 2476 ± 992 ng/ml, P = 0.007). Cortisol concentrations were also significantly higher in girls with AN than in controls on Cluster analysis (mean cortisol, 8.6 ± 2.0 vs. 6.0 ± 1.1 µg/dl, P < 0.0001; nadir cortisol, 5.5 ± 2.3 vs. 3.5 ± 1.2 µg/dl, P = 0.002; and AUC for cortisol, 6112 ± 1467 vs. 4150 ± 809 µg/dl, P < 0.0001), as were values of urine free cortisol normalized for creatinine and surface area (UFC/cr.SA) (0.035 ± 0.021 vs. 0.020 ± 0.008 µg/mg·m2, P = 0.03). E2 values were lower in AN (16.7 ± 6.6 vs. 21.9 ± 9.0 pg/ml, P = 0.03).

Regional body composition

Figure 1Go demonstrates differences in regional body composition between the two groups. Percent trunk fat was significantly lower in adolescents with AN compared with healthy controls (33.5 ± 3.9 vs. 38.6 ± 5.6%, P = 0.001), whereas percent extremity fat did not differ in the two groups (56.8 ± 4.0 vs. 55.9 ± 4.6%, P not significant) (Fig. 1AGo), suggesting a preferential loss of truncal fat with weight loss. The T/E fat ratio was lower in AN (0.60 ± 0.10 vs. 0.70 ± 0.16, P = 0.01). Thus, girls with AN had 13.2% lower percent trunk fat and a 14.3% reduction in T/E fat compared with controls.



View larger version (20K):
[in this window]
[in a new window]
  		FIG. 1. Regional fat mass and lean mass in adolescent girls with AN (black bars) and healthy controls (white bars). A, Lower percent trunk fat in girls with AN (P = 0.001), with a decreased T/E fat (P = 0.01). B, Higher trunk lean mass (P = 0.01), lower extremity lean mass (0.009), and a higher T/E lean (P = 0.009) in AN compared with controls. *, P ≤ 0.01; **, P ≤ 0.001.

 
Although total lean mass did not differ between the groups, girls with AN had significantly lower percent lean mass in the extremities (42.5 ± 1.7 vs. 44.1 ± 2.1%, P = 0.009) and significantly higher percent lean mass in the trunk (49.9 ± 1.9 vs. 48.1 ± 2.5%, P = 0.01) compared with healthy adolescents (Fig. 1BGo). The ratio of T/E lean was higher in girls with AN (1.18 ± 0.09 vs. 1.10 ± 0.10, P = 0.009). Percent trunk lean mass was 3.7% higher, percent extremity lean 3.6% lower, and T/E lean 7.3% higher in AN compared with healthy adolescents.

Predictors of regional fat distribution

Table 1Go describes results of correlational analysis comparing measures of GH concentration with regional fat mass in girls with AN and controls. Strong inverse correlations were observed between GH concentration measures and percent trunk fat and T/E fat in controls but not in AN. A positive correlation was observed between GH concentration and percent extremity fat in controls, but no such correlation could be demonstrated in AN. However, when girls with AN were divided into two groups based on whether serum GH levels from frequent sampling were above or below the median GH value (4.45 ng/ml) for this group, we observed lower T/E fat in girls with AN whose GH levels were above the median than in those whose GH values were below the median (percent trunk fat, 31.9 ± 2.7 vs. 34.9 ± 3.2%, P = 0.05; T/E fat, 0.55 ± 0.07 vs. 0.64 ± 0.09, P = 0.03). Figure 2Go demonstrates the correlation between total AUC for GH and percent trunk fat, percent extremity fat and T/E fat in girls with AN, and controls.


View this table:
[in this window]
[in a new window]
  		TABLE 1. Relationship between GH concentration and regional fat mass in girls with AN and healthy adolescents

 


View larger version (32K):
[in this window]
[in a new window]
  		FIG. 2. Correlations between total area under the curve for GH [AUC (GH)] and percent trunk fat (A), percent extremity (extr) fat (B), and the ratio of T/E (extr) fat (C) in healthy adolescents (gray circles) and girls with AN (black circles). In controls, AUC (GH) correlated inversely with percent trunk fat and T/E fat and positively with percent extremity fat. AUC (GH) did not correlate with percent trunk fat, percent extremity fat, or T/E fat in girls with AN.

 
Neither trunk fat nor extremity fat values correlated with IGF-I, cortisol levels (serum or urinary), or with E2 (data not shown). We then examined regional body composition in girls with AN based on median values for mean serum cortisol and UFC/cr.SA. Indeed, girls with AN whose cortisol levels were greater than the median value of serum cortisol in this group (8.7 µg/dl) had mean T/E fat that trended higher than girls with AN whose serum cortisol levels were less than 8.7 µg/dl (0.65 ± 0.11 vs. 0.56 ± 0.09; P = 0.06). Similarly, percent trunk fat trended higher in girls with AN whose serum cortisol levels were higher than 8.7 µg/dl vs. those with serum cortisol values less than 8.7 µg/dl (35.2 ± 4.2 vs. 32.5 ± 3.0%) (Wilcoxon rank sum test, P = 0.09). When girls with AN were separated into those with median UFC/cr.SA values above the median (0.029 µg/mg·m2) and below the median for this group, T/E fat was higher in AN girls with UFC/cr.SA values greater than 0.029 µg/mg·m2 vs. those with values less than 0.029 µg/mg·m2 but did not reach statistical significance (percent trunk fat, 34.9 ± 3.3 vs. 32.9 ± 4.3%; T/E fat, 0.638 ± 0.092 vs. 0.579 ± 0.113). These results were not observed in healthy adolescents.

We then examined regional fat distribution in girls with AN who had low GH and high cortisol levels (n = 11) vs. those with high GH and low cortisol levels (n = 9) vs. controls (using median values of GH and cortisol). Three girls with AN had either high GH and high cortisol or low GH and low cortisol, and these girls were not included in this analysis. A significant difference was noted between T/E fat in the three groups, such that AN girls with high GH and low cortisol had the lowest T/E fat (Table 2Go). AN girls with high cortisol and low GH had trunk fat and T/E fat that although lower did not differ significantly from controls, although BMI, fat mass, and percent fat mass were significantly lower than healthy adolescents, suggesting some preservation of trunk fat in this group, despite low weight and low overall fat mass.


View this table:
[in this window]
[in a new window]
  		TABLE 2. Regional body composition by GH and cortisol levels

 
In addition, seven of the 11 girls with high cortisol but low GH subsequently recovered weight (all within 6 months of the baseline visit), whereas only two of the nine girls with high GH but low cortisol recovered weight in the follow-up period of the study (one by 6 months and one at 12 months follow-up). Two other girls who recovered weight over the follow-up period had high GH and cortisol levels. Girls who recovered weight early in the course of the study and thus may have had less severe AN, i.e. the girls with high cortisol and low GH were also less severely affected with regards to regional fat distribution. High GH levels may, in fact, be a marker for increased severity of disease with poorer chances for early weight recovery.

Predictors of regional lean mass distribution

Table 3Go demonstrates the correlation between measures of cortisol concentration and regional lean mass distribution in girls with AN and controls. Cortisol concentration correlated directly with T/E lean in girls with AN but not in controls. An inverse correlation was observed between nadir cortisol concentration and percent extremity lean mass in AN but not in controls. Figure 3Go shows the correlations between nadir cortisol concentration and regional lean mass in girls with AN and controls. Neither percent trunk lean mass nor percent extremity lean mass correlated with GH concentration, IGF-I, or E2 (data not reported).


View this table:
[in this window]
[in a new window]
  		TABLE 3. Relationship between cortisol concentration and regional lean mass in girls with AN and healthy adolescents

 


View larger version (28K):
[in this window]
[in a new window]
  		FIG. 3. Correlations between nadir cortisol concentration and T/E lean (A), percent extremity (extr) lean mass (B), and the ratio of T/E (extr) lean mass (C) in healthy adolescents (gray circles) and girls with AN (black circles). In girls with AN, nadir cortisol concentration correlated directly with percent trunk lean mass and T/E lean mass, and inversely with percent extremity lean mass. Nadir cortisol did not correlate with percent trunk lean mass, percent extremity lean mass, or T/E lean mass in healthy adolescents.

 
When we compared girls with AN who had low GH and high cortisol levels vs. those with high GH and low cortisol levels vs. controls, girls with AN with low GH and high cortisol levels had the highest T/E lean and lowest percent extremity lean mass (Table 2Go).

Effects of weight recovery

Weight recovery in 11 girls with AN resulted in a significant increase in percent trunk fat (35.7 vs. 32.2%, P = 0.0003) and T/E fat (0.63 vs. 0.57, P = 0.008). Percent extremity lean mass increased to a lesser extent (43.9 vs. 42.7%, P = 0.04), with a trend toward a decrease in the T/E lean (1.11 vs. 1.17, P = 0.06). GH nadir decreased with weight recovery (1.41 vs. 2.3 ng/dl, P = 0.06). Mean and total AUC for GH, mean cortisol, AUC, and nadir cortisol decreased minimally but not significantly with weight recovery, suggesting that a longer period of recovery may be required before significant changes are observed in these hormonal parameters. GH normalization may precede normalization of cortisol values.

Baseline cortisol AUC correlated positively with percent change in percent fat mass over 12 months in all girls with AN (r = 0.63, P = 0.01), and particularly in girls with AN who recovered weight over the follow-up period (r = 0.79, P = 0.02), such that girls with higher cortisol values at baseline had greater increases in fat mass. In addition, higher baseline cortisol AUC predicted a greater increase in percent trunk fat in these girls with AN at weight recovery (r = 0.60, P = 0.08).


   Discussion
Top
 Abstract
 Introduction
 Subjects and Methods
 Results
 Discussion
References
 
We demonstrate that in healthy adolescents, although not in AN as a group, GH concentration predicts percent trunk and extremity fat, with higher GH concentrations being associated with lower trunk fat and greater extremity fat. In girls with AN, but not in healthy adolescents, high cortisol concentrations predict lower lean mass in the extremities and increased lean mass in the trunk. Lack of associations among GH, cortisol, and regional fat mass in AN may be a consequence of opposing effects of cortisol and GH on trunk fat. Girls with AN who have high cortisol and low GH values are more likely to recover weight early and have regional fat mass that does not differ from controls. Girls with AN who have high GH and low cortisol levels are less likely to recover weight over 1 yr of follow-up and have lower trunk fat than controls.

GH deficiency causes an increase in fat mass and a decrease in lean mass both in children and in adults, which reverses after treatment with rhGH (5, 6, 7, 8, 9, 38). These changes in body composition are a consequence of the lipolytic effects of GH (4) and a reduction of triglyceride accumulation via inhibition of lipoprotein lipase activity (39). In particular, there is a decrease in truncal fat mass with rhGH. Consistent with these effects, we observed an inverse correlation between GH concentration and trunk fat in our healthy controls, with higher levels of GH correlating with lower levels of percent trunk fat and T/E fat. In addition, GH concentrations in healthy adolescents correlated directly with percent extremity fat, suggesting that higher GH levels in adolescent girls favor a redistribution of body fat resulting in a reduction in trunk fat at the expense of extremity fat.

T/E fat was significantly lower in girls with AN compared with controls, and these adolescents also had higher GH concentrations than healthy adolescents. One may speculate that higher GH levels in AN are responsible for reduced trunk fat in this population. However, we could find no relationship between GH concentrations and percent trunk fat or T/E fat in girls with AN as a whole. One possible reason for these findings is that fat mass is decreased to such a significant extent as a consequence of undernutrition that a linear relationship can no longer be observed. When we divided girls with AN into two groups based on median GH values, girls with AN whose GH values were greater than the median value did indeed have lower T/E fat than girls whose GH values were below the median value. Also, in agreement with the fact that mature adipocytes express receptors for GH, but not for IGF-I, we found no relationship between IGF-I levels and fat distribution in either controls or in girls with AN. These data suggest that GH acts on adipocytes by direct effects and not via IGF-I.

Our subjects with AN had significantly higher cortisol concentrations than healthy adolescents of comparable maturity. High cortisol concentrations, both stress related and in Cushing syndrome, are associated with increased truncal fat (20, 21), and this effect on body composition is related to the antilipolytic action of cortisol (4). Treatment of Cushing syndrome results in a decrease in fat mass, especially truncal fat (40). Cushing syndrome is also associated with a decrease in muscle mass, and in particular extremity muscle mass, as demonstrated by a decrease in arm muscle area (23) and decreased muscle in legs and arms by DXA (24).

Our subjects with AN did not have increased fat mass or increased trunk fat, although levels of cortisol were higher than in controls. These findings are different from those in women with stress-related hypercortisolemia, who do have increased truncal fat (20). In contrast, both total and trunk fat were significantly reduced in girls with AN compared with controls. We postulate that the lack of effects of cortisol on fat distribution is related to very low levels of available substrate in AN, a condition associated with significant weight loss, whereas other conditions of stress-related hypercortisolemia are not associated with such weight loss. In addition, it is likely that expected effects of elevated cortisol levels in AN on fat distribution are offset by concomitant high GH levels in this disorder, which would cause a decrease in truncal fat. Indeed, when we divided girls with AN into two groups based on median cortisol values, girls with AN with values of cortisol above the median did have T/E fat that trended higher than girls with AN whose cortisol levels were below the median for the group. The opposing effects of high cortisol and high GH levels on fat distribution may also account for the lack of correlation between regional body composition measures and levels of these hormones in AN.

To better distinguish the opposing effects of high GH and cortisol in AN, we compared girls with AN who had high cortisol and low GH levels vs. those with high GH and low cortisol levels (based on median values for GH and cortisol). We were intrigued to note that for the most part, there was not an overlap between girls having high GH vs. high cortisol values. Girls with high cortisol and low GH levels appeared to be less severely affected with AN in that the majority (seven of 11) recovered weight within 6 months of the baseline visit, and had percent trunk fat and T/E fat that although lower did not significantly differ from controls. Conversely, girls with AN who had high GH and low cortisol levels at baseline appeared to be more severely affected in that only two of the nine girls in this category recovered weight subsequently and did so later in the course of follow-up. This group had the lowest T/E fat (significantly lower than controls). This dichotomy also explains why a correlation between cortisol or GH with regional fat measures was not noted in girls with AN.

Our data also suggest that higher cortisol levels in the presence of lower GH levels may be responsible for some preservation of body fat, particularly trunk fat, in AN, and with increasing severity of AN, increases in GH occur, causing a decrease in trunk fat.

Interestingly, in adult women with AN, weight recovery is associated with increased truncal adiposity related to cortisol levels, with greater increases in trunk fat in women with higher basal cortisol levels (41), suggesting that availability of substrate may indeed be necessary for the manifestation of antilipolytic effects of cortisol. In controls, cortisol concentrations did not predict fat distribution, likely because normocortisolemia, unlike high cortisol levels, does not have a significant effect on fat metabolism. Consistent with the data from adult studies, higher baseline cortisol predicted greater increases in fat mass and also trunk fat in weight recovered AN. These data support our supposition that lack of substrate could account for significant correlations between cortisol and regional fat mass in AN, and with increased availability of substrate, this relationship may become more evident. Of note, weight recovery did not result in marked decreases in cortisol levels, suggesting that hypercortisolemia may persist for variable periods of time even with weight recovery. Nadir GH, conversely, did decrease with weight recovery. Our data suggest that normalization of GH secretion precedes normalization of cortisol secretion during weight recovery in AN.

Lean body mass was preserved in girls with AN unlike fat mass, and inverse correlations were noted in this group between cortisol concentrations and percent extremity lean mass, consistent with the effects of high cortisol levels on muscle mass. In addition, girls with AN who had high cortisol and low GH levels had the lowest extremity lean mass and highest trunk lean mass and T/E lean ratio. Our data suggest a redistribution of lean body mass with high levels of cortisol, resulting in greater trunk than extremity lean mass compared with controls.

Conditions of hypoestrogenism, such as menopause, are associated with increased trunk fat at the expense of extremity fat (android fat distribution pattern), although total fat mass remains unchanged (27). Studies suggest an absence of estrogen receptors on adipocytes (42), and a decrease in SHBG levels with increased bioavailable testosterone has been postulated to be the mechanism responsible for these changes in regional body composition Indeed, hormone replacement therapy in postmenopausal women results in a reversal of these changes (28) and is associated with an increase in SHBG and a decrease in bioavailable testosterone (43). In our subjects, we could find no relationship between E2 levels and regional fat distribution. Girls with AN are hypogonadal and have low levels of estrogen (1, 2). However, levels of free testosterone are also low in this condition (2). This may explain why low levels of estrogen in girls with AN do not favor an android distribution of body fat.

Thus, we demonstrate that higher GH levels are associated with a lower T/E fat in controls but not in AN, whereas cortisol levels are associated with a redistribution of lean mass in AN but not in controls, favoring a reduction in lean mass in the extremities. Regional body composition is not predicted by estrogen or IGF-I levels. Although both GH and cortisol are elevated in AN, expected effects of these hormones on regional fat mass are not observed, possibly because of severe substrate deficiency in this disorder and because of the opposing effects of these hormones on regional fat distribution.


   Acknowledgments
 
We thank the skilled nursing staff of the General Clinical Research Center and Ellen Anderson and her Bionutrition staff for the care provided to our subjects at the GCRC. We also thank our study volunteers because without their participation, this study would not have been possible.


   Footnotes
 
This work was supported by National Institutes of Health Grants M01-RR-01066, DK 062249, and K23 RR018851.

First Published Online February 15, 2005

Abbreviations: AN, Anorexia nervosa; AUC, area under the curve; BMI, body mass index; DXA, dual energy x-ray absorptiometry; E2, estradiol; rhGH, recombinant human GH; T/E fat, trunk to extremity fat ratio; T/E lean, trunk to extremity lean mass ratio; UFC/cr.SA, urine free cortisol normalized for creatinine and surface area.

Received October 14, 2004.

Accepted January 28, 2005.


   References
Top
 Abstract
 Introduction
 Subjects and Methods
 Results
 Discussion
 References
 

  1. Misra M, Miller K, Bjornson J, Hackman A, Aggarwal A, Chung J, Ott M, Herzog D, Johnson M, Klibanski A 2003 Alterations in growth hormone secretory dynamics in adolescent girls with anorexia nervosa and effects on bone metabolism. J Clin Endocrinol Metab 88:5615–5623[Abstract/Free Full Text]
  2. Soyka L, Misra M, Frenchman A, Miller K, Grinspoon S, Schoenfeld D, Klibanski A 2002 Abnormal bone mineral accrual in adolescent girls with anorexia nervosa. J Clin Endocrinol Metab 87:4177–4185[Abstract/Free Full Text]
  3. Misra M, Soyka L, Miller K, Grinspoon S, Levitsky L, Klibanski A 2003 Regional body composition in adolescents with anorexia nervosa and changes with weight recovery. Am J Clin Nutr 77:1361–1367[Abstract/Free Full Text]
  4. Ottosson M, Lonnroth P, Bjorntorp P, Eden S 2000 Effects of cortisol and growth hormone on lipolysis in human adipose tissue. J Clin Endocrinol Metab 85:799–803[Abstract/Free Full Text]
  5. Hoffman AR, Kuntze JE, Baptista J, Baum HBA, Baumann GP, Biller BMK, Clark RV, Cook D, Inzucchi SE, Kleinberg D, Klibanski A, Phillips LS, Ridgway EC, Robbins RJ, Schlechte J, Sharma M, Thorner MO, Vance ML 2004 Growth hormone (GH) replacement therapy in adult-onset GH deficiency: effects on body composition in men and women in a double-blind, randomized, placebo-controlled trial. J Clin Endocrinol Metab 89:2048–2056[Abstract/Free Full Text]
  6. Biller B, Sesmilo G, Baum H, Hayden D, Schoenfeld D, Klibanski A 2000 Withdrawal of long-term physiological growth hormone (GH) administration: differential effects on bone density and body composition in men with adult-onset GH deficiency. J Clin Endocrinol Metab 85:970–976[Abstract/Free Full Text]
  7. Bengtsson BA, Eden S, Lonn L, Kvist H, Stokland A, Lindstedt G, Bosaeus I, Tolli J, Sjostrom L, Isaksson OG 1993 Treatment of adults with growth hormone (GH) deficiency with recombinant human GH. J Clin Endocrinol Metab 76:309–317[Abstract]
  8. Eiholzer U, l’Allemand D, van der Sluis I, Steinert H, Gasser T, Ellis K.2000 Body composition abnormalities in children with Prader-Willi syndrome and long-term effects of growth hormone therapy. Horm Res 53:200–206.
  9. Carrel AL, Myers SE, Whitman BY, Allen DB 2002 Benefits of long-term GH therapy in Prader-Willi syndrome: a 4-year study. J Clin Endocrinol Metab 87:1581–1585[Abstract/Free Full Text]
  10. Douyon L, Schteingart D 2002 Effect of obesity and starvation on thyroid hormone, growth hormone, and cortisol secretion. Endocrinol Metab Clin North Am 31:173–189[CrossRef][Medline]
  11. Roemmich J, Huerta M, Sundaresan S, Rogol A 2001 Alterations in body composition and fat distribution in growth hormone-deficient prepubertal children during growth hormone therapy. Metabolism 50:537–547[CrossRef][Medline]
  12. Johannsson G, Marin P, Lonn L, Ottosson M, Stenlof K, Bjorntorp P, Sjostrom L, Bengtsson B 1997 Growth hormone treatment of abdominally obese men reduces abdominal fat mass, improves glucose and lipoprotein metabolism, and reduces diastolic blood pressure. J Clin Endocrinol Metab 82:727–734[Abstract/Free Full Text]
  13. Karlsson C, Stenlof K, Johannsson G, Marin P, Bjorntorp P, Bengtsson B, Carlsson B, Carlsson L, Sjostrom L 1998 Effects of growth hormone treatment on the leptin system and on energy expenditure in abdominally obese men. Eur J Endocrinol 138:408–414[Abstract]
  14. Counts D, Gwirtsman H, Carlsson L, Lesem M, Cutler G 1992 The effect of anorexia nervosa and refeeding on growth hormone-binding protein, the insulin-like growth factors (IGFs), and the IGF-binding proteins. J Clin Endocrinol Metab 75:762–767[Abstract]
  15. Golden N, Kreitzer P, Jacobson M, Chasalow F, Schebendach J, Freedman S, Shenker I 1994 Disturbances in growth hormone secretion and action in adolescents with anorexia nervosa. J Pediatr 125:655–660[CrossRef][Medline]
  16. Argente J, Caballo N, Barrios V, Munoz M, Pozo J, Chowen J, Morande G, Hernandez M 1997 Multiple endocrine abnormalities of the growth hormone and insulin like growth factor-I axis in patients with anorexia nervosa: effect of long- and short-term weight recuperation. J Clin Endocrinol Metab 82:2084–2092[Abstract/Free Full Text]
  17. Scacchi M, Pincelli A, Caumo A, Tomasi P, Delitala G, Baldi G, Cavagnini F 1997 Spontaneous nocturnal growth hormone secretion in anorexia nervosa. J Clin Endocrinol Metab 82:3225–3229[Abstract/Free Full Text]
  18. Stoving R, Veldhuis J, Flyvbjerg A, Vinten J, Hangaard j, Koldkjaer O, Kristiansen J, Hagen C 1999 Jointly amplified basal and pulsatile growth hormone secretion and increased process irregularity in women with anorexia nervosa: indirect evidence for disruption of feedback regulation within the growth hormone-insulin like growth factor-I axis. J Clin Endocrinol Metab 84:2056–2063[Abstract/Free Full Text]
  19. Ottosson M, Vikman-Adolfsson K, Enerback S, Olivecrona G, Bjorntorp P 1994 The effects of cortisol on the regulation of lipoprotein lipase activity in human adipose tissue. J Clin Endocrinol Metab 79:820–825[Abstract]
  20. Epel E, McEwen B, Seeman T, Matthews K, Castellazzo G, Brownell K, Bell J, Ickovics J 2000 Stress and body shape: stress-induced cortisol secretion is consistently greater among women with central fat. Psychosom Med 62:623–632[Abstract/Free Full Text]
  21. Garrapa G, Pantanetti P, Arnaldi G, Mantero F, Faloia E 2001 Body composition and metabolic features in women with adrenal incidentaloma or Cushing’s syndrome. J Clin Endocrinol Metab 86:5301–5306[Abstract/Free Full Text]
  22. Lonn L, Kvist H, Ernest I, Sjostrom L 1994 Changes in body composition and adipose tissue distribution after treatment of women with Cushing’s syndrome. Metabolism 43:1517–1522[CrossRef][Medline]
  23. Pirlich M, Biering H, Gerl H, Ventz M, Schmidt B, Ertl S, Lochs H 2002 Loss of body cell mass in Cushing’s syndrome: effect of treatment. J Clin Endocrinol Metab 87:1078–1084[Abstract/Free Full Text]
  24. Wajchenberg B, Bosco A, Marone M, Levin S, Rocha M, Lerario A, Nery M, Goldman J, Liberman B 1995 Estimation of body fat and lean tissue distribution by dual energy x-ray absorptiometry and abdominal body fat evaluation by computed tomography in Cushing’s disease. J Clin Endocrinol Metab 80:2791–2794[Abstract]
  25. Marin P, Oden B, Bjorntorp P 1995 Assimilation and mobilization of triglycerides in subcutaneous abdominal and femoral adipose tissue in vivo in men: effects of androgens. J Clin Endocrinol Metab 80:239–243[Abstract]
  26. Woodhouse L, Gupta N, Bhasin M, Singh A, Ross R, Phillips J, Bhasin S 2004 Dose-dependent effects of testosterone on regional adipose tissue distribution in healthy young men. J Clin Endocrinol Metab 89:718–726[Abstract/Free Full Text]
  27. Tremollieres F, Pouilles J, Ribot C 1996 Relative influence of age and menopause on total and regional body composition changes in postmenopausal women. Am J Obstet Gynecol 175:1594–1600[CrossRef][Medline]
  28. Arabi A, Garnero P, Porcher R, Pelissier C, Benhamou C, Roux C 2003 Changes in body composition during post-menopausal hormone therapy: a 2 year prospective study. Hum Reprod 18:1747–1752[Abstract/Free Full Text]
  29. Veldhuis J, Johnson M 1986 Cluster analysis: a simple, versatile, and robust algorithm for endocrine pulse detection. Am J Physiol 250:E486–E493
  30. Misra M, Miller K, Almazan C, Ramaswamy K, Lapcharoensap W, Worley M, Neubauer G, Herzog D, Klibanski A 2004 Alterations in cortisol secretory dynamics in adolescent girls with anorexia nervosa and effects on bone metabolism. J Clin Endocrinol Metab 89:4972–4980[Abstract/Free Full Text]
  31. Greulich W, Pyle S 1959 Radiographic atlas of skeletal development of the hand and wrist. 2nd ed. Stanford, CA: Stanford University Press
  32. Kelly T, Berger N, Richardson T 1998 DXA body composition: theory and practice. Appl Radiat Isot 49:511–513[CrossRef][Medline]
  33. Visser M, Fuerst T, Lang T, Salamone L, Harris T 1999 Validity of fan-beam dual energy x-ray absorptiometry for measuring fat-free mass and leg muscle mass. Health, aging, and body composition study: Dual Energy X-ray Absorptiometry and Body Composition Working Group. J Appl Physiol 87:1513–1520[Abstract/Free Full Text]
  34. Sopher AB, Thornton JC, Wang J, Pierson Jr RN, Heymsfield SB, Horlick M 2004 Measurement of percentage of body fat in 411 children and adolescents: a comparison of dual-energy x-ray absorptiometry with a four-compartment model. Pediatrics 113:1285–1290[Abstract/Free Full Text]
  35. Elberg J, McDuffie JR, Sebring NG, Salaita C, Keil M, Robotham D, Reynolds JC, Yanovski JA 2004 Comparison of methods to assess change in children’s body composition. Am J Clin Nutr 80:64–69[Abstract/Free Full Text]
  36. Pintauro S, Nagy T, Duthie C, Goran M 1996 Cross-calibration of fat and lean measurements by dual-energy x-ray absorptiometry to pig carcass analysis in the pediatric body weight range. Am J Clin Nutr 63:293–298[Abstract/Free Full Text]
  37. van der Sluis IM, de Ridder MAJ, Boot AM, Krenning EP, de Muinck Keizer-Schrama SMPF, Mughal Z 2002 Reference data for bone density and body composition measured with dual energy x-ray absorptiometry in white children and young adults. Arch Dis Child 87:341–347[Abstract/Free Full Text]
  38. van der Sluis I, Boot A, Hop W, De Rijke Y, Krenning E, de Muinck Keizer-Schrama S 2002 Long-term effects of growth hormone therapy on bone mineral density, body composition, and serum lipid levels in growth hormone deficient children: a 6-year follow-up study. Horm Res 58:207–214[CrossRef][Medline]
  39. Ottosson M, Vikman-Adolfsson K, Enerback S, Elander A, Bjorntorp P, Eden S 1995 Growth hormone inhibits lipoprotein lipase activity in human adipose tissue. J Clin Endocrinol Metab 80:936–941[Abstract]
  40. Ueland T, Kristo C, Godang K, Aukrust P, Bollerslev J 2003 Interleukin-1 receptor antagonist is associated with fat distribution in endogenous Cushing’s syndrome: a longitudinal study. J Clin Endocrinol Metab 88:1492–1496[Abstract/Free Full Text]
  41. Grinspoon S, Thomas L, Miller K, Pitts S, Herzog D, Klibanski A 2001 Changes in regional fat redistribution and the effects of estrogen during spontaneous weight gain in women with anorexia nervosa. Am J Clin Nutr 73:865–869[Abstract/Free Full Text]
  42. Bronnegard M, Ottosson M, Boos J, Marcus C, Bjorntorp P 1994 Lack of evidence for estrogen and progesterone receptors in human adipose tissue. J Steroid Biochem Mol Biol 51:275–281[CrossRef][Medline]
  43. Gower B, Nyman L 2000 Associations among oral estrogen use, free testosterone concentration, and lean body mass among postmenopausal women. J Clin Endocrinol Metab 85:4476–4480[Abstract/Free Full Text]



This article has been cited by other articles:


Home page
 Eur J EndocrinolHome page
M. T. G. De Alvaro, M. T. Munoz-Calvo, V. Barrios, G. Martinez, G. A. Martos-Moreno, F. Hawkins, and J. Argente
Regional fat distribution in adolescents with anorexia nervosa: effect of duration of malnutrition and weight recovery
Eur. J. Endocrinol., October 1, 2007; 157(4): 473 - 479.
[Abstract] [Full Text] [PDF]

Home page
 J. Clin. Endocrinol. Metab.Home page
J. J. Puder, S. Varga, M. Kraenzlin, C. De Geyter, U. Keller, and B. Muller
Central Fat Excess in Polycystic Ovary Syndrome: Relation to Low-Grade Inflammation and Insulin Resistance
J. Clin. Endocrinol. Metab., November 1, 2005; 90(11): 6014 - 6021.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
90/5/2580    most recent
Author Manuscript (PDF)
Right arrow Submit a related Letter to the Editor
Right arrow Purchase Article
Right arrow View Shopping Cart
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Copyright Permission
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Misra, M.
Right arrow Articles by Klibanski, A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Misra, M.
Right arrow Articles by Klibanski, A.
Right arrowPubmed/NCBI databases
*Compound via MeSH
*Substance via MeSH
Medline Plus Health Information
*Eating Disorders
Hazardous Substances DB
*ESTRADIOL
*HYDROCORTISONE
Related Collections
Right arrow Female Endocrinology
Right arrow Metabolism
Right arrow Neuroendocrinology and Pituitary
Right arrow Pediatric Endocrinology

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Endocrinology Endocrine Reviews J. Clin. End. & Metab.
Molecular Endocrinology Recent Prog. Horm. Res. All Endocrine Journals