Skip to main content
The Journal of Clinical Endocrinology and Metabolism logoLink to The Journal of Clinical Endocrinology and Metabolism
. 2009 Nov 19;95(1):369–374. doi: 10.1210/jc.2009-1730

Fibroblast Growth Factor-21 May Mediate Growth Hormone Resistance in Anorexia Nervosa

Pouneh K Fazeli 1, Madhusmita Misra 1, Mark Goldstein 1, Karen K Miller 1, Anne Klibanski 1
PMCID: PMC2805486  PMID: 19926712

Abstract

Context: Anorexia nervosa (AN), a state of chronic nutritional deprivation, is characterized by GH resistance with elevated GH levels and decreased levels of IGF-I. Fibroblast growth factor (FGF)-21, a hormone produced in the liver and adipocytes, is induced in the liver by fasting and peroxisome proliferator-activated receptor-α agonists. In a transgenic mouse model, FGF-21 reduces IGF-I levels by inhibiting signal transducer and activator of transcription-5, a mediator of the intracellular effects of GH.

Objective: The objective of the study was to investigate the relationship between FGF-21, GH, and IGF-I in AN.

Design: This was a cross-sectional study.

Setting: The study was conducted at a clinical research center.

Patients: Patients included 23 girls: 11 with AN (16.5 ± 0.6 yr) and 12 normal-weight controls (15.7 ± 0.5 yr).

Interventions: There were no interventions.

Main Outcome Measures: We measured fasting FGF-21, glucose, insulin, IGF-I, and total area under the curve for GH (GH-AUC) and leptin during 12-h overnight frequent sampling.

Results: FGF-21 levels were significantly higher in AN compared with controls, and there was a positive correlation between FGF-21 and GH-AUC (P = 0.03) after controlling for percent body fat and insulin resistance. In subjects with elevated FGF-21 levels, there was a strong inverse association between FGF-21 and IGF-I (R = −0.88, P = 0.004). FGF-21 strongly correlated with total area under the curve for leptin (R = 0.67, P = 0.02).

Conclusions: FGF-21 levels are higher in AN independent of the effects of percent body fat and insulin resistance. The positive association between FGF-21 and GH-AUC and the inverse association between elevated FGF-21 levels and IGF-I suggests that above the normal range, FGF-21 may mediate a state of GH resistance in AN.


In women with anorexia nervosa, FGF-21 may be a mediator of GH resistance.


Anorexia nervosa (AN), a primary psychiatric disorder characterized by extreme self-imposed starvation, is the third most common chronic illness in adolescent girls (1). Disruption of the normal GH-IGF-I axis is characteristic of undernutrition (2,3,4). Adolescent girls with AN have increased basal and pulsatile secretion of GH and low levels of IGF-I (4). Elevations in GH secretion coupled with low IGF-I levels have led to the concept of a nutritionally mediated state of acquired GH resistance in AN in which endogenous GH is unable to stimulate IGF-I production as effectively as in a fed state. However, the mechanism of GH resistance is unclear.

Fibroblast growth factor (FGF)-21 is a member of the fibroblast growth factor family of proteins. Primarily produced in the liver (5), FGF-21 has recently been shown to be produced in human adipocytes as well (6). Serum FGF-21 levels have been shown to correlate positively with body mass index (BMI) (6) and have been shown to be higher in overweight and obese individuals (7). Serum FGF-21 levels have been shown to correlate with FGF-21 mRNA expression in sc fat (6), suggesting that the higher levels of FGF-21 in overweight and obese individuals may be due to adipocyte-derived FGF-21. FGF-21 transgenic mice have been shown to have lower fasting glucose and leptin levels compared with wild-type littermates (8). Similarly, administration of FGF-21 to ob/ob and db/db mice led to improved glucose clearance coupled with significantly lower insulin levels due to FGF-21’s stimulation of insulin-independent, glucose transporter-1-dependent glucose uptake in adipocytes (8). However, in a cross-sectional study of lean, overweight, and obese humans (6), a positive association between FGF-21 and fasting insulin has been reported, even after adjusting for BMI, suggesting that in states of insulin resistance there may a compensatory increase in FGF-21 levels.

In mouse models, starvation and agonists of peroxisome proliferator-activated receptor-α, a nuclear receptor activated by fatty acids and a key regulator of ketogenesis (9), induce FGF-21 production in the liver (10). FGF-21 levels have also been shown to increase in humans in response to a very low-calorie diet (7). FGF-21 has also been shown to induce ketogenesis (10,11), and transgenic mice have a reduced core body temperature and decreased locomotor activity compared with wild-type mice (10). These findings suggest that elevations in hepatic FGF-21 may be an adaptive mechanism to induce ketogenesis and decrease metabolic activity and energy expenditure in a fasting state.

Recently Inagaki et al. (12) demonstrated that FGF-21 transgenic mice have elevated GH concentrations and a greater than 50% decrease in IGF-I levels compared with wild-type mice. They demonstrated that FGF-21 overexpression leads to a state of GH resistance by inhibiting signal transducer and activator of transcription (STAT)-5, a key transcription factor in the GH signaling cascade (12). Given the state of chronic starvation in AN, we hypothesized that girls with AN would have elevated levels of FGF-21 and that FGF-21 would be a mediator of GH resistance characteristic of AN.

Subjects and Methods

Subjects

Twenty-three adolescent girls were studied: 11 with AN (16.5 ± 0.6 yr, ±1 sem) diagnosed by criteria of the Diagnostic and Statistical Manual of Mental Disorders, fourth edition, and 12 healthy controls of comparable age (15.7 ± 0.5 yr). Demographic and baseline hormonal data have been previously reported (4,13). Girls with AN had a mean BMI of 16.4 ± 0.3 kg/m2 and the mean BMI of the healthy controls was 22.2 ± 1.0 kg/m2. None of the healthy controls had a history of an eating disorder or significant dieting behavior. All subjects (AN and healthy controls) had TSH, FSH, LH, and prolactin levels within the normal range as well as a hematocrit greater than 30%, a potassium level greater than 3 mmol/liter and a glucose level greater than 50 mg/dl. The subjects with AN were recruited through referrals from primary care providers, nutritionists, psychiatrists, and therapists as well as in-patient and day eating disorder treatment programs in Massachusetts, New Hampshire, and Maine. Healthy controls were recruited through mass mailings to primary care providers and advertisements in community newspapers and also within the Partners HealthCare network. Girls with AN were interviewed by a psychiatrist specializing in eating disorders to confirm the Diagnostic and Statistical Manual of Mental Disorders, fourth edition, diagnosis of AN. For healthy controls, absence of psychiatric disorders or disordered eating was ascertained by the investigator performing the history and physical examination based on self-report by the subject and parents. When there was any indication of a history or current symptoms consistent with a psychiatric disorder or disordered eating, the study psychiatrist further evaluated the subject and excluded these diagnoses in our population of healthy controls. Subjects with AN were all enrolled in integrated eating disorder programs at study initiation. The Institutional Review Board of the Partners Health Care system approved the study, and informed assent and consent were obtained from all study subjects and their parents. The clinical characteristics and data on total area under the curve for GH (GH-AUC) have been previously reported for these subjects (4,13).

Experimental protocol

After a screening visit at which a history, physical examination, and laboratory studies to determine eligibility were performed, subjects were admitted for an overnight stay at the Clinical Research Center at Massachusetts General Hospital. Body composition assessment and a bone age were obtained on the day of admission. Subjects were served dinner before 1930 h, and frequent sampling for GH and leptin was performed every 30 min from 2000 h on the night of admission until 0800 h the next morning. After frequent sampling, a fasting blood sample was obtained for glucose, insulin, FGF-21, and IGF-I.

Anthropometric measurements

Subjects were weighed wearing a hospital gown on an electronic scale. A single stadiometer at the Clinical Research Center was used to measure heights of subjects using an average of triplicate measurements. BMI was calculated using the formula [weight (kilograms)/height (meters)2]. A single investigator (M.M.) assessed bone age using the methods of Greulich and Pyle (14). The same investigator, a pediatric endocrinologist, determined Tanner stage in all of the subjects.

Biochemical assessment

Glucose levels were measured by the hospital laboratory using published methods (15). Glucose levels may be converted to SI units (millimoles per liter) by dividing by 18. GH was measured using an immunoradiometric assay (Nichols Institute Diagnostics, San Juan Capistrano, CA) with a detection limit of 0.05 ng/ml and an intraassay coefficient of variation of 2.4–9.4%. An immunoradiometric assay (Nichols Institute Diagnostics) was used to measure serum IGF-I with a detection limit of 30 μg/liter and coefficient of variation of 3.1–4.6%. Leptin was measured with a RIA (Linco Diagnostics, Inc., St. Charles, MO) with a sensitivity of 0.5 μg/liter and a coefficient of variation of 4.6–5.7%. A sandwich ELISA (Bio Vendor, Brno, Czech Republic) was used to measure FGF-21 with a detection limit of 7 pg/ml and an intraassay coefficient of variation of 3–4.1%. RIA was used to measure insulin (Diagnostics Products Corp., Los Angeles, CA; coefficient of variation 4.7–7.7%) and the homeostasis model of assessment of insulin resistance (HOMA-IR) was used as the measure of insulin resistance. HOMA-IR was calculated using the formula: [fasting glucose (millimoles per liter) × fasting insulin (milliunits per milliliter)]/22.5 (16).

Analysis of GH and leptin concentration

The computerized mathematical algorithm, Cluster, was used to determine GH-AUC and total area under the curve for leptin (leptin-AUC) (17).

Body composition

Body composition, including validated measures of fat mass and fat-free mass, was determined by whole-body dual-energy x-ray absorptiometry (DXA; QDR 4500; Hologic, Bedford, MA) (18,19). Fat-free mass refers to soft tissue only and does not include bone. The precision error (sd) of DXA has been reported to be 425 g for whole-body fat and fat-free mass (18) with a correlation coefficient 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 (19).

Statistical methods

The data were analyzed using the JMP program (version 5; SAS Institute, Inc., Cary, NC). All data are presented as mean ± 1 sem. A Student’s t test was used to calculate differences between the means. Where data were not normally distributed, a logarithmic transformation was performed to approximate a normal distribution or nonparametric analyses were performed. We performed univariate and multivariate analyses to determine predictors of FGF-21 and investigate the association between FGF-21, GH, and IGF-I.

Results

Clinical characteristics

The clinical characteristics of the study subjects are presented in Table 1. Subjects with AN had lower weight, BMI, percent body fat, and IGF-I levels compared with controls.

Table 1.

Clinical characteristics of subjects with anorexia nervosa and healthy controls

AN Healthy controls P value
Age (yr) 16.5 ± 0.6 15.7 ± 0.5 NS
Weight (kg) 45.2 ± 1.3 57.5 ± 2.6 0.0007
BMI (kg/m2) 16.4 ± 0.3 22.2 ± 1.0 0.0001
Body fat (by DXA) (%) 18.6 ± 1.4 29.6 ± 1.2 <0.0001
FGF-21 (fasting) (pg/ml) 296.4 ± 70 293.1 ± 53.1 NSa
Leptin (μg/liter) 4.8 ± 1.0 15.9 ± 1.8 <0.0001
IGF-I (fasting) (μg/liter) 315.9 ± 43.7 530.3 ± 41 <0.002
Glucose (fasting) (mg/dl) 82.1 ± 2.2 85 ± 1.4 NS
Insulin (fasting) (μU/ml) 7.0 ± 0.8 14.8 ± 1 <0.0001

Data are reported as mean ± 1 sem. Bold values indicate P < 0.05. NS, Not significant. 

a

P = 0.03 after controlling for percent body fat and insulin resistance. 

Positive predictors of FGF-21

Positive predictors of FGF-21 were determined for the individual groups and the group as a whole. In AN, positive predictors of FGF-21 included fasting glucose (R = 0.80, P = 0.01) and percent body fat (R = 0.64, P = 0.03) (Table 2). Fasting glucose was also a positive predictor of FGF-21 in the group as a whole (R = 0.46, P = 0.04) (Table 2). HOMA-IR was a positive predictor of FGF-21 in healthy controls (R = 0.78, P = 0.003) (Table 2).

Table 2.

Positive predictors of FGF-21

AN
Healthy controls
All subjects
R P value R P value R P value
Body fat (by DXA) (%) 0.64 0.03 −0.28 0.40 0.15 0.51
Glucose (fasting) 0.80 0.01 0.11 0.73 0.46 0.04
HOMA-IR 0.30 0.43 0.78 0.003 0.42 0.06
Leptin (total area under the curve) 0.67 0.02 −0.012 0.97 0.11 0.62
Leptin (mean value) 0.72 0.01 −0.03 0.93 0.12 0.60

Bold values indicate P < 0.05. NS, Not significant. 

FGF-21 levels in AN and healthy controls

FGF-21 levels were not different in AN compared with healthy controls (AN: 296.4 pg/ml vs. healthy control: 293.1 pg/ml, P = 0.97). After controlling for insulin resistance and percent body fat, FGF-21 levels were significantly higher in AN compared with controls (P = 0.03). Because FGF-21 is produced in adipocytes and the liver, this suggests increased production of liver-derived FGF-21 in AN.

Relationship of FGF-21 to leptin

FGF-21 was found to be significantly and positively correlated with leptin-AUC (R = 0.67, P = 0.02) in AN (Table 2). This relationship was not observed in the healthy controls or in the group as a whole.

Relationship of FGF-21 to GH and IGF-I

After controlling for HOMA-IR and percent body fat, there was a positive correlation between GH-AUC and log FGF-21 (P = 0.03). We also investigated the relationship between IGF-I and elevated FGF-21 levels (>300 pg/ml). A value of greater than 300 pg/ml was chosen based on previously published data in healthy controls demonstrating that values greater than 300 pg/ml are approximately 1 sd or more above the mean value for FGF-21 in these populations (6,7,20,21). In subjects with elevated FGF-21 levels, there was a significant inverse association between FGF-21 and IGF-I (R = −0.88, P = 0.004). This relationship remained significant after controlling for HOMA-IR and percent body fat (P = 0.01).

Discussion

We have shown that FGF-21 levels are elevated in AN compared with healthy controls after controlling for percent body fat and insulin resistance and that there is a strong positive correlation between FGF-21 and leptin-AUC in AN. We have also shown that there is a positive association between GH-AUC and FGF-21 and an inverse correlation between IGF-I and elevated levels of FGF-21, suggesting that FGF-21 is a mediator of GH resistance in AN.

AN is a state of extreme, self-induced nutritional deprivation characterized by GH resistance. Studies have shown elevated GH levels in girls and women with AN and low IGF-I levels (4). This state of GH resistance is thought to be nutritionally mediated and occurs at the level of the liver, the site of IGF-I induction by GH.

FGF-21 is a member of the FGF family of proteins. It has been shown to be produced by the liver (5), and more recently, adipocytes have been found to be an important source of FGF-21 production (6). Studies suggest a dichotomy in the secretion of FGF-21 by fat or the liver, depending on body weight and energy availability. Obese and overweight individuals have elevated levels of FGF-21 compared with controls, and serum levels of FGF-21 correlate with FGF-21 mRNA expression in sc fat (6), suggesting that the elevated FGF-21 levels observed in obesity are likely due to increased levels of adipocyte-derived FGF-21. In contrast, mouse models demonstrate that starvation induces increased production of liver-derived FGF-21 (10). Serum FGF-21 levels may thus be a function of both absolute fat mass and the extent of starvation.

Recently Mráz et al. (7) demonstrated that serum FGF-21 levels also increase in humans on a very low-calorie diet and measured mRNA expression of FGF-21 in sc and visceral fat as well as the liver. Because mRNA expression was measured in a cross-sectional manner, it is not known how liver mRNA expression of FGF-21 would respond to the very low-calorie diet. Also, although mRNA expression of liver FGF-21 was found to be 100-fold higher than FGF-21 mRNA expression in adipose tissue, it is not known how mRNA expression relates to serum levels of FGF-21 (7). It is possible that a significant portion of the FGF-21 produced in the liver is immediately metabolized. Therefore, further studies are needed in humans to further understand the expression of adipose-derived FGF-21 compared with liver-derived FGF-21 in response to various experimental states.

Inagaki et al. (12) also recently demonstrated that FGF-21 mediates a state of GH resistance in a transgenic mouse model. Mice overexpressing FGF-21 were found to have elevated levels of GH and low levels of IGF-I (12). The mechanism by which FGF-21 induces this state of GH resistance was shown to be decreased phosphorylation of signal transducer and activator of transcription-5 (12). Our data demonstrating a positive correlation between FGF-21 and GH-AUC and an inverse association between FGF-21 and IGF-I support the idea that FGF-21 mediates the GH resistance in AN, a state of chronic starvation.

There are limited data regarding FGF-21 levels in AN. The only other study examining FGF-21 in AN reported lower FGF-21 levels in adult women with AN than in controls (20). In contrast, our data demonstrate similar levels of FGF-21 in adolescent girls with AN and controls before adjusting for insulin resistance and percent body fat and higher levels in AN after controlling for these variables, which are known to impact FGF-21. The fact that FGF-21 is produced in adipocytes and individuals with AN have significantly lower fat depots compared with controls suggests that our finding of comparable levels of FGF-21 in girls with AN and controls may be attributable to elevated liver-derived FGF-21 offsetting the decrease in fat derived FGF-21. One explanation as to why we found similar levels of FGF-21 in AN and healthy controls whereas Dostálová et al. (20) found significantly lower levels of FGF-21 in AN compared with controls is that their population of women with AN was leaner than our population of adolescent girls and the difference in mean BMI between their AN and healthy control groups (mean BMI of AN group: 15.9 ± 0.33 kg/m2 and mean BMI of healthy controls: 22.9 ± 0.41 kg/m2) was greater than the difference in BMI between our two groups (mean BMI of AN group: 16.4 ± 0.3 kg/m2 and mean BMI of healthy controls: 22.2 ± 1.0 kg/m2). Although these differences in BMI appear modest, an important factor is the age of the subjects in relation to BMI. The differences in BMI are actually larger than they appear because Dostálová et al. studied adults and our study population consisted of adolescents. A BMI of 15.9 kg/m2 in a 20-yr-old (or an adult) would be a much lower BMI (sd score of −3.0) than a BMI of 16.4 in a 16.5-yr-old (BMI sd score of −2.0), based on 2000 Centers for Disease Control and Prevention growth data.

In comparing FGF-21 levels in AN and healthy controls, we also controlled for insulin resistance because serum FGF-21 levels have been shown to be positively associated with fasting insulin levels (6), and FGF-21 is an important mediator of insulin-independent glucose uptake (8). Based on these known effects of insulin resistance on FGF-21, we would have expected FGF-21 levels to be lower in AN compared with healthy controls, as was demonstrated by Dostálová et al. (20). Yet after controlling for insulin resistance, we found that FGF-21 levels are in fact higher in AN compared with healthy controls, suggesting that there is an independent effect of the state of chronic starvation that leads to increases in FGF-21 in this population, similar to the liver-derived induction of FGF-21 observed in mouse models of chronic starvation (10).

Similar to Dostálová et al. (20), we found FGF-21 levels to be positively associated with leptin, which is also secreted by fat. Of importance, in the study by Dostálová et al. (20), FGF-21 levels decreased with refeeding, suggesting that with refeeding, liver-derived FGF-21 decreases, potentially before adipocyte-derived FGF-21 is able to increase. Importantly, the study by Dostálová et al. (20) did not examine the relationship between FGF-21 and GH secretion.

The limitations of our study include the cross-sectional design and the fact that we cannot therefore determine causality based on these data. The small size of our study population is also a limitation. Yet the fact that significant associations were demonstrated in such a small group reflects the strength of the associations. Also, it is currently impossible to differentiate between adipocyte-derived FGF-21 and liver-derived FGF-21 in the human model. Our data, coupled with the data in the obesity model (6) suggest that there may be differential actions of liver-derived and adipocyte-derived FGF-21.

In conclusion, our data demonstrate that FGF-21 levels are significantly higher in AN compared with healthy controls after controlling for percent body fat, a significant determinant of FGF-21 levels in the obese population (6), and insulin resistance. We have also shown a significant positive association between FGF-21 and GH-AUC as well as a significant inverse association between elevated FGF-21 levels and IGF-I levels. This suggests that there may be a threshold above which FGF-21 mediates GH resistance in AN.

Acknowledgments

We acknowledge Michael Mannstadt, M.D., for his helpful discussion regarding the biology of FGF hormones. We also thank the nurses and bionutritionists of the Massachusetts General Hospital Clinical Research Center for their expert care.

Footnotes

This work was supported in part by National Institutes of Health Grants R01 DK062249 and T32 DK007028 and National Center for Research Resources Grant ULI RR025758 to the Harvard Clinical and Translational Science Center. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Center for Research Resources or the National Institutes of Health.

Disclosure Summary: The authors have no conflicts to declare.

First Published Online November 19, 2009

Abbreviations: AN, Anorexia nervosa; BMI, body mass index; DXA, dual-energy x-ray absorptiometry; FGF, fibroblast growth factor; GH-AUC, area under the curve for GH; HOMA-IR, homeostasis model of assessment of insulin resistance; leptin-AUC, AUC for leptin.

References

  1. Lucas AR, Beard CM, O'Fallon WM, Kurland LT 1991 50-year trends in the incidence of anorexia nervosa in Rochester, Minn.: a population-based study. Am J Psychiatry 148:917–922 [DOI] [PubMed] [Google Scholar]
  2. Scacchi M, Pincelli AI, 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 [DOI] [PubMed] [Google Scholar]
  3. Støving RK, Veldhuis JD, Flyvbjerg A, Vinten J, Hangaard J, Koldkjaer OG, Kristiansen J, Hagen C 1999 Jointly amplified basal and pulsatile growth hormone (GH) secretion and increased process irregularity in women with anorexia nervosa: indirect evidence for disruption of feedback regulation within the GH-insulin-like growth factor I axis. J Clin Endocrinol Metab 84:2056–2063 [DOI] [PubMed] [Google Scholar]
  4. Misra M, Miller KK, Bjornson J, Hackman A, Aggarwal A, Chung J, Ott M, Herzog DB, Johnson ML, 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 [DOI] [PubMed] [Google Scholar]
  5. Nishimura T, Nakatake Y, Konishi M, Itoh N 2000 Identification of a novel FGF, FGF-21, preferentially expressed in the liver. Biochim Biophys Acta 1492:203–206 [DOI] [PubMed] [Google Scholar]
  6. Zhang X, Yeung DC, Karpisek M, Stejskal D, Zhou ZG, Liu F, Wong RL, Chow WS, Tso AW, Lam KS, Xu A 2008 Serum FGF21 levels are increased in obesity and are independently associated with the metabolic syndrome in humans. Diabetes 57:1246–1253 [DOI] [PubMed] [Google Scholar]
  7. Mráz M, Bartlova M, Lacinova Z, Michalsky D, Kasalicky M, Haluzikova D, Matoulek M, Dostálová I, Humenanska V, Haluzik M 2009 Serum concentrations and tissue expression of a novel endocrine regulator fibroblast growth factor-21 in patients with type 2 diabetes and obesity. Clin Endocrinol (Oxf) 71:369–375 [DOI] [PubMed] [Google Scholar]
  8. Kharitonenkov A, Shiyanova TL, Koester A, Ford AM, Micanovic R, Galbreath EJ, Sandusky GE, Hammond LJ, Moyers JS, Owens RA, Gromada J, Brozinick JT, Hawkins ED, Wroblewski VJ, Li DS, Mehrbod F, Jaskunas SR, Shanafelt AB 2005 FGF-21 as a novel metabolic regulator. J Clin Invest 115:1627–1635 [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Rodriguez JC, Gil-Gómez G, Hegardt FG, Haro D 1994 Peroxisome proliferator-activated receptor mediates induction of the mitochondrial 3-hydroxy-3-methylglutaryl-CoA synthase gene by fatty acids. J Biol Chem 269:18767–18772 [PubMed] [Google Scholar]
  10. Inagaki T, Dutchak P, Zhao G, Ding X, Gautron L, Parameswara V, Li Y, Goetz R, Mohammadi M, Esser V, Elmquist JK, Gerard RD, Burgess SC, Hammer RE, Mangelsdorf DJ, Kliewer SA 2007 Endocrine regulation of the fasting response by PPARα-mediated induction of fibroblast growth factor 21. Cell Metab 5:415–425 [DOI] [PubMed] [Google Scholar]
  11. Badman MK, Pissios P, Kennedy AR, Koukos G, Flier JS, Maratos-Flier E 2007 Hepatic fibroblast growth factor 21 is regulated by PPARα and is a key mediator of hepatic lipid metabolism in ketotic states. Cell Metab 5:426–437 [DOI] [PubMed] [Google Scholar]
  12. Inagaki T, Lin VY, Goetz R, Mohammadi M, Mangelsdorf DJ, Kliewer SA 2008 Inhibition of growth hormone signaling by the fasting-induced hormone FGF21. Cell Metab 8:77–83 [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Misra M, Miller KK, Herzog DB, Ramaswamy K, Aggarwal A, Almazan C, Neubauer G, Breu J, Klibanski A 2004 Growth hormone and ghrelin responses to an oral glucose load in adolescent girls with anorexia nervosa and controls. J Clin Endocrinol Metab 89:1605–1612 [DOI] [PubMed] [Google Scholar]
  14. Greulich W, Pyle S 1959 Radiographic atlas of skeletal development of the hand and wrist. 2nd ed. Stanford, CA: Stanford University Press [Google Scholar]
  15. 1992 Case records of the Massachusetts General Hospital. Weekly clinicopathological exercises. Normal reference laboratory values. N Engl J Med 327:718–724 [DOI] [PubMed] [Google Scholar]
  16. Radziuk J 2000 Insulin sensitivity and its measurement: structural commonalities among the methods. J Clin Endocrinol Metab 85:4426–4433 [DOI] [PubMed] [Google Scholar]
  17. Veldhuis JD, Johnson ML 1986 Cluster analysis: a simple, versatile, and robust algorithm for endocrine pulse detection. Am J Physiol 250:E486–E493 [DOI] [PubMed] [Google Scholar]
  18. Kelly TL, Berger N, Richardson TL 1998 DXA body composition: theory and practice. Appl Radiat Isot 49:511–513 [DOI] [PubMed] [Google Scholar]
  19. Visser M, Fuerst T, Lang T, Salamone L, Harris TB 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 [DOI] [PubMed] [Google Scholar]
  20. Dostálová I, Kaválková P, Haluziková D, Lacinová Z, Mráz M, Papezová H, Haluzik M 2008 Plasma concentrations of fibroblast growth factors 19 and 21 in patients with anorexia nervosa. J Clin Endocrinol Metab 93:3627–3632 [DOI] [PubMed] [Google Scholar]
  21. Li H, Bao Y, Xu A, Pan X, Lu J, Wu H, Lu H, Xiang K, Jia W 2009 Serum fibroblast growth factor 21 is associated with adverse lipid profiles and γ-glutamyltransferase but not insulin sensitivity in Chinese subjects. J Clin Endocrinol Metab 94:2151–2156 [DOI] [PubMed] [Google Scholar]

Articles from The Journal of Clinical Endocrinology and Metabolism are provided here courtesy of The Endocrine Society

RESOURCES