| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Pediatric Endocrinology Unit (D.M.-M.), Pediatric Psychosomatic Department (D.S., A.Y., M.F.), Institute of Endocrinology (C.P., R.H., H.K.), Adult Eating Disorders Department (A.R.), and Tissue Typing Laboratory (R.L.), Sheba Medical Center, Tel Hashomer, 52621 Israel; Sackler School of Medicine (D.M.-M., D.S., A.Y., R.L.), Tel-Aviv University, Tel-Aviv, 69978 Israel; and The Mina and Everard Goodman Faculty of Life Sciences (E.Y.), Bar-Ilan University, Ramat-Gan, 52900 Israel
Address all correspondence and requests for reprints to: Dalit Modan-Moses, M.D., Division of Pediatrics, Chaim Sheba Medical Center, Tel-Hashomer, 52621 Israel. E-mail: dmodan{at}sheba.health.gov.il.
 |
Abstract |
|---|
 Top Abstract IntroductionPatients and MethodsResultsDiscussionReferences |
|---|
Objective: Our objective was to assess adiponectin levels and the distribution of its different isoforms in AN patients before and after long-term refeeding, and to relate them to alterations in body mass index, leptin, insulin sensitivity, and additional endocrine parameters.
Design, Setting, and Participants: We conducted a longitudinal controlled study of 38 female adolescent malnourished AN inpatients, with 13 young, lean, healthy women serving as controls. Blood samples were obtained upon admission and thereafter at 1, 3, and 5 months (at target weight).
Main Outcome Measures: Changes in body mass index, leptin, adiponectin, insulin sensitivity, and adiponectin multimeric forms were measured.
Results: At admission, leptin levels of AN patients were significantly lower, whereas insulin sensitivity (assessed by homeostasis model assessment-insulin resistance), adiponectin levels, and the ratio of high molecular weight (HMW) adiponectin to total adiponectin were significantly higher compared with controls. During weight recovery, leptin levels and homeostasis model assessment-insulin resistance increased significantly, whereas adiponectin and HMW adiponectin/total adiponectin ratio decreased significantly, to levels similar to controls. An initial increase in adiponectin levels was observed after 1 month of refeeding. There was no correlation between adiponectin and either T4 or cortisol levels.
Conclusions: Our study demonstrates hyperadiponectinemia, increased adiponectin HMW isoform, and increased insulin sensitivity in adolescent AN female patients and reversal of these findings with weight rehabilitation. We hypothesize that increased adiponectin levels may have a protective role in maintaining energy homeostasis during extreme malnourishment.
|
Introduction |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
Anorexia nervosa (AN) is associated with multiple endocrine perturbations, including hypogonadism, hypothyroidism, hypercortisolism, GH resistance, altered glucose metabolism, insulin sensitivity (4), and hypoleptinemia (5, 6). Several studies demonstrated hyperadiponectinemia in underweight AN patients (5, 6, 7, 8, 9, 10), whereas others reported that adiponectin levels were either lower (11) or not different (6) from controls. Furthermore, data regarding the dynamics of changes in adiponectin during refeeding of AN patients is limited and contradicting (7, 9, 11).
In the present study we assessed alterations in adiponectin levels in a large group of AN patients during long-term refeeding and weight restoration and related them to body mass index (BMI), leptin levels, insulin sensitivity, and additional endocrine parameters. We also describe, for the first time, the changes in distribution of adiponectin isoforms during weight gain.
|
Patients and Methods |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
A total of 38 female AN (restricting type) patients, admitted to the Pediatric Psychosomatic Department at the Sheba Medical Center, were prospectively enrolled in this study, which consisted of two substudies.
Study 1, comprising 20 consecutive AN patients, aged 16 ± 5.3 yr, with a mean BMI of 15.9 ± 1.8 kg/m2 (group 1), was designed to evaluate adiponectin levels in AN patients before, during, and after weight rehabilitation and to determine the relation of adiponectin to weight, leptin, and hormonal changes.
Study 2, comprising another 18 AN patients, aged 15.7 ± 1.7 yr, with a mean BMI of 15.9 ± 0.9 kg/m2 (group 2), was designed to evaluate changes in insulin sensitivity during nutritional rehabilitation.
None of the patients had evidence of bulimia nervosa, bipolar, schizophrenic spectrum, or substance abuse disorders, or any significant medical or neurological illness. The patients were hospitalized for the entire duration of the study and were discharged after reaching their target weight and maintaining it for at least 2 wk. Blood samples were obtained from group 1 patients upon admission and thereafter at 1, 3, and 5 months after admission, and from group 2 at the time of admission and after weight rehabilitation (5.6 ± 3 months).
The study protocol was approved by the institutional ethics review committee, and a written informed consent was obtained from all patients and, in the case of minors, from their parents or legal guardians.
Controls
The control group included 13 volunteer young, lean, healthy women (age, 28.8 ± 3.2 yr; BMI, 20.2 ± 2.1 kg/m2).
Height and weight measurements
Standing height was measured to the nearest 0.1 cm and body weight to the nearest 0.1 kg. Measurements were taken during the morning hours by a single investigator, using standardized procedures.
Biochemical assessment
Glucose was measured within 1 h of blood withdrawal by the glucose oxidase method (Olympus AU2700; Olympus, Hamburg, Germany). For other measurements, serum was recovered immediately and frozen at –30 C. Adiponectin and leptin were determined by RIA (Linco, St. Charles, MO), and insulin, 17ß-estradiol, FSH, LH, TSH, free T4, total T3, and cortisol by a chemiluminescent immunometric method (Immulite 2000, Diagnostic Products Corporation, Los Angeles, CA).
Insulin resistance index was calculated from fasting glucose and insulin values, according to the homeostasis model assessment (HOMA) method (12).
Adiponectin multimers
Adiponectin multimers in serum samples were analyzed under nonreducing and nondenaturing conditions as described before (13). The relative proportion of each multimer to total adiponectin was obtained by dividing band densitometry of either multimer by total density in each lane.
Data analysis
Data are reported as mean ± SD. Leptin values were transformed to log scale for improving symmetry of the distributions. The repeated measures model was used for comparison of the transformed values between various time points. Within-patient factor was time after hospitalization. Comparison between SA index at admission and after weight recovery was performed using the Student’s t test. All tests were two-sided and P values < 0.05 were considered significant. Calculations were performed using SPSS 11.0 (SPSS Inc., Chicago, IL).
|
Results |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
Weight gain
Patients’ weight increased significantly (P < 0.001) during hospitalization: group 1, from 39.1 ± 5.3 to 48.5 ± 3.8 kg; group 2, from 40.9 ± 4.5 to 50 ± 4.9 kg. Similarly, BMI increased significantly (P < 0.001) during hospitalization: group 1, from 15.9 ± 1.8 to 19.5 ± 1.1 kg/m2 (Fig. 1A
); group 2, from 15.9 ± 0.9 to 19.3 ± 0.7 kg/m2.
|
During hospitalization, cortisol levels decreased (20.7 ± 7.1 vs. 12.6 ± 5.1 µg/dl; P = 0.009), and total T3 increased (0.97 ± 0.32 vs. 1.55 ± 0.45 nmol/liter; P < 0.001). Expected changes were seen in LH, FSH, and 17ß-estradiol (data not shown).
Leptin and adiponectin
As depicted in Fig. 1B
, upon admission, leptin levels in group 1 were lower compared with the control group (2.5 ± 1.7 vs. 11.2 ± 6.2 ng/ml; P < 0.001), and increased significantly (P < 0.001) during the follow-up period, reaching 9.1 ± 3.6 ng/ml, which was not significantly different from the controls (P = 0.052). In contrast to leptin, adiponectin levels upon admission (Fig. 1C) were significantly higher compared with the control group (16.8 ± 0.8 vs. 11.9 ± 3.3 µg/ml; P < 0.001). Adiponectin levels increased significantly (P = 0.04) during the first month of hospitalization to 19.7 ± 5.6 µg/ml, and thereafter decreased gradually and significantly (P < 0.001), reaching 11.5 ± 0.7 µg /ml after 5 months of refeeding (Fig. 1C), similar to those of controls (P = 0.228). Likewise, in group 2, patients’ adiponectin levels at admission were significantly higher than levels at discharge (16 ± 5.2 vs. 10.5 ± 4.1 µg /ml; P = 0.001). Noteworthy, upon admission, leptin to BMI ratio was significantly (P < 0.001) lower, whereas adiponectin to BMI ratio was significantly (P < 0.001) higher compared with controls. At the 5-month time point, the same trend was observed, but the differences were no longer significant. As shown in Fig. 1, adiponectin levels were inversely correlated with leptin levels and BMI during hospitalization, although these correlations failed to reach statistical significance.
There was no correlation between adiponectin and T4 or cortisol levels, either at admission or by the end of follow-up, in accordance with previous studies (14, 15), but unlike another study (16).
Insulin sensitivity (group 2)
On admission, mean glucose and insulin were 79.8 ± 10.5 mg/dl and 6.4 ± 3.7 mU/liter, respectively. After weight recovery, mean glucose was 84 ± 6.2 mg/dl, whereas mean insulin was 8.9 ± 5.3 mU/liter. Mean HOMA-IR increased significantly (P = 0.05) during hospitalization from 1.28 ± 0.19 to 1.84 ± 0.25.
Adiponectin multimeric isoforms
At admission, HMW/total adiponectin ratio (SA index) was 48.8 ± 9.7%, significantly (P = 0.035) higher compared with the control group (36.4 ± 11.4%). By the time of discharge, the patients’ SA index decreased significantly (P = 0.035) to 33.6 ± 10.3%, not significantly different than the control group. As depicted in Fig. 2, the decrease in SA index resulted mainly from a marked decrease in the HMW multimer.
|
|
Discussion |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
Adiponectin is low in obesity, whereas weight reduction results in an increase of adiponectin levels (2). Hence, we hypothesized that adiponectin in the malnourished state of AN will be elevated and will decline during weight gain. Indeed, adiponectin in the malnourished state was significantly higher compared with the control group. This difference was further amplified when adiponectin levels were normalized for BMI. Our findings are in accordance with previous studies (5, 6, 7, 8, 9, 10), however, other investigators reported that adiponectin in AN patients was lower than (11) or not different from controls (6). The increased adiponectin levels in the malnourished state may be attributed to fat mass depletion, as well as to increased vagal tone observed in AN (18), because ß-adrenergic stimulation has been shown to inhibit adiponectin gene expression (19). Hyperadiponectinemia could contribute to the increased insulin sensitivity observed in AN patients in our study, as well as by others (5, 6, 7, 8, 9, 10, 11). Moreover, because adiponectin blocks B-lymphopoiesis, suppresses macrophage function, and interferes with TNF-
signaling, its elevated levels could contribute to the immune suppression observed in malnourished AN patients (20). Furthermore, it was recently suggested that adiponectin might have a role in maintaining energy homeostasis under energy shortage conditions (21).
Only a few studies investigated longitudinal changes in adiponectin levels during refeeding of AN patients, with conflicting results (7, 9, 11). Similar to our observations, Iwahashi et al. (9) described an initial increase, followed by a gradual decrease in adiponectin levels during the follow-up of a single patient. On the other hand, Tagami et al. (11) reported an increase in adiponectin levels during weight gain in 13 extremely malnourished AN patients, whereas Bosy-Westphal et al. (7) demonstrated a nonsignificant decrease in adiponectin levels in 11 patients after 12 wk of refeeding. The apparent disparity might be attributed to differences in both the severity of malnourishment at admission and the follow-up duration. When integrating the results of the above studies with our own data, obtained in a considerably larger group of patients monitored for a period of 5 months (Fig. 1C), the following model can be proposed: in severely malnourished patients, refeeding results in an initial increase in adiponectin, reflecting early fat accumulation. This was observed in the current study during the first month of weight gain (Fig. 1C), by Tagami et al. (11) during an increase in BMI from 13.8 to 15.9 kg/m2, and by Iwahashi et al. (9). Once a critical threshold of adipose mass is reached, subsequent weight gain is associated with a decrease in adiponectin, as would be expected in a normal population, and as observed by us and by Iwahashi (9) during the final months of hospitalization. We speculate that Bosy-Westpal et al. (7) observed only a nonsignificant decrease in adiponectin due to a shorter follow-up period.
Adiponectin expression may be more closely related to adipocyte size than adipose tissue mass or body weight (22). In addition, the early stage of adipocyte differentiation and development is characterized by an increase in adiponectin gene expression, whereas aged adipocytes display reduced expression (23). Thus, initial weight gain in AN patients might be associated with an increased proportion of young vs. old adipocytes, resulting in an increase in adiponectin, whereas subsequent weight gain is associated with adipocyte hypertrophy, leading to a decrease in adiponectin. Alternatively, because adiponectin levels are inversely correlated with waist-to-hip ratio (24), the changes in adiponectin levels during refeeding may also be attributed to alterations in body fat distribution, because an increase in the amount of visceral fat was observed after weight normalization in AN patients (25). Finally, the initial increase in adiponectin may reflect a lag-phase between changes in fat mass and adiponectin levels. A similar delay between weight loss and alteration in adiponectin was recently reported after bariatric surgery (22).
In the present study, we demonstrated an increased ratio of HMW to total adiponectin (SA) in malnourished AN patients, reflecting mainly higher levels of the HMW form. After weight gain, the distribution of adiponectin multimeric forms was similar to that of controls. These findings are consistent with recent studies that reported an increase in SA index during weight reduction in obese subjects (22). This is the first description of adiponectin multimers in AN patients before and after nutritional rehabilitation. Furthermore, only a handful of studies addressed alterations in adiponectin forms during changes in body weight, and to the best of our knowledge, changes in HMW adiponectin and SA index during weight gain have not been previously described. The decrease in the SA index during the study period was paralleled by a decrease in insulin sensitivity as reflected by HOMA-IR, consistent with studies showing an increase in HMW adiponectin and insulin sensitivity after weight loss (22).
One potential weakness of this study is the age difference between patients and controls. However, our patients were all postpubertal, and previous studies showed no significant age effect on adiponectin levels in postpubertal women (26). Secondly, similar to others (6, 10, 11), we demonstrated increased insulin sensitivity in the malnourished state, using HOMA-IR. However, different results were obtained when direct measures of glucose disposal were used (5).
In conclusion, our study demonstrates hyperadiponectinemia, increased HMW/total adiponectin ratio and increased insulin sensitivity in AN patients, along with reversal of these findings with weight recovery. We hypothesize that increased adiponectin levels may have a protective role in maintaining energy homeostasis during extreme malnourishment. Further studies are warranted to elucidate whether our findings are secondary to the low weight per se or to changes specific to AN.
|
Footnotes |
|---|
First Published Online February 27, 2007
Abbreviations: AN, Anorexia nervosa; BMI, body mass index; HMW, high molecular weight; HOMA, homeostasis model assessment; HOMA-IR, HOMA-insulin resistance; LMW, low molecular weight; MMW, medium molecular weight; SA index, ratio of HMW adiponectin to total adiponectin.
Received August 3, 2006.
Accepted February 20, 2007.
|
References |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  L. Ibanez, G. Sebastiani, A. Lopez-Bermejo, M. Diaz, M. D. Gomez-Roig, and F. de Zegher Gender Specificity of Body Adiposity and Circulating Adiponectin, Visfatin, Insulin, and Insulin Growth Factor-I at Term Birth: Relation to Prenatal Growth J. Clin. Endocrinol. Metab., July 1, 2008; 93(7): 2774 - 2778. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 |
