| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Endocrinology Unit (L.I., M.D.), Hospital Sant Joan de Déu, University of Barcelona, 08950 Esplugues, Barcelona, Spain; Diabetes, Endocrinology and Nutrition Unit (A.L.-B.), Dr. Trueta Hospital, 17007 Girona, Spain; Endocrinology Unit (M.V.M.), Hospital de Terrassa, 08227 Terrassa, Spain; and Department of Woman & Child (F.d.Z.), University of Leuven, 3000 Leuven, Belgium
Address all correspondence and requests for reprints to: Lourdes Ibáñez, M.D., Ph.D., Endocrinology Unit, Hospital Sant Joan de Déu, University of Barcelona, Passeig de Sant Joan de Déu, 2, 08950 Esplugues, Barcelona, Spain. E-mail: libanez{at}hsjdbcn.org.
| Abstract |
|---|
|
|
|---|
Setting, Design, Patients, Intervention: Prepubertal LBW girls with PP (mean body weight 2.4 kg; age 7.9 yr; body mass index 18.4 kg/m2) were studied. Girls were randomly assigned to remain untreated (n=19) or receive metformin for 4 yr (n = 19; 425 mg/d for 2 yr, then 850 mg/d for 2 yr).
Main Outcomes: At the start and after 4 yr, height, weight, fasting insulin, glucose, IGF-I, testosterone, lipids, leptin, high molecular weight adiponectin, body composition by absorptiometry, abdominal fat partitioning (only 4 yr) by magnetic resonance imaging, and menarcheal status were determined.
Results: Metformin-treated girls gained on average 5.5 kg (or
50%) less fat, after 4 yr were less insulin resistant and less hyperandrogenic, had lower IGF-I levels and a less atherogenic lipid profile, and were less likely to be post-menarcheal than untreated girls, whereas their gain in height, lean mass, and bone mineral density were similar. After 4 yr, untreated girls had more visceral fat, a higher ratio of visceral-to-sc fat, and a higher leptin-to-high molecular weight adiponectin ratio (all
50% higher) than metformin-treated girls.
Conclusion: Long-term metformin treatment appears to reduce total and visceral fat in LBW-PP girls, and to delay menarche without attenuating linear growth, thereby opening the perspective that adult height may be increased.
| Introduction |
|---|
|
|
|---|
LBW girls with precocious pubarche (PP) (pubic hair < 8 yr) are known to be at particular risk for developing an adipose state of hyperinsulinemic androgen excess that leads toward an early menarche with growth arrest (4, 5, 6). After using metformin in post-PP adolescents (7, 8) and in post-menarcheal LBW-PP girls (9), we initiated a first randomized study exploring the early and prolonged use of metformin in LBW-PP girls, thereby focusing on body composition and on pubertal growth and maturation. We reported the prepubertal results after 6-month study (10) and the first pubertal results after 2 yr (11); here, we unveil the first post-menarcheal results after a 4-yr study.
| Subjects and Methods |
|---|
|
|
|---|
The study population consisted of 38 LBW-PP girls, distributed into two well-matched subgroups of 19 girls each, whose baseline characteristics were reported (10). In the total population, birth weight (mean ± SEM) was 2.4 ± 0.1 kg after 38.6 ± 0.4 wk, age at diagnosis of PP was 6.8 ± 0.2 yr, age at study start 7.9 ± 0.1 yr, bone age 9.0 ± 0.1 yr, height 129.4 ± 1.2 cm, weight 31.0 ± 0.9 kg, body mass index (BMI) 18.4 ± 0.3 kg/m2, dehydroepiandrosterone sulfate (DHEAS) at PP diagnosis 102 ± 6 µg/dl, and post-ACTH 17-hydroxyprogesterone 274 ± 16 ng/dl.
As described (10), the inclusion criteria were: 1) PP due to exaggerated adrenarche, as judged by high-serum DHEAS and/or androstenedione levels; 2) weight less than 2.9 kg at term birth (38–41 wk) or below –1 SD for gestational age at preterm birth (33–37 wk); 3) BMI less than 22 kg/m2, which corresponds to the +2 SD cutoff in girls aged approximately 8 yr (12); and 4) prepuberty (Tanner B1). None of the girls had a family or personal history of diabetes mellitus, or presented evidence for thyroid dysfunction, glucose intolerance, or late-onset congenital adrenal hyperplasia; none was receiving a medication known to affect gonadal function or carbohydrate metabolism.
The study was registered as ISRCTN84749320 and was approved by the Institutional Review Board of Barcelona University, Hospital of Sant Joan de Déu. Informed consent was obtained from parents and assent from the girls.
Study design and assessments
Girls were randomly assigned, as described (10, 11), to remain untreated or receive metformin, once daily at dinnertime (425 mg for 2 yr, then 850 mg for 2 yr).
Clinical examination was performed six monthly by the same investigator (L.I.); assessment of serum insulin, fasting blood glucose, SHBG, testosterone, DHEAS, a lipid profile, and body composition were each performed six monthly for 2 yr, and yearly thereafter. Serum leptin was measured at 0, 2, and 4 yr, and high molecular weight (HMW) adiponectin [the physiologically most relevant isoform of this adipokine (13)] was assessed at 0 and 4 yr. Height was measured with a Harpenden stadiometer (Holtain Ltd., Crosswell, Crymych, UK). Menarcheal timing was derived by six-monthly history.
Body composition was assessed by dual-energy x-ray absorptiometry with a Lunar Prodigy coupled to Lunar software (Lunar Corp., Madison, WI), as described (11).
After 4 yr, sc and visceral adipose tissue areas in the abdominal region were assessed by magnetic resonance imaging (MRI) using a multiple-slice MRI 1.5 Tesla scan (Signa LX Echo Speed Plus Excite; General Electric, Milwaukee, WI), as described (3). All scans were performed by the same operator (blinded to treatment allocation), and all images were analyzed by the same radiologist (also blinded to treatment allocation).
Variations in dietary intake and physical activity level were assessed yearly with validated questionnaires (14, 15). Frequency and quantity of food intake were recorded by each girl, under parental guidance, for 7 d before the annual control visit (14). Assisted by another questionnaire, parents reported the daily activities of their child during the week before the annual control visit. Girls were classified into one of four activity levels (15): 1, low (no regular physical activity); 2, moderate (sporadically involved in physical activities); 3, high (recreational activity
3 times per week, 30–60 min per session); and 4, vigorous (intense physical activity
4 times per week, > 60 min per session).
Hormone assays, calculations, and statistics
Serum glucose was measured by the glucose oxidase method. Serum immunoreactive insulin, SHBG, testosterone, DHEAS, and IGF-I were assayed as described (8, 9). Fasting insulin sensitivity was estimated from fasting insulin and glucose levels using the homeostasis model assessment (HOMA) (HOMA-CIGMA Calculator program v2.00; Diabetes Research Laboratory, Oxford, UK). Leptin was measured by RIA (LINCO Research, Inc., St. Charles, MO), and HMW adiponectin was measured by sandwich ELISA (LINCO Research), as described (16, 17). Samples from both subgroups were kept frozen until assay and were assayed concomitantly.
Statistical analyses were performed with SPSS 12.0 (SPSS, Inc., Chicago, IL). t tests were used to compare the changes within each subgroup and the 0- to 4-yr changes between subgroups. Differences in longitudinal data between subgroups were tested by repeated measures ANOVA. The between-subgroup difference in menarcheal prevalence was assessed by
2. For uniformity, all results are expressed as mean ± SEM. The level of statistical significance was set at P < 0.05.
| Results |
|---|
|
|
|---|
50% higher) than metformin-treated girls. Between 42 and 48 months, annualized height velocity started to be lower in untreated (and mostly post-menarcheal) girls than in treated girls (4.0 ± 0.8 cm/yr vs. 5.0 ± 0.6 cm/yr; P = 0.04), an observation corroborating the perspective that metformin treatment may be followed by a taller stature in adulthood.
|
|
|
Metformin was well tolerated. Pill counts at each study visit indicated that the treatment compliance was good, the mean number of missed doses being 1 per month. Indices of hepatic and renal function remained unchanged throughout treatment. None of the patients dropped out of the study.
| Discussion |
|---|
|
|
|---|
After a 4-yr study, untreated LBW-PP girls had an abdominal ratio of visceral-to-sc fat that was elevated, certainly when compared with such a ratio in 12-yr-old girls with simple overweight or obesity (19, 20). Reassuringly, after 4 yr, metformin treatment was accompanied by a lower amount of visceral fat and by a strikingly lower ratio of visceral-to-sc fat in the abdomen, and, thus, by an evolution toward a more favorable metabolic phenotype (20).
The mechanisms underpinning the apparent effects of metformin in LBW-PP girls are poorly understood. Among the long-known actions of metformin are an improvement in insulin sensitivity in muscle and liver, a decrease in hepatic glucose production via gluconeogenesis, an increase in peripheral glucose use (mainly via a stimulation of insulin-mediated muscle glucose uptake and glycogen synthesis), and positive effects on insulin receptor expression and tyrosine kinase activity; in addition, metformin suppresses the gluconeogenic effects of glucagon and increases the translocation of glucose transporters to the cell surface (21, 22). Metformin also appears to have a direct effect on ovarian steroidogenesis, specifically to reduce both androgen and estradiol production (23, 24). LKB1 and AMP-activated protein kinase were recently identified as being among the prime targets of metformin's actions (25, 26). Finally, microarray studies showed that metformin is capable of inducing a hepatic gene expression pattern that mimics the profile induced by calorie restriction (27).
Our long-term study (spanning about one third of the girls' lifetime, so far) had no placebo-controlled design. This limitation implies that the observed reductions in fat gain are attributable either to metformin and/or some placebo effect that, in turn, could be conferred directly by the medication or indirectly, for example, by unreported lifestyle changes in diet and/or exercise.
In conclusion, long-term metformin treatment appears to reduce total and visceral fat markedly in LBW-PP girls and to delay menarche without attenuating linear growth, thereby opening the perspective that adult height may be increased.
| Acknowledgments |
|---|
| Footnotes |
|---|
First Published Online March 4, 2008
Abbreviations: BM, Body mass index; DHEAS, dehydroepiandrosterone sulfate; HMW, high molecular weight; HOMA, homeostasis model assessment; LBW, low birth weight; MRI, magnetic resonance imaging; PP, precocious pubarche.
Received January 3, 2008.
Accepted February 25, 2008.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
L. Ibanez, A. Lopez-Bermejo, M. Diaz, M. V. Marcos, P. Casano, and F. de Zegher Abdominal Fat Partitioning and High-Molecular-Weight Adiponectin in Short Children Born Small for Gestational Age J. Clin. Endocrinol. Metab., March 1, 2009; 94(3): 1049 - 1052. [Abstract] [Full Text] [PDF] |
||||
![]() |
R. L. Barbieri Update in Female Reproduction: A Life-Cycle Approach J. Clin. Endocrinol. Metab., July 1, 2008; 93(7): 2439 - 2446. [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 |