Search

Among girls with precocious pubarche (PP), those with low birth weight (LBW) are, even if nonobese, at risk for progression to polycystic ovary syndrome (PCOS) including hyperinsulinemic hyperandrogenism, dyslipidemia, dysadipocytokinemia, and central fat excess. Recently, we disclosed the efficacy of insulin sensitization with metformin to disrupt progression from PP to PCOS in formerly LBW girls who were postmenarche. In LBW-PP girls, we have now extended the exploration of early insulin sensitization therapy in two directions: 1) metformin therapy was started before puberty; and 2) we assessed the effects of metformin discontinuation in girls who had started metformin treatment after menarche. Prepubertal LBW-PP girls (n = 33; mean age, 8.0 yr; body mass index, 18.5 kg/m(2)) were randomly assigned to remain untreated or to receive metformin (425 mg/d) for 6 months. Postpubertal LBW-PP girls (n = 24; age, 12.4 yr; body mass index, 21.0 kg/m(2)) had been randomized (at -12 months) to remain untreated or to receive metformin (850 mg/d) for 12 months, at which time (0 month) a treatment cross-over was performed for 6 months. Fasting blood glucose and serum insulin, SHBG, dehydroepiandrosterone sulfate, androstenedione, testosterone, lipid profile, IL-6, and adiponectin were assessed at 0 and 6 months, as was body composition (by dual x-ray absorptiometry). In the prepubertal study (group A), comparisons of untreated vs. treated girls disclosed normalizing effects of metformin on SHBG, androstenedione, dehydroepiandrosterone sulfate, low and high density lipoprotein cholesterol, triglycerides, IL-6, adiponectin, total and abdominal fat mass, and lean body mass. In the postpubertal study (group B), treatment cross-over at 0 month was in each subgroup followed by a striking reversal in the course of the endocrine-metabolic state, adipocytokinemia, and body composition; all changes pointed to normalizing effects of metformin treatment. In conclusion, these two studies provide the first evidence that 1) prepubertal metformin therapy has normalizing effects on PCOS features in high risk girls with a combined history of LBW and PP; and 2) in adolescence, metformin's normalizing effects are reversed as soon as metformin therapy is discontinued.

Adolescent girls born small for gestational age (SGA) are at risk for anovulation, hyperinsulinism, subclinical hyperandrogenism, dyslipidemia, and central adiposity. Hyperinsulinemic insulin resistance has been proposed as a key pathogenetic factor underpinning these associations. We have tested this hypothesis in an intervention study by assessing the effects of insulin sensitization (metformin treatment, 850 mg/d for 3 months) in eumenorrheic, nonobese, anovulatory SGA adolescents [n = 13; mean birth weight, 2.3 kg; age, 15 yr; body mass index (BMI), 20.5 kg/m(2); >or=3 yr post-menarche] who were in a steady state (over approximately 6 months) for BMI, hyperinsulinism, subclinical hyperandrogenism, and dyslipidemia, and who presented a deficit of lean body mass and an excess of (truncal and abdominal) fat mass. Metformin treatment was accompanied by a drop in fasting insulin and serum androgens and by a less atherogenic lipid profile (all P

Low birth weight associations with hyperinsulinemia and other adulthood disease risk factors have been described in several cohorts, including girls who present with precocious pubarche (pubic hair <8 yr). We hypothesized that these associations might be influenced by the insulin gene (INS) variable number of tandem repeat (VNTR), a common polymorphism related to INS transcription levels. In 141 Caucasian girls, who presented with precocious pubarche, hyperinsulinemia was assessed from mean insulin levels during an oral glucose load (MSI), and insulin sensitivity was determined from fasting glucose and insulin levels. Fasting blood lipid profiles were also measured. DNA was genotyped for INS VNTR allele class (I or III) in precocious pubarche girls and in 140 age- and body mass index-matched control girls. INS VNTR genotype distribution was similar in precocious pubarche and control girls. However among precocious pubarche girls, INS VNTR genotype was related to the severity of phenotype; I/I and I/III genotypes had lower birth weights (P < 0.01), higher MSI (P < 0.005), and lower insulin sensitivity (P < 0.005) than III/III girls. In precocious pubarche girls, birth weight was also inversely related to MSI (r = -0.29; P < 0.0005), total cholesterol (r = -0.38; P < 0.0005), and low density lipoprotein cholesterol (r = -0.44; P < 0.0005). Using logistic regression, additive adverse effects of I/* genotype and low birth weight were seen on MSI (P = 0.03 and P = 0.004, respectively) and total cholesterol levels (P = 0.01 and P < 0.0001). In summary, in girls who presented with precocious pubarche, hyperinsulinemia and dyslipidemia were related to both low birth weight and INS VNTR class I alleles. A similar interaction between genotype and intrauterine growth restraint may underlie other reported links between low birth weight and adulthood disease risks.

Precocious pubarche in girls is often preceded by low weight at birth and followed by hirsutism, ovarian hyperandrogenism, and oligomenorrhea in adolescence, the latter usually being accompanied by dyslipidemia and hyperinsulinism, which are, in turn, two major risk factors for cardiovascular disease in later life. We hypothesized that insulin resistance may be a key pathogenetic factor in this sequence. We tested the hypothesis by assessing the effects of an insulin-sensitizing agent, metformin, given at a daily dose of 1275 mg for 6 months to 10 nonobese adolescent girls (mean age, 16.8 yr; body mass index, 21.9 kg/m2; birth weight, 2.7 kg) with hirsutism, ovarian hyperandrogenism (diagnosis by GnRH agonist test), oligomenorrhea, dyslipidemia, and hyperinsulinemia after precocious pubarche. Before the metformin trial, longitudinal studies in these girls had shown that hyperinsulinism was present at prepubertal diagnosis of precocious pubarche, and that it increased markedly in late puberty or early postmenarche. Metformin treatment was well tolerated and was accompanied by a marked drop in hirsutism score, insulin response to oral glucose tolerance test, free androgen index, and baseline testosterone, androstenedione, dehydroepiandrosterone, and dehydroepiandrosterone sulfate levels (all P < 0.01). During metformin treatment, the LH and 17-hydroxyprogesterone hyperresponses to GnRH agonist were attenuated (P < 0.01); serum triglyceride, total cholesterol, and low density lipoprotein cholesterol levels decreased; and high density lipoprotein cholesterol rose. All girls reported regular menses within 4 months. Withdrawal of metformin treatment was followed, within 3 months, by a consistent reversal toward pretreatment conditions. In conclusion, metformin treatment reduced hyperinsulinemia, hirsutism, and hyperandrogenism; attenuated the LH and 17-hydroxyprogesterone hyperresponses to GnRH agonist; improved the atherogenic lipid profile; and restored eumenorrhea in nonobese adolescent girls with a history of precocious pubarche. These observations corroborate the idea that insulin resistance may indeed be a prime factor underpinning the sequence from reduced fetal growth, through precocious pubarche, to adolescent endocrinopathies that are reminiscent of so-called polycystic ovary syndrome.