Main

An increasing body of evidence suggests that the spectrum of variation for numerous endocrine and metabolic variables in adult life is related to fetal growth(1). Pronounced adrenarche and PP in girls, defined as the appearance of pubic hair before the age of 8 y(2), are among the endocrine variants related to reduced fetal growth(3,4). In turn, PP in girls was recently found to be associated with dyslipidemia and low serum IGFBP-1 concentrations(5,6). Accordingly, we have now studied whether the presence of dyslipidemia and low serum IGFBP-1 in PP girls is also linked to reduced prenatal growth and may therefore have a prenatal origin.

METHODS

Study population. The study population consisted of 187 girls, 83 of whom had no history of PP (age, 12.1 ± 2.7 y; range, 6-17 y) and 104 of whom had PP (age, 11.5 ± 3.4 y; range, 5-18 y). Clinical characteristics of prepubertal, pubertal, and postmenarchal subgroups(7) are displayed in Table 1.

Table 1 Clinical characteristics of the study population
Full size table

Girls with PP were only included in the study when PP was secondary to pronounced adrenarche, i.e. after exclusion of nonclassic adrenal hyperplasia(8–11). None of the girls presented evidence for thyroid dysfunction, Cushing syndrome, hyperprolactinemia, diabetes mellitus, or hypertension, nor received medication affecting lipid metabolism. All subjects had normal glucose tolerance(12); insulin responses to an oral glucose load, the lipid profile, and the IGFBP-1 levels in 185 of these girls have been reported within a different context(5,6).

The Institutional Review Board of the Barcelona Hospital approved the study protocol. Informed consent was obtained from the parents and assent from the girls.

Experimental methods. Birth weight and gestational age data were obtained from hospital records and transformed into SD scores, as previously described(4).

Serum lipids and lipoproteins were measured from blood samples after an overnight fast. Samples were centrifuged, and serum was separated and frozen at -20°C until assay. LDL-cholesterol was calculated by the Friedewald formula. Lipoprotein fractionation was performed using ultracentrifugation with a Centrikon T-1035, rotor type TFT 45.6 ultracentrifuge (Kontron International, Zurich, Switzerland). Total HDL was isolated by selective precipitation with phosphotungstic acid-magnesium chloride and measured by the CHOD-PAP method(13). Serum triglycerides were measured using CHOD-PAP- and GPO-PAP-based methods in a Hitachi 917 autoanalyzer (Boehringer Mannheim, Mannheim, Germany). The mean intra- and interassay coefficients of variation were 2.2% and 2.0%, respectively. Serum triglycerides were considered high when >95 mg/dL in girls aged 5-10 y and >110 mg/dL in girls aged 11-18 y(14). LDL/HDL ratios >1.8 were considered high(6).

Serum IGFBP-1 was measured by a quantitative immunometric assay (Medix-Biochemma, Oulu, Finland), as described previously(5). Values were considered low if they were <2.8 µg/L before puberty and <1.5 µg/L in pubertal or postmenarchal girls(5).

Results are expressed as mean ± SEM. Mann-Whitney U test was used for statistical comparisons, with significance at p < 0.05.

RESULTS

Birthweight SD scores of girls with PP (-0.76 ± 0.12) were lower (p < 0.0001) than those of girls without PP (0.38 ± 0.08); these scores correspond to respective birthweights of 3024 ± 60 g (after 39.08 ± 0.17 wk) and 3501 ± 39 g (after 39.06 ± 0.12 wk).

Among girls who had PP (Fig. 1), those with dyslipidemia and low IGFBP-1 had strikingly lower (p < 0.0001) birthweight SD scores (-2.02 ± 0.23 corresponding to 2485 ± 129 g) than those with normal lipids, lipoproteins, and IGFBP-1 (-0.37 ± 0.15 for 3284 ± 57 g), whereas girls with an intermediate number of abnormalities had intermediate birthweight SD scores (-0.80 ± 0.18 for 2914 ± 105 g).

Figure 1
figure 1

Birthweight SD scores of 187 girls either without PP (-) or with PP (+); the latter groups are graded according to an increasing number of abnormal values for serum triglycerides, LDL/HDL ratio, or IGFBP-1.

Full size image

DISCUSSION

The presence of dyslipidemia or low serum IGFBP-1 in girls who had PP was found to depend on weight gain before birth. Previous associations are herewith extended in two directions. First, the link between dyslipidemia and reduced fetal growth, initially recognized in 64-year-old men(15), appears to be already detectable in girls with PP. Second, the spectrum of pediatric endocrinopathies and metabolic anomalies that are somehow related to reduce fetal growth is further broadened to include dyslipidemia and low circulating IGFBP-1 levels, as well as insulin resistance and hyperinsulinism(4,16), growth failure(17,18), exaggerated adrenarche with PP(3,4), ovarian hyperandrogenism and anovulation(4,19), and male subfertility(20). However, the distinct or common mechanisms underlying each of these associations remain poorly understood.

Circulating IGFBP-1 levels correlate inversely with plasma insulin levels, and are considered to be a useful marker for hyperinsulinemia or insulin resistance in subjects with intact endogenous insulin secretion(21–23). Hyperinsulinemia and suppressed IGFBP-1 levels may enhance the availability and insulin-like actions of free IGF-I(24). Increased plasma IGF-I concentrations in low birthweight children showing postnatal catch-up growth has been suggested, together with hyperinsulinemia, as one of the mechanisms linking reduced fetal growth to cardiovascular disease in adult life(25,26).

Increased fasting triglyceride, high LDL, and decreased HDL concentrations are independent risk factors for the early development of cardiovascular disease(27,28). Hyperinsulinemia is the presumed link between dyslipidemia and the other components of syndrome X(29–31). The present results support previous studies in children indicating that this cluster of metabolic abnormalities can already be detected at a young age(6,32,33) and may indeed have a prenatal origin.

In conclusion, the presence of dyslipidemia and low serum IGFBP-1 in girls who had PP was found to be related to prenatal weight gain. The pathophysiologic mechanisms underlying these relationships remain to be elucidated.