Strong associations between low birth weight and insulin resistance have been described. However, most of these studies have been retrospective. We aimed to determine whether infants born small for gestational age (SGA: birth weight <5th percentile for gestational age) have decreased insulin sensitivity, compared with appropriate for gestational age (AGA: birth weight >10th percentile) at 1 yr of age. We studied blood lipids, fasting insulin levels, other markers of insulin sensitivity, and insulin secretion during an iv glucose tolerance test in a cohort of 85 SGA and 23 AGA 1-yr-old infants. In addition, SGA infants were stratified according to catch-up growth (CUG) in weight (WCUG) or length (LCUG) during the first year of life. At 1 yr, SGA infants had a clear tendency to higher triglycerides. Fasting insulin was significantly higher in SGA infants with WCUG, compared with those who did not catch up and AGA infants (mean +/- SEM, 32.6 +/- 4.6 vs. 14.9 +/- 2.3 vs. 21.4 +/- 3.3 pM, respectively; P < 0.05). Length increment (in SD score) was the principal determinant of postload insulin secretion (R(2) = 0.1, P < 0.01). We conclude that insulin secretion and sensitivity are closely linked to patterns of rapid WCUG and LCUG during early postnatal life. Fasting insulin sensitivity is more related to WCUG and current body mass index, whereas insulin secretion seems to be directly related to LCUG.

Aims/hypothesis: Insulin resistance and type 2 diabetes risk in human subjects who were small-for-gestationalage (SGA) at birth may be a consequence of rapid early postnatal weight gain. Materials and methods: We prospectively studied early changes in fasting insulin sensitivity and insulin secretion, assessed by a short intravenous glucose tolerance test that was conducted several times from birth to 3 years of age in 55 SGA (birthweight below fifth percentile) newborns and in 13 newborns with a birthweight appropriate for gestational age (AGA). Results: Most SGA infants showed postnatal upward weight centile crossing and by 3 years were similar in size to AGA infants. SGA infants had lower pre-feed insulin levels at postnatal age 48 h than AGA infants (median 34.4 vs 59.7 pmol/l, p<0.05), but by the age of 3 years they had higher fasting insulin levels (median 38.9 vs 23.8 pmol/l, p<0.005), which were related to rate of weight gain between 0 and 3 years (r=0.47, p=0.0003). First-phase insulin secretion did not differ between SGA and AGA infants, but SGA infants had a lower glucose disposition index (beta cell compensation) (median 235 vs 501 min mmol, p=0.02), which persisted after allowing for postnatal weight gain (p=0.009). Conclusions/interpretation: SGA infants showed a marked transition from lower pre-feed insulin and increased insulin sensitivity at birth to insulin resistance over the first 3 years of life. This transition was related to rapid postnatal weight gain, which could indicate a propensity to central fat deposition. The additional observation of reduced compensatory beta cell secretion underlines the need for long-term surveillance of glucose homeostasis in all SGA subjects, whether or not they show postnatal catch-up growth.

Adiponectin, a novel adipocytokine with insulin sensitizing properties, is inversely related to obesity and insulin resistance in adults. We recently reported large variations in weight gain and insulin sensitivity during the first year in infants born small for gestational age (SGA) or appropriate for gestational age (AGA). We now determined whether adiponectin levels were related to postnatal growth and insulin sensitivity in a prospective cohort followed from birth to two years old (n = 85) (55 female/30 male, 65 SGA/20 AGA). Serum adiponectin levels at one year and two years were higher compared to reported levels in adults and older children, and decreased from one year (21.6 +/- 0.6 microg/ml) to two years (15.7 +/- 0.7 microg/ml) (p < 0.05). At two years adiponectin levels were lower in females (15.3 +/- 0.4 microg/ml) than males (16.4 +/- 0.6 microg/ml) (p < 0.05), but no gender difference was seen in leptin or insulin levels. No differences in adiponectin levels were seen between SGA and AGA infants at one or two years. However, in SGA infants changes in adiponectin between one to two years old were inversely related to weight gain (r = -0.310, p < 0.05). Changes in leptin levels between one to two years were positively related to weight gain in both SGA and AGA infants (r = 0.450 and r = 0.500 respectively, both p < 0.05). Adiponectin levels were unrelated to insulin levels at one or two years, nor to change in insulin levels between one to two years. In multiple regression analysis, adiponectin levels were related only to postnatal age; omitting age from the model, the determinants of higher adiponectin levels were male gender (p = 0.03), lower postnatal body weight (p < 0.001), and higher birth weight SD score (p = 0.004). In conclusion, fall in serum adiponectin levels during the first two years of life were related to increasing age and greater weight gain SGA infants, but were unrelated to insulin sensitivity.

In early postnatal life, SGA infants display an increased insulin sensitivity with respect to glucose disposal but not with respect to suppression of lipolysis, ketogenesis, and hepatic production of IGFBP-1. It will be important to determine how these differential sensitivities to insulin vary with increasing age.

IGF-I levels increased rapidly from birth in SGA, but not AGA children. During the key first-year growth period, IGF-I levels were related to beta-cell function and longitudinal growth. In contrast, by 3 yr, when catch-up growth was completed, IGF-I levels were related to body mass index and IR, and these higher IGF-I levels in SGA infants might indicate the presence of relative IGF-I resistance.