Insulin resistance is not due to persistently elevated serum tumor necrosis-alpha levels in small for gestational age, premature, or twin children.

Abstract

In pregnancies with small for gestational age (SGA) fetuses, elevated amniotic fluid tumor necrosis factor-alpha (TNF-alpha) levels have been reported. TNF-alpha has been shown to induce insulin resistance in rodents and humans. We hypothesized that an adverse fetal or early neonatal environment for SGA, twin, and premature children leads to persistently elevated TNF-alpha levels that induce insulin resistance in each of these groups. The study group consisted of 16 SGA, 14 premature, 53 twin subjects, and the control group of 40 normal subjects (10 short-stature and 30 normal-stature). All subjects were prepubertal and non-obese. Insulin sensitivity (S(I)) was measured in all but the normal-statured control subjects. Fasting plasma TNF-alpha and cortisol levels were measured in all subjects. The study group had reduced S(I)[SGA 18.5 +/- 3, premature 17.8 +/- 2, twin 12.7 +/- 0.7 (x10(-4)/min/ microU/mL)] compared to the short normal control subjects (43 +/- 8 x 10(-4)/min/ microU/mL, p < 0.001). Plasma TNF-alpha levels were lower in the insulin-resistant study group when compared to the control group (2.9 +/- 0.1 vs. 5.0 +/- 0.2 pg/mL, p < 0.001). An association was present between reduced S(I) and low plasma TNF-alpha levels in the study group (p = 0.01, r = 0.4). Fasting plasma cortisol was lower in the study compared to the control group (266 +/- 16 vs. 341 +/- 28 nmol/L, p < 0.01) due to the influence of the twin study subgroup. There was no relationship between plasma cortisol and TNF-alpha levels (p = 0.3). SGA, premature, and twin children are insulin resistant and have low plasma TNF-alpha and cortisol levels. We speculate that the mechanism leading to insulin resistance in these subjects is also suppressing plasma TNF-alpha and cortisol concentrations.