Anthropometric parameters, serum
cortisol levels, plasma
CBG,
CBG binding and
insulin sensitivity (using the frequently sampled intravenous
glucose tolerance test with minimal model analysis) were measured in a group of 38 healthy subjects (19 men, mean age 36.2 +/- 1.9; body mass index (BMI) 28.8 +/- 1.2, range 22.2-35.7), and 19 women, age 34.9 +/- 1.4; BMI 28.1 +/- 0.8, range 19-37.9)].
RESULTS: Plasma
CBG levels did not differ between men and women. In men,
CBG binding was associated with several parameters of the
insulin resistance syndrome, including area under the curve for
glucose during an oral
glucose tolerance test (MBG, r = 0.45, P = 0.04), fasting
insulin (r = 0.66, P = 0. 002), plasma
triglycerides (r = 0.75, P < 0.0001), VLDL-
triglycerides (r = 0.59, P = 0.007), fasting FFA (r = 0.72, P = 0.002),
uric acid (r = 0.57 (P = 0.01) and
insulin sensitivity (SI, r = - 0.58, P = 0.008). Free
cortisol (estimated as the ratio of
cortisol to
CBG) was not associated with waist-to-hip ratio (WHR) or parameters of
insulin sensitivity. In contrast to men,
CBG binding was not associated with MBG, fasting
insulin, plasma
triglycerides, VLDL-
triglycerides, FFA,
uric acid or SI (all P = NS) in women. Serum free
cortisol, however, correlated positively with WHR (r = 0. 62, P = 0.02) and negatively with SI (r = - 0.68, P = 0.01) in obese women. A multiple linear regression to predict
CBG binding was constructed, with plasma
CBG concentration and
insulin sensitivity as independent variables. In this model, only SI entered the equation at a statistically significant level (P = 0.0012) contributing to 52% of the variance in
CBG binding in men. When plasma FFA levels were added to the model, both SI (P = 0.04) and FFA levels (P = 0.039) contributed to 66% of the variance of
CBG binding in men. In women, both plasma
CBG concentration (P = 0.0005) and
insulin sensitivity (P = 0.047) entered the equation at a statistically significant level, contributing to 60% of the variance in
CBG binding. When plasma FFA levels were added to the model, only plasma
CBG concentration (P = 0.043) was found to significantly contribute to 38% of the variance in
CBG binding. The latter finding suggests that FFA levels constituted a confounding variable in the association between SI and
CBG binding in women.
CONCLUSIONS: