Circulating IL-18 concentration is associated with insulin sensitivity and glucose tolerance through increased fat-free mass.

AIMS/HYPOTHESIS: Knowledge of the factors which simultaneously contribute to insulin-resistance-related inflammation may contribute to early therapeutic targeting. IL-18 has recently been described as one of the factors which, in addition to insulin resistance, may also contribute to atherosclerosis. However, the source of IL-18 is not well characterised.

MATERIALS AND METHODS

We aimed to study body composition (bioelectric impedance), glucose tolerance (OGTT) and insulin sensitivity (minimal model method) in relation to serum IL-18 (ELISA) concentration in 144 otherwise healthy men aged 51.9+/-12.5 years.

RESULTS

In contrast to previous observations in women, circulating IL-18 was not significantly associated with BMI (r=0.12, p=0.1) or WHR (r=0.08, p=0.3). IL-18 was also not associated with absolute or percent fat mass (bioelectric impedance, p>0.20) but, interestingly, it was significantly linked to fat-free mass (p=0.03). Serum IL-18 increased with each quartile of fat-free mass, corresponding to values of < or = 64.2; >64.2 to < or = 71.6; >71.6 to < or = 80.9; and > or = 80.9 kg (ANOVA, p<0.0001). IL-18 was more closely associated with postload glucose during an OGTT (p=0.04) rather than with fasting glucose (p=0.1). HbA1c (p=0.03), HDL-cholesterol (p=0.04) and serum triglycerides (p=0.03) and parameters of systemic inflammation (C-reactive protein, p=0.02) were also significantly associated with circulating IL-18. Insulin sensitivity (minimal model analysis) was linked to circulating IL-18 (p=0.01). In a multiple linear regression analysis this relationship remained significant after controlling for BMI, age and glucose tolerance status. In another model, both fat-free mass and insulin sensitivity contributed to 10% of IL-18 variance.

CONCLUSIONS/INTERPRETATION: Fat mass does not seem to influence circulating IL-18, as initially proposed. In contrast, the fat-free mass compartment (a well-known confounder in the evaluation of insulin sensitivity) may significantly contribute to the relationship between IL-18 and insulin action.