Elevated Levels of the Anti-Inflammatory Interleukin-1 Receptor Antagonist Precede the Onset of Type 2 Diabetes

RESEARCH DESIGN AND METHODS—

We present results from a nested case-control study within the Whitehall II cohort, which was established in 1985 and included 10,308 civil servants age 35–55 years (11). Phase 3 (1991–1994) of the study was the first phase where glucose tolerance was assessed by a 75-g oral glucose tolerance test and is the baseline for the current study (n = 7,537). Participants were followed through postal questionnaires at 2.5-year intervals (phases 4–8) and further clinical examinations (including an oral glucose tolerance test) in 1997–1999 (phase 5) and 2003–2004 (phase 7) (12). Individuals without type 2 diabetes at baseline and with incident type 2 diabetes during the follow-up period of 11.5 ± 3.0 years served as case subjects (n = 181). The control subjects (n = 376), with normal glucose tolerance at baseline and during the follow-up, were frequency matched to case subjects for age (5-year bands), sex, and BMI (5 kg/m² bands). Further details on the selection criteria for this nested case-control study and information on the collection of anthropometric, metabolic, socioeconomic, and immunological variables and on statistical analysis are given in an online appendix (available at http://dx.doi.org/10.2337/dc08-1161).

RESULTS—

Characteristics of case and control subjects are shown in Table A1 of the online appendix. The comparison of the included and excluded diabetic case subjects revealed only a few significant differences: selected case subjects were less likely to be women and smokers, were more likely to be ex-smokers, and had a marginally lower BMI. Other comparisons between the two groups were not significant (data not shown). Characteristics of control subjects in our selection mainly reflected the selection criteria (normal glucose tolerance throughout the study and matching for age, sex, and BMI to case subjects): they were slightly older and had a higher BMI, waist circumference, and diastolic blood pressure but lower fasting and 2-h glucose compared with the rest of the cohort who were nondiabetic at baseline and follow-up (data not shown).

IL-1Ra concentrations at baseline were higher in case subjects (median 232.8 pg/ml [25th–75th percentiles 180.7–342.2]) than in control subjects (207.6 pg/ml [159.3–274.8]) (P = 0.0006). A 1-SD increase of IL-1Ra (157.7 pg/ml) was associated with incident type 2 diabetes in models that adjusted for multiple potential confounders, including age, sex, waist circumference, cardiovascular risk factors, socioeconomic status, proinflammatory mediators, and fasting glycemia (Table 1). Further inclusion of BMI had virtually no impact on odds ratio (model 2 plus BMI odds ratio 1.41 [95% CI 1.13–1.77]; P = 0.0027). However, addition of 2-h glucose led to reduced effect sizes and loss of statistical significance (Table 1).

CONCLUSIONS—

Elevated levels of IL-1Ra were associated with an increased risk of developing type 2 diabetes in this nested case-control study. We found a slight attenuation of this association when adjusting for waist circumference, which is consistent with IL-1Ra production in adipose tissue (13). It is remarkable that the association was stable to adjustment for a range of further potential confounders, including fasting glucose and insulin. However, adjustment for 2-h glucose attenuated the association. This finding could be interpreted to indicate that increased IL-1Ra levels are a reaction to and not a cause of early postprandial hyperglycemia before the onset of diabetes. Studies with measurements of glycemic markers and IL-1Ra at multiple time points and analysis of their trajectories will be needed to clarify this point.

Our findings expand observations from cross-sectional studies that reported elevated levels of IL-1Ra in the circulation of individuals with obesity and insulin resistance (8–10). Thus, individuals who will develop type 2 diabetes are characterized not only by an upregulation of proinflammatory immune mediators (1–3) but also by the upregulation of at least one anti-inflammatory immune marker. Because animal studies and a recent clinical trial indicated that administration of IL-1Ra attenuates subclinical inflammation, supports β-cell function/insulin secretion, and may also improve insulin sensitivity (6,14), it is tempting to speculate that elevated IL-1Ra levels in individuals at risk of type 2 diabetes may be an attempt to counteract the proinflammatory effects of interleukin-1β and to preserve insulin secretion and insulin sensitivity—an effort that eventually fails. However, our data cannot rule out the alternative interpretation that IL-1Ra has additional metabolic effects beyond the inhibition of interleukin-1β that could lead to insulin resistance and type 2 diabetes.

As a potential limitation of the study, it should be mentioned that point estimates and CIs are derived from nonweighted data from a nested case-control study and, therefore, their statistical inference may be restricted and may not represent the best available estimate within the context of the original cohort. Thus, further studies will be needed to support our hypothesis that individuals with high risk of type 2 diabetes are characterized by the presence of an early compensatory, anti-inflammatory response that precedes the development of the disease. This hypothesis could be tested by the analysis of further, mainly anti-inflammatory immune mediators, such as interleukin-10 or transforming growth factor-β in additional cohorts.