Serum plasminogen activator inhibitor (PAI)-1, adiponectin, resistin, and retinol-binding protein 4 (RBP4) levels may help predict progression from normal glucose tolerance to prediabetes or type 2 diabetes (T2D), according to study results published in The Journal of Clinical Endocrinology & Metabolism.
As the prevalence of T2D is increasing dramatically, it is important to have measures to predict the development of abnormal glucose homeostasis. The goal of the current study was to explore whether cytokine levels in people without known T2D could be used to predict changes in glucose metabolism over 10 years.
The prospective longitudinal observational study included patients aged 40 to 69 years from the Ansung cohort of the Korea Genome Epidemiology Study, an ongoing community-based cohort study established to examine trends in T2D and risk factors. The present study included 912 individuals with available data from a 75-g oral glucose tolerance test and 8 cytokine levels measured at baseline (2001-2002), including PAI-1, resistin, interleukin (IL)-6, leptin, monocyte chemoattractant protein-1 (MCP-1), tumor necrosis factor α (TNF-α), RBP4, and adiponectin.
After exclusion of patients with preexisting T2D at baseline or missing data, the study cohort included 241 individuals with normal glucose tolerance and 330 with prediabetes at baseline. The patients were regularly examined every 2 years for 10 years. For each patient who converted from normal glucose tolerance to prediabetes or T2D after the 10-year follow-up, the researchers selected a matched control patient who had not developed prediabetes or T2D.
Multinomial logistic regression analysis was used to identify predictors of new-onset prediabetes and T2D at 10-year follow-up. The data were adjusted for traditional risk factors.
Of 241 individuals with normal glucose tolerance at baseline, 38 (15.8%) developed prediabetes and 82 (34.0%) developed T2D at 10-year follow-up examination. Of 330 individuals with prediabetes at baseline, 228 (69.1%) went on to develop T2D.
In people without T2D at baseline, resistin and RBP4 levels were significant predictors of T2D development at 10 years. In patients with resistin levels in the highest tertile, the risk of developing T2D was more than 2-fold higher (odds ratio [OR], 2.34; 95% CI, 1.43-3.81; P =.001) compared with patients who had levels in the lowest tertile. Levels of RBP4 within the highest tertile were associated with a 2.64 times greater risk for progression to T2D (95% CI, 1.59-4.37; P <.001) compared with levels in the lowest tertile.
In patients with normal glucose tolerance at baseline, PAI-1 levels in the highest tertile were associated with a significantly increased risk for prediabetes (OR, 3.23; 95% CI, 1.13-9.22; P =.028) compared with levels in the lowest tertile. The risk for prediabetes was more than 3-fold higher in patients with adiponectin levels in the middle (OR, 3.65; 95% CI, 1.13-11.76; P =.030) and lowest (OR, 3.37; 95% CI, 0.98-11.56; P =.053) tertiles compared with patients who had levels in the highest tertile. With regard to RBP4 concentrations, levels in the lowest tertile were associated with a 5.48 times greater likelihood of developing T2D (95% CI, 1.87-16.03; P =.002) than levels in the lowest tertile.
In the group with prediabetes at baseline, the highest tertiles of resistin (OR, 2.94; 95% CI, 1.27-6.84; P =.012) or RBP4 (OR, 2.43; 95% CI, 1.10-5.34; P =.028) were associated with an increased risk for progression to T2D compared with the lowest tertiles.
The researchers noted several study limitations, including its observational design, enrollment of only middle-aged adults in a rural area of Korea, and the use of only a single measurement for cytokine levels.
“We suggest that measuring the levels of various adipokines and cytokines in healthy adults without diabetes may be helpful for predicting the risk of future diabetes,” concluded the researchers.
Cho NH, Ku EJ, Jung KY, et al. Estimated association between cytokines and the progression to diabetes: 10-year follow-up from a community-based cohort. J Clin Endocrinol Metab. 2020;105(3):1-9.