Inflammatory biomarkers and prediction of insulin resistance in Congolese adults

Reine Freudlendrich Eboka-Loumingou Sakou; Benjamin Longo-Mbenza; Mûnka Nkalla-Lambi; Etienne Mokondjimobe; Henry Germain Monabeka; Donatien Moukassa; Ange Antoine Abena; Mia Pamela Mekieje Tumchou; Venant Tchokonte-Nana

doi:10.1016/j.heliyon.2021.e06139

Inflammatory biomarkers and prediction of insulin resistance in Congolese adults

Heliyon. 2021 Feb 2;7(2):e06139. doi: 10.1016/j.heliyon.2021.e06139. eCollection 2021 Feb.

Authors

Affiliations

¹ National Public Health Laboratory, Brazzaville, Republic of the Congo.
² Faculty of Health Sciences, Marien Ngouabi University, Brazzaville, Republic of the Congo.
³ Lomo University of Research, Kinshasa, Democratic Republic of the Congo.
⁴ Department of Internal Medicine, University of Kinshasa, Democratic Republic of the Congo.
⁵ Municipal Clinic Albert Leyono, Brazzaville, Republic of the Congo.
⁶ University Hospital Centre, Brazzaville, Republic of the Congo.
⁷ General Hospital Edith Lucie Bongo Ondimba, Oyo, Republic of the Congo.
⁸ Faculty of Health Sciences, University des Montagnes, Bangangte, Cameroon.

Abstract

Several studies have shown that low levels of adiponectin (ADP) and high levels of alpha tumor necrosis factor (NFT) increase the risk or severity of many cardiometabolic diseases associated with insulin resistance. The main objective of this study was to evaluate the association between plasma adipokines and IR measured by HOMA-IR. The secondary objective was to determine the biomarker of the potential inflammation to predict IR in Congolese melanoderm subjects residing in Brazzaville. This cross-sectional study was conducted on 234 apparently healthy participants over the age of 18. Socio-demographic and clinical data were collected. Biological data, including the total ADP and NFT dosage, were measured using the ELISA method. Participants were categorized into two groups according to HOMA-IR ≥ 2.5. Univariate and multivariate logistic regression analyses were conducted to identify risk factors for insulin resistance. An optimized model was obtained after the logistic regression. The analysis of the receptor's operating characteristics (OCR) was performed to determine the optimal threshold value and diagnostic characteristics, as well as the area under the curve (ASC). ADP averages were significantly low (11.49 ± 7.61 ng/mL; P < 0.001) while those of TNF were significantly higher (96.03 ± 44.09 pg/mL) in the HOMA-IR group ≥ 2.5. There was a positive and significant correlation (p < 0.05) between BMI, TT, CRPhs, TNF and HOMA-IR. And a negative and significant correlation was noted between ADP and HOMA-IR (r = - 0.39; P < 0.01). Similarly, a negative and significant correlation (p < 0.01) was noted between BMI, TT, TNF, CRPhs and ADP. The optimal threshold value of the total ADP for predicting IR was 17.52 ng/mL with a sensitivity of 89% [IC 95% (0.83-0.95)], 56% specificity [IC 95% (0.47-0.65)] and a CSA of 0.76 [IC 95% (0.69-0.81)]. After logistic regression, the CSA of the optimized model was 0.84 [IC 95% (0.79-0.89)]. ADP can be used as a highly plausible IR prediction biomarker.

Keywords: Adiponectin; HOMA-IR; Insulin resistance; Melanodermic; Tumor necrosis factor.