Associations of Predicted Body Fat Mass and Fat Percentage with Insulin Resistance among a Middle-aged and Elderly Non-diabetic Population

RUAN Haiyan,LI Liying,ZHENG Yi,CHEN Xiaoping,HE Sen

DOI: https://doi.org/10.12114/j.issn.1007-9572.2022.0181

2022-01-01

Chinese General Practice

Abstract:Background Identifying insulin resistance (IR) is of important clinical value as it is a pathological basis of multiple diseases, such as metabolic syndrome, type 2 diabetes, hypertension, and atherosclerosis. Numerous previous studies have suggested that excessive gain of fat is closely associated with IR. Dual-energy X-ray absorptiometry is often clinically used as the golden standard for measuring human body fat, but its health economic cost is high. Objective To assess the associations of predicted fat mass (pFM) and predicted fat percent (pFP) with IR in a middle-aged and elderly population without diabetes in Chengdu. Methods In May 2007, a cross-sectional survey was conducted by trained medical workers in Chenghua District, Chengdu, Sichuan. A total of 711 middle-aged and elderly individuals were selected as surveyees by random cluster sampling, and 570 of them were included for analysis after excluding 141 with diabetes. The survey includes filling in a standardized questionnaire, a physical examination, and a laboratory test. The homeostasis model assessment of IR (HOMA-IR) , pFM, and pFP were calculated. Results Among the 570 subjects, the range of HOMA-IR was from 0.01 to 5.50, with a median value of 1.05 (0.72, 1.53) ; the range of pFM was from 2.34 to 41.41 kg, with a value of mean (SD) of (18.59±6.47) kg; the range of pFP was from 13.95% to 48.46%, with a median value of 28.97 (23.77, 37.39) %. The prevalence of IR was 20.17% (115/570) . The results suggested the HOMA-IR value and IR prevalence increased with the increase of pFM and pFP (P<0.001) . Multivariate Logistic regression analysis indicated that both pFM〔OR=3.22, 95%CI (2.45, 4.33) 〕 and pFP〔OR=3.51, 95%CI (2.61, 4.84) 〕 were associated with IR (P<0.05) . For predicting IR, the area under the receiver operating characteristic curve (AUC) was 0.777〔95%CI (0.733, 0.822) 〕with 38.10 kg as the optimal cut-off value for pFM, and 0.719〔95%CI (0.668, 0.769) 〕with 17.74% as the optimal cut-off value for pFP, respectively. And the AUC for four common traditional anthropometric indicators, namely waist circumference, body mass index, waist-to-height ratio (WHtR) and a body shape index (ABSI) was 0.749〔95%CI (0.700, 0.798) 〕, 0.753〔95%CI (0.706, 0.801) 〕, 0.768〔95%CI (0.722, 0.815) 〕, and 0.618〔95%CI (0.561, 0.675) 〕, respectively. Especially, the AUC of pFM or pFP was greater than that of ABSI (Z=4.452, P<0.001; Z=2.486, P=0.013) . Conclusion In this middle-aged and elderly non-diabetic population of Chengdu, pFM and pFP were independently associated with IR, and also could well discriminate IR. But pFM and pFP was only superior to ABSI.