Carneiro Isabella Barbosa Pereira, Sampaio Helena Alves de Carvalho, Carioca Antônio Augusto Ferreira, Pinto Francisco José Maia, Damasceno Nágila Raquel Teixeira
Centro de Ciências da Saúde, Universidade Estadual do Ceará, Fortaleza, CE, Brasil.
Faculdade de Saúde Pública, Universidade de São Paulo, São Paulo, SP, Brasil.
Arq Bras Endocrinol Metabol. 2014 Nov;58(8):838-43. doi: 10.1590/0004-2730000003296. Epub 2014 Nov 1.
Despite the importance of insulin resistance (IR) on chronic diseases development, its diagnosis remains invasive. Thus, it's necessary to develop alternative methods to predict IR on clinical practice, and the anthropometric indices are a good alternative to it. Given that, this study's purpose is to evaluate these indices behavior in relation to HOMA-IR (Homeostasis Model Assessment of Insulin Resistance).
We collected weight, height and waist circumference from 148 adolescents. Through these indices, we calculated the body mass index (BMI), inverted body mass index (iBMI), waist-to-height ratio (WHtR) and conicity index (C index). We also collected data from body composition (body fat percentage - %BF), through electric impedance, and biochemical data (fasting glucose and insulin levels) employed on the HOMA-IR calculation. The HOMA-IR cutoff adopted was of 2.39±1.93. The statistical analysis involved the Spearman correlation analysis, multiple linear regression models and ROC (Receiver Operating Characteristic) curves construction, using 95% CI. We used the statistic pack SPSS v.18, considering p<0.05 as the significance level.
All anthropometric indices were statistically and positively correlated to HOMA-IR. The ROC curve showed that WC, WHtR and C index, in this order, were the most efficient to predict IR.
Among the indicators studied, those related to central fat accumulation seem the most suitable for predicting IR.
尽管胰岛素抵抗（IR）对慢性疾病发展具有重要意义，但其诊断仍具侵入性。因此，有必要开发替代方法以在临床实践中预测IR，人体测量指标是很好的替代选择。鉴于此，本研究旨在评估这些指标与HOMA-IR（胰岛素抵抗稳态模型评估）的关系。
我们收集了148名青少年的体重、身高和腰围。通过这些指标，我们计算了体重指数（BMI）、反体重指数（iBMI）、腰高比（WHtR）和锥度指数（C指数）。我们还通过电阻抗收集了身体成分数据（体脂百分比-%BF），以及用于HOMA-IR计算的生化数据（空腹血糖和胰岛素水平）。采用的HOMA-IR临界值为2.39±1.93。统计分析包括Spearman相关性分析、多元线性回归模型和ROC（受试者工作特征）曲线构建，使用95%置信区间。我们使用统计软件包SPSS v.18，将p<0.05视为显著性水平。
所有人体测量指标与HOMA-IR均呈统计学正相关。ROC曲线显示，WC、WHtR和C指数按此顺序是预测IR最有效的指标。
在所研究的指标中，与中心性脂肪堆积相关的指标似乎最适合预测IR。
