Nutrients (Nov 2019)

Obesity- and Lipid-Related Parameters in the Identification of Older Adults with a High Risk of Prediabetes According to the American Diabetes Association: An Analysis of the 2015 Health, Well-Being, and Aging Study

  • Robinson Ramírez-Vélez,
  • Miguel Ángel Pérez-Sousa,
  • Katherine González-Ruíz,
  • Carlos A. Cano-Gutierrez,
  • Jacqueline Schmidt-RioValle,
  • María Correa-Rodríguez,
  • Mikel Izquierdo,
  • Jesús Astolfo Romero-García,
  • Adriana Yolanda Campos-Rodríguez,
  • Héctor Reynaldo Triana-Reina,
  • Emilio González-Jiménez

DOI
https://doi.org/10.3390/nu11112654
Journal volume & issue
Vol. 11, no. 11
p. 2654

Abstract

Read online

This study evaluated the predictive ability of 11 obesity- and lipid-related parameters, including body mass index (BMI), waist circumference (WC), waist-to-height ratio (WtHR), body roundness index (BRI), “A” body-shape index (ABSI), conicity index (C), visceral adiposity index (VAI), triglyceride-to-glucose fasting index (TyG), triglyceride-to-glucose fasting related to BMI (TyG-BMI), triglyceride-to-glucose fasting related to WC (TyG-WC), and triglyceride-to-glucose fasting related to WtHR (TyG-WtHR), to identify patients from an elderly Colombian population with a high risk of prediabetes according to the 2016 American Diabetes Association criteria. The data were obtained from the 2015 Colombian Health and Wellbeing and Aging Survey. A total of 3307 elderly Colombian individuals (aged over 60 years) were included. Anthropometric data, fasting plasma glucose, blood lipid profiles, family history, and health-related behaviors were assessed, and prediabetes was defined as a fasting plasma glucose of 100 to 125 mg/dL. The areas under the receiver operating characteristic (ROC) curves (AUCs) were calculated for each anthropometric indicator, using the prediabetes classification to identify their sensitivity and specificity, and these indicated that the prevalence of prediabetes was 25.3% in this population. After adjusting for potential confounding factors, the TyG index was strongly associated with the odds of having prediabetes in both sexes, and multivariate logistic regression analysis showed that the ORs for prediabetes increased across quartiles (p < 0.001). The TyG index was best able to identify prediabetes in either sex (AUC and optimal cut-off = 0.700 and 8.72, and 0.695 and 8.92 for men and women, respectively), suggesting that compared to the other parameters, the TyG index has the best discriminative power to predict prediabetes in the whole population. Thus, we propose the TyG index be used as a complementary marker for assessing prediabetes in older adults.

Keywords