Conicity Index as an Indicator of Abdominal Obesity in Renal Individuals Undergoing Hemodialysis: An Analysis of Latent Classes

OBJECTIVES: To identify the cutoff point of the conicity index in renal individuals undergoing hemodialysis. METHODS: Observational, cross-sectional study with 953 individuals undergoing hemodialysis between February and September 2019 in a metropolitan region in southeastern Brazil. The conicity index (CI) was calculated using the following equation: waist circumference/0.109 x √Weight/Height. To determine the cut-off point, the analysis of latent classes was used by cross-validation through a latent variable (VL) of abdominal obesity. The LV was defined using the pattern of responses of the observed anthropometric variables considering the presence and absence of abdominal obesity: waist circumference, waist-to-height ratio and body shape index. The information criteria: Akaike information criterion (AIC), Bayesian information criterion (BIC) and entropy indicated the most adequate model of two latent classes. The identified cutoffs were elucidated by the area under the curve (AUC), sensitivity and specificity. RESULTS: The cutoff points were similar between the two sexes, with the CI cutoff point being 1275 for men with sensitivity and specificity of 83% and 83.60%, respectively, and 1285 for women with sensitivity and specificity of 78.60% and 89.30%, respectively, with an AUC of 0.921 (P < 0.001) for both sexes. Abdominal obesity, determined by the CI, showed a total of 59.5% of individuals with excess weight (p-value: 0.001). CONCLUSIONS: The cutoff points for the CI for men and women showed good accuracy for the identification of abdominal obesity in renal patients on hemodialysis. In this way, we demonstrate the reliability of its use in clinical practice. FUNDING SOURCES: This study received financial support from the Espírito Santo Research and Innovation Support Foundation (FAPES) No. 35,081.543.19306.18042018. Notice no. 03/2018 - Research Program for the SUS (PPSUS).