Study of association and diagnostic utility of novel anthropometric and lipid indices for nonalcoholic fatty liver disease: A hospital-based study.

Abstract

Given the prevalence of nonalcoholic fatty liver disease (NAFLD) of 25% and an expected prevalence of 55.4% by 2040 and its associated complications, it is essential to prioritize large-scale screening programs in the general population. 1. To examine the association of the conventional and novel anthropometric and lipid indices in NAFLD and 2. To determine diagnostic ability, their cutoff values, and best parameter/model for detecting NAFLD. A case-control study was done among 262 NAFLD cases and 186 non-NAFLD controls in a tertiary care hospital. BMI, WC/Ht, LAP, VAI, TC, TG, HDL, LDL, VLDL, Non-HDL-C, TG: HDL-C, TC: HDL-C, LDL-C/HDL-C, TyG, CMI, RC, AIP, and AC were measured/calculated by recommended methods. Binary univariate and multivariate logistic regression analyses and ROC analysis to determine the association, best statistical model, and diagnostic validity of study variables. The individuals with BMI >30 had highest risk (OR = 9.636) of NAFLD followed by those with TG >500 mg/dl (7.054), LAPM (3.157) and LAPF (2.975), WC (M) (3.07), WC: Ht >0.6 (2.893), TyG (2.361), VAI-M (2.053), and VAI-F (2.253) in decreasing order. On ROC analysis, anthropometric parameters outperformed the lipid indices, among which BMI [AUC = 0.702; <i>P</i> < 0.001] and WC [0.671; <i>P</i> < 0.001] showed the highest AUC, followed by WC: Ht [0.657; <i>P</i> < 0.001]. Among lipid indices, LAP had the highest AUC [0.666; <i>P</i> < 0.001], followed by CMI [0.629; <i>P</i> < 0.001] and TyG [0.621; <i>P</i> < 0.001]. BMI, WC, LAP, WC/Ht, VAI, CMI, and TyG are the best simple and reliable indices for predicting/detecting NAFLD in the decreasing order of their diagnostic abilities which can identify the population at risk of NAFLD.