The Association Between Hyperuricemia and Obesity Metabolic Phenotypes in Chinese General Population: A Retrospective Analysis

PURPOSE: Serum uric acid (UA) not only affects the development of obesity but also alters the metabolic status in obese subjects; thus we investigated the relationship between serum UA and the overweight/obese metabolic phenotypes. METHODS: The demographic, biochemical, and hematological data were collected for 12,876 patients undergoing routine physical examination, and 6,912 participants were enrolled in our study. Participants were classified into four obesity metabolic phenotypes according to their BMI and the presence of metabolic syndrome: metabolically healthy overweight/obese (MHOO), metabolically healthy and normal weighted (MHNW), metabolically abnormal and overweight/obese (MAOO), and metabolically abnormal but normal weighted (MANW). Univariate and multivariate logistic regression analysis, stratified analysis, and also interaction analysis were conducted to analyze the relationship between serum UA and obesity metabolic phenotypes. RESULTS: Multivariable logistic regression analysis showed that hyperuricemia was positively associated with MHOO, MANW, and MAOO phenotypes relative to MHNW. After adjusting for the confounding factors, the odds ratios (OR) for individuals with hyperuricemia to be MHOO, MANW, and MAOO phenotypes were 1.86 (1.42–2.45), 2.30 (1.44–3.66), and 3.15 (2.34–4.24), respectively. The ORs for having MHOO, MANW, and MAOO increased 6% [OR: 1.06 (1.05–1.07), P < 0.0001], 5% [OR: 1.05 (1.03–1.07), P < 0.0001], and 11% [OR: 1.11 (1.10–1.13), P < 0.0001] for each 10 unit (μmol/L) of increase in serum UA level. Stratification analysis as well as an interaction test showed that sex and age did not interfere with the association of hyperuricemia with each metabolic phenotype. In terms of the components of the metabolic syndrome, after adjusting for other confounding factors including all of the metabolic indicators except itself, hyperuricemia was positively associated with increased BMI [OR: 1.66 (1.32–2.09), P < 0.0001], hypertriglyceridemia [OR: 1.56 (1.21–2.02), P = 0.0006], and hypertension [OR: 1.22 (1.03–1.46), P = 0.0233], while it had no significant association with hyperglycemia and low HDL-C (all P > 0.05). CONCLUSION: In our study, we discovered that hyperuricemia was positively associated with MHOO, MANW, and MAOO phenotypes, and this relationship was independent of sex and age.