Insulin resistance (IR) has been witnessing an upward trend in incidence rates, with a high comorbidity rate with obesity, dyslipidemia and hyperuricemia. However, the correlation between IR, obesity/dyslipidemia and SUA levels is not well understood.

To explore the dose-response correlation between homeostatic model assessment for IR (HOMA-IR) and SUA levels, and to evaluate the mediating effect of HOMA-IR on obesity/lipids and SUA levels.

We conducted a retrospective study that a total of 3 928 adults who underwent physical examination in the Seventh Affiliated Hospital, Southern Medical University from November 2021 to December 2022 were selected and divided into males and females, non-IR and IR subgroups. The dose-response correlation between HOMA-IR and SUA was analyzed by restricted cubic spline (RCS) . The mediating effects of HOMA-IR on correlation between obesity, total cholesterol (TC) , triacylglycerol (TG) , low-density lipoprotein cholesterol (LDL-C) , high-density lipoprotein cholesterol (HDL-C) and SUA levels were analyzed. Linear regression model was used to analyze the correlation between SUA levels and HOMA-IR, fasting blood insulin (FINS) , and fasting blood glucose (FBG) .

Among the 3 928 patients, 3 026 were males and 902 were females. Among the 3 026 males, 757 were IR patients and 2 269 were non-IR patients. The levels of systolic blood pressure (SBP) , diastolic blood pressure (DBP) , SUA, HOMA-IR, FINS, FBG, BMI, waist-hip ratio (WHR) , TG, TC and LDL-C were higher in IR patients than in non-IR patients, age and HDL-C levels were lower in IR patients than in non-IR patients (P<0.05) . Among the 902 females, 226 were IR patients and 676 were non-IR patients. The levels of SBP, DBP, SUA, HOMA-IR, FINS, FBG, BMI, WHR, TG, LDL-C were higher in IR patients than in non-IR patients, HDL-C levels were lower in IR patients than in non-IR patients (P<0.05) . The RCS results showed that there was a non-linear dose-response correlation between HOMA-IR and SUA levels after adjusting covariates, with gender differences presented in this correlation. The SUA levels in males exhibited an approximate ∩-shaped trend (P_nonlinear=0.002) with the increase of HOMA-IR. While in females, SUA levels showed a trend of increasing and then constant (P_nonlinear=0.002) . Whereas, as SUA levels increased, HOMA-IR showed a trend of decreasing and constant in males (P_nonlinear=0.002) , while HOMA-IR showed a U-shaped trend in females (P_nonlinear=0.002) . The results of mediation analysis showed that HOMA-IR partially mediates the correlation between obesity, central obesity, TG and SUA, and the ratio of mediation effect to the total effect was 9.5%, 7.9% and 5.3% in males, 36.1%, 19.9% and 6.5% in females, respectively. HOMA-IR completely mediated the correlation between HDL-C and SUA, and the ratio of mediating effect to total effect was 20.0% in males and 22.0% in females. The patients were stratified by SUA=7.28 mg/dL in males and 4.70 mg/dL in females, respectively. After adjusting covariates, in males, the results of linear regression model analysis showed a negative correlation between SUA and FBG levels (P<0.05) , and individuals SUA≥7.28 mg/dL showed a positive correlation between SUA and FINS levels (P<0.05) ; in females, individuals SUA≥4.70 mg/dL showed a positive correlation between HOMA-IR, FINS and SUA (P<0.05) .

There is a nonlinear dose-response correlation between HOMA-IR and SUA levels. HOMA-IR has a partial mediating effect on the correlation between obesity, TG and SUA levels, and a complete mediating effect on the correlation between HDL-C and SUA levels. FBG is negatively correlated to SUA levels in males.