Different‐shaped curves for serum uric acid with and without diabetes: Results from China Health and Retirement Longitudinal Study

• Published in: Journal of diabetes Vol. 11; no. 6; pp. 431 - 439
• Main Authors: Cheng, Fang, Yin, Xin, Duan, Wenhou, Ye, Runze, Zhu, Yakun, Jia, Chongqi
• Format: Journal Article
• Language: English
• Published: Melbourne Wiley Publishing Asia Pty Ltd 01.06.2019; John Wiley & Sons, Inc
• Subjects: Biomarkers - blood; Blood Glucose - analysis; China - epidemiology; Cross-Sectional Studies; Diabetes; Diabetes Mellitus - blood; Diabetes Mellitus - epidemiology; Diabetes Mellitus - physiopathology; fasting plasma glucose; Female; Follow-Up Studies; Glucose; Glycated Hemoglobin A - analysis; HbA1c; Hemoglobin; Humans; Longitudinal Studies; Male; Middle Aged; Prevalence; Prognosis; restrictive cubic spline regression; Retirement; Risk Factors; Sex Factors; Uric acid; Uric Acid - blood; 尿酸; 空腹血糖; 糖化血红蛋白; 限制性立方样条回归; China; 尿酸; 限制性立方样条回归; 空腹血糖; restrictive cubic spline regression; fasting plasma glucose; HbA1c; uric acid; 糖化血红蛋白
• ISSN: 1753-0393; 1753-0407; 1753-0407
• DOI: 10.1111/1753-0407.12863

Background Several studies have revealed conflicting results regarding the relationship between serum uric acid (SUA) and blood glucose concentrations. The aim of this study was to identify the shape of the curve between SUA and either fasting plasma glucose (FPG) or HbA1c in diabetic and non‐diabetic participants. Methods This cross‐sectional study of the relationship between SUA and diabetes and non‐diabetes was conducted in 10 266 participants. Data from the national baseline survey of the China Health and Retirement Longitudinal Study (CHARLS; 2011‐12) were used. Restrictive cubic spline regression and two piecewise linear regression models were used to detect non‐linear trends and calculate the slope of each segment. Results After adjusting for all confounding factors in all subjects, an L‐shaped relationship was found between FPG and SUA in males (β2 = 3.47, P < 0.01) and females (β2 = 5.84, P < 0.01), as well as between HbA1c and SUA in males (β2 = 0.10, P = 0.03) and females (β2 = 0.12, P = 0.02). In non‐diabetic individuals, there was a positive linear correlation between FPG and SUA in females (β = 0.62, P = 0.02). After a diagnosis of diabetes, a negative linear relationship was found between SUA and blood glucose in males, and a nearly L‐shaped curve was found in females. Conclusions An L‐shaped association was found between SUA and blood glucose in both genders. Stratified analysis revealed an L‐shaped curve only in diabetic females. Further studies are warranted to investigate the mechanisms of the association between uric acid and blood glucose levels. 摘要 背景 目前一些文献研究血清尿酸(SUA)与血糖浓度之间关系显示结果不一致。本研究的目的是确定糖尿病和非糖尿病患者的SUA与空腹血糖(FPG)或糖化血红蛋白(HbA1c)之间的曲线关系类型。 方法 本横断面研究探讨了血清尿酸浓度与糖尿病和非糖尿病患者之间的关系，共纳入10266名受试者。数据来自中国健康与养老追踪基线调查(China Health and Retirement Longitudinal Study，CHARLS；2011‐12)的结果。采用了限制性立方样条回归检测非线性趋势，采用分段线性回归模型进行线性拟合。 结果 控制所有可能的混杂因素后，我们在总人群中发现L型曲线关系不仅存在于FPG与SUA之间:男性(β2 = 3.47，P < 0.01)和女性(β2 = 5.84，P < 0.01)，也存在于HbA1c与SUA之间:男性(β2 = 0.10，P = 0.03)和女性(β2 = 0.12，P = 0.02)。分层结果显示:非糖尿病个体中，仅女性FPG与SUA呈线性正相关(β= 0.62，P = 0.02)；在糖尿病个体中，男性中SUA与血糖呈负线性关系，而女性中呈近似L型曲线。 结论 男性和女性中的SUA与血糖关系均呈L型分布。分层分析后显示仅在糖尿病女性中L型曲线仍存在。在后续的研究中我们需要进一步探讨尿酸与血糖水平之间的机制。 Highlights To investigate the relationship between serum uric acid (SUA) and blood glucose, a national cross‐sectional study of SUA was conducted in those with and without diabetes. Stratified analysis showed an L‐shaped curve between SUA and blood glucose only in diabetic females, regardless of whether fasting plasma glucose or HbA1c were used to measure blood glucose.