Abstract：
BACKGROUND: Long-chain free fatty acids (FFAs) are associated with risk of incident diabetes. However, a comprehensive assessment of the associations in normoglycemic populations is lacking.
OBJECTIVE: Our study aimed to comprehensively investigate the prospective associations and patterns of FFA profiles with diabetes risk among normoglycemic Chinese adults.
METHODS: This is a prospective nested case-control study from the China Cardiometabolic Disease and Cancer Cohort (4C) study. We quantitatively measured 53 serum FFAs using a targeted metabolomics approach in 1707 incident diabetes subjects and 1707 propensity score-matched normoglycemic controls. Conditional logistic regression models were employed to estimate odds ratios (ORs) for associations. Least Absolute Shrinkage and Selection Operator (LASSO) penalty regression and quantile g-computation (qg-comp) analyses were implemented to estimate the association between multi-FFA exposures and incident diabetes.
RESULTS: The majority of odd-chain FFAs exhibited an inverse association with incident diabetes, wherein the ORs per SD increment of all 7 saturated fatty acids (SFAs), monounsaturated fatty acid (MUFA) 15:1, and polyunsaturated fatty acid (PUFA) 25:2 were ranging from 0.79 to 0.88 (95% CIs ranging between 0.71 and 0.97). Even-chain FFAs comprised 99.3% of total FFAs and displayed heterogeneity with incident diabetes. SFAs with 18-26 carbon atoms are inversely linked to incident diabetes, with ORs ranging from 0.81 to 0.86 (95% CIs ranging between 0.73 and 0.94). MUFAs 26:1 (OR: 0.85; 95% CI: 0.76, 0.94), PUFAs 20:4 (OR: 0.84; 95% CI: 0.75, 0.94), and 24:2 (OR: 0.87; 95% CI: 0.78, 0.97) demonstrated significant associations. In multi-FFA exposure model, 24 FFAs were significantly associated with incident diabetes, most of which were consistent with univariate results. The mixture OR was 0.78 (95% CI: 0.61, 0.99; P = 0.04159). Differential correlation network analysis revealed pre-existing perturbations in intraclass and interclass FFA coregulation before diabetes onset.
CONCLUSIONS: These findings underscore the variations in diabetes risk associated with FFAs across chain length and unsaturation degree, highlighting the importance of recognizing FFA subtypes in the pathogenesis of diabetes.
OBJECTIVE: Our study aimed to comprehensively investigate the prospective associations and patterns of FFA profiles with diabetes risk among normoglycemic Chinese adults.
METHODS: This is a prospective nested case-control study from the China Cardiometabolic Disease and Cancer Cohort (4C) study. We quantitatively measured 53 serum FFAs using a targeted metabolomics approach in 1707 incident diabetes subjects and 1707 propensity score-matched normoglycemic controls. Conditional logistic regression models were employed to estimate odds ratios (ORs) for associations. Least Absolute Shrinkage and Selection Operator (LASSO) penalty regression and quantile g-computation (qg-comp) analyses were implemented to estimate the association between multi-FFA exposures and incident diabetes.
RESULTS: The majority of odd-chain FFAs exhibited an inverse association with incident diabetes, wherein the ORs per SD increment of all 7 saturated fatty acids (SFAs), monounsaturated fatty acid (MUFA) 15:1, and polyunsaturated fatty acid (PUFA) 25:2 were ranging from 0.79 to 0.88 (95% CIs ranging between 0.71 and 0.97). Even-chain FFAs comprised 99.3% of total FFAs and displayed heterogeneity with incident diabetes. SFAs with 18-26 carbon atoms are inversely linked to incident diabetes, with ORs ranging from 0.81 to 0.86 (95% CIs ranging between 0.73 and 0.94). MUFAs 26:1 (OR: 0.85; 95% CI: 0.76, 0.94), PUFAs 20:4 (OR: 0.84; 95% CI: 0.75, 0.94), and 24:2 (OR: 0.87; 95% CI: 0.78, 0.97) demonstrated significant associations. In multi-FFA exposure model, 24 FFAs were significantly associated with incident diabetes, most of which were consistent with univariate results. The mixture OR was 0.78 (95% CI: 0.61, 0.99; P = 0.04159). Differential correlation network analysis revealed pre-existing perturbations in intraclass and interclass FFA coregulation before diabetes onset.
CONCLUSIONS: These findings underscore the variations in diabetes risk associated with FFAs across chain length and unsaturation degree, highlighting the importance of recognizing FFA subtypes in the pathogenesis of diabetes.
摘要：
背景：长链游离脂肪酸（FFA）与糖尿病发病风险相关。然而,缺乏对血糖正常人群的关联的全面评估.
目的：我们的研究旨在全面调查血糖正常的中国成年人FFA谱与糖尿病风险的前瞻性关联和模式。
方法：这是一项来自中国心血管代谢疾病和癌症队列（4C）研究的前瞻性巢式病例对照研究。我们使用靶向代谢组学方法在1707例糖尿病患者和1707例倾向评分匹配的正常血糖对照中定量测量了53例血清FFA。使用条件逻辑回归模型来估计关联的比值比（OR）。实施了最小绝对收缩和选择算子（LASSO）惩罚回归和分位数g计算（qg-comp）分析，以估计多FFA暴露与糖尿病之间的关联。
结果：大多数奇数链FFA与糖尿病发病呈负相关，其中所有7种饱和脂肪酸(SFA)的OR/SD增量，单不饱和脂肪酸（MUFA）15：1和多不饱和脂肪酸（PUFA）25：2的范围为0.79至0.88（95CIs的范围为0.71至0.97）。偶数链FFA占总FFA的99.3％，并显示出糖尿病的异质性。具有18至26个碳原子的SFA与糖尿病的发病成反比，ORs范围为0.81至0.86（95CIs范围为0.73至0.94）。MUFA26:1(OR[95CI]:0.85[0.76-0.94])，PUFAs20:4(0.84[0.75-0.94])和24:2(0.87[0.78-0.97])显示显著关联。在多FFA暴露模型中，24个FFA与糖尿病发病显著相关，其中大部分与单变量结果一致.混合物的OR为0.78[0.61-0.99](P=0.04159)。差异相关网络分析显示，在糖尿病发作之前，在类内和类间FFA共调中存在预先存在的扰动。
结论：这些发现强调了与FFA相关的糖尿病风险在链长度和不饱和程度上的变化，强调认识FFA亚型在糖尿病发病机制中的重要性。
目的：我们的研究旨在全面调查血糖正常的中国成年人FFA谱与糖尿病风险的前瞻性关联和模式。
方法：这是一项来自中国心血管代谢疾病和癌症队列（4C）研究的前瞻性巢式病例对照研究。我们使用靶向代谢组学方法在1707例糖尿病患者和1707例倾向评分匹配的正常血糖对照中定量测量了53例血清FFA。使用条件逻辑回归模型来估计关联的比值比（OR）。实施了最小绝对收缩和选择算子（LASSO）惩罚回归和分位数g计算（qg-comp）分析，以估计多FFA暴露与糖尿病之间的关联。
结果：大多数奇数链FFA与糖尿病发病呈负相关，其中所有7种饱和脂肪酸(SFA)的OR/SD增量，单不饱和脂肪酸（MUFA）15：1和多不饱和脂肪酸（PUFA）25：2的范围为0.79至0.88（95CIs的范围为0.71至0.97）。偶数链FFA占总FFA的99.3％，并显示出糖尿病的异质性。具有18至26个碳原子的SFA与糖尿病的发病成反比，ORs范围为0.81至0.86（95CIs范围为0.73至0.94）。MUFA26:1(OR[95CI]:0.85[0.76-0.94])，PUFAs20:4(0.84[0.75-0.94])和24:2(0.87[0.78-0.97])显示显著关联。在多FFA暴露模型中，24个FFA与糖尿病发病显著相关，其中大部分与单变量结果一致.混合物的OR为0.78[0.61-0.99](P=0.04159)。差异相关网络分析显示，在糖尿病发作之前，在类内和类间FFA共调中存在预先存在的扰动。
结论：这些发现强调了与FFA相关的糖尿病风险在链长度和不饱和程度上的变化，强调认识FFA亚型在糖尿病发病机制中的重要性。
