Dyslipidemia associates with and usually precedes the onset of chronic kidney disease (CKD), but a comprehensive assessment of molecular lipid species associated with risk of CKD is lacking. Here, we sought to identify fasting plasma lipids associated with risk of CKD among American Indians in the Strong Heart Family Study, a large-scale community-dwelling of individuals, followed by replication in Mexican Americans from the San Antonio Family Heart Study and Caucasians from the Australian Diabetes, Obesity and Lifestyle Study. We also performed repeated measurement analysis to examine the temporal relationship between the change in the lipidome and change in kidney function between baseline and follow-up of about five years apart. Network analysis was conducted to identify differential lipid classes associated with risk of CKD. In the discovery cohort, we found that higher baseline level of multiple lipid species, including glycerophospholipids, glycerolipids and sphingolipids, was significantly associated with increased risk of CKD, independent of age, sex, body mass index, diabetes and hypertension. Many lipid species were replicated in at least one external cohort at the individual lipid species and/or the class level. Longitudinal change in the plasma lipidome was significantly associated with change in the estimated glomerular filtration rate after adjusting for covariates, baseline lipids and the baseline rate. Network analysis identified distinct lipidomic signatures differentiating high from low-risk groups. Thus, our results demonstrated that disturbed lipid metabolism precedes the onset of CKD. These findings shed light on the mechanisms linking dyslipidemia to CKD and provide potential novel biomarkers for identifying individuals with early impaired kidney function at preclinical stages.

血脂异常与慢性肾病 (CKD) 相关，并且通常先于慢性肾病 (CKD) 发病，但缺乏对与 CKD 风险相关的分子脂质种类的全面评估。在这里，我们试图在强心脏家庭研究（一项大规模的社区居民个人研究）中确定与美洲印第安人 CKD 风险相关的空腹血脂，随后在圣安东尼奥家庭心脏研究的墨西哥裔美国人和来自白种人的白种人中进行复制。澳大利亚糖尿病、肥胖和生活方式研究。我们还进行了重复测量分析，以检查基线和相隔约五年的随访之间脂质组变化和肾功能变化之间的时间关系。进行网络分析以确定与 CKD 风险相关的不同脂质类别。在发现队列中，我们发现多种脂质种类（包括甘油磷脂、甘油脂和鞘脂）的较高基线水平与 CKD 风险增加显着相关，与年龄、性别、体重指数、糖尿病和高血压无关。许多脂质种类在至少一个外部队列中在单个脂质种类和/或类别水平上被复制。调整协变量、基线脂质和基线率后，血浆脂质组的纵向变化与估计肾小球滤过率的变化显着相关。网络分析确定了区分高风险人群和低风险人群的不同脂质组学特征。因此，我们的结果表明，脂质代谢紊乱先于 CKD 发病。 这些发现揭示了血脂异常与 CKD 之间的联系机制，并为在临床前阶段识别早期肾功能受损的个体提供了潜在的新型生物标志物。