Chronic kidney disease (CKD) is a major global public health problem, and its increasing prevalence is closely linked to atherosclerosis. The atherogenic index of plasma (AIP), a simple lipid-based indicator, has demonstrated predictive value for cardiovascular diseases. However, studies examining the association between AIP and the risk of developing CKD remain limited.

To investigate the association between the AIP and new-onset CKD.

This prospective cohort study utilized data from 101 510 employees of the Kailuan Group in Tangshan, Hebei Province, who underwent health examinations between June 2006 and October 2007. After applying inclusion and exclusion criteria, 85 253 participants were enrolled. Participants were stratified into four groups (Q1 to Q4) based on baseline AIP quartiles: Q1 (AIP <-0.58), Q2 (-0.58≤AIP <-0.17), Q3 (-0.17≤AIP <0.29), and Q4 (AIP≥0.29). Follow-up continued until December 31, 2021, with new-onset CKD as the endpoint. The cumulative incidence of CKD was plotted using the Kaplan-Meier method, and intergroup differences were assessed with the log-rank test. The association between AIP and CKD was evaluated using Cox proportional hazards regression models.

A total of 18 175 patients developed CKD during 13.97 (13.53, 14.17) years of follow-up. With increasing AIP, the cumulative incidence of CKD in the Q1-Q4 groups was 16.87%, 21.49%, 22.31%, and 24.47%, with incidence densities of 13.48/1 000 person-years, 17.83/1 000 person-years, 18.56/1 000 person-years, and 20.77/1 000 person-years, respectively. After correcting for relevant confounders, the HR (95%CI) for incident CKD was 1.24 (1.18-1.29), 1.26 (1.21-1.33) and 1.51 (1.43-1.59) in the Q2-Q4 groups compared with the Q1 group (P_trend<0.001). Further analysis showed that when patients with CKD events at the first 2 years of follow-up, patients with all-cause mortality events during follow-up, patients taking antihypertensive, hypoglycaemic, and lipid-lowering medications at baseline, and patients with myocardial infarction or stroke during follow-up were excluded, the risk of Q4 group was similar to the results of the main analysis and the results were more robust when compared with Q1 group. Subgroup analyses showed that significant interactions were observed for age, sex, BMI, history of hypertension, and history of smoking (P_interaction<0.001), and AIP had a more significant risk in the subgroups of age<60 years, sex, BMI≥28 kg/m², and history of smoking (P<0.05).

An elevated AIP is an independent risk factor for new-onset CKD and serves as an early predictor of its development.

慢性肾脏病（CKD）是全球范围内日益严峻的公共卫生问题，其发病率逐年上升，并且与动脉粥样硬化的发生密切相关。血浆致动脉粥样硬化指数（AIP）作为一种基于血脂的简便指标，已被证明能够有效预测心血管事件的风险。然而，关于AIP与CKD发生风险之间关系的研究尚不充分，仍需进一步探索。

探究AIP与新发CKD的相关性。

采用前瞻性队列研究，选取2006年6月—2007年10月河北省唐山市开滦集团职工健康查体人群（n=101 510，其中男81 110例、女20 400例，年龄18~98岁）为研究对象。经过纳入和排除标准筛选后，共85 253例纳入研究，根据基线AIP四分位数将研究对象分为4组，即Q1组（AIP <-0.58）、Q2组（-0.58≤AIP <-0.17）、Q3组（-0.17≤AIP <0.29）、Q4组（AIP≥0.29）。随访至2021-12-31，观察终点事件为新发CKD。采用Kaplan-Meier法绘制累积发病率曲线，并采用Log-rank检验比较组间差异。AIP与CKD的相关性采用Cox比例风险回归模型分析。

随访13.97（13.53，14.17）年，期间共有18 175例患者罹患CKD。随着AIP增加，Q1~Q4组CKD累积发病率分别为16.87%、21.49%、22.31%、24.47%，发病密度分别为13.48/千人年、17.83/千人年、18.56/千人年、20.77/千人年。在校正相关混杂因素后，Cox比例风险回归分析结果显示，与Q1组相比，Q2~Q4组发生CKD的HR（95%CI）分别为1.24（1.18~1.29）、1.26（1.21~1.33）、1.51（1.43~1.59）（P_趋势<0.001）。进一步分析发现，分别排除在第一个2年随访期发生CKD事件的患者，随访期发生全因死亡事件的患者，基线服用降压药、降糖药、降脂药的患者，随访期间发生心肌梗死、脑卒中的患者后，与Q1组相比，Q4组的风险与主分析结果相似，结果较稳健。亚组分析显示，AIP在年龄、性别、BMI、高血压史、吸烟史亚组中存在明显的交互作用（P_交互<0.001），AIP在年龄<60岁、男性、BMI≥28 kg/m²、有吸烟史等亚组中存在更显著的风险（P<0.05）。

高AIP为新发CKD的独立危险因素，能更早的预测新发CKD的发病风险。