Department of Health Management, Health Management Research Center of Central South University, The Third Xiangya Hospital, Central South University, Changsha, 410013, Hunan Province, China.
Department of Psychiatry & Mental Health Institute, The Second Xiangya Hospital, Central South University, Changsha, Hunan Province, China.
Int Urol Nephrol. 2022 Jul;54(7):1629-1639. doi: 10.1007/s11255-021-03045-9. Epub 2021 Nov 1.
As health screening continues to increase in China, there is an opportunity to integrate a large number of demographic as well as subjective and objective clinical data into risk prediction modeling. The aim of this study was to develop and validate a prediction model for chronic kidney disease (CKD) in Chinese health screening examinees with type 2 diabetes mellitus (T2DM).
We conducted a retrospective cohort study consisting of 2051 Chinese T2DM patients between 35 and 78 years old who were enrolled in the XY3CKD Follow-up Program between 2009 and 2010. All participants were randomly assigned into a derivation set or a validation set at a 2:1 ratio. Cox proportional hazards regression model was selected for the analysis of risk factors for the development of the proposed risk model of CKD. We established a prediction model with a scoring system following the steps proposed by the Framingham Heart Study.
The mean follow-up was 8.52 years, with a total of 315 (23.20%) and 189 (27.27%) incident CKD cases in the derivation set and validation set, respectively. We identified the following risk factors: age, gender, body mass index, duration of type 2 diabetes, variation of fasting blood glucose, stroke, and hypertension. The points were summed to obtain individual scores (from 0 to 15). The areas under the curve of 3-, 5- and 10-year CKD risks were 0.843, 0.799 and 0.780 in the derivation set and 0.871, 0.803 and 0.785 in the validation set, respectively.
The proposed scoring system is a promising tool for further application of assisting Chinese medical staff for early prevention of T2DM complications among health screening examinees.
随着中国健康筛查的不断增加，有机会将大量人口统计学以及主观和客观临床数据纳入风险预测模型。本研究旨在为中国健康筛查的 2 型糖尿病（T2DM）患者建立和验证预测慢性肾脏病（CKD）的模型。
我们进行了一项回顾性队列研究，纳入了 2051 名年龄在 35 至 78 岁之间的中国 T2DM 患者，这些患者于 2009 年至 2010 年期间参加了 XY3CKD 随访项目。所有参与者以 2:1 的比例随机分配到推导集或验证集中。Cox 比例风险回归模型用于分析该拟议的 CKD 风险模型中发生 CKD 的发展的危险因素。我们按照 Framingham 心脏研究提出的步骤建立了一个具有评分系统的预测模型。
平均随访时间为 8.52 年，推导集和验证集中分别有 315（23.20%）和 189（27.27%）例 CKD 事件。我们确定了以下危险因素：年龄、性别、体重指数、2 型糖尿病病程、空腹血糖变化、中风和高血压。将各点相加得到个人分数（0 至 15 分）。推导集和验证集中，3、5 和 10 年 CKD 风险的曲线下面积分别为 0.843、0.799 和 0.780，0.871、0.803 和 0.785。
该评分系统是一种有前途的工具，可进一步用于帮助中国医务人员对健康筛查受检者的 T2DM 并发症进行早期预防。
