Applicability of Different Estimated Glomerular Filtration Rate Equations in Chinese Healthy Population: A Cross-Sectional Study

Objective: To estimate kidney function with aging by equations based on serum creatinine (Scr) and investigate the applicability of these equations in estimated glomerular ltration rate (eGFR) in healthy Chinese population. Patients and methods: A total of 34,416 healthy individuals were enrolled in this study, eGFR was calculated by CKD-EPI, FAS and Xiangya equations. Plotted scatter diagrams were adopted to show the difference between eGFR by above three equations. Agreement analysis between these equations was quantied by the linear weighted κ statistic. Results: eGFRs by CKD-EPI and Xiangya equation had a constant decline with aging. eGFR by FAS equation was approximately 116.5 mL/min/1.73m 2 under 40 years, then decreased with a rate of 1.27 mL/min/1.73 m 2 /year. The difference was relatively small, and substantial agreement was observed between CKD-EPI and FAS equations with Scr levels above 0.9 mg/dL in males (κ =0.798) or 0.7 mg/dL in females (κ =0.745). The agreement between CKD-EPI and FAS equations was moderate to near-perfect in non-elderly (κ 0.617-0.901 for males and 0.536-0.861 for females), whereas was fair to moderate in older adults (κ 0.494-0.513 for males and 0.373-0.526 for females). eGFR by CKD-EPI was slightly higher than which by FAS equation in older population. The agreement between Xiangya and CKD-EPI or FAS equation was all slight (κ (cid:0) 0.2). Conclusion: FAS equation could be used to evaluate declining rate of eGFR in healthy population. We recommend FAS or CKD-EPI equation in non-elderly adults, and FAS equation in older adults to calculate eGFR. More verication is needed, especially in older adults and population with low Scr level. glomeruli with aging, which results in the slower declining rate of eGFR in males.


Page 3/11
total cholesterol (TC) were in normal range according to guideline based Chinese. Exclusion criterion included: (1) Lack of Scr value; (2) Pregnant women.

Laboratory Measures
Basic demographic information was collected using a physician-administered questionnaire. The questionnaire was evaluated for incomplete or inconsistent answers, by nurses and our staff physicians. Body mass index (BMI) was calculated as measured weight (kg) divided by height (m) squared.
All the fasting blood samples were assayed on AU5800 automatic analyzer (Beckman Coulter, Inc., USA), in strict accordance with the instructions of the apparatus. Scr was determined using the enzymatic method (Shanghai Kehua Dongling Diagnostic Products Co., Ltd, China) with a reference range 44-133umol/L. eGFR was calculated by CKD-EPI, FAS and Xiangya equations, which were shown in Table S1.

Statistical Analyses
All participants were divided into six subgroups according to age: 18-29 years, 30-39 years, 40-49 years, 50-59 years, 60-69 years and ≥ 70 years. Continuous variables were represented as mean ± standard deviation (SD), categorical variables were represented as N/%. Welch's ANOVA test was used to compare differences of variables between six subgroups, differences between males and females were tested with the independent-samples t test, p < 0.05 were considered statistically signi cant. We used partial correlation analysis to evaluate the relationship between the relevant factors and the differences of the three equations, and plotted scatter diagrams to directly re ect the relationship under different Scr and age levels. Agreement analysis between different equations was quanti ed using the linear weighted κ statistic. The criteria were as follows: 0 ≤ κ value < 0. 21

Characteristics of subjects
A total of 34,416 healthy people were enrolled in this study. The mean age of total population was 39.0 years and 45.23% of which were males (N = 15568). The mean Scr was 0.75 mg/dL, the level of it showed stable before 40 years, then increased with aging. Other observation indices were shown in Table 1. eGFR estimated glomerular ltration rate by CKD-EPI, FAS and Xiangya respectively.

Differences between equations strati ed by Scr level
Partial correlation analysis of the difference between three equations and important related factors showed Δ (CKD-EPI, FAS), Δ (CKD-EPI, Xiangya) and Δ (FAS, Xiangya) were all signi cantly correlated with Scr and age. We plotted scatter diagrams to show the difference between these equations under different Scr level (Fig. 3). Δ (CKD-EPI, FAS) reduced with increasing Scr level when Scr was less than 0.9 mg/dL in males or 0.7 mg/dL in females, while was relatively stable when Scr levels more than above cut-off point. Δ (CKD-EPI, Xiangya) and Δ (FAS, Xiangya) were considerable with elevating Scr level. We performed agreement analysis by adopting 0.9 mg/dL in males and 0.7 mg/dL in females as the cut-off point, a substantial agreement between CKD-EPI and FAS equations was seen with Scr ≥ 0.9mg/dL in males (κ 0.798) and Scr ≥ 0.7mg/dL in females (κ 0.745). Whereas agreement between Xiangya and other two equations were all slight (κ<0.2). Agreement between equations was poor for subjects with Scr < 0.9mg/dL in males and Scr < 0.7mg/dL in females (Table S2 and Table S3).
The black, orange and blue circle represents the total, females and males, respectively. The solid line in the scatter plot represents the trend line.
Differences between equations strati ed by age As shown in Fig. 4, Δ (CKD-EPI, FAS), Δ (CKD-EPI, Xiangya) and Δ (FAS, Xiangya) existed differences across age groups. Δ (CKD-EPI, FAS) was relatively stable with aging. Δ (CKD-EPI, Xiangya) and Δ (FAS, Xiangya) have marked difference in females and males. In addition, we divided into six age groups, and used weighted κ statistic to quantify agreement between equations. We observed moderate to near-perfect in age 18-59 years (κ 0.617-0.901 for males and 0.536-0.861 for females) between CKD-EPI and FAS equations, whereas fair to moderate in age ≥ 60 years (κ 0.494-0.513 for males and 0.373-0.526 for females). However, the agreement in every age group was no more than moderate between Xiangya and CKD-EPI equations or FAS equations. In addition, CKD-EPI equation had higher estimates than FAS and Xiangya equations in GFR categories (Table S4 and Table S5).
The black, orange and blue circle represents the total, females and males, respectively. The solid line in the scatter plot represents the trend line.

Discussion
In this study, we investigated the variety tendency of eGFR with aging using CKD-EPI, FAS and Xiangya equations and analyze the difference between these equations in healthy population.  [17]. On the contrary, another study that used clearances of inulin described a slower fall in GFR with age in females [18]. Some previous studies have revealed the protective role of osetragen in kidney function [19], which contribute to the higher eGFR level in females under 50 years. In the term of declining rate of eGFR, one possible explanation is that kidney function is preserved as healthy subjects age at the expense of a gradual decrease in the kidney functional reserve [20]. Men may have a higher reserved kidney function to compensate for the accumulation of sclerotic glomeruli with aging, which results in the slower declining rate of eGFR in males.
In our scatter plot related to creatinine, we observed the considerable differences between three equations with Scr levels less than 0.7 mg/dL in males and 0.9 mg/dL in females, and agreement between equations was poor.  [21]. We also tested the consistency between EKFC and other equations, and observed that the consistency between EKFC and CKD-EPI equations was better than that between CKD-EPI and FAS equations. However, the accuracy in evaluating eGFR of EKFC was unde ned and need more veri cation. Thus, further research is needed to evaluate the eGFR estimating equations in subjects with low Scr value. When Scr levels more than above cut-off point, Δ (CKD-EPI, FAS) was relatively small and stable. We quanti ed agreement among the three equations, the result showed greater agreement between CKD-EPI and FAS equation with Scr ≥ 0.7 in females and Scr ≥ 0.9 mg/dL in males. Therefore, according to creatinine level, we tend to suggest the application of the FAS or CKD-EPI equation to evaluate eGFR in healthy females with Scr ≥ 0.7 and healthy males with Scr ≥ 0.9 mg/dL.
In our scatter plot related to age, Δ(CKD-EPI, FAS) was relatively stable with aging. Agreement analysis showed moderate to near-perfect agreement between CKD-EPI and FAS equations in non-elderly subjects, but worse agreement between these equations in older adults. The agreement of CKD-EPI and FAS equation was slightly better in males compared with this in females. In older population, most of the eGFR values calculated by CKD-EPI equation were higher than that by FAS equation. A Meta-Analysis by Eriksen et al. [22] distributed a slightly higher eGFR by CKD-EPI equation than mGFR both males and females. We also performed agreement analysis among subgroups according to age and Scr level, which is consistent with above results. The agreement was slight between Xiangya equation and CKD-EPI or FAS equation. Therefore, the clinicial Figures Figure 1 The screening owchart for the eligible participates.

Supplementary Files
This is a list of supplementary les associated with this preprint. Click to download. Tablesupplement13.docx TableS4.xlsx TableS5.xlsx