This study is a retrospective cross-sectional study on Chinese adults (age≧18 years) at the Health Management Center of The First Affiliated Hospital of Nanjing Medical University from January 2018 to January 2020. This study was approved by the ethics committee of the First Affiliated Hospital of Nanjing Medical University and approved with a waiver of informed consent as the research involves no more than minimal tangible or intangible risk to the subjects (No. 2018-SR-181).

The inclusion criteria were: (1) BMI:18.5-30 kg/m2, (2) No history of hypertension or blood pressure (BP) < 140/90 mmHg, (3) No history of diabetes or fasting blood glucose (FBG) < 7.0 mmol/L, (4) No history of kidney disease, absence of proteinuria and hematuria, SCr ≤ 133.0μmol/L, (5) No history of the cardio-cerebrovascular disease, malignant tumor, thyroid disorders, and systemic disease, and (6) Serum uric acid (UA), alanine aminotransferase (ALT), low-density lipoprotein cholesterol (LDL-C), high-density lipoprotein cholesterol (HDL-C), and total cholesterol (TC) were in reference range according to guidelines for the Chinese. The exclusion criteria were absence of SCr value and being pregnant.

Basic demographic information (gender, age, and past medical history) was collected using a physician-administered questionnaire by nurses and staff physicians. After the subject sat quietly for 5 min at least, nurses used a sphygmomanometer to measure the blood pressure of the left brachial artery one times. Body mass index (BMI) was calculated as measured weight (kg) divided by measured height (m) squared.

Blood samples were collected in the morning after an overnight fast of at least 8 h. All the fasting blood samples were assayed on Beckman AU5800 automatic bio-chemical analyzer (Beckman Co., Ltd., USA), in strict accordance with the instructions of the apparatus. SCr was measured using the enzymatic method (sarcosine oxidase-PAP, Shanghai Kehua Bio-engineering Co., Ltd., China) with a reference range of 0.5–1.5 mg/dL, which is traceable to National Institute Standardized Technology creatinine standard reference material 909b. The intra- and inter-assay coefficients of variations for creatinine assay were consistently ≦6%, with stable quality control. All enzyme activities were measured at 37 °C.

eGFR was evaluated by the following equations: (1) eGFRCKD-EPI: eGFR = 141 × (SCr/0.9)−0.411 × 0.993age (males and SCr ≤ 0.9 mg/dL), eGFR = 141 × (SCr/0.9)−1.209 × 0.993age (males and SCr > 0.9 mg/dL), eGFR = 144 × (SCr/0.7)−0.329 × 0.993age (females and SCr ≤ 0.7 mg/dL), eGFR = 144 × (SCr/0.7)−1.209 × 0.993age(females and SCr > 0.7 mg/dL). (2) eGFRFAS: eGFR = 107.3/(SCr/QSCr) (for 2 ≤ age ≤ 40 years), eGFR = 107.3/(SCr/QSCr) × 0.988(age−40) (for age > 40 years) (female: QSCr = 0.70 mg/dl; male: QSCr = 0.90 mg/dl). (3) eGFRXiangya: eGFR = 2374.78 × SCr−0.54753 × age−0.25011 × (0.8526126 if female). SCr was written in mg/dL when calculating eGFR using CKD-EPI and FAS formulas but was converted to µmol/l using Xiangya equation.

All participants were stratified into six age groups: 18–29, 30–39, 40–49, 50–59, 60–69, and ≥ 70 years old. Continuous variables were represented as mean (standard deviation (SD)), and categorical variables were represented as proportion. T-test and Welch’s ANOVA test were adopted to compare the differences of continuous variables. General linear regression analysis was applied to calculate the annual eGFR decline rate. We plotted the scatter diagrams to directly reflect the different values in subgroups according to SCr.

Agreement between these equations was analyzed by Bland–Altman diagrams and intraclass correlation coefficient (ICC). Bland–Altman plots display for individual the difference between two of the three equations against their mean. The mean difference was used as bias to compare the equations with each other. 95% Limits of Agreement (LoA) between the equations were defined as mean difference ± 1.96 SD of the differences and were used as a measure of the variability of the bias. These values represented the range within which 95% of the differences were included. ICC values of < 0.5, 0.5–0.75, 0.75–0.9, and > 0.90 indicate poor, moderate, good, and excellent agreement, respectively. Considering the slight differences in the equations between males and females, we further conducted the above statistical analysis in subgroups according to gender. Finally, we calculated the reference intervals using mean ± 1.96 SD in age-gender subgroups. Two-tailed P < 0.05 was considered statistically significant.

Data were missing for BMI (5846, 5.62%), blood pressure (5581, 5.37%), FBG (2935, 2.82%), and UA (172, 0.17%). Missing values were imputed by EM algorithm. Statistical analyses were performed using the SPSS software version 25.0 (IBM Inc., USA) and MedCalc for Windows (version 19.6.1.0; MedCalc Software, Mariekerke, Belgium). The graphic drawing was implemented in Graphpad Prism 8.0.2 (GraphPad Software Inc., San Diego, CA, USA).