Skip to main content

Table 4 Detailed performance of the five GFR-Estimating Equations

From: Applicability of Creatinine-based equations for estimating glomerular filtration rate in elderly Chinese patients

 

Bias

Precision

Accuracy

MAE

GFR category

Misclassification(%)

Median difference

IQR

P30(%)

RMSE

All Participants(N = 612)

 CKD-EPI

0.795a

16.2

64.9

14.5

10.7

37.4

 LMR

−3.015a

13.4

69.9*

13.3

9.79

37.4

 BIS1

−0.050

12.7

73.9**

12.9

9.26

35.3

 FAS

−1.765a

13.1

72.4**

12.1

9.48

36.3

 EKFC

−1.970a

13.5

69.0*

13.3

9.71

36.9

Participants Aged 65–79 Years(N = 444)

 CKD-EPI

0.290

15.3

65.1

14.2

10.40

38.1

 LMR

−3.140a

12.7

69.1*

13.5

9.80

38.3

 BIS1

0.150

12.7

73.0**

13.1

9.33

36.5

 FAS

−1.670a

12.7

72.1**

13.2

9.46

37.2

 EKFC

−2.180a

13.0

68.9*

13.5

9.68

37.4

Participants Aged ≥80 Years(N = 168)

 CKD-EPI

3.135a

18.0

64.3

15.1

11.30

35.7

 LMR

−2.615a

14.2

72.0*

12.9

9.76

35.1

 BIS1

−1.095

13.8

76.2*

12.1

9.11

32.1

 FAS

−1.810

13.5

73.2*

12.7

9.54

33.9

 EKFC

−1.800a

14.8

69.0

13.0

9.82

35.7

Male(N = 386)

 CKD-EPI

0.665

15.7

68.4

14.2

10.34

36.0

 LMR

−3.890a

14.1

73.3*

13.5

9.91

39.4

 BIS1

−0.770

14.0

75.1*

12.8

9.37

35.2

 FAS

−1.810a

13.5

75.1

12.9

9.42

36.5

 EKFC

−2.145a

13.6

71.5*

13.4

9.73

38.3

Female(N = 226)

 CKD-EPI

1.260a

18.0

58.8

15.0

10.37

39.8

 LMR

−1.950

13.7

64.2

13.1

9.93

34.1

 BIS1

0.370a

13.1

71.7**

12.9

9.39

35.4

 FAS

−1.590

12.8

67.7*

13.4

9.45

35.8

 EKFC

−1.815a

14.0

64.6*

13.3

9.76

34.5

mGFR≥60 ml/min/1.73m2(N = 158)

 CKD-EPI

1.590

23.8

80.4

17.6

13.96

39.9

 LMR

−7.860a

20.5

81.6

18.9

14.37

44.9

 BIS1

−8.285a

19.9

79.7

18.8

14.57

47.5

 FAS

−6.480a

21.4

77.2

18.9

14.60

46.8

 EKFC

−5.760a

21.7

81.6

18.3

13.81

44.9

mGFR<60 ml/min/1.73m2(N = 454)

 CKD-EPI

0.720a

15.1

59.5

13.2

9.50

36.6

 LMR

−2.065a

12.3

65.9*

10.8

8.20

34.8

 BIS1

1.220a

11.9

71.8**

10.0

7.42

31.1

 FAS

−0.640

11.7

70.7**

10.3

7.70

32.6

EKFC

−1.120a

13.1

64.5**

11.1

8.29

34.1

Diabetic (N = 293)

 CKD-EPI

0.730

16.3

65.5

13.6

10.15

36.9

 LMR

−3.150a

12.9

69.6

12.7

9.36

37.9

 BIS1

−0.370

12.8

74.1*

11.9

8.71

35.8

 FAS

−1.820a

12.3

71.7*

12.0

8.83

36.2

 EKFC

−1.990a

12.5

70.0*

12.5

9.17

36.5

Non-diabetic(N = 319)

 CKD-EPI

0.960a

16.7

64.3

15.2

11.1

37.9

 LMR

−2.830a

14.0

70.2*

14.0

10.2

37.0

 BIS1

0.110

13.0

73.7**

13.7

9.8

34.8

 FAS

−1.760

13.9

73.0**

14.0

10.1

36.4

 EKFC

−1.950a

14.7

68.0*

14.0

10.2

37.3

  1. Median difference in bias is the difference between equation biases (estimated GFR minus measured GFR); P30 Percentage of estimates within 30% of the measured value, IQR The inter-quartile range of difference, RMSE Root mean square error, MAE Mean absolute error, CKD-EPI Chronic kidney disease epidemiology collaboration, LMR Lund-Malmö Revised, BIS1 Berlin Initiative Study 1, FAS Full age spectrum equation, EKFC European Kidney Function Consortium
  2. a P < 0.05, compared with mGFR; * P < 0.05, **P < 0.001,compared with CKD-EPI