Table 2.

Bias, precision, and accuracya

mGFRbnBiasPrecisionAccuracy
Total/RelativeTotal/RelativeWithin ±15%Within ±30%Within ±50%
CGcMDRDdCGMDRDCGMDRDCGMDRDCGMDRD
≥90807−14.6f/−11.7f−23.4f/−19.5f24.4/20.921.8/17.841.429.980.968.798.498.3
60 to 89.9512−0.1/0−6.1f/−7.8f21.4/28.217.0/22.641.840.875.479.794.197.5
30 to 59.94844.0f/9.8f−0.6/−0.714.6/32.913.0/28.739.537.668.873.488.694.8
15 to 29.92754.9f/23.0f0.4/1.610.5/48.88.3/38.226.932.454.956.078.287.6
<151306.6e/78.0e3.3f/42.0f6.4/93.95.2/68.819.220.831.540.844.665.4
≥601319−9.0f/−7.2f−16.7f/−15.0f24.3/24.721.7/20.641.634.178.872.996.798.0
<608894.7f/23.9f0.3/6.3f12.5/56.010.9/42.532.633.559.163.279.088.3
Overall2208−3.5f/5.3f−9.8f/−6.4f21.5/43.120.0/33.038.033.970.869.088.894.1
  • a Bias defined as total (ml/min/1.73 m2) and relative (%) mean difference between estimated and measured glomerular filtration rate (mGFR). Precision defined as one SD of total and relative bias. Accuracy defined as percentage of GFR estimations within ±15, ±30, and ±50% range of respective GFR measurements.

  • b mGFR, measured GFR by renal inulin clearance in ml/min/1.73 m2.

  • c CG, Cockcroft-Gault formula estimated GFR (ml/min/1.73 m2).

  • d MDRD, Modification of Diet in Renal Disease formula estimated GFR (ml/min/1.73 m2).

  • e P < 0.05.

  • f P < 0.001 (estimated versus mGFR).