Table 4.

Effects of predictor variables on mean GFR in multivariable analysisa

% Change in Measured iGFR (95% CI)% Change in eGFRMDRD% Change in eGFRCG
CKD group (n = 828)
    serum creatinine
        10% decrease from 1.5 mg/dl14.3 (13.5 to 15.1)12.911.1
        10% decrease from 5.5 mg/dl8.6 (7.7 to 9.6)12.911.2
    race (black)6.6 (0.3 to 13.3)211.4
    gender (female)−22.9 (−26.7 to −19.0)−25.8−10.2
age (10% increase)−2.3 (−3.0 to −1.7)−1.9−5.0
    weight (10% increase)0.6 (−0.4 to 1.5)05.6
Kidney donor group (n = 459)
    serum creatinineb 10% decrease from 0.8 mg/dl3.2 (4.2 to 2.2)12.911.4
    race (black)−1.7 (−5.9 to 2.7)210.4
    gender (female)b−4.8 (−8.7 to −0.6)−25.8−11.5
    age (10% increase)−1.8 (−2.3 to −1.2)−1.9−3.6
    weight (10% increase)−0.1 (−0.9 to 0.7)05.0
  • a Shown are the results of multiple regression analyses relating iGFR, eGFRMDRD, and eGFRCG to serum creatinine, race, gender, age, and weight conducted on the log scale. Separate regression models were estimated for each of the three dependent variables (iGFR, eGFRMDRD, and eGFRCG) and for the CKD and kidney donor groups. Regression coefficients are expressed as the percentage change in iGFR, eGFRMDRD, and eGFRCG, which are associated with black race (compared with nonblack), female gender (compared with male), a 10% decrease in serum creatinine, and 10% increases in age, and weight. CI, confidence interval.

  • b Significant difference between the regression coefficient of iGFR between the kidney donors and CKD patients (P < 0.05).