Table 5.

Multivariable analysis for severe AKI within 7 days

VariableUnadjustedAdjusted
Clinical Model 1Model 1Clinical Model 2Model 2
OR (95% CI)P ValueOR (95% CI)P ValueOR (95% CI)P ValueOR (95% CI)P ValueOR (95% CI)P Value
Early Cr elevation18.84 (5.71 to 62.15)<0.00110.96 (3.15 to 38.06)<0.00112.86 (3.52 to 46.97)<0.001
Nonrenal SOFA1.18 (1.07 to 0.32)0.0021.11 (0.99 to 1.25)0.061.05 (0.94 to 1.18)0.371.09 (0.95 to 1.24)0.211.09 (0.95 to 1.25)0.21
sCr, admission5.65 (3.01 to 10.61)<0.0014.70 (2.33 to 9.50)<0.0013.12 (1.50 to 6.50)0.0024.84 (2.53 to 9.25)<0.0012.98 (1.50 to 5.91)0.002
No. of AKI risk factors1.28 (1.08 to 1.51)0.0041.04 (0.85 to 1.26)0.730.98 (0.80 to 1.20)0.87
Hazard tier1.74 (1.16 to 2.63)0.0081.20 (0.73 to 1.99)0.460.77 (0.44 to 1.32)0.34
  • Exploratory univariate analysis for several variables was first performed to identify possible confounders associated with severe AKI for potential inclusion in the multivariable analysis. Because all factors significant in univariate analysis were part of the definition for the hazard tranches, the final models included early Cr elevation, nonrenal SOFA score (per 1-point increase), admission sCr (transformed to the natural log), and either the number of AKI risk factors or hazard tier. A multivariable logistic regression analysis was then performed, and variables were forced into the model (enter mode). Sensitivity analyses using nonrenal APACHE II or SAPS II in place of nonrenal SOFA yielded similar results. OR, odds ratio; sCr, serum creatinine.