Skip to main content
. 2015 Jun 12;59(7):3935–3943. doi: 10.1128/AAC.00102-15

TABLE 2.

Population PK model developmenta

Model description Population modelb OFVc ΔOFV
Base model and univariable analysisd
    Volume of distribution (liters) V = θV × wt 1,022.74
        ECMO V = θV × wt × 1.41ECMO(= 1 or 0) 1,008.22 −14.52
        Hemofiltration V = θV × wt × 1.54HMFLTR(= 1 or 0) 1,012.13 −10.61
    Clearance (liters/h) CL = θCL × wt 1,022.74
        Creatinine CL = θCL × wt × (creatinine/0.4)−0.29 950.84 −71.91
        Hemofiltration CL = θCL × wt × 0.66HMFLTR 1,019.18 −3.56
Multivariable analysis
    CL = f(SCR), V = f(ECMO)e V = θV × wt × 1.39ECMO 936.95 −13.89f
CL = θCL × wt × (creatinine/0.4)−0.29
    CL = f(SCR), V = f(ECMO,HMFLTR) V = θV × wt × 1.30ECMO × 1.34HMFLTR 930.70 −6.25g
CL = θCL × wt × (creatinine/0.4)−0.29
    CL = f(SCR, HMFLTR), V = f(ECMO, HMFLTR) V = θV × wt × 1.30ECMO × 1.34HMFLTR 930.33 −0.37g
CL = θCL × wt × (creatinine/0.4)−0.29 × 0.90HMFLTR
Final model
    V = f(ECMO) V = θV × wt × 1.39ECMO
    CL = f(SCR) CL = θCL × wt × (creatinine/0.4)−0.29 936.95 −13.89f
a

All coefficients in the models are estimated parameters from the respective model.

b

V, volume of distribution; wt, weight; HMFLTR, hemofiltration; CL, clearance.

c

OFV, objective function value.

d

Only models of covariates that significantly improved the OFV are shown in the univariable analysis. For a full list of covariates tested, see Materials and Methods.

e

SCR, serum creatinine.

f

ΔOFV calculated relative to CL∼creatinine model (950.84).

g

ΔOFV calculated relative to preceding model.