Skip to main content

Table 3 PK parameter estimates for the structural base model, the categorical model which includes the effect of citrate (RCA) versus non-citrate (non-RCA) systems, and presence of vasopressor; and sample models including continuous dialysis covariates

From: Exploring population pharmacokinetic models in patients treated with vancomycin during continuous venovenous haemodiafiltration (CVVHDF)

 

Parameter

Mean

SD

CV%

Median

Base CL model

Base model

CL (L h−1)

2.59

0.49

18.99

2.70

V (L)

80.98

16.89

20.86

73.72

Model based on categorical (RCA or non-RCA) covariate (CL1 reflects intervals on RCA)

Category applied to CL

CL1 (L h−1)

2.73

0.25

9.14

2.58

CL2 (L h−1)

2.54

0.57

22.53

2.50

V (L)

81.13

15.83

19.51

73.38

Model based on categorical (vasopressor presence) covariate (CL1 reflects intervals with vasopressor use)

Category applied to CL

CL1 (L h−1)

CL2 (L h−1)

2.73

0.72

26.48

2.77

 

2.53

0.33

12.97

2.53

V (L)

77.94

12.44

15.96

75.25

Sample models including continuous dialysis covariates

Qeff applied to CL, linear model

CL1 (L h−1)

CL2 (L h−1)

1.606

0.89

1.1

0.74

68.33

83.1

1.42

0.82

V (L)

79.69

12.13

15.22

76.17

Flux applied to CL, exponential model

CL1 (L h−1)

CL2 (L h−1)

1.88

0.36

0.83

0.33

44.43

92.22

2.12

0.23

V (L)

79.62

12.31

15.47

73.99

  1. Although the model based on continuous dialysis covariates resulted in significant differences in −2LL, there was little improvement in other model comparison metrics (see Additional file 1: Table S2), a higher CV% than base model parameter estimates and a high (> 0.9) correlation between CL1 and CL2 estimated values; prompting caution in their interpretation. CL1 and CL2 represent RCA and non-RCA, or vasopressor and non-vasopressor use, respectively, in the categorical covariates added to the base model. For continuous covariate models, CL1 and CL2 represent the parameters as presented in the sample equations in Additional file 1: Table S1; For the linear model CL1 is the intercept and CL2 is the coefficient associated with change in covariate relative to median value, for the exponential model CL1 is the coefficient and CL2 is the constant part of the exponent which varies with the change in covariate relative to the median value
  2. CL, Clearance; CV, coefficient of variability (CV = SD/mean); Qeff, effluent flow rate; SD, standard deviation; V, volume of distribution