Tamara van Donge

Simplification of methadone dosing in preterm neonates 83 5 on V d in a linear relationship improved the model fit, described by Equation 4, where Pi represents the parameter in the i th individual, ϴ 1 is the population parameter estimate (typical value for a patient) and ϴ 2 is the exponent (fixed to 1 for linear model). ! = " ∗ & #$%!&' ) ( ! Equation 4 Clearance for both (R)- and (S)-methadone increased with advancing GA and was observed to be higher for (R)-methadone compared to (S)-methadone (0.244 versus 0.167 L/h) (Figure 2). Important to mention is that bodyweight and gestational age are highly correlated in this investigated population (r=0.82). Since the range of PNA was limited (15 days) it was not possible to identify an effect of PNA on estimated parameters. No further covariates were identified or included in the final model. Gestational age (weeks) R−Methadone clearance (L/hr) 0.0 0.2 0.4 0.6 0.8 1.0 1.2 26 28 30 32 34 36 Gestational age (weeks) S−Methadone clearance (L/hr) 0.0 0.2 0.4 0.6 0.8 1.0 1.2 26 28 30 32 34 36 Figure 2: (R)/(S)-methadone clearance values for different weeks of gestational age. (R)-methadone and (S)-methadone values are presented by blue and green circles, respectively. Medians for (R)- methadone and (S)-methadone clearance are presented by solid lines. Prediction intervals (95%) are obtained by model-based simulations based on all included subjects and presented for (R)- and (S)-methadone. Final population pharmacokinetic model A 1-compartment model with first order absorption and elimination kinetics best described the data, including GA on CL by a power function and on V d in a linear relationship. Final parameter estimates of developed models for (R)-methadone and (S)-methadone are shown in Table 1 Supporting Information. The population parameter values for apparent V d and apparent CL for (R)-methadone and (S)-methadone corresponded to 26.9 L and 0.244 L/h and 18 L and 0.167 L/h, respectively.