Covariate analysis
Age, height, LBW, FFM, BMI and centre of inclusion were explored as
covariates. In univariate analysis - using the structural model without
application of allometric scaling – the separately weight-related
covariates body weight, FFM and LBW were significantly related to CL and
V1. We chose however to allometrically scale these parameters with body
weight as it is easy and routinely measured, as opposed to LBW and FFM.
Incorporation of a separate residual error for one haemophilia treatment
centre improved the model (p<0.05), but was not significant
after the backward selection reprocess (p<0.01). Therefore,
all centres were described by the same residual error model. CL
decreased with age; the latter was the only covariate that improved the
fit of the model to the data. On basis of this relationship typical
clearance of a 73 kg patient would decrease from 1.89 dL/h at an age of
20 years to 1.36 dL/h at an age of 70 years.