Blood pressure corrected evoked flow responses overlapping confidence bounds of parameter estimations. This shows that for mean CBFV parameters changed in 25-41% of the cases and for systolic CBFV responses in 15-31% of cases. Individual parameters therefore do change signifi- cantly and this could be of influence on findings earlier presented on NVC parameters. Evaluating the precision of parameter estimations by the value of CP shows a significant decrease in confidence bound width for all NVC parameters. This means parameters are estimated more precisely leading to less intra-individual variation. This could enhance the power to detect differences between different patient groups, but could also lead to less pronounced effects shown with analy- ses performed on uncorrected responses. But, showing that blood pressure variation via cerebral autoregulation can have a significant improvement on NVC response fits advocates incorporation of cerebral autoregulation in the model used to fit these responses. Recently, several investigators evaluated reproducibility of TCD-based evalua- tion of cerebral hemodynamic function 5, 8, 10, 23, 30. Good reproducibility is of key importance for the clinical application of these tests, especially when used to evaluate therapeutic interventions or monitor disease course. Combining mechanisms such as neurovascular coupling, cerebral autoregulation and CO2 reactivity might help to disentangle the specific contributions of each mechanism and improve the reproducibility of the estimated parameters describing each process. Paneraiet al 17 presented a promising approach using autoregressive models whereas we combined two known models for neurovascular coupling and cerebral autoregulation. Both are attempts for more reliable parameter estimation and reduction of unexplained variance. We also measured end-tidal CO2-data but did not yet use this as an extra model input. A different approach might be to build a more physiological based model that consists of components for different regulatory mechanisms: metabolic, neurogenic and myogenic. Payne 18 described such a model, which is a combination of a hemodynamic model and different feedback mechanisms adjusting arterial compliance and resistance. A similar approach is evaluated by Spronck 28. In this study cerebral autoregulation is modeled on the stimulus averaged re- sponses of blood pressure and cerebral blood flow velocity. However, cerebral autoregulation is basically not a visual stimulus related mechanism but is con- tinuously reacting to blood pressure variability. Therefore it might be considered to evaluate the effect of cerebral autoregulation on the original non averaged raw signals. This needs to be investigated in further research. 93
proefschrift gommer inhoud met kaft.indd
To see the actual publication please follow the link above