Chapter 5 Table 5 Percentage of unreliable NVC parameter estimates. N=71, * p<0.01 Before After Before After correction correction correction correction (mean CBFV) (mean CBFV) (systolic CBFV) (systolic CBFV) ! rad/s 17 % 3 % * 13 % 6 % 13 % 3 % * 8 % 4 % Tv s 34 % 15 % * 31 % 23 % K 15 % 10 % 8 % 7 % Discussion The averaged NVC response can be described more precisely when including cerebral autoregulation in a model of neurovascular coupling. The fit errors for the RGCA model with blood pressure correction are significantly lower compared to the RG model. This holds for both averaged responses of mean and systolic CBFV signals, whereas fit error reduction is larger for the mean values. Before blood pressure correction no difference exists between fit errors for mean and systolic values, but after correction fit errors for mean CBFV values are signifi- cantly lower compared to systolic fit errors. Blood pressure correction is more effective for averaged NVC responses based on mean CBFV values. This might be explained by the fact that the mechanism of cerebral autoregulation is basi- cally controlling mean cerebral blood flow, which is more comparable with mean CBFV than with systolic CBFV values. Rosengartenet al compared systolic and diastolicCBFV responses 22 and concluded that both could be used to determine parameters of the control system model, although the values differed for the two conditions. They preferred using the systolic response, because this would be more robust and less susceptible to artifacts. They did not report results using mean CBFV responses. Comparison of NVC parameters for the whole group did not show significant differences for the parameters except for a small difference in for mean CBFV values. This suggests that although fit errors are significantly lower after blood pressure correction, NVC parameter values are not affected. However, parame- ters can be both increased or decreased after blood pressure correction which could result in no average change, whereas individual parameters change signifi- cantly. This was confirmed by the absolute differences between corrected and uncorrected parameters as shown in figure 5. To investigate if individual pa- rameters changed significantly their changes were evaluated based on non- 92
proefschrift gommer inhoud met kaft.indd
To see the actual publication please follow the link above