Page 75

proefschrift gommer inhoud met kaft.indd

Cerebral autoregulation in AD and MCI parameters with spontaneous blood pressure variations in supine rest. Using the results from this study the group size needed to show significant differences between AD and C could be estimated. A sample size analysis for GainLF shows that two to three times as many subjects are needed to show possible significant differences (p<0.05) between AD and C with 80 % power. The windkessel model parameters that were fitted with the transfer function results show that the differences between AD, MCI and C groups are specific for the peripheral vasculature, since the arterial resistance Ra did not differ between the groups. The peripheral resistance Rp is increased in AD compared to controls. This is in accordance with the increased CVRi we found. Batemanet al 2 also found increased resistance in AD using MRI to quantify cerebral haemodynam- ics. They also found compliance to be significantly lower (about 20 %) in AD compared to controls. We found compliance in AD to be 17 % lower than in controls, which was not significant. Our results therefore do not provide evi- dence for additional benefit of using the windkessel model over simply deter- mining CVRi. CVRi is much easier to determine and could prove its value in the multidimensional diagnosis of AD. With a correlation between CVRi and Rp of 0.97 the parameters show to be almost identical. The autoregulation parameters phase and ARI have positive correlations with blood pressure parameters, whereas autoregulation gain does not correlate with blood pressure. Apart from autoregulation, vascular compliance also elicits flow phase lead with respect to pressure due to fundamental hemodynamic behaviour of elastic tubes. As Zhanget al showed, in the WKM for low values of Cp a posi- tive correlation with phase exists and for high values of Cp it is negatively corre- lated with phase 31. We found no correlation between Cp and blood pressure and therefore the correlation between phase, ARI and blood pressure seems to be determined by direct influence of blood pressure. So, the higher blood pressure the faster autoregulation performs showing cerebral autoregulation seems to adjust adequately to higher levels of blood pressure. CVRi and Rp also positively correlate with blood pressure. For CVRi this can be easily explained by the formula used to calculate it. Rp, which is the only windkessel model parameter that significantly differs between AD and C, is strongly correlated with CVRi and therefore seems to reflect best the differences in CVRi among these groups also indicating that vascular pathology manifests predominantly in the smaller arteri- oles. CVRi and the windkessel parameters Ra and Cp positively correlate specifi- cally with right medial temporal lobe atrophy, whereas CVRi and Ra also corre- late with the MTA average, but less strong. This side specific correlation is difficult to explain, but could indicate lateralisation of pathological cognitive function. 73


proefschrift gommer inhoud met kaft.indd
To see the actual publication please follow the link above