LAUSR

The study of Dynamic Cerebral Autoregulation (DCA) in essential hypertension has received considerable attention because of its clinical importance. Several studies have examined the dynamic relationship between spontaneous beat-to-beat arterial blood pressure data and contemporaneous cerebral blood flow velocity measurements (obtained via transcranial Doppler at the middle cerebral arteries) in the form of a linear input-output model using Transfer Function analysis. This analysis is more reliable when the contemporaneous effects of changes in blood CO2 tension are also taken into account, because of the significant effects of CO2 Dynamic Vasomotor Reactivity (DVR) upon cerebral flow. In this paper, we extract such input-output predictive models from spontaneous time-series hemodynamic data of 24 patients with essential hypertension and 20 normotensive control subjects under resting conditions, using the novel methodology of Principal Dynamic Modes (PDMs) that achieves improved estimation accuracy over previous methods for relatively short and noisy data. The obtained data-based models are subsequently used to compute indices/markers that quantify DCA and DVR in each subject/patient and, therefore, can be used to assess the effects of essential hypertension. These model-based DCA and DVR indices were properly defined to capture the observed effects of DCA/DVR and found to be significantly different (p<0.05) in hypertensive patients. We also found significant differences between patients and controls in the relative contribution of three PDMs to the model output prediction - a finding that offers the prospect of identifying the physiological mechanisms affected by essential hypertension when the PDMs become interpreted in terms of specific physiological mechanisms.