Feedback Control of Arterial Pressure Using Deep Brain Stimulation: Transfer Function Analysis and Simulation Study.

BACKGROUND

Deep brain stimulation (DBS) is an established treatment for Parkinson disease. Emerging evidence suggests that DBS of the subthalamic nuclei may be a target for neuromodulation therapy for orthostatic hypotension due to efferent baroreflex failure. To develop a central algorithm for an artificial baroreflex system, we first estimated the dynamic responses of arterial pressure to DBS, and then determined the feedback correction factors for DBS.

METHODS

A total of 14 Parkinson disease patients treated with DBS participated in submaximal stimulation screening tests for 20 s to identify responders exhibiting a pressor reaction of ≥10 mm Hg in systolic arterial pressure. In the responders, we randomly performed stimulation while tonometrically measuring the systolic arterial pressure. The transfer function from stimulation-to-systolic arterial pressure was computed using a fast Fourier transform algorithm. Finally, the feedback correction factors were determined via numerical simulations.

RESULTS

The transfer function was identifiable in all 3 responders. The average stimulation-to-systolic arterial pressure transfer function exhibited low-pass characteristics, with a steady-state gain of 8.24 mm Hg/stimulation, corner frequency of 0.02 Hz, and pure delay of 2.5 s. Based on these results, the proportional and integral correction factors were set to 0.12 and 0.018, respectively, for the feedback control of the stimulation. Computer simulations revealed that the system could quickly and effectively attenuate a sudden drop in systolic arterial pressure induced by external disturbances, such as head-up tilt.

CONCLUSIONS

An artificial baroreflex system with the feedback correction of stimulation of DBS may be a novel therapeutic approach for efferent baroreflex failure.