Blood pressure control during surgical operations

Abstract

In order to reduce intraoperative blood loss and spare blood transfusion, we developed a blood pressure control system using a state-predictive controller. Using adult mongrel dogs, the mean arterial pressure (MAP) was recorded from a femoral artery while trimethaphan camsilate was infused at constant rates. A pure delay plus a first-order delay model was then derived from the dose-response curves and the values of plant parameters (gain, time-constant, dead-time, and so on) were estimated based on the experimental data. For this model, a state-predictive servo system was designed to cope with the pure delay existing in the model, and simulated. In order to evaluate the accuracy and reliability of this system, we experimented on dogs. With a reference MAP set at 60 mmHg, the MAP reached the reference level in 5.8 to 26.5 min. The duration of error from the reference MAP (+/- 10%) was 2.3 +/- 3.9 min/h (n = 7). These results indicated the safety and stability of our system.