Biofeedback-driven dialysis: where are we?

The progressive increase in the mean age and the growing conditions of co-morbidity, especially of cardiovascular pathologies and diabetes, have significantly worsened the patients' clinical status and tolerance to the hemodialysis (HD) treatment. On the other hand, the demand for short treatment times enhances the risk for hemodynamic instability as well as for inadequate depuration. The traditional management of the dialysis session, setting of predefined treatment parameters, with active therapeutic interventions only in the event of complications, is definitely unsuitable for short-lasting treatments, often complicated by hemodynamic instability, especially in critical patients. The first step to improve the management of the dialysis session is the utilization of continuous and uninvasive monitoring systems for hemodynamic or biochemical parameters involved in the dialysis quality. Special sensors for the continuous measurement of blood volume, blood temperature, blood pressure, heart rate, electrolytes, have been realized throughout the last 10 years. As a second step, some of these devices have been implemented in the dialysis instrumentation, mainly with a view to preventing cardiocirculatory instability but also to control the dialysis efficiency (biofeedback control systems). The basic components of a biofeedback system are: the plant, the sensors, the actuators and the controller. The plant is the biological process that we need to control, while the sensors are the devices used for measuring the output variables. The actuators are the working arms of the controller. The controller is the mathematical model that continuously sets the measured output variable against the reference input and modifies the actuators in order to reduce any discrepancies. Yet, in practice there are a number of conceptual, physical and technological difficulties to be overcome. In particular, the behavior of what is to be controlled may be non-linear and time-varying, with interactions between the actuators and the controlled variable. In these cases, more sophisticated control systems are needed, which must be capable of identifying the behavior of the process, and continuously update information data while the control is on. These complex systems are called adaptive controllers. In dialysis, over the last few years, it has been relatively easy to realize some biofeedback systems since a series of sensors have been developed for online monitoring. Three biofeedback devices are routinely used with the aim of improving the cardiovascular instability, one of the main problems limiting the tolerance to treatment by the patient and the quality of HD in itself - the first is the biofeedback control of blood volume, the second is the biofeedback control of thermal balance, and the third is the biofeedback control of blood pressure.