Search

In the first days of life, the daily evaporative loss from premature neonates can reach up to 20% of body mass. Such loss can be reduced by increasing the air humidity inside the incubator. Neither passive humidification nor open loop systems allow high humidity rates to be maintained or easily controlled: at 34 degrees C, the maximum levels vary with the system from 40% to 77% of relative humidity. The skin evaporative exchanges between the neonate and the environment are directly proportional to the water vapour partial pressure difference between the neonate's skin and the air. An active closed loop system has been designed, which permits reliable and accurate control of humidity according to the water vapour partial pressure set, between 1 and 6 kPa, in an air temperature range of 28-39 degrees C. It is characterised by variations of about 0.05 kPa around the set value and a maximum humidification speed of 0.25 kPa min-1. The algorithm is based on optimal control and the dynamic programming principles. Test results place this active system above usual systems for its power, precision and adaptability. It is an exploitable tool in fundamental and clinical research, to precisely study the humidity effects on neonatal comfort and thermo-regulation evolution.

The anatomy, physiology, and biochemistry of the human skin have been studied for a long time. A special interest has been shown in the water permeability of the premature infant's skin, which is known to be an important factor in the maintenance of a controlled water and heat balance. The rate of evaporative heat exchange between the skin surface of a very premature infant and the surrounding incubator air may be so high that evaporative heat loss alone may exceed the infant's total metabolic heat production. However, it has been demonstrated in several investigations published in recent years that basal evaporative water loss can be consistently reduced by increasing the ambient humidity. Nevertheless, the passive humidification system (water reservoir) used in most incubators cannot achieve high and steady humidity levels. In this paper, we propose an active humidification system. The algorithm is based on a combination of optimal control theory and dynamic programming approach. The relative-humidity (R.H.) regulation is performed in range of 35-90% at 33 degrees C with small oscillations (+/- 0.5% R.H.) around the reference value (i.e., prescribed R.H.).