2000 OPEN FORUM Abstracts
DELIVERY OF SUB-ATMOSPHERIC OXYGEN CONCENTRATIONS DURING MECHANICAL VENTILATION OF CHILDREN WITH HEART DISEASE
David M. Dolcini BS RRT, Philip C. Smith MD PhD, Timothy R. Myers BS RRT Robert L. Chatburn, RRT FAARC, University Hospitals of Cleveland, OH
INTRODUCTION: The immediate survival of infants with hypoplastic left heart syndrome (HLHS) is dependent upon the success in achieving several therapeutic goals. One of these therapeutic goals is to establish and maintain a balance between systemic and pulmonary blood flow at or near unity. Because air-oxygen blenders have a lower limit of 21% oxygen, one technique to deliver sub-atmospheric oxygen concentrations is to add a measured flow of nitrogen to the inhaled gas. But this technique is imprecise. Flows as low as 0.1 L/min can change the oxygen concentration by 2 or 3 percent. The purpose of this study was to evaluate the feasibility of delivering sub-atmospheric oxygen concentrations through the ventilator blender. Specifically, we sought to determine the relation between blender reading and delivered oxygen concentration when the blender was supplied with nitrogen and air instead of oxygen and air.
Methods: A Teledyne T-190 oxygen analyzer was calibrated with precision blended gases at 5 different oxygen concentrations (Respir Care. 1999; 44:1226) prior to use and between each ventilator brand studied. Infant ventilators (Infant Star, Bear Cub, and Bird VIP) were used for this study (three of each brand). All nine blenders on the ventilators were calibrated at 21 and 100% prior to study implementation. A k-tank of nitrogen (50 psi) was attached to each air inlet and air (50 psi) was attached to each oxygen inlet of the ventilators. Ventilators were connected to a test lung with settings that created a mean airway pressure of 11-12 cmH2O. The Teledyne analyzer was placed in-line before the humidifier. Ventilator blenders were adjusted to achieve desired (ie, measured) oxygen concentrations of 0-21%.
Results: Sub ambient oxygen concentrations could be set precisely with all ventilators. The maximum difference between predicted setting (from regression line below) and the actual setting on an individual ventilator was 5% for the Infant Star and 2% for the other ventilators.
CONCLUSION: Delivering sub-atmospheric concentrations of oxygen can be reliably achieved during mechanical ventilation by connecting the blender to nitrogen and air sources instead of oxygen and air. We recommend this technique instead of using a flowmeter to mix nitrogen into the inspiratory limb of the ventilator circuit. (See Original for Figure)