The Science Journal of the American Association for Respiratory Care

2008 OPEN FORUM Abstracts


Kevin Crezee1, Tim Macknight1, Jeff Hoydu1, Troy Lynch1

Background: Primary Children's Medical Center (PCMC) in Salt Lake City, Utah is a 252-bed pediatric tertiary center. PCMC's Pediatric Intensive Care Unit admits approximately 600 patients annually for care of congenital heart defects requiring surgical intervention. A subgroup of this population are patients with a single-ventricle anatomy dependent upon a narrow range of blood oxygen and blood carbon dioxide levels to maintain acceptable pulmonary blood flow. Currently, these patients are at times treated with sub-ambient oxygen therapy to control pulmonary blood flow. However, concerns exist regarding decreased cerebral saturation and its effects. We sought to develop a system that would allow for precise control of inspired carbon dioxide (FiCO2) through the Evita XL ventilator (Drager, Telford, PA) allowing for a constant level of FiCO2 and FiO2 >= 0.21, regardless of changes in minute ventilation.

Methodology: A low-flow blender's 50 pound per square inch (PSI) air line is connected to an air wall source and a 50 PSI line to an E-cylinder of 100% CO2. A 50 PSI line with CO2/Air is connected from the low-flow blender outlet to the high-flow blender air inlet. The high-flow blender's 50 PSI O2 line is attached to a wall source. A 50 PSI line was connected from the high-flow blender outlet to the O2 inlet of the ventilator. The air line on the ventilator is attached to a wall source. A CapnoCheck Plus monitoring device (BCI, Waukesha, WI) was placed at the ventilator wye to measure CO2 in millimeters of mercury (mmHg) and analyze FiO2. The CO2 monitoring device of the Evita XL was also placed pre-humidifier to measure CO2 in the inspiratory limb of the ventilator circuit. FiO2 on the ventilator was set to 1.0 and internal O2 monitoring was disabled. FiCO2 levels were controlled with the low-flow blender and FiO2 with the high-flow blender. Our goal was to achieve a CO2 reading of 18 mmHg or 2.8% CO2, based upon typical barometric pressure readings at our altitude. Set FiO2 on the O2 blender was maintained at 0.3

Results: See Table 1 for detailed results. There was little or no variation in set vs. measured FiO2 and it is not included in our table. There was little or no variation in measured CO2 across a wide spectrum of ventilator support.

Conclusion: Given our findings, we feel that inspired CO2 can be safely delivered and precisely controlled with this configuration. We suggest prominent labeling of all hoses and blenders.