2002 OPEN FORUM Abstracts
TIDAL VOLUME DELIVERY DECREASES WITH INCREASING FiO2 DURING VOLUME AND PRESSURE VENTILATION IN THREE OF FOUR PEDIATRIC VENTILATORS.
John W. Berkenbosch1, Ryan Grueber2, Osauma Dabbagh3, Andrew McKibben3. Departments of Child Health1, Respiratory Care2, and Internal Medicine3, The University of Missouri-Columbia, Columbia, MO, USA
INTRODUCTION: It is well recognized that alterations in gas density can impact ventilator and pneumotach function. While it has been assumed that the small difference in density between room air and 100% oxygen would have negligible effect on these functions, we have been unable to find published data to confirm this.
Methods: We investigated the effect of FiO2 (0.2, 0.4, 0.6, 0.8, 1.0) on delivered tidal volume (VT) during volume and pressure ventilation in 4 pediatric ventilators, the BIRD VIP and VIP Gold and the Siemans Servo 900C and Servo 300. During volume ventilation, set tidal volumes (VT) ranged from 20-250 mL. During pressure ventilation, inspiratory pressures were chosen to deliver tidal volumes in the same range. Inspiratory VT?s displayed on the ventilator (displayed VT) were recorded and compared with the inspiratory VT displayed during ventilation with room air (expected VT Dis). Actual delivered VT?s were measured with a Neonatal Bicore connected to the side port of a ?bag-in-box? spirometer, making volume measurements independent of inspired gas properties. This volume (delivered VT) was compared to the VT delivered by the ventilator during ventilation with room air (expected VT Del). For each ventilator, ratios of displayed VT and delivered VT/expected VT (Dis or Del) were calculated and compared at each FiO2 with a Kruskal-Wallis analysis of variance. A p value <0.05 was considered significant.
Results: During volume ventilation, there was a small decrease (2.2%) in displayed VT with increasing FiO2 with the VIP Gold. Displayed VT was unaffected by FiO2 with the VIP, Servo 900, and Servo 300 ventilators. There was a progressive decrease in delivered VT with increasing FiO2 with the VIP (6.5%), VIP Gold (5.4%), and Servo 900C (13%) (p<0.001). This decrease was present over the range of tidal volumes measured, and is similar to the difference in density between room air and 100% O2 (10.8%). Delivered VT was unaffected by changes in FiO2 with the Servo 300 ventilator.
During pressure ventilation, there was a small but significant increase in displayed VT with increasing FiO2 with the VIP (3.0%), VIP Gold (4.5%) and Servo 900C (6.3%) ventilators (p<0.005). In contrast, displayed VT decreased 3.2% with increasing FiO2 with the Servo 300 ventilator (p<0.001). As with during volume ventilation, delivered VT decreased with increasing FiO2 with the VIP (4.3%), VIP Gold (2.4%) and Servo 900C (2.3%) ventilators (p<0.01). However, the magnitude of this decrease was less than during volume ventilation. There was no significant effect of FiO2 on actual delivered VT with the Servo 300 ventilator.
DISCUSSION: This is the first study to demonstrate a decrease in tidal volume delivery with increasing FiO2 in 3 of 4 ventilators commonly used in pediatric patients. This decrease was both ventilator and ventilation-mode specific. Changes were most pronounced during volume ventilation but the effects did not appear to be sensed by the ventilator. The absence of an effect with the Servo 300 is most likely a result of its gas module design. While the clinical significance of this effect is likely minimal in most scenarios, the practitioner should consider this when making large changes in FiO2.