Variation of inhaled nitric oxide concentration with the use of a continuous flow ventilator

Objective: To investigate the homogeneity of nitric oxide con centrations at different ventilator settings in a delivery system using a continuous flow ventilator. Design: A prospective bench study using a nitric oxide delivery system, mixing a nitrogen/nitric oxide gas mixture in the ventilator circuit with two types of continuous flow ventilators (Babylog 8000(TM), Draeger, Luebeck, Germany; Infant Star(TM), Infrasonics, San Diego, CA). Setting: A biomedical laboratory. Interventions: A nitrogen/nitric oxide gas mixture was injected at three different sites in the ventilator circuit (just before and just behind the humidifier, and 20 cm before the Y-connector). Ventilator flow (12, 15, and 20 L/min) and rates (30 to 110 breaths/min with increments of 10 breaths/min) were changed as well as the compliance of the test lung (0.36, 0.5, and 1.0 ml/cm H2O). Carbon dioxide, instead of nitrogen/nitric oxide, was injected at the same points in the circuit. Measurements and Main Results: The mean nitric oxide concentration increased significantly (p < .001) with increasing ventilator rates (although the flow ratio of the ventilator gas and the nitrogen/nitric oxide gas mixture was kept constant) when the nitrogen/nitric oxide injection site was near to the Y-connector of the ventilator circuit with both ventilators. The mean nitric oxide concentration did not change significantly when the nitrogen/nitric oxide gas mixture was mixed to the ventilator gas at the inlet of the humidifier, using the Babylog 8000 ventilator. Analysis of ventilator circuit flow patterns showed fluctuations during the respiratory cycle. The magnitude of the flow changes was different at the three injection sites in the ventilator circuit. Real-time measurements of the CO2 concentration showed fluctuations during the distinct respiratory phases that differed at the separate injection sites. Mean CO2 concentrations showed a similar pattern as compared with the mean nitric oxide concentration data at the same settings. Conclusions: Mixing a nitrogen/nitric oxide gas mixture 20 cm before the Y-connector results in an increase of the mean nitric oxide concentration with increasing ventilator rates. This phenomenon does not occur with the nitrogen/nitric oxide gas mixture mixed at the inlet of the humidifier, using a ventilator with a through out constant flow at the inspiratory outlet of the ventilator. The fluctuations of the main ventilator circuit flow result in changing ratios of nitrogen/nitric oxide gas mixture and the ventilator gas flow. We speculate this changing flow ratio produces the increase in mean nitric oxide concentration within the ventilatory circuit. To ensure a constant concentration of nitric oxide by blending a nitrogen/nitric oxide gas mixture in the ventilator circuit of a continuous flow ventilator, the site of injection of the nitrogen/nitric oxide gas mixture should be at the point where ventilator circuit flow fluctuations are minimal.