Attempt to normalize simulated exhaled nitric oxide according to ventilatory settings.

BACKGROUND

Simulated exhaled nitric oxide (eNO) depends on ventilatory settings used in different experimental conditions.

OBJECTIVES

To normalize the simulated minute exhaled nitric oxide according to different ventilatory settings.

WORKING HYPOTHESIS

Different ventilatory settings influence the concentrations of exhaled nitric oxide and these results can be normalized. METHODOLOGY AND STUDY DESIGN: We used a rubber lung model (50 ml) with an orifice through which a 3 mm endotracheal tube was introduced. The NO, which simulated that of endogenous production, was delivered through the base of the lung using a unidirectional rotameter and obtaining a concentration of around 25 ppb. The sample of gas was recorded through a 6 F arterial catheter introduced into the endotracheal tube to its tip. The ventilator used was a Babylog 8000. Air delivered was compressed and filtered and had an NO content of under 0.3 ppb. The NO level assessed was the plateau value given by the software of the Sievers NOA apparatus. Each experiment involved sampling during 1 min, three times. Normalization was done using a multiple cubic regression formula.

RESULTS

An increase in respiratory frequency or in peak of inspiratory pressure were accompanied by a decrease in eNO (ppb). Minute volume was adjusted for the percentage of leakage given by the ventilator. Normalization was obtained analyzing 518 respirations with different ventilatory settings. The coefficient of variation fell from 15.5% to 0.27%. Validation of the normalization formula was performed in other three groups (320, 372, and 372 respirations) with different simulated NO concentrations (25, 16, and 50 ppb), resulting in reduction of the coefficient of variation from 42.7% to 9.3%, from 42.3% to 10.6% and from 45.2% to 9.6%, respectively.

CONCLUSIONS

Normalization of simulated minute eNO according to ventilatory settings is possible using the equipment and experimental set-up reported. Extrapolation to patients is not possible without constraints.