Exhaled nitric oxide and carbon monoxide in mechanically ventilated brain-injured patients

The inflammatory influence and biological markers of prolonged mechanical-ventilation in uninjured human lungs remains controversial. We investigated exhaled nitric oxide (NO) and carbon monoxide (CO) in mechanically-ventilated, brain-injured patients in the absence of lung injury or sepsis at two different levels of positive end-expiratory pressure (PEEP). Exhaled NO and CO were assessed in 27 patients, without lung injury or sepsis, who were ventilated with 8ml kg(-1) tidal volumes under zero end-expiratory pressure (ZEEP group, n = 12) or 8 cm H2O PEEP (PEEP group, n = 15). Exhaled NO and CO was analysed on days 1, 3 and 5 of mechanical ventilation and correlated with previously reported markers of inflammation and gas exchange. Exhaled NO was higher on day 3 and 5 in both patient groups compared to day 1: (PEEP group: 5.8 (4.4-9.7) versus 11.7 (6.9-13.9) versus 10.7 (5.6-16.6) ppb (p < 0.05); ZEEP group: 5.3 (3.8-8.8) versus 9.8 (5.3-12.4) versus 9.6 (6.2-13.5) ppb NO peak levels for days 1, 3 and 5, respectively, p < 0.05). Exhaled CO remained stable on day 3 but significantly decreased by day 5 in the ZEEP group only (6.3 (4.3-9.0) versus 8.1 (5.8-12.1) ppm CO peak levels for day 5 versus 1, p < 0.05). The change scores for peak exhaled CO over day 1 and 5 showed significant correlations with arterial blood pH and plasma TNF levels (r(s) = 0.49, p = 0.02 and r(s) = -0.51 p = 0.02, respectively). Exhaled NO correlated with blood pH in the ZEEP group and with plasma levels of IL-6 in the PEEP group. We observed differential changes in exhaled NO and CO in mechanically-ventilated patients even in the absence of manifest lung injury or sepsis. These may suggest subtle pulmonary inflammation and support application of real time breath analysis for molecular monitoring in critically ill patients.