Nitric oxide concentrations in the exhaled gas (NOe) increases during various inflammatory conditions in humans and animals. Little is known about the sources and factors that influence NOe. NOe at end expiration was measured by chemiluminescence in an isolated, blood-perfused rabbit lung. The average end-expiratory concentration over 10 breaths was used. The effect of positive end-expiratory pressure (PEEP), flow rate, pH, hypoxia, venous pressure, and flow pulsatility on NOe were determined. At constant blood flow, increasing PEEP from 1 to 5 cm H2O elicited a reproducible increase in NOe from 49 +/- 7 to 53 +/- 8 parts per billion (ppb) (p < 0.05). When blood pH was increased from 7.40 to 7.74 by breathing low CO2 gas, NOe rose from 45 +/- 7 to 55 +/- 7 ppb (p < 0.001). Hypoxia caused a dose-dependent decrease in NOe from 37 +/- 3 during baseline to 23 +/- 2 during ventilation with 0% O2 (p < 0.01). Venous pressure elevation from 0 to 5 and 10 mm Hg decreased NOe from 32 +/- 5, to 26 +/- 5 and 24 +/- 5 ppb, respectively (p < 0.05). Switching from steady to pulsatile flow (same man flow) resulted in a small, albeit significant reduction in NOe; 30 +/- 4 to 28 +/- 4 ppb (p < 0.05). Changes in flow rate between 200 and 20 ml/min were associated with small changes in NOe; however, when flow was stopped, NOe rose substantially to 56 +/- 6 ppb (p < 0.05). The changes in NOe were rapid (1 to 2 min) and reversible. The results suggest that NOe is influenced by ventilatory and hemodynamic variables, pH, and hypoxia. We suggest that caution must be taken when interpreting changes in exhaled NO in humans or experimental animals. Changes in total and regional blood flow, capillary blood volume, ventilation, hypoxia, and pH should not be overlooked.