We have previously shown that nitric oxide, an important mediator of pulmonary vasodilation at birth, reduces lung liquid production (Pediatr Res 1994; 329A) and that this effect is not mimicked by infusing cGMP directly into the pulmonary circulation (Pediatr Res 1996; 329A). To see whether cGMP might act directly on the pulmonary epithelium, we studied the effects of cGMP instilled into the lung liquid of six chronically instrumented fetal sheep. Sixteen experiments were done at an average gestational age of 129 ± 4 d. Net lung luminal liquid production (Jv) was measured by plotting the change in lung luminal liquid concentration of radiolabelled albumin, an impemeant tracer that was mixed into the lung liquid at the beginning of each study. Jv was measured during a 1-2 h baseline, then for 2-3 h following instillation of 8-Br-cGMP (5-20 mg), a stable cGMP analog. Following cGMP instillation (n=8), Jv decreased from 17 ± 6 ml/h to -4 ± 15 ml/h (P=.001); left pulmonary artery blood flow and pulmonary and systemic arterial pressures were not significantly affected. Control (saline) instillations (n=8) caused no significant change in any variable.

In summary, instillation of cGMP into the fetal lung liquid decreased net lung liquid production, and in some cases resulted in net resorption of lung liquid; this effect was not associated with any significant change in pulmonary hemodynamics. In contrast, we have previously shown that infusing cGMP into the pulmonary circulation increases pulmonary blood flow but has no effect on Jv. Thus, the effect of nitric oxide on lung liquid production can be mimicked by cGMP, via a direct effect on the pulmonary epithelium, and independent of its effect on the pulmonary circulation.

Funded by AHA Grant-in-Aid #94-317 and the J.H. Cummings Foundation