RESUMEN
In order to test gas exchange in lungs with unilateral injury, when mechanical ventilation is needed, three groups (G-I, G-II, G-III) of seven dogs each were studied. Injury was induced in the left lung by injecting 0,lN, 1.0 ml per kg of body weight of hydrochloric acid, pH = 2.00. For groups I and II a conventional volumetric artificial ventilator was used. G-I was the control group. In group II a bilateral positive end-expiratory pressure (PEEP) was introduced during the last phase (phase 3) of the experiment, and in group III, a prototype of a volume cycled ventilator, with separated bellows, was used delivering tidal volumes separately to each lung through a Carlens' tube. In G-III, PEEP was introduced only to the injured lung in the last phase of the experiment. Phase l (Fl), basal phase, was similar in the three groups; phase 2 (F2) was the period after instillation of hydrochloric acid and phase 3 was the period after a 5 cmH2O bilateral PEEP was used in G-II, and a selective PEEP to the left lung was used in G-III. In each phase of the experiment, hemodynamic and gas exchange variables were obtained. Our results allowed the following conclusions: the differential lung ventilation technique maintained the alveolar ventilation of the dogs and the differential lung ventilation with unilateral PEEP was better for gas exchange (phase 3) than the conventional ventilation with bilateral PEEP (CPPB).