1. Ventilation—perfusion balance in the presence of airway obstruction will depend on the efficiency of hypoxic pulmonary vasoconstriction beyond obstructed airways and the matching of redistributed blood flow and ventilation to the rest of the lung. This study investigated the relative importance of these mechanisms in man during experimental bronchial occlusion. 2. The bronchus to the left lower lobe was temporarily occluded with a balloon-tipped catheter during fibreoptic bronchoscopy in eight supine normal volunteers. Respiratory gas tensions were measured within the occluded lobe with a respiratory mass spectrometer. The distribution of ventilation and perfusion was assessed under control conditions and after 5 min of bronchial occlusion by computer analysis of the regional distribution of radioactivity during inhalation of 81m Kr gas and following injection of 99m Tc-labelled macroaggregated albumin respectively. 3. Respiratory gas partial pressures within the occluded lobes rapidly stabilized at mixed venous gas tensions: P o 2 43.4 ± 2.2 (SEM) mmHg, P co 2 40.2 ± 1.8 mmHg. During occlusions the arterial oxygen saturation fell from a baseline of 96.3 ± 0.46% to a nadir of 92.1 ± 0.43%. Bronchial occlusion produced underventilation in the left lung relative to perfusion, both in the region of the occluded lower lobe and at the lung apex. Relative overventilation occurred in the right lung. 4. It is concluded that arterial hypoxaemia during lobal bronchial occlusion is caused primarily by shunting of mixed venous blood, though the shunt fraction is reduced by approximately 50% by hypoxic pulmonary vasoconstriction. In lung adjacent to obstructed regions reduced compliance may impair ventilation more than perfusion to contribute to hypoxaemia. It seems likely that redistribution of ventilation and perfusion to unobstructed regions during lobar bronchial occlusion is dependent on mechanical factors rather than O 2 - or CO 2 -dependent changes in bronchial or vascular tone.

1. Acute hypoxic pulmonary vasoconstriction is important in the restoration of ventilation—perfusion balance in the presence of regional alveolar hypoventilation. However, the magnitude and time course of this response in man has not been adequately characterized in regions smaller than an entire lung. We have studied the effectiveness of hypoxic vasoconstriction in diverting blood from hypoxic lobes in normal supine subjects, and have documented the redistribution of pulmonary blood flow under these conditions. 2. Lobar hypoxia was induced for 80–300 s by placing occluding balloon-tipped catheters in lobar bronchi during fibreoptic bronchoscopy in 10 normal subjects. Respiratory gas partial pressures within occluded lobes were measured with a mass spectrometer. The percentage reduction in blood flow to the hypoxic lobes was assessed after injection of 99m Tc-labelled albumin by γ-scintigraphy, and compared with a control scan performed 1 week later. A computer program was used to analyse changes in regional pulmonary perfusion. 3. During lobar bronchial occlusion respiratory gas partial pressures rapidly approached reported values for mixed venous partial pressures. After a mean time of occlusion of 3.5 min lobar blood flow was reduced by 47 ± 5%. During occlusions pulmonary blood flow was not evenly redistributed, but was preferentially redistributed to more cranial lung regions. 4. We conclude that acute hypoxic pulmonary vasoconstriction in occluded lobes is more effective at rapidly diverting pulmonary blood flow away from hypoxic lung regions than has previously been reported in man during unilateral hypoxia of an entire lung. Non-uniform redistribution of pulmonary blood flow in the supine subject is likely to be due to compression of the lung bases by the diaphragm in the supine position.

1. Pulmonary function tests, including alveolar mixing efficiency by the single-breath and multi-breath methods, and ventilation scans were performed on 16 volunteer subjects. The tests were repeated after the inhalation of a methacholine aerosol in sufficient dosage to increase airways resistance. 2. After inhalation of methacholine there was a significant fall in mean series dead space of 31 ml ( P < 0.05), and mean multi-breath alveolar mixing efficiency fell from 68% to 36% ( P < 0.001), a fall occurring in all subjects. Mean single-breath alveolar mixing efficiency measured on the first breath of the nitrogen washout fell from 76% to 70%, but this change did not reach statistical significance (0.1 > P > 0.05). 3. In eight of the subjects, technically adequate lung scans and pulmonary function tests were obtained both before and not more than 30 min after methacholine inhalation. In seven there were obvious visible defects on the ventilation scans, and in five of these the computer-calculated underventilation score became abnormal. 4. Thus inhalation of methacholine causes maldistribution of ventilation, a fall in alveolar mixing efficiency and a fall in series dead space, presumably brought about by bronchoconstriction. The parallel component of this maldistribution of ventilation, as judged by 81m Kr ventilation scanning, does not of itself seem to be sufficient to explain the fall in alveolar mixing efficiency, and therefore a degree of diffusion limitation is probably involved as well.