Abstract

The alveolo-arterial gradient of Pco₂ was measured and the percentage of ventilated but unperfused alveoli was derived from Severinghaus's formula in 312 patients divided into five groups according to clinical symptoms, biological, radiographic, scintigraphic or pathological data—87 patients with proved thromboembolic disease, 41 with probable thromboembolic disease, 67 with possible thromboembolic disease, 101 with non-embolic pulmonary disease, and 16 cases of miscellaneous cardiac diseases. After all capnographic curves without an alveolar plateau had been eliminated, 223 capnograms (71%) were examined.

In 59% of the patients with proved thromboembolic disease, the percentage of ventilated but unperfused alveoli was abnormal (>15%). In this group the mean percentage of ventilated but unperfused alveoli (16·3%) was significantly different from the value obtained in the remaining groups. This test was positive in 40% of the patients with probable thromboembolic disease but it was also positive in 22·5% of the cases of miscellaneous pulmonary and cardiac non-embolic diseases.

Compared to the lung scan, this method was less sensitive but also less equivocal in patients with preexisting cardiopulmonary disorders. Estimates of the pulmonary vascular defect by these two methods did not always correspond.