Researchers have long sought a noninvasive bedside test for diagnosing acute pulmonary embolism in critically ill surgical patients. Because pulmonary embolism increases the physiologic dead space in the lungs, researchers have tested various ways of using this fact as a diagnostic tool. The tests developed often failed because of the presence of another, coexisting lung disease. A previous study successfully differentiated healthy patients, patients with pulmonary embolism and patients with chronic obstructive pulmonary disease by measuring the late pulmonary dead space fraction (Fd_late) by carbon dioxide (CO₂) expirogram. The Fd_late is the calculated difference between arterial CO2 and the concentration of CO₂ in an exhalation of 15 percent of total lung capacity.

The CO₂ expirogram measures CO₂ concentration (in an exhaled breath) versus the volume of a single exhaled breath. The concentration of exhaled CO₂ normally reaches the level of the exhaled volume when 15 percent of the patient's total lung capacity has been exhaled. The CO₂ concentration of the exhaled breath eventually equals the arterial partial pressure of CO₂. In patients with pulmonary embolism, the CO₂ concentration of the exhaled breath does not rise to the arterial CO₂ level (at the 15 percent exhaled volume level) because of the increase in physiologic dead space. A high Fd_late value is indirect evidence of a pulmonary embolism. Patients with chronic obstructive pulmonary disease exhibit different slopes on the expirogram, but the CO₂ concentration level of the exhaled breath still approaches arterial CO₂ levels when a patient exhales to the 15 percent level.

In a recent prospective study, Anderson and colleagues evaluated Fd_late levels as a diagnostic tool in a population of post-trauma surgical inpatients. A total of 14 angiograms were performed on 12 patients who exhibited clinical signs of pulmonary embolism. Of the 14 tests, nine were negative for pulmonary embolism. CO₂ expirograms were obtained, and all but one of the nine patients had a negative expirogram with an Fd_late value below the predetermined cutoff value. All five patients with angiograms positive for pulmonary embolism had an Fd_late value in the range consistent with pulmonary embolism.

Overall, the CO₂ expirogram showed 100 percent sensitivity and 89 percent specificity. These results support previous data and distinguish the CO₂ expirogram from other methods of calculating the Fd_late at the bedside. The CO₂ expirogram can be determined using the patient's tidal volume, which enables it to be used with ventilated patients and accounts for other lung disease and changes seen on pulmonary function tests. Recent technologic advances make the CO₂ expirogram more accessible and make Fd_late calculation promising as a screening test for pulmonary embolism.