Sinclair et al1 hypothesized that the effects of hypercapnia on gas exchange in lung injury were largely unknown and may be beneficial, although they subsequently found no significant change. In their model, hypercapnia was induced by adding CO2 to inspired gases or adjusting minute ventilation through changes in rate rather than tidal volume (keeping the mean airway pressure constant). In our institution, we have used another method, the addition of dead space, to induce hypercapnia in ARDS patients.,5 This approach also has the advantage that mean airway pressure is unaltered. No significant alteration was observed in Pao2, Pao2/fraction of inspired oxygen ratio, or shunting when acute hypercapnia was compared to normocapnia.,5 Sinclair et al1 postulated that lung injury produced by the homogeneous depletion of a surfactant may abrogate the beneficial effects of CO2 on gas exchange that is observed in normal lungs; however, hypercapnia may be advantageous in patients with heterogeneous lung injury such as that caused by pneumonia. Although only two patients in our study had pneumonia as a cause of ARDS, neither showed improvement in gas exchange when they were acutely hypercapnic.