One of the claimed advantages of intermittent mandatory ventilation (IMV) over assisted mechanical ventilation (AMV) (assist-control) is the avoidance or correction of acute respiratory alkalosis, ostensibly by allowing patients to achieve normal alveolar ventilation (VA) and PaCO2 through the function of an intact ventilatory drive. However, although respiratory alkalosis in patients being hyperventilated with controlled mechanical ventilation (CMV) can be corrected by a change to IMV, CMV is seldom appropriate for patients with acute respiratory failure, and whether IMV affects respiratory alkalosis in patients triggering the ventilator in the AMV mode has not previously been tested. We studied 26 patients with acute respiratory alkalosis (pH greater than or equal to 7.48) while receiving AMV. Measurements of arterial blood gases and CO2 production (VCO2), and calculation of VA, were performed after 30 min of AMV, repeated after 30 min of IMV at a mandatory rate one half the previous AMV rate, and then repeated again 30 min after a return to the original AMV settings. Mean arterial pH decreased slightly from 7.51 during AMV to 7.48 during IMV, and returned to 7.51 on resumption of AMV (p less than 0.05 for both changes); corresponding mean values for PaCO2 were 28.6, 29.7, and 27.5 mmHg, respectively. These changes were related to an increase in VCO2 during IMV as compared with AMV (p less than 0.05), without a significant alteration in VA. When the mandatory rate was further reduced during IMV from one half to one fourth the prior, triggered AMV rate in 10 patients, no additional reduction in pH occurred.(ABSTRACT TRUNCATED AT 250 WORDS)