We compared the effects of almitrine and doxapram on the arterial blood gases and ventilation of patients with chronic respiratory insufficiency and chronic hypercapnia and hypoxemia. Sixteen long-term in-patients were randomly allocated to one of the following treatment groups: the first group (8 patients) received IV almitrine 0.5 mg/kg and the second group (8 patients) IV doxapram 1 mg/kg by IV perfusion during 30 min. All gave their informed consent. Arterial blood gases and ventilation were measured 10 min and 5 min before treatment, at the 5th, 15th and 25th min of perfusion time, and 5, 10 and 15 min after infusion. There was a marked increase in paO2 in almitrine-treated patients, which was maximum at the 25th min of infusion (+ 14.6 mm Hg, p < 0.001), but only a slight improvement was observed in the doxapram group (+ 3.3 mm Hg, p < 0.05). After almitrine the maximum mean paCO2 decrease was at the 10th min after perfusion (-6.9 mm Hg, p < 0.001); after doxapram the maximum decrease, although highly significant, was much less (-2.8 mm Hg, p < 0.01). Thus, at the presently used and well-tolerated doses, almitrine is much more efficient than doxapram in improving gas exchange in patients with chronic hypoxemia and hypercapnia. However, complementary studies using higher dosage of doxapram are warranted.