Patients with chronic obstructive pulmonary diseases (COPD) and/or central sleep apnea are sometimes treated with the carbonic anhydrase inhibitor acteazolamide to improve blood gas values. Studies have shown that this agent may have a complicated effect on lung ventilation, because carbonic anhydrase has a widespread distribution within the body, particularly in tissues involved in the control of breathing. To investigate whether acetazolamide may have (neuro)muscular effects on respiration, we measured the responses of ventilation, phrenic nerve activity, and transpulmonary pressure to changes in arterial PCO2 before and after intravenous administration of a low-dose (4.6 +/- 0.2 mg x kg(-1), mean +/- SEM) of this inhibitor in anesthetized spontaneously breathing rabbits. The agent decreased the mean resting end-tidal PCO2 by 1 kPa and increased ventilation from 258 +/- 15 to 292 +/- 14 ml x min(-1) x kg(-1) (p < or = 0.05). The ventilatory and tidal volume responses to CO2 were reduced, and the response curves were shifted to lower PCO2 values. At the level of phrenic activity, however, the response was shifted leftward without altering CO2 sensitivity. With an unchanged lung compliance, the slopes of the relationships between tidal volume and phrenic activity and that between the tidal change in transpulmonary pressure and phrenic amplitude were both reduced by about 40%, indicating an action of acetazolamide on (neuro)muscular level. The results raise the suggestion that treatment of some hypercapnic COPD patients with acetazolamide may have undesired clinical implications, particularly in those with already weakened respiratory muscles.