To determine how and why different bronchodilators affect ventilation-perfusion (V/Q) relationships differently in experimental asthma, 10 anesthetized dogs were exposed 3 times each (2-1/2 h apart) to aerosolized 1% methacholine over 3 to 5 min. Mechanical, hemodynamic, gas exchange, and catecholamine concentration measurements were made before challenge, and 10, 20, 30, and 135 min after challenge. Fifteen min after challenge the dogs were exposed to aerosolized epinephrine, isoproterenol, salbutamol, or isotonic saline in random order. The worsened V/Q inequality after isoproterenol paralleled increased cardiac output (QT) and decreased pulmonary vascular resistance (PVR). Because of increased QT, arterial PO2 (PaO2) did not fall in spite of the increased V/Q inequality. After administration of isoproterenol, improvement in air-flow obstruction was always associated with significantly less increase in mean PaO2 and more V/Q inequality than after saline. After epinephrine and salbutamol a similar significant decrease in airflow obstruction was associated with only slightly more V/Q mismatching than after saline. The effects of epinephrine on low V/Q ratio areas, shunt, PaO2, QT, heart rate, and PVR were less and of shorter duration than those of isoproterenol. Salbutamol resulted in changes similar to but more persistent than those induced by epinephrine. All variables approached prechallenge values within 2 h after bronchodilator administration, regardless of agent used. These studies show that in a repeatable canine asthma model, randomized administration of epinephrine, isoproterenol, and salbutamol produce different gas exchange responses in spite of similar improvements in air-flow obstruction.(ABSTRACT TRUNCATED AT 250 WORDS)