OBJECTIVETo compare the efficacy of inhaled salbutamol with salmeterol for the treatment of arterial hypoxaemia in anaesthetized horses.STUDY DESIGNProspective, randomized, clinical study.ANIMALSA total of 108 client-owned horses (American Society of Anesthesiologists status I-V) anaesthetized for elective and emergency procedures.METHODSHorses were premedicated with acepromazine [intramuscularly 0.1 mg kg-1 or intravenously (IV) 0.05 mg kg-1] and xylazine (0.6 mg kg-1 IV). Midazolam (0.06 mg kg-1 IV) and ketamine (2.2 mg kg-1 IV) were combined to induce anaesthesia, and isoflurane in oxygen/air mixture (inspired oxygen fraction 0.7) was used for maintenance of anaesthesia. Mechanical ventilation was initiated without delay using the following ventilator settings: tidal volume 10 mL kg-1, respiratory rate 8 breaths minute-1, inspiratory-to-expiratory time ratio 1:2, no positive end-expiratory pressure. If arterial blood gas analysis revealed PaO2 < 100 mmHg (13.3 kPa), the administration of either inhaled salbutamol (2 μg kg-1) or salmeterol (0.5 μg kg-1) was randomly assigned Blood gas analysis was repeated 15 and 30 minutes after treatment. The intervention was considered successful when PaO2 after treatment ≥ 1.2 × PaO2 before treatment (i.e. ≥20% increase). PaO2 at 15 and 30 minutes was compared between groups using Mann-Whitney U test; p < 0.05 was considered significant.RESULTSOf the 108 horses, 60 were administered salbutamol, 65% and 60% responded successfully at 15 and 30 minutes, increasing their initial PaO2 by 38% and 44%, respectively. The other 48 horses were administered salmeterol, 35% responded successfully at 15 and 30 minutes, increasing their initial PaO2 by 3% and 4%, respectively. PaO2 was significantly higher after salbutamol than after salmeterol at 15 and 30 minutes.CONCLUSIONS AND CLINICAL RELEVANCEUsing the described protocol, inhaled salbutamol was more effective than salmeterol in improving PaO2 in anaesthetized horses with value < 100 mmHg (13.3 kPa).