This clinical study analysed the anaesthetic sparing effect of a medetomidine constant rate infusion (CRI) during isoflurane anaesthesia in horses. Forty healthy horses undergoing different types of orthopaedic and soft tissue surgeries were studied in a randomized trial. Orthopaedic surgeries were primarily arthroscopies and splint bone extractions. Soft tissue surgeries were principally castrations with one ovariectomy. All horses received 0.03 mg kg−1 acepromazine IM 1 hour prior to sedation. Group A (11 orthopaedic and nine soft tissue surgeries), was sedated with 1.1 mg kg−1 xylazine IV, group B (13 orthopaedic and seven soft tissue surgeries) with 7 µg kg−1 medetomidine IV. Anaesthesia was induced in both groups with 2.2 mg kg−1 ketamine and diazepam 0.02 mg kg−1 IV. Maintenance of anaesthesia was with isoflurane (ISO) in 100% oxygen, depth of anaesthesia was always adjusted by the first author. Group B received an additional CRI of 3.5 µg kg−1 hour−1 medetomidine. Respiratory rate (RR), heart rate (HR), mean arterial blood pressure (MAP), Fe′ISO and Fe′CO2 were monitored with a methane insensitive monitor (Cardiocap 5, Ohmeda, Anandic, Diessenhofen) and noted every 5 minutes. Arterial blood was withdrawn for gas analysis (PaO2, PaCO2) 5 minutes after the induction of anaesthesia and every 30 minutes thereafter. Dobutamine (DOB) was given as a CRI to maintain mean arterial blood pressure above 70 mm Hg. Data were averaged over time (sum of measurements/number of measurements) and tested for differences between groups by unpaired t-tests. There were no significant differences between the groups in terms of body mass (group A, 508 ± 73.7 kg; group B, 529.25 ± 78.4 kg) or duration of anaesthesia (group A, 125.5 ± 36 minutes; group B, 121.5 ± 48.4 minutes). The mean Fe′ISO required to maintain a surgical plane of anaesthesia was significantly higher in group A (1.33 ± 0.13%) than in group B (1.07 ± 0.19%; p = 2.78 × 10−5). Heart rate was different between the two groups (group A, 42.2 ± 8.3; group B, 32.6 ± 3.5; p = 8.8 × 10−5). Dobutamine requirements were higher in group A (group A, 0.72 ± 0.24 μg kg−1 minute−1; group B, 0.53 ± 0.23 μg kg−1 minute−1; p = 0.023). Respiratory rate, Fe′CO2, PaO2, PaCO2 were not different between the groups. Adjustment of anaesthetic depth subjectively was easier with the medetomidine infusion and isoflurane (group B) than with isoflurane as a sole agent (group A). In group A 12 horses and in group B five horses showed purposeful movements on 27 (A) and 12 (B) occasions. They were given thiopental (group A, 0.0114 mg kg−1 minute−1; group B, 0.0023 mg kg−1 minute−1). In group A, a further 17 horses were given ketamine to deepen anaesthesia (52 occasions, 0.00426 mg kg−1 minute−1) whereas in group B only nine horses needed ketamine (34 occasions, 0.00179 mg kg−1 minute−1). An infusion of 3.5 µg kg−1 MED during ISO anaesthesia resulted in a significantly reduced ISO requirement.