Objective-To compare cardiopulmonary responses during anesthesia maintained with halothane and responses during anesthesia maintained by use of a total intravenous anesthetic (TIVA) regimen in horses. Animals-7 healthy adult horses (1 female, 6 geldings). Procedure-Each horse was anesthetized twice. Romifidine was administered IV, and anesthesia was induced by IV administration of ketamine. Anesthesia was maintained for 75 minutes by administration of halothane (HA) or IV infusion of romifidine, guaifenesin, and ketamine (TIVA). The order for TIVA or HA was randomized. Cardiopulmonary variables were measured 40, 60, and 75 minutes after the start of HA or TIVA. Results-Systolic, diastolic, and mean carotid arterial pressures, velocity time integral, and peak acceleration of aortic blood flow were greater, and systolic, diastolic, and mean pulmonary arterial pressure were lower at all time points for TIVA than for HA. Pre-ejection period was shorter and ejection time was longer for TIVA than for HA. Heart rate was greater for HA at 60 minutes. Minute ventilation and alveolar ventilation were greater and inspiratory time was longer for TIVA than for HA at 75 minutes. The PaCO2 was higher at 60 and 75 minutes for HA than for TIVA. Conclusions and Clinical relevance-Horses receiving a constant-rate infusion of romifidine, guaifenesin, and ketamine maintained higher arterial blood pressures than when they were administered HA. There was some indication that left ventricular function may be better during TIVA, but influences of preload and afterload on measured variables could account for some of these differences.