A comparison of the effects of three techniques for maintaining anesthesia in a continuous intravenous infusion of medetomidine hydrochloride-butorphanol tartrate (Tributeran). To compare the effects of three techniques for maintaining anesthesia in a continuous intravenous infusion of medetomidine hydrochloride-butorphanol tartrate in cattle. Prospective, randomized study. Fifty-two healthy nonpregnant, nonlactating (n = 25) or lactating (n = 27) cows. Cows were assigned randomly to receive a 5-mg/kg loading dose of medetomidine and 0.01 mg/kg/min loading dose of butorphanol (IV), 5 mg/kg loading dose of medetomidine and 1 mg/kg/hr loading dose of butorphanol (SC), or 0.007 mg/kg/min loading dose of butorphanol (IM). Medetomidine and butorphanol were administered continuously by an infusion pump (VetPRO: SmartSite Inc., Boise, ID) and a 50-mL polyvinyl chloride (PVC) syringe (Becton-Dickinson Animal Health Systems, Inc., Franklin Lakes, NJ) during a 3-day study. Two observers evaluated visual and tactile responses and degree of resistance to removal and placement of electrodes on the left side of the body. In the first 12 hours, scores for physical function, mental status, eye and head condition, muscle condition, ear condition, attitude, tail position, and reaction to noise were compared between groups. In the second 12 hours, data on the degree of anesthesia and reaction to noise were recorded. Cows were considered anesthetized when they failed to respond in response to noise or manipulation. Cows were considered anesthetized when they failed to respond in response to manipulation. Data were analyzed using analysis of variance (Proc-NIX: SAS Institute Inc, Cary, NC). Significance was established at P < 0.05. The IV group received the loading dose of butorphanol SC, IM, or IV. Medetomidine, butorphanol, and detomidine were purchased from Novartis Animal Health (Larch Grove, IA). Data were collected during the time of day when cows had a voluntary full complement of teeth. There were no significant differences between groups at any time point during the experiment. Cows receiving IV treatment displayed less responsiveness in terms of tail movement and reaction to noise (P < 0.05). This response continued throughout the 2-hour observation period. Cows receiving SC and IM treatments were less responsive for the first 12 hours, but the difference was not significant at any time point. Only cows in the IV group achieved a reliable level of sedation. There was no difference in degree of sedation between groups. The IM butorphanol rate resulted in a more reliable response in terms of sedation, and this technique was preferred by the staff. However, the IV rate resulted in a greater magnitude of sedation. These results provide guidelines for maintenance of sedation in cows that are either undergoing castration or that have been castrated and remain in a stocker facility for a short period of time.