A retrospective review of the records of 115 dogs with 13 types of malignant tumors was performed in an attempt to identify factors that influence the degree of emesis observed within 24 hours after cisplatin therapy. Six groups were established on the basis of dosage of cisplatin and on the route of administration: 10 mg/m2 IV (n = 17), 40 mg/m2 IV (n = 10), 50 mg/m2 IV (n = 19), 60 mg/m2 IV (n = 7), 70 mg/m2 IV (n = 36), and 70 mg/m2 intra-arterially (IA; n = 26). Age, gender, weight, dose of cisplatin, route of administration (IV vs IA), and duration of infusion were evaluated. Increasing doses of cisplatin (P less than 0.01) and the IV route of drug administration (P less than 0.0006) were associated with an increased occurrence of vomiting in response to the first treatment. Anesthesia, performed in dogs that were given cisplatin IA, was a confounding variable that may have reduced the emetic potential of the drug after the dogs were awakened. If the dog vomited in response to the first treatment, there was a greater tendency to vomit after subsequent cisplatin treatments (P = 0.08). Multivariable analysis, after correcting for the effects of age, gender, and weight, indicated that high doses of cisplatin (P less than 0.001) and lower weight of the dog (P less than 0.04) were significantly associated with an increased likelihood of vomiting.(ABSTRACT TRUNCATED AT 250 WORDS)