SUMMARY Objective To determine the incidence of gastrointestinal (GI) tract bleeding in dogs undergoing spinal surgery with adjunct corticosteroid treatment, and to determine the protective efficacy of cimetidine, sucralfate, and misoprostol against such bleeding in these dogs. Animals 40 dogs that underwent spinal surgery. Procedures Myelography and surgery were performed on the first or second day of hospitalization. Methylprednisolone sodium succinate was given at a dosage of 30 mg/kg of body weight prior to myelography, followed by a second full or half dose 2 to 4 hours later at clinician discretion. Spinal surgery was performed in conventional manner, postoperative administration of analgesics was done, and dogs were fed a diet lacking red meat. Dogs were assigned at random to 1 of the 3 treatment groups or to the control group. Dogs of the treatment groups received cimetidine, sucralfate, or misoprostol. Physical examination and determination of PCV and serum total protein values were performed daily. A fecal sample was examined daily for gross and occult blood. Results 36 of 40 dogs had GI tract bleeding during a hospitalization period of 3 to 6 days. There was no significant difference in development of bleeding between the control group and any of the treatment groups. Conclusions Gastrointestinal tract bleeding occurred in 90% of dogs undergoing spinal surgery combined with administration of methylprednisilone sodium succinate, a higher rate than that found in previous studies. This bleeding was not life-threatening. Prophylactic benefit from any of the GI protectants tested was not found. (Am J Vet Res 1997;58:1320–1323)

Medical records of 52 dogs born with portal venous anomalies (PVA) were reviewed and separated into 2 groups, based on anatomic location of the shunt (intrahepatic vs extrahepatic). Statistical analysis was used to compare the groups with respect to presurgical clinicopathologic data, as well as to compare presurgical clinicopathologic data with postoperative mortality. Three presurgical predictors of location were identified. Mean body weight at admission was statistically different (P = 0.0001) between dogs with an intrahepatic PVA (mean, 15 kg) and those with an extrahepatic PVA (mean, 4.8 kg). Blood glucose concentration was significantly (P = 0.028) higher in dogs with an intrahepatic PVA (mean, 99.8 mg/dl) than in dogs with an extrahepatic PVA (mean, 85.1 mg/dl). The third significant (P = 0.036) predictor of location was degree of elevation of serum alkaline phosphatase concentration. The intrahepatic location resulted in a concentration 3.2 times that of the reference value, whereas an extrahepatic location resulted in a 1.7 times increase. Two predictors of early postoperative mortality were identified. In comparing the 2 groups, significant differences in PCV before surgery (P = 0.04) and in rectal temperature after surgery (P = 0.002) were found. Dogs that survived had a mean presurgical PCV of 36% and a mean postoperative rectal temperature of 35.7 C, whereas nonsurvivors had a mean presurgical PCV of 40.6% and a mean temperature of 34.3 C. There was no significant difference between nonsurvivors and survivors with respect to postligation portal pressure or change in portal pressure after ligation. Overall mortality was 15%.