Blue heliotrope (Heliotropium amplexicaule) poisoning in cattle.

proximal rectum, as no macroscopic evidence of infarction was present at necropsy. The mare’s progressive deterioration in condition following resection and anastomosis resulted from leakage of faecal material through the anastomosis, into the peritoneal cavity. All anastomoses leak to some extent (Peacock 1984). but in the majority of cases containment and resolution occurs with few systemic signs. In this mare, sufficient leakage occurred to result in acute, diffuse peritonitis with concomitant signs of septic shock. Therefore, a diverting colostomy was performed to reduce peritoneal contamination. Colostomies have received little attention from equine surgeons in the past. Stashak and Knight (1978) reported the successful management of a grade 3 rectal tear with an end colostomy. Post-operative management included intra-colostomic enemas, regular cleaning and application of protectant ointment to the stoma, antibiotics, and modification of the diet. Herthel (1974) used a similar surgical technique with success in treating a grade 4 rectal tear. In both reports, atrophy of the distal stump is mentioned as a complication which makes closure difficult. Herthel (1974) stated that it was necessary to leave 60 cm of bowel proximal to the rectal tear to facilitate closure of the colostomy. Azzie (1975) and Spiers et a1 (1980), in treating horses with rectal tears, performed loop colostomies. Azzie (1975) treated 2 cases, with Case 2 dying from haemorrhage associated with prolapse of the colostomy. In order to avoid this, nylon tape tension sutures were placed in the skin across the colostomy site in Case 1. In the latter, no prolapse occurred and treatment was successful. Spiers et a1 (1980) performed loop colostomies on 3 horses. Two of the 3 had complications associated with the colostomies (severe mucosal oedema at the colostomy site, with herniation through the colostomy stoma in one case and mucosal prolapse in another). Mucosal prolapse was the major complication associated with colostomy in this mare. Birnbaum and Ferrier (1952) list 7 factors involved in the prolapse of a colostomy. Of these factors, an excessively large incision in the abdominal wall was considered the most important in this case. However, the relative contribution of other factors is difficult to determine in the absence of sufficient case reports in horses. Saha et a1 (1973) stated that in man, a colostomy wound which admits more than 2 fingers is liable to prolapse. We are not able to state with certainty, but it is possible that mucosal prolapse is more likely to occur in horses than man. If this is true then further work is required to develop optimal techniques for the creation of colostomies in horses. The progression from acute, diffuse peritonitis on presentation to a localised, fibrinous peritonitis at necropsy and the absence of faeces passed via the anus confirmed that the colostomy diverted faecal material from the anastomosis. At necropsy intestinal obstruction, secondary to entrapment of a loop of small intestinal by fibrin in the region of the anastomosis, was the cause of acute abdominal pain, which necessitated destruction of the animal. Of the factors listed by Peacock (1984) infection was regarded as the major cause of adhesions. Thus, a colostomy performed earlier may have reduced the tendency to form adhesions. In conclusion, the prognosis for types 3 and 4 rectal prolapse remains poor. Successful treatment relies on an intact vascular supply to the remaining small colon and rectum following resection and anastomosis of the compromised portion. In addition, colostomies will be useful in the short term to divert faecal material from compromised bowel or to enable healing of the anastomosis, once techniques for creation and management are improved. The authors wish to thank K G Johnson and M A Brownlow for assistance in preparation of the manuscript and E A Laing for performing the necropsy.