Successful management of jejunojejunal anastomosis dehiscence by extra-abdominal exteriorization and bandaging in a cat with septic peritonitis

Abstract

Duodenal and jejunal resections were performed in a cat with septic peritonitis due to small intestinal perforations by a linear foreign body. Three days later jejunal resection and anastomosis were repeated due to dehiscence of the anastomosis site. This segment of intestine was exteriorized through the body wall and managed with bandages for 5 days before it was surgically replaced into the abdomen. The cat made a full recovery.

Subcutaneous pouch exteriorization or externalization of colonic anastomoses has been widely reported in human surgery, mainly as an alternative to temporary colostomy, which is used in the unprepared colon during emergency management of serious colonic trauma (1–3). A similar subcutaneous pouch technique of exteriorized anastomosis was recently described in a dog for the management of a dehisced ileocolic anastomosis in order to facilitate wound healing and improve prognosis (4). Linear foreign bodies (LFBs) are common emergency surgical conditions of the gastrointestinal tract in cats. Linear foreign bodies are often wrapped around the base of the tongue or fixed in the pylorus resulting in intestinal pleating and partial intestinal obstruction (5,6). Intestinal perforation and secondary septic peritonitis or permanent fibrosis are serious complications associated with LFBs (5–7). In these cases intestinal resection and anastomosis is required to improve prognosis (7). Pre-existing peritonitis at the time of initial intestinal resection and anastomosis in cats does have a negative effect on intestinal healing and may result in anastomotic dehiscence (5,8,9).

To the authors’ knowledge this is the first report describing an exteriorized intestinal anastomosis in a cat. The purpose of this study is to report a successful exteriorization of a jejunojejunal anastomosis in a cat following dehiscence associated with peritonitis due to a linear foreign body.

Case description

A 6-month-old, male domestic short hair cat weighing 2.5 kg was presented with a history of anorexia, depression, and intermittent vomiting of 6 d duration. No improvement was noted following intravenous fluid [normal saline solution at 2 mL/kg body weight (BW) per hour] or metoclopramide antiemetic therapy. On physical examination the cat was underweight, had a heart rate of 200 beats/min, rectal temperature of 39°C, and a thread was found around the base of the tongue. No further abnormalities were detected on physical examination. Laboratory investigation revealed neutrophilic leukocytosis [29 500/μL; reference range (RR): 3000 to 13 400/μL] and azotemia (blood urea nitrogen 22.8 mmol/L; RR: 3.2 to 11.4 mmol/L). Abdominal radiography showed mild distention of intestinal loops with no gas/fluid patterns present; no plication or displacement of the intestines was identified. Loss of serosal detail indicating a peritoneal effusion was also detected. Abdominal ultrasound confirmed the presence of peritoneal effusion. Turbid fluid was collected by abdominocentesis and submitted for cytological analysis, which identified degenerate neutrophils and intracellular and extracellular bacteria. The abdominal fluid was also submitted for culture and sensitivity and yielded Escherichia coli. A diagnosis of intestinal obstruction and septic peritonitis associated with a linear foreign body was made.

Following stabilization with IV fluids (normal saline 0.9%, 4 mL/kg BW per hour), the cat was premedicated with acepromazine (Acepromazine maleate; Boehringer Ingelheim Vetmedica, Ellinico, Greece) 0.04 mg/kg BW, IM, and pethidine (Pethidine hydrochloride; Molteni, Florence, Italy), 4 mg/kg BW, IM, and anesthetized with propofol (Propofol; Fresenius, Bad Homburg, Germany), 5 mg/kg BW, 5 mg/kg IV and 2% isoflurane (Isoflurane-vet; Merial, Harlow, Essex, UK) in oxygen. Cefazolin (Vifazolin; Vianex, Patras, Greece), 20 mg/kg BW, IV was given at induction of anesthesia and continued after surgery every 8 h for 2 wk. A ventral midline celiotomy extending from xiphoid to pubis was performed. A moderate amount of fluid was suctioned from the abdominal cavity and marked plication of the small intestine was noted. Multiple perforations were visible in the mesenteric wall of the distal duodenum and jejunum associated with a thread emerging through them. A 3-cm length of the distal duodenum was removed and a duodeno-duodenal anastomosis was performed using 2 continuous 4/0 polydioxanone sutures. A 13-cm length of mid jejunum was also resected and a jejunojejunal anastomosis was performed in a similar fashion. Overall 70% of the small intestine was resected. Following completion of the anastomosis normal saline was infused through the intestine to test for leakage. The abdomen was lavaged with 2 L of warmed normal saline (0.9%), the anastomotic sites were omentalized and the celiotomy was closed with a simple continuous 3-0 polydioxanone suture. Subcutis and skin were closed in a routine manner. Postoperative analgesia was provided with pethidine 2 mg/kg IM every 2 h for 16 h. Canned food (i/d; Hill’s Pet Nutrition Watford, UK) was offered 24 h after surgery.

A slowly progressive increase in appetite and diarrhea were noticed 3 d after surgery. Anorexia, pain on abdominal palpation, and pyrexia (body temperature 40.5°C) during the third post-operative day were highly suggestive of a complicated intestinal healing. An abdominal ultrasound revealed a small amount of effusion. Cytological analysis of fluid collected by abdominocentesis showed numerous intra- and extra-cellular cocci and degenerate neutrophils. Septic peritonitis related to intestinal leakage was diagnosed and an emergency celiotomy was done. The cat had a heart rate of 200 beats/min, mean blood pressure of 70 mmHg, and BUN of 10.7 mmol/L. The cat was stabilized with Lactated Ringer’s solution (4 mL/kg BW per hour) and was premedicated with dexmedetomidine (Dexdomitor; Zoetis, Sao Paulo, Brazil), 10 μg/kg BW, IM, pethidine (Molteni), 4 mg/kg BW, IM, and ketamine (Imalgene; Merial, Lyon, France) 10 mg/kg BW, IM. Anesthesia was induced with propofol (Fresenius), 4 mg/kg BW, IV, and maintained with 2% isoflurane. A midline celiotomy through the previous surgical incision was performed and a moderate amount of greenish fluid was observed upon entering the abdomen. Careful inspection revealed that the duodeno-duodenal anastomosis was healing normally but the jejunojejunal anastomosis showed dehiscence and was leaking intestinal contents. The dehisced portion was resected with 2 cm margins of healthy tissue and an end to end anastomosis was made as in the previous surgery. The abdomen was lavaged with 2 L of warm normal saline (0.9%). A 3-cm paramedian incision through the abdominal wall was made 5 cm lateral to and left of the midline. A 4-cm intestinal loop was exteriorized. The abdominal wall incision was partially closed with 2 simple interrupted 3/0 polydioxanone sutures to accommodate the exteriorized intestinal portion and to avoid strangulation of the exteriorized intestine and prevent self-reduction of the loop into the abdominal cavity. The more proximal of the 2 interrupted sutures used to close the paramedian abdominal incision passed through the seromuscular portion of the exteriorized intestinal loop to secure it outside the abdominal wall. The intestinal anastomosis was then exteriorized through a 3-cm incision in the skin. A sterile nonadherent dressing (Melolin, Smith and Nephew, Hull, UK) soaked in normal saline was applied around the exteriorized intestine. Wet to dry laparotomy sponges were used to completely cover the exteriorized intestine followed by a sterile disposable napkin to keep the intestinal loop moist. The celiotomy was closed routinely and the abdominal wall was covered with sterile bandages. A sterile urinary catheter attached to a closed collection system was placed to prevent urine contamination of the bandage. The cat made a good recovery from anesthesia and was fed orally the day after surgery with a small quantity of canned food (i/d; Hills). Blood for hematology and serum biochemistry analysis was obtained every 2 d and blood pressure and urine output were monitored. Cefazoline was continued based on susceptibility of the E. coli recovered from the abdominal fluid. The bandage was changed under propofol anesthesia (5 mg/kg BW, IV) 24 h after surgery to allow assessment of the anastomosis and placement of new laparotomy sponges. The skin of the surrounding abdominal region was surgically prepared before every bandage change. This procedure was repeated daily for the next 5 d and showed that the anastomosis had healed uneventfully (Figure 1). The cat’s appetite improved progressively and the consistency of the feces changed from liquid to semi-solid. Complete blood cell count and cytology of fluid collected through abdominocentesis 5 d after the second surgery showed no abnormalities. The exteriorized intestine was reduced into the abdominal cavity 5 d after the second surgery. Reduction was performed under the previously described anesthetic protocol. The anastomosis site was reduced following removal of the abdominal wall sutures, the abdominal cavity was lavaged with normal saline solution via the intestinal exit wound and the abdominal wall was closed routinely in 3 layers. Antibiotic therapy was continued and analgesia was provided with meloxicam (Metacam; Boehringer Ingelheim), 0.1 mg/kg BW, IV, q24h. The cat was discharged 6 d after the second surgery. Oral cefuroxime (Zinadol; GlaxoSmithKline, Chalandri, Greece), 20 mg/kg BW, q8h was prescribed for 10 d.

Figure 1.

The exteriorized anastomosis on day 5 just before reduction and closure of the abdominal incision. The anastomosis site shows uncomplicated healing.

On re-examination at about 40 d after surgery the cat was in good physical condition and had loose feces. Because the loose feces could be due to possible short bowel syndrome, oral ursodeoxycholic acid (Ursofalk; Galenica, Kifisia, Greece), 15 mg/kg BW per day was prescribed for 15 d. Eleven months after surgery the cat remained well and had soft but formed feces.

Discussion

In longstanding cases, as the fixed LFB is forced against the intestinal wall, mucosal edema and ischemia in the mesenteric border of the intestines may be created and the LFB may eventually cut through at several sites resulting in perforations, leakage of intestinal contents, and septic peritonitis (5,7,10). Affected cats are best managed with extensive intestinal resection and anastomosis (11). In the case presented here, 2 anastomoses were performed to avoid complications of a more extensive resection (11,12). However, decreased survival was reported in cats that had more than 1 intestinal procedure (13). Septic peritonitis secondary to intestinal perforation associated with LFBs has a 50% mortality rate in cats (5). In our case septic peritonitis was managed with intestinal resection, thorough peritoneal lavage, primary abdominal closure, and appropriate antibiotics. Primary closure is an acceptable method for the management of septic peritonitis in cats and dogs (14–16). Intestinal dehiscence and secondary septic peritonitis are the most serious complications following intestinal surgery in cats (9). Risk factors for anastomotic dehiscence and leakage include pre-existing peritonitis, intestinal foreign body, and serum albumin concentration ≤ 25 g/L (8). Anastomotic dehiscence and mortality rates in cats following intestinal surgery seem to be lower than in dogs (13). In our cat dehiscence in 1 of the 2 anastomoses may have been associated with pre-existing peritonitis, as serum albumin concentration was within normal limits (5).

In the study reported here the decision to proceed with extra-abdominal intestinal exteriorization was based on the prevention of re-dehiscence associated with the septic environment of pre-existing peritonitis, the increased mortality related to abdominal sepsis and the higher cost of other treatment options including abdominal drainage with closed-suction drains or vacuum-assisted peritoneal drainage (4,17,18). An intestinal anastomosis has been successfully exteriorized and placed in a subcutaneous pouch in a dog with an intestinal foreign body and septic peritonitis and was performed in this case as an alternative to the conventional intraperitoneal healing (4,19). Healing of jejunojejunostomies is impaired in an extraperitoneal position compared to an intra-abdominal position and this is attributed to the beneficial effect of mesothelium (19). In another study, the bursting pressure of omentalized extraperitoneal anastomoses was increased compared to non-omentalized anastomoses 3 d after surgery (20). Although the exteriorized jejunojejunal anastomosis described here was not omentalized, exteriorization provided a non-septic environment for intestinal healing and allowed daily monitoring of the anastomosis by direct visualization and palpation. Healing of the enterotomy is weakest at the end of the lag of the inflammatory phase of healing (0 to 4 d) and the time after surgery until dehiscence is usually 3 to 5 d (21). Our intestinal anastomosis loop was maintained exteriorized for 5 d and reduction to the abdominal cavity was based on the appearance of the anastomosis and resolution of peritonitis. Our technique was different from the one previously reported (4); the anastomosis was not exteriorized into a subcutaneous pocket to prevent adhesion formation but in a position over the skin and completely covered with moistened laparotomy sponges and bandages to avoid exposure serositis and contamination from the surrounding skin surface. The technique used in the present study is not much different from open abdominal drainage in which the abdomen remains open using a continuous loose suture placed in the external sheath of the rectus muscle. The abdominal wall in our case was closed with sutures leaving a gap to accommodate the exteriorized intestinal loop and there was a communication with the external environment so abdominal effusion would have been draining. Potential limitations of the technique described here include increased morbidity associated with open peritoneal drainage, daily general anesthesia required for bandage changes of the exteriorized intestine, the risk of eventration of intestine from the abdominal cavity, and the bandage removal and self-traumatization of the exteriorized intestine (4). A prospective study should be performed to compare this technique with others for the management of intestinal anastomosis failures in cats due to a septic environment.

Extensive small intestinal resection in 7 cats ranging from 70% to 90% has been reported to result in persistent diarrhea and has been compatible with short bowel syndrome in 2 cats (11). However, minimal information was available concerning the extent of intestinal resection in these 2 cats (11). In an experimental study in normal cats, in which 85% of the small intestine including all jejunum and most of the ileum was resected, severe diarrhea and weight loss was reported, signs compatible with short bowel syndrome (12). Administration of ursodeoxycholic acid in these cats improved recovery by decreasing intestinal transit time (12). In the present study, the cat had 70% of the small intestine resected, including part of the duodenum and jejunum, and kept producing soft but formed feces 10 mo after surgery.

This is the first description of a successful exteriorization of an intestinal anastomosis outside the abdomen to manage a dehiscence associated with septic peritonitis due to a LFB in a cat. This technique might be considered as an alternative for the management of intestinal anastomoses in a septic environment. CVJ