The loss of fluid and electrolytes from a high-output ileostomy (>1200 ml/day) can quickly result in dehydration and if not properly managed may cause acute renal failure. The management of a high-output ileostomy is based upon three principles: correction of electrolyte disturbance and fluid balance, pharmacological reduction of ileostomy output, and treatment of any underlying identifiable cause. There is an increasing body of evidence to suggest that Clostridium difficile may behave pathologically in the small intestine producing a spectrum of enteritis that mirrors the well-recognised colonic disease manifestation. Clinically this can range from high-output ileostomy to fulminant enteritis. This report describes two cases of high-output ileostomy associated with enteric C difficile infection and proposes that the management algorithm of a high-output ileostomy should include exclusion of small bowel C difficile.
Statistics from Altmetric.com
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.
Seven to ten days following the formation of a new ileostomy, ileal adaptation increases the absorptive capacity of the ileum for water, reducing and thickening the effluent output. However, a high-output ileostomy, with effluent volumes persistently in excess of 1200 ml/day may result from a short bowel, intra-abdominal sepsis, enteritis (eg, Crohn disease or radiation enteritis), incomplete intestinal obstruction, or mineralocorticoid deficiency.1 A high-output stoma is a more common entity than it is widely acknowledged, complicating up to 17% of newly fashioned ileostomies.2 Management is multimodal with exclusion of identifiable causes, antimotility and antisecretory drugs, restriction of hypotonic oral fluids and replacement with glucose–saline solutions.1
In recent years, the frequency and severity of Clostridium difficile infection has been on the increase worldwide.3 Infection control protocols devised to combat the C difficile “epidemic” have increased the awareness of healthcare workers to the importance of infection by the bacterium in patients with diarrhoea. However, small bowel enteritis secondary to C difficile infection is not widely acknowledged and therefore the potential for C difficile to produce a high-output ileostomy may not be recognised. We report two cases of C difficile enteritis presenting with high-output ileostomies, and review the literature supporting an emerging role for C difficile as a cause of high-output ileostomy.
A 65-year-old woman with quiescent Crohn disease (medication free) was admitted with peritonitis. At laparotomy, she was found to have perforated a gangrenous small bowel loop, which had volved around a band adhesion. There was extensive peritoneal soiling, and therefore a small bowel resection was performed with defunctioning ileostomy (170 cm distal to the duodenojejunal flexure) and mucus fistula (60 cm from ileocaecal junction). Following this she was administered a 5-day course of intravenous cefuroxime and metronidazole. Eight days after surgery she was noted to have a high-output ileostomy (fig 1) that was managed with a combination of loperamide, codeine phosphate and oral fluid restriction, but ultimately required total parenteral nutrition. Stool samples were sent for C difficile ELISA analysis and the following day the presence of C difficile toxin in the ileostomy effluent was confirmed and treated with a 2-week course of oral metronidazole. Following treatment her ileostomy, output decreased and subsequent testing of ileostomy effluent was negative for C difficile toxin.
A 43-year-old immunosuppressed woman, taking steroids and azothioprine for active Crohn disease, underwent an emergency laparotomy for perforation of the distal ileum proximal to a terminal ileal stricture with widespread peritoneal soiling. A right hemicolectomy with terminal ileal resection was performed with formation of an ileostomy (300 cm distal to the duodenojejunal flexure) and mucus fistula. Following surgery she developed severe sepsis that was treated with intravenous cefuroxime and metronidazole, and multiorgan dysfunction requiring prolonged inotropic support, ventilation and enteral feeding. Eleven days after surgery she developed a purulent high-output ileostomy from which C difficile toxin was isolated. The following day she was started on a 10-day course of oral metronidazole, administered via a nasogastric tube. The stoma output remained high but normalised with a subsequent course of oral vancomycin (fig 2) and further samples of ileal effluent tested negative for C difficile toxin.
We present two cases of a high-output ileostomy associated with enteric C difficile infection. The concept of small bowel colonisation/infection with C difficile is not new. In 1973 a study by Vince et al, to establish the development of ileostomy flora, isolated clostridia from the ileal effluent of two of the three patients studied.4 Furthermore, in 1986 Testore et al postulated that the human jejunum may act as a reservoir for C difficile, when the bacterium was isolated from 3% (3/100) of jejunal specimens taken postmortem from patients who had died without diarrhoea or gastrointestinal symptoms.5
The isolation of C difficile from human small bowel alone is insufficient to support a pathogenic role for the organism at this anatomical site. However, the ability of gastrointestinal pathogens such as C difficile to adhere to the intestinal epithelium is crucial for infection of the gastrointestinal tract.6 This was elegantly established with flow cytometry studies by Drudy et al when they demonstrated the adherence of C difficile to primary and cultured human small intestinal epithelial cells in vitro.6 Other in vitro studies have confirmed the ability of C difficile to precipitate inflammatory small bowel enteritis in infected laboratory mammals. Lyerly et al demonstrated mucosal inflammation and the accumulation of haemorrhagic fluid in the caecum and small intestine of hamsters and mice following intragastic administration of C difficile toxin A.7 In a perfused rabbit ileal loop model, C difficile toxin A has been shown to induce severe inflammatory enteritis, with toxin exposure resulting in increased epithelial permeability, neutrophil infiltration, epithelial cell necrosis and destruction of villous architecture.8 Furthermore, as in humans, the use of antibiotics in rabbits appears to facilitate the proliferation of C difficile. In a study by Guandalini et al, treatment with oral ampicillin resulted in colonic C difficile colonisation and diarrhoea in 14 of 25 rabbits, eight of which also had ileal C difficile.9
There is clinical evidence to support the ability of C difficile to colonise the small bowel and in vitro evidence of inflammation and damage to the small bowel that mirrors colonic C difficile infection. To date there have been 22 reported clinical cases of small bowel C difficile infection. Eighteen cases followed abdominal surgery and in 20 cases patients had antibiotics administered prior to the diagnosis of enteric C difficile. The overall mortality for reported cases of small bowel C difficile is 45.5% (10/22) although 13 patients had severe disease with sepsis or organ dysfunction. In contrast, seven cases report high-output ileostomy in which treatment with either metronidazole or vancomycin resulted in resolution of symptoms, a feature in common with the two cases reported here.
Clinical clostridial infection most frequently arises following the use of antibiotics, particularly cephalosporins and penicillin derivatives.
Clostridium difficile infection rates have increased in recent years with the emergence of more virulent strains
C difficile has been identified as a small bowel commensal but can also behave pathologically producing a small bowel enteritis.
Enteral C difficile infection has been identified as a cause for high-output ileostomy.
The management algorithm for high-output ileostomy should now include exclusion and treatment of enteral C difficile.
The pathogenesis of C difficile enteritis is not understood, but it is likely to involve mucosal damage and inflammation mediated by enterotoxins produced during the proliferation of C difficile similar to that seen with colitis. Bacteria such as lactobacillus and group D enterococci are known to antagonise the proliferation of C difficile, and eradication of these organisms by antibiotics is thought to play a significant role in the subsequent development of colonic C difficile infection, diarrhoea and colitis.10 Most antibiotics in common usage have, at some stage, been implicated in causing C difficile diarrhoea.10 Traditionally this has most frequently followed treatment with clindamycin or ampicillin (and other derivatives of penicillin), but recently has been more frequently associated with the use of cephalosporins; this trend is likely to mirror changes in prescribing practice.11 Antibiotic usage was prevalent prior to the development of enteric C difficile in 20 of the 22 reported cases and therefore appears to be important in the development of colonic and enteric C difficile. Previous gastrointestinal surgery was also common among reported cases of small bowel C difficile infection and this too has been identified as a risk factor for colonic C difficile.11
There has been increased reportage of small bowel C difficile infection, with nine cases reported in the last year. The recent emergence of more virulent strains and epidemic outbreaks of C difficile may have resulted in the apparent rise in enteric C difficile. Alternatively, it could simply reflect an increase in awareness and testing for small bowel C difficile. Whichever is the case, it is our belief that in the current climate, treatment for a high-output ileostomy should now include the detection and, if present, treatment of enteric C difficile.
Competing interests None.
Patient consent Obtained.
Provenance and peer review Not commissioned; externally peer reviewed.