Bacteremia due to clostridium difficile: case report and review of the literature.

OBJECTIVE
The purpose of this study is to report a case of C. difficile bacteremia in a Crohn's disease patient and to review the literature on previously reported cases.


METHODS
Searches of MEDLINE and PubMed databases were made.


RESULTS
We report the first case of C. difficile bacteremia in a Crohn's disease patient. There are 15 other reported cases of C. difficile bacteremia reported in the literature. We found that the majority of patients (10 of 15 patients) had polymicrobial bacteremia and that the overall mortality rate is significant, with 6 of 15 reported patients dying.


CONCLUSION
In conclusion, we find that C. difficile bacteremia is associated with a significant mortality rate and it would seem prudent to consider aggressive antibiotic therapy.


Introduction
Clostridium diffi cile is the primary cause of pseudomembranous colitis and a major cause of antibioticassociated diarrhea. 1 In the original report of C. diffi cile published in 1935 the bacterium was named "the diffi cult clostridium" because early attempts at isolation were unsuccessful and it grew slowly in culture. 3 C. diffi cile produces an enterotoxin (toxin A) and a cytotoxin (toxin B). Toxin A has been shown to be the cause of diarrhea and pseudomembranous colitis. 4 C. diffi cile has rarely been reported to cause extraintestinal disease. 1 The role of toxins A and B in extracolonic manifestations of C. diffi cile remains unclear. We report a case of C. diffi cile bacteremia in a Crohn's disease patient and review the literature on previously reported cases.

Methods
A review of the published literature on C. diffi cile bacteremia was done using MEDLINE and PubMed databases. Searches were conducted to fi nd articles from 1966-2008. Medical subject headings used to search the databases included C. diffi cile, including subheadings of bacteremia, extraintestinal disease and Crohn's disease, as well as a keyword search using "C. diffi cile bacteremia." Titles and abstracts of potentially relevant articles were reviewed by a single author.

Case Report
We describe the case of a 50-year-old white male with small bowel Crohn's disease initially admitted with nausea and abdominal distention secondary to a small bowel obstruction. The patient has a 30 year history of Crohn's disease involving the jejunum and terminal ileum with multiple proximal small bowel strictures. He has had an appendectomy and back surgery in the past but was never treated surgically for his Crohn's disease.
He was on maintenance therapy with Azulfi dine and Azathioprine. The patient was started on Infl iximab in November 2005 after multiple admissions for small bowel obstruction. The patient was changed to Adalimumab in May 2007 for patient convenience and diffi culty related to obtaining regular intravenous access. He denied any recent antibiotic use.
Computerized tomography scanning on admission demonstrated a small bowel obstruction with thickened and edematous small bowel in the right lower quadrant 8-10 cm from the ileocecal valve with a small amount of ascites and no evidence of abscess. A nasogastric tube was placed for decompression and the patient was placed on solumedrol 20 mg intravenous (iv) every eight hours along with aggressive iv hydration and pain management with hydromorphone. The patient initially improved on hospital day number one. On the morning of hospital day number 2, the patient was reported to be febrile to 39.4 ο C and tachycardic. The patient was complaining of increased pain and on exam had significantly increased tenderness with absent bowel sounds. At that time, the blood culture drawn on admission was reported as growing Escherichia coli, Enterococcus fecalis and Klebsiella oxytoca. The patient was started on intravenous antibiotics (ampicillin/sulbactam and gentamicin) and taken for an emergent laparotomy. He was found to have a perforation with a free abdominal abscess and a partial small bowel obstruction of the jejunum. The patient underwent a small bowel resection with jejunojejunal anastamosis and a right hemicolectomy with ileocolonic anastamosis and ileostomy. The pathology revealed a T4N1 poorly differentiated adenocarcinoma of the jejunum. The patient did well clinically post-op however a routine blood culture drawn for fever on post-op day number one grew Clostridium diffi cile. The patient denied signifi cant diarrhea. Subsequent stool studies sent for Clostridium diffi cile toxins A/B were negative. At that time, the patient had received four days of antibiotics. The patient was maintained on pipercillin-tazobactam. All other follow-op blood cultures were unremarkable. The patient received a total of 21 days of intravenous antibiotics (6 days of ampicillin/sulbactam and gentamicin followed by 15 days of pipercillintazobactam). The remainder of his post-op course was unremarkable and he made a full recovery.

Discussion
We report the fi rst case of C. diffi cile bacteremia in a Crohn's disease patient. There are 15 other reported cases of C. diffi cile bacteremia reported in the literature that are summarized in Table 1.
The prevailing theory regarding the pathophysiology of C. diffi cile bacteremia is that the colonic wall infl ammation associated with pseudomembranous colitis permits transient bacteremia to develop. The overall mortality rate is signifi cant, with 6 of 15 reported patients dying. In terms of the demographics, 11 of the 15 patients were male and the age range was from neonate to age 69. A high proportion (4 of the 15 patients) had an underlying malignancy. Unfortunately, C. diffi cile stool toxin was sent in only 7 of the patients. The stool toxin was positive in 5 out of the 7 patients. C. diffi cile Associated Diarrhea (CDAD) was reported in 6 out of the 7 patients. Two out of the 5 patients with positive stool toxin died. The majority of patients (10 of 15 patients) had polymicrobial bacteremia. Four of the reported cases occurred postoperatively.
Recent antibiotic use was found to be a signifi cant risk factor. Antibiotic use leads to an alteration of the intestinal microfl ora, leading to overgrowth of endogenous C. diffi cile or allowing colonization by nosocomial C. diffi cile. Only 12 of the case reports comment on antibiotic exposure (9 of 12 patients had antibiotic exposure). Cephalosporins were the most common class of antibiotics that these patients were exposed to.
Information on therapy is available on 11 of the patients. The activity of various drugs against C. diffi cile according to the Manual of Clinical Microbiology is summarized in Table 2. Among the cases reviewed, 4 were treated with metronidazole (2 of which died). The specifi cation of oral versus intravenous therapy was incomplete. It is important to note that the two patients who died were both neutropenic leukemic patients. There were 7 patients that were treated (at least in part) with vancomycin and they all survived except one. Two of these patients were treated exclusively with oral vancomycin and they both survived. One of the patients was treated only with intravenous vancomycin and survived. There was also one patient treated with both oral and intravenous vancomycin who survived. The remaining three patients were treated with a regimen that included other antibiotics. In our case, the patient was successfully treated with pipercillin-tazobactam.
C. diffi cile is a ubiquitous organism that can be found in the environment and exposure to the organism is common. It has been estimated that 15%-25% of adults become colonized after admission to the hospital. 16 There is also growing literature to support a strong link between inflammatory bowel disease and C. difficile infection. Previous studies have demonstrated that 5%-20% of patients admitted with an IBD fl are will have C. diffi cile infection. 16 The growing literature support for the link between IBD fl ares and C. diffi cile infection along with the signifi cant mortality associated with C. diffi cile bacteremia highlight the importance of this topic. In conclusion, we fi nd that C. diffi cile bacteremia is associated with a signifi cant mortality rate. C. difficile Associated Diarrhea (CDAD) was reported in 6 out of the 15 patients. Not surprisingly, the majority of patients had recent antibiotic exposure. We found a high proportion of patients were male. We also found that the majority of patients had a polymicrobial bacteremia. Therefore, it is unclear if C. diffi cile is the primary pathogen. In terms of treatment, it would seem prudent to consider aggressive antibiotic therapy given the high mortality rate.