Case Studies

56 GHS Proc. May 2016; 1 (1): 56-5956

Clostridium difficile is an opportunis-tic pathogen that multiplies when more benign intestinal flora is lacking and pro-

duces a damaging, intensely inflammatory toxin capable of transmural permeation.1 C difficile colitis (CDC) has rapidly increased in incidence over recent decades and is now the most common cause of nosocomial diarrhea.2 It complicates the hospital stays of an increasingly frail geriatric population, frequently resulting in septic shock, organ dysfunction, and even death, despite max-imal medical management. Thus surgical therapy remains the mainstay of salvage intervention. However, after failing various pharmaceuticals, many patients have deteriorated into multi-sys-tem organ dysfunction, rendering them less than ideal candidates for the conventional operation: an extensive, highly morbid total abdominal col-ectomy with permanent end-ileostomy. In 2011, Neal and colleagues at the University of Pitts-burgh presented a less drastic alternative: mini-mally invasive creation of a loop ileostomy to allow for antegrade large-volume polyethylene glycol (PEG) and vancomycin enemas, accompanied by systemic intravenous metronidazole.3 Patients in that study undergoing this gentler option exhib-ited less than half the mortality of those receiving colectomy in short-term follow-up. As a result, for many general surgeons, temporary laparoscopic loop ileostomy with colonic lavage has become the preferred first option in a staged salvage approach

to refractory CDC, reserving colectomy only for those that fail. We ourselves have found success creating many such loop ileostomies for fulmi-nant CDC; however, we have since encountered a devastating potential complication of reversing these stomas: recurrent C difficile invading the small bowel, otherwise known as Clostridium dif-ficile enteritis (CDE), an infection associated with overwhelming sepsis and a mortality of 30%.4

Case DescriptionGC was an 81-year-old Caucasian man with a his-tory of coronary artery disease, dyslipidemia, and gastroesophageal reflux, for which he took multi-ple medications, including proton pump inhibi-tors. He was in otherwise good health and highly functioning, living at home with his spouse. He presented to his primary care physician com-plaining of unilateral facial tenderness and swell-ing, was given a diagnosis of parotiditis, and completed an empiric course of clindamycin. His symptoms, however, did not resolve, and further work-up ultimately revealed a malignant parotid tumor, likely metastatic melanoma.

In the interim, the patient developed diarrhea that persisted despite a 3-week course of oral ciprofloxacin and metronidazole at home, neces-sitating hospital admission for dehydration with acute kidney injury. Highly sensitive polymerase chain reaction (PCR) testing (Xpert C difficile/

Fatal Clostridium difficile Enteritis as a Delayed Complication of Loop Ileostomy for Fulminant C difficile ColitisSummer N. Rochester, DO; Robert Farrar, MD; R. Jared Sanders, MD; Megan K. Straughan, MD; and Meghann L. Kaiser, MD

From the Department of Surgery, Greenville Health System, Greenville, SC (S.N.R., R.J.S., M.K.S, M.L.K.), and Department of Pathology, Greenville Health System, Greenville, SC (R.F.)

AbstractThe recent introduction of loop ileostomy with colonic lavage represents a powerful surgical inter-vention for fulminant Clostridium difficile colitis, but is not without potential hazards. We present a case of fatal recurrent C difficile small bowel enteritis, a heretofore unreported potential complication of ileostomy reversal. A review of the literature suggests potential mechanisms and means to prevent this devastating outcome.

57

RECURRANT FATAL C DIFFICLE AFTER ILEOSTOMY TAKEDOWN

GHS Proc. May 2016; 1 (1): 56-59 57

Epi, Cepheid Inc, Sunnyvale, Calif.) revealed a particularly virulent B1/NAP/027 strain of C difficile. He was initially treated with systemic intravenous metronidazole as well as vancomycin per oral and rectum. Nonetheless, his condition continued to deteriorate, and surgical consulta-tion was sought. The computed tomography (CT) scan obtained was suggestive of severe CDC (Fig. 1). After lengthy discussion, taking into account the patient’s advanced age, comorbidities, and worsening clinical status, the patient elected to proceed with emergent loop ileostomy on hospital day 3. The colon was closely inspected laparoscop-ically and found to be boggy and congested but otherwise viable. We accordingly proceeded with laparoscopic creation of loop ileostomy within a few centimeters of the ileocecal valve, allowing for easy cannulation, through which large-volume PEG and a 10-day course of vancomycin colonic flushes were administered, along with systemic intravenous metronidazole. Despite a complicated 30-day hospital course, the patient did exhibit complete clinical resolution of CDC and was dis-charged to a skilled nursing facility. Infectious disease consultants advised he be maintained on a regimen of enteric cholestyramine, to bind and inactivate C difficile toxin, and Saccharomy-ces boulardii probiotic supplements, to maintain healthy colonic flora. Repeat PCR testing of rectal output revealed no further evidence of C difficile.

After undergoing radical parotidectomy, compli-cated by a superficial wound infection, the patient was treated with a course of cephalexin and expe-rienced no return of diarrhea or other gastroin-testinal complaints. Multiple discussions were had with both patient and family throughout this period. The patient acknowledged that ileostomy reversal was not medically necessary; neverthe-less, he felt the stoma adversely affected his quality of life and ardently desired reversal. Five months after initial ileostomy creation, and 1 month fol-lowing his parotidectomy, the patient underwent elective ileostomy reversal. One gram of prophy-lactic cefoxitin was administered preoperatively, and a stapled anastomosis was created without incident. The patient was extubated immediately thereafter and transferred to a monitored floor bed. By postoperative day (POD) 2 the patient was tolerating an oral diet and ambulating with-out complaints. On POD3, however, he acutely decompensated into respiratory distress with frank peritonitis. Broad-spectrum antibiotics, including intravenous metronidazole, were initi-ated. CT (Fig. 2) was concerning for possible small and large bowel ischemia, and the patient returned

Figure 1Initial CT scan showing edematous, fluid-filled colon (white arrow) and juxtaposed normal-appearing small bowel (black arrow), consistent with diagnosis of C difficile colitis.

Figure 2CT scan showing edematous fluid-filled loops of both small and large bowel.

to the operating room for emergent exploratory laparotomy. Once again, the colon was noted to be boggy and friable, as was the ileum. The anasto-mosis was intact without biliary staining, and the mesenteric vessels surveyed via Doppler trans-mitted strong signals throughout. With no other obvious explanation for the patient’s clinical find-ings, we elected to proceed with resection of the involved bowel, which entailed a total abdominal colectomy and partial distal small bowel entrec-tomy. The specimens grossly exhibited mucosal pseudomembranes with cobblestoning not only throughout the colon but also extending a signif-icant distance proximally past our anastomoses and into the resected ileal portion (Fig. 3, Page 58). Formalin fixation prevented definitive testing for the C difficile toxin. However, vessels were noted to be patent, making primary ischemia an unlikely explanation. Multiple pathologists judged CDE the most likely pathology (Fig. 4, Page 58).

Intraoperatively the patient continued to deterio-rate, and we elected to leave him in discontinuity with an open abdomen to expedite aggressive fluid resuscitation in the intensive care unit. Overnight, despite heroic efforts, including vancomycin per rectum and nasogastric tube, severe shock unre-

58 GHS Proc. May 2016; 1 (1):56-59

sponsive to fluids, pressors, and inotropes ensued. The family elected to withdraw support in accor-dance with the patient’s previously expressed wishes, and he expired shortly thereafter on POD4.

DiscussionCDE was once felt to be a rare entity with dis-mal prognosis, affecting primarily postcolectomy patients suffering from inflammatory bowel dis-ease (IBD). Between 1980 and 2000, only 9 cases were reported in the literature.2 Subsequently, a relative multitude have erupted, many associated with neither IBD nor colectomy.4 This exponen-tial rise is likely the product of heightened clinical suspicion, coupled with the prevalence of several factors now recognized to increase susceptibility, including advanced age, immunosuppression, recent antibiotic exposure, hospitalization, white race, gastric acid suppression, and the B1/NAP/027 strain of C difficile.5 Our patient exhibited many of these risk factors, but the last 2 may be particu-larly damaging. Clinically significant infection of the small bowel likely requires a greater burden of

Figure 3Lumen of the excised terminal ileum, revealing cobblestoning and pseudomembranes, consistent with a diagnosis of C difficile enteritis.

Figure 4Histology of excised terminal ileum, revealing a dense inflammatory infiltrate and sloughing pseudomembranes (bracket) with transmural necrosis.

both pathogen and toxin. While a more neutral pH encouraged proliferation of the C difficile micro-organism,6 the B1/NAP/027 genotype simultane-ously enabled far greater toxin production.7

A recent study found that ileostomy effluent col-lected from 16% of asymptomatic patients carried the C difficile toxin.8 A survey of jejunal specimens harvested at autopsy from patients without known gastrointestinal pathology likewise revealed a 3% rate of C difficile colonization.9 These findings sug-gest that, far from being the innocent bystander once thought, the small bowel may in fact rep-resent a silent reservoir for recurrent C difficile infections. Ileostomy reversal and right hemicol-ectomy are especially linked to CDC,10 suggesting that breach of the ileocecal barrier may release the pathogen. Our case study supports this hypoth-esis. Following loop ileostomy and vancomycin enemas, our patient was asymptomatic and rectal specimen tested negative for persistent C difficile toxin, but an ileal sample was not obtained. More-over, the patient received a maintenance regimen of toxin-binders and probiotics, which may have rendered small bowel colonization subclinical. Later, however, stapled anastomosis of a very dis-tal ileostomy using a long, 75 mm cartridge likely extended across and thereby compromised his ile-ocecal valve. This insult, coupled with additional antibiotic exposure and the immunosuppression of malignancy with recent surgery allowed the patho-gen to re-enter and flourish in the colon. Thus cir-cumstances set the stage for a fulminant and ulti-mately fatal infection encompassing both large and small bowel. In the future, potential precautionary steps we are considering include 1) sampling ileos-tomy effluent to identify asymptomatic carriers, 2) performing hand-sewn end-to-end ileal anastomo-ses to preserve the integrity of the ileocecal valve, and 3) administering newer pharmaceuticals such as fidaxomicin that are associated with signifi-cantly lower recurrence rates than vancomycin.11

ConclusionThe recent introduction of loop ileostomy with colonic lavage has improved short-term survival rates among patients with fulminant CDC. Nev-ertheless, increasingly virulent strains affecting an elderly, complicated patient population compel surgeons to suspect the small bowel as a poten-tial source of ongoing colonization and CDE as a potentially fatal ultimate outcome. Research efforts are certainly warranted to determine what steps should be taken prior to and during ileos-tomy reversal to ensure our patients’ continued wellbeing.

59

RECURRANT FATAL C DIFFICLE AFTER ILEOSTOMY TAKEDOWN

GHS Proc. May 2016; 1 (1): 56-59

References1. Killeen S, Martin ST, Hyland J, O’ Connell PR, Winter

DC. Clostridium difficile enteritis: a new role for an old foe. Surgeon. 2014;12:256-62.

2. Dineen SP, Bailey SH, Pham TH, Huerta S. Clostrid-ium difficile enteritis: A report of two cases and sys-tematic literature review. World J Gastrointest Surg. 2013;5:37-42.

3. Neal MD, Alverdy JC, Hall DE, Simmons RL, Zuck-erbraun BS. Diverting loop ileostomy and colonic lavage: an alternative to total abdominal colectomy for the treatment of severe, complicated Clostridium difficile associated disease. Ann Surg. 2011;254:423-7; discussion 427-9.

4. Beal EW, Bass R, Harzman AE. Two patients with fulminant Clostridium difficile enteritis who had not undergone total colectomy: a case series and review of the literature. Case Rep Surg. 2015;2015:957257.

5. Kim JH, Muder RR. Clostridium difficile enteritis: a review and pooled analysis of the cases. Anaerobe. 2011;17:52e5.

6. Dial S, Delaney JA, Barkun AN, Suissa S. Use of gas-tric acid-suppressive agents and the risk of communi-

ty-acquired Clostridium difficile-associated disease. JAMA. 2005;294:2989-95.

7. Lavalle e C, Laufer B, Pe´pin J, Mitchell A, Dube´ S, Labbe´ AC. Fatal Clostridium difficile enteritis caused by the BI/NAP1/027strain: a case series of ileal C. dif-ficile infections. Clin Microbiol Infect. 2009;15:1093e9.

8. Tsiouris A, Neale JA, Reickert CA, Times M. Clostrid-ium difficile of the ileum following total abdominal colectomy,with or without proctectomy: who is at risk? Dis Colon Rectum. 2012;55:424e8.

9. Testore GP, Nardi F, Babudieri S, Giuliano M, Di Rosa R, Panichi G. Isolation of Clostridium difficile from human jejunum: identification of a reservoir for dis-ease? J Clin Pathol. 1986;39:861e2.

10. Randall JK, Young BC, Patel G, Fitzgerald A, George BD. Is Clostridium difficile infection a particular problem after reversal of ileostomy? Colorectal Dis. 201;13:308-11.

11. Louie TJ, Miller MA, Mullane KM, et al; OPT-80-003 Clinical Study Group. Fidaxomicin versus vancomy-cin for Clostridium difficile infection. N Engl J Med. 2011;364:422e31.

CorrespondenceAddress to: Meghann L. Kaiser, MD Greenville Health System, Dept of Surgery 3rd Floor Support Tower 701 Grove Rd Greenville, SC 29605 (mkaiser@ghs.org)