Int J Colorectal Dis (2003) 18:455–458 DOI 10.1007/s00384-003-0502-3 Baran Tokar Sultan Aydoǧdu Özgül Paşaoǧlu Huseyin Ilhan Emine Kasapoǧlu Accepted: 9 April 2003 Published online: 16 May 2003 © Springer-Verlag 2003 B. Tokar (✉) · S. Aydoǧdu · Ö. Paşaoǧlu H. Ilhan · E. Kasapoǧlu Departments of Pediatric Surgery, Pediatrics, and Pathology, School of Medicine, Osmangazi University, 26480 Eskisehir, Turkey e-mail: btokar@ogu.edu.tr Tel.: +90-222-2392979 ext 3255 Fax: +90-222-2290110 C A S E R E P O RT Neutropenic enterocolitis: is it possible to break vicious circle between neutropenia and the bowel wall inflammation by surgery? Abstract Background: Neutropenic enterocolitis is a devastating bowel wall inflammation in patients with protracted neutropenia. The approach for diagnosis and treatment is still controversial, and it is difficult and challenging to decide on what should be the next step in the management. Case presentation: We report a 10-year-old boy who developed neutropenic enterocolitis in the course of the conservative treatment for aplastic anemia. Oral mucositis and the perianal fissure with an ulcer were important indicators for what was happening on the colonic mucosa. Colonoscopy and biopsy confirmed the diagnosis. A fast recovery was achieved with a right hemicolectomy and ileostomy. Conclusion: Retrospective analysis of the long-term follow-up of our patient suggests that defunctioning the colon by ileostomy breaks the vicious circle between neutropenia and bowel wall inflammation, and an early surgical intervention could be considered as an adjunctive approach to the conservative management of persistent cases. Keywords Neutropenic enterocolitis · Aplastic anemia · Ileostomy Introduction Case report Neutropenic enterocolitis (NE) is an inflammatory process characterized by bowel wall inflammation with mucosal damage and intramural infection. The cause is poorly understood, and as an acute and life-threatening condition it occurs predominantly in children with prolonged or severe neutropenia [1, 2]. NE is a devastating disease process, and there is no significant consensus in the literature for the treatment and the follow-up. The steps in the diagnostic evaluation and management, including the decision whether to operate, involve some controversial and challenging points. We report a case of aplastic anemia with NE and discuss how the surgical intervention and diversion of fecal flow by stoma from the diseased bowel affect the protracted neutropenic disease process. A 10-year-old boy with aplastic anemia was receiving conservative treatment with total parenteral nutrition, broad-spectrum antibiotics (meropenem, vancomycin, metronidazole, fluconazole, and trimethoprim/sulfamethoxazole), and granulocyte/macrophage colony-stimulating factor (rHu GM-CSF; Molgramostim). At the end of 1 month of this medical therapy his complaints of fever, fatigue, and poor oral intake were still continuing and also diarrhea, nausea and vomiting, and a diffuse abdominal pain appeared, while findings such as oral mucositis, a progressive deep perianal fissure with an ulcer extending into the rectum (Fig. 1), and a diffuse abdominal tenderness were determined on the physical examination. The patient’s clinical condition deteriorated and during this period; he had a protracted neutropenia with an absolute neutrophil count (ANC) of less than 500/µl. The hemoglobin concentration was 6.2 g/dl and C-reactive protein 28.6 mg/dl. Urine and blood cultures were negative, and no pathogenic bacteria or parasites were detected in the stools. Plain abdominal radiography showed a mild degree cecal and ascending colon distension. Abdominal ultrasonography and computed tomography revealed only nonspecific colonic wall thickening. Stool analysis was later negative for Clostridium difficile toxin. We performed colonoscopy and found two patchlike significant ulcerous lesions (5–6 cm in 456 Fig. 1 Deep perianal fissure (arrow) with an ulcer extending into the rectum Fig. 3 Section of bowel wall shows fungal hyphae and spores. Hematoxylin and eosin stain, original magnification ×400 raphy showed an intestinal perforation. The exploratory laparotomy found a sigmoid colon perforation and irregular bowel wall thickening on the both ulcer locations determined by colonoscopy. The serosal inflammation of the colon was unremarkable. Since the histopathological diagnosis was not yet confirmed, and the patient had a total colonic involvement including the anorectal area following a simple plication of the site of perforation, the ulcerous segment on the hepatic flexura was resected, the rest of the colon was preserved, and a double-barreled colostomy was performed on the resection site. In the postoperative period the colostomy did not work efficiently, and on the fourth postoperative day a second operation was decided following discussion of whether supportive medical treatment should continue or a second exploration be done. The second laparotomy revealed only dilated and edematous bowel without severe inflammation or gangrene. Following a right hemicolectomy end ileostomy was performed, and the distal end beginning from the midtransverse colon was closed over as a “Hartman’s closure.” In the early postoperative period the ileostomy began to function effectively; a fast recovery was achieved in the clinical condition of the patient and the ANC was determined as 2000/µl. Enteral feeding started on the sixth postoperative day, and the proportion of the enteral feeding in the total nutritional support increased gradually. The patient was discharged on postoperative day 21 after full recovery. The ileostomy was closed 3 months later with an ileocolic anastomosis. At follow-up examinations, the laboratory tests including hematological and biochemical profile were all in normal ranges, and the patient remained clinically well with normal enteral feeding, weight gain, and bowel habit 1 year after the second operation. Fig. 2 Panaromic view of the lesion, showing ischemic and necrotic changes in mucosa (a), submucosa (b), and circular muscle layers (c) of large intestine. There is a dense inflammatory infiltrate in the submucosa extending into the muscle layers. Hematoxylin and eosin stain, original magnification ×40 Discussion diameter), one on the sigmoid and the other on the hepatic flexura of the transverse colon and also colonic microulcerous foci. Biopsy was performed, and microscopic examination showed inflammation with mucosal necrosis, mononuclear cell infiltration, no neutrophils, and extensive vascular thrombi of Gram-positive bacilli and fungi (Figs. 2, 3). The histopathological diagnosis was NE. On the day after the colonoscopic examination the patient developed acute abdominal symptoms, and plain abdominal radiog- NE is most frequently observed after chemotherapy for hematological and solid tissue malignancies. Cytotoxic chemotherapy induced mucosal injury may play an important role in the majority of cases, but this does not explain NE observed in association with aplastic anemia, organ transplantation, or human immunodeficiency virus 1 infection [3, 4]. For the pathogenesis of NE several factors leading to mucosal compromise and infection 457 have been suggested. These include shock, aggressive antibiotic therapy, treatment-induced necrosis of intestinal leukemic infiltrates, traumatic mucosal erosions, hemorrhage with subsequent mucosal necrosis, and the administration of prednisone or cytotoxics [2, 5, 6, 7]. Although the cause is not clearly understood, profound neutropenia appears to be a trigger for the enterocolitis [3]. Neutropenia is thought to cause colonic mucosal infection and inflammation foci, and this mucosal pathology may support the neutropenia. The protracted neutropenia aggravates the colonic mucosal damage, and this vicious circle continues until one of the contributing factors disappears. Although most of the abnormalities determined in the clinical presentation and radiological findings of our case were nonspecific, retrospective evaluation shows that oral mucositis and the perianal fissure with an ulcer extending into the rectum were important indicators for what was happening on the colonic mucosa. NE usually involves the cecum and right colon [3, 5, 8]. Other parts of the gastrointestinal tract may also have some macroscopic and microscopic changes, including oral mucositis, mucosal ulceration, edema, inflammation, necrosis, and perforation [4, 9, 10, 11]. Total colon is reported to be involved in only 6% of cases, and rectal involvement seems to be even more rare [1, 9]. In our patient we found significant mucosal ulcerous lesions on the rectum, sigmoid, and transverse colon, microulcerous foci on the entire colonic mucosa, and oral mucositis. If abdominal pain, fever, and watery or bloody diarrhea develops in a patient with a protracted neutropenia, the possibility of NE should be considered. The findings determined in the laboratory and radiological diagnostic procedures could be nonspecific. In this case a complete colonoscopy with biopsy may lead to the diagnosis [9]. The primary differential diagnosis is pseudomembranous colitis, which has the same clinical presentation. However, neutropenia is not a prominent feature for pseudomembranous colitis, and stool specimens assayed for toxin of C. difficile help in differentiation [3]. A concomitant ulcerative colitis or Crohn’s disease, cholecystitis, ischemic colitis, leukemic or lymphomatous infiltration of the bowel wall, and infectious colitis from bacterial or other opportunistic pathogen, such as cytomegalovirus should also be considered in the differential diagnosis [3, 12, 13]. The diagnosis is often delayed if a predisposing condition is not recognized. Every available diagnostic procedure should be considered before deterioration of the patient’s status. In our case intestinal perforation was most likely in view of the colonoscopic examination or the biopsy specimen taken from the sigmoid colon. Colonoscopy is an invasive procedure and may cause a perforation due to colonic inflammation and air insufflation, but it provides early diagnostic clues for differentiation of the pathology, and it is the only procedure that shows the severity of the mucosal changes. Many reports suggest a poor prognosis for NE; reported mortality rates range from 50% to 100% [2, 14]. Fear of increasing morbidity and mortality by a surgical intervention favors a decision for more conservative management. A successful conservative approach and reserving laparotomy for perforation, life-threatening hemorrhage, and the differential diagnosis of other surgical problems have been reported by several authors [2, 11, 15, 16]. Depending on the underlying disease the patient may respond to medical therapy consisting of bowel rest, broad spectrum antibiotics, and intravenous fluid resuscitation. However, the question of whether an early surgical intervention to the source of infection and inflammation could facilitate the treatment of neutropenia must be discussed. Several reports recommend an earlier surgical intervention [8, 17]. After the decision to intervene surgically another controversial and challenging step is to decide on what operation should be performed, especially when there is no significant inflammatory and gangrenous change on the gross appearance of the colon during the laparotomy. In general a right hemicolectomy, with or without primary anastomosis, or, depending on the pathology, total abdominal colectomy has been advocated [1, 8, 11, 14, 17]. As an alternative procedure a divided ileostomy for defunctioning the colon has been suggested [18]. Patients with primary ileocolic anastomosis usually have a worse postoperative course than those with a right hemicolectomy and ileostomy [8]. This suggests that along with the right hemicolectomy which removes the septic focus, an ileostomy must also have an important role in obtaining a good postoperative recovery, otherwise the right hemicolectomy with a primary ileocolic anastomosis could work as effective as the procedure combined with an ileostomy. In our case we tried to limit the resection to a minimum. In the second operation, although there were significant ulcerous lesions on the sigmoid colon and rectum, a fast recovery was achieved with right hemicolectomy and ileostomy. In addition to the improvement in the clinical condition and the laboratory tests of the patient, the ulcers on the distal preserved colon had healed totally before closure of the ileostomy. This suggests that if there is no significant gross inflammatory and gangrenous change, a resection may not be necessary, and defunctioning the colon with an ileostomy may be the only needed procedure. Fecal flow in a dilated colon of an ill patient having an impairment of bowel motility due to inactivity and the disease process may increase colonic distension and promote the persistence of micro-organisms in the bowel walls. Distension and infection may cause and support bacterial translocation and submucosal ischemia [19]. An ileostomy diverting the fecal flow may prevent this 458 sequence. Although our initial purpose for surgical intervention in this case was an intestinal perforation, a longterm follow-up of this patient taught us that in a protracted disease state with NE an early surgical intervention without evidence of any surgical emergency may provide a fast recovery of the general status of the patient and neutropenia, and that resting the colon by ileostomy may heal even the severe mucosal ulcerous changes. References 1. Larsen TK, Qvist N, Bak M (2001) Delayed neutropenic enterocolitis in a 12-year-old girl treated with total colectomy and J-pouch reservoir. J Pediatr Surg 36:1066–1067 2. Baerg J, Murphy JJ, Anderson R, Magee JF (1999) Neutropenic enteropathy: a 10-year review. J Pediatr Surg 34:1068–1071 3. Davila AD, Willenbucher RF (1998) Other diseases of the colon and rectum. In: Feldman M, Sleisenger MH, Scharschmidt BF (eds) Sleisenger and Fordtran’s gastrointestinal and liver disease. Saunders, Philadelphia, pp 1977–2006 4. Katz JA, Wagner ML, Gresik MV, Mahoney DH, Fernbach DJ (1990) Typhlitis. An 18-year experience and postmortem review. Cancer 65:1041–1047 5. Moir DH, Bale P (1976) Necropsy findings on childhood leukaemia emphasizing neutropenic enterocolitis and cerebral calcification. Pathology 8:247–258 6. Amromin GD, Solomon RD (1962) Necrotizing enteropathy: a complication of treated leukemia or lymphoma patients. JAMA 182:23–29 7. Varki AP, Armitage JO, Feagler JR (1979) Typhlitis in acute leukemia. Successful treatment by early surgical intervention. Cancer 43:695–697 8. Koea JB, Shaw JH (1989) Surgical management of neutropenic enterocolitis. Br J Surg 76:821–824 9. Kulaylat M, Doerr R, Ambrus J (1997) A case presentation and review of neutropenic enterocolitis. J Med 28:1–19 10. Jain Y, Arya LS, Kataria R (2000) Neutropenic enterocolitis in children with acute lymphoblastic leukemia. Pediatr Hematol Oncol 17:99–103 11. Shamberger RC, Weinstein HI, Delorey MJ, Levey RH (1986) The medical and surgical management of typhlitis in children with acute nonlymphocytic (myelogenous) leukemia. Cancer 57:603–609 12. Gorschluter M, Glasmacher A, Hahn C, Leutner C, Marklein G, Remig J, Schmidt-Wolf IG, Sauerbruch T (2001) Severe abdominal infections in neutropenic patients. Cancer Invest 19:669–677 13. Scott-Conner CE, Fabrega AJ (1996) Gastrointestinal problems in the immunocompromised host. A review for surgeons. Surg Endosc 10:959–964 14. Wade DS, Nava HR, Douglas HO (1992) Neutropenic enterocolitis, clinical diagnosis and treatment. Cancer 69:17–23 15. Gomez L, Martino R, Rolston KV (1998) Neutropenic enterocolitis: spectrum of the disease and comparison of definite and possible cases. Clin Infect Dis 27:695–699 16. Shaked A, Shinar E, Freund H (1983) Neutropenic typhlitis: a plea for conservatism. Dis Colon Rectum 26:351–352 17. Alt B, Glass NR, Sollinger H (1985) Neutropenic enterocolitis in adults. Review of the literature and assessment of surgical intervention. Am J Surg 149:405–408 18. Moir CR, Scudamore CH, Benny WB (1986) Typhlitis: selective surgical management. Am J Surg 151:563– 566 19. Ikard RW (1981) Neutropenic typhlitis in adults. Arch Surg 116:943–945