Secondary Logo

Journal Logo

Case Reports

Polymyxin B–Immobilized Fiber Hemoperfusion in a Patient with Active Ulcerative Colitis

Nakamura, Tsukasa*; Kawagoe, Yasuhiro*; Ueda, Yoshihiko; Koide, Hikaru

Author Information
doi: 10.1097/01.mat.0000169079.37644.48
  • Free


Inflammatory bowel disease, including ulcerative colitis, involves a complex interplay of environmental, genetic, microbial, immune, and nonimmune factors.1 Endogenous interstitial bacterial flora plays an important role in the initiation and perpetuation of the disease. Commensal bacteria express fimbrial and nonfimbrial adhesions, which enable them to attach to the mucus layer and to intestinal epithelial cells. Increased adherence of commensal bacteria to inflamed mucosa may enhance the exposure of the mucosal immune system to intestinal bacteria or bacterial components, resulting in sustained inflammation.2 By in situ hybridization, Kleessen et al.3 confirmed bacterial invasion of mucosa in 83% of colonic specimens obtained from ulcerative colitis patients. Systemic endotoxemia was found in 88% of patients with ulcerative colitis.4 Administration of endotoxin to human volunteers induces many systemic changes in inflammatory bowel disease.5 A better understanding of the endotoxins and the mechanisms of their action may help us develop more effective treatments for ulcerative colitis.

The polymyxin B–immobilized fiber (PMX-F) cartridge has been commercially available since 1994 and is now an approved therapeutic device for patients with sepsis.6 We reported previously that PMX-F treatment is effective in reducing endotoxin and various mediators and in reducing the mortality rate in septic patients.7,8 However, little is known about the effect of PMX-F treatment in ulcerative colitis patients with endotoxemia. We describe here the effect of PMX-F treatment in a patient with ulcerative colitis and endotoxemia.

Case Report

A 31-year-old woman was hospitalized for general fatigue, low-grade fever, abdominal pain, and severe diarrhea. On examination, her blood pressure was 120/80 mm Hg, and she was anemic. Neither rales nor heart murmur was heard. Her blood urea nitrogen concentration was 16 mg/dl, serum creatinine was 0.9 mg/dl, and hemoglobin was 7.8 g/dl. Her urine was negative for casts, protein, and occult blood. Fresh blood was recognized in the stool. Her antinuclear antibody level and serum complement C3 and C4 levels were normal. Her serum alanine aminotransferase and lactate dehydrogenase levels were also normal. Chest radiography showed no abnormalities. Abdominal ultrasonography revealed edematous colon. Serum proteinase-3-antineutrophil cytoplasmic antibody (ANCA) and myeloperoxidase-ANCA results were negative. Colonoscopy showed multiple ulcers, edematous mucosa, bleeding, and polyposis in the colon. The diagnostic criteria for ulcerative colitis set by the Experts Committee on Inflammatory Bowel Disease of the Ministry of Health and Welfare of Japan were met.9

The patient was treated with prednisolone (initial dose 40 mg/day, then tapered) and sulfasalazine (3.0 g/day), and her symptoms disappeared within 1 month. Over the next 3 months, she was treated with sulfasalazine alone, and her condition remained stable for 4 years. She then suffered abdominal pain, severe diarrhea (up to 20 bowel movements per day), and melena and was admitted. Prednisolone (40 mg/day) was again administered, but her symptoms did not improve. Colonoscopy revealed multiple ulcers, bleeding, and polyposis in the colon (Figure 1A). Histopathologic examination of the colonic biopsy specimen revealed classic features of active ulcerative colitis with diffuse bloody erosion, massive leukocyte and lymphocyte infiltration, and decreased goblet cells (Figure 2). Blood endotoxin concentrations were determined by the Endospecy test, as described previously6; the upper limit of normal is 9.8 pg/ml, and the lower detection limit is 0.5 pg/ml. The patient’s blood endotoxin level was 38.0 pg/ml. Tumor necrosis factor-α (TNF-α) and interleukin-6 (IL-6) concentrations were determined by enzyme immunoassay. The detection limit of the assay was 5 pg/ml for TNF-α and 6 pg/ml for IL-6. The patient’s blood TNF-α level was 20.0 pg/ml and blood IL-6 level was 250 pg/ml.

Figure 1.
Figure 1.:
(A) Endoscopic view of the colon before polymyxin B–immobilized fiber hemoperfusion. There are multiple ulcers with bleeding, edematous mucosa, and polyposis. (B) Endoscopic view of the colon after two treatments of polymyxin B–immobilized fiber hemoperfusion 24 hours apart at a flow rate of 80–100 ml/min. The mucosa is no longer edematous and shows revascularization. The ulcers are remarkably improved.
Figure 2.
Figure 2.:
Colonic histopathology. The mucosa has the classic features of ulcerative colitis with diffuse bloody erosion and massive leukocyte and lymphocyte infiltration. Goblet cells are decreased in number.

Polymyxin B–immobilized fiber treatment was performed twice, with a 24-hour interval between treatments. PMX-F and the hemoperfusion column have been described previously.6–8 We reported previously that a 24-hour interval was effective for PMX-F treatment in septic patients.7,8 We obtained circulatory access for direct hemoperfusion with PMX-F by means of a double-lumen catheter (Arrow International, Inc., Reading, PA) inserted into the femoral vein by the Seldinger method. Direct hemoperfusion was performed for 2 hours at a flow rate of 80–100 ml/min. Blood samples were obtained immediately before the first PMX-F treatment and immediately after the second PMX-F treatment. Blood endotoxin levels decreased to5.0 pg/ml after each treatment. Blood TNF-α levels decreased to below the detection limit, and blood IL-6 levels decreased to 20 pg/ml after treatment. Two weeks after the PMX-F treatment, the patient’s symptoms had completely resolved. Her stool output was reduced from 20 bowel movements per day to one per day, and the bloody stools and abdominal pain had disappeared. The prednisolone was reduced gradually from 40 mg/day to 5 mg/day. An endoscopic photograph of the patient’s colon after treatment showed disappearance of the edema, revascularization of the mucosa, and improvement of the ulcers (Figure 1B). Histopathologically, the colonic biopsy specimen showed that goblet cell depletion was improved, and massive cell infiltration was absent.


Quantitative and qualitative disturbances in fecal flora suggest a role for enteric bacteria and their products in the pathogenesis of inflammatory bowel disease including ulcerative colitis. Gardiner et al.4 reported that the presence of systemic endotoxemia, its correlation with disease activity, disease extent, and endotoxin core antibody concentration, and the presence of circulating TNF-α all support a pathogenic role for endotoxins in inflammatory bowel disease. Our patient had high blood endotoxin, TNF-α, and IL-6 concentrations. Gut-associated lymphoreticular tissue (GALT) protects the host against infectious agents and/or noxious toxins. In GALT, protective immune responses are well orchestrated and do not harm the normal intestinal mucosa.10 In patients with ulcerative colitis, however, the GALT immune response is exaggerated, producing a prolonged, severe inflammatory state in the intestinal mucosa.11 Bacterial antigens may account for the release of excessive amounts of proinflammatory cytokines detected in mucosal specimens and in stools of patients with inflammatory bowel disease.10 Recently, Kleesen et al.3 reported that colonic specimens obtained from patients with ulcerative colitis were colonized by a variety of organisms, such as bacteria belonging to the gamma subdivision of Proteobacteria, Enterobacteriaceae, the Bacteroides/Prevotella cluster, the Clostridium histolyticum/Clostridium lituseburense group, and the Clostridium coccoides/Eubacterium rectale group. In the present case, however, we did not examine bacterial species.

During sepsis, release of various endotoxins from microorganisms activates the immune response, including release of proinflammatory cytokines.12 PMX-F, which was developed to remove endotoxin selectively, has been available for treatment of sepsis.6–8 There is strong evidence that the septic state is due to a combination of direct and indirect effects of endotoxin. Endotoxin releases a vast array of mediators, and it is reasonable to remove or to antagonize endotoxin when treating patients with sepsis.12 Endotoxin may also play an important role in the pathogenesis of ulcerative colitis. Therefore, we performed PMX-F treatment in a patient with active ulcerative colitis. Blood endotoxin, TNF-α, and IL-6 concentrations decreased after PMX-F treatment. In addition, clinical symptoms and changes in the colon viewed endoscopically were ameliorated by PMX-F treatment.

In summary, PMX-F treatment may be beneficial for the management of ulcerative colitis with endotoxemia.


1. Farrell RJ, LaMont JT: Microbial factors in inflammatory bowel disease. Gastroenterol Clin N Am 31: 41–62, 2002.
2. Schultsz C, Van den Berg FM, Ten Kate, FW, Tytgat GN, Dankert J: The intestinal mucus layer from patients with inflammatory bowel disease harbors high numbers of bacteria compared with controls. Gastroenterol 117: 1089–1097, 1999.
3. Kleessen B, Kroesen AJ, Buhr HJ, Blaut M: Mucosal and invading bacteria in patients with inflammatory bowel disease compared with controls. Scan J Gastroenterol 37: 1034–1041, 2002.
4. Gardiner KR, Halliday MI, Barclay GR, et al: Significance of systemic endotoxaemia in inflammatory bowel disease Gut 36: 897–901, 1995.
5. Sturk A, van Deventer, SJH, Wortel CH, et al: Detection and clinical relevance of human endotoxemia. Z Med Lab Diagn 31: 147–158, 1990.
6. Aoki M, Kodama M, Tani T, Hanasawa K: Treatment of sepsis by extracorporeal elimination of endotoxin using polymyxin B-immobilized fiber. Am J Surg 167: 412–417, 1994.
7. Nakamura T, Ushiyama C, Suzuki Y, Shoji H, Osada S, Shimada N, Koide H: Effect of polymyxin B-immobilized fibre on various mediators in patients with hypothermic sepsis. Clin Intensive Care 12: 223–228, 2002.
8. Nakamura T, Kawagoe Y, Matsuda T, Ebinara I, Koide H: Effects of polymyxin B-immobilized fiber hemoperfusion on amino acid imbalance in septic encephalopathy. Blood Purif 21: 282–286, 2003.
9. Shimoyama T, Sawada K, Hiwatashi N, et al: Safety and efficacy of granulocyte and monocyte adsorption apheresis in patients with acute ulcerative colitis: A multicenter study. J Clin Apher 16: 000–9, 2001.
10. Caradonna L, Amati L, Magrune T, Pellegrino NM, Jirillo E, Coccavo D: Enteric bacteria, lipopolysaccharides and related cytokines in inflammatory bowel disease: Biological and clinical significance. J Endotoxin Res 6: 205–214, 2000.
11. Macdermott RP: Immunology of inflammatory bowel disease. Curr Opin Gastroenterol 14: 54–67, 1998.
12. Kanno Y, Nemoto H, Nakamoto Y, et al: Selection of hemoperfusion therapy for patients with septic shock on the basis of the primary disease. J Artif Organs 6: 205–210, 2003.
Copyright © 2005 by the American Society for Artificial Internal Organs