Gastrointestinal side effects frequently complicate antimicrobial therapy. Of the adverse effects, diarrhea and an occasionally lethal pseudomembranous colitis have been the subject of numerous reports. Pseudomembranous colitis occurred in the pre-antimicrobial era, usually as a complication of an abdominal operation.
In 1977, based on studies in a hamster model, it was reported that clindamycin-associated colitis was caused by a toxin-producing organism, Clostridium difficile. Most cases of antimicrobial-associated diarrhea can be classified as C. difficile associated or idiopathic. The role of other pathogens, such as Candida albicans, as a cause of antimicrobial-associated, C. difficile-negative disease requires more evidence.
Nearly all patients with C. difficile disease report use of an antimicrobial within the prior 6 weeks. Almost all antimicrobials have been implicated in cases of C. difficile disease; those most commonly involved are the cephalosporins, ampicillin or amoxicillin, and clindamycin. Most cases (total numbers) are associated with cephalosporin use, although the incidence rates are probably highest following clindamycin. Rarely, some antineoplastic drugs have been involved, such as fluorouracil and methotrexate. C. difficile should be considered as a cause of both community-acquired and nosocomial diarrhea; this organism is the most frequent cause of nosocomial diarrhea. In the hospital setting, numerous outbreaks of diarrhea as well as sporadic cases have been associated with C. difficile. The organism has been recovered from the hands of hospital personnel, and transmission via the hands appears to be an important mode of spread from patient to patient. In one study, use of vinyl gloves was associated with a fivefold decline in the incidence of C. difficile diarrhea.
C. difficile is an anaerobic gram-positive rod. Disease is localized to the colon. Pseudomembranes consisting of fibrin, mucus, epithelial cells, and leukocytes may be present adhering to the underlying mucosa. Only 3% of healthy adults will be culture-positive for C. difficile, in contrast to colonization rates of 50% in newborns. Colitis results from toxin production by the organism. Pathogenesis appears to involve four factors: (a) alteration in the intestinal flora secondary to antimicrobial therapy, (b) presence of C. difficile, usually from an exogenous source but sometimes in the patient’s endogenous flora, (c) presence of an organism capable of producing toxins A and B, and (d) age-related susceptibility (the illness is uncommon in children, and the elderly are at an increased risk).
The spectrum of disease ranges from asymptomatic to life-threatening. The typical patient notes profuse watery diarrhea with abdominal pain 4 to 9 days after starting to take an antimicrobial. Diarrhea may begin after the antimicrobial has been discontinued. Fever and leukocytosis are often present. Fever and abdominal pain may occur without diarrhea. Fecal leukocytes are present in about half the patients, and the stool guaiac test result may be positive. A leukemoid reaction and hypoalbuminemia may occur. If the disease is untreated, complications include toxic megacolon, colonic perforation, and shock. The death rate may be 10% in elderly debilitated patients.
The diagnosis depends on demonstrating disease by endoscopy and/or C. difficile toxin by specific assay. Sigmoidoscopy is adequate in 67% of cases, but in about 33% of cases, disease involves only the right side of the colon. Computed tomography may suggest the diagnosis, demonstrating a characteristic thickening of the colon. A barium enema is best avoided because of the possibility of colonic perforation. Diagnosis is usually established by demonstrating the presence of toxin B by tissue culture assay. Endoscopy may be helpful when the result of the toxin assay for C. difficile is negative. Results of tissue culture assays for C. difficile toxin are positive in 95% to 100% of patients with antimicrobial-associated pseudomembranous colitis, 15% to 25% of patients with antimicrobial-associated diarrhea without confirmed pseudomembranous colitis, 2% to 8% of patients with antimicrobial exposure without diarrhea, and none of healthy adults. False-negatives occur in about 10% of patients. Stool culture for C. difficile is available but is used mainly for epidemiologic studies. Other available tests include a latex particle agglutination test, enzyme immunoassays, polymerase chain reaction, and a dot immunobinding assay. The latex agglutination test lacks the sensitivity (60%) of the tissue culture assay but has a specificity of 96%. One enzyme immunoassay had a sensitivity of 85%, in comparison with 94% for the tissue culture assay. The specificity of the enzyme immunoassay was excellent (98%). Further data are needed to define the role of the dot immunobinding assay and polymerase chain reaction in the diagnosis.
Therapy for antimicrobial-associated colitis consists of stopping the implicated antimicrobial and providing supportive care. If an antimicrobial is still needed to treat the underlying infection, an agent should be selected that is infrequently associated with this disease, such as a quinolone or an aminoglycoside. Although data are lacking, avoid use of a cephalosporin, ampicillin, or clindamycin. Oral vancomycin (125 mg four times daily) or metronidazole (500 mg thrice daily) should be selected for patients with severe disease. Some patients with mild-to-moderate disease will respond if the implicated antimicrobial is discontinued and supportive therapy with fluids is provided. An advantage of metronidazole is its low cost. Treatment should be given usually for 7 to 10 days. Antidiarrheal agents must be avoided because they promote toxin retention in the colon. Corticosteroids are not indicated. Patients who are unable to take oral vancomycin should receive it by nasogastric tube. For patients who cannot take oral medications, parenteral metronidazole should be given along with oral vancomycin via a nasogastric tube or by a long tube. More data are needed on the optimal management of patients who cannot take oral medications. Most patients respond to oral vancomycin with defervescence within 24 hours and a reduction in diarrhea and abdominal cramps within 4 to 5 days. Response is poorer in patients with a toxic megacolon or other causes of an ileus.
Approximately 20% of patients will relapse within 4 weeks of completing therapy. Most relapses occur within 3 to 10 days after discontinuation of therapy. The frequency of relapse is the same whether vancomycin or metronidazole is used as initial therapy. Relapse does not depend on the duration of treatment, and repeated toxin assays are not indicated. Treatment of patients who relapse with oral vancomycin or metronidazole is usually effective. Bartlett has suggested two regimens for patients who have multiple relapses: (a) vancomycin plus rifampin for 10 to 14 days, or (b) vancomycin or metronidazole orally for 10 to 14 days followed by a 3-week course of cholestyramine, or cholestyramine plus lactobacilli, or vancomycin orally every other day. Other methods to restore the normal colon flora include administration of Saccharomyces boulardii orally for 1 month or of Lactobacillus preparations. (N.M.G.)
Bartlett JG. The 10 most common questions about Clostridium difficile-associated diarrhea/colitis. Infect Dis Clin Pract 1992;1:254.
Answers to questions regarding C. difficile issues.
Bartlett JG. Clostridium difficile infection: pathophysiology and diagnosis. Semin Gastrointest Dis 1997;8:12–21.
A review of diagnostic tests for the detection of C. difficile toxins. Latex particle agglutination lacks sensitivity, and enzyme-linked immunosorbent assays and dot immunoblot assays may have a role in rapid diagnosis.
Bartlett JG, et al. Antibiotic-associated pseudomembranous colitis due to toxinproducing clostridia. N Engl J Med 1978;298:531.
A toxin-producing clostridial species (C. difficile) resistant to clindamycin is the likely cause of antimicrobial-associated pseudomembranous colitis.
Climo MW, et al. Hospital-wide restriction of clindamycin: effect on the incidence of Clostridium difficile-associated diarrhea and cost. Ann Intern Med 1998;128:989–995.
Use of other antibiotics with anaerobic activity, such as cefotetan, ticarcillin-clavulanate, and imipenem-cilastatin, and a decreased use of clindamycin were associated with a decreased incidence of C. difficile diarrhea.
DiPersio JR, et al. Development of a rapid enzyme immunoassay for Clostridium difficile toxin A and its use in the diagnosis of C. difficile-associated disease. J Clin Microbiol 1991;29:2724.
A direct enzyme immunoassay test for C. difficile toxin A had a sensitivity of 85% and specificity of 98%. Results were available in 2.5 hours, compared with 24 to 48 hours for the tissue culture assay for toxin.
Do AN, et al. Risk factors for early recurrent Clostridium difficile-associated diarrhea. Clin Infect Dis 1998;26:954–959.
Relapse is more common than reinfection. Risk factors for relapse include a history of renal insufficiency and leukocytosis (cell count ³15,000/mm3).
Fekety R. Guidelines for the diagnosis and management of Clostridium difficile-associated diarrhea and colitis. Am J Gastroenterol 1997;92:739–750.
Review. Unnecessary use of antibiotics should be avoided within the first 2 months after treatment of an episode of C. difficile infection.
Fekety R, et al. Recurrent Clostridium difficile diarrhea: characteristics of and risk factors for patients enrolled in a prospective, randomized, double-blinded trial. Clin Infect Dis 1997;24:324–333.
Factors associated with recurrent C. difficile disease included (a) onset of initial disease in the spring, (b) prior episodes of C. difficile disease, (c) use of antibiotics for another infection during or shortly after the C. difficile episode, (d) female sex, and (e) certain strains of C. difficile.
Fishman EK, et al. Pseudomembranous colitis: CT evaluation of 26 cases. Radiology 1991;180:57.
The computed tomographic findings are nonspecific and demonstrate an increase in bowel wall thickness with the “accordion sign.”
George RH, Symonds JM, Dimock F. Identification of Clostridium difficile as a cause of pseudomembranous colitis. Br Med J 1978;1:695.
C. difficile was isolated from the stools of patients with pseudomembranous colitis.
Gerding DN, et al. Infect Control Hosp Epidemiol 1995;16:459–477.
A review of the diagnosis, epidemiology, infection control, and treatment of C. difficile disease. Testing stools of asymptomatic patients for C. difficile, including testing for cure, is not indicated.
Ho M, et al. Increased incidence of Clostridium difficile-associated diarrhea following decreased restriction of antibiotic use. Clin Infect Dis 1996;23(Suppl 1):S102–S106.
C. difficile disease increased after broad-spectrum antibiotics were removed from formulary restriction status.
Hutin Y, Casin I, Lesprit P, et al. Prevalence of and risk factors for Clostridium difficile colonization at admission to an infectious diseases ward. Clin Infect Dis 1997; 920–924.
Clostridium difficile was detected in the stool of 13% of patients on admission and was usually related to use of antibiotics in the prior month.
Jacobs J, et al. Eur J Clin Microbiol Infect Dis 1996;15:561–566.
In hospitalized patients with fewer than six stools per day, the prevalence of C. difficile is low (3%); in patients with more than six stools per day, the yield is 27%.
Johnson S, Gerding DN. Clostridium difficile-associated diarrhea. Clin Infect Dis 1998;26:1027–1036.
Review. The pathogenesis of C. difficile colitis involves exogenous acquisition of the organism and antibiotic exposure.
Johnson S, et al. Prospective, controlled study of vinyl glove use to interrupt Clostridium difficile nosocomial transmission. Am J Med 1990;88:137.
Hand carriage of C. difficile by hospital personnel is an important means of spread of this organism. The use of vinyl gloves can reduce the incidence of disease.
Johnson S, et al. Treatment of asymptomatic Clostridium difficile carriers (fecal excretors) with vancomycin or metronidazole. Ann Intern Med 1992;117:297.
Asymptomatic excretion of C. difficile should not be treated.
Katz DA, Lynch ME, Littenberg B. Clinical prediction rules to optimize cytotoxin testing for Clostridium difficile in hospitalized patients with diarrhea. Am J Med 1996;100:487–495.
Patients without a history of antibiotic use within the past month and without either diarrhea (at least three watery stools) or abdominal pain are unlikely to have a positive result on C. difficile toxin assay.
Keighly MRB, et al. Randomized controlled trial of vancomycin for pseudomembranous colitis and postoperative diarrhea. Br Med J 1978;2:1667.
Oral vancomycin (125 mg every 6 hours) was effective for pseudomembranous colitis caused by toxigenic strains of C. difficile.
Kelly CP, LaMont JT. Clostridium difficile infection. Annu Rev Med 1998;49:375–390.
Review. Response usually occurs within 3 days after start of therapy.
Kreutzer EW, Milligan FD. Treatment of antibiotic-associated pseudomembranous colitis with cholestyramine resin. Johns Hopkins Med J 1978;143:67.
Cholestyramine (4 g thrice daily), an anion-binding resin that may bind the toxin, was effective in 12 patients with pseudomembranous colitis.
Lemann F, et al. Arbitrary primed PCR rules out Clostridium difficile cross-infection among patients in a haematology unit. J Hosp Infect 1997;35:107–115.
Polymerase chain reaction can be helpful to type strains in the evaluation of an outbreak.
Manabe YC, et al. Clostridium difficile colitis: an efficient clinical approach to diagnosis. Ann Intern Med 1995;123:835–840.
Predictors of a positive C. difficile toxin assay included a positive fecal leukocyte test, semiformed stool, use of a cephalosporin, and onset of diarrhea 6 days after start of antibiotic therapy.
McFarland LV, et al. Nosocomial acquisition of Clostridium difficile infection. N Engl J Med 1989;320:204.
Fifty-nine percent of health care workers had a positive hand culture for C. difficile after contact with an infected patient.
Renshaw AA, Stelling JM, Doolittle MH. The lack of value of repeated Clostridium difficile cytotoxicity assays. Arch Pathol Lab Med 1996:120:49–52.
In only 1% of cases was a repeated C. difficile toxin assay useful if performed within a 7-day period.
Rifkin GD, et al. Antibiotic-induced colitis: implication of a toxin neutralized by Clostridium sordellii antitoxin. Lancet 1977;2:1103.
A clostridial toxin was isolated from two patients with pseudomembranous colitis. Oral vancomycin (500 mg every 6 hours) was given for 10 days, with resolution of the illness.
Salcedo J, et al. Intravenous immunoglobulin therapy for severe Clostridium difficile colitis. Gut 1997;41:366–370.
IV immunoglobulin may have a role in patients with C. difficile colitis who fail standard antibiotic therapy.
Stanley RJ, Melson GL, Tedesco FJ. The spectrum of radiographic findings in antibiotic-related pseudomembranous colitis. Radiology 1974;111:519.
Plaquelike mucosal lesions on barium enema are highly suggestive.
Tedesco FJ, Barton RW, Alpers DH. Clindamycin-associated colitis: a prospective study. Ann Intern Med 1974;81:429.
The authors noted a 21% incidence of diarrhea and a 10% incidence of pseudomembranous colitis in patients receiving clindamycin.