Case records of the Massachusetts General Hospital. Weekly clinicopathological exercises. Case 17-1996. A 48-year-old man with the acquired immunodeficiency syndrome, abdominal pain, and bloody diarrhea

Coexisting cytomegalovirus infection in immunocompetent patients with Clostridium difficile colitis

Cytomegalovirus (CMV) colitis usually occurs in immunocompromised patients with human immunodeficiency virus infection, organ transplantation, and malignancy receiving chemotherapy or ulcerative colitis receiving immunosuppressive agents. However, CMV colitis is increasingly recognized in immunocompetent hosts. Notably, CMV colitis coexisting with Clostridium difficile infection (CDI) in apparently healthy individuals has been published in recent years, which could result in high morbidity and mortality. CMV colitis is a rare but possible differential diagnosis in immunocompetent patients with abdominal pain, watery, or especially bloody diarrhea, which could be refractory to standard treatment for CDI. As a characteristic of CDI, however, pseudomembranous colitis may be only caused by CMV infection. Real-time CMV-polymerase chain reaction (PCR) for blood and stool samples may be a useful and noninvasive diagnostic strategy to identify CMV infection when treatment of CDI eventually fails to show significant benefits. Quantitative CMV-PCR in mucosal biopsies may increase the diagnostic yield of traditional histopathology. CMV colitis is potentially life-threatening if severe complications occur, such as sepsis secondary to colitis, massive colorectal bleeding, toxic megacolon, and colonic perforation, so that may necessitate pre-emptive antiviral treatment for those who are positive for CMV-PCR in blood and/or stool samples while pending histological diagnosis.

Systemic antibody response to Clostridium difficile in colonized patients with and without symptoms and matched controls

It has been proposed that patients who develop Clostridium difficile-associated disease (CDAD) do so because they are unable to mount an adequate immune response. Serum was collected from three groups of elderly in-patients: (i) cases (n=21) of CDAD, being toxin A/B-positive; (ii) carriers (n=21) asymptomatic for CDAD (no diarrhoea) but at least toxin or culture positive; and (iii) controls (n=26) asymptomatic for CDAD and negative for both C. difficile toxin and culture. The age and gender of each group were compared, and the colonizing strains were ribotyped and toxinotyped. Serum antibodies (IgG and IgM) were measured by ELISA using different antigen preparations: EDTA extract (containing cell-surface proteins and carbohydrates), guanidine hydrochloride extract (surface-layer proteins), aqueous phenol-extracted lipocarbohydrate (LC); crude toxin (dialysis culture supernatant) and purified toxin A. LPS from Escherichia coli was used as a control antigen. Antibodies were also tested for toxin neutralization on tissue monolayers and for binding to EDTA-extracted antigens by Western blotting. IgG antibody measurements to cytomegalovirus (CMV) were included as an indicator of potential immunosenescence. Results showed that the patient groups were well matched by age and gender, and the colonizing strains were similar in cases and carriers, being predominantly ribotype 001 and toxinotype 0. By ELISA, IgG levels to most of the antigens were highest in the cases and lowest in the controls, with the exception of antibodies to the LC, which were higher in the controls than the cases. Levels in the carriers tended to be of intermediate level or similar to the controls. For all antigens, the levels of IgM were not significantly different among cases, carriers and controls. Serum from all groups was able to neutralize the cytotoxic action of toxin on both Vero and Caco2 cells, and all to a similar extent. Western blots showed an overall higher level of IgG antibodies to the EDTA-extracted antigens in the cases. The results of the CMV ELISA showed that specific IgG was detected in more cases (78%) than carriers and controls (both 65%), but this difference in seropositivity was not significant. The conclusion is that, during symptomatic infection, patients respond to protein antigens of C. difficile in a manner typical of a secondary antibody response, with no evidence that an inability to respond predisposes to the appearance of symptoms.