RESUMO
A cross-sectional study was conducted (April 96 to July 97) to investigate the prevalence of Clostridium difficile infection in immunocompromised patients at 3 health care institutions. Standard bacteriological procedure, tissue culture and enzyme immunoassay were used to investigate faecal specimens from immunocompromised inpatients (N = 113) including 21 patients treated with immunosuppressive drugs, 39 patients had received radiotherapy and 53 patients who were not treated with immunosuppressive drugs. The overall prevalence of C. difficile infections was 14.2 percent (16/113). All the C. difficile isolates were identified as toxigenic strains. The prevalence of C. difficile infection in patients who had received immunosuppressive drugs (23.8 percent, 5/21) did not differ significantly from those who had not (20.7 percent, 11/53), but was significantly lower in patients who received radiotherapy (p<0.01). The occurrence of C. difficile was not significantly associated with antibiotics in patients who were on immunosuppressive drugs (2/13, 15.4 percent vs 3/8, 37.5 percent) or patients who were not (7/32, 21.8 percent vs 4/21, 19.0 percent). Eighty percent (4/5) of C. difficile isolates from patients on immunosuppressive drugs were from those whose regimens include cytotoxic drugs. Similarly, anti-tubercolosis drugs were the antibiotics most frequently associated with C. difficile infections (4/8, 50 percent). C. difficile infections are likely to occur in patients treated with cytotoxic drugs and certain antibiotics.(AU)