Mortality due to C difficile colitis in elderly people has been underestimated

BMJ 1996; 312 doi: http://dx.doi.org/10.1136/bmj.312.7033.778a (Published 23 March 1996) Cite this as: BMJ 1996;312:778
  1. M Lesna,
  2. D M Parham
  1. Consultant pathologist Consultant pathologist Royal Bournemouth Hospital, Bournemouth BH7 7DW

    EDITOR,—M Impallomeni and colleagues report the frequency of diarrhoea due to Clostridium difficile in elderly patients receiving cefotaxime.1 Having seen two cases of fatal toxic megacolon after such infection, we reviewed all the obtainable notes for the 47 patients diagnosed in our hospital during December 1993 as having C difficile diarrhoea. Only 30 sets of notes were available for review. The diagnosis was made on the basis of detection of the toxin and the results of stool culture. The patients were all elderly (average age 84 (range 72-98)) and had a variety of diseases, some having multiple diseases. Some had been treated with antibiotics before admission, many having been given them as part of the management of stroke.

    Only three had had diarrhoea at home (induced by laxatives in one patient, during treatment with co-trimoxazole in one, and as part of a presumed infectious illness in one). All received more than one antibiotic in the hospital: most received cephalosporins, but erythromycin, co-amoxiclav, and amoxycillin were also given. The diagnosis was made by colonic biopsy in one case and at necropsy in one.

    Thirty of the 47 patients with microbiologically diagnosed C difficile diarrhoea died, 29 within three weeks of the microbiological diagnosis and one (with persistent colitis) 50 days later. Twelve patients died within one week of the diagnosis, and five died before the microbiological result became available. Pseudomembranous colitis or C difficile infection was not mentioned in the death certificates.

    To assess the mortality due to C difficile colitis in elderly people we looked at the outcome in 57 controls matched for age and sex who had been treated in the hospital at about the same time as the patients. All had received a range of antibiotic treatment. Ten (18%) of the 57 controls compared with 29 (62%) of the 47 patients with C difficile diarrhoea died 2-30 days after admission (three of the controls had disseminated malignancies).

    We conclude that the mortality from pseudomembranous colitis caused by C difficile in elderly patients with multiple diseases has been underestimated. Furthermore, morbidity may be considerably increased in those patients with C difficile colitis who survive. Our observation about the association between the cephalosporins, and particularly the mixture of various antibiotics, and pseudomembranous colitis in elderly people is in keeping with Impallomeni and colleagues' conclusions, and a further increase in hospital acquired C difficile colitis may be predicted if widespread use of broad spectrum antibiotics continues unabated.


    1. 1.