Risk of diarrhoea due to Clostridium difficile during cefotaxime treatmentBMJ 1996; 312 doi: https://doi.org/10.1136/bmj.312.7033.778 (Published 23 March 1996) Cite this as: BMJ 1996;312:778
Cefotaxime compares favourably with other third generation cephalosporins
- Eugene Rothschild,
- Alain Rauss,
- Gaby Danan
- Head of safety unit for marketed drugs Pharmacoepidemiologist Head of department Corporate Drug Safety and Epidemiology Department, Roussel Uclaf, 102 Route de Noisy, 93235 Romainville, France
EDITOR,—The retrospective study by M Impallomeni and colleagues shows an increased risk of diarrhoea due to Clostridium difficile in elderly patients receiving cefotaxime.1 The validity of their results is limited by the lack of precise data and by the questionable methodology used.
Firstly, the authors do not specify the number of cases of C difficile diarrhoea relating to each antibiotic used. Secondly, they do not state the extent to which all cases were correctly identified. Thirdly, exposure to different antibiotics was estimated from pharmacy expenditure rather than by direct review of patients' records. Fourthly, the absence of precise figures precludes reliable calculation of the frequency of C difficile diarrhoea in patients treated with cefotaxime or of the relative risk. Finally, the authors admit that in their univariate analysis they ignored other potential risk factors. Multivariate analysis of risk would have been more appropriate for identifying potential confounding factors, with particular account being taken of single versus combination antibiotic treatment and any use of antibiotics before admission. The fact that the authors calculated the relative risks for different antibiotics implies that not all cases of C difficile diarrhoea occurred in association with cefotaxime. The authors' figure, however, relates monthly expenditure on cefotaxime to all cases of C difficile diarrhoea. Such observations based on drug use do not show cause and effect, either during the outbreak or after the use of cefotaxime was stopped.
The figure shows that the ward faced an epidemic of C difficile diarrhoea from November 1993 onwards. The incidence of superinfection with C difficile depends on nosocomial factors2 and not directly on individual antibiotic regimens. The apparent prevalence associated with any particular antibiotic, however, may be distorted by the policy on use of antibiotics. Oversimplification should be avoided in assessments of the role of any individual or class of antimicrobial agent in C difficile disease in a nosocomial epidemic, especially when infection control measures are incomplete and no information on typing of C difficile is provided.
The authors' conclusions should be set against extensive experience gained with cefotaxime. With limited influence on normal intestinal flora,3 cefotaxime compares favourably with other third generation cephalosporins in terms of the rate of acquisition of C difficile in healthy volunteers and the occurrence of diarrhoea (1.2%) and C difficile disease (pseudomembranous colitis 0.14%) in patients.4 In terms of risk-benefit analysis, these characteristics have prompted expert panels worldwide to cite cefotaxime as a suitable first line drug for the hospital management of severe community acquired pneumonia.5