Background Several studies indicate that 96% of patients with symptomatic C. difficile infection had received antibiotics within 14 days prior to the onset of diarrhoea. Complications include dehydration, electrolyte disturbances, toxic megacolon, hypotension, renal failure, systemic inflammatory response syndrome, sepsis and death.

Purpose To analyse patients admitted to our hospital testing positive for toxins A + B of C. difficile over 4½ years and to determine the associated mortality.

Material and methods Descriptive retrospective study conducted between 1 January 2010 and 30 June 2014 that included all hospitalised patients testing positive for C. difficile toxins A + B. Dara recorded were antimicrobial treatments before detection of toxins, condition for which they were prescribed, treatment to eradicate C. difficile and its duration. The percentage of patients who died during that admission or within 10 days of discharge was also determined.

Results Results were positive in 60 samples from 54 patients, 51.7% women and 48.3% men, with a mean age of 76.2 years. The most common pathologies for which antibiotics were prescribed were respiratory infection (36.4%), urinary tract infection (27.3%), intra-abdominal infection (16.4%), unspecified febrile syndrome (7.3%) and others (12.6%). 35 patients (64.8%) received more than one antimicrobial prior to detection of C. difficile, mainly ß-lactams (63.0%) and fluoroquinolones (27.8%).

Treatments prescribed for eradication of C. difficile were:

• Metronidazole (63.3%): <7 days in 4 isolates, 7–10 days in 18 of them, >10 days for 15 and five patients died during treatment.

• Oral vancomycin (20%): <7 days in 1 patient, 7–10 days in 3 isolates and >10 days for 8 of them.

• No treatment: 10 patients (16.7%).

Of the patients evaluated, 18 died (32.7%).

Conclusion The number of patients in whom toxins A + B of C. difficile were positively identified was low, probably due to the low degree of clinical suspicion and not too satisfactory sensitivity of the technique. Prescribing for treatments for eradication was not appropriate in 27.8% of patients (10 received no treatment and 5 for insufficient time). The mortality rates found in our study agree with data from the literature (22–40%).

References and/or acknowledgements No conflict of interest.