Role of previous systemic antibiotic therapy on the probability of recurrence after an initial episode of infection treated with vancomycin.

OBJECTIVES

To investigate the role of previous antibiotic therapy in the risk of recurrence after a infection (CDI) treated with vancomycin.

METHODS

Multicentre observational study. Patients with a CDI episode achieving clinical cure with oral vancomycin and followed up 8 weeks were included. Previous antibiotic exposure up to 90 days was collected. Multivariate analysis of predictors of recurrence adjusted by the propensity score (PS) of being previously treated with each non-CDI antibiotic was performed.

RESULTS

Two hundred and forty-one patients were included; 216 (90%) had received systemic antibiotics. Fifty-three patients (22%) had a CDI recurrence. Rates of recurrence were lower in those treated with piperacillin/tazobactam in the last month when compared with those not receiving piperacillin/tazobactam [3 (7%) versus 50 (25%);  = 0.01], whereas higher rates were seen in those treated with cephalosporins in the last month [26/87 (30%) versus 27/154 (17%);  = 0.03]. In multivariate analysis controlled by the inverse probability of treatment weighting by PS, receiving 5 days of piperacillin/tazobactam in the last month as the last antibiotic regimen prior to CDI was independently associated with a lower risk of recurrence [adjusted OR (AOR) 0.13; 95% CI: 0.06-0.29;  < 0.0001] whereas exposure for 5 days to cephalosporins (versus piperacillin/tazobactam) was associated with an increased risk (AOR 10.9; 95% CI: 4.4-27.1;  < 0.0001).

CONCLUSIONS

Recent use of piperacillin/tazobactam might be associated with a lower risk of CDI recurrence, while recent use of cephalosporins might promote an increased risk. These findings should be considered when treating hospitalized patients.