Antimicrobial regime for cardiac surgery: the safety and effectiveness of short-course flucloxacillin (or teicoplanin) and gentamicin-based prophylaxis.

OBJECTIVES: To determine if changing from multidose cefuroxime-based to flucloxacillin (or teicoplanin) and gentamicin-based antibiotic prophylaxis for cardiac surgery was as effective at preventing infections without increasing postoperative renal impairment.
METHODS: Outcomes in consecutive patients from two 18-month periods with the different antibiotic regimes. Group 1 (1725 patients)-cefuroxime 1.5 g at induction and postoperatively. Group 2 (1695 patients)-flucloxacillin (or teicoplanin) and gentamicin at induction, valve procedures received further dose on weaning bypass. Primary end-points: new/worsening renal impairment, surgical site infection (SSI), Clostridium difficile infection (CDI). Multivariate logistic regression and interrupted time series segmented regression analysis were used.
RESULTS: Demographics were similar (age, EuroSCORE, gender, preoperative renal impairment). There were fewer wound infections in group 2: SSI 3.2% (group 1) versus 2.7% (group2) (p=NS); sternal infections 2.7% versus 2.0% (p=NS). New or worsening renal impairment was less frequent with gentamicin (4.3% group 1 vs. 3.4% group 2, p=NS). Mean postoperative stay 9.4 days (group 1) versus 8.7 days (group 2) (p=0.05). Logistic regression identified: diabetes, EuroSCORE associated with increased risk of renal and infective complications; female gender, pre-existing renal impairment associated with increased risk of acute renal impairment; bypass time associated with increased risk of wound infection. There were nine CDIs in group 1 compared with one in group 2 (p=0.02).
CONCLUSIONS: The change from multidose cephalosporin prophylaxis to short-course flucloxacillin (or teicoplanin) and gentamicin was not associated with an increase in renal complications, and resulted in significantly fewer CDIs, with no significant change in the incidence of wound infections.