A seven-year survey of management of coagulase-negative staphylococcal sepsis in the neonatal intensive care unit: vancomycin may not be necessary as empiric therapy.

BACKGROUND

The typical empiric therapy for coagulase-negative staphylococcal (CONS) sepsis includes vancomycin. In our neonatal intensive care unit, we have consistently avoided the use of vancomycin to treat CONS sepsis, except for specific cases, and have used instead cefazolin as empiric agent.

OBJECTIVES

The clinical outcome of infants with CONS sepsis was evaluated in relation to the susceptibility of CONS blood isolates to cefazolin over a period of 7 years.

METHODS

Clinical characteristics, symptoms of sepsis and antibiotic use were studied retrospectively. Susceptibility of CONS blood isolates to cefazolin was determined by E-test.

RESULTS

Of 163 infants with proven CONS sepsis, 121/140 (86%) infants with a cefazolin-susceptible (minimum inhibition concentration (MIC) ≤8 mg/l) and 21/23 (91%) with a cefazolin-resistant (MIC ≥32 mg/l) blood isolate were treated with cefazolin. 21 (13%) infants were switched to vancomycin, in only 3 of them CONS had become resistant to cefazolin. The majority (81%) of the infants with a good response to cefazolin had the indwelling central venous catheter removed, in contrast to only 22% of the infants with cefazolin treatment failure. Median cefazolin MIC values were 0.75-2 mg/l during the study period.

CONCLUSIONS

The great majority of infants with CONS sepsis was successfully treated with cefazolin. The use of vancomycin could be restricted to specific cases. Despite the consistent use of cefazolin in neonatal CONS sepsis over an extended period of time, cefazolin MIC values remained low and in the susceptible range. Removal of the central venous catheter in infants with clinical symptoms of sepsis is an important therapeutic measure.