Cefazolin Therapy for Staphylococcus aureus Bacteremia

Sir—Cefazolin remains one of the most frequently prescribed parenteral antibiotics. Its favorable pharmacokinetics, good activity against gram-positive and gram-negative organisms, and convenient low price are the reasons for its extensive use. Although in vivo studies conducted in the 1970s showed that rabbits with experimental Staphylococcus aureus endocarditis treated with cefazolin responded more slowly than did those treated with penicillin and that methicillin reduced the number of microorganisms in cardiac vegetations more rapidly than did cefazolin [1, 2], many still believe that both antibiotics are equivalent for treatment of staphylococcal infection. Herein, we report the failure of cefazolin to eradicate S. aureus bacteremia and endocarditis. A 74-year-old diabetic man with communicating hydrocephalus was admitted to the hospital because of fever. He had a ventriculoperitoneal shunt in place, and 2 weeks before admission, an endoscopic resection of a bladder cancer had been performed. The patient was thin, febrile (temperature, 38.5°C), and obtunded, without focal signs of infection. The lungs were clear, and no murmurs were heard. Treatment with amoxycillin-clavulanate was started. Blood cultures yielded S. aureus (MIC, 0.5 µg/mL), and treatment was switched to cloxacillin (2 g q4h iv). Fever subsided after 48 h, and the results of new sets of blood cultures were negative. Because urinary tract infection due to Klebsiella pneumoniae occurred, cloxacillin therapy was stopped after 6 days of administration. Treatment with cefazolin (1 g q8h iv; MIC for S. aureus, 1 µg/mL) was started. Three days later, a relapse of fever occurred, and 3 new blood cultures were again positive for S. aureus. A diastolic murmur was noted, and a transthoracic echocardiogram revealed a fibrocalcified aortic valve with vegetations. Combined therapy with cloxacillin plus gentamicin was started. Fortunately, fever and bacteremia subsided, and the patient had an uneventful clinical course. Six months later, he was doing well. Almost 30 years ago, some authors reported the relative lesser activity of cefazolin in staphylococcal endocarditis, compared with methicillin; this was attributed to inactivation of the drug by β-lactamase produced in cardiac vegetations [3–5]. More recent in vitro and in vivo studies have shown that, when a large inoculum is used, the MIC of cefazolin for S. aureus increased, and a rapid and complete degradation of the drug occurred that was associated with regrowth in a time-kill experiment [6, 7]. DNA sequencing of the β-lactamase gene revealed type A β-lactamase, which is known to efficiently inactivate cefazolin [7]. If those studies provide evidence for a gap in the activity of cefazolin against S. aureus in experimental models of endocarditis, our experience accurately illustrates the limited activity of the drug on the clinical ground. In a celebrated editors' note, Sherwood Gorbach commented that, on the occasion of a meeting in Los Angeles, California, during the question-and-answer session, someone asked Dr. Marvin Turck for his opinion about a new, rather controversial antibiotic [8]. “In a display of unrepentant scorn, Marv declared: that's a drug for my mother-in-law but not for my mother” (p. 46–47). A word of caution must be given: in our view, the use of cefazolin for the treatment of S. aureus endocarditis should be regarded as “a regimen for mothers-in-law.” Potential conflicts of interest. All authors: no conflicts.