Evidence demonstrates that antimicrobial therapy with an anti-staphylococcal β-lactam antibiotic is the gold standard for treatment of SAB caused by MSSA, but confusion about which agent—particularly cefazolin versus an anti-staphylococcal penicillin—is preferred remains an unanswered question
Highlighted Study: Beganovic M, Cusumanno JA, Lopes V, LaPlante KL, Caffery, AR. Comparative effectiveness of exclusive exposure to nafcillin or oxacillin, cefazolin, piperacillin/tazobactam, and fluoroquinolones among a national cohort of veterans with methicillin-susceptible Staphylococcus aureus bloodstream infection. Open Forum Infect Dis. 2019;6(7):ofz270. doi: 10.1093/ofid/ofz270.
The appropriate treatment regimen for Staphylococcus aureus bacteremia (SAB) has been a hotly debated topic, especially in the last few years. Both methicillin-resistant S aureus (MRSA) and methicillin-sensitive S aureus (MSSA) are serious infections that have reported mortality rates between 20-40%.1 SAB can often cause metastatic infections such as infective endocarditis, meningitis, and osteomyelitis, which are very difficult to treat and require weeks of antimicrobial therapy. Studies have found conflicting results on the most effective therapy for both MRSA and MSSA bacteremia.
Recently, the focus on determining the most effective antimicrobial therapy for MSSA bacteremia has become popular in discussion. Evidence demonstrates that antimicrobial therapy with an anti-staphylococcal β-lactam antibiotic is the gold standard for treatment of SAB caused by MSSA, but confusion about which agent—particularly cefazolin versus an anti-staphylococcal penicillin—is preferred remains an unanswered question.2-3 Two meta analyses found that cefazolin may be safer and more effective than anti-staphylococcal penicillins, but there were limiting factors in both studies that prevent them from answering the question completely.4,5
A recent study by Beganovic and colleagues aimed to answer this question by comparing patients who had MSSA bacteremia and were treated exclusively with nafcillin, oxacillin, cefazolin, piperacillin/tazobactam, or fluoroquinolones (either moxifloxacin or levofloxacin).
This was a retrospective cohort study of patients who were hospitalized at a Veterans Affairs (VA) medical center and diagnosed with MSSA bacteremia from January 1, 2002, to October 1, 2015. The patients included were 18 years and older, remained admitted for more than 2 days, survived for longer than 2 days, and were treated exclusively with the antibiotics of interest. Patients were excluded if they received combination therapy or had a change in antimicrobial therapy.
Comparisons were made between the following groups: (1) nafcillin/oxacillin versus cefazolin, (2) nafcillin/oxacillin/cefazolin versus piperacillin/tazobactam, and (3) nafcillin/oxacillin/cefazolin versus levofloxacin/moxifloxacin. Outcomes evaluated were all-cause 30-day mortality from the first positive MSSA blood culture, length of hospital stay, length of intensive care unit stay, inpatient mortality, 30-day readmission, and 30-day S aureus reinfection. A total of 428 patients met inclusion criteria, with good representation among groups. One hundred and five patients were in the nafcillin/oxacillin group (24.5%), 107 in the cefazolin group (25%), 113 in the piperacillin/tazobactam group (26.4%), and 103 in the fluoroquinolone group (24.1%).
Baseline demographics were similar between groups, including the Charlson comorbidity index (median, 2 vs 2; P = 0.61). Notable differences were higher rates of chronic kidney disease (27.6% vs 41.1%; P = 0.04), older age (mean, 60.3 vs. 64; P = 0.03), and earlier antimicrobial treatment initiation from culture (median, 2 vs 1 days; P = 0.009) in nafcillin/oxacillin versus cefazolin, respectively.
However, baseline characteristics were balanced via propensity score matching (N = 44). Clinical outcomes were similar in both groups including: 30-day mortality 2/44 (4.6%) versus 3/44 (6.8%) (hazard ratio [HR], 0.67; 95% confidence interval [CI], 0.11-4.00); discharge (HR, 0.80; 95% CI 0.44-1.44); and 30-day readmission (HR, 0.75; 95% CI 0.26-2.16) in nafcillin/oxacillin versus cefazolin, respectively.
The piperacillin/tazobactam group was older (66.6 vs. 62.2 years, P = 0.003), had higher comorbidity scores (3 vs 2, P = 0.001), higher APACHE scores (34.5 vs 24.0, P < 0.0001), more ICU admissions (13.3% vs. 5.2%, P = 0.01), and higher previous hospitalizations (24.8% vs 15.1%, P = 0.03) compared with the nafcillin/oxacillin/cefazolin group. Again, the baseline characteristics were balanced via propensity score and resulted in N = 48 for both groups. The authors found that 30-day mortality was higher in the piperacillin/tazobactam group (HR, 0.10; 95% CI, 0.01-0.78).
Comorbidity burden was similar between groups, however patients treated with fluoroquinolones were older (69 vs 62.2 years, P < 0.0001), and had higher APACHE scores (32 vs 24, P = 0.0002). Again, groups were balanced after propensity score matching with a final N = 32. There were no differences in 30-day mortality between nafcillin/oxacillin/cefazolin and the fluoroquinolones (HR, 1.3; 95% CI, 0.30-5.96). This finding surprised the authors.
A limitation of this study is that it only included patients who had exposure to 1 antibiotic, which is not representative of clinical scenarios where antibiotics are adjusted, and usually include multiple empiric antimicrobials. The authors note that possible confounding by indication could be a limitation. This limitation is particularly in the case of nafcillin/oxacillin as these 2 antibiotics are rarely used empirically. Confounding by comedication was another major concern, but the authors utilized a highly specified exposure definition to overcome the problem. A final limitation is that safety data was not collected, which could have helped strengthen evidence that cefazolin may be a safer choice than nafcillin/oxacillin in relation to adverse effects.
The study helps antimicrobial stewardship efforts by not only providing more evidence of the effectiveness of narrow spectrum agents, but by indicating that broad spectrum beta lactams, in this case piperacillin/tazobactam, may not be as effective. More data is needed showing efficacy or even superiority of broader spectrum β-lactams, prior to recommending or justifying their use over the narrow spectrum agents. Although efficacy of fluoroquinolones for MSSA bacteremia was shown, their usage should not be considered first line. The abundance of literature showing adverse effects, including a black box warning from the FDA, should limit this class to a last line of therapy, for very specific patient populations.
Zimmer is an ID/Antimicrobial Stewardship pharmacist at CoxHealth in Springfield, MO. Drew enjoys fly-fishing, hunting, the St. Louis Cardinals and all things ID. You can find him on Twitter @zimrx17.
*He is an active member of the Society of Infectious Disease Pharmacists.