This article evaluates the point in microbiological workup at which results of respiratory secretion and blood cultures could be interpreted as excluding MRSA and, thus, no longer warrant the empirical vancomycin therapy.
Time to first culture positivity among critically ill adults with methicillin-resistant growth in respiratory or blood cultures.
Melling PA, Noto MJ, Rice TW, Semler MW, Stollings JL. Ann Pharmacother. 2019;:1060028019877937. doi: 10.1177/1060028019877937.
Empiric antibiotic selection for treatment of serious and health care-associated infections, including pneumonia and bacteremia, often includes coverage of methicillin-resistant Staphylococcus aureus (MRSA), as these infections are associated with high mortality rates.1,2 Vancomycin is frequently the drug of choice; however, its use is marked by concern for serious adverse effects. Several guidelines recommend discontinuation of empirical MRSA coverage if cultures representative of the infection being treated (eg blood or respiratory) do not grow the organism.2,3
Melling and colleagues conducted a prospective, observational cohort study among adult patients admitted to 5 intensive care units (ICU) at Vanderbilt University Medical Center to evaluate the point in microbiological workup at which results of respiratory secretion and blood cultures could be interpreted as excluding MRSA and, thus, no longer warrant the empirical vancomycin therapy.4 This was an ancillary study conducted with patient data collected from the Solutions and Major Adverse Renal Events Trial (SMART) study which was a randomized trial of adult patients in the ICU comparing balanced crystalloids (eg lactated ringers solution) with saline for intravenous fluid therapy over a nearly 2 year period (June 2015 to April 2017).5 Patients in the SMART study who had sputum or blood cultures collected, that were on intravenous vancomycin, and were being empirically treated for MRSA pneumonia or bacteremia were included in this analysis. The primary outcomes of the study were time from culture collection to S aureus growth on respiratory cultures and preliminary identification of gram-positive cocci on gram stain of positive blood cultures. Secondary outcomes included time to growth of S aureus on blood cultures and prevalence of inappropriate vancomycin discontinuation.
A total of 6553 patient admissions received intravenous vancomycin and were included; 178 and 85 of these admissions had culture data consistent with MRSA pneumonia and bacteremia, respectively. Time to identification are summarized in the Table. In respiratory cultures, the median time to identification of S aureus was 33.3 hours, and the longest time to MRSA identification was 113 hours. Median time to MRSA identification from blood cultures was 45.5 hours, the longest time to identification was 90 hours gram-positive cocci, however, were identified on blood culture gram stain at a median of 21.3 hours.
Vancomycin was de-escalated based on gram stain or preliminary results; however, it was continued in all MRSA-positive patients. Notably, this included the 5 pneumonia and 2 bacteremic patients whose cultures took longer than 72 or 48 hours, respectively, as these 7 patients all had another indication for continuing vancomycin. In this large cohort of critically ill adults on vancomycin, the rates of positive cultures described above may inform the safety of discontinuation of empirical MRSA therapy after 48 to 72 hours of negative respiratory and blood cultures among critically ill adults.
This study has several notable limitations. All patients were critically ill and from a single site. This patient population may have had other indications for continuing vancomycin, so the possibly number of patients that could have been de-escalated may be over-estimated if looking only at culture reports. The institution’s microbiology practices (e.g. BACTEC for blood cultures) may not be able to be extrapolated to other institutions with markedly different practices. No rapid diagnostic technologies were utilized in the present study, and this has been shown to decrease time to pathogen identification, de-escalation, and time to effective therapy.6 MRSA nares has been shown to be a useful tool in the campaign to de-escalate MRSA coverage, though this testing was not utilized in the present study.7 These results may be useful to institutions who do not regularly utilize MRSA nares screening or in those who utilize culture and not PCR screening.
This ancillary analysis of patients from the SMART study allowed for a relatively large number of patients with positive cultures to be reviewed. This report is the first to include data on the timeliness of growth of MRSA in respiratory cultures and the role for possible de-escalation of empirical vancomycin. Appropriate de-escalation is one of the core tenets of antimicrobial stewardship programs; this study provides a more concrete timeline for the opportunity for de-escalation of empirical MRSA coverage in patients with culture not positive for MRSA.
Brian R. Raux, PharmD, BCPS, BCIDP is an infectious diseases/antimicrobial stewardship clinical pharmacy specialist at MUSC Health in Charleston, SC.* Raux is an active member of the Society of Infectious Disease Pharmacists.