By Richard R. Watkins, MD, MS, FACP, FIDSA

Associate Professor of Internal Medicine, Northeast Ohio Medical University; Division of Infectious Diseases, Cleveland Clinic Akron General, Akron, OH

Dr. Watkins reports that he has received research support from Allergan.

SYNOPSIS: A meta-analysis determined that nares screening for methicillin-resistant Staphylococcus aureus (MRSA) has a high specificity and negative predictive value for MRSA pneumonia. MRSA nasal screening can be a useful tool for antimicrobial stewardship personnel to de-escalate empiric anti-MRSA therapy.

SOURCE: Parente DM, Cunha CB, Mylonakis E, Timbrook TT. The clinical utility of methicillin-resistant Staphylococcus aureus (MRSA) nasal screening to rule out MRSA pneumonia: A diagnostic meta-analysis with antimicrobial stewardship implications. Clin Infect Dis 2018;67:1-7.

Although methicillin-resistant Staphylococcus aureus (MRSA) pneumonia is a serious illness associated with significant morbidity and mortality, its overall prevalence is low, especially as a cause of community-acquired pneumonia (CAP). Thus, clinicians frequently must deal with the dilemma of when to use empiric anti-MRSA therapy (e.g., vancomycin and linezolid), factoring in the inherent drawbacks, such as cost, adverse reactions, toxicities, and the promotion of antimicrobial resistance, associated with these agents. To address this concern, Parente and colleagues sought to determine the value of MRSA nasal screening in the management of MRSA pneumonia.

The investigators conducted a meta-analysis that included studies with information about both rates of positive MRSA nasal screening and the rates of MRSA pneumonia that were confirmed by culture. All classes of pneumonia were included: CAP, hospital-acquired pneumonia (HAP), healthcare-associated pneumonia (HCAP), and ventilator-associated pneumonia (VAP). Studies that used MRSA surveillance cultures from sites other than the nares were excluded from the analysis. Researchers employed a bivariate random-effects model to calculate the sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV).

There were 22 studies with 5,163 patients who met the inclusion criteria. Of these, 18 (81.1%) were retrospective, three were prospective (13.6%), and one was not reported. The diagnostic criteria for MRSA pneumonia differed in the various studies, with most using radiographic, microbiological, and clinical criteria to determine the diagnosis. The MRSA nares surveillance method differed in the studies as well, with 11 (50%) using polymerase chain reaction (PCR), four (18.35%) using culture, one study (4.5%) using both methods, and for the remaining six studies (27.3%) the method was not described. The timing for obtaining MRSA screening was reported in most of the studies (95.5%).

For all pneumonia types, the sensitivity of MRSA nares screen to predict pneumonia was 70.9% (95% confidence interval [CI], 58.8-80.6%], specificity was 90.3% [95% CI, 86.1-93.3%], PPV was 44.8%, and NPV was 96.5%. For CAP and HCAP, the values were 85% sensitivity [95% CI, 59.7-95.6%], 92.1% specificity [95% CI, 81.5-96.9%], 56.8% PPV, and 98.1% NPV. For VAP, the values were somewhat different, with a sensitivity of 40.3% [95% CI, 17.4-68.4%], specificity of 93.7% [95% CI, 77.1-98.4%], PPV of 35%, and NPV of 94.8%. There was a low probability of publication bias as determined by funnel plot testing.


The overuse of vancomycin is a serious concern in clinical practice. The study by Parente and colleagues is welcome because it provides solid evidence that can help antibiotic stewardship efforts in reducing the amount of anti-MRSA antibiotics prescribed for pneumonia. Even though a positive MRSA screen was not diagnostic, if the screen result was negative, pneumonia essentially could be ruled out in instances of CAP/HCAP. The sensitivity and NPV were lower for VAP, which the authors blamed on artificial airways providing a secondary source of MRSA besides the nares. However, in the absence of risk factors for MRSA and the presence of a negative MRSA nasal screen, it seems reasonable to stop anti-MRSA therapy and then observe closely. Further clinical studies will need to be conducted to determine outcomes in patients with pneumonia whose therapy is modified based on the results of MRSA nasal screening.

There are a couple of exceptions for which MRSA nasal screening might not be reliable to predict pneumonia. These include patients who were decolonized recently, those with a MRSA infection in the preceding 30 days, those with structural lung disease such as cystic fibrosis or bronchiectasis, and those who are critically ill.

As with all meta-analyses, the strength of the findings is directly proportional to the robustness of the studies that are included. That being said, most of the studies (81.8%) in the present meta-analysis had a retrospective design, which makes them subject to confounding by indication and sampling bias. Moreover, verification bias is a concern because nasal screening results often influence culture collection and clinical diagnosis. It is notable that the confidence interval for the sensitivity associated with VAP was particularly wide (17.4-68.4%), which may reduce the value of this variable. Finally, not all of the studies clearly defined the time that nasal swabs were collected compared to when the sputum cultures were taken.

The meta-analysis by Parente and colleagues has shown that a negative nasal screen for MRSA is a rapid, easy, and inexpensive way to exclude MRSA pneumonia. This will allow the discontinuation of anti-MRSA antibiotics to occur sooner, thus sparing patients unnecessary therapy and reducing costs. One potential strategy is to allow pharmacists to order MRSA nasal screens whenever a patient is prescribed vancomycin or linezolid for pneumonia. A negative result then could be discussed with the prescribing physician. Whether future pneumonia guidelines incorporate these new data remains to be seen.