MRSA screens and MRSA pneumonia
This topic comes up so frequently in my ICU. I’m surprised I didn’t think to write about this earlier. Shout out to our clinical pharmacist (T.A.) for being such an effective steward in our antibiotic choices, lab tests, anticoagulation and more!
I would say the majority of our patients end up having fevers and hypotension or some kind of hemodynamic instability to trigger “sepsis thoughts” in our head. Maybe we see a CXR read that says b/l pulmonary edema/infiltrates, cannot rule out superimposed pneumonia/pneumonitis, etc, etc. Then we start treating for CAP or HCAP. Then we jump to the sputum cx, blood cx, UA with cx reflex, MRSA screen, RVP, apap, and then vanc + pip-tazo or vanc + cefepime. Or we think we see a vegetation or AoR abscess on TTE/TEE, which also warrants empiric broad spectrum abx. The list can go on.
So let’s talk about the effectiveness of screenings. I discussed briefly about PCT on another post. But I would like to know how to utilize MRSA screenings/PCR more effectively. From my review of lit, you have to break down MRSA swab PCR assay (nasal colonization) for predictive value of MRSA pneumonia vs MRSA bacteremia. It’s important to cover organisms but also to recognize when to de-escalate antibiotic therapy.
One retrospective cohort study published by Dangerfield, et al in 2014 at a tertiary care facility looked at 435 patients in the final analysis. The patient population included lymphoma, stem cell transplant recipients, solid organ transplant patients, and some intensive care patients (for clinical diagnostic purposes and not for screening). The respiratory cultures were sputum, induced sputum, and BAL). Briefly it showed that the positive predictive value (PPV) was 35.4% (PCR pos and culture pos) and the negative predictive value (NPV) was 99.2% (PCR neg and culture neg). For the sub-groups: there was NPV 100% in HCAP patients, 98.4% in CAP and 97.7% in HAP patients. Criteria for health care associated pneumonia was radiologic evidence of infiltrate or cavitation and hypo/hyperthermia, RR>20, cough, and hypoxia<90%, inc sputum production, and leukopenia/leukocytosis.
This group had a great NPV for MRSA pneumonia however there are some limitations that I can come up with. Risk factors for infection including medical and social variables, PCR swab technique which effects sensitivity, timing of MRSA swab screening, etc. Also it is conceivable to believe that MRSA targeting antibiotics were given prior to the MRSA swab which would affect the false negative patients with infection. But this study is a good start. It’s also important to note that 30 day mortality and length of hospital stay was not statistically different in their cohort between those who received empiric abx and those who did not. But this non significant difference can also be related to the small sample size they had.
Another meta-analysis of 22 studies with 5163 patients by Parente, et al published in Clin Infect Dis in 2018 showed calculated PPV of 44.8% and 96.5% NPV, particularly in cases of MRSA HCAP/CAP.
Most other literature have the same type of results. This tells me that if the MRSA screen is neg, the patient has a high likelihood that he or she does not have MRSA pneumonia. In that case, the patient could be taken off vancomycin or other MRSA-active agents. And this is good because piperacillin-tazobactam cover MSSA.
DK
