To the Editor—When healthcare-associated pneumonia (HAP) is suspected, broad-spectrum antibiotics, often including vancomycin, are recommended to include coverage for methicillin-resistant Staphylococcus aureus (MRSA). Reference Kalil, Metersky and Klompas1 However, a primary goal of antimicrobial stewardship is to rapidly de-escalate these broad-spectrum antibiotics, based on culture results or other clinical data. Reference Redfield and Khabbaz2 A causative organism can be identified in some HAP cases and the antibiotics can be appropriately tailored, but in many cases no pathogen is identified, making de-escalation more challenging. To combat this situation, intranasal MRSA screening has been suggested as a stewardship tool because the absence of MRSA in the nares has been found to have a high negative predictive value for MRSA pneumonia. Reference Parente, Cunha, Mylonakis and Timbrook3
Universal decolonization by applying chlorhexidine gluconate (CHG) to skin and mupirocin to the nares has been shown in large multicenter studies to reduce MRSA and other bloodstream infections. Reference Lee, Bartsch and Wong4 Due to concerns that widespread use of mupirocin may lead to increasing resistance, many institutions have elected instead to use the antiseptic povidone-iodine (P-I) due to its antistaphylococcal properties and similar outcomes. Reference Hill and Casewell5–Reference Block, Robenshtok, Simhon and Shapiro7
Although sensitivity of culture-based MRSA screening may decline in the setting of antimicrobial or antiseptic use, PCR can also detect nonviable bacteria. Reference Shenoy, Noubary and Kim8 However, after our institution adopted universal nasal decolonization in addition to CHG bathing in our intensive care units (ICUs), staff raised concerns that MRSA nasal screening would no longer be accurate, and data were unable to be identified on this topic in the literature. The goal of this study was to determine whether nasal decolonization with P-I diminishes the utility of polymerase chain reaction (PCR)–based nasal MRSA screening.
Methods
We conducted a prospective cohort proof-of-concept study at our 1,200-bed community-based academic healthcare system from February to July 2019, with an enrollment goal of 20 participants for a convenience sample. Participants were eligible if they were aged ≥18 years, had been admitted to a medical ICU or stepdown unit, and had undergone baseline MRSA nasal screening by PCR (GeneXpert, Cepheid, Sunnyvale, CA) as ordered by their provider that was positive for MRSA. We excluded patients whose expected ICU or stepdown unit length of stay was <48 hours, those whose initial MRSA screen was performed after ≥2 doses of nasal P-I, and those who did not have nasal P-I decolonization performed for any reason (eg, allergy, patient refusal). Participants could only be enrolled once in the study. Verbal informed consent was obtained from each participant.
Intranasal P-I (Aplicare, 7.5%) was applied twice daily for 5 days or until ICU discharge, according to the protocol. Due to availability, Medline P-I (10%) was used from June 2019 until the end of the study. All positive PCR results underwent confirmatory testing via nonquantitative culture using MRSA-specific media (CHROMagar). All baseline PCR-positive results were confirmed via culture. Follow-up MRSA PCR tests were obtained after 4–6 days, immediately prior to intranasal P-I application, or at least 8 hours after the most recent application. All follow-up samples also underwent confirmatory culture. We used descriptive statistics to define patient demographics and calculated sensitivity of MRSA PCR, using culture as the comparative standard for our study. We conducted all analyses in SAS version 9.4 software (SAS Institute, Cary, NC). This study was approved by the ChristianaCare Institutional Review Board.
Results
Of the 25 patients initially enrolled, 20 completed the study. Reasons for dropping out included refusal of P-I (n = 2) and inability to collect a follow-up MRSA screen (n = 5). Of these 20 patients, the median age was 72 years and most were female. Most (70%) were admitted with active infection suspected, and 40% had a known MRSA history. Most (60%) were admitted with suspected respiratory infections (Table 1). Patients underwent a mean of 8.1 nasal P-I applications (range, 4–13) prior to follow-up testing, with no significant difference in the number of doses between MRSA-positive and MRSA-negative results at follow-up (8.3 and 7.4, respectively). At follow-up, 16 (80%) of 20 remained MRSA positive via both PCR and culture. Of the 4 patients with negative follow-up results, 1 was negative by both PCR and culture, and 3 were PCR positive and culture negative. All 4 had received ≥1 doses of vancomycin, and 1 patient had received ≥1 doses of linezolid. The sensitivity of follow-up MRSA PCR testing was 100%.
Table 1. Demographics and Clinical Characteristics
Note. COPD, chronic obstructive pulmonary disease; HAP, healthcare-associated pneumonia; MRSA, methicillin-resistant Staphylococcus aureus; SD, standard deviation.
a Patients may have had >1 admitting diagnosis.
b Included anticholinergic syndrome (n=1), alcohol-related diagnoses (n=2), acute kidney injury (n=1), and shock esophagitis (n=1).
Discussion
In this study, PCR remained highly sensitive for nasal MRSA colonization even after multiple applications of P-I; therefore, prior receipt of nasal P-I should not deter MRSA screening for stewardship or other purposes. The limitations of this study included the small sample size, loss to follow-up due, change of P-I formulations during the study, inconsistent number of P-I doses prior to final MRSA screening, and lack of quantitative MRSA cultures to determine the dose response to P-I application. However, the fact that most patients remained culture positive after 4–13 applications raises concerns that P-I may be less effective than mupirocin for clearing nasal colonization or that off-the-shelf P-I is less effective than specific nasal formulations. Reference Rezapoor, Nicholson, Tabatabaee, Chen, Maltenfort and Parvizi9
We could not identify any direct comparisons of P-I and mupirocin for decolonization of human nares, although P-I has been shown to successfully decolonize the nares in the immediate treatment period. Reference Anderson, David and Scholz10 Notably, 80% of our study participants remained positive; thus, larger studies using quantitative cultures to further investigate the effectiveness of both nasal P-I and mupirocin are needed. But as even culture-negative samples remained PCR positive, stewardship teams should feel comfortable de-escalating anti-MRSA antibiotics based on a negative MRSA screen, even in the setting of nasal P-I use.
Acknowledgments
The authors thank the physician and nursing leadership of the ChristianaCare Medical ICU, Wilmington ICU and Transitional Medical Unit, as well as all the staff nurses, who assisted us in the planning and execution of this project.
Financial support
No financial support was provided relevant to this article.
Conflicts of interest
All authors report no conflicts of interest relevant to this article.
