Enhanced Medical and Community Face Masks with Antimicrobial Properties: A Systematic Review

Face masks help to limit transmission of infectious diseases entering through the nose and mouth. Beyond reprocessing and decontamination, antimicrobial treatments could extend the lifetime of face masks whilst also further reducing the chance of disease transmission. Here, we review the efficacy of treatments pertaining antimicrobial properties to medical face masks, filtering facepiece respirators and non-medical face masks. Searching databases identified 2113 studies after de-duplication. A total of 17 relevant studies were included in the qualitative synthesis. Risk of bias was found to be moderate or low in all cases. Sixteen articles demonstrated success in avoiding proliferation (if not elimination) of viruses and/or bacteria. In terms of methodology, no two articles employed identical approaches to efficacy testing. Our findings highlight that antimicrobial treatment is a promising route to extending the life and improving the safety of face masks. In order to reach significant achievements, shared and precise methodology and reporting is needed.


Introduction
The present pandemic due to coronavirus disease (COVID-19) significantly impacted the health of millions of people and highlighted weaknesses in the personal protective equipment (PPE) supply chain around the world. Face masks have been used in medical settings for infection prevention for decades, being one of the most important countermeasures in mitigating high risk of droplet and aerosol transmission of pathogens in health care settings [1]. Medical masks and FFP2 or N95 respirators are recommended for Health Workers (HWs) when providing care to suspected or confirmed COVID-19 patients, especially if performing aerosol generating procedures, which requires respirators to be worn continuously [2]. Patients must wear medical face masks for in person care to control sources of infection [3]. Expanded use of masks has resulted in increased wear time and use without training. Although there is precedent for mask wearing among the general public in Asian countries such as China, South Korea and Japan, this only became a global strategy in reaction to the spread of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). Mask wearing is generally considered a low risk and low-cost approach for disease control among the general public, facilitating source control from asymptomatic persons unknowingly transmitting virus. It is accepted that non-medical (community) masks are generally sufficient for this purpose, with members of the public having a lower transmission risk relative to that of HWs. However, certain designs of community masks are considered unfit for use (e.g., those containing respiratory exhalation valves do not 2 of 16 offer source control) and use of medical grade PPE by the general population contributed to shortages [4].
In this context, many researchers are investigating the reprocessing and reuse of masks and filtering facepiece respirators (FFPs/FFRs), focusing on meeting regulatory standards and using readily available equipment present in hospitals. Decontamination is of paramount importance to any PPE re-use. Research in this area grew so fast in reaction to COVID-19 that literature reviews became available within one year [5][6][7][8]. Reprocessing or decontamination must ensure devices keep their original properties (e.g., filtering and breathability), functional integrity, shape and there must be no residual toxicity for the wearer [9,10]. Antimicrobial enhanced fabrics could be used to engineer masks and respirators allowing not only a longer lifespan of the mask, but also the possibility to exploit novel routes for mask decontamination. Antimicrobial systems can be broadly grouped into categories, as reported in a recent revision of the literature on PPE for health applications: [11] metal oxides and nanoparticles; salt compounds; graphene-based materials; quaternary ammonium compounds (QACs); N-halamine-based compounds; and naturally derived antimicrobial agents. Of these, nanoparticles, such as copper oxide, graphene oxide nanosheets [12] and plant extracts [13] have been investigated for decontamination methods.
There is no comprehensive systematic review available for masks fabricated with antimicrobial properties, their efficacy against bacteria or viruses or the possibility of reprocessing with the necessary durability and safety. Current standard methods for testing efficacy of antimicrobials are typically specific to the pathogen, i.e., bacterial (ISO 20743, AATCC TM100), viral (ISO 18184) or fungal (ISO 13629). General chemical safety assessments such as REACH chemical safety have limited application to modified fabrics for masks. Any recommendations concerning the development and production of PPE requires access to the best available evidence. Therefore, the present work provides a systematic review of the literature addressing antimicrobial materials and treatments for medical face masks, FFRs and community masks. The aim is to provide evidencebased recommendations on antimicrobial treatments for respirators and masks, especially regarding efficacy, reliability and safety to the wearer, and their possible role in facing the current pandemic and future healthcare crises.

Search Strategy and Selection Criteria
This study was conducted in accordance with the preferred reporting items for systematic reviews and meta-analyses (PRISMA) guidelines [14]. The protocol was designed prospectively and is available upon request. Studies were included if they met the following criteria: original peer-reviewed studies describing "augmented" masks by antimicrobial agents reporting at least one laboratory test in order to assess the efficacy.
MEDLINE and Embase databases were searched to identify studies from 1 January 2010 till 1 January 2021. Bibliographies of relevant articles were assessed as a secondary source of studies. The literature search was performed and verified by two independent reviewers using the index terms grouped in three categories: device (respirator OR mask OR filtering OR nonwoven OR fabric OR electro AND spun OR textile OR personal AND protection AND equipment), type of active augmentation (antimicrobial OR antiviral OR nanoparticles OR nanotechnology OR viricidal OR biocidal OR bactericidal OR inactivation) and organism affected (COVID-19 OR bioaerosol OR airborne OR coronavirus OR virus OR respiratory AND infections).
Two authors independently evaluated all retrieved studies against the eligibility criteria and divergent opinions were resolved, achieving consensus through discussion with a third author. Articles were excluded if there was not sufficient documentation. Reviews, duplicate publications and editorials were also excluded.

Data Analysis
Data were extracted independently by two reviewers and entered into standardized spreadsheets. Any disagreement was resolved, achieving consensus via discussion with a third reviewer. The following data were extracted: type of article, publication year, type of substrate, type of antimicrobial system, integration methodology, viruses and/or bacteria tested and methods, antimicrobial efficacy results and secondary outcomes.

Outcomes
The main outcome was antimicrobial efficacy of antimicrobial treatments applied to PPE, measured by Logarithmic Reduction Value (LRV) of Colony Forming Units (CFU). This review also focused on methodological aspects, such as methods employed to assess antimicrobial efficacy, antimicrobial technology employed, how the treatment was integrated with the production process, which comparator was employed in each study, which pathogen was used and how it was applied to the PPE. Finally, a number of secondary outcomes were systematically investigated, aiming at assessing the impact of antimicrobial treatment on PPE fundamental properties, including: breathability; filtering capacity; reusability; impact on PPE cost/production; stability and durability of treatment; and safety for the wearer (e.g., toxicity via inhalation of antimicrobial treatment substances, skin irritations or respiratory inflammation).

Risk of Bias
Due to lack of standardized tools for assessing study results and risk of bias in this field, we employed previously developed objective assessment criteria [10,15]. This consisted of a predetermined evaluation matrix, containing information on study design, methodological consistency, population heterogeneity, sampling bias and selective reporting (Table S3 in Supplementary Materials). We adapted the original matrix, removing categories not relevant to this review. The assessment was made by two reviewers independently and divergences were overcome with consensus.

Role of Funding Source
The funder was involved in defining the scope of the work. Study design, data collection, data interpretation and report writing were completed independently of the funder.

Results
Searching MEDLINE and Embase yielded 2364 titles. After duplicate removal, 2116 titles/abstracts were screened, with 1982 excluded. Among the remaining 55 full-texts screened for compliance with the eligibility criteria, only 17 studies met the eligibility criteria and reported sufficient experimental results and methodological details for inclusion in the final analysis. Excluded full texts and reasons for exclusion are provided in Table S1. A flow chart representing the screening process is given in Figure 1. Table 1 contains a summary of the characteristics of the included studies covering three main areas: types of antimicrobial systems used, antimicrobial efficacy testing and results. A detailed account of all extracted information is provided in Table S2. Risk of bias was moderate or low in all studies, covering the following areas: study design, methodological consistency, population heterogeneity, sampling bias and selective reporting (Tables S3 and S4).
Antimicrobial properties of salt compounds were evaluated by two studies [27,30], in both studies transmission electron microscopy (TEM) highlighted structural damage and morphological changes in test pathogens, attributed to contact with the natural salt recrystallization process. Quan et al. applied a sodium chloride (NaCl) salt coating to a surgical mask polypropylene filtering layer [30]. Aerosolized viral strains: H1N1, PR/34 H1N1 and VN/04 H5N1 applied to coated fabric were determined to be inactive by TEM, due to hyperosmotic stress upon the viral envelope. A second study also evaluated NaCl, alongside potassium sulphate (K 2 SO 4 ) and potassium chloride (KCl) [27]. Bacteria strains: Klebsiella pneumoniae, methicillin-resistant S. aureus, E. coli, Pseudomonas aeruginosa and Streptococcus pyogenes showed time-dependent inactivation for all salt coatings. Best performance was seen with a three-layer NaCl filter, granting a 4log reduction within 30 min of aerosol exposure. Bacterial inactivation was confirmed in vivo using a mouse infection model.
In one study, laser-induced graphene (LIG) was synthesized into fabric using Polyimide substrate [28]. The antibacterial properties were tested alongside commercial samples of activated carbon face masks (activated carbon fibre (ACF)) and surgical masks (meltblown fabric, MBF), by submerging in E. coli suspension. A CFU assay showed 0.73 log reduction of E. coli after 8 h. The antibacterial activity of LIG was found to be high, 81.57%, compared to ACF (2.00%), and MBF (9.13%). Hydrophobic and hydrophilic LIG were compared and exhibited similar antibacterial activity. The authors propose this happens through different mechanisms; the former due to the abundant oxygen-containing functional groups such as −COOH and −OH, that may cause loss of intracellular substances due to charge transfer, the latter due to dehydration.
Two QACs were evaluated [19,21]. Tseng et al. [19] applied Goldshield 5 (QAC based commercial detergent), to a surgical face mask. Over 99.3% antibacterial efficiency was seen when aerosolized bacteria (Acinetobacter baumannii, Enterococcus faecalis and S. aureus) challenged the mask surface. Xiong et al. [21] modified the PP layer of a surgical mask using QAC//Hexagonal Boron Nitride/PP (QAC/h-BN/PP), forming a nanocomposite, activated surface. E. coli and S. aureus were incubated with test fabric samples following standard methods (ISO 22196 and JIS Z 2801). Antimicrobial rates of the QAC/h-BN/PP samples were 99.3% (E. coli) and 96.1% (S. aureus), based on optical density of recovered bacteria. A so-called 'contact killing' mechanism was confirmed by zone-of-inhibition testing, i.e., the system did not release biocidal compounds.
Two studies investigated naturally derived antimicrobial agents [29,31]. Duong-Quy et al. [29] reported a novel face mask (Lamdong Medical College (LMC) mask) containing an antimicrobial agent derived from the leaf oil of Folium Plectranthii amboinicii (Lour), a traditional Vietnamese medicinal plant used to treat upper respiratory infections, bronchitis and gastrointestinal infections. The LMC mask and a four-layer activated carbon surgical mask (positive control) were worn by randomized volunteers, followed by laboratory analysis of recovered bacterial growth. Antibacterial activity of the LMC mask was not statistically different to the positive control. Woo et al. [31] investigated dialdehyde starch (DAS), seeking alternatives to aldehyde antimicrobials, which are highly toxic to humans. Commercial filters, such as two cellulose filters (CFs) commonly used for air cleaning and a polypropylene FFR (PF), were modified with DAS aqueous suspension at different concentrations. The antimicrobial assessment was made using a nebulized solution of MS2 bacteriophage and artificial saliva, to emulate aerosols produced from sneezing or coughing. Relative survivability (RS) of MS2 viruses on filters treated with different concentrations of DAS suspension showed a clear biocidal effect for all the filters, with RS decreasing with increasing concentration of DAS. Table 2 contains a summary of the secondary outcomes addressed in the study pool. Three studies did not evaluate secondary outcomes [20,24,32]. The frequency of secondary outcomes investigated is given in Figure 3. No significant difference [22,27,28] or very slight increase [26] in pressure drop across treated vs. untreated substrates was seen, taken to suggest acceptable levels of breathability. Woo et al. [31] compare similarly treated CF and PF substrates. CFs showed improvement in both pressure drop (lowered) and filtration efficiency (increased) when treated with DAS, with no improvement seen for PFs. A single study reported reduced filtration efficiency after coating of commercially available NIOSH FFRs (N95 and P95) [25]. Two studies found no change in filtration efficiency following antimicrobial treatment [23,26], whereas salt coating increased filtration efficiency in one report [27]. Three studies evaluated toxicity [17,18,29]. One study evaluated particles released from the antimicrobial system, the authors demonstrated that copper eluted from their test mask was within the permissible exposure limit [17]. In contrast, Li et al. asked 20 volunteers to wear their nanoparticle-treated facemask, finding no reports of inflammation or itching after wearing. One study utilised a biomarker for respiratory inflammation to detect any inflammation potentially caused by wearing [29]. No formal cost analysis was offered, but several studies made statements about advantageous lowcost or easy production [17,20]. Durability/stability in terms of shelf life or storage was evaluated in four studies [19,26,27,30]. Tseng et al. found that their GS5 'decontamination effect' lasted a week after initial coating, concluding this would reduce cleaning costs and increase feasibility [19]. Kumar et al. claimed 'self-cleaning' properties (via nonwetting surface properties) reduced risk of exposure to pathogens on disposal [26]. Three studies investigated stability of the antimicrobial system under varied environmental conditions, to address storage considerations [16,27,30]. All systems were found to be stable to high temperature and humidity, taken to suggest safe long-term storage and reuse.
Fourteen studies considered secondary outcomes of which just three deal with a very important parameter which is safety for the wearer.  CFs but not PF RH = room humidity. MC = 1-Chloro-2,2,5,5-tetramethyl-4-imidazolidinone (a N-halamine monochlorinated compound). LIG = laser induced graphene. DAS = dialdehyde starch. CF = cellulose filter. PF = polypropylene filter. Fourteen studies considered secondary outcomes of which just three deal with a very important parameter which is safety for the wearer.

Discussion
The purpose of this literature review was to determine the efficacy of antimicrobial treatments applied to medical or community face masks. We focused on antimicrobial efficacy, methodological procedure and the impact of modifications on essential PPE properties. A total of 17 studies were included; of the excluded texts, 22 did not concern PPE, while 16 did not report antimicrobial efficacy. Overall, antimicrobial treatments were found to be effective. Yet, it is crucial to recognize the huge heterogeneity among studies, including technology employed, integration method, efficacy testing methods, challenge pathogens and control masks/fabrics. This is likely due to the fact that this is the first study systematically reviewing literature focusing on antimicrobial treatments for PPE. It is urgent to achieve standard methodology, in order to regulate community masks claiming

Discussion
The purpose of this literature review was to determine the efficacy of antimicrobial treatments applied to medical or community face masks. We focused on antimicrobial efficacy, methodological procedure and the impact of modifications on essential PPE properties. A total of 17 studies were included; of the excluded texts, 22 did not concern PPE, while 16 did not report antimicrobial efficacy. Overall, antimicrobial treatments were found to be effective. Yet, it is crucial to recognize the huge heterogeneity among studies, including technology employed, integration method, efficacy testing methods, challenge pathogens and control masks/fabrics. This is likely due to the fact that this is the first study systematically reviewing literature focusing on antimicrobial treatments for PPE. It is urgent to achieve standard methodology, in order to regulate community masks claiming to possess antimicrobial properties, becoming increasingly available on the market. Evidently, there is no standard method for assessing antimicrobial properties.
The very limited number of articles relating directly to antimicrobial systems for masks, relative to many focused on more general antimicrobial modified fabric, is noteworthy. We observed many articles investigating antimicrobial properties of augmented materials, postponing considerations of the final application to future research [33]. Other applications included skin wound care [34,35], water purification [36,37], air filters [38,39] and antimicrobial surfaces [40]. Unsurprisingly, most studies were published following outbreaks of epidemic-prone respiratory pathogens such as Avian influenza virus (H5N1) (2003) [41], H1N1 swine flu (2009/10) [42] and SARS-CoV-2 [43,44]. The majority of antimicrobial agents identified carry a body of evidence supporting their mechanism and utility [45]. The antimicrobial activity of copper and silver, especially in nanoparticle form, is well documented, including towards coronaviruses [46][47][48]. Indeed, N-halamines, graphene and QACs have proved effective against a broad spectrum of microorganisms, with long-term stability and durability [49][50][51]. The mechanism of graphene's antimicrobial action remains controversial, as discussed in Seifi et al.'s review of antibacterial properties of graphene. Several possible mechanisms were identified, including membrane stress, charge transfer, entrapment, oxidative stress, self-killing and photothermal, which may occur alone or in combination [51]. Photoactive chemicals producing reactive oxygen species are also considered effective and durable candidates for fabrication of antimicrobial materials [52,53].
The relative scarcity of research tackling antimicrobial systems for masks is compounded by a lack of shared procedures for evaluating efficacy and safety. Indeed, test methods were highly heterogenous amongst the identified studies, with antimicrobial assays varying from study to study (e.g., Sandwich test, ASTM Method F2101.01, CFU assay, zone of inhibition assay, etc.). This was also reflected in the challenge pathogens, varying greatly among bacteria, viruses and virus surrogates. Although all methods were scientifically valid, heterogeneity prevented direct comparison. In total, 16 of the 17 articles reported high efficacy of their augmented mask systems in either preventing proliferation or directly eliminating viral or bacterial pathogens. Of all systems considered, only one did not prove effective.
Interestingly, no studies attempted complete analysis of the mask system, covering fundamental properties of an airway protection device, i.e., filtering capacity, breathability (permeability to air, pressure drop), toxicity and longevity of the antimicrobial system and cost. Many studies evaluated some of these properties, while others provided antimicrobial efficacy results alone, leaving further analysis entirely to future work. An important aspect, safety of the product for the wearer, was addressed only in three studies, each with unique approaches. Only one study considered the possibility of reprocessing, authors claimed that their photoactive mask was able to 'self-sterilize' under solar irradiation, whilst maintaining its antiviral properties. Several articles discussed shelf life and reuse of their mask system, granted through the persistently active antimicrobial system, even when subject to variable humidity and temperature.
To improve homogeneity, considering the availability of community masks claiming antimicrobial efficacy, it is paramount that standards for testing antimicrobial efficacy and safety are created. Such standards should include well-defined results reporting, including appropriate comparators, considering not only the maximum value of LRV of CFU, but also reporting LRV at different time points, i.e., giving indications on the efficacy during wearing. Additionally, the maximum duration of the antimicrobial protection should always be addressed. The antimicrobial efficacy should be considered against viruses and bacteria in aerosolized and inoculated forms. All identified studies focused on reduction of transmission of airborne pathogens, none addressed transmission from contact with fluid secretions, i.e., touching of masks with contaminated hands, a major issue for non-professional users.
Moreover, safety is paramount, to avoid toxicity or adverse effects to the wearer, pre-existing standards must be met (e.g., REACH regulations from EU). Further, impact of modifications on key properties must be understood, namely filtration efficiency (e.g., ISO 21501-4; EN 14683:2019, EN 13274-7:2019; ASTM F2299) and breathing resistance (e.g., EN 14683:2019 (Annex C); ASTM D737; ISO 9237:1995). Lastly, treatment persistence should be considered and quantified (including limiting number of decontamination cycles in the case of reusable systems), including indications of storage/shelf life under normal conditions. Supplementary Materials: The following are available online at https://www.mdpi.com/article/ 10.3390/jcm10184066/s1, Table S1. A list of excluded studies and reasons for their exclusion, Table S2: Detailed information extracted from included studies, Table S3. Risk of bias tool, defining low, moderate and high risk of bias for each category, Table S4. Results of risk of bias assessment for each area and overall risk of bias score.
Author Contributions: F.B., L.P., U.B. and K.S., designed the study. F.B. and L.P. coordinated and supervised the study. F.B., U.B. and R.P. designed the data collection methodology. All authors acquired data, screened records, extracted data and assessed risk of bias. K.S. and F.B. wrote the original draft. All authors provided critical conceptual input; analysed and interpreted data; and critically revised and edited the manuscript. All authors have approved the final article. All authors have read and agreed to the published version of the manuscript.