A Pilot Metagenomic Study Demonstrating Virtual Reality Head Mounted Displays Utilized in Medical Education Are Reservoirs of Viable Pathogenic Microbes

Abstract

Virtual reality (VR) devices are increasingly being utilized within operating theaters and intensive care units where appropriate sanitation is vital to ensure that patients do not unnecessarily acquire hospital‐associated infections. The morphology of VR devices in conjunction with the variety of materials and internal components provides challenges to their repurposing. This study aimed to evaluate the microorganisms remaining on VR headsets following sanitation by laboratory staff in a medical education anatomy teaching facility. The external components and internal facial interface were swabbed and separately cultured on four AGAR plates (Horse Blood, Nutrient, bile Esculin, and Mannitol Salt). Colonies were counted, sampled, pooled and subsequently processed for shotgun metagenomic sequencing. A higher number of colonies were present on surfaces closest to the eyes and facial interface compared to the external components. Metagenomic analysis identified 27 pathogenic bacteria including 4 “ESKAPE” pathogens (Enterobacter sp., Staphylococcus aureus, Klebsiella spp. and, Escherichia coli) and numerous organisms associated with ocular infections. A broad range of antimicrobial resistance genes were identified conveying resistance to Methicillin, Aminoglycosides, Macrolides, Tetracyclines, and Polymixins. Further research is required to ensure that current sanitization practices of VR head mounted displays are appropriate within high‐risk hospital settings.

Metagenomic analysis identified 27 pathogenic bacteria including 4 "ESKAPE" pathogens (Enterobacter sp., Staphylococcus aureus, Klebsiella spp., and Escherichia coli) and numerous organisms associated with ocular infections. A broad range of antimicrobial resistance genes were identified conveying resistance to Methicillin, Aminoglycosides, Macrolides, Tetracyclines, and Polymixins.

1. Introduction

Virtual reality (VR) is being increasingly utilized within healthcare for the purposes of medical education, clinical interventions, and research (Goldsworthy et al. 2023a, 2023b). Whilst an increasing body of evidence from clinical trials and systematic reviews supports the fact that VR devices are effective for the management of psychological symptoms and pain in some populations, a lack of research has been undertaken on aspects relating to its implementation such as cleaning and infection control policies and procedures (Goldsworthy et al. 2025). Hospital‐acquired infections (HAIs) represent a significant cause of morbidity and mortality whilst placing an increasing strain on healthcare systems physical, economic, and human resources (Poudel et al. 2023). In response, robust antimicrobial stewardship programs and policies have been developed inclusive of key strategic aims targeting the prevention of patients acquiring healthcare‐associated infections thereby reducing the need for antibiotic prescription and preventing the development of antimicrobial resistance (WHO 2021).

In order to facilitate the widespread deployment of VR devices within healthcare, it is important to ensure that devices are able to be sanitized appropriately (Goldsworthy et al. 2024). This is especially important when considering the proximity of VR HMDs to the eyes, nares, and oral cavity during use and its increasing use within high‐risk populations within intensive care, palliative care, and cancer or high‐risk settings such as operating theaters (Goldsworthy et al. 2024). Hospital‐acquired infections are commonly caused by a broad range of bacterial pathogens inclusive of both aerobic and anaerobic organisms. Whilst Staphylococcus species remain a major cause of ocular infections, Gram negative bacilli, streptococci, and anaerobic bacteria contribute significantly to HAIs (Sandu et al. 2025). Anaerobes in particular contribute to surgical site infections. As a result, it is important to ensure that VR HMDs do not serve as reservoirs for these pathogens considering they are already being utilized by both surgeons and patients during surgery.

The Spaulding's Classification System seeks to appropriately dictate what medical equipment requires high or low levels of sanitization based on if they come in contact with intact skin, mucous membranes or sterile tissues (e.g., scalpels during surgery) (A Rational Approach to Disinfection and Sterilization: Centers for Disease Control and Prevention 2025). According to this classification system VR head‐mounted displays (HMDs) are currently commonly considered non‐critical devices that only come in contact with intact skin and therefore only require low‐level disinfection (Moore et al. 2021). In line with this recommendation, commonly employed strategies to sanitize VR HMDs to date have included alcohol sprays, sanitization wipes, and ultraviolet‐C (UV‐C) devices (Roberts et al. 2022). Additionally, the use of third‐party add‐on equipment such as silicone or disposable facial interfaces are commonly utilized within VR interventions within medical settings to reduce risk and improve ease of sterilization (Goldsworthy et al. 2023a, 2023b, 2024; Roberts et al. 2022). However, a lack of evidence exists to indicate if current cleaning protocols utilized within healthcare to sanitize VR head‐mounted displays are efficacious (Goldsworthy et al. 2024; Moore et al. 2021; Høeg et al. 2025). Previously published cleaning protocols and reviews have called for further research to investigate the microbial presence on VR‐HMDs and to ensure that risk management strategies are appropriate (Goldsworthy et al. 2023a, 2024; Moore et al. 2021). This pilot study seeks to explore if VR HMDs utilized within a medical education facility present an infection risk to users.

2. Methods

A pilot study of five VR HMDs was randomly selected from a medical education teaching facility within Queensland, Australia. The sample size (n = 5) was determined by availability of devices and included all devices utilized within an anatomy teaching laboratory. Devices were commonly utilized by staff and students and were cleaned utilizing 70% isopropyl alcohol wipes according to the laboratory protocols prior to storage.

2.1. Sampling of VR HMD

Devices were swabbed using sterile culture EZ II swabs (Becton Dickson) pre‐moistened with sterile saline. For sample collection, gloves were worn and changed between each swab to prevent cross‐contamination. Two samples of each VR HMD were undertaken. The first swab sampled external surfaces, likely to come in contact with the hands during donning and doffing but unlikely to come in direct contact the face (external surfaces). The second swab sampled surfaces which either come in direct contact with the skin or may come in contact with the eyelashes, eyelids, and perioral area during donning and doffing (internal surfaces). Specifically, the facial interface, the lenses, and the areas surrounding the lens (Figure 1).

Figure 1.

Internal and external surfaces.

2.2. Culture of VR HMD

Swabs were immediately cultured on four types of media (horse blood agar, nutrient agar, bile esculin agar, mannitol salt agar) at 37°C for 48 h with 5% CO₂. Plates were labeled to enable identification of the VR HMD for post‐culture analysis. Following incubation, individual plates were photographed and, colonies pertaining to the same VR HMD collected and pooled according to the VR HMD and sample location (external, internal) within a liquid broth medium. Pooled samples were subsequently frozen at −80°C. Colony‐forming units (CFUs) were counted and recorded (Table 1). Indirect (post culture) metagenomic analysis was undertaken in favor of direct sequencing of swabs to investigate if VR HMDs are indeed fomites which can be contaminated with viable pathogenic microorganisms. Post culture (indirect) sequencing enables enrichment of viable microbial populations prior to DNA extraction by increasing the biomass available for downstream metagenomic sequencing, thereby improving DNA yield and reducing the influence of low‐abundance or degraded environmental DNA that may not reflect viable microorganisms. Whilst indirect sequencing confirms the viability of identified organisms, it restricts the identification of organisms to those which are culturable under the utilized conditions (NA, MSA, EBA, HBA) MRSA (ma. As a result, direct sequencing, without prior cultures of swabs, would identify a larger array of bacterial species present on HMDs.

Table 1.

Images of cultured agar plates from the external and internal surfaces of the five VR HMDs.

	Horse blood agar	Nutrient agar	Bile esculin	Mannitol salt agar
HMD 1 External
HMD 1 Internal
HMD 2 External
HMD 2 Internal
HMD 3 External
HMD 3 Internal
HMD 4 External
HMD 4 Internal
HMD 5 External
HMD 5 Internal

Abbreviation: HMD, Head‐mounted display.

2.3. DNA Extraction

DNA extraction of the 10 samples (five internal and five external) was undertaken at the Australian Center for Ecogenomics (ACE), Brisbane, Australia via bead beating with 0.1 mm diameter glass beads (BioSpec Products, Bartlesville, OK, USA) on a Powerlyser 24 homogenizer (MoBio, Carlsbad, CA, USA). Briefly, samples were transferred to a bead tube and 800 µL of bead solution (Qiagen, Germantown, MD USA) was added before subsequently bead‐beating for 5 min at 2000 rpm prior to centrifugation at 10,000 g for 1 min. Following the addition of 60 µL of cell lysis buffer, tubes were vortexed and then heated at 65°C for 10 min whilst being mixed at 1000 rpm, before being vortexed for another 30 s and stored at −20°C in preparation for DNA extraction. Prior to DNA extraction, samples were thawed at room temperature, vortexed and centrifuged for 1 min at 10,000 g. The resulting lysate was transferred to a new collection tube and DNA extraction carried out using DNeasy Powersoil Pro Kit (Qiagen), as per manufacturer protocol with a final elution volume of 50 µL within a sterile, EDTA‐free elution buffer.

2.4. Metagenomic Sequencing and Bioinformatic Analysis

Libraries were prepared according to the manufacturer's protocol using Nextera DNA Flex Library Preparation Kit (Illumina San Diego, CA, USA). Preparation and bead clean‐up were run on the Mantis Liquid Handler (Formulatrix) and Epmotion (Eppendorf) automated platform. On completion of the library preparation protocol, each library was quantified and, quality control (QC) was performed using the Quant‐iT dsDNA HS assay kit (Invitrogen, Carlsbad, CA, USA) as per manufacturer′s protocol. Library pooling, C and loading Nextera DNA Flex libraries were pooled at equimolar amounts of 2 nM per library to create a sequencing pool. The library pool was quantified in triplicates using the Qubit dsDNA HS assay kit (Invitrogen). Sequencing was carried out on the NextSeq. 500 (Illumina) using NextSeq. 500/550 High Output v2 2 × 150 bp paired end chemistry according to the manufacturers protocol. The post sequencing derived data was transferred into Illumina base space platform (https://basespace.illumina.com).

Following the sequencing runs, data as demultiplexed FASTQ files were uploaded and analyzed within the CosmosID platform for proper mining using highly dynamic comparator databases (GenBook®). Using this reference database, the resultant data were subject to a specific multi‐kingdom resolutive taxonomic identification analysis. Additionally, based on internal statistical scores from CosmosID, data were filtered to enable a listing of microbial or gene identification without further validation to determine their presence in each sample. Analysis retrieved from the CosmosID platform included taxonomic and relative abundance estimates for the microbial NGS datasets for the microbial population (bacteria, viruses, fungi, and protists) and genes (virulent factor and antibiotic‐resistant genes). A subgroup analysis of the “richness” (number of all distinct microbial strains on the HMDs) and the number of “hits” (number of sequencing reads assigned to a taxon/gene) was undertaken.

3. Results

3.1. Culture of Internal and External Surfaces

All VR HMDs were found to be contaminated with viable microorganisms following 48 h of culture (Table 1). Significantly more CFUs were cultured on internal surfaces (72%) than external surfaces (28%) (Figure 2). On average 95.4 colonies grew from swabs on external surfaces compared to 241 colonies from swabs of internal surfaces (Table 2). Additionally, there was a high degree of variability of colonies between the VR HMDs with HMD#3 accounting for 1011 colonies representing ~60% of all CFUs (Figure 3). Of note horse blood agar demonstrated diverse hemolytic patterns including beta hemolysis. Additionally, the bile esculin agar indicated the presence of enterococci and streptococcus species.

Figure 2.

Distribution of CFUs according to HMD surface.

Table 2.

Colony‐forming units of the external and internal surfaces of the five devices.

	Horse blood agar	Nutrient agar	Bile esculin	Mannitol salt agar	Total	Mean ± SD
HMD 1 External	21	20	7	7	55	11 ± 9.14
HMD 1 Internal	198	36	19	1	254	50.8 ± 83.6
HMD 2 External	22	8	11	2	43	8.6 ± 8.7
HMD 2 Internal	59	28	2	0	89	17.8 ± 25.91
HMD 3 External	284	10	10	1	305	61 ± 124.75
HMD 3 Internal	590	21	80	15	706	141.2 ± 252.72
HMD 4 External	15	13	8	5	41	8.2 ± 6.06
HMD 4 Internal	82	21	8	14	125	25 ± 32.79
HMD 5 External	13	13	6	1	33	6.6 ± 6.27
HMD 5 Internal	16	9	5	1	31	6.20 ± 6.53
Total	1300	179	156	47	1682
Mean ± SD	260 ± 185.81	35.8 ± 9.07	31.2 ± 23.07	9.4 ± 5.6

Abbreviation: HMD, head mounted display.

Figure 3.

Number of CFUs for each HMD.

3.2. Overview of Metagenomic Sequencing

Samples were sequenced to an average depth of 2.58 Gb/s (range: 2.05–3.21) corresponding to an average 19.75 million reads per sample. Metagenomic sequencing identified the presence of 114 distinct bacteria strains (richness) across the five HMDs with 186 overall hits. Despite displaying a greater number of CFUs on the interior surfaces richness (number of distinct microbial strains within the sample) and hits (number of sequencing reads assigned to a taxon/gene) were both lower (richness:65 and hits:86) on the interior HMD samples than was exterior HMD samples (richness: 80 and hits: 100). In total the 22 Gram negative bacterial strains (19%) and 92 Gram positive strains (81%) were identified. The distribution of sequencing classifications hits was similar, with 16% of assigned reads mapping to Gram negative bacteria and 84% to Gram positive bacteria. Bacteria associated with the face, gastrointestinal tract and urogenital tract were all identified (Figure 4). Thirty‐one unique antimicrobial resistant genes were identified with a total of 109 hits, 53 on the interior and 56 on the exterior. A total richness of 139 virulence factor genes was identified with 103 and 210 hits on the interior and exterior surfaces, respectively. In addition to bacteria, two fungi Balansia obtecta and Melampsora pinitorqua were identified.

Figure 4.

Bacteria associated with facial commensals, the oral cavity and lower gastrointestinal tract.

3.3. Bacteria Associated With the Gastrointestinal Tract

Six bacteria from four HMDs were identified (Streptococcus salivarius, Prevotella buccae, Porphyromonas gingivalis, Granulicatella elegans, Gemella sanguinis, and Actinomyces oris) which are commonly associated with the oral cavity (Delorme et al. 2015; Reyes 2021; Ohara‐Nemoto et al. 2005; Könönen and Gursoy 2022; Cavalcanti et al. 2017; Torres‐Morales et al. 2023). Four of these bacteria were found on internal surfaces with three on external surfaces. An additional nine bacteria (Prevtolla copri, Phocaeicola vulgatus, Bacteroides uniformis, Bacteroides stercoris, Alistipes putredinis, Alistipes finegoldii, Agathobacter rectalis, Klebsiella spp., Escherichia coli, Enterobacter hormaechei) were identified which are commonly associated with the lower gastrointestinal tract (Radka et al. 2020; Wang et al. 2025; Morita et al. 2023; Ryu et al. 2024). The majority of which are considered normal, beneficial bacterial commensals which contribute to healthy gut microbiota by assisting in fiber fermentation, short‐chain fatty acid production, immune modulation and, prevention of pathogenic bacterial colonization through microbial antagonism (Radka et al. 2020; Wang et al. 2025; Morita et al. 2023). However, when present within sterile areas may cause life‐threatening infections (Wang et al. 2025).

3.4. Potentially Pathogenic Organisms and Associated Antimicrobial Resistance

A total of 27 known pathogens were identified. Four of the six ESKAPE pathogens, Enterobacter sp., Staphylococcus aureus, Klebsiella spp., and Escherichia coli, were identified (Woh and Zhang 2025). Two HMDs were colonized with Klebsiella spp. with Escherichia coli and Enterobacter spp. colonizing the external surface of a single HMD. Staphylococcus aureus was identified on 4/5 HMDs on both internal and external surfaces. Of note Staphylococcus aureus was co‐located (located within the same sample but not necessarily within the same bacteria) with an antimicrobial resistant gene (ARG) responsible for providing resistance to methicillin suggesting the presence of MRSA (mecA gene) as well as ARGs associated with increased transcriptional expression of mecA (mecR1 gene) and encoding beta lactamase (blaZ gene) (Touaitia et al. 2025).

Genes associated with targeted bacterial protein synthesis were commonly identified (total hits 72, internal 40). These genes primarily convey resistance to five common classes of antimicrobials including Macrolides (total hits 36, internal 20), Aminoglycosides (e.g., aac(6′)‐aph(2”), total hits 10, internal 5), Tetracyclines (total hits 9, internal 5), Mupirocin (total hits 8, internal 4), Fusidic acid (total hits 6, internal 4) and Phenicols (total hits 3, internal 2). Cell wall targeting antibiotics associated with beta lactams and Fosfomycin were also identified (total hits 2, internal 10). Of note one HMD displayed the bcrC gene associated with conferring polymyxin resistance.

In addition to Staphylococcus aureus, several bacteria associated with eye infections were identified including Bacillus spp., Klebsiella spp., Escherichia coli, and a high number of coagulase‐negative staphylococci (CoNS) (Tilahun et al. 2025; Li et al. 2025; Natarajan et al. 2025). These CoNS, inclusive of S. haemolyticus, S. warnerii, S. hominis, S. capitis, S. epidermidis, S. capreae, and S. auricularis. These bacteria were commonly co‐located with numerous genes associated with the treatment of eye infections including Methicillin (e.g., mecA, hits 4), Aminoglycosides (richness 3, hits 10), Macrolides (richness 10, hits 36), Tetracyclines (richness 3, hits 9) and a gene associated with polymyxins (bcrC) were identified across the five HMDs. This pattern of co‐location of these bacteria with these respective genes raises the possibility for not only methicillin‐resistant Staphylococcus aureus and Staphylococcus epidermidis, a common concern for ocular infections, but also the presence of multidrug‐resistant pathogens.

Spore‐forming facultative anaerobic bacteria were found on HMDs and include several different species of avirulent environmental bacilli such as Bacillus anthracis, Bacillus thuringiensis, Bacillus wiedmannii, Bacillus paranthracis, Bacillus licheniformis, Bacillus velezensis, Bacillus haynesii, Bacillus altitudinis, Bacillus mobilis, Bacillus albus, Bacillus tropicus, Brevibacillus agri, Brevibacillus parabrevis, and Rummeliibacillus pycnus (richness 14, hits 26). In combination with the identification of a gene (alo) encoding cholesterol‐dependent cytolysin pore‐forming toxins, which promote cell lysis through the insertion of large B‐barrel pores in the host cell membranes resulting in a disruption of osmotic pressure and cell lysis. In addition to ARGs, a gene associated with quaternary ammonium efflux pump (qacC) in Staphylococcus aureus was located on all HMDs. On 3/5 HMDs an avirulent environmental strain of Bacillus anthracis, which lacked key virulence genes to support endotoxin production and immune evasion required to cause anthrax, was identified alongside numerous virulence factor genes (richness 55, hits 108) (Table 3).

Table 3.

Details of samples relating to Bacillus anthracis.

Sample	Taxonomic Id	% Unique Match	Read frequency
RB5809‐S7	1234146	1.09	8315
RB5809‐S7	637380	1.96	24,707
RB5815‐S13	2653217	18.18	1,482,342

Total unique match percentage relates to k‐mers that are unique to Bacillus anthracis divided by the total number of pre‐calculated unique k‐mers available for Bacillus anthracis in the CosmosID database. All data was filitered utilizing internal statistical scores utilizing %Unique match within the CosmosID platform assist in preventing false positive calls.

4. Discussion

This pilot study provides the first metagenomic evidence that VR HMDs used in a medical education setting may act as reservoirs for viable microorganisms, including potentially pathogenic and antimicrobial‐resistant organisms. Despite adherence to existing cleaning protocols by laboratory staff outside of the research team using isopropyl alcohol wipes, all five HMDs sampled were found to be contaminated with viable bacteria. Internal surfaces, those more likely to come into direct contact with users' skin, eyelids, or eyelashes, yielded significantly more CFUs than external surfaces, highlighting a key transmission point that may not be sufficiently mitigated by current sanitization practices. Indirect (post culture) sequencing of colonies demonstrated the potential for VR HMDs to be colonized by bacteria associated with both the oral, nasal, and lower intestinal tract as well as ocular pathogens. Of note, the use of indirect post culture sequencing restricted the identification of microbes to those culturable under the utilized conditions therefore underestimating the diversity of microbes present of the VR HMDs.

Our findings align with and extend upon a recent systematic review which highlighted the heterogeneity and potential inadequacy of VR cleaning protocols currently in use across the healthcare industry (Goldsworthy et al. 2024). These protocols often rely on the Spaulding Classification System, which categorizes medical devices based on the type of tissue they contact (intact skin, mucous membranes, or sterile tissue) (Shenoy et al. 2025). However, since its introduction in the 1950s medical technology has evolved necessitating updates to this classification system to prevent healthcare‐associated infections (Rowan et al. 2023). Under the Spaulding's Classification system, VR HMDs are typically classified as “non‐critical” and thus only require low‐level disinfection (Moore et al. 2021). However, this classification may fail to capture the real‐world manner in which VR devices are used in healthcare, particularly in high‐risk environments such as intensive care units, operating theaters, or oncology wards. During donning and doffing, HMD lenses may contact eyelashes and eyelids, structures adjacent to the conjunctival mucous membranes, potentially allowing for pathogen transmission. Moreover, users may contaminate their hands when adjusting the device, which combined with touching their eyes and face after use poses a route for infection.

This concern is exacerbated by our metagenomic findings, which identified four of the six ESKAPE pathogens (Staphylococcus aureus, Escherichia coli, Enterobacter hormaechei, and Klebsiella spp.), organisms well‐recognized for their multidrug resistance and clinical severity (Miller and Arias 2024). The presence of Staphylococcus aureus on four out of five HMDs, in conjunction with the detection of mecA, mecR1 and blaZ genes, suggests that methicillin‐resistant S. aureus (MRSA) may be present on these devices (Niemeyer et al. 1996). Notably, Bacillus anthracis, a pathogen of high concern, was detected on 3/5 HMDs along with numerous virulence factors. As this was identified via indirect (post culture) metagenomic sequencing it suggests metabolically active live cells, or spores capable of germinating were present on the HMDs at the time of sampling. Despite this examination of the genomic data revealed a lack of genes encoding virulence factors such as pag, cya, and lef responsible for toxin production or capB, capC, capA, and dep associated with capsule synthesis and depolymerization (Koehler 2002). As a result the organism was likely avirulent and not the causative agent of anthrax (Koehler 2002).

Strict anaerobes were identified on the HMDs including Prevotella copri, Prevotella buccae, Porphyromonas ginigvalis, Phocaeicola vulgatus, Phocaeicola massiliensis, Moraxella osloensis, and Moraxella cinereus. Whilst these bacteria are considered commensal gastrointestinal and oral flora they are known to cause invasive infections when mucosal membranes are disrupted or in immunocompromised individuals. The presence of these bacteria therefore highlight that, within healthcare settings especially, care should be taken to thoroughly sanitize VR HMDs between users.

The detection of bacteria associated with both the oral cavity and lower gastrointestinal tract further supports the hypothesis that VR HMDs may be cross‐contaminated via hand contact, aerosolisation, or inadequate cleaning. Additionally, 31 antimicrobial resistance genes (ARGs) were detected, including those conferring resistance to aminoglycosides, macrolides, tetracyclines, polymyxins, and disinfectants (e.g., qacC, linked to reduced susceptibility to quaternary ammonium compounds) (Zhang et al. 2015). These findings raise important questions about the adequacy of existing low‐level disinfection protocols in preventing the transmission of healthcare‐associated pathogens within higher risk healthcare settings.

However, the development of universal guidelines for VR HMDs is complicated by the heterogenous designs, and comprising of mixed materials including plastic, foam, and fabric and glass. Some newer models (e.g., Meta Quest 3) have incorporated less porous materials in their facial interfaces, potentially improving cleanability. However, many current models retain porous, absorbent components that may harbor microbes even after surface disinfection. Internal structures of headsets, such as vents and lenses, are rarely sanitized yet may serve as microbial reservoirs or dissemination points. In parallel technologies such as augmented reality (AR) devices used in surgery, internal fans intended to cool the device, and user may inadvertently circulate pathogens toward open surgical fields. A more recent classification system for determining sanitization requirements proposed by Kremer et al. acknowledges the need for healthcare providers and policy makers to both consider the use case and the complexity of the devices when determining the appropriate sanitization standards. Both of these classification systems however, fail to outline best practice relating to equipment storage. Many current HMDs come with boxes lined by fabric which is difficult to clean appropriately, may harbor and accumulate microbes over time and is therefore unsuitable for commercial use.

Overall, these findings underscore the need for updated, evidence‐based VR HMD disinfection guidelines that reflect their real‐world use in healthcare, particularly in high‐risk settings. Within healthcare settings these protocols should consider the entire clinical workflow which also commonly incorporates the use of other contaminated fomites such as tablets or mobile phones which enable providers to monitor the patient's experience (Olsen et al. 2023a, 2023b, 2025). Whilst various ultraviolet‐C (UV‐C) devices are currently commercially available they still require validation prior to widespread integration as part of a standard sanitization strategies within healthcare settings. This validation should not only ensure that microorganisms are effectively deactivated, and genetic material denatured within areas where shadows may prevent adequate disinfection but to also ensure that UV‐C resistant genes do not develop with repeated use.

4.1. Limitations and Future Recommendations

The current study is limited primarily by its low number of VR HMD samples and the lack of heterogeneity in the VR HMDs utilized which prevents the current findings from being generalized beyond this study. Whilst this pilot study successfully highlights that VR HMDs have the potential to act as fomites the indirect (post culture) sequencing approach restricts the identification of organisms to those culturable on the specific culture conditions. As a result, they inherently do not identify a large portion of the microbiome (e.g., fastidious bacteria, viruses, anaerobes) which are important considerations considering the close proximity of the VR HMDs to the eyes, nares and oral cavity. Additionally, the depth of sequencing has likely limited the specific identification of some virulence factor genes and ARGs which would provide further insights into potential pathogenicity of the identified bacteria.

Future studies should expand upon these findings by evaluating a larger number of devices, including additional VR HMD models, across varied clinical settings, comparing the efficacy of different disinfection methods, and determining whether viable pathogens with clinical relevance persist despite current cleaning practices utilizing direct culture methods. Until such evidence is available, it may be prudent for healthcare institutions to adopt precautionary protocols and policies when deploying VR in high‐risk environments or with patients under droplet or contact precautions.

5. Conclusion

This pilot study provides strong evidence that VR HMDs used in healthcare settings can act as reservoirs for viable, potentially pathogenic, and antimicrobial‐resistant microorganisms, including MRSA and multiple ESKAPE pathogens. Despite adherence to standard alcohol‐based disinfection protocols, widespread contamination, particularly on internal surfaces, was identified, raising serious concerns about the adequacy of current cleaning practices. Given the close contact with users' skin, eyes, and mucous membranes, and the frequent use of these devices in high‐risk environments, current low‐level disinfection classifications may be insufficient. These findings underscore the urgent need for robust, evidence‐based sanitization guidelines. Healthcare providers should be diligent when undertaking sanitization of HMDs and consider the risk benefit when utilizing devices with patient's requiring droplet or contact precautions.

Author Contributions

Conceptualization: Adrian Goldsworthy, Matthew Olsen, Lotti Tajouri. Data curation: Adrian Goldsworthy. Formal Analysis: Adrian Goldsworthy, Matthew Olsen, Lotti Tajouri. Funding acquisition: Adrian Goldsworthy, Matthew Olsen, Mohd Fairuz Shiratuddin, Simon McKirdy, Kok Wai Wong, Lotti Tajouri. Investigation: Adrian Goldsworthy, Matthew Olsen, Lotti Tajouri. Methodology: Adrian Goldsworthy, Matthew Olsen, Lotti Tajouri. Project administration: Adrian Goldsworthy, Matthew Olsen, Mohd Fairuz Shiratuddin, Kok Wai Won, Lotti Tajouri. Resources: Adrian Goldsworthy, Matthew Olsen, Mohd Fairuz Shiratuddin, Kok Wai Wong, Lotti Tajouri. Software: Lotti Tajouri. Supervision: Matthew Olsen, Mohd Fairuz Shiratuddin, Simon McKirdy, Rashed Alghafri, Abiola Senok, Hamda Alfalasi, Kok Wai Wong, Lotti Tajouri. Validation: Adrian Goldsworthy, Matthew Olsen, Rashed Alghafri, Abiola Senok, Lotti Tajouri. Visualization: Adrian Goldsworthy, Matthew Olsen, Lotti Tajouri. Writing – original draft: Adrian Goldsworthy. Writing – review and editing: Adrian Goldsworthy, Matthew Olsen, Mohd Fairuz Shiratuddin, Simon McKirdy, Rashed Alghafri, Abiola Senok, Hamda Alfalasi, Kok Wai Wong, Lotti Tajouri.

Ethics Statement

The authors have nothing to report.

Conflicts of Interest

The authors declare no conflicts of interest.

Data Availability Statement

Data is available upon reasonable request to the corresponding author.