Microbial linkage between arthropods and patients

Six studies used methods designed to demonstrate microbial colonization or infection of both patients and arthropods captured in the hospital. ^{Reference Almeida-Nunes, Marcilio and Oliveira12–Reference Hanrahan, Schouten, Fyffe, Jencson and Stroever17} However, only three were able to infer an association between arthropods and patients. Lima et al. cultured samples from patients and ants captured in a hospital in Brazil. ^{Reference Marques, Rodrigues and Rebelo13} Cultures from ants yielded 41 different species of bacteria and 18 from patients. Common isolates included Coagulase negative Staphylococcus spp. (both), Acinetobacter baumannii (both), Acinetobacter lwoffii (ant), Staphylococcus aureus (ant), and Klebsiella pneumoniae (patients). Though both sources were positive for similar bacteria, direct linkage was not demonstrated.

Fotedar and colleagues collected samples from both arthropods (houseflies) and patients with infected wounds (India, 1992). ^{Reference Fotedar, Banerjee and Samantray14} Klebsiella spp. was isolated from 36.7% of captured houseflies and 28.1% of sampled wound infections. Advanced methods to demonstrate genetic similarities were not available in 1991, limiting the ability to demonstrate a definite link. However, antibiograms for bacteria isolated demonstrated a high degree of resistance (82% in houseflies and 96.3% of Klebsiella spp. isolated from patients), while the antibiogram among controls showed only 8.7% of the samples were similarly resistant.

Almeida-Nunes et al. assessed healthcare transmission of dengue virus in Brazil. ^{Reference Almeida-Nunes, Marcilio and Oliveira12} Four patients hospitalized more than 10 days before exhibiting symptoms and demonstrating positive serology were identified. Previous studies had already demonstrated transmission via needle stick injuries, blood transfusions, transplantation, and other mucosal exposures. The investigators found mosquitoes within the emergency department positive for dengue, as well as breeding sites and viable larvae. They concluded that the likely source of infection was from the mosquitoes in the hospital.

A study from South Africa by Cotton et al. was the first to genetically demonstrate similarity between cockroaches captured in a neonatal intermediate care unit and patient isolates. ^{Reference Cotton, Wasserman and Pieper15} During an investigation of an outbreak of extended spectrum beta-lactamase (ESBL)-producing Klebsiella pneumoniae, the team observed cockroaches on “work surfaces, medical equipment such as incubators, in sharps containers, and even on infants.” Investigators collected environmental samples from the unit and nasal/rectal swabs from healthcare workers (possible carriers). The areas observed with cockroaches were positive for ESBL-producing K. pneumoniae, while the specimen from healthcare workers were negative. Bacterial isolates from cockroaches and patients demonstrated indistinguishable patterns on pulsed-field gel electrophoresis. The authors assert the evidence is not robust enough to prove a causal connection. However, the circumstantial evidence related to the outbreak, as well as negative cultures in controls, strongly support the likelihood of either direct (cockroaches crawling on infants) or indirect (cockroaches contaminating environmental surfaces) transmission via arthropod.

A similar study conducted by Hanrahan et al. identified a genetic association between cockroaches in an ICU and patients with HAI due to MDR Enterobacter cloacae (MRE). ^{Reference Hanrahan, Schouten, Fyffe, Jencson and Stroever17} Upon identification of the outbreak, the team implemented robust infection prevention measures and screened all new and existing patients in the unit for MRE. They cultured environmental surfaces, multidose vials, protein supplements, and hands of healthcare workers (HCWs). They also required prolonged closure of rooms after terminal cleaning to ensure rooms were negative for MRE prior to housing a new patient. Environmental surfaces were culture positive for MRE, and hands of HCWs and medications/supplements were negative.

Cockroaches were implicated as possible sources of MRE after being observed in the ICU. As further evidence supporting this theory, rooms left empty that were initially negative for MRE would turn positive while still closed. This suggested contamination by a source other than healthcare workers and patients. Cockroaches caught in the ICU were positive for MRE, confirmed to be genetically similar to patient isolates and environmental samples on PFGE. Again, the study design did not permit a direct causal link between the cockroach and patient. However, the authors concluded cockroaches contributed to environmental contamination and cross-transmission to patients. Enhanced pest management resulted in the resolution of the outbreak.

Finally, Hassan et al. used advanced molecular analyses to demonstrate clear linkages between arthropods, hospital surfaces (HS), and patients (Pakistan, 2021). ^{Reference Hassan, Ijaz and Khan16} The investigators collected samples from surgical site infections (SSIs), hospital surfaces, and arthropods to isolate MDR Enterobacteraciae in a public hospital of Pakistan. They sought specific resistance genes, determined antibiotic sensitivity profiles, and conducted whole genome sequencing to identify specific genotypes of bacteria and phylogenetic relationships between isolates. They also used single nucleopeptide polymorphism (SNP) analysis to study transmission events. Though it was impossible to determine temporal precedence across the various samples, investigators did find a high degree of similarity and genetic linkages between surgical site infections, the environment, and arthropods. For example, “the majority of E. coli linkages occurred between arthropods and SSIs (n = 14) and, to a lesser extent, between HSs and SSIs (n = 4).” ^{Reference Hassan, Ijaz and Khan16} The authors acknowledge that the study was not designed to show causality, rather the interconnectedness of clinically relevant pathogens across the hospital. However, it is the most rigorous study to demonstrate the likelihood of arthropods as vectors of transmission to date.

Microbial carriage of arthropods in healthcare settings

The majority of studies that met inclusion criteria for this review only assessed carriage of arthropods. Supplemental Tables 2 and 3 detail the pathogens identified from the alimentary tract or external surfaces of arthropods captured in the HE ^{Reference Faulde and Spiesberger2,Reference Almeida-Nunes, Marcilio and Oliveira12,Reference Cotton, Wasserman and Pieper15–Reference Faulde, Sobe, Burghardt and Wermter43,Reference Lise, Garcia and Lutinski45–Reference Elgderi, Ghenghesh and Berbash56} or in conjunction with a community control. ^{Reference Fotedar, Banerjee and Samantray14,Reference Khodabandeh, Shirani-Bidabadi, Madani and Zahraei-Ramazani30,Reference Rahuma, Ghenghesh, Ben Aissa and Elamaari57–Reference Felizatti, Ceccato, Cintra-Socolowski and Zeni Beretta77} All studies isolated microbes from arthropods. Only one study sought to isolate enteroviruses from cockroaches and failed to do so. ^{Reference Chadee and Le Maitre55}

Forty-five studies isolated Staphylococcus spp., nine of which isolated methicillin-resistant Staphylococcus aureus (MRSA). ^{Reference Hanrahan, Schouten, Fyffe, Jencson and Stroever17,Reference Sobur, Islam and Haque25,Reference Abdolmaleki, Mashak and Dehkordi27,Reference Pai, Chen and Peng35,Reference Gliniewicz, Czajka and Laudy36,Reference Aquino, Silveira and Pessoa48,Reference Russo53,Reference Šrámová, Daniel and Absolonová75} MRSA was isolated from cockroaches (German and American), various species of fly, and ants. Studies also examined multiple species of Enterococcus including clinically relevant E. faecalis and E. faecium. ^{Reference Marques, Rodrigues and Rebelo13,Reference Beatson Susan39,Reference Rupprecht, Moter and Wiessener42,Reference Fotedar, Banerjee, Singh and Verma63,Reference Fotedar, Banerjee, Samantray, Nayar and Verma64,Reference Davari, Khodavaisy and Ala66,Reference Jalil, Abdulhadi and Al-Ameer73,Reference Šrámová, Daniel and Absolonová75} Numerous studies identified Enterococcus to the genus level but did not distinguish by species (Supplemental Table 2). These bacteria were isolated from ants, cockroaches, houseflies, moth flies, spiders, beetles, and wasps.

Streptococcus spp. were isolated from houseflies, cockroaches, ants, and moth flies. ^{Reference Faulde and Spiesberger2,Reference Tilahun, Worku, Tachbele, Terefe, Kloos and Legesse22,Reference Kassiri, Kassiri and Kazemi28,Reference Zarchi and Vatani32,Reference Chehelgerdi and Ranjbar34–Reference Boiocchi, Davies and Hilton40,Reference Lise, Garcia and Lutinski45,Reference Do Nascimento, Amaral and Ferreira46,Reference Oliva, Díaz and Fuentes González54,Reference Oliva, Díaz and Fuentes González54,Reference Rahuma, Ghenghesh, Ben Aissa and Elamaari57,Reference Chiuya, Masiga, Falzon, Bastos, Fèvre and Villinger58,Reference Naher, Afroz and Hamid62–Reference Fotedar, Banerjee, Samantray, Nayar and Verma64,Reference Davari, Khodavaisy and Ala66,Reference Jalil, Abdulhadi and Al-Ameer73,Reference Rangel and De-Simone78} Once again, most were not identified to the species level. However, several studies isolated S. pyogenes from cockroaches and ants ^{Reference Chehelgerdi and Ranjbar34,Reference Beatson Susan39,Reference Fotedar, Banerjee, Singh and Verma63,Reference Pai, Chen and Peng71} and S. pneumoniae from cockroaches. ^{Reference Jalil, Abdulhadi and Al-Ameer73} Supplemental Table 2 details other Gram-positive cocci (GPC) isolated from various arthropods. However, few are noted to be clinically relevant except as opportunistic pathogens in immunocompromised patients.

Not unexpectedly, cockroaches, moth flies, cigarette beetles, and ants were found to harbor the foodborne pathogen Bacillus cereus. ^{Reference Faulde and Spiesberger2,Reference Watanabe, Takaku, Yokota, Hayashi, Tamaki and Kokeguchi33,Reference Beatson Susan39,Reference Rupprecht, Moter and Wiessener42,Reference Fotedar, Banerjee, Singh and Verma63,Reference Jalil, Abdulhadi and Al-Ameer73} Environmental and opportunistic pathogen, Acinetobacter baumanii, was isolated from houseflies, moth flies and ants. ^{Reference Faulde and Spiesberger2,Reference Marques, Rodrigues and Rebelo13,Reference Marques, Rodrigues and Rebelo13,Reference Heiden, Kurz and Bohnert24,Reference Lise, Garcia and Lutinski45} Additional species of Acinetobacter were also cultured including A. lwolffii, A. calcoaceticus, and A. haemolyticus. ^{Reference Faulde and Spiesberger2,Reference Marques, Rodrigues and Rebelo13,Reference Le Guyader, Rivault and Chaperon41,Reference Lise, Garcia and Lutinski45,Reference Daniel, Srámová and Absolonová76,Reference Felizatti, Ceccato, Cintra-Socolowski and Zeni Beretta77} Each of these bacteria have been implicated in HAIs and can exhibit MDR. ^{Reference Diwan, Khatal, Kelkar, Singh and Deoker79–Reference Ehelepola and Wijesinghe82} Other Gram positives isolated from ants, moth flies, and cockroaches included Clostridium spp. (C. perfringens isolated from ants), ^{Reference Beatson Susan39} Corynebacterium spp., ^{Reference Faulde and Spiesberger2,Reference Pai, Chen and Peng35,Reference Lise, Garcia and Lutinski45,Reference Do Nascimento, Amaral and Ferreira46,Reference Daniel, Srámová and Absolonová76,Reference Felizatti, Ceccato, Cintra-Socolowski and Zeni Beretta77} Nocardia spp., ^{Reference Faulde, Sobe, Burghardt and Wermter43,Reference Lise, Garcia and Lutinski45,Reference Oliveira, Souza and Campos52} Wolbachia spp., ^{Reference Watanabe, Takaku, Yokota, Hayashi, Tamaki and Kokeguchi33} Listeria monocytogenes, ^{Reference Do Nascimento, Amaral and Ferreira46} and numerous other non-clinically relevant Gram positives. ^{Reference Lise, Garcia and Lutinski45}

GNs were isolated from almost every study in this review and all studied arthropods except mosquitoes. Pseudomonas spp., including MDR strains, was isolated from cockroaches, ants, houseflies, moth flies, and cluster flies in thirty-six studies. Other GNs included Enterobacter spp. (E. cloacae and E. aerogenes), Citrobacter spp. (C. freundii) Klebsiella spp. (K. pneumoniae, K. oxytoca), Serratia spp. (S. marcescens, S. liquefaciens), E. coli Salmonella spp. and Shigella spp. ^{Reference Tilahun, Worku, Tachbele, Terefe, Kloos and Legesse22,Reference Beatson Susan39,Reference Oliva, Díaz and Fuentes González54,Reference Chiuya, Masiga, Falzon, Bastos, Fèvre and Villinger58,Reference Fotedar, Banerjee, Singh and Verma63,Reference Jalil, Abdulhadi and Al-Ameer73} Two other important environmental pathogens isolated included Stenotrophomonas maltophilia and Burkholderia cepacia. ^{Reference Faulde and Spiesberger2,Reference Marques, Rodrigues and Rebelo13,Reference Mehainaoui, Menasria, Benouagueni, Benhadj, Lalaoui and Gacemi-Kirane20,Reference Rupprecht, Moter and Wiessener42,Reference Faulde, Sobe, Burghardt and Wermter43,Reference Rahuma, Ghenghesh, Ben Aissa and Elamaari57} Flies, including cluster and moth flies, were the most common source of Stenotrophomonas, followed by ant and cockroach, while Burkholderia was isolated from ants. ^{Reference Marques, Rodrigues and Rebelo13} Many other GNs isolated from arthropods are listed in Supplemental Table 2.

Gram-negative cocci were not identified as abundantly as Gram-negative rods. However, Neisseria spp., including N. meningitidis, were found among ants and moth flies in this review, ^{Reference Faulde and Spiesberger2,Reference Marques, Rodrigues and Rebelo13,Reference Beatson Susan39,Reference Do Nascimento, Amaral and Ferreira46} as were Haemophilus spp. ^{Reference Fotedar, Banerjee, Singh and Verma63,Reference Pai, Chen and Peng71} There were additional Gram-variable bacteria isolated and many environmental bacteria less well-known in the healthcare environment.

Five studies isolated and identified viruses from mosquitoes in the HE and surrounding areas. Ehelepola et al. and Almeida-Nunes et al. captured mosquitoes positive for dengue viruses 1–4 in healthcare facilities in Sri Lanka and Brazil, respectively. ^{Reference Almeida-Nunes, Marcilio and Oliveira12,Reference Lemos, Lemos and Prado83} Chiuya et al. identified several viruses from mosquitoes captured in their community controls in Kenya but only the Sindbis virus was identified when captured in the HE. ^{Reference Adegoke, Okon and Monday59} Finally, Krokovsky et al found Zika and Chikungunya viruses among mosquitoes captured in HEs of Brazil. ^{Reference Kappel, Oliveira, Silva and Pelli51}

Lastly, 19 studies isolated and identified fungi and yeast from arthropods captured in their HEs (Supplemental Table 3). Aspergillus spp. (A. flavus, A. fumigatus, A. niger) were highly prevalent among cockroaches, flies, and ants. ^{Reference Merad, Belkacemi and Merad21,Reference Madani, Safari, Khodabandeh, Izadi and Zahraei-Ramazani26,Reference Kassiri, Kassiri and Kazemi28,Reference Khodabandeh, Shirani-Bidabadi, Madani and Zahraei-Ramazani30,Reference Pai, Chen and Peng35,Reference Lise, Garcia and Lutinski45,Reference Prado, Gir, Pereira, Reis and Pimenta49,Reference Krokovsky, Guedes and Santos50,Reference Chadee and Le Maitre55,Reference Fotedar, Banerjee, Samantray, Nayar and Verma64,Reference Davari, Khodavaisy and Ala66,Reference Nazari, Mehrabi, Hosseini and Alikhani67,Reference Pai, Chen and Peng71,Reference Daniel, Srámová and Absolonová76,Reference Fotedar and Banerjee84,Reference Alhmidi, John and Mana85} Candida spp., including C. albicans, C. glabrata, C. parapsilosis, and others were also identified in many of the same studies.