Demodex Folliculorum and Bacterial Co-Infection in Patients With Blepharitis

Background Demodex mites are obligate ectoparasites found in the pilosebaceous follicle and sebaceous gland of many mammals. The aim of the study was to examine prevalence of bacterial infection in Demodex folliculorum infected patients with blepharitis. Methods The study included patients with a diagnosis of blepharitis with D. folliculorum (n=128) and healthy volunteers (n=103). Patients were questioned about their age, sex, social habits, and literacy status and per capita income. The examination consisted of examining the vision of the patient with and without ocular correction, tonus in both eyes and a careful examination of the anterior segment of both eyes. The bacterial culture of the conjunctival sac was obtained by inoculating the samples on Columbia agar with 5% sheep blood and identied based on morphological characteristics and stained by the Gram method. Species identication was performed by coagulase test or VITEK Compact.


Background
Demodex mites are obligate ectoparasites found in the pilosebaceous follicle and sebaceous gland of many mammals. Two species of Demodex, D. folliculorum (Simon, 1842) and D. brevis (Akbulatova, 1963) are found in human body. Demodex folliculorum occurs in the hair follicle and Zeiss glands, usually forming clusters of several individuals. Demodex brevis is most often isolated as separate specimens in the depths of sebaceous glands in the skin of the face, in the Meibomian glands, and in the eyelids [1]. Both Demodex species are present in the face skin, mainly around the nose, around the eyes, on the forehead and chin. In addition to facial skin, these mites can be found in other parts of the body, including the hairy part of the head, auditory canals, skin in the neckline area, genitals, hands and feet, and nipples [2][3][4]. It has been noted that D. folliculorum is more numerous, but D. brevis occupies larger areas of the skin.
The transmission routes of Demodex spp. have not yet been fully investigated. It is likely that infection with Demodex spp. occurs through direct contact, use of common toiletries or towels, or through dust [5]. Skin colonization occurs during childhood or adolescence; no mites are found in the skin of newborns [1,6]. The prevalence of mite infestation increases with age, and ∼58% of individuals with mites will have evidence of chronic blepharitis [7].
Demodecosis usually causes symptoms in the skin and eyes, but it can also be asymptomatic. The symptoms of ocular demodecosis are non-speci c. Patients infected with Demodex spp. have reported tearing, burning, foreign body sensations, eyelid margin hyperemia, eyelash loss, itching, eye redness, blurred vision, and conjunctivitis with excessive sensitivity to light [4,5]. Burning and itching of eyelid margins are more frequently reported on warmer days [8]. Therefore, D. folliculorum was supposed to cause chronic anterior blepharitis, while D. brevis has been considered to cause posterior blepharitis [9].
It was found that Demodex mites participate in the transmission of pathogens, which may play a key role in the pathogenesis of demodecosis [10]. Symptomatic demodecosis often occurs with simultaneous bacterial infection, which is con rmed by a decrease in the number of Demodex spp. after administration of tetracycline to people with acne [11]. Demodex folliculorum contributes to the development of rosacea by transmitting bacteria from insensitive areas to sensitive areas where in ammation may occur.
Together with their food, which is mainly epidermal cells and the secretions of sebaceous glands, Demodex mites can take up microorganisms from the surface of the skin. Then, through their digestive tract, microbes are transported to the hair follicles of the host [1,12]. In addition, the disintegration of Demodex spp. inside the hair follicle can lead to release of the transmitted bacteria and the development of local immune response [13].
The most common bacteria colonizing conjunctival sacs are Staphylococcus spp., Streptococcus spp., Micrococcus spp., and Corynebacterium spp., occurring in about 70%, 26%, 22%, and 7% of patients, respectively. Most common among obligate anaerobic bacteria are Propionibacterium acnes (~44%), Peptostreptococcus spp. (~6%), Lactobacilllus spp. (~2%), and Clostridium spp. (1%) [14]. The composition of the bacterial ora of the conjunctival sac depends on many factors, including patient age and the presence of chronic disease. For example, Propioniobacterium spp. is more common in adults, whereas Streptococcus spp. is more common in children [15]. Chronic diseases, including diabetes, may increase the number of coagulase-negative staphylococci compared to healthy patients. These bacteria produce substances inhibiting the development of pathogenic bacteria, stimulating local immunological processes and the exfoliation and regeneration of epithelial cells of the eye [16]. The aim was to examine the relationship between the incidence of bacteria infection in patients with and without D. folliculorum infection.

Materials And Methods
The study was carried out between October 2015 and May 2018, and was approved by the Bioethics Committee of the Pomeranian Medical University in Szczecin (KB-0012/82/15). It conformed to the principles outlined in The Declaration of Helsinki as revised in 2008.

Characteristics of groups
The study involved 231 patients from Poland. Patients classi ed in the studied group were selected from the Ophthalmology Department (n=5) and the Ophthalmology Department of the Regional Hospital (n=45) in Kołobrzeg, the Ophthalmology Department of the Independent Public Complex of Health Care Centres in Gry ce (n=49) and residents of the Social Welfare Home (SWH) in Jaromin (29 male aged 41 to 80; mean age 59.0). In total, 128 participants diagnosed with blepharitis and infected with D. folliculorum (23-85 years of age) were chosen. The patients with blepharitis included those with at least two symptoms of blepharitis, such as burning sensation in the eyes, tearing, eyelid hyperemia, foreign body sensation, and excessive loss of eyelashes. The healthy control group included 103 non-infected patients (24-81 years of age). Exclusion criteria were as follows: using topical ophthalmic medications (except arti cial tears) over the previous 3 months before the study started, a history of ocular or eyelid trauma and surgery in the last 6 months, previous diagnosis of chemical burns, Stevens-Johnson syndrome, ocular cicatricial pemphigoid, with eyelid malpositions such as entropion, ectropion, and distichiasis, signs of active ocular infection, or in ammation other than blepharitis, and general treatment with oral antibiotics.
Both groups were asked to ll the informed consent to participate in this protocol, which was followed by an interrogation carried out to capture information and slit-lamp evaluation with a magni cation of ×25.

Demodex spp. examination
A total of eight eyelashes were excised per patient, four eyelashes per eye. They were extracted with ne forceps and placed separately on each end of a slide. A coverslip was placed on the top of the eyelashes after coating them with Hoyer medium. The presence and counting of Demodex were performed in the samples by light microscopy with a magni cation of 4x, 10x and 40x. Infection was de ned as the presence of eggs, larvae, or mature forms of D. folliculorum on the eyelashes.

Clinical examination
Participants took part in an ophthalmic interview regarding eye problems and their personal and familial history of eye diseases. The residents from the Social Welfare Centre were not examined in detail due to their limited cooperation during the examination. The ophthalmological examination consisted of testing of uncorrected and best corrected distance visual acuity (VA) using Snellen charts. The examination was performed at a distance of 4 m in a room providing the same lighting for all examinations. The result of the best corrected visual acuity was recorded and conversed to the LogMAR scale (decimal logarithm from the minimum angle of resolution). Intraocular pressure (IOP) was measured with Mackay-Marg Tono-Pen AVIA applanation tonometer (Reichert, USA). The measurement was performed three times and the average used for the analysis. Anterior segment examination was performed using a Haag-Streit L0185 slit lamp (Nikon, Japan).

Microbiological examination
The samples for microbiological examination were obtained from the conjunctival sac using a sterile swab and AMIES transport medium. The identi cation methods used in this paper correspond to those used in routine bacteriological diagnostics. Samples were plated on basic microbiological media: Columbia agar with 5% sheep blood, Chapman, MacConkey, and Sabouraud and then incubated at 37 °C for 24-48 hours. Strains were identi ed based on morphological evaluation of the colonies on the media and preparations stained by the Gram method.
The identi cation of Staphylococcus spp. was performed by determination of hemolytic capacity of colonies on Columbia agar medium with 5% sheep's blood and by evaluation of growth on Chapman medium, allowing for differentiation of staphylococci into mannitol-positive and mannitol-negative strains. All strains showing the ability to ferment mannitol were analyzed for the presence of clumping factor A, protein A, and coagulase. The presence of all three factors indicated Staphylococcus aureus.
MacConkey medium was used to isolate and identify strains of Gram-negative bacteria. Due to the lack of pathogenicity of this group of microorganisms in conjunctivitis, only growth morphology on the medium was evaluated, dividing bacteria into lactose-positive and lactose-negative strains. Species identi cation was performed by VITEK Compact (bioMerieux, Poland).
All the microorganisms showing growth characteristic of Corynebacteria on Columbia agar with 5% sheep's blood were analyzed by Gram staining. Gram-positive cells with a characteristic club-like shape were considered to be Corynebacterium spp.
Using the disk diffusion test, the drug susceptibility of isolated strains was determined. Antybiogram was perfomed for Staphylococcus aureus strains as the pathogen caused conjunctivitis. From single colonies grown after 18-24 h, a suspension of density 0.5 according to McFarland scale (1x108 cfu/ml) was prepared and inoculated into Mueller-Hinton agar medium (bioMerieux, Poland). Subsequently, antibiotic discs with erythromycin (15 µl), clindamycin (2 µl), gentamicin (10 µl), neomycin (10 µl), tetracycline (10 µl), and trimethoprim/sulfamethoxazole (1.25/23.75 µl) were placed onto the culture medium. Methicillinresistant Staphylococcus aureus (MRSA) was determined using cefoxitin 30-μg disks. Assessment of the growth inhibition zone around the discs and analysis of the results were performed according to the guidelines of the National Reference Centre for Microbial Susceptibility (www.antybiotyki.pl).

Statistical analysis
Statistical studies were performed using Stat Soft Statistica 10.0 PL. The nonparametric Mann-Whitney test was used to evaluate the differences between Demodex spp. infection and mean IOP and VA in the analyzed groups. In order to establish possible relationships between D. folliculorum infection and the occurrence of eye diseases and symptoms in patients from the two groups, the Chi 2 independence test was used. Differences were deemed statistically signi cant at p<0.05.

Results
In Table 1 was shows mean IOP and VA in uninfected and infected with D. folliculorum participants. In patients with D. folliculorum infection mean IOP and VA were signi cantly lower than in uninfected patients.
The ocular microbiota was found in 9 (8.7%) uninfected participants and in all patients infected with D. folliculorum. The aerobic and facultative anaerobic bacterial ora colonizing conjunctival sacs of the examined patients were found to be Bacillus subtilis, Corynebacterium spp., Haemophilus in uenzae, Micrococcus spp., Staphylococcus spp. and Streptococcus spp. Four (3.1%) patients infected with D. folliculorum had Corynebacteriaceae, three (2.3%) of which also had chalazia.
Staphylococcus aureus (Fig. 1C) was observed in an adult male patient with no chronic diseases. In microscopic examination, mature forms of D. folliculorum were observed on the eyelashes (Fig. 1A). Ophthalmic examination showed that VA of the right eye and of the left eye were 1.0 and IOP of the right eye was 11.7 mmHg and 10.7 mmHg, respectively. Anterior segment examination using a slit-lamp showed cylindrical dandruff on the upper eyelid and blockage of the Meibomian glands. Staphylococcus aureus was also found in adult female patient with arterial hypertension and diabetes mellitus. Microscopic examination of her eyelashes revealed numerous larval and mature forms of D. folliculorum. Ophthalmic examination showed that VA of the right eye was 0.9 and in the left eye was 1.0. Intraocular pressure was 12.0 mmHg, and 13.0 mmHg, respectively. The patient suffered from irritation of the eye and conjunctiva. Staphylococcus aureus sensitive to erythromycin, clindamycin, gentamicin, neomycin, tetracycline, and trimethoprim/sulfamethoxazole were observed in two patients with mature forms of D.
folliculorum. An adult female patient with thrombocytopenia without ophthalmic symptoms had hyperopia corrected by glasses. Ophthalmic examination showed that VA to be 0.2 in both eyes and IOP was 14.0 and 15.0 mmHg, respectively. Slit lamp examination showed slight follicular irritation of the conjunctiva. In an adult female patient with hyperopia corrected with glasses, VA was 0.6 in the right eye and 0.8 in the left eye. Intraocular pressure was 17.0 mmHg in both eyes. Slit lamp examination showed a cylindrical dandruff on the upper eyelid. Staphylococcus aureus was found in adult male patient from SWH with myopia corrected with eyeglasses. Microscopic examination of his eyelashes revealed mature forms of D. folliculorum. Due to poor cooperation, visual acuity of the eyes was not examined. Intraocular pressure was 14.0 in the right eye and 15.0 mmHg in the left eye. Examination with a slit lamp showed a single cylindrical dandruff on the upper eyelid. Staphylococcus aureus was found in patient from SWH. Microscopic examination revealed isolated mature forms of D. folliculorum. Similarly, visual acuity was not examined due to lack of cooperation, IOP was 15.0 mmHg in both eyes. Slit lamp examination showed irritation of the conjunctiva close to the upper and lower eyelids, and both the upper and lower eyelids were swollen. Additionally, S. aureus was found in an adult SWH resident. Around the patient's eyelashes isolated mature forms of D. folliculorum were found in microscopic examination. Visual acuity was 1.0 in both eyes, while IOP was 9.0 in the right eye and 11.0 mmHg in the left. Examination using a slit-lamp showed no speci c symptoms.
Methicillin-resistant S. aureus was found in an adult male patient with hyperopia corrected by glasses, hypertension, and atrial brillation. Numerous mature forms of D. folliculorum were observed in the patient. Ophthalmic examination showed that VA in both eyes was 1.0 and IOP to be 21.0 in the right eye and 18.0 mmHg in the left. On the upper eyelid, we observed cylindrical dandruff and blockage of the Meibomian glands; eyelashes were glued together. MRSA was also found in an adult HIV-infected patient from SWH. Microscopic examination showed isolated mature forms of D. folliculorum. Intraocular pressure was found to be 11.0 mmHg in both eyes. Slit lamp examination showed pale conjunctiva.
Acinetobacter baumannii was isolated from an adult patient with hypertension. The patient had numerous eggs (Fig. 1B), as well as larval and mature forms of D. folliculorum. Best corrected distance visual acuity was found to be 0.7 in the right eye and 0.6 in the left eye. Intraocular pressure was 20 mmHg in the right eye and 17 mmHg, respectively. The slit lamp test revealed cylindrical dandruff on the upper eyelid.
Streptococcus pneumoniae was found in an adult patient with mature forms of D. folliculorum. Due to poor cooperation, visual acuity of the eyes was not examined. Intraocular pressure was 12.0 in the right eye and 9.0 mmHg in the left eye. The slit lamp study showed irritation, conjunctival hyperemia, cylindrical dandruff, and Meibomian gland dysfunction.
Klebsiella oxytoca were found in an adult patient with hypertension. During microscopic examination of eyelashes, the patient was found to have numerous eggs along with larval and mature forms of D. folliculorum. Ophthalmological examination showed that VA was 0.3 in the right eye and 0.4 in the left eye. Intraocular pressure was 21.0 mmHg in the right eye and 18.0 mmHg in the left eye. In addition, the patient had hyperopia corrected with glasses In the ophthalmological examination, cylindrical dandruff was observed, the eyelashes were stuck together, and the Meibomian glands were found to be obstructed with oily secretion.
Bacillus spp. was found in adult man without chronic diseases. The patient presented with single mature forms of D. folliculorum during microscopic examination. Due to poor cooperation from the patient, vision was not examined, whereas IOP was 14.0 mmHg in the right eye and 21.0 mmHg in the left eye. Slit lamp examination did not show any changes.

Discussion
Demodicosis is a problem in both dermatology and ophthalmology due to the chronic nature of the infection. Most studies concern the Demodex infections as cause of skin diseases, including pityriasis folliculorum, perioral dermatitis, scabies-like eruptions, facial pigmentation, eruptions of the bald scalp, demodicosis gravis, and even basal cell carcinoma [17]. Some researchers indicate that these mites participate in the etiopathogenesis of eye diseases, while others disagree [18,19]. However, there are studies which offer evidence that D. folliculorum and D. brevis can cause anterior blepharitis associated with disorders of eyelashes, and posterior blepharitis with meibomian gland dysfunction and keratoconjunctivitis, respectively [20][21][22]. Severe cases of blepharitis can arise from co-infection of Demodex spp. and bacteria. In the study, we compared occurrence of bacteria in the healthy patients and blepharitis subjects with Demodex spp. infection.
Microbiome was found in conjunctival sac swab in all patients infected with Demodex spp., and in 9% uninfected D. folliculorum participants. This may indicate that Demodex spp. promotes colonization of the conjunctival sac with microbiota. Zhu et al. [23] found bacteria in 54 patients with blepharitis (45 of them were also infected with Demodex spp.) and 37 controls. The colony counts and the incidence of Propinibacterium acnes from Demodex folliculorum-infected patients were signi cantly higher than of non-infected patients.
Spickett [24] showed that D. folliculorum may be a vector for Mycobacterium leprae. Demodex mites may also transmit Staphylococcus spp. and Streptococcus spp. on its surface. In a study conducted on patients, staff, and visitors of the Optometry Clinic in Oklahoma, S. aureus and S. epidermidis were found in 16.8% and 75.8% of participants, respectively [25]. In the study, two or more mites (11.6% and 5.2%, respectively) were reported more frequently in patients infected with S. aureus than in uninfected patients. Staphylococcus aureus was found in 21.9% of patients aged 1-29 years, 13.1% aged 30-59, and 15.1% aged 60-89. In another study, Türk et al. [26] found S. aureus in two D. folliculorum infected patients with blepharitis. In our study, S. aureus was isolated from 7% of patients D. folliculorum infected, including 14.3% of residents of the Social Welfare House. One resident of the Social Welfare House in Jaromin had coinfection with Demodex spp. and methicillin-resistant Staphylococcus aureus. We did not nd S. aureus in the uninfected D. folliculorum participants.
Lee et al. [27] found no differences in the presence or distribution of bacteria on eyelashes between uninfected and infected with Demodex spp. patients. Coagulase negative Staphylococcus spp., Corynebacterium diphtheriae, and S. aureus were found in patients of both groups. There were no differences in the occurrence of MRSA on eyelids between those uninfected and infected with Demodex spp. patients. Zhu et al. [23] did not observe differences in colonies of S. aureus and S. epidermidis between Demodex spp. infected and uninfected patients. Bezza Benkaouha et al. [28] also did not nd a difference for bacterial ora and Demodex spp. infection; however authors conducted a study in a small number of subjects.
Acinetobacter baumannii is one of the most common etiological factors of hospital-acquired infections. It shows natural mechanisms of resistance to antibiotics and chemotherapy. In the present study, A. baumannii was isolated from the conjunctival sac of a patient infected with D. folliculorum.
Lacey et al. [29] isolated Bacillus oleronius from a D. folliculorum extracted from face of a patients with papulopustular rosacea, and stated that found two speci c antigens (62 and 83 kDa) protein produced this bacteria can stimulate and be responsible for in ammation of the hair follicle. Li et al. [30] on serum from 59 patients with diagnosed rosacea showed a statistically signi cant correlation between ocular Demodex infestation and serum immunoreactivity to 62 and 83 kDa B. oleronius proteins. Results of a study by O'Reilly et al. [31] showed that proteins derived from B. oleronius may be a neutrophil-activating factor. Such activation of neutrophils could take place if B. oleronius proteins released from mites entered the tissues surrounding the hair follicle. This, in turn, could result in the development of local in ammation in the perifollicular tissue. In our study, Bacillus spp. were isolated from the conjunctival sac of a patient with D. folliculorum infection. Szkaradkiewicz et al. [32] isolated 23 strains of B. oleronius eyelashes from 18 patients Demodex related chronic blepharitis. The authors observed more severe symptoms of blepharitis in patients with B. oleronius infection. However, B. oleronius was also found in ve uninfected participants, which may undermine its role in the development of blepharitis. The authors concluded that these bacteria, living inside the intestines of the Demodex mites as symbionts, can be excreted by these mites onto the surface of human skin. Due to the fact that B. oleronius plays a signi cant role in the process of digestion in termites, it seems that these bacteria may play a similar role in Demodex spp. [29,33].
Streptococcus pneumoniae can cause in ammation of the middle ear, paranasal sinuses, and conjunctiva and cornea of the eye, as well as pneumonia. Streptococcus pneumoniae infection can cause severe or chronic complications [34,35]. In the presented study S. pneumoniae was reported in a resident of a Social Welfare Home infected with D. folliculorum.
The present study was a preliminary study that demonstrated co-infection of bacteria and Demodex spp., but it has some limitations that should be addressed. We recruited only healthy patients (non-infected and without blepharitis) and patients with blepharitis who were also infected with Demodex spp. Blepharitis can be caused not only by Demodex spp. but also by various bacterial infection, and that's why in the future study patients with blepharitis without Demodex spp. infection should also be included.
In the study, we did not distinguish between mixed and single type (e.g. anterior or posterior) of blepharitis; Rynerson and Perry [36] observed disruption of the bio lm in the eyelids in different types of blepharitis. Moreover, in the future study, more patients should be included, which allows deep analyses on the subtype of bacteria and Demodex spp.

Conclusions
Demodex spp. can collect microorganisms found on the surface of the skin and transport them to the host's hair follicles. Transmission of bacteria from non-susceptible sites to sensitive areas can contribute to the development of in ammatory reactions. Therefore, patients infected with Demodex spp. should also undergo microbiological examination of conjunctival swabs. The treatment of each patient should be individualized, adapted to the clinical condition, and in cases of bacterial co-infection, an antibiotic and/or a topical steroid drug should be additionally prescribed. Furthermore, daily hygiene of the eyelid margins should be recommended.

Declarations
Ethics approval and consent to participate The study was approved by the Bioethics Committee of the Pomeranian Medical University in Szczecin (KB-0012/82/15). It conformed to the principles outlined in The Declaration of Helsinki as revised in 2008.

Consent for publication
Not applicable.

Data Availability
The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Supplementary Files
This is a list of supplementary les associated with this preprint. Click to download.