Infestation with Demodex mites has been associated with acne vulgaris. The aim of this study was to explore the association between Demodex infestation and severe acne vulgaris in outpatients seen at Hospital Regional Lambayeque in Chiclayo, Peru.
Material and methodsWe conducted a cross-sectional study of 46 patients with severe acne and 92 patients with nonsevere acne. Severe acne vulgaris was diagnosed if the score was 3 or more on the Spanish Acne Severity Scale (EGAE, in its Spanish acronym). Demodex infestation was diagnosed when a skin surface biopsy showed more than 5mites/cm2.
ResultsThe patients had a median age of 18 years (interquartile range, 15–20 years), 60.9% were male, 81.9% lived in an urban area, and 29.7% were infested with Demodex mites. In the bivariate analysis, severe acne vulgaris was significantly associated with Demodex infestation (P=.001), sex (P=.003), residence (P=.015), a paternal history of acne (P=.045), a maternal history of acne (P=.045), and type of skin (P<.001). In the multivariate analysis, after adjustment for male sex, urban residence, previous treatment, maternal and paternal history of acne vulgaris, and an oily skin type, patients with Demodex infestation were 4.2 times more likely to have severe acne vulgaris (95% CI: 1.6–10.9, P=.003).
ConclusionDemodex infestation was associated with severe acne vulgaris in outpatients at our hospital.
Existen antecedentes de asociación de la infestación por Demodex sp. y el acné vulgar. El objetivo fue evaluar si la infestación por Demodex sp. se asocia a acné vulgar grave en consultas externas del Hospital Regional Lambayeque.
Material y métodosEstudio transversal en 46 pacientes con acné grave y 92 pacientes con acné no grave. Se definió como acné vulgar grave al de los participantes que tuvieron un grado ≥ 3 con la escala de gravedad del acné en español (EGAE). La infestación por Demodex sp. se determinó mediante biopsia cutánea superficial, considerándose infestación si existieron más de cinco ácaros por cm2.
ResultadosEl 60,9% de los participantes fueron del sexo masculino, con una mediana de edad de 18 años y un rango intercuartílico de 15 a 20 años, provenían del sector urbano (81,9%) y se encontró infestación por Demodex sp. en el 29,7%. En el análisis bivariado, se encontró asociación estadísticamente significativa entre acné vulgar grave e infestación por Demodex sp. (p = 0,001), sexo (p = 0,003), procedencia (p = 0,015), antecedente paterno de acné (p = 0,045), antecedente materno de acné (p = 0,045) y tipo de piel (p < 0,001). En el análisis multivariado, la infestación por Demodex sp. estuvo 4,2 veces más asociada a acné vulgar grave (IC 95%: 1,6-10,9; p = 0,003) ajustado por sexo, procedencia urbana, tratamiento previo, antecedentes paterno y materno de acné vulgar y presencia de piel grasa.
ConclusionesLa infestación por Demodex sp. se asocia al acné vulgar grave.
Acne vulgaris is one of the most common skin diseases and it has a major impact on quality of life. According to one study conducted in several European countries, it affects 58.7% of people aged 15 to 24 years.1 Acne is also the second greatest contributor to the global burden of disease measured in disability-adjusted life years.2 Other studies have linked it to anxiety, depression, and even suicidal ideation.3 Acne vulgaris is thus a significant skin disease in terms of both prevalence and impact on quality of life.
Acne vulgaris is a multifactorial disease that has been associated with a paternal and maternal history of acne, consumption of chocolate and dairy products, and a high glycemic index diet.1,4,5Cutibacterium acnes also has a central role in the pathophysiology of acne vulgaris through activation of the innate immune system.6 Acne is thus caused by genetic and lifestyle factors.
Demodex mites infest the sebaceous glands on the face. There are 2 species: Demodex folliculorum and Demodex brevis. Both can be found on the face, although D. folliculorum is the predominant species. Demodex mites are transmitted through close contact. Several publications have shown an association between Demodex infestation and a number of skin diseases that affect the face, including rosacea, blepharitis, and acne vulgaris.7–15Demodex mites have several mechanisms of action in the hair follicle, including obstruction, perforation leading to a foreign body–type inflammatory reaction, the release of cytokines and free radicals that induce inflammation, and alteration of the microbiome.16–21 The same mechanisms are involved in the development of acne vulgaris and severe acne vulgaris.
Studies linking acne vulgaris to Demodex mites have been conducted in Turkey and China.11,17,22 In our review of the literature, we found no such studies in Latin America.23 The aim of this study was to evaluate the association between Demodex mites and severe acne vulgaris in patients seen at the outpatient clinic at Lambayeque Regional Hospital in northern Peru.
Material and MethodsWe designed a prospective, cross-sectional, observational study of patients aged 12 to 30 years who received medical care at the Dermatology outpatient department of Hospital Regional Lambayeque, a high-complexity hospital in northern Peru, between May 2019 and April 2020.
Due to the little information available on the association between Demodex infestation and severe acne vulgaris, we conducted a pilot study with 30 patients to calculate the minimum sample size required; 60% of the patients with severe acne vulgaris and 33.3% of those with nonsevere acne vulgaris had a Demodex infestation. Using the Epidat 3.1 software program, and applying a confidence level of 95%, a power of 80%, and a ratio of 1:2 for people with and without the event, we calculated a minimum sample size of 138 patients: 46 with the event and 92 without.
Participants were recruited in the outpatient clinic at the end of routine practice. We included patients with severe and nonsevere acne vulgaris graded using the Spanish Acne Severity Scale (EGAE, in its Spanish acronym). We excluded patients who had been receiving acne treatment for more than 1 month, patients with a history of polycystic ovary syndrome, and patients under treatment with systemic glucocorticoids, lithium, phenytoin, isoniazid, complex B, or halogenated compounds.
The details of the study were explained to all potential participants, who were assured that participation was voluntary and would not interfere with their care before, during, or after the study. Informed signed consent was obtained from all participants older than 18 years and from the parents or guardians of those aged 12 to 17 years. Each patient was then given a form containing 3 separate sections. The first contained a set of multiple-choice questions to be completed by the patient. The second was completed by a dermatologist at the outpatient clinic, while the third was completed by a microbiologist from the hospital laboratory blinded to the other information.
The first section was designed to collect information on sex, age (current age according to national identification card), highest level of education, place of residence (urban, rural, suburban), family history of acne vulgaris (father, mother, siblings), previous treatment of acne vulgaris (yes, no, type), personal history of rosacea, use of facial cleansers (yes, no), and number of times they washed their face a day (0, 1, 2, 3, >3).
In the second section, the dermatologist made a note of acne severity and skin type. Severity was assessed using the EGAE, which has an interobserver reliability of 0.773 (Kendall concordance coefficient) and a concurrent validity of 0.889 (Spearman correlation coefficient); it is also sensitive to change (P<.001). As the EGAE is a visual scale, acne lesions located in 3 areas of the body (face, chest, and back) were compared with the EGAE photographs ordered by degree of severity.24 In this scale, severity is graded from 1 to 4 for lesions on the face and 1 to 3 for those on the chest and back. Severe acne vulgaris thus was diagnosed for EGAE scores of 3 or 4 accordingly. Skin type was classified as dry, mixed, or oily.
The third section of the form was used to record data on Demodex infestation. The tests were carried out independently and in a blinded fashion by a microbiologist at the hospital's parasitology laboratory. They therefore did not involve any risk for the patient. A standard superficial skin biopsy was performed25 to collect a sample from the superficial stratum corneum and hair follicle. Cyanoacrylate adhesive placed on a microscope slide was applied to 1-cm2 areas on the cheek, forehead, chin, and nose. Immersion oil was then placed on the slide and the Demodex mite count was calculated using an Olympus CX41RF binocular microscope. Infestation was considered to exist if at least 5mites were observed in a low-magnification field (10×) or more than 5 mites were observed in a 1-cm2 area.26
The data were entered into an Excel database and analyzed in Stata version 14. Normality of distribution for quantitative variables was checked using the Shapiro–Wilk test. Quantitative data are presented as median and interquartile range and qualitative variables as absolute and relative frequencies.
The χ2 and Fisher tests were used in the bivariate analysis and logistic regression in the multivariate analysis. In both cases, significance was set at 5%.
The project was approved by the Research Ethics Committee at Hospital Regional de Lambayeque (code 0321-035-19 CIEI) and authorized for execution at said hospital. The authors supplied the study materials, while the hospital provided access to laboratory facilities.
ResultsOverall, 138 patients who met the selection criteria were included; they were predominantly male (84, 60.9%), had a median age of 18 years (interquartile range, 15–20 years), largely lived an urban area (113, 81.9%), and were more likely to have a secondary level of education (74, 53.6%). Forty-one patients (29.7%) had a Demodex infestation, in all cases involving D. folliculorum (Table 1).
Participant Characteristics According to Acne Severity.
Variable | No. | % | Not severe | Severe | P |
---|---|---|---|---|---|
Demodex infestation | |||||
No | 97 | 70.3 | 73 (79.4) | 24 (52.2) | .001a |
Yes | 41 | 29.7 | 19 (20.7) | 22 (47.8) | |
Age (median). y | 18 (15–20) | 17 (15–20) | 18 (16–21) | .265 | |
Sex | |||||
Female | 54 | 39.1 | 44 (47.8) | 10 (21.7) | .003a |
Male | 84 | 60.9 | 48 (52.2) | 36 (78.3) | |
Level of education | |||||
No schooling or primary only | 2 | 1.5 | 1 (1.1) | 1 (2.2) | .435 |
Secondary | 74 | 53.6 | 47 (51.1) | 27 (58.7) | |
Higher | 62 | 44.9 | 44 (47.8) | 18 (39.1) | |
Place of residence | |||||
Rural | 10 | 7.3 | 3 (3.3) | 7 (15.2) | .015b |
Suburban | 15 | 10.9 | 8 (8.7) | 7 (15.2) | |
Urban | 113 | 81.9 | 81 (88.0) | 32 (69.6) | |
Previous treatment | |||||
No | 73 | 52.9 | 54 (58.7) | 19 (41.3) | .054 |
Yes | 65 | 47.1 | 38 (41.3) | 27 (58.7) | |
Type of treatment | |||||
Topical | 13 | 20.0 | 9 (23.7) | 4 (14.8) | .768 |
Doxycycline | 32 | 49.2 | 18 (47.4) | 14 (51.9) | |
Minocycline | 6 | 9.2 | 4 (10.5) | 2 (7.4) | |
Isotretinoin | 14 | 21.5 | 7 (18.4) | 7 (25.9) | |
Treatment failure | |||||
No | 20 | 30.8 | 15 (39.5) | 5 (18.5) | .102 |
Yes | 45 | 69.2 | 23 (60.5) | 22 (81.5) | |
Rosacea | |||||
No | 135 | 97.8 | 91 (98.9) | 44 (95.7) | .258 |
Yes | 3 | 2.2 | 1 (1.1) | 2 (4.4) | |
Paternal history of acne vulgaris | |||||
No | 88 | 63.8 | 64 (69.6) | 24 (52.2) | .045a |
Yes | 50 | 36.2 | 28 (30.4) | 22 (47.8) | |
Maternal history of acne vulgaris | |||||
No | 112 | 81.2 | 79 (85.9) | 33 (71.7) | .045a |
Yes | 26 | 18.8 | 13 (14.1) | 13 (28.3) | |
History of acne vulgaris in siblings | |||||
No | 89 | 64.5 | 56 (60.9) | 33 (71.7) | .208 |
Yes | 49 | 35.5 | 36 (39.1) | 13 (28.3) | |
Facial cleansers | |||||
No | 76 | 55.1 | 54 (58.7) | 22 (47.8) | 0.226 |
Yes | 62 | 44.9 | 38 (41.3) | 24 (52.2) | |
Hand washing | |||||
Once a day | 25 | 18.1 | 17 (18.5) | 8 (17.4) | .987 |
2–3 times a day | 95 | 68.8 | 63 (68.5) | 32 (69.6) | |
≥3 times a day | 18 | 13.0 | 12 (13.0) | 6 (13.0) | |
Skin type | |||||
Mixed | 62 | 44.9 | 52 (56.5) | 10 (21.7) | <.001a |
Oily | 76 | 55.1 | 40 (43.5) | 36 (78.3) |
Sixty-five patients had received previous treatment for acne vulgaris (47.1%), the most common being doxycycline (32/65, 49.2%). Forty-five (69.2%) had not responded to treatment. Three patients (2.2%) reported a history of rosacea, but no clinical signs were observed at the time of this study (Table 1).
In the bivariate analysis, acne severity was significantly associated with Demodex infestation (P=.001), male sex (P=.003), urban residence (P=.015), a paternal or maternal history of acne vulgaris (P=.045) and an oily skin type (P<.001) (Table 2). In the multivariate analysis, patients with Demodex infestation were 4.2 times more likely to have severe acne vulgaris compared with those without an infestation (95% CI, 1.6–10.9; P=.003) following adjustment for age, sex, place of residence, previous treatment, a family history of acne (paternal or maternal), and skin type (Table 2).
Risk Factors for Severe Acne Vulgaris.
Variable | Crude | Adjusted | ||||
---|---|---|---|---|---|---|
OR | CI | P | OR | CI | P | |
Demodex sp. | ||||||
No | ||||||
Yes | 3.5 | 1.6–7.6 | 0.001 | 4.2 | 1.6–10.9 | .003 |
Age | 1.1 | 1–1.2 | 0.133 | 1.1 | 1.0–1.3 | .036 |
Sex | ||||||
Female | ||||||
Male | 3.3 | 1.5–7.4 | 0.004 | 5.1 | 1.8–14.7 | .003 |
Level of education | ||||||
No schooling or primary | ||||||
Secondary | 0.6 | 0.0–9.6 | 0.699 | |||
Higher | 0.4 | 0.0–6.9 | 0.535 | |||
Place of residence | ||||||
Rural | ||||||
Suburban | 0.4 | 0.1–2.0 | 0.256 | 0.2 | 0.0–1.9 | .170 |
Urban | 0.2 | 0.0–0.7 | 0.014 | 0.2 | 0.0–0.8 | .028 |
Previous treatment | ||||||
No | ||||||
Yes | 2.0 | 1.0–4.1 | 0.055 | 1.2 | 0.5–3.1 | .644 |
Type of treatment | ||||||
Topical | ||||||
Doxycycline | 1.8 | 0.4–6.9 | 0.423 | |||
Minocycline | 1.1 | 0.1–8.9 | 0.911 | |||
Isotretinoin | 2.3 | 0.5–10.9 | 0.313 | |||
Treatment failure | ||||||
No | ||||||
Yes | 2.9 | 0.9–9.2 | 0.077 | |||
Rosacea | ||||||
No | ||||||
Yes | 4.1 | 0.4–46.9 | 0.252 | |||
Paternal history of acne vulgaris | ||||||
No | ||||||
Yes | 2.1 | 1.0–4.4 | 0.047 | 2.2 | 0.9–5.6 | .088 |
Maternal history of acne vulgaris | ||||||
No | ||||||
Yes | 2.4 | 1.0–5.7 | 0.049 | 2.8 | 0.9–8.4 | .073 |
History of acne vulgaris in siblings | ||||||
No | ||||||
Yes | 0.6 | 0.3–1.3 | 0.210 | |||
Facial cleansers | ||||||
No | ||||||
Yes | 1.6 | 0.8–3.2 | 0.227 | |||
Hand washing | ||||||
Once a day | ||||||
2–3 times a day | 1.1 | 0.4–2.8 | 0.874 | |||
≥3 times a day | 1.1 | 0.3–3.9 | 0.927 | |||
Skin type | ||||||
Mixed | ||||||
Oily | 4.7 | 2.1–10.6 | P<.0001 | 3.3 | 1.3–8.6 | .013 |
We detected a significant association between Demodex infestation and severe acne vulgaris in our study population. Significant differences between patients with and without severe acne vulgaris were observed for Demodex infestation, male sex, urban residence, a maternal and paternal history of acne vulgaris, and an oily skin type. Our findings are consistent with those of several previous studies.11,12,16,25
The participants largely lived in an urban area, probably because the study was performed in a tertiary care hospital, to which urban residents are more likely to be referred. Three participants reported a personal history of rosacea, but there were no clinical signs of the disease at the time of the study.
Our observation of an association between Demodex mites and oily skin supports findings by Porta25 and Lacey et al.,18 who found that oily skin favors the proliferation of these mites, greater inflammation, and, as a result, more severe acne vulgaris.
No significant associations were observed between acne severity and use of facial cleansers or frequency of face washing. Yuan et al.,16 by contrast, found that facial cleansers used over a 7-day period reduced Demodex infestation in patients with mild or moderate acne. The differences between their findings and ours could be due to the fact that our patients had severe acne and that we did not analyze duration of facial cleanser use.
The association observed between Demodex mites and severe acne vulgaris is partly consistent with findings by Akçınar et al.11 in Turkey. Using the Global Acne Grading System (GAGS), they found that patients with severe acne vulgaris had a higher density of Demodex mites. The association, however, was not significant, possibly because they used a different scale to assess acne severity and a different cutoff for Demodex infestation.
Several studies have suggested that Demodex mites could favor higher rates of Cutibacterium acnes and Staphylococcusaureus17,19,20,27 by acting as a vector for these bacteria.26Demodex mites can also induce the expression of toll-like 2 receptors and inflammatory infiltrates (macrophages, mast cells, T helper cells types 1 and 17, and eosinophils) and the release of inflammatory cytokines, such as tumor necrosis factor α and interleukin 1 and 6.17,19,27
Confirmation that these likely mechanisms are involved in acne vulgaris as they are in rosacea would mean that Cutibacterium acnes is not the only agent involved in the onset of acne, and would also explain why patients do not respond adequately to treatment.
One of the strengths of this study is that the EGAE scale has high validity and interobserver reliability. A range of methods exist for evaluating acne severity, but they have not been validated or undergone assessment of intraobserver or interobserver agreement.28
Our study also has some limitations. First, we excluded patients with a history of polycystic ovarian syndrome, which, in patients with a certain history, has been linked to Demodex infestation. We were therefore unable to evaluate the possible effect of polycystic ovarian syndrome on the association between Demodex infestation and severe acne vulgaris in our population. Second, our findings for a family history of acne and response to previous acne treatments may be influenced by recall bias or patient expectations, or in the case of treatments, subjective evaluations. Third, because of our study design, we were unable to determine whether Demodex infestation occurred before or after the onset of severe acne vulgaris. This is important and should be taken into account when interpreting results, especially in a disease such as acne, which has varying levels of severity.
Despite the above limitations, our study represents a preliminary approach to the evaluation of the association between Demodex mites and acne vulgaris in Latin America, where studies are lacking. The only study we identified in this regard, performed at our hospital, examined facial skin diseases.
Cohort studies are needed to evaluate possible correlations between acne severity (measured using properly validated quantitative scales) and both Demodex infestation and other potentially relevant variables. Clinical trials could also be designed to investigate the effects of Demodex treatment on acne severity
In conclusion, Demodex infestation is associated with severe acne vulgaris, a paternal and maternal history of acne vulgaris, age, sex, level of education, and skin type.
FundingHospital Regional Lambayeque provided access to the laboratory facilities, and the authors supplied the materials.
Conflicts of InterestNone.