Introduction

Acne vulgaris is a chronic inflammatory skin disease manifested by open or closed comedones and inflammatory elements (papules, pustules and nodes) as well as post-inflammatory changes of varying severity (hyperpigmentation, scars) [1, 2]. According to approximate data, about 9.4% of the world population suffer from this disease [3]. Patients with acne showed a significant decrease in the life quality and social activity, as well as frequent comorbidity with psychological disorders [4]. Thus, in a recent meta-analysis summarizing the results of 42 studies, D.V. Samuels et al. [5] showed that acne was significantly associated with depression (r= 0.22, 95% CI 0.17—0.26; p<0.00001) and anxiety (r= 0.25, 95% CI 0.19—0.3; p<0.00001).

According to modern concepts, acne is considered as a multifactorial dermatosis where significant pathogenesis role is played by an increase in sebum production, follicular hyperkeratosis, colonization of Propionibacterium acnes (P. acnes) and inflammatory mediators released in the skin [1, 2]. Genome-wide associative studies have revealed more than 20 genetic specifics that determine an increased risk of acne, especially its complicated forms [6—8]. At the same time, other factors predisposing the disease are to be actively studied.

Helicobacter pylori (H. pylori) — microaerophilic spiral-shaped gram-negative bacteria colonizes the mucous membrane of the human stomach; it’s leading factor in gastroduodenal diseases development (chronic gastritis, gastric ulcer and duodenal ulcer, MALT lymphoma of the stomach and adenocarcinoma) [9, 10]. Recent reviews and meta-analyzes show that more than 40% of population in developing countries is infected with this microorganism [11, 12]. In general, this epidemiological trend is also typical for Russia with a tendency towards a gradual decrease in prevalence [13]. So, according to D. Bordin et al. [14] obtained in all Russian federal regions by using the ¹³C—urease breath test the prevalence of H. pylori infection was 42.5% in 2017 and 35.3% in 2019.

Currently, a number of meta-analysis have established the associations of H. pylori with some skin diseases including rosacea (OR 3.12, 95% CI 1.92—5.07; p<0.0001), chronic urticaria (OR 1, 66, 95% CI 1.12—2.45; p= 0.01) and psoriasis (OR 1.19, 95% CI 1.15—2.52; p= 0.008) [15—17]. The clinical significance and mechanisms of such associations remain controversial [18]. However, the association of H. pylori and acne has not been investigated in meta-analytical studies. In the literature, there is evidence that patients with severe acne have both a higher frequency of H. pylori and significantly increased titer of antibodies to this microorganism compared with control individuals and patients with mild and moderate disease [19—21].

Objective. To systematize data on association between H. pylori and acne.

Material and methods

Data Search

Data search were carried out in the electronic databases MEDLINE/PubMed, EMBASE, Cochrane, Google Scholar, the Russian Science Citation Index (RSCI) from January 1, 1990 to November 2020. In these databases it was analyzed the titles and abstracts. To search the MEDLINE/PubMed database the following keyword combinations were used: acne (Title/Abstract) helicobacter (Title/Abstract), or acne (Title/Abstract) pylori (Title/Abstract), or acne vulgaris (Title/Abstract) helicobacter (Title/Abstract), as well as their analogs in Russian (for searching in the RSCI database).

Study selection criteria

Criteria for inclusion in meta-analysis: relevant publications in peer-reviewed periodicals in English or Russian as well as publications with detailed descriptive statistics allowing the data to be included in meta-analysis. The object of the study was acne patients; the frequency of H. pylori detection was analyzed. Studies devoted to patients with rosacea were excluded from the analysis. If duplicate results were found in two publications (from different or one electronic database) only one of them was selected for the final analysis.

Data Extraction

Two researchers (D.N.A. and R.M.U.) independently performed data extraction using standardized forms. The year of publication, country, methodology for H. pylori diagnosing, the total sample size of acne patients, the number of H. pylori-infected patients, the number of controls, and the number of H. pylori-infected among controls were analyzed. Any disagreement was resolved by negotiations.

Statistical analysis

Statistical data processing was carried out using special software MedCalc 19.5.3 (Belgium) in Microsoft Windows 10 (USA). Results are presented as pooled rates of H. pylori in acne patients (%) and 95% confidence intervals (95% CI). Heterogeneity between different studies was assessed using Cochrane's Q and I² criteria. At p<0.05 and I²>50 significant heterogeneities were observed. The probability of publication error was assessed by constructing a funnel—shaped scatter plot as well as calculating the Begg-Mazumdar correlation test and Egger's regression test.

Results

Data Search

A search in electronic databases identified 32 studies for further analysis. Out of these 32 studies, 18 papers were excluded since they were not original ones (15 — reviews, 1 — experimental study, 2 — other irrelevant works). The selected 14 studies were analyzed in details for compliance with inclusion criteria, and 9 studies were excluded afterwards (Fig. 1). As a result, the remaining 5 original studies met the criteria and were included in this meta-analysis (see Table) [19, 20, 22—24].

Fig. 1. CONSORT diagram detailing the study selection strategy.

Characteristics of the selected studies

Characteristics of included studies

The final analysis included 5 studies with 629 acne patients; the studies performed in Russia (n=2) [22, 24], Ukraine (n=1) [23], Iran (n=1) [19] and Egypt (n=1) [20]. In 2 studies the data were compared with control groups [19, 20]. All studies used heterogeneous methods for diagnosing H. pylori including blood ELISA for IgG [23, 24], breath test with kinetic assessment of exhaled ammonia [22], ¹³C-urease breath test [19] as well as analysis of microorganism antigen in feces [20].

Prevalence of H. pylori in acne patients

The generalized H. pylori incidence in acne patients was 77.844% (95% CI 54.769–94.232) (Fig. 2). In the analysis a random effect model was used since there was significant heterogeneity between the results (p<0.0001, I²=97.34%). In the control group the generalized H. pylori incidence was 48.821% (95% CI 40.189—57.489). There was a statistically significant association between H. pylori infection and acne (OR 1.744, 95% CI 1.077—2.822; p=0.024).

Fig. 2. Forest diagram showing the generalized incidence of H. pylori in patients with acne.

The probability of publication bias was estimated by constructing a funnel-shaped scatter plot as well as calculating the Begg-Mazumdar test and Egger's regression test. Visual analysis of the funnel-shaped scattering diagram (Fig. 3) did not reveal significant asymmetry. In addition, the presence of statistically significant publication bias was excluded according to the results of the Begg-Mazumdar test (p=0.4386) and Egger's regression test (p=0.5638).

Fig. 3. Funnel scatter plot to assess the likelihood of publication bias.

Discussion

Acne is one of the most common skin diseases worldwide that occurs in many age groups, it significantly affects the patients’ life quality and often associated with anxiety and depression [4, 5]. The factors that contribute to the disease predisposition or modification of its course (more severe one) are not well understood. At the epidemiological level it has been shown that H. pylori is often associated with skin diseases such as rosacea, chronic urticaria, and psoriasis, although the clinical significance of these associations remains uncertain [15—17].

In the studies conducted up to the moment the detection rate of H. pylori in acne patients varies over a wide range from 47.9% to 100% [19, 20, 22—24]. The presented meta-analysis of 5 studies showed that the generalized H. pylori infection in acne patients was 77.844% (95 CI 54.769—94.232) while in control subjects it was 48.821% (95% CI 40.189—57.489). At the same time, a statistically significant association was found between H. pylori and acne (OR 1.744, 95% CI 1.077—2.822; p=0.024). In general, the data obtained may indicate a potential role of H. pylori in this skin disease predisposition or modification of its clinical course. A number of studies have noted that in patients with severe acne compared with control persons and patients with mild and moderate acne not only a higher incidence of H. pylori is detected but also a significantly increased titer of fecal antigen or antibodies to this microorganism were identified [19— 21]. Thus, in a recent study by R. Saleh et al. [20] with gradation of acne patients in accordance with the Global Acne Grading Scale it was shown that in the severe acne there were significantly higher levels of fecal antigen and titer of antibodies to H. pylori compared with patients suffering from mild and moderate forms, as well as control groups (p<0.001). Similar data were obtained in the work of S.A. Khashaba et al. [21] using ELISA.

The exact mechanisms of associations between H. pylori and cutaneous manifestations remain unknown [18]. It is assumed that chronic gastritis caused by H. pylori leads to increase in permeability of the stomach epithelial lining as well as the vessels walls of the mucous membrane, and a higher exposure of bacterial and alimentary antigens in the systemic circulation that can induce both the local release of inflammatory mediators in tissues and systemic immunological reactions (autoimmune and inflammatory processes; the formation of immune complexes induced by molecular mimicry and cross-reactive antibodies) [25–27]. It has been shown that cytotoxin CagA and H. pylori urease are capable to initiate the disorganization of cells’ tight junctions by increasing phosphorylation of myosin light chain as well as inactivation of the PAR1 family protein kinase that regulates the function of cell cytoskeleton [28, 29].

It should be noted, there are several leakages in our work. First, the studies included in the meta-analysis are represented by a small pool of investigations carried out in a limited number of countries (Russia, Ukraine, Iran, Egypt). Secondly, the analyzed studies are characterized by a significant heterogeneity of the methods used for H. pylori diagnosis. However, this is the first meta-analysis summarizing the results of various studies with H. pylori detection rate in acne patients.

Conclusion

A meta-analysis has shown that H. pylori is quite common in acne patients and may represents a risk factor for this skin disease. However, additional large homogeneous studies are needed to verify the results obtained and to investigate the potential causal association between H. pylori and acne.

Authors’ contributions:

The concept and design of the study: Andreev D.N., Maev I.V.

Collecting and interpreting the data: Andreev D.N., Umyarova R.M.

Statistical analysis: Andreev D.N.

Drafting the manuscript: Andreev D.N., Umyarova R.M.

Revising the manuscript: Maev I.V.

The authors declare no conflict of interest.