Introduction

Tick-borne diseases (TBDs) are a growing public health concern in Mongolia and a cause of significant disease burden in humans because ticks serve as vectors in the transmission of pathogens [1–3]. Tick-borne borreliosis (TBB; also known as Lyme disease) is caused by the spirochete Borrelia burgdorferi sensu lato, whereas tick-borne rickettsiosis (TBR) is caused by a gram-negative intracellular bacterium; TBB and TBR constitute TBDs worldwide, with clinical manifestations [1].Tick-borne borreliosis causes several neurological and arthritic symptoms, such as headache, paralysis, and erythema migrans [2]. Tick-borne rickettsiosis usually manifests as mild fever, muscle ache, rash, cough, and nausea [3].Several studies have evaluated the seroprevalence of Borrelia andRickettsia in herders [3,4], blood donors in Mongolia [5], forestry workers in various Europeancountries [6–8], farmers in Poland [6], and blood donors in California [9]; however, no large-scale study has comprehensively investigated this aspect in all Mongolian provinces, while differentiating between different occupations.

Seroprevalence surveys can provide valuable data regarding the epidemiology and diagnosis of TBDs. First, as antibodies persist in the blood for years after symptomatic and asymptomatic infection, seropositivity indicates the cumulative occurrence of infection in individuals. Second, knowledge of the baseline seroprevalence is crucial for appropriate interpretation of antibody test results in terms of their predictive diagnostic value [9].Serum antibody detection constitutes the standard method for diagnosing TBB and TBR [7,9,10]. Immunoglobulin M (IgM) istypically used as a marker for the serodiagnosis of current infection, while immunoglobulin G (IgG) is used to identify recent infection.However, diagnosis is complicated by several factors; in the early stages of infection (up to 8 weeks), antibodies may be absent, whereas antibodies may persist for years after recovery from disease [10].Therefore, to have therapeutic value, the clinical presentation and background seroprevalence should be considered in interpreting serological test results[11].

The seroprevalence of IgG antibodies against B. burgdorferi sensu lato and Rickettsia has been evaluated in different countries [3,4,6,7,9]. In blood donors in southern Norway, the seroprevalence of serum IgG antibodies against Borrelia and Rickettsia was 22% and 4.2%, respectively [10].The seroprevalence of Borrelia ranged from 20% to 62% in forestry workers in Europe [7], 18.2% to 50.7% in Polish farmers [6], 5% to 23% in Italian forestry workers, and 14% to 20% in French forestry workers [7].The global seroprevalence of B. burgdorferi is estimated at 14.5%, and the three regions with the highest seroprevalence are Central Europe (20.7%), East Asia (15.9%), and Western Europe (13.5%) [12,13].

In previous Mongolian studies that assessed the seroprevalence of IgG antibodies against B. burgdorferi sensu lato, the seroprevalence was 1.9% in the Selenge and Bulgan Provinces, 3% in the Tuv Province, 13.9% in the Dornogobi Province, and 25% in the Zavkhan Province [5,14]. The seroprevalence of IgG antibodies against Rickettsia spp. was 17.5% in herders, 19.5% in non-herders, and 20.4% in livestock [3,4].

In Mongolia, Borrelia and Rickettsiahave been detected in ticks [1,15],humans [3–5,14], and small mammals [16]. Determining the human seroprevalence of IgG antibodies against TBDs plays a vital role in clinical diagnosis and facilitates the enhancement of public health awareness [12]. Therefore, this study was conducted to estimate the seroprevalence of TBB and TBR to inform public health approaches and future research on Borrelia and Rickettsia.

Materials and methods

Results

Seroprevalence of tick-borne rickettsiosis

The seroprevalence of Rickettsia was 29.2% (95% CI: 27.4–31.2%) and 25.8% (95% CI: 23.6–28.1%) in female and male participants, respectively (Table 1). From 2016 to 2020, the TBR seroprevalence showed a curve and differed significantly by year (P < 0.001), with a peak of 46% in 2018 (Fig 1A).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 1. Overall Rickettsia seroprevalence trends by age.

Rickettsia seroprevalence is plotted based on generalised additive model trend curves, with 95% confidence intervals assessed from cluster bootstrapping 1000 by year (A) and age (B). Three-dimensional age-specific contour maps for Rickettsia seroprevalence probability for the target year (C).

https://doi.org/10.1371/journal.pone.0289274.g001

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Stratified seroprevalence of IgG antibodies against Rickettsia spp. detected by ELISA in participants aged 0–75 years and the results of binary logistic regression analysis of potential risk factors for seropositivity from 2016–2020 in Mongolia.

https://doi.org/10.1371/journal.pone.0289274.t001

Seroprevalence of tick-borne borreliosis

Among the 4724 serum samples, 162 were positive for IgG antibodies against B. burgdorferi sensu lato. Table 2 shows the seroprevalence of TBB stratified by sex, occupation, age group, and province. The seroprevalence of Borrelia was 3.78% (95% CI 3.1–4.5%) and 3.17% (95% CI: 2.4–4.0) in female and male participants, respectively (Table 2).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 2. Stratified seroprevalence of IgG antibodies against Borrelia burgdorferi sensu lato detected by ELISA in participants aged 0–75 years and the results of binary logistic regression analysis of potential risk factors for seropositivity from 2016–2020 in Mongolia.

https://doi.org/10.1371/journal.pone.0289274.t002

Regarding co-infections, 7 participants were positive for Rickettsia and Borrelia. All seven co-infected patients were female. The 0–4 age group accounted for one case of co-infection, 5–9 age group for three cases, 25–29 age group for one case, 45–49 age group for one case, and 55–59 age group for one case. On the basis of occupation, the co-infection was observed in two employed participants, three school children, and one herder. Two cases of co-infection were noted in Bayankhongor province, and one case each in Dornod, Dundgobi, Khovd, and Selenge provinces.

Age trend analysis.

The seroprevalence of Rickettsia was nonlinearly related to age in all participants (Fig 1). From age 0–19 years, it decreased slowly (Fig 1B) with the maximum value (35.1%) observed between 25 and 29 years, followed by a gradual increase from 25.8% to 31.6% between the ages of 30 and 54 years. From age 55–64 years, it decreased slowly (Fig 1C).

To explore the potential for synergistic statistical interaction between age and Rickettsia seroprevalence, without arbitrarily categorising either continuous variable, generalized additive models that employ 2D smoothing curve on age and year, were used to plot a graph of the three-dimensional association (Fig 1).

The association of age and Rickettsia seroprevalence from the trend analysis held across all years in all participants (Fig 1B). The steeper slope of the age–seroprevalence association, relative to the year–seroprevalence association, implies, as the risk factor analysis did, that age had a stronger association with Rickettsia seroprevalence year. Age being the conditional variable, the association of year with Rickettsia seroprevalence was nearly nonlinear. Combinations of age and year that synergistically increased (or decreased) the conditional probability of TBR infection appeared on the three-dimensional risk surface.

Spatial distribution

The seroprevalence of Rickettsia and TBB was mapped according to province (Figs 2 and 3, respectively). Rickettsia was detected in all 21 provinces, with the highest seroprevalence in the Gobi-Altai Province (79.2%) and the lowest seroprevalence in the Dornogobi Province (7.6%) (Fig 2). Borrelia was detected in 16 of the 21 provinces, with the highest seroprevalence in the Zavkhan Province (13.8%) and the lowest seroprevalence in the Uvs Province (1.2%) (Fig 3). The geographic distribution ofTBR and TBB differed, suggesting a difference in their areas of endemicity. The seroprevalence of both Rickettsia and Borrelia was 10–15% higher in western provinces than in the eastern provinces.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 2. Spatial distribution of the seroprevalence of Rickettsia in the participants.

The colour of each province corresponds to the seroprevalence of Rickettsia (darker colours indicate a higher seroprevalence as shown in the legend).

https://doi.org/10.1371/journal.pone.0289274.g002

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 3. Spatial distribution of the seroprevalence of Borrelia in the participants.

The seroprevalence evaluated for each province is mapped for Borrelia. The shade of each province indicates the seroprevalence of Borrelia (darker colours indicate a higher seroprevalence).

https://doi.org/10.1371/journal.pone.0289274.g003

Discussion

In this study, 4724 sera from a wide geographical range in Mongolia were analysedfor Rickettsia and Borrelia seroprevalence.We observed a high number of tick-borne infections due to the Rickettsia and Borrelia epidemics that swept through Mongolia, peaking between 2016 and 2020.

Von Fricken et al. [14] previously reported a high seroprevalence of Borrelia antibodies in Zavkhan (25%) and Selenge (20%) Provinces after the 2007–2017 Borrelia epidemic in Mongolia. In this study, the seroprevalence of IgG antibodies against B. burgdorferi sensu lato was 13.8% and 5.3% in Zavkhan and Selenge Provinces, respectively.

In a previous study, the seroprevalence of Rickettsia in Mongolia was concentrated in the 0–18-year (21%), 19–49-year (28%), and 40–49-year (33%) age groups [3]. However, in this study, we observed that the seroprevalence of Rickettsia was higher in the 25–29-year age group than in the other age groups. This increase may be explained by the fact that in rural areas of Mongolia, 25 to 29-year-olds spend more time outdoors and are more likely to engage in outdoor activities such as herding their domestic animals in grasslands, camping, gardening, and horse-riding than other age groups. Furthermore, the seroprevalence of Rickettsia was higher in the 5–9- and 10–14-year age groups than in the 35–39- and 40–44-year age groups. It is possible that young children cannot adequately protect themselves from tick bites or infestation. Although the usefulness of conducting inspections of the body for ticks has not been confirmed, there is evidence that removal of Ixodes scapularis within 36 hours after attachment will reduce the risk of TBB [19,20]. Contrarily,Steiner-Khamsi and Gerelmaa foundthat about 40% of herder families wantedtheir children to liveinschooldormitories. Afterthe first school year, many herder families withdrew their children because they performed poorly or because theywere required to repeat a grade due to an poor health dormitory [21]. More girls than boys from herder families are admitted to school. The imbalance is greatest at the upper secondary level, where 60% of girls are admitted to school. Accordingtoan earlier report on male child labour in agriculture and animal husbandry, the majority of male children in rural areas never attend school or drop out [22]. Sincetheyaremore likely to be outside and handle animals, children from herder families maybe more susceptible to TBDs.

Rickettsia and Borrelia serprevalence was higher among herders and students than among those with other occupations. Mongolian herders herd their livestock in common open pastures, whereas nomadic or semi-nomadic herders move from one grazing area to another with their livestock [23].Herders, as a professional group, work in different ecological areas, and their surroundings are a natural ecosystem for ticks. Sex differences in seroprevalence may be attributed to lifestyle or could be attributed to sampling bias if women have a higher tendency to seek medical attention in hospitals than men [24–26]. Tick checks are alsoassociatedwith higher levels of knowledge, according to researchers from The Netherlands [27]. The children ofMongolian livestock herders help their parents raise their young domestic animals. Mongolia is one of the few countries in which nomadic and semi-nomadic lifestyles still exist [3,23]. Consequently, Mongolian herder families, especially children, live in close proximity with domestic animals; therefore, TBDs and zoonoses are major public health problems in this population [1,3–5]. Herders’ children are exposed to tick bites when working with wool and cashmere, wool-baiting, combing goats, and shearing animals.

A study in Mongolia assessed the association between symptoms and tick bites based on serological evidence of Rickettsia and Borrelia infection; however, as the study was conducted in a hospital setting, a selection bias potentially existed because participation was limited to those who presented to the hospital for treatment, and the nomadic herder population may have been underrepresented [14].The prevention of TBDs in Mongolia largely depends on the beliefs, practices, and awareness of members of the herder community. Nomadic and semi-nomadic herders with less than 100 domestic animals are often isolated [23] because they herd in open spaces, and it is difficult for them to access public services, especially healthcare and education. This might explain the large difference between experiencing symptoms of disease and seeking treatment [28]. Not only do Mongolian [3,4] herders have a high seroprevalence of TBR and TBB, herders in China [29], Ghana [30], North Cameroon [31], and Jordan [32] also have a higher seroprevalence of Rickettsia and Borrelia than those in other occupations. According to studies in Europe, the seroprevalence of B. burgdorferi sensu lato is higher in agricultural, farming, and forest industry workers than in the general population [2,6].

Tick-borne diseases are prevalent in the human population and may become an important and persistent threat to human health [8]. Humans are occasional hosts of ticks but do not play a role in maintaining tick-borne agents in nature [33]. In one study, spotted fever group Rickettsia was detected in 19.5% (73/374) of humans [3] and 20.4% (478/2342) of livestock [4] in Mongolia. TBDs can directly affect the lives of Mongolians by causing illness in humans and, indirectly, can cause economic losses due to zoonoses in livestock. Therefore, numerous epidemiological studies have been conducted to determine the seroprevalence of antibodies against the Rickettsia spotted fever group [4,34,35], B. burgdorferi sensu lato [3,5,36,37], Anaplasma spp. [38,39], and molecular identification of tick-borne encephalitis virus [40,41] in Mongolia; however, most previous studies were limited by small sample sizes and limited geographic coverage.

In this study of a hospital-based population between 2016 and 2020, the seropositivities of Rickettsia in all provinces and Borrelia in 16 provinces were 21.8% and 3.4%, respectively.However, previous studies (with sample sizes of 316, 374, 335, and 150) might not have been adequately facilitated [3–5,14]. It is often difficult to compare field surveys on TBR and TBB because of varying sampling methods and sample sizes. Our study confirmed a higher seroprevalence of Rickettsia than that of Borrelia. In one study, all of the Dermacentor spp. ticks tested in Mongolia were positive for Rickettsia [33,42]. In another study, Dermacentor nuttalli was detected in all the ecological areas and provinces of Mongolia [15]. Ixodes persulcatus ticks are the known vectors of borreliosis in Mongolia, whereas Dermacentor and Haemaphysalis spp.ticks are vectors carriers of Rickettsia [43]. Differences in biological niches of ticks from northern to southern provinces may partially explain the ecological and geographical differences in the seroprevalence of Rickettsia and Borrelia.The burden of TBDs in the Selenge Province has been reported previously [5]; however, in this study, we detected a high seroprevalence of Borrelia in the Zavkhan Province. The high seroprevalence of Borrelia in Zavkhan Province is of concern because of the high volume of native and foreign visitors who pass through the province. We infer that the lower seroprevalence of Rickettsia and Borrelia among humans in the Gobi-Altai and Gobi-Sumber provinces, which are situated adjacent to the desert, is attributable to decreased grazing.

According to our study, Rickettsia seroprevalence was higher in the western provinces than in the eastern provinces. Altantogtoh et al. [44] previously reported that Rickettsia was more prevalent in tick pools in the western and eastern provinces of Mongolia than in those in the other provinces. This shows that there is considerable geographic variability in the ecology of Rickettsia spp. pathogens. These differences may reflect differences in human contact with ticks and infected vectors, but more research is needed to clarify these differences.

In previous studies, co-infection with Borrelia burgdorferi and Anaplasma phagocytophylilum was found [45]. Moreover, this was detected with Babesia caballi and Babesia equi [46] infections, according to the relative frequency of pathogen detection in different ticks, as described earlier for I. persulcatus in Mongolia [45,46]. In addition, co-infection with Borrelia and Babesia was found, including infection with D. nuttali (1.1%), in I. persulcatus (28.6%), and H. asiaticum (2.3%) [15]. In this study, Rickettsia and Borrelia co-infection was found in humans (0.14%). These findings emphasize the importance of preventing tick bites and monitoring for multiple infections in both ticks and humans.

The strengths of this study include the large sample size comprising a population with a wide age range, the sensitivity and specificity of the diagnostic tests used to measure anti-Rickettsia and Borrelia antibodies, and the robust analysis of age trends and risk factors. Nevertheless, this study had some limitations. First, the 4724 participants were patients who sought treatment at health facilities; therefore, we were unable to determine the seroprevalence of asymptomatic infection. Collectively, this findings demonstrate the importance of further research in this area to gain a better understanding of this situation.Second, this study did not investigate participants’ knowledge, attitudes, beliefs, and practices with regard to TBDs. Third, the study focused on enrolment at hospitals when nomadic movement could lead to exposure elsewhere, especially given the prolonged period for which antibodies can be detected.

In summary, this study determined the seropositivity of Rickettsia and Borrelia and the risk factors for infection among children and adults in a hospital-based population from Mongolia.The seroprevalence of Rickettsia was high, especially in herders, and seroprevalence studies on Rickettsia and Borrelia should be considered for inclusion in the future. Rickettsia was found in all provinces, whereas Borrelia was found in only 16 provinces of Mongolia. Furthermore, we found that occupation and age were risk factors for rickettsial infections.

Supporting information

Acknowledgments

We thank the local staff of the National Centre for Zoonotic Diseases in all the provincial branches, especially the Zavkhan Province General Hospital for collecting whole blood samples.