Seroprevalence of six pathogens transmitted by the Ixodes ricinus ticks in asymptomatic individuals with HIV infection and in blood donors

The objective of our study was to estimate the seroprevalence of six pathogens transmitted by ticks in HIV-infected persons and blood donors in Poland (B. burgdorferi s.l., A. phagocytophilum, Ehrlichia spp., Babesia spp., Rickettsia spp. Bartonella henselae) to assess the frequency of exposure to such microorganisms in immunocompetent and immunocompromised individuals in endemic regions for I. ricinus ticks. Serum samples were collected from 227 HIV-infected patients and 199 blood donors. All samples were analyzed for antibodies against six tick-borne pathogens and seroprevalence rates were statistically compared between two tested group as well as age, sex and lymphocyte T CD4+ level in HIV infected patients. The seroprevalence of tick-borne infections in HIV-infected patients is higher than that of the healthy population in Poland, although no association between serological status of patients and lymphocyte CD4+ T cell level has been observed. The frequency of tick-borne coinfections and doubtful results of serological tests were significantly higher in HIV-positive individuals. In Poland, the possibility of tick-borne diseases transmission with blood is rather negligible.

Doubtful results. The frequency of doubtful results was significantly higher in HIV-infected patients than in blood donors ( Table 2). In HIV-positive patients the doubtful results for A. phagocytophilum were 9.7% and 6.2%, for E. chaffeensis-4.4% and 6.6%, for B. henselae-2.6% and 2.2% for IgM and IgG tests, respectively, whereas the doubtful results for these pathogens were not noted in blood donors. For B. burgdorferi IgM ELISA test, the amount of doubtful results was similar in both tested groups of patients/participants. In HIV-positive patients, the doubtful results for B. burgdorferi IgG ELISA test were observed more often (2.6% vs. 0.5%); however, this difference was not statistically significant.

Discussion
The large population of immunocompromised patients, including those with HIV/AIDS, is constantly growing. Analysis of the published reports of tick-borne infections shows that the disease in immunocompromised patients is far more severe, prolonged and more likely to be fatal 9,14,28 . Thus far, investigations of the epidemiology of tick-borne pathogen infections involving serological studies have concentrated on inhabitants of endemic regions who were healthy and whose immunological function was normal. By contrast to those studies or to reviews of clinical findings concerning hospitalized patients, our molecular and retrospective study had a different objective: establishing both the incidence and seroprevalence of tick-borne infections in patients with HIV, compared with the control group of healthy blood donors without clinical symptoms. Both groups were based in an endemic area of I. ricinus ticks, namely Poland. The Polish Society of Epidemiology and Infectious Diseases has issued a recommendation regarding diagnostics of tick-borne diseases to which this study particularly conforms.  In HIV-positive patients, borreliosis is rarely reported as co-infection, with only a few cases so far 28,42 . Most of them were identified early, with neuroborreliosis confirmed in Dutch and Swedish patients. Our recent molecular study confirmed the asymptomatic B. garinii infection in an HIV-postitive patient with no signs of early or late-stage Lyme borreliosis 40 . Borrelia burgdorferi was the predominant pathogen in our study, and about 30% and 5% of HIV-infected individuals were recorded as positive for IgM and IgG, respectively. In this group of patients, the IgM seroprevalence rate was significantly higher compared to blood donors. The significantly higher B. burgdorferi seroprevalence was observed in HIV-infected patients aged >35 years and with a median lymphocyte CD4+ T cell less than 300/μl. To the best of our knowledge, as of today, only one serological study of Borrelia infection in HIV-postitive patients was conducted, and the results were comparable -a total of 33% sera were positive 43 . Antigens of Borrelia and Treponema spirochetes often cross-react with each other. False positive results of serologic tests have been obtained in neurological patients with infections with other bacteria in the same group, for instance T. pallidum 42,44 . Therefore, our positive results of IgM and IgG ELISA tests were confirmed by Western Blot according to the European guidelines 45 in order to exclude false positive results. Only half of positive ELISA results were confirmed in WB in HIV-positive patients. Accordingly, our data strongly recommends confirmatory testing in HIV-infected patients with ELISA positive Lyme screening.
Borrelia seroprevalence among blood donors in Europe does not exceed 10% [46][47][48] . In our study, the IgM and IgG seroplevalence in ELISA tests were rather similar (13.1% and 5.0%), and 23% (IgM) and 80% (IgG) of positive/doubtful results were confirmed in WB, respectively. All blood donors with anti-Borrelia IgG antibodies underwent borreliosis in the last 10 years. It is difficult to determine whether tested blood donors were spirochetemic at the time of blood giving, yet most of them declared multiple blood donation in the last decade. Till now, there are only limited and conflicting data on the number of spirochetes in the blood of spirochetemic Lyme disease patients 49,50 . However, as spirochetemia may occur in Lyme disease, the potential for transfusion-transmitted B. burgdorferi exists but has yet to be reported 51 . Ginzburg and co-authors have postulated that host-adapted B. burgdorferi survives poorly under blood storage conditions, particularly if the number of organisms per milliliter of human blood is low. We have not confirmed the presence of B. burgdorferi DNA in the blood of the tested group of donors (data not published) and, therefore, we conclude that the risk of transfusion-transmitted B. burgdorferi is rather negligible.
Babesia is transmitted primarily by ixodid ticks, although blood transfusion is another important cause of infection 52 . To date, babesiosis in HIV-infected individuals has been observed in the United States of America 53,54 , yet one case has been also recorded in Europe (Spain) 29 . In those cases, the phase of infection was determined as chronic since it lasted several months. Relapses of babesiosis occurred despite treatment which because of high parasitemia involved blood transfusions 54 . It is worth to note that recent studies have identified the first case of false-positive HIV serology that was associated with active babesiosis, yet after successful treatment of babesiosis, the positive HIV serology turned negative 55 . Currently, there is no data on Babesia seroprevalence in HIV-positive individuals. In our study, the IgM B. microti seroprevalence in HIV-positive patients was significantly higher than among blood donors whose estimated IgM and IgG seroprevalence (<2%) was comparable to that reported for B. microti among blood donors in Europe [56][57][58] . Nevertheless, IgG seroprevalence among HIV-positive patients and blood donors was similar, and we have not confirmed the presence of Babesia DNA in tested blood samples (data not published). Human babesiosis appears to be rare in Europe 9 , including Poland 11 , which is consistent with a low I. ricinus infection rate (1.6-2.8% in Poland, 0.5% in Slovakia, 0.7% in Switzerland) 4,59-61 .
Human ehrlichiosis and anaplasmosis are acute febrile tick-borne diseases caused by various species from the genera Ehrlichia and Anaplasma. A. phagocytophilum, causing human granulocytotropic anaplasmosis (HGA), and Ehrlichia chaffeensis, the etiologic agent of human monocytotropic ehrlichiosis (HME), are considered as an emerging zoonosis with clinical manifestations ranging from a mild febrile illness to a fulminant disease characterized by multi-organ system failure, especially in immunocompromised individuals 62 . HME and HGA have similar clinical presentations and both pathogens could be transmitted by transfusion or organ transplantation 63,64 . Till now, only one case of asymptomatic A. phagocytopilum infection in HIV-positive patient was confirmed 40 . In our study, HIV-infected patients and blood donors presented a similar rate of A. phagocytophilum seroprevalence (<3%). In Europe, among blood donors, antibodies to A. phagocytophilum vary from 5% in Belgium 65 to 22% in Greece 66 . The low A. phagocytophilum seroprevalence rate found in our study is compatible with the small number of clinical cases identified so far 67 as well as the low prevalence in I. ricinus ticks in Poland 5 .
In persons infected with HIV, ehrlichiosis caused by E. chaffeensis is often life-threatening 68 . However, the disease responds well to specific, antibiotic therapy, particularly when antibiotics are used at an early stage of infection 69 . The E. chaffensis seroprevalence in HIV-infected patients in the US was estimated at 1.7% and was similar to the one observed in healthy persons 70 . Nonetheless, in this group of patients the fatal false-negative results of serological test were reported. Accordingly, the incidence of ehrlichiosis in this population seems to be underestimated 68,69 . The severe pathology and multi-organ involvement in fatal ehrlichiosis that mimics toxic shock-like syndrome was observed in patients who are immunocompromised due to other infections, such as HIV or chemotherapy, and is thought to be related to dysregulation of the host immune response and immunopathologic mechanism that leads to tissue damage and multi-system organ failure 71,72 . In our study, the rate of IgG anti-Ehrlichia antibodies was almost three times higher in HIV-infected individuals than among blood donors. Although our positive patients have not declared clinical manifestation characteristic of tick-borne diseases, the diagnosis of HME in HIV-positive patients is complicated since the symptoms of HME often mimic typical findings commonly associated with HIV-infection 70  to survive in stored blood with the potential for transfusion-associated infection has been shown 73 . Bartonella infection presents varied clinical symptoms, mainly in HIV-infected patients whose bacillary anigiomatosis and hepatic peliosis are classically associated with AIDS 74 . Bartonellosis in this group of patients seems to be less frequent today, possibly because of earlier recognition of HIV serostatus and the lesser number of individuals with CD4 lymphocyte cell counts below 50 cells/mm 3 75 . Our study demonstrated low B. henselae seroprevalence rate among HIV-infected patients and it was similar to that observed in a healthy population of blood donors (<4.5%). Bartonella seroreactivity rates varied between 2% and 30% in studies from Europe and the US with higher prevalence rates in intravenous drug users, homeless people, cat owners and veterinarians 76,77 . Among HIV-infected patients from Europe, anti-Bartonella antibodies were significantly higher than in our study and varied from 16% to 41% of individuals 78,79 . The likelihood of false-negative serological results are high in heavy immunocompromised patients with active Bartonella infection 80 . However, the majority of patients tested in our study have good immunological status -in 81% of tested patients, the level of lymphocyte CD4+ T cells was higher than 300/µl and most of the patients (82%) were on HAART. Furthermore, in contrast to the previous study where the prevalence of anti-Bartonella antibodies is inversely proportional to the number of CD4 lymphocytes 74,81 , we have not found significant association between the Bartonella serological status and lymphocyte CD4+ T cell level in HIV-positive patients. Our study presented only 19% of patients with low CD4 levels, which could significantly alter our results. Even though the rate of Bartonella infection in ticks in Poland is rather low (<2% 82 ), the serological studies have proved that B. henselae and B. quintana are present and widely distributed in Poland in such specific risk groups as: alcoholics, veterinarians and cats' owners 76 . Therefore, the further study determining the Bartonella prevalence among HIV-infected individuals, especially those with a fever of unknown origin, is needed. Intracellular bacteria of the Rickettsiaceae family (spotted fever group (SFG) are responsible for tick-borne rickettsiosis. In humans, the syndrome manifests clinically mainly through rash, fever and 'tache noire' , i.e. eschar developing at the site of the tick bite 83 . In HIV-positive patients, only isolated cases of Mediterranean spotted fever (MSF) have been reported and found to be caused primarily by R. monacensis and R. conorii 39 . Occassionally, MSF presentation is mimicked by primary HIV infection 84 . Till now, only few serological studies of Rickettsia infections among HIV-infected patients in Europe were performed. Nogueras et al 85 . have shown that seroprevalences of R. typhi and R. felis infections do not exceed 7% in this group of patients and were similar to those obtained in healthy subjects from the same region. In our study, the estimated the IgG seroprevalence of SFG Rickettsia was comparable and did not exceed 2% in both groups. Nevertheless, the SFG Rickettsia seroprevalence in Poland seems to be rather high in occupationally exposed populations (forestry and agricultural workers -36%) 86 . Moreover, Rickettsia infection in ticks is high and varies from 4% to 53% depending on the tick species (I. ricinus vs. D. reticulatus, respectively) 5,87,88 . It is likely that, similarly to Bartonella infection, the serologic response in HIV-infected patients with good immunological status could be comparable to that of a healthy population. Therefore, the further study estimating the Rickettsia prevalence among HIV-infected individuals, especially with heavy immunosuppression, should be conducted.
In our study, no significant association was noted between the serological status of patients and their age, sex or lymphocyte CD4+ T cell level in HIV-positive patients. Such medical information about patients was available only for 65% individuals, and it was the primary limitation in our study. However, the previous studies, despite the fact that they possessed all the data, did not find any statistical relations between seropositivities and the assessed variables as well 43,79,85,89 .
Very few HIV-infected patients and blood donors were seropositive for two of the six studied pathogens. The single infection and coinfections were significantly more often noted in HIV-infected patients whose immunodeficiency significantly increases the risk of infection caused by pathogens. Simultaneous seropositivity for A. phagocytophilum and E. chaffeensis was observed in 59% of all coinfections. It is likely that cross-reactive antigens, shared by Ehrlichia and Anaplasma that induce cross-reactive antibodies, may affect the high rate of false-positive results in serological tests 62 . Due to this cross-reactivity among ehrlichial species, sera should be tested against both E. chaffeensis and A. phagocytophilium antigens when ascribing a specific etiology. Coinfections with B. burgdorferi and B. microti were the second most frequent combination noted only in HIV-infected patients. Dunn et al. 90 observed that coinfection with B. burgdorferi and B. microti significantly increases B. microti parasitemia in mice and that larval ticks become infected with B. microti in greater numbers when fed on coinfected hosts. A possible explanation is that the host immune response to disseminating spirochetes is not restricted to the skin and may interfere with the splenic immune response, which is critical for the control and clearance of B. microti infection. Initial case reports suggested that concurrent Lyme disease and babesiosis are associated with severe illness 91 . Accordingly, concurrent babesiosis should be considered for any patient with Lyme disease who experiences more severe illness symptoms than expected, especially when the patient does not respond well to recommended antibiotic therapy.
In our study, the frequency of doubtful results for Anaplasma, Ehlichia and Bartonella was significantly higher in HIV-infected patients than in blood donors. The results of the previous studies have shown that the serological evaluation of the presence of IgG Ehrlichia antibodies in patients in a single sample is not sufficient to confirm ehrlichiosis. Non-specific reactions, which are more often observed in the group of patients infected with HIV, may result from the dysfunction of an immunological system, and they might be the consequence of the cross-reaction with other pathogens 92 . In consideration of the above, examination of many samples using concurrent serological and molecular tests might be crucial to confirm infection with the specific grade of pathogen (for example Ehrlichia or Anaplasma) 69,92 . Similarly, in the case of serological diagnostic of Bartonella, in patients infected with HIV, the non-specific or false positive results of serological tests are explained by the cross-reaction with the following pathogens: Coxiella burnetti or Chlamydia trachomatis 93,94 .
In conclusion, our study confirmed that the seroprevalence of tick-borne infections in HIV-infected patients is higher than that of the healthy population in Poland, however no association between serological status of patients and lymphocyte CD4+ T cell level has been observed. The frequency of tick-borne coinfections and doubtful results of serological tests seems to be higher in HIV-positive individuals. Although the advent of HAART had a considerable impact on the incidence of AIDS-associated opportunistic infections, the further studies of tick-borne infection in HIV-infected patients, particularly in patients who do not regain immunological function despite well controlled HIV replication on effective HAART, should be performed 95 . In this group of patients, tick-borne pathogens may cause chronic, debilitating opportunistic infection and even death. Thus, in clinical care of HIV-positive individuals, detailed history of tick bites in endemic tick areas should be collected via directed anamnesis. The low seroprevalence and negative results of molecular studies of tick-borne pathogens in blood donors (data not published) have suggested that the possibility of tick-borne diseases transmission with blood is rather negligible, which is consistent with the lack of reported cases of transfusion-transmitted tick-borne infections in Poland. Nevertheless, there is still a clear need to further such studies in order to maintain a balance between consideration of the real risk of disease transmission with blood and excessively restrictive approach that eliminates blood donors.

Methods
Selection and recruitment of patients/participants and serum samples. In 2016, serum samples were collected from 199 blood donors (representing the control group) who were diagnosed in AmerLab Ltd. Diagnostic Laboratory of Parasitic Diseases and Zoonotic Infections. Subjects with immunodeficiency were excluded. All participants signed informed consent and obtained a standardized, anonymous questionnaire to record data including age, gender, immunological status, place of residence, the history of tick bite, as well as previously diagnosed borreliosis or other tick-borne diseases. Blood samples obtained from blood donors were stored at room temperature and centrifuged immediately or within a maximum of 12 h after collection. Sera were frozen at −20 °C until further analysis.
The retrospective study was conducted on HIV-positive patients who did not have any known history of tick bite, nor any clinical manifestation characteristic for tick-borne diseases. In 2013, serum samples were collected from 227 patients routinely followed at the HIV Outpatients' Clinic of the Hospital for Infectious Diseases in Warsaw.
The study protocol followed ethical guidelines of the 2013 Declaration of Helsinki and the study was approved by the Internal Review Board of the Warsaw Medical University (no. AKBE/24/16). Informed consent was obtained from all individual participants included in the study. All ethical approvals for the study have been obtained in accordance with the Polish regulations. (5) Spotted Fever Rickettsia IgG EIA antibody kit (Fuller Laboratories, California, the USA; the cut-off calibrator is set and the index value for each serum is derived. Indices from 0.9 to 1.1 absorbance units may be considered equivocal, while those above 1.1 are considered positive and those below 0.9 are considered negative) for Rickettsia spp.; (6) Bartonella henselae IFA Human IgM and IgG antibody Kit (Fuller Laboratories, California, the USA; positive cut-off 1/512) with the manufacturer's interpretation criteria.
Statistical analysis. Statistical analysis was performed using IBM SPSS Statistics v. 23.0 software. For the analysis of the results, doubtful serological results of tested pathogens were classified as negative. A descriptive analysis of the participants was included and calculations of seroprevalence rates for each pathogen were performed. Seroprevalence rates were compared with the tested group (HIV infected patients/blood donors), age, sex, lymphocyte T CD4+ level (HIV infected patients) using Maximum Likelihood techniques based on log-linear analysis of contingency tables (HILOGLINEAR).
Ethics approval and consent to participate. The study protocol followed ethical guidelines of the 2013 Declaration of Helsinki. The study was approved by the Internal Review Board of the Warsaw Medical University (No. AKBE/24/16). Informed consent was obtained from all individual participants included in the study. All ethical approvals for the study have been obtained according to Polish regulations.

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