Seroprevalence of Babesia Infections in Humans Exposed to Ticks in Midwestern Germany

Transcription

1 JOURNAL OF CLINICAL MICROBIOLOGY, July 2002, p Vol. 40, No /02/$ DOI: /JCM Copyright 2002, American Society for Microbiology. All Rights Reserved. Seroprevalence of Babesia Infections in Humans Exposed to Ticks in Midwestern Germany Klaus-Peter Hunfeld, 1 * Annette Lambert, 1 Helge Kampen, 2 Sabine Albert, 1 Christian Epe, 3 Volker Brade, 1 and Astrid M. Tenter 3 Institute of Medical Microbiology, University Hospital of Frankfurt, D Frankfurt/Main, 1 Institute for Medical Parasitology, University of Bonn, D Bonn, 2 and Department of Parasitology, School of Veterinary Medicine Hannover, D Hannover, 3 Germany Received 25 February 2002/Returned for modification 11 April 2002/Accepted 22 April 2002 Babesiosis is considered to be an emerging tick-borne disease in humans worldwide. However, most studies on the epidemiology of human babesiosis to date have been carried out in North America, and there is little knowledge on the prevalence of infection and frequency of disease in other areas. The aim of this study was to investigate the prevalence of Babesia infections in a human population in Germany. A total of 467 sera collected between May and October 1999 from individuals living in the Rhein-Main area were tested for the presence of immunoglobulin G (IgG) and IgM antibodies to antigens of Babesia microti and Babesia divergens by indirect fluorescent-antibody (IFA) tests. These sera were derived from 84 Lyme borreliosis patients suffering from erythema migrans, 60 asymptomatic individuals with positive borreliosis serology, and 81 individuals with a history of tick bite. Cutoff values for discrimination between seronegative and seropositive results in the IFA tests were determined using sera from 120 healthy blood donors and 122 patients suffering from conditions other than tick-borne diseases (malaria, n 40; toxoplasmosis, n 22; syphilis, n 20; Epstein-Barr virus infection, n 20; and presence of antinuclear antibodies, n 20). The overall specificities of the IFA tests for B. microti and B. divergens were estimated to be >97.5%. Positive IgG reactivity against B. microti antigen (titer, >1:64) or B. divergens antigen (titer, >1:128) was detected significantly more often (P < 0.05) in the group of patients exposed to ticks (26 of 225 individuals; 11.5%) than in the group of healthy blood donors (2 of 120 individuals; 1.7%). IgG antibody titers of >1:256 against at least one of the babesial antigens were found significantly more often (P < 0.05) in patients exposed to ticks (9 of 225) than in the control groups (1 of 242). In the human population investigated here, the overall seroprevalences for B. microti and B. divergens were 5.4% (25 of 467) and 3.6% (17 of 467), respectively. The results obtained here provide evidence for concurrent infections with Borrelia burgdorferi and Babesia species in humans exposed to ticks in midwestern Germany. They also suggest that infections with Babesia species in the German human population are more frequent than believed previously and should be considered in the differential diagnosis of febrile illness occurring after exposure to ticks or blood transfusions, in particular in immunocompromised patients. Babesiosis, which is caused by intraerythrocytic parasites of the protozoan genus Babesia, is one of the more common diseases of free-living animals worldwide and is gaining increasing attention as an emerging tick-borne zoonosis in humans (5, 11, 15, 16). The parasites are named after the Romanian scientist Victor Babes, who in 1888 first identified pear-shaped, Plasmodium-like parasites as a cause of febrile hemoglubinuria in cattle (1). To date, about 100 species of Babesia have been described worldwide, based on intraerythrocytic stages detected in mammals (20). However, their pleomorphism in different species of mammalian hosts and the results of recent molecular systematic studies render some of these descriptions dubious. It is now assumed that some species of Babesia are less host specific than believed previously and that the number of valid species in the genus will be gradually reduced as more information becomes available on them (5, 11, 16). Since the late 1950s, two species of Babesia in particular, i.e., the cattle species Babesia divergens in Europe and the rodent species * Corresponding author. Mailing address: Institute of Medical Microbiology, University Hospital of Frankfurt, Paul-Ehrlich-Str. 40, D Frankfurt/Main, Germany. Phone: Fax: K.Hunfeld@em.uni-frankfurt.de. Babesia microti in North America, have been shown to cause a significant number of infections in humans (9, 15). In addition, other Babesia species of unknown identity, presently designated WA1 type, MO1, and CA1 to CA4, have been found to be pathogens of considerable concern for humans in the United States (10, 15, 25, 29), with recent molecular phylogenetic studies suggesting that some of these parasites may be derived from dogs or wildlife (15). It has also been suggested that species of Babesia infecting rhesus monkeys, formerly described as belonging to the genus Entopolypoides, may have zoonotic potential (4). The frequent reports of human babesiosis from North America are in contrast to only sporadic reports of the disease from Europe and other parts of the world. However, in many non-american regions microbiological investigations and awareness of the clinical presentation of babesiosis in humans lag far behind those in North America. Moreover, diagnostic tools such as indirect fluorescent-antibody (IFA) tests or PCRbased assays designed for specific and reliable detection of the pathogens are not readily available to diagnostic laboratories in Europe. Nevertheless, recent seroepidemiological studies suggested that Babesia infections may occur more frequently than previously believed in patients exposed to ticks in Europe, 2431

2 2432 HUNFELD ET AL. J. CLIN. MICROBIOL. with seroprevalences ranging between 4 and 13% in the populations investigated (9, 12). While the specific vectors of many species of Babesia are still unknown, those of zoonotic potential are known to be transmitted to their vertebrate hosts by ixodid ticks (11, 15, 16). Recent molecular studies in some European countries based on DNA sequence analyses have shown that tick-borne pathogens other than Borrelia burgdorferi, such as species of Ehrlichia, Babesia, and Bartonella, are prevalent in the three-host tick Ixodes ricinus (2, 7, 26), whose larvae, nymphs, and adults feed on different hosts, including virtually any warm-blooded animal and humans. This tick is the most common tick in western and central Europe (12, 32). It is also regarded as the most important vector for tick-borne diseases in humans in these regions, and, although it is not known for certain, there is circumstantial evidence that I. ricinus transmits at least B. divergens to humans (6, 16). Consequently, it has been suggested that babesiosis in humans is an underdiagnosed disease in the European part of the Northern hemisphere. To contribute to the ongoing discussion of whether Babesia infections are common in European human populations or not, we have used antigens of the two Babesia species most frequently reported to occur in humans, i.e., B. microti and B. divergens, as surrogate markers in a seroepidemiological study to detect anti-babesia antibodies in patients exposed to ticks and human control groups in midwestern Germany. MATERIALS AND METHODS Serum samples. The serum samples tested in this study initially were submitted to a diagnostic microbiological laboratory and stored at 20 C until used for serological testing. In total, 467 sera collected between May and October 1999 were selected for retrospective analysis for reactivity with babesial antigens. All sera were obtained from residents of the Rhein-Main area (midwestern Germany) with different histories of tick exposure and from patients with infectious or chronic diseases other than Lyme borreliosis. (i) Patients exposed to ticks. Group I consisted of serum samples obtained from 84 Lyme borreliosis patients (median age, 43.3 years; male/female ratio, 1.2:1) suffering from clinically diagnosed erythema migrans (EM). Group II consisted of serum samples obtained from 60 individuals (median age, 50.6 years; male/female ratio, 1:1.6) randomly taken from the Lyme screening program, which is routinely performed on newly admitted patients by some general practitioners. These samples displayed positive Lyme serology as determined by a commercially available enzyme-linked immunosorbent assay (ELISA) based on whole-cell extract (DadeBehring, Marburg, Germany) and a whole-cell immunoblot (Viramed, Planegg, Germany), but the respective patients lacked clinical symptoms of active Lyme disease, i.e., serum scars. Group III consisted of serum samples from 81 individuals (median age, 42.2; male/female ratio, 1:1) with recorded tick bites in their recent medical history (2 weeks to 3 months before serum collection) but without any clinical manifestation of Lyme disease. (ii) Blood donor sera. Serum samples from 120 age-matched healthy blood donors living in the Rhein-Main area were provided by the blood bank of Frankfurt/Main, Germany (group IV). These sera were used to determine cutoff titers for discrimination between positive and negative reactions in the IFA tests for B. microti and B. divergens, taking into account the local epidemiological situation; i.e., serum samples were taken at random, but blood donors with a known history of tick bite or clinical manifestation of Lyme disease were excluded from this group. (iii) Sera from patients with conditions other than tick-borne diseases. To evaluate the usefulness of the cutoff titers determined for the IFA tests for B. microti and B. divergens as described above with respect to seroepidemiological studies in Germany, a control set of sera was derived from 122 individuals with conditions other than borreliosis and no history of tick bite. These included 20 individuals with active or recent Epstein-Barr virus (EBV) infection, 22 individuals with active or recent toxoplasmosis, 20 individuals with active or recent syphilis as diagnosed by experienced physicians according to clinical symptoms and laboratory data, and 20 individuals with a positive result for antinuclear antibodies (ANA). In addition, sera were obtained from 40 patients with recent malaria to investigate potential cross-reactivity of the B. microti and B. divergens antigens with anti-plasmodium antibodies. These samples were provided by one of the German reference laboratories for malaria at the Institute for Medical Parasitology, University of Bonn, Bonn, Germany. B. microti IFA test. A commercially available test kit (B. microti-ifa IgM/IgG; MRL Diagnostics, Cypress, Calif.) was used according to the manufacturer s specifications to diagnose serum antibodies to B. microti (12, 14). For detection and semiquantification of specific antibodies, this assay uses golden hamster erythrocytes infected with the B. microti GI strain, which was originally isolated from a human patient who acquired the infection on the east coast of the United States (23). B. divergens IFA test. For diagnosis of antibodies to B. divergens, a laboratoryderived IFA test, previously developed for diagnosis of B. divergens-specific antibodies in cattle, was adapted for the examination of human sera (14). IFA antigens were prepared from the blood of jirds (Meriones unguiculatus) experimentally infected with a B. divergens isolate that was originally obtained from naturally infected cattle in northern Germany. Blood was collected from the jirds within 5 days after experimental infection at a parasitemia of about 20 to 25% using sodium citrate as an anticoagulant and then washed three times in potassium-free phosphate-buffered saline (ph 7.2) according to the method of Tenter and Friedhoff (28). The antigen suspension was diluted in phosphate-buffered saline and dispensed onto individual fields of microscopic slides so that 15 to 20 parasitized erythrocytes were distributed within one field of view (diameter, 180 m). The slides were air dried at 27 C and stored at 70 C until used in the IFA tests. Serological testing. The presence or absence of seroreactivity in the IFA tests for B. microti and B. divergens was studied using fluorescein-conjugated goat anti-human immunoglobulin M (IgM) and IgG antibodies (MRL Diagnostics). All sera were titrated in triplicate on different days, and titers are reported as geometric mean titers from three separate experiments. In both assays, positive and negative controls provided by the manufacturer were used to ensure accurate test performance. In addition, sera that were tested in IgM assays were preincubated with rheumatoid factor absorbance reagent (MRL Diagnostics). Measurement of antibodies started at a serum dilution of 1:20 (IgM) or 1:32 (IgG). Sera were then diluted serially in twofold steps to determine end point titers. Specific IgG cutoff titers for the B. microti and B. divergens IFA tests were first determined with the 120 sera from healthy blood donors and were set at the 98th percentile. Cutoff titers were then evaluated with the 122 sera from patients with conditions other than tick-borne diseases. Borrelia burgdorferi ELISA and immunoblot. All sera from the groups of patients exposed to ticks and healthy blood donors (groups I to IV) were tested for anti-borrelia IgG and IgM antibodies in a whole-cell extract ELISA (Dade- Behring) (12). Likewise, serum samples obtained from individuals with disorders other than Lyme borreliosis that were positive for anti-babesia antibodies were tested for anti-borrelia burgdorferi antibodies. Serum samples positive for IgG and/or IgM were then confirmed by a specific whole-cell lysate immunoblot (Viramed) with Borrelia afzelii strain Pko as an antigen. Interpretation of the immunoblot test results was done using criteria previously developed on the basis of clinical testing and a recently published mathematical analysis (32). Statistics. The Fischer-Yates exact test was used for statistical analysis of serological results obtained for the various groups of patients. RESULTS Determination of cutoff titers. For the B. microti and B. divergens IFA tests used in this study, cutoff titers were determined as follows. In the B. microti IgG assay, eight serum samples (6.7%) from the 120 randomly chosen healthy blood donors from the Rhein-Main area (group IV) were positive at a serum dilution of 1:32, and two of these sera (1.7%) were positive at a dilution of 1:64. In the B. divergens IgG assay, 11 serum samples (9.2%) were positive at a dilution of 1:32; 4 (3.3%) of these were positive at a dilution of 1:64, and 1 (0.8%) was positive at a dilution of 1:128. Reactivity with both babesial antigens in the IgG assays was found for one sample (0.8%). In order to attain a high specificity for the IFA tests used in this study, cutoff titers for discrimination between seronegative and

3 VOL. 40, 2002 SEROPREVALENCE OF BABESIA INFECTION IN GERMANY 2433 TABLE 1. Evaluation of specificities of IFA tests for B. microti and B. divergens with serum samples from 122 patients with disorders other than tick-borne infections Condition of patients (n) No. (%) of seropositive samples a B. microti IgG assay B. divergens IgG assay Positive EBV serology (20) 1 (5.0) 1 (5.0) Positive ANA serology (20) 0 (0) 1 (5.0) Active or recent malaria (40) 0 (0) 1 (2.5) Active or recent Toxoplasma 0 (0) 2 (9.1) infection (22) Active or recent syphilis (20) 1 (5.0) 0 (0) Total (122) 2 (1.6) 5 (4.1) a Seropositivity was defined using the following cutoff titers: B. microti IgG assay, 1:64; B. divergens IgG assay, 1:128. All of the 122 samples were negative in the IgM assays with B. microti or B. divergens antigens at a serum dilution of 1:20. seropositive reactions were set at the 98th percentile according to World Health Organization guidelines (33), i.e., at 1:64 for the B. microti IgG assay at 1:128 and for the B. divergens IgG assay. Tests for anti-babesia IgM antibodies are known to frequently give false-positive results (11). Therefore, to obtain a higher specificity in this study and to minimize false-positive interpretation of the IgM assays, all sera were initially tested for anti-babesia IgG antibodies only. Only those sera that were positive for Babesia-specific IgG antibodies were then also tested for the presence of anti-babesia IgM antibodies, thereby indicating a more recent infection. For the IgM assays with B. microti or B. divergens antigens, a cutoff titer of 1:20 was used, because all 120 sera derived from healthy blood donors were negative with the babesial antigens at this serum dilution. Application of these cutoff titers to sera from 122 patients with infectious diseases other than borreliosis (EBV, syphilis, malaria, and toxoplasmosis) or with autoimmune diseases (ANA-positive patients) resulted in detection of IgG seroreactivity with the B. microti antigen in two (1.6%) and with the B. divergens antigen in five (4.1%) of the problematic samples (Table 1). Nonspecific reactions directed against erythrocytes or increased background reactivity was seen with only a few of these sera and could easily be distinguished from specific fluorescence as displayed by seropositive samples. All sera of this group that were positive in IgG assays were negative in IgM assays with babesial antigens and were also negative for antibodies to Borrelia burgdorferi. When the sera from healthy blood donors and the samples from diseases not due to tick infection were combined, the overall specificity of the B. microti IFA test was estimated to be 98.6% for detection of IgG antibodies. For the B. divergens IFA test, the overall specificity was estimated to be 97.5% for detection of IgG antibodies. For comparison, the specificities of the ELISA used in this study for detection of anti-borrelia burgdorferi antibodies in patients exposed to ticks were 95% for IgG antibodies and 99.2% for IgM antibodies (Table 2). Patients with EM. The prevalences of IgG and IgM antibodies to Borrelia burgdorferi in the sera of the 84 Lyme borreliosis patients with clinically diagnosed EM (group I) were 26.2 and 70.2%, respectively (Table 2). Examination of these serum samples in the IFA tests with babesial antigens revealed IgG antibody titers of 1:64 for B. microti in eight cases (9.5%) and IgG titers of 1:128 for B. divergens in five cases (6.0%). IgG titers to the babesial antigens ranged up to 1:256 (Table 3). An additional IgM response was detected in three of the B. microti- and two of the B. divergens-positive patients (Tables 2 and 3). Immunoreactivity with both babesial antigens was present in two of these cases. Five of the 11 Babesia-positive samples also showed IgG and/or IgM reactivity against Borrelia burgdorferi (Table 3). Individuals with positive borreliosis serology but lacking clinical symptoms of active Lyme disease. IgG antibody titers of 1:64 or 1:128 against B. microti were found in four sera (6.7%) obtained from the 60 individuals of group II. Two of these serum samples were also positive for IgM antibodies to B. microti, and one (1.7%) of these samples also showed IgG reactivity with the B. divergens antigen (Tables 2 and 3). Patients with a history of tick bite. IgG and IgM antibodies to Borrelia burgdorferi were detected in 13 (16.0%) and 16 (19.8%) of the 81 patients known to have had a recent tick infestation (group III). Anti-B. microti IgG antibodies were found in nine (11.1%) and anti-b. divergens IgG antibodies were found in five (6.2%) of these samples (Table 2). Reactions with both babesial antigens were present in three cases. IgG antibody titers in this group ranged up to 1:2,048 for B. TABLE 2. Results obtained for serum samples from 225 patients exposed to ticks and 120 healthy controls from the Rhein-Main area No. (%) of seropositive samples a Group(s) (n) Borrelia burgdorferi B. microti B. divergens IgG assay IgM assay Only IgG assay IgG and IgM assays Only IgG assay IgG and IgM assays B. microti and B. divergens I, patients with EM (84) 22 (26.2) 59 (70.2) 8 (9.5) 3 (3.6) 5 (6.0) 2 (2.4) 2 (2.4) II, asymptomatic individuals seropositive 56 (93.3) 17 (28.3) 4 (6.7) 2 (3.3) 1 (1.7) 0 (0) 1 (1.7) for Lyme borreliosis (60) III, patients with history of tick bite (81) 13 (16.0) 16 (19.8) 9 (11.1) 3 (3.7) 5 (6.2) 0 (0) 3 (3.7) I III, all patients exposed to ticks (225) 91 (40.4) 92 (40.9) 21 (9.3) 8 (3.6) 11 (4.9) 2 (0.9) 6 (2.7) IV, healthy blood donors (120) 5 (4.2) 1 (0.8) 2 (1.7) 0 (0) 1 (0.8) 0 (0) 1 (0.8) I IV, total (345) 96 (27.8) 93 (27.0) 23 (6.7) 8 (2.3) 12 (3.5) 2 (0.6) 7 (2.0) a Seropositivity was defined using the following cutoff titers: B. microti IgG assay, 1:64; B. divergens IgG assay, 1:128; B. microti IgM assay, 1:20; B. divergens IgM assay, 1:20.

4 2434 HUNFELD ET AL. J. CLIN. MICROBIOL. TABLE 3. Comparison of antibody titers in serum samples reactive against at least one of the babesial antigens Condition and sample no. Antibody titer with babesial antigen B. microti B. divergens Reactivity in assays for Borrelia burgdorferi IgG assay IgM assay IgG assay IgM assay IgM assay IgG assay EM (group I) 11 a 1:256 b 72 1:256 1:128 1: :128 1: : :128 1: :128 1: : : :64 1: :128 1: :64 Seropositive for Borrelia burgdorferi (group II) 134 1:128 1:20 1: : : :64 1:80 History of tick bite (group III) 1 1: :128 1: :256 1: :128 1: : : :64 1: : :128 1:1, : :2,048 1:1280 Healthy blood donors (group IV) 438 1: :256 1:128 EBV positive 9 1: :64 ANA positive (9) 1:128 Malaria (MID806) 1:128 Toxoplasmosis : :128 Syphilis (7456) 1:64 a, negative result. b, positive result. microti and up to 1:1,024 for B. divergens (Table 3). An additional IgM response was detected in three of the B. microtipositive individuals. Four of the 11 Babesia-positive patients also had IgM or IgG antibodies to Borrelia burgdorferi (Table 3). One patient in this group had an IgG titer of 1:2,048 and an IgM titer of 1:1,280 against B. microti. This patient was negative for anti-b. divergens and anti-borrelia burgdorferi antibodies but reported five tick bites and a subsequent flu-like illness that occurred 3 weeks before the first serum sample was collected. However, we were not able to detect babesial stages in Giemsa-stained blood smears or babesial DNA in a B. microtispecific PCR (7) carried out on an EDTA-blood sample that was taken from this patient 8 weeks after tick infestation. Statistical analysis. Specific IgG antibody titers reflecting an infection with B. microti (titer, 1:64) or B. divergens (titer, 1:128) were observed significantly more often (P 0.05) in the patients exposed to ticks (groups I to III; 26 [11.5%] out of 225) than in healthy blood donors (2 [1.7%] out of 120 individuals) (Tables 2 and 3). Moreover, IgG titers of 1:256

5 VOL. 40, 2002 SEROPREVALENCE OF BABESIA INFECTION IN GERMANY 2435 against at least one of the babesial antigens were found with significantly greater frequency (P 0.05) in the patients exposed to ticks (9 [4%] out of 225 individuals) than in the control groups (1 [0.4%] out of 242 individuals), with the highest antibody titers to babesial antigens being observed in patients with a history of recent tick infestation (Table 3). DISCUSSION Thus far, diagnosis of human babesiosis in Europe has been based mainly on the detection of the parasites in blood smears of patients with clinical symptoms of disease. One problem with diagnosing Babesia-specific antibodies in human sera is that serological tests for Babesia are not commercially available to diagnostic laboratories in Europe and are not standardized. Along with lack of clinical experience with the disease in humans, nonspecific or subclinical disease manifestations exacerbated by a lack of available diagnostic tests may explain why virtually nothing is known to date about disease caused by B. microti and B. divergens in immunocompetent humans in Germany. A pilot study on a small group of patients in 1998 suggested seroreactivity with babesial antigens in 11.8% of Lyme borreliosis patients from the Rhein-Main area (midwestern Germany) as diagnosed by an IFA test using B. microti as a surrogate antigen, although serological testing was performed without particular adaptation of cutoff values or evaluation of test specificity with respect to the local situation (12). Here, we have used IFA tests for the detection of anti-b. microti and anti-b. divergens antibodies in a larger group of individuals under test conditions that have been optimized for seroepidemiological studies in Germany. IFA tests are known to be sensitive, specific, and reproducible for diagnosis of B. microti-specific antibodies in human sera (11, 17). Here, we evaluated and adapted a commercially available IFA test that is based on a North American isolate of B. microti for the detection of antibodies in sera of individuals exposed to ticks from midwestern Germany. In addition, we have adapted and standardized an IFA test previously developed for detection of B. divergens-specific antibodies in cattle for use in a human diagnostic laboratory. In this study, the use of cutoff titers of 1:64 for B. microti and 1:128 for B. divergens provided excellent overall specificities of 98.6% for detection of IgG antibodies in the B. microti IFA test and 97.5% for detection of IgG antibodies in the B. divergens IFA test, as revealed by testing 120 sera from healthy blood donors and 122 sera from patients with autoimmune disorders or infections other than tick-borne diseases. These specificities correspond well with results from other studies that reported specificities ranging from 90 to 100% for comparable test systems (17). Moreover, titers of 1:32 to 1:160 have been reported to be both diagnostic and specific, with positive predictive values of 69 to 100% and negative predictive values of 96 to 99% (17). For B. microti, sporadic reports of human infections from France and Germany have claimed to show asymptomatic seropositive individuals, but so far no systematic studies have been carried out in these countries (9). Both B. microti and B. divergens have been isolated from ticks, rodents, and cattle in Germany (13, 21, 31). Moreover, the potential relevance of Babesia species for individuals exposed to ticks in European countries was very recently substantiated by demonstrating the presence of B. microti, B. divergens, and closely related species in 9.6% of I. ricinus ticks in Slovenia by PCR and subsequent nucleotide sequence analysis of the small-subunit rrna gene (6). In Germany, I. ricinus is widely distributed and is regarded as the main vector of tick-borne infections to humans (12, 32). It is interesting that the seroprevalence of infections with Babesia species (11.5%) in the humans exposed to ticks examined here is similar to that found in German companion animals. Thus, a recent seroepidemiological study reported a seroprevalence of 15% for B. microti in dogs in Germany (22). The prevalence of antibodies to either B. microti or B. divergens in the individuals exposed to ticks tested here (26 of 225; 11.5%) was significantly higher than that in the control group of healthy blood donors (2 of 120; 1.7%). Interestingly, titers of 1:256, possibly indicating a more recent infection with the pathogen, were found with significantly greater frequency (9 versus 1; P 0.05) in patients exposed to ticks than in the control groups. The seroprevalences of Babesia infections in the German population examined here correlate well with those in a study of Swedish Lyme borreliosis patients, in which 13% of the individuals were found to be seropositive for B. divergens (30), and with those in a study in western France that indicated 0.5% asymptomatic seropositive individuals out of 408 investigated when a more conservative cutoff titer of 1:80 was used (A. Gorenflot, M. Marjolet, L. Hostis, A. Coutarmanac h, and A. Marchand, Abstr. 3rd Int. Conf. Malaria Babesiosis, p. 134, 1987). Moreover, the overall prevalence of antibodies to babesial antigens (26 of 225; 11.5%) in the individuals exposed to ticks tested here is similar to the positivity rate for Babesia species observed in ticks (9.6 to 16.3%) in Europe as recently determined by molecular biological methods (6, 27). Any discussion of seroepidemiological data concerning babesiosis must take into account interspecies reactivity of antigenic components within the genus Babesia and cross-reactivity with other bacterial or parasitic agents (11, 12, 14). The results obtained with sera from patients with active or recent toxoplasmosis, malaria, or syphilis did not demonstrate increased reactivity in the IFA tests used here. If cross-reactions between Babesia and Borrelia were to occur, one would expect corresponding interactions with Treponema pallidum, to which Borrelia burgdorferi shows a close antigenic relationship (12, 32). The phenomenon of coinfection with Babesia and other tickborne pathogens, particularly with Borrelia burgdorferi, has caused growing concern. In Europe there are only few reports on potential coinfection with B. divergens, as determined by seroreactivity (asymptomatic infection), and Borrelia burgdorferi (8). By contrast, it has been estimated that as many as 13% of Lyme disease patients are coinfected with B. microti in areas of endemicity in the United States (3, 18, 19). Furthermore, it has been speculated that the increasing B. microti seropositivity seen during the past 30 years in the United States is consistent with the increased incidence of Lyme disease (19). When dealing with the actual frequency of infections in European countries, it must be considered that in immunocompetent individuals babesiosis is probably mild and self-limiting. Therefore, it is likely that most cases take a subclinical course and that undiagnosed carriers exist. Moreover, the initial symptoms of both human babesiosis and Lyme borreliosis are known to

6 2436 HUNFELD ET AL. J. CLIN. MICROBIOL. overlap significantly; both diseases cause nonspecific symptoms, such as fever, fatigue, and flu-like illness (11, 25). Thus, patients with an inadequate response to appropriate therapy for proven or suspected Lyme disease following a tick bite should be examined for infections with other tick-borne agents, including human granulocytic ehrlichiae, tick-borne encephalitis virus, and Babesia species (12). The apparent existence and high prevalence of chronic babesial infections may become increasingly important because asymptomatic but chronically infected blood donors are now known to be a source of transfusion-transmitted babesiosis in areas where Babesia species with zoonotic potential are endemic (24). As a consequence, seroepidemiological and molecular epidemiological studies are required to determine the true distribution and medical relevance of babesial pathogens in the various parts of Europe. Several hundred infections with B. microti in humans in the coastal areas of the New England states of the United States have been reported (11, 15). By contrast, reports on human infections with B. microti in Europe have been sparse, and little is known about their frequency and importance in that part of the northern hemisphere (9, 12, 21, 25). In the past, human babesiosis has been considered to occur rarely in Europe, with only about 30 reported cases, and all but four clinical cases have been attributed to B. divergens, which has been diagnosed mainly in splenectomized patients in France and Great Britain (8). However, these data are unlikely to accurately reflect the true epidemiological situation of Babesia infections or distribution of the pathogens in the European human population. Rather, it is likely that increasing scientific and medical interest in human babesiosis worldwide will result in larger numbers of reported cases in Europe and other parts of the world and that different clinical pictures of the disease in immunocompetent hosts will emerge (11). ACKNOWLEDGMENT We thank I. Gutgesell for excellent technical assistance. REFERENCES 1. Babes, V Sur l hemoglobinurie bacterienne boeuf. C. R. Acad. Sci. 107: Baumgarten, B. U., M. Röllinghoff, and C. Bogdan Prevalence of Borrelia burgdorferi and granulocytic and monocytic ehrlichiae in Ixodes ricinus ticks from southern Germany. J. Clin. Microbiol. 37: Benach, J. L., and G. S. Habicht Clinical characteristics of human babesiosis. J. Infect. Dis. 144: Bronsdon, M. A., M. J. Homer, J. M. H. Magera, C. Harrison, R. G. Andrews, J. T. Bielitzki, C. L. Emerson, D. H. Persing, and T. R. Fritsche Detection of enzootic babesiosis in baboons (Papio cynocephalus) and phylogenetic evidence supporting synonymy of the genera Entopolypoides and Babesia. J. Clin. Microbiol. 37: Cox, F. E. G Babesiosis and malaria, p In S. R. Palmer, E. J. L. Soulsby, and D. I. H. Simpson (ed.), Zoonoses: biology, clinical practice, and public health control. Oxford University Press, Oxford, United Kingdom. 6. Duh, D., M. Petrovec, and T. Avsic-Zupanc Diversity of babesia infecting European sheep ticks (Ixodes ricinus). J. Clin. Microbiol. 39: Eskow, E. S., P. J. Krause, A. Spielman, K. Freeman, and J. Aslanzadeh Southern extension of the range of human babesiosis in the eastern United States. J. Clin. Microbiol. 37: Gorenflot, A., K. Moubri, E. Precigout, B. Carcy, and T. P. Schetters Human babesiosis. Ann. Trop. Med. Parasitol. 92: Granström, M Tick-borne zoonoses in Europe. Clin. Microbiol. Infect. 3: Herwaldt, B., D. H. Persing, E. A. Precigout, W. L. Goff, D. A. Mathiesen, P. W. Taylor, M. L. Eberhard, and A. F. Gorenflot A fatal case of babesiosis in Missouri: identification of another piroplasm that infects humans. Ann. Intern. Med. 124: Homer, M. J., I. Aguilar-Delfin, S. R. Telford 3rd, P. J. Krause, and D. H. Persing Babesiosis. Clin. Microbiol. Rev. 13: Hunfeld, K.-P., R. Allwinn, S. Peters, P. Kraiczy, and V. Brade Serologic evidence for tick-borne pathogens other than Borrelia burgdorferi (TOBB) in Lyme borreliosis patients from midwestern Germany. Wien. Klin. Wochenschr. 110: Huwer, M., A. Schwarzmaier, H. D. Hamel, and R. Will The occurrence of B. divergens in the Freiburg i. Br. District and piroplasmosis prevention trials in cattle. Berl. Munch. Tierarztl. Wochenschr. 107: Kampen, H., E. Maltezos, M. Pagonaki, K.-P. Hunfeld, W. A. Maier, and H. M. Seitz Individual cases of autochthonous malaria in Evros Province, northern Greece: serological aspects. Parasitol. Res. 88: Kjemtrup, A. M., and P. A. Conrad Human babesiosis: an emerging tick-borne disease. Int. J. Parasitol. 30: Krause, P. J., and S. R. Telford 3rd Babesiosis, p In H. M. Gilles (ed.), Protozoal diseases. Arnold, London, United Kingdom. 17. Krause, P. J., S. R. Telford 3rd, R. Ryan, P. A. Conrad, M. Wilson, J. W. Thomford, and A. Spielman Diagnosis of babesiosis: evaluation of a serologic test for the detection of Babesia microti antibody. J. Infect. Dis. 169: Krause, P. J., S. R. Telford 3rd, R. J. Pollack, R. Ryan, P. Brassard, L. Zemel, and A. Spielman Babesiosis: an underdiagnosed disease of children. Pediatrics 89: Krause, P. J., S. R. Telford 3rd, A. Spielman, V. Sikand, R. Ryan, D. Christianson, G. Burke, P. Brassard, R. Pollack, J. Peck, and D. H. Persing Concurrent Lyme disease and babesiosis: evidence for increased severity and duration of illness. JAMA 275: Levine, N. D The protozoan phylum Apicomplexa, vol. 2. CRC Press, Boca Raton, Fla. 21. Mehlhorn, H., W. Raether, E. Schein, M. Weber, and M. Uphoff Lichtund elektronenmikroskopische Untersuchung der intraerythrozytären Stadien von Babesia microti. Dtsch. Tierärztl. Wochenschr. 93: Metz, W Einheimische und importierte zeckenübertragene Infektionen bei Hunden in Deutschland, p Inauguraldissertation, Tierärztliche Hochschule, Hannover, Germany. 23. Piesman, J., S. J. Karakashian, S. Lewengrub, M. A. Rudzinska, and A. Spielman Development of Babesia microti sporozoites in adult Ixodes dammini. Int. J. Parasitol. 16: Popovsky, M. A Transfusion-transmitted babesiosis. Transfusion 31: Scharan, K. P., and P. J. Krause Babesiosis, p In B. A. Cunha (ed.), Tick-borne infectious diseases: diagnosis and management. Marcel Decker, Inc., New York, N.Y. 26. Schouls, L. M., I. van de Pol, S. G. T. Rijpkema, and C. S. Schot Detection and identification of Ehrlichia, Borrelia burgdorferi sensu lato, and Bartonella species in Dutch Ixodes ricinus ticks. J. Clin. Microbiol. 37: Skotarczak, B., and A. Cichocka Isolation and amplification by polymerase chain reaction DNA of Babesia microti and Babesia divergens in ticks in Poland. Ann. Agric. Environ. Med. 8: Tenter, A. M., and K. T. Friedhoff Serodiagnosis of experimental and natural Babesia equi and B. caballi infections. Vet. Parasitol. 20: Thomford, J. W., P. A. Conrad, S. R. Telford 3rd, D. Mathiesen, B. H. Bowman, A. Spielman, M. L. Eberhard, B. L. Herwaldt, R. E. Quick, and D. H. Persing Cultivation and phylogenetic characterization of a newly recognized human pathogenic protozoan. J. Infect. Dis. 169: Uhnoo, I., O. Cars, D. Christensson, and C. Nystroem-Rosander First documented case of human babesiosis in Sweden. Scand. J. Infect. Dis. 24: Walter, G Zur Übertragung und zum Parasitämieverlauf von Babesia microti (Stamm Hannover ) bei Rötelmaus (Clethrionomys glareolus) und Erdmaus (Microtus agrestis). Acta Trop. 41: Wilske, B., L. Zöller, V. Brade, H. Eiffert, U. B. Göbel, G. Stanek, and H. W. Pfister Lyme Borreliose, p In H. Mauch and R. Lütticken (ed.), Qualitätsstandards in der mikrobiologisch-infektiologischen Diagnostik. Urban & Fischer Verlag, Munich, Germany. 33. World Health Organization Veterinary Public Health Unit WHO Workshop on Lyme Borreliosis Diagnosis and Surveillance. Document no. WHO/CDS/VPH/ World Health Organization, Geneva, Switzerland.