Human ehrlichiosis is a tickborne infectious disease caused by Ehrlichia sp. that has primarily been detected in the United States. Common clinical manifestations of human ehrlichiosis are fever, headache, myalgia, and malaise. Leukopenia and thrombocytopenia often occur. Symptoms range from mild fever to severe illness with multiple organ dysfunction, which is occasionally fatal (1). In a retrospective analysis, we show serologic evidence for human ehrlichiosis in 3 febrile patients in Japan.

Figure

Figure. Western blots using serum samples from a febrile patient (case 1) in Wakayama Prefecture in study showing serologic evidence of human exposure to ehrlichiosis agents in Japan. Serum samples were...

In case 1, a male patient, who was 48 years of age and worked in the manufacturing industry, sought care at a primary care clinic in 2015 for high fever (>40°C) and headache ≈1 month after hiking in the mountains. The clinic physician prescribed levofloxacin and acetaminophen, but the treatment was not effective. Therefore, the patient was seen at the Japanese Red Cross Wakayama Medical Center. The day before onset of high fever, the patient found a small rash on the left side of his abdomen. This date was considered day 0, although there might have been symptoms that the patient was unaware of before that time. The rash was an erythema migrans–like lesion that expanded on day 5. The patient was hospitalized, and borreliosis or tick-associated rash illness, which is similar to Lyme borreliosis–like erythema migrans, was suspected (2); however, a tick bite or eschar was not observed. After intravenous administration of minocycline (200 mg/d), the patient’s fever abated, but the lesion expanded and was accompanied by puritis. On day 10, the patient was discharged from the hospital, after which the rash gradually disappeared. Diagnostic tests for borreliosis were negative. We retrospectively performed immunofluorescence assays (IFAs) and Western blot (Appendix) using patient serum samples collected on days 2 and 17. We showed seroconversion to antibodies against Ehrlichia chaffeensis antigens by IFA and the presence of IgM and IgG against Ehrlichia sp. P28 protein by Western blot (Table; Figure). We suspected the patient had ehrlichiosis and tick-associated rash illness.

In case 2, a male patient, who was 66 years of age and worked as a truck driver, sought care at the Ise Red Cross Hospital in 2018 for fever (38°C), annular erythema, and malaise. The patient had renal impairment and jaundice. The principal physician suspected leptospirosis, but diagnostic tests for leptospirosis were negative. The physician suspected other bacterial infections, including Japanese spotted fever (JSF) or anaplasmosis. The patient was treated intravenously with minocycline (200 mg/d) and sulbactam/ampicillin (6 g/d) for 4 days. Subsequently, amoxicillin (1.5 g/d) was administered orally for 14 days, and the patient recovered. Diagnostic tests for JSF were negative. We retrospectively analyzed patient serum samples collected on days 14, 32, and 60 after onset of illness. We showed seroconversion to antibodies against E. chaffeensis antigens by IFA and the presence of IgM and IgG against Ehrlichia sp. P28 protein by Western blot (Table; Appendix Figure 1). The IFA titers for both IgM and IgG decreased on day 60.

In case 3, a female patient, who was 69 years of age and owned a Japanese-style accommodation, sought care at the Ise Red Cross Hospital in 2018 for mild fever, generalized edema and rash, headache, and malaise. The principal physician suspected JSF and treated the patient with oral minocycline (200 mg/d) and levofloxacin (500 mg/d) for 10 days; the patient recovered. Diagnostic tests for JSF were negative. We retrospectively analyzed patient serum samples collected on days 5, 58, and 115 by IFA and Western blot and found seroconversion to antibodies against E. chaffeensis antigens by IFA and the presence of both IgM and IgG against Ehrlichia sp. P28 protein antigens by Western blot (Table; Appendix Figure 2). In this case, the IgM titer increased in the convalescent-phase serum on day 58 but decreased on day 115. However, the IgG titer increased on days 58 and 115 after onset of illness. In addition, we detected antibodies against Anaplasma phagocytophilum by IFA and A. phagocytophilum–specific P44 surface antigen by Western blot. We detected only IgG antibodies against A. phagocytophilum in all 3 serum samples, suggesting a past infection with A. phagocytophilum.

The 3 patients lived on the Kii peninsula of Japan (Appendix Figure 3), which is known to be a JSF-endemic area, especially in Wakayama and Mie Prefectures (3,4). In addition, anaplasmosis exists in those areas (5). Previously, we revealed the presence of ticks infected with A. phagocytophilum and Ehrlichia sp. that could potentially infect humans in Mie prefecture (6,7). In particular, members of the Ehrlichia sp. genotype 2 group, including Ehrlichia sp. MieHl92 and MieHl94, were considered candidate organisms that might cause human ehrlichiosis in Japan (6).

In conclusion, we provide serologic evidence of autochthonous cases of human ehrlichiosis in Japan. We recommend that ehrlichiosis should be considered as a clinical cause of febrile illness in this country.

Top