A 62-year-old physician living in Huntington, Long Island, New York, USA, was in good health until June 9, 2013, when he felt unwell and fever, chills, headache, myalgias, fatigue, sweats, joint pain, poor appetite, and conjunctivitis developed. On the third day of illness, he noted dark urine that lasted for several days. On June 13, he was seen by his family physician, who noted fever but no other abnormality.

A complete blood count (CBC) showed a hemoglobin level of 13.9 g/dL (reference range 13 g/dL‒18 g/dL); a hematocrit of 40.8% (40%‒54%); a leukocyte count of 4,700 cells/μL (4,500 cells/μL‒11,000 cells/μL) with 55% neutrophils (54%‒62%) and 28% lymphocytes (25%‒33%); a platelet count of 68,000 cells/μL (150,000 cells/μL‒400,00 cells/μL); an aspartate aminotransferase level of 65 U/L (1 U/L‒36 U/L); and an alanine aminotransferase level of ALT 70 U/L (1 U/L‒45 U/L). He was told to return for reevaluation if symptoms did not resolve and was seen again on June 16. At that time, he had a fever (temperature 103°F). He was given doxycycline (100 mg orally every 12 h) but did not improve.

Laboratory results on June 16 showed that a B. microti IgM result was strongly positive (IgM titer >1:1,024, IgG titer <1:16), but Anaplasma phagocytophilum antibody was absent (Table). He was then given atovaquone (750 mg, 2×/d for 2 wks) and azithromycin (500 mg, 1×/d for 2 wks). He recovered completely 3 weeks after symptoms began.

Subsequent attempts to perform Babesia whole-genome sequencing on a residual blood sample obtained 3 days after the start of treatment identified B. microti DNA, but it was insufficient to perform full-genomic sequencing. At a follow-up visit to his physician on July 5, a CBC and tests for aspartate and alanine aminotransferase levels showed results within references ranges.

Three years later, on June 19, 2016, the patient had a fever (temperature 100°F), chills, sweats, headache, myalgias, anorexia, and difficulty concentrating develop. He also noted dark urine for several days. On June 22, he was seen by his physician, who obtained a CBC, which showed a hemoglobin level of 12.6 g/dL and hematocrit of 38.3%; a leukocyte count of 4,700 cells/μL with 55% neutrophils and 30% lymphocytes; and a platelet count of 41,000 cells/μL. A blood smear showed a Babesia parasitemia level of 1%. We amplified B. microti DNA by using PCR. Results for B. microti antibody were positive (IgM titer 1:256, IgG titer >1;1,024) (Table). He was then given atovaquone (750 mg every 12 h) and azithromycin (500 mg on day 1 and then 250 mg 1×/d) for 10 days, at which time symptoms had resolved. A repeat blood smear did not show any parasites. The patient has subsequently been in good health.

This patient had 2 separate episodes of B. microti babesiosis 3 years apart. He lived in an area where B. microti was hyperendemic and showed typical symptoms of Babesia infection during each episode, including dark urine that is indicative of hemoglobinuria (8). In the first episode, he did not have a blood smear or PCR performed, but a high B. microti IgM response was suggestive of B. microti infection (1,9). B. microti infection was subsequently confirmed by identification of B. microti DNA. In the second episode, B. microti infection was confirmed by blood smear and PCR.

Both clinical and laboratory evidence support reinfection rather than relapse of infection for this patient. He was repeatedly exposed to ticks in an area where babesiosis is commonly reported (1). He was in good general health without evidence of immunosuppression, whereas all cases of relapsing babesiosis have been reported in immunocompromised persons. After the first episode of babesiosis, he had complete clinical recovery 2 weeks after the onset of infection and did not experience the second episode until 3 years later. In contrast, those persons who have had relapsed B. microti infection have all been immunocompromised, experienced relapses of infection 2 weeks to 3 months after the previous episode, and usually lack full clinical recovery between relapses (3–7). Finally, our patient had a robust B. microti IgM response 2 weeks after the onset of his first infection and an anamnestic antibody response with a high IgG titer on day 4 of the second infection, which is characteristic of reinfection rather than relapse. Patients who have experienced relapse have conditions that impair antibody response (e.g., B cell lymphoma, rituximab therapy, HIV/AIDS). A minimal or absent B. microti antibody response has been demonstrated in patients who have had relapsing babesiosis and have been tested for B. microti antibody (5,7).

Previous studies describe the persistence of human B. microti infection and clinical immunity. In a prospective study of babesiosis patients who were tested for B. microti DNA by PCR every 3 months after acute illness until infection cleared, parasitemia persisted up to 13 months in 22 antimicrobial drug‒treated patients and up to 27 months in 23 untreated patients (2). In another study, a patient was reported as having relapsing infection that persisted for 27 months (5). The immediate host response to B. microti infection is provided by innate immune elements that include the spleen, macrophages, and neutrophils. In contrast, long-term clearance of B. microti parasites depends in large part on antibody (2–7). Studies of the duration of B. microti antibody have demonstrated persistence for as little as 6 months and as long as 6 years (2,6,10–11).

Although there is strong evidence that our patient experienced reinfection, we do not have absolute confirmation, and it is possible that he could have had persistent B. microti infection that relapsed after 3 years. No B. microti PCR (or blood smear) was obtained after the initial infection. We attempted to further distinguish between relapse and reinfection by genetic sequencing of B. microti DNA from blood samples obtained from both episodes of infection. Unfortunately, we were unable to obtain sufficient DNA from the first episode for sequencing because a blood sample was only available 3 days after antimicrobial drug therapy was initiated, leaving few viable parasites.

In summary, our study shows evidence of reinfection after successful treatment of a B. microti infection. Although the evidence is highly supportive, it is not definitive. Whether our patient experienced reinfection or relapse 3 years after the initial infection, investigation of similar patients could provide useful information about the immune response to B. microti infection. Patients who have experienced babesiosis, and their healthcare professionals, need to be aware that babesiosis reinfection might occur, as for Lyme disease (12,13). Tickborne disease preventive measures should be maintained for patients with or without a history of babesiosis (14).