Background

Suffolk County, a suburban county on Long Island (LI), New York, has a population of 1.8 million people and annually reports the highest absolute number of tick-borne diseases in NY. In 2017, there were 523 cases of Lyme disease, 55 cases of Ehrlichiosis, 31 cases of Anaplasmosis, and 138 cases of Babesiosis [1]. A series of cases in the Northeast US described a new Borrelia species, Borrelia miyamotoi, first reported to cause human infections in 2013 [2]. B. miyamotoi is closely related to the relapsing fever family of Borrelia spp (e.g. B. hermsii); however, it is transmitted by Ixodes scapularis, the same tick that transmits Borrelia burgdorferi (Lyme disease), Anaplasma phagocytophilum and Babesia microti [3]. The geographic expansion of B. miyamotoi in the US extends to other Lyme endemic areas such as the Midwest of North America [4]. B. miyamotoi infection can clinically present during warm months as a flu-like syndrome similar to Lyme disease, Anaplasmosis, or Babesiosis. The diagnosis of B. miyamotoi infection is complicated by the overlap in clinical manifestations caused by other TBD and the need to order specific diagnostic tests that may not be familiar to general practitioners even in hyperendemic areas for TBD such as LI [5]. In addition, little is known about co-infections between B. miyamotoi and the other TBD. Although 3–5% of ticks on LI have been found to be infected by B. miyamotoi and up to 74% have B. burgdorferi [6], there is no evidence that co-infections have occurred in humans who have been diagnosed with early Lyme disease in New York [7]. The goal of this study was to describe the recent epidemiology of B. miyamotoi by performing a retrospective chart review in all patients diagnosed with B. miyamotoi infection in Stony Brook Medicine (SBM) system, Suffolk County, NY.

Methods

A retrospective study was conducted at Stony Brook Medicine (SBM) Hospitals (Stony Brook University Hospital–SBUH- and Southampton Hospital-SHH) between January 1, 2013 and December 31, 2017. SBM is the only tertiary medical center in Suffolk County, NY. The case search was performed from only positive tests results from the laboratory database. A positive result for B. miyamotoi was determined by either a positive real-time qPCR in the blood or IgG antibody detected by EIA using B. miyamotoi glycerophosphodiester phosphodiesterase recombinant antigen (rGlpQ) using previously described assays [8, 9] (performed at Oxford Immunotec, Norwood, MA).

Results

At SHH, a total of 8575 PCR tests were performed for both B. burgdorferi and for B. miyamotoi, with positive results for 19 (0.2%) and 17 (0.19%) of these organisms, respectively. Other TBD results were; Babesia PCR in blood was 1.5% (263/17,626), A. phagocytophilum was 0.4% (80/17501), and Ehrlichia chaffeensis was 1% (172/16955). At SBUH, less than 200 PCR and IgG EIA tests for B. miyamotoi were performed during the study period. All PCR tests were negative at SBUH. For IgG EIA, 8 were positive at SHH (total tested = 38) and 11 were positive at SBUH (total tested = 60). A total of 28 cases were positive for either IgG EIA (n = 19) or PCR (n = 9) (8 other PCR-positive cases were not included because clinical information was not available).

All 9 PCR-positive cases (median age: 67; range: 22–90 years) had clinical findings suggestive of acute or relapsing infection. Of these 9 cases, 8 were men (88%), 3 were diagnosed in the outpatient clinic (33.3%), while the remaining 6 (66.6%) were diagnosed through the emergency room and required hospitalization. Demographics, clinical manifestations, and laboratory results on patients who had a positive blood B. miyamotoi PCR test are described in Table 1. None of these 9 cases had evidence for active co-infections with other TBD (all had negative blood PCR results for Ehrlichia, Anaplasma, Babesia, and Borrelia burgdorferi). One case had a positive IgM for Lyme with band 41 present in immunoblot but no erythema migrans, and another case had a positive IgG for B. burdogferi with 10 bands in the immunoblot.

Table 1 Demographics, clinical manifestations and laboratory results on patients with Borrelia miyamotoi PCR positive in the blood
Full size table

Discussion and conclusions

The positivity rate of B. miyamotoi PCR in this area of NY in a 5-year study period is 0.19% (17/8575). We were able to review clinical records for 9 of these PCR positive cases. B. miyamotoi PCR was ordered in these patients at SHH was because they had clinical manifestations compatible with a TBD in the summer and this PCR test is part of a TBD panel offered by a commercial laboratory. In contrast, at SBUH, diagnostic tests for B. miyamotoi (PCR and/or EIA) were rarely performed, most likely because the commercial TBD panel was not included in the routine test catalog; most of these tests were ordered as part of Infectious Disease service consultations, based on specific clinical findings.

It is possible that most of the patients in this study were thought to have Lyme disease after a deer tick bite, prompting the healthcare provider to order the TBD panel which includes a B. miyamotoi PCR test. One B. miyamotoi PCR positive case had a positive IgM for Lyme with the band 41 positive, which is not a serological criterion for acute Lyme. The presentation of this case was not consistent with acute Lyme, but with the other tick-borne disease such as Ehrlichia, Anaplasma or B. miyamotoi infection (leukopenia, transaminitis, thrombocytopenia). The true incidence of B. miyamotoi on LI is unknown; however, when compared to other TBD tested in the same healthcare system, the percentage of positive tests for B. miyamotoi was similar to that of other TBD present in this region. Lack of knowledge about laboratory tests for B. miyamotoi may hinder its detection. Whereas most clinicians are aware of the two-tiered algorithm for Lyme serology (despite its relatively poor sensitivity) [10], many clinicians are not aware that PCR performed on blood is the optimal test for detecting acute B. miyamotoi infection or that serologic diagnosis requires the use of rGlpQ-based EIA (the GlpQ protein is present in the relapsing fever group of Borrelia spp, but is not found in B. burgdorferi). B. miyamotoi infection often produces a positive result on a whole-cell lysate Lyme EIA or the Lyme C6 peptide antibody test [11, 12], but in most cases the Lyme immunoblot is negative, a result that is also common in early Lyme disease. Based on the partial cross-reactivity between B. burgdorferi and B. miyamotoi, a negative result on a Lyme serology assay does not rule out B. miyamotoi infection.

Limitations of this study include the retrospective nature and lack of convalescent titers in some patients. However, this study may serve first as a baseline to consider B. miyamotoi infection in some flu-like syndromes during the summer after a deer tick bite and second, to avoid misidentifying a case as early Lyme disease when in fact it could be a B. miyamotoi infection whose long-term symptoms may be similar to post-Lyme disease syndrome (PLDS) [13]. Whether some of the PLDS cases reported in clinical practice may have been related to B. miyamotoi infections or other TBD remains an open question.