Case Report: Anaplasmosis in Canada: Locally Acquired Anaplasma phagocytophilum Infection in Alberta

INTRODUCTION

Anaplasma phagocytophilum is an obligate intra‐granulocytic parasite that is predominately transmitted by the ticks Ixodes scapularis and Ixodes pacificus in North America. We report on the second laboratory‐confirmed case of A. phagocytophilum acquired within the province of Alberta, Canada. The first confirmed case was detected in 2009.¹

CASE REPORT

The patient was a 67-year-old woman who lives on a rural property close to a provincial park in the Edmonton health zone. On day zero (May 2017), after spending time outdoors gardening and cleaning her dog’s kennel, a tick check revealed a partially engorged tick attached on her left upper arm. A circular, non-painful, non-pruritic erythematous patch, approximately 2 cm in diameter, surrounded the attached tick. On failing to remove the entire tick at home, she went to the local emergency department (ED) to have the tick removed via forceps. She did not receive antibiotics.

By day 2, the erythematous patch had resolved. On day 6, the patient developed progressive fatigue, night sweats, and myalgia. Episodes of alternating, stabbing temporal headaches ensued for the next 7 days, motivating her to return to the ED on day 13. In the ED, she was found to be febrile at 38°C and thrombocytopenic (89 × 10⁹/L). She was discharged from the ED, without receiving antibiotics, with a provisional diagnosis of fever of unknown origin. However, the ongoing severity of her symptoms, notably drenching night sweats, prompted her to return to her family physician on day 16, and she was referred back to the ED. A repeat complete blood count showed a mild anemia (112 g/L), elevated white blood cells (13.2 × 10⁹/L) with neutrophilia (11.5 × 10⁹/L), and lymphopenia (0.9 × 10⁹/L). Her thrombocytopenia had resolved. Her physical examination was unremarkable except for tachycardia and an irregular pulse. An electrocardiogram demonstrated new onset atrial fibrillation, and she was admitted to hospital for further management. Based on the history of tick exposure and her clinical presentation, Lyme disease and Rocky Mountain spotted fever (RMSF) were suspected. Baseline serologies for these two agents were collected, and she was started empirically on doxycycline. Her night sweats abated within 2 days of the start of antibiotics, and over the next 2 months, she was incrementally able to return to her daily routine without experiencing periods of tiredness or arrhythmia.

The attached tick was identified as a partially engorged female I. scapularis, and it was positive for A. phagocytophilum and Borrelia burgdorferi DNA by real-time PCR testing performed at the Public Health Agency of Canada, Winnipeg, Canada, as part of a national surveillance program. The tick was negative, by PCR, for Babesia microti and Borrelia miyamotoi. Serology testing, using an indirect immunofluorescent assay for A. phagocytophilum IgG antibodies, was performed retrospectively on blood samples collected at days 13, 31, and 52. A greater than 4‐fold increase in A. phagocytophilum IgG titers, from less than 1:64 on day 13 to 1:2,048 on days 31 and 52, was observed. Serologies for RMSF and Lyme disease (C6 antigen) were both negative on day 13.

On day 31, follow-up Lyme disease serology was positive by enzyme immunoassay. Confirmatory immunoblot testing demonstrated one positive band on IgM immunoblot (p41) and two bands positive on IgG immunoblot (p41 and p28), consistent with a negative result. Repeat Lyme disease serology on day 52 was negative.

DISCUSSION

A number of clinical and diagnostic challenges contributed toward the delay in the recognition of this patient’s infection. First, this is only the second laboratory‐confirmed case of A. phagocytophilum identified in Alberta since 2009 and, although it is a recognized tick‐borne infection, the degree of clinical suspicion and awareness is significantly lower than that for other tick-borne infections such as B. burgdorferi and Rickettsia rickettsii.¹ Second, populations of the competent tick vectors for this agent, I. scapularis and possibly I. pacificus, are not yet established in Alberta,^2,3 which further reduces the perception of an infection due to A. phagocytophilum. In Alberta, passive tick surveillance for blacklegged ticks and their associated pathogens occurs for ticks removed from companion animals and humans. These ticks are presumably introduced annually into the province by migratory birds, given their low population numbers that, so far, have not been increasing and the limited variety of tick life stages (especially nymphs) present in the environment.^2,3 The program which focuses on companion animals has been in place since 2007, but ticks are screened only for B. burgdorferi. However, blacklegged ticks removed from humans have been routinely tested for A. phagocytophilum, and before 2013, 3–6% of these ticks collected were infected.^4,5 From 2013 to 2019, 8% of blacklegged ticks (six of 75) collected from people in Alberta were infected, and one or two A. phagocytophilum–infected ticks were collected in most of these years (LRL, unpublished data). Thus, there is a very low, but persistent, risk of human exposure to “bird-borne” A. phagocytophilum–infected blacklegged ticks in Alberta.

Third, initial manifestations of A. phagocytophilum infection are often mild and nonspecific, such as fever, chills, myalgias, and headaches, leading to difficulties in diagnosis. In addition, the characteristic erythema migrans rash, a frequently noted feature of B. burgdorferi, or the maculopapular rash typical for R. rickettsii infection, is absent in A. phagocytophilum infections. The lack of a rash may reduce a clinician’s concern of a tick-borne disease, especially in an area where previous cases of A. phagocytophilum have not been described or are very rare. Antibiotic stewardship policies promote an observational approach in areas where tick‐borne infections are not endemic, as most signs and symptoms following a tick exposure will be a consequence of reactions to the tick saliva and will subside either without treatment or with use of over-the-counter antihistamines. As in this case, early removal of the tick may also provide false reassurance that the risk is low. Delayed transmission occurs for both B. burgdorferi and A. phagocytophilum; however, transmission of A. phagocytophilum has been reported, in animal models, to occur more rapidly than that of the Lyme disease agent.⁶

Fourth, misdiagnosis of A. phagocytophilum infection can occur because of the delay of antibody appearance to A. phagocytophilum, the persistent presence of antibodies in patients with prior infection, or cross-reaction with similar organisms. While antibody response to A. phagocytophilum generally occurs 7–10 days after the onset of symptoms, outliers do occur.⁷ In addition, low antibody titers can persist for years following infection.⁸ Therefore, repeat serological testing is warranted 2–4 weeks after initial testing to assess for seroconversion or the change in antibody titers in patients who have initially tested positive. A persistently low titer is suggestive of remote infection, whereas titers that have changed more than 4-fold are suggestive of an acute infection. In our case, the patient was serologically negative at day 13, but on day 31, both A. phagocytophilum serology and B. burgdorferi C6 serology became positive. Confirmatory B. burgdorferi testing with Western blot was negative, and therefore, the B. burgdorferi serology was interpreted as negative. A follow-up C6 Lyme antibody test was negative at day 52, whereas A. phagocytophilum remained positive with high titers. Whether the positive C6 Lyme antibody test represents cross-reaction with A. phagocytophilum or an antibody response that was subsequently blunted by doxycycline remains unknown. However, it is speculated that A. phagocytophilum may potentially cross-react with B. burgdorferi by ELISAs and demonstrate positive bands on B. burgdorferi IgM and IgG immunoblot.⁹ In our case, B. burgdorferi IgM and IgG immunoblot was positive for p41, corresponding to a flagella antigen from B. burgdorferi. In one study, 80% of patients with confirmed A. phagocytophilum infection had a positive p41 band on B. burgdorferi IgM immunoblot, which suggests that antibodies to p41 is potentially present in both A. phagocytophilum and B. burgdorferi infections.⁹ Cross-reaction of B. burgdorferi serological tests from A. phagocytophilum infections may cause misdiagnosis, particularly because B. burgdorferi is much more commonly tested for than A. phagocytophilum. In this case, microscopic examination of peripheral blood films and/or molecular testing of whole blood may have helped provide an answer, and this testing is offered in Canada. However, because of the low index of suspicion, this testing was not requested.

At the present time, human granulocytic anaplasmosis (HGA) is not nationally notifiable in Canada; hence, it is difficult to gauge the extent and frequency of human infections occurring across the country. A study examining A. phagocytophilum seropositivity among Ontario residents, with no recent travel history, who were tested for suspected HGA was 10.8%.¹⁰ However, it is unknown how many of these patients were infected during remote travel to endemic areas outside Ontario. The outcomes of passive tick surveillance in Alberta highlight that a small percentage (< 10%) of blacklegged ticks detected in the province from companion animals or people without history of travel are infected with the agent of HGA.^4,5 As mentioned, these ticks were presumably dispersed in Alberta by migratory birds.^2,3 In our case, the patient lives in an area close to a provincial park, which is along a major migratory bird flyway. Hence, we hypothesize that the infected tick was delivered to this location by a migratory bird and the patient was the unfortunate recipient of the infected tick bite.

It remains to be determined whether blacklegged tick populations will become established in Alberta, but it would seem prudent to maintain ongoing tick surveillance programs to monitor for this possibility. In addition, physicians in Alberta should be mindful that there is a low, but persistent, risk of locally acquired cases of HGA, despite the apparent lack of resident vector populations.