As Difficult as it Gets: Challenges in Managing Early Lyme Disease With Anaplasmosis Co-Infection

Abstract. Lyme disease (LD), the predominant vector-borne illness in the United States, affects approximately 476,000 individuals annually, according to the CDC. As its geographic footprint expands, healthcare professionals increasingly encounter challenges in diagnosis and treatment, particularly in the early stage of the disease. Despite being a nationally notifiable disease since 1991, more effective management strategies are still needed. In this clinical case, we highlight some challenges and offer practical insights for the enhanced management of early LD with anaplasmosis co-infection.

Introduction. Lyme disease is the predominant vector-borne ailment in the United States, afflicting approximately 476,000 individuals annually. Notably, 95% of LD cases reported to the CDC emanate from 14 states spanning the northeastern, mid-Atlantic, and north-central regions of the country. These states include Connecticut, Delaware, Maine, Maryland, Massachusetts, Minnesota, New Hampshire, New Jersey, New York, Pennsylvania, Rhode Island, Vermont, Virginia, and Wisconsin. The causative agent of this infection is primarily Borrelia burgdorferi (Bb), and its prevalence peaks during late spring, summer, and early fall. The black-legged deer tick, identified as Ixodes scapularis or Ixodes pacificus, serves as the common vector.

Detection and removal of adult female ticks, which are large and reddish, help prevent transmission. Infected questing nymphs pose a higher risk of infection than adult female ticks. Additionally, Ixodes ticks transmitting B. burgdorferi can harbor six other disease-causing infectious organisms in North America, with Anaplasma phagocytophilum (Ap) and Babesia microti being the most identified. Anaplasmosis, also known as human granulocytic anaplasmosis (HGA), is a bacterial infection caused by Ap, carried by the black-legged tick (Ixodes scapularis) in the Northeast and Midwest, and the western black-legged tick (Ixodes pacificus) along the West Coast.

Tickborne rickettsial diseases (TBRD) such as anaplasmosis can lead to severe illnesses and even fatalities in individuals who are otherwise in good health, despite the widespread availability of effective and affordable treatment options. The primary obstacle lies in promptly diagnosing these diseases and administering appropriate treatment. Given the escalating prevalence of LD and other tick-borne illnesses, coupled with the potential for significant long-term health repercussions, healthcare providers need better guidelines for diagnosis and treatment. Existing clinical reference tools often lack the necessary detail or efficiency to address this growing need, leaving primary care providers grappling with the challenges. This paper aims to spotlight some of the obstacles and important considerations in the management of early LD, particularly when co-infection with anaplasmosis is suspected.

Case Description. In late July, a 45-year-old man from rural central Pennsylvania presented to the clinic with a fever that had begun a week earlier, reaching 102.5°F. His symptoms included a headache, muscle aches, joint pain, and general malaise, but he did not experience typical upper respiratory or gastrointestinal symptoms. The fever had subsided 3 days before his visit. He had no significant past medical history, worked in healthcare, had no recent sick contacts, and had not traveled recently. He lived in a heavily wooded 5-acre property with his wife, three children, and a dog, and reported a minor tick bite a week before his symptoms started. At the visit, he was afebrile, and the physical examination was normal. COVID-19 testing was negative, and Lyme and anaplasma serology tests were ordered. He declined doxycycline treatment initially. Four days after presentation, Lyme serology tests revealed positive IgM and negative IgG results, and he reported a recurrence of fever, headache, fatigue, and body aches the previous night. He was then prescribed doxycycline 100 mg twice daily for 10 days, which resolved his symptoms after three doses. Less than a week later, anaplasma serology results showed a positive IgM titer of >1:256 and an IgG titer of 1:640, suggesting a recent or past infection. Despite completing the doxycycline course, his symptoms returned 24 hours later. He was then given doxycycline 100 mg twice daily for an additional 21 days, which resolved his symptoms within 24 hours. At his follow-up visit, he reported no recurrence of symptoms and declined further testing.

Discussion. The intricacies revealed in this seemingly straightforward case underscore several significant challenges encountered in clinical practice. Through this case, we aim to shed light on these complexities.

Second course of doxycycline: what is the optimal duration of treatment? Although there are uncertainties about the choice of an antibiotic and treatment duration for many presentations of LD, the 2020 Infectious Diseases Society of America-American Academy of Neurology-American College of Rheumatology (IDSA-AAN-ACR) Guidelines for erythema migrans (EM) recommend treating the patient with a 10-day course of Doxycycline.¹ Disease severity at onset predicts the risk for late-stage manifestations, and therefore, guides treatment duration. As disease progresses beyond EM lesion, treatment must be extended. doxycycline for 14 to 21 days is recommended for early or early disseminated phase patients who do not have neurological involvement.^1,2

For anaplasmosis treatment, the 2006 IDSA guideline recommends doxycycline 100 mg orally two times a day for 10 days in adults for both anaplasmosis alone or with LD co-infection.³

Our patient exhibited a prompt response to the initial doxycycline treatment, but all symptoms recurred within 24 hours after he finished 10 days of treatment. A possible explanation is that our patient presented 1 week after the onset of symptoms, which may have led to then further dissemination of LD or perhaps co-infections may have altered the disease severity and the effectiveness of the treatment. In a case when LD and anaplasmosis co-infection is suspected, especially in a case with late presentation, a longer doxycycline treatment course might be prudent. Patients with early disseminated LD can present with constitutional symptoms such as fever, headache, and arthralgia. However, our patient did not exhibit signs of other characteristic LD symptoms, such as cranial neuritis, radiculoneuritis, meningitis, carditis, or acute arthritis. It was challenging to discern or test for these symptoms, especially when the patient was experiencing a flu-like syndrome with fever, headache, arthralgia, and myalgia. Consequently, we opted for a 21-day course of Doxycycline treatment instead of a 14-day course following the recurrence of illness.

Lyme disease, anaplasmosis, or co-infection: challenges in diagnosis. In an endemic area, co-infection with anaplasmosis and babesiosis should be considered in patients who develop fever after an Ixodes tick bite, especially in patients who have high-grade fever for greater than 48 hours.³ The CDC’s general reference standard for diagnosis of TBRD such as anaplasmosis is the indirect immunofluorescence antibody (IFA) testing with paired serum samples 2 to 4 weeks apart. A greater than or eqal to four-fold rise in antibody titer is considered confirmatory evidence of acute infection.⁴ A “probable case” is defined as a patient with a compatible illness with a single positive IFA titer or the visualization of morulae in leukocytes.⁵

For LD, the officially recommended test for diagnosis is an indirect serology testing with standard two‑tiered testing and modified two‑tiered testing using an enzyme-linked immunosorbent assay and a Western Blot to measure the presence of anti-Bb antibodies. Deciphering serology test results involves navigating complexities that arise from temporal dynamics and potential antibody persistence. The test has major issues with sensitivity and specificity, leading to false-positive or false-negative results. Sensitivity is compromised in the early LD due to factors such as the “window period” (latency time of the immune response after infection, cross-reactivity of antibodies from common antigens of other diseases, or insufficient antibody production due to co-infections.) Sensitivity is below 50% in early LD, and this highlights a potential limitation in capturing early infections through serological assessments. Serology tests lack the ability to distinguish a past but cleared infection from an active one as both IgM and IgG antibody responses can persist for years or even decades after the infection has been eradicated. IgM antibody usually wanes and may become undetectable in late active disease, but the IgG antibody persists. The IgM antibody response may also persist for years, so the presence of IgM antibodies does not necessarily indicate an infection of short duration or early-stage infection. Therefore, the serology is unsuitable for monitoring response to treatment or identifying new infections.^6,7

A positive Lyme serology with IgM alone or IgG plus IgM may not conclusively indicate an acute LD infection as both LD and Anaplasmosis can manifest similar flu-like symptoms. Determining the active status of LD in the context of co-infection poses challenges. Fortunately, doxycycline is the drug of choice for both diseases.

To maximize specificity, IgM reactivity alone should not be used to diagnose LD if signs or symptoms of early LD have been present longer than 1 month (the “1-month rule”). But such patients may have only a positive IgM without a fully evolved IgG antibody response even after 1 month of illness.⁸

According to the CDC's Lyme disease pretest probability algorithm, testing is unlikely to provide useful insights when the pretest probability is low. Serological testing alone is not sufficient to make nor exclude a LD diagnosis and test results must always be correlated with the clinical presentation. Therefore, the test should not be used in patients with non-specific symptoms and a low pre-test probability.⁹

In general, the CDC guideline recommends complete blood cell count (CBC), comprehensive metabolic panel, and peripheral blood smear when considering a diagnosis of TBRD. For anaplasmosis diagnosis, the CDC recommends lab testing if a patient lives or visited an area with anaplasmosis, the illness occurs during spring, summer, or fall when tick activity is high, and the patient remembers a tick bite. The optimal diagnostic test depends on the timing of symptom onset. A PCR test is most sensitive in the first week of illness and within 48 hours of starting doxycycline treatment. A negative PCR result does not rule out the disease and treatment. Antibody serology test is often negative in the first week of illness. IgG is more important and paired acute and convalescent samples collected 2 to 4 weeks apart demonstrating a ≥ four-fold increase in titer is the best evidence of recent infection.⁴ A positive IgM alone does not prove acute infection and is likely less specific than IgG antibody. A blood-smear microscopy during the first week of illness might reveal morulae (microcolonies of anaplasmae) in granulocytes, which is highly suggestive of a diagnosis. Some characteristic laboratory findings can include thrombocytopenia, leukopenia, and elevated hepatic transaminase, increased immature neutrophils, and mild anemia. Response to treatment offers clues on anaplasmosis as well.⁴

Looking at our case, following the CDC's pretest probability algorithm, our patient resides in an endemic area, has experienced tick exposure, and exhibited symptoms typical of LD (acute flu-like syndrome). Hence, the pretest probability was moderate-to-high, justifying the Lyme test, with a positive result aligning with LD. If the patient had remained symptomatic during his visit, a comprehensive workup, including PCR, serology, blood smear, blood cell counts, complete metabolic panel, and tests for potential co-infections, would have been appropriate. However, our patient presented 1 week after symptom onset with symptoms resolving for 3 days (and 5 asymptomatic days before symptom recurrence). Therefore, we opted for focused LD and anaplasmosis serology tests, as these are the top two tick-borne illnesses in Pennsylvania. Additionally, we offered our patient Doxycycline. The patient's clinical symptoms, delayed medical attention, positive serology, and swift response to doxycycline support a diagnosis of LD with anaplasmosis co-infection.⁴ Declining further testing, the patient could only be classified as a probable case of anaplasmosis.⁵

This case underscores the ease of overlooking or delaying the diagnosis of LD with or without anaplasmosis or another TBRD co-infection, emphasizing the importance of maintaining suspicion and familiarity with appropriate testing and treatment protocols. If there had been no response within 48 hours of Doxycycline treatment, reconsideration of anaplasmosis diagnosis and testing for B. microti co-infection would be warranted. Additionally, had the patient remained asymptomatic post-visit, treatment would not have been recommended for anaplasmosis even with confirmed anaplasmosis serology.³

Flu-like symptoms: challenges in LD staging. Often referred to as "the great imitator," LD presents a diagnostic challenge due to its diverse and overlapping symptoms, mimicking various other medical conditions. Early LD symptoms include fever, chills, headache, fatigue, neck stiffness, myalgia, joint pain, and swollen lymph nodes. The non-specific nature of these signs and symptoms, coupled with their broad differentials, makes early LD diagnosis challenging.

Generally, LD is classically divided into three stages: early localized, early disseminated, and late disseminated.¹⁰ While most patients exhibit the early localized disease marked by the presence of an EM lesion, up to 30% of individuals either do not develop an EM lesion or its presence is overlooked.⁹ Both early localized disease of EM and early-disseminated LD may have concomitant constitutional symptoms.¹¹ Failure to recognize EM or flu-like presentations can lead to missed diagnosis and delayed antibiotic treatment, and an increased risk of late manifestations.⁹

Notably, clinical guidelines and common clinical reference tools may disproportionately emphasize EM as a hallmark of early LD, potentially overlooking patients who present solely with flu-like symptoms without EM or other LD indicators. In the case of using doxycycline for LD treatment, the limitations of such tools become evident, emphasizing the need for a more comprehensive approach that considers the diverse presentations of LD beyond the conventional emphasis on EM.

Testing for LD and anaplasmosis: considerations for co-infections. Determining the appropriate timing and scope of testing for co-infections in patients with LD poses a critical clinical challenge. In endemic geographic regions, early LD can have co-infection with anaplasmosis or B. microti or both. Patients with LD who have a high fever > 48 hours should be tested for co-infection with anaplasmosis and/or B. microti infection.³ And, when patients with LD experience persistent fever for > 1 day while on Doxycycline, testing for B. microti is recommended.¹ Suspecting Anaplasmosis in patients with LD becomes pertinent when characteristic laboratory abnormalities are evident, which may include thrombocytopenia, leukopenia, neutropenia, and/or anemia. Additionally, signs of hemolysis, such as elevated indirect bilirubin levels, anemia, and increased lactate dehydrogenase, strongly suggest babesiosis.¹

There are practical challenges to implementing these recommendations. For example, laboratory turnaround time is long; routine labs are typically not ordered for people with flu-like symptoms; the patient’s symptoms can be severe and disturbing; and there is danger for disease progression. Therefore, it might be more practical to lower the threshold for co-infection testing at the first patient encounter. It should be noted that currently, there is no definitive test of cure for LD. When symptoms recur shortly after competing treatment, a longer course of treatment should be considered.

Prompt initiation in suspicion of LD and Anaplasmosis. As mentioned above, one barrier to care is the laboratory turnaround time, which is generally long for LD and even longer for anaplasmosis (more than 3 days for LD and 8 days for anaplasmosis in our case.) In the context of extended laboratory turnaround times, the pivotal question arises: should antibiotic treatment, such as doxycycline, be initiated while awaiting laboratory results?

Recognizing that antibiotic treatment significantly shortens the duration of the disease and prevents the progression to late-stage LD, it is imperative to commence antibiotics promptly when LD is suspected. Delaying treatment to await test results may risk allowing the disease to advance.¹²

Anaplasmosis typically presents with fever, malaise, muscle pain, headache, and gastrointestinal issues. However, severe cases can escalate to life-threatening conditions such as sepsis, renal failure, rhabdomyolysis, and multiorgan failure, which necessitates immediate treatment even before confirmatory PCR testing is completed.¹³ The emphasis on initiating empirical treatment promptly aims to prevent disease progression and avoid serious complications. According to CDC guidelines, early empiric antibiotic therapy in suspected TBRD can mitigate severe outcomes or fatalities, with Doxycycline being the preferred treatment for both adults and children.⁴ In cases of mild anaplasmosis or when doxycycline is contraindicated, such as in patients with a severe allergy or during pregnancy, rifampin may be used instead. Given the risk of complications and the typical presentation of unexplained fever, chills, and headache, all symptomatic individuals suspected of having anaplasmosis should receive antimicrobial therapy.³

Use of guidelines and clinical reference tools. In our bustling clinical settings, we frequently rely on various clinical reference tools to assist us in making medical decisions. However, the original published clinical guidelines often prove inefficient due to their extensive length, complexity, and limited accessibility. Within these guidelines, the ranking of recommendation strength may still leave significant decision-making responsibilities to individual healthcare providers. For instance, the 2020 IDSA-AAN-ACR Guidelines, tailored for physicians treating LD, including primary care providers, only covers limited aspects of LD management. While commonly used, highly efficient clinical reference tools can sometimes oversimplify disease management and overlook crucial details that we covered above.

The tick (if caught): consideration on testing. Frequently encountered in clinical practice, patients often present with ticks they've removed, accompanied by the question: “Should I send the tick out for testing?” The tick removed from patients can be sent for species identification. However, sending Ixodes ticks for Bb testing is not recommended as the result does not reliably predict clinical infection. Apart from species identification, there is no recommended role for testing ticks for infection,¹¹ and testing black-legged ticks for human pathogens could yield misleading results.¹⁰