Disclaimer

Our medical policies are designed for informational purposes only and are not an authorization, or an explanation of benefits, or a contract.  Receipt of benefits is subject to satisfaction of all terms and conditions of the coverage.  Medical technology is constantly changing, and we reserve the right to review and update our policies periodically.

Description

Lyme disease (LD) is a multisystem inflammatory disease caused by the spirochete Borreliaburgdorferi and transmitted by the bite of an infected ixodid tick endemic to Northeastern, North Central, and Pacific coastal regions of the United States. The disease is characterized by stages, beginning with localized infection of the skin (erythemamigrans), followed by dissemination to many sites. Manifestations of early disseminated disease may include lymphocytic meningitis, facial palsy, painful radiculoneuritis, atrioventricular nodal block, or migratory musculoskeletal pain. Months to years later, the disease may be manifested by intermittent oligoarthritis, particularly involving the knee joint, chronic encephalopathy, spinal pain, or distal paresthesias. While most manifestations of LD can be adequately treated with oral antibiotics, intravenous (IV) antibiotics are indicated in some patients with neurologic involvement or atrioventricular heart block. However, overdiagnosis and overtreatment of LD are common due to its nonspecific symptoms, a lack of standardization of serologic tests, and difficulties in interpreting serologic test results. In particular, patients with chronic fatigue syndrome or fibromyalgia are commonly misdiagnosed as possibly having LD and undergo inappropriate IV antibiotic therapy. The purpose of this policy is to provide diagnostic criteria for the appropriate use of IV antibiotic therapy. The following paragraphs describe the various manifestations of LD that may prompt therapy with IV antibiotics and the various laboratory tests that are used to support the diagnosis of LD.

Neurologic Manifestations of Lyme Disease (Neuroborreliosis) 

Lymphocytic meningitis, characterized by head and neck pain, may occur during the acute disseminated stage of the disease. Analysis of the cerebrospinal fluid (CSF) is indispensable for the diagnosis of Lyme meningitis. If the patient has LD, the CSF will show a lymphocytic pleocytosis(lymphocyte count greater than normal) with increased levels of protein. Intrathecal production of antibodies directed at spirochetal antigens is typically present. A normal CSF analysis is strong evidence against Lyme meningitis. Treatment with a 2- to 4-week course of IV antibiotics, typically ceftriaxone or cefotaxime, is recommended.

Cranial neuritis, most frequently Bell’s palsy, may present early in the course of disseminated LD, occasionally prior to the development of antibodies, such that an LD etiology may be difficult to rule in or out. While Bell’s palsy typically resolves spontaneously with or without treatment with oral antibiotics, some physicians have recommended a lumbar puncture and a course of IV antibiotics if pleocytosis in the CSF is identified, primarily as a prophylactic measure to prevent further neurologic symptoms.

A subacute encephalopathy may occur months to years after disease onset, characterized by subtle disturbances in memory, mood, sleep, or cognition accompanied by fatigue. These symptoms may occur in the absence of abnormalities in the electroencephalogram (EEG), magnetic resonance imaging (MRI), or CSF. In addition, the symptoms are nonspecific and overlap with fibromyalgia and chronic fatigue syndrome. Thus diagnosis of Lyme encephalopathy may be difficult and may be best diagnosed with a mental status exam or neuropsychological testing. However, treatment with IV antibiotics is generally not indicated unless CSF abnormalities are identified.

Much rarer, but of greater concern, is the development of encephalomyelitis, characterized by spastic paraparesis, ataxias, cognitive impairment, bladder dysfunction, and cranial neuropathy. CSF examination reveals a pleocytosis and an elevation in protein. Selective synthesis of anti-spirochetal antigens can also be identified. A course of IV antibiotics with 3 to 4 weeks of ceftriaxone is suggested when CSF abnormalities are identified.

A variety of peripheral nervous system manifestations of LD have also been identified. Symptoms of peripheral neuropathy include paresthesias, or radicular pain with only minimal sensory signs. Patients typically exhibit electromyographic (EMG) or nerve conduction velocity abnormalities. CSF abnormalities are usually seen only in those patients with a coexistent encephalopathy.

Cardiac Manifestations of Lyme Disease 

Lyme carditis may appear during the early dissemination stage of the disease; symptoms include atrioventricular heart block, tachyarrhythmias, and myopericarditis. Antibiotics are typically given, although no evidence proves that this therapy hastens the resolution of symptoms. Both oral and IV regimens have been advocated. Intravenous regimens are typically used in patients with a high-degree atrioventricular block or a PR interval on the electrocardiogram of greater than 0.3 second. Patients with milder forms of carditis may be treated with oral antibiotics.

Lyme Arthritis 

Lyme arthritis is a late manifestation of infection and is characterized by an elevated IgG response to B. burgdorferi and intermittent attacks of oligoarticular arthritis, primarily in the large joints such as the knee . Patients with Lyme arthritis may be successfully treated with a 30-day course of oral doxycycline or amoxicillin, but care must be taken to exclude simultaneous central nervous system (CNS) involvement, requiring IV antibiotic treatment. In the small subset of patients that do not respond to oral antibiotics, an additional 30-day course of oral or IV antibiotics may be recommended.

Fibromyalgia and Chronic Fatigue Syndrome 

Fibromyalgia pain and chronic fatigue syndrome are the diseases most commonly confused with LD. Fibromyalgia is characterized by musculoskeletal complaints, multiple trigger points, difficulty in sleeping, generalized fatigue, headache, or neck pain. The joint pain associated with fibromyalgia is typically diffuse, in contrast to Lyme arthritis, which is characterized by marked joint swelling in one or a few joints at a time, with few systemic symptoms. Chronic fatigue syndrome is characterized by multiple subjective complaints, such as overwhelming fatigue, difficulty in concentration, and diffuse muscle and joint pain. In contrast to LD, both of the above conditions lack joint inflammation, have normal neurological test results, or have test results suggesting anxiety or depression. Neither fibromyalgia nor chronic fatigue syndrome has been shown to respond to antibiotic therapy.

Serologic Tests 

The antibody response to infection with B. burgdorferi follows a typical pattern. During the first few weeks after the initial onset of infection, there is no antibody production. The specific IgM response characteristic of acute infection peaks between the third and sixth week. The specific IgG response develops only after months and includes antibodies to a variety of spirochetal antigens. IgG antibodies produced in response to LD may persist for months or years. Thus detection of IgG antibodies only indicates exposure, either past or present. In LD endemic areas, underlying asymptomatic seropositivity may range up to 5%–10%. Thus, as with any laboratory test, interpretation of serologic tests requires close correlation with the patients’ signs and symptoms. For example, patients with vague symptoms of LD, chronic fatigue syndrome, or fibromyalgia may undergo multiple serologic tests over many weeks to months in an effort to establish the diagnosis of LD. Inevitably, in this setting of repeat testing, one enzyme-linked immunosorbent assay (ELISA) or test, whether IgG or IgM, may be reported as weakly positive or indeterminate. These results most likely represent false positive test results in the uninfected patient who has had long-standing symptoms from a different condition and previously negative test results.

Currently, the Centers for Disease Control and Prevention (CDC) recommend a 2-step method for the serologic diagnosis of LD:

1. Enzyme-Linked Immunosorbent Assay (ELISA) for: Borreliaburgdorferi Antibodies 

This test is a screening serologic test for LD. ELISA tests are available to detect IgM or IgG antibodies or to detect both antibody types together. More recently developed tests using recombinant or synthetic antigens have improved diagnostic sensitivity. For example, the FDA-approved C6 ELISA is highly sensitive to infection, and is under study as an indicator of antibiotic therapy efficacy. A positive or indeterminate ELISA test result alone is inadequate serologic evidence of LD. All of these tests must be confirmed with an immunoblot test. In addition, results must be correlated with the clinical picture.

2. (Western) Immunoblot 

This test is used to confirm the serologic diagnosis of LD in patients with positive or indeterminate ELISA tests. In contrast to the standard ELISA test, the immunoblot investigates the specific antibody response to the different antigens of B. burgdorferi Typically, several clinically significant antigens are tested. According to CDC criteria, the test result is considered positive if 2 of the 3 most common IgM antibody bands to spirochetal antigens are present, or 5 of the 10 most frequent IgG antibody bands are present. Because the CDC criteria were developed for surveillance, they are conservative and may miss true LD cases. Some support the use of more liberal criteria for a positive result in clinical diagnosis; however, alternative criteria have not been well validated. Criteria for interpreting immunoblot results are different in Europe than in the United States due to differences in prevalent Borellia species causing disease.

Other tests include:

Polymerase Chain Reaction (PCR) 

In contrast to the above 2 serologic tests, which only indirectly assess prior or present exposure to B. burgdorferi, PCR directly tests for the presence of the spirochete. Because PCR technology involves amplification of DNA from a portion of B. burgdorferi, there is a high risk of exogenous contamination, resulting in false positive results. Positive results in the absence of clear clinical indicators or positive serology are not definitive for diagnosis. In addition, the test cannot distinguish between live spirochetes or fragments of dead ones. The PCR technique has been studied using a variety of specimens. PCR has the best detection rates for skin biopsies from patients with erythema migrans and for synovial tissue (and synovial fluid, to a lesser extent) from patients with Lyme arthritis. CSF may be positive by PCR during the first two weeks of infection, but thereafter the detection rate is low. PCR is not recommended for urine or blood specimens.

Borrelia PCR also provides information on which of the 3 major species pathogenic for humans has been found in the specimen tested (genotyping).

T-Cell Proliferative Assay 

T-lymphocyte proliferation assays are not recommended as diagnostic tests; they are difficult to perform and standardize, and their sensitivity is not well characterized.

Evaluation of Cerebrospinal Fluid (CSF) 

Aside from the standard evaluation of CSF for pleocytosis, protein levels, and glucose levels, various tests are available to determine whether anti- B. burgdorferi antibodies are being selectively produced within the central nervous system. Techniques include a variety of immunoassays. For example, intrathecal antibody production can be detected by the CSF/serum index of B. burgdorferi antibodies. CSF and serum samples diluted to match the total IgG concentration in CSF are run in parallel in an IgG ELISA. Excess : Borrelia -specific antibody in CSF indicates a positive result. As noted, PCR can also be used to detect the spirochete in the CSF, most successfully within the first 2 weeks of infection.

Treatment of Lyme Disease 

As noted above, treatment with IV antibiotics is generally indicated only in patients with symptoms and laboratory findings consistent with CNS or peripheral neurologic involvement, and in a small subset of patients with heart block or documented Lyme arthritis who have not responded to oral antibiotics. Typical IV therapy consists of a 2- to 4-week course of ceftriaxone or cefotaxime, both third-generation cephalosporins, or penicillin or chloramphenicol. No data suggest that prolonged or repeated courses of IV antibiotics are effective. Lack of effect should suggest an incorrect diagnosis or slow resolution of symptoms, which is commonly seen in LD. In addition, some symptoms may persist after treatment, such as Lyme arthritis; this phenomenon may be related to various self-sustaining inflammatory mechanisms rather than persistent infection.

---

Policy

Treatment of Lyme Disease (LD) consists of oral antibiotics, except for the following indications: 

I.A 2- to 4-week course of IV antibiotic therapy may be considered medically necessary in patients with neuroborreliosis with objective neurologic complications of documented LD (see the following for methods of documentation). 

Objective neurologic findings include: 

 

Lyme disease may be documented either on the basis of serologic testing or by clinical findings of erythema migrans in early infection. Documentation of CSF abnormalities is required for suspected CNS infection, as indicated above.

Serologic documentation of infection requires:

 

1. Positive or indeterminate ELISA test, AND

2. Positive immunoblot blot by CDC criteria.

Documented CSF abnormalities include ALL of the following:

• Pleocytosis;
• Evidence of intrathecal production of B. burgdorferi antibodies in CSF; and
• Increased protein levels.

Polymerase chain reaction (PCR)-based direct detection of B. burgdorferi in CSF samples may be considered medically necessary and may replace serologic documentation of infection in patients with a short duration of neurologic symptoms (less than 14 days) during the window between exposure and production of detectable antibodies.

II. A 2- to 4-week course of IV antibiotics may be considered medically necessary in patients with Lyme carditis, as evidenced by positive serologic findings (defined above) and associated with a high degree of atrioventricular block or a PR interval of greater than 0.3 second. Documentation of Lyme carditis may include PCR-based direct detection of B. burgdorferi in the blood when results of serologic studies are equivocal. 

III. A single 2- to 4-week course of IV antibiotic therapy may be considered medically necessary in the small subset of patients with well-documented Lyme arthritis who have such severe arthritis that it requires the rapid response associated with IV antibiotics. Documentation of Lyme arthritis may include PCR-based direct detection of B. burgdorferi in the synovial tissue or fluid when results of serologic studies are equivocal. 

Intravenous antibiotic therapy is considered not medically necessary in the following situations: 

• Patients with symptoms consistent with chronic fatigue syndrome or fibromyalgia;
• Patients with seronegative LD in the absence of CSF antibodies;
• Initial therapy in patients with Lyme arthritis without coexisting neurologic symptoms;
• Cranial nerve palsy (e.g. Bell’s palsy) without clinical evidence of meningitis;
• Antibiotic-refractory Lyme arthritis (unresponsive to 2 courses of oral antibiotics or to 1 course of oral and 1 course of intravenous antibiotic therapy);
• Patients with vague systemic symptoms without supporting serologic or CSF studies;
• Patients with a positive ELISA test, unconfirmed by an immunoblot or Western blot test (see definition above);
• Patients with an isolated positive serologic test in the setting of multiple negative serologic studies;

 

IV. Repeat or prolonged courses (greater than 4 weeks) of intravenous antibiotic therapy are considered not medically necessary.

V. Repeat PCR-based direct detection of B. burgdorferi is considered investigational in the following situations:

• as a justification for continuation of IV antibiotics beyond 1 month in patients with persistent symptoms
• as a technique to follow therapeutic response

VI. PCR-based direct detection of B. burgdorferi in urine samples is considered investigational in all clinical situations.

VII. Genotyping or phenotyping of B. burgdorferi is considered investigational.

Policy Guidelines

Three different CPT codes describe direct detection of B. burgdorferi: direct probe (87475), amplified probe technique (87476), and quantification (87477). When these codes were introduced in 1998, for the sake of consistency, the same grouping of 3 codes was used for a wide variety of different organisms. However, only the amplified probe technique (87476) is used clinically for the detection of B. burgdorferi. The direct probe technique (87475) is not clinically useful due to the small numbers of organisms present. The quantification technique (87477) has no clinical role at this time since treatment decisions are not based on the quantification of organisms present. Therefore, codes 87475 and 87477 would be considered investigational.

Benefit Application

The diagnostic performance of PCR-based technology to identify B. burgdorferi is subject to false positive results unless strict measures are instituted to prevent cross contamination between samples. As laboratory performance is variable, plans may want to assess the diagnostic performance of local laboratories.

Rationale

Direct Detection of B. burgdorferi with Polymerase Chain Reaction (PCR) Technology
While diagnosis of Lyme disease is generally based on the clinical picture and demonstration of specific antibodies, PCR-based technology can detect the spirochete in the CSF in cases of neuroborreliosis, in the synovial fluid of cases of Lyme arthritis, and rarely in skin biopsy specimens of those with atypical dermatologic manifestations. (1) However, a PCR-based test is generally considered a second tier test, performed only when the results of serologic tests and clinical evaluation are equivocal. For example, while PCR-based tests can identify organisms in skin biopsy specimens of patients with dermatologic manifestations (i.e., erythema migrans), this diagnosis is typically made clinically and antibiotic therapy started empirically. A skin biopsy is rarely necessary. Similarly, diagnosis of Lyme arthritis is based on clinical and serologic studies without the need for synovial tissue or fluid. Finally, intrathecal antibody production is considered a more sensitive test than PCR-based CSF detection in patients with suspected neuroborreliosis, but a PCR-based technique may be useful in patients with a short duration of disease (i.e., less than 14 days) during the window between exposure and the emergence of detectable levels of antibodies in the CSF. (2) However, it should be noted that that the test cannot distinguish between live and dead organisms. PCR-based detection is typically not performed in the urine due to the variable presence of endogenous polymerase inhibitors that have an impact on the test's sensitivity.

PCR-based technology has been used as one step in the genotypic analysis of B. burgdorferi. B. burgdorferi was originally characterized as a single species (B. burgdorferi sensu lato), but genotypic analysis has revealed that this group represents 3 distinct species and genomic groups. Of these, the following have been isolated from patients with Lyme disease: B. burgdorferi sensu stricto, B. garinii, and B. afzelii. The prevalence of these different genospecies may vary among populations and may be associated with different clinical manifestations. (3) However, no data were found in the published literature regarding whether or how knowledge of the genotype or phenotype of B. burgdorferi could be used to improve patient management and outcomes. In the U.S., B. burgdorferi sensu stricto is the only human pathogenic species, but in Europe all 3 species cause infection. Recently a new human pathogenic species, B. spielmanii, was found in a small number of European patients; therefore, criteria for interpreting immunoblot results are different in Europe than in the U.S. (4)

Role of Intravenous (IV) Antibiotics

A diagnosis of Lyme disease (LD) requires appropriate epidemiologic data, supporting clinical observation (including exposure to ixodid ticks in an endemic area), and supporting laboratory findings. However, overdiagnosis and overtreatment of LD is common (5-7). Intravenous antibiotic therapy in patients with presumed LD would be inappropriate in the following situations: an incorrect diagnosis; a history of prolonged or repeated courses of IV antibiotics; and use of IV antibiotics when oral antibiotics are adequate. An incorrect diagnosis of LD includes those patients with positive serologies without characteristic signs or symptoms of LD, or those with non-specific symptoms and no known exposure to ticks in an endemic area, or those without supporting serologic evidence.

In 1993 the American College of Rheumatology published a position paper on IV antibiotic treatment for LD, which concluded that “empiric treatment of patients with nonspecific chronic fatigue or myalgia on the basis of positive serologic results alone will result in many more instances of antibiotic toxicity than cures of atypically symptomatic true Lyme disease...In patients whose only evidence for Lyme disease is a positive immunologic test, the risks for empiric IV antibiotic treatment outweigh the benefits….” (7) Other studies have also supported the use of oral, not IV, antibiotics in patients with LD without neurologic involvement. (8-10)

Practice guidelines regarding the treatment of Lyme disease and including discussion of supportive evidence have been issued by the Infectious Diseases Society of America (IDSA). (11) The IDSA also endorsed the American Academy of Neurology evidence-based practice parameter for the treatment of nervous system Lyme disease. (12) The IDSA guidelines recommend IV antibiotics only in the following situations (note: none of the recommendations suggest longer than a 1-month course of IV antibiotics):

 

• Early neurologic disease
  • Meningitis or radiculopathy 14–28 days
• Cardiac disease
  •  Acute onset of varying degrees of intermittent atrioventricular heart block, sometimes in association with clinical evidence of myopericarditis: 14–21 days
• Late disease
  • Persistent or recurrent arthritis after initial oral regimen: 14–28 days (a second, 4-week oral regimen may also be used)
  • CNS or peripheral nervous system disease: 14–28 days

In the particular case of cranial nerve palsy associated with LD (most commonly Bell’s palsy, also known as 7th nerve palsy) and without clinical evidence of meningitis, the evidence indicates that oral antibiotic therapy is satisfactory. (11,12) Cranial nerve palsy may, in fact, resolve without treatment, but treatment should be administered to avoid late complications of LD.

In addition, guidelines recommend symptomatic treatment for symptoms that persist after appropriate antibiotic therapy. For example, in a small number of patients with known prior infection, arthritis may persist despite negative B. burgdorferi DNA by PCR in synovial fluid or tissue. Such persistent arthritis is termed “antibiotic-refractory Lyme arthritis,” defined as “persistent synovitis for at least 2 months after completion of a course of intravenous ceftriaxone (or after completion of two 4-week courses of an oral antibiotic for patients unable to tolerate cephalosporins), in conjunction with negative results of PCR.” (11) Symptomatic treatment, rather than additional antibiotic treatment, is recommended. (11)

The evidence generally does not support persistent B. burgdorferi infection in patients with well-documented infection who have received recommended antibiotic therapy. (12) Blinded, randomized controlled trials of extended antibiotic therapy versus placebo in such patients have shown no differences in outcomes (summarized in the Table). Therefore, prolonged courses of antibiotic therapy, which may be associated with adverse events, are not recommended.

2009 Update

A literature update performed through May 2009 found no new evidence that would affect the policy statements, which remain unchanged.

Table.
Summary of randomized, controlled trials of prolonged antibiotic therapy in patients with well-documented, previously treated Lyme disease.

Study	N	Patient description	Experimental treatment	Control treatment	Results
Klempner et al. 2001 (13)	78	1) Positive for IgG Abs to B. burgdorferi; persistent symptoms that interfered with patient function 2) Negative for IgG Abs to B. burgdorferi; else, as above	IV ceftriaxone daily for 30 days, oral doxycycline for 60 days	IV and oral placebos	No significant difference in QoL outcomes for 1) and 2). Studies terminated after interim analysis indicated that it was highly unlikely that a significant difference in treatment efficacy would be observed.
Kaplan et al. 2003 (14)	129	Same trial as Klempner et al. 2001 (13)			Both treatment and control arms showed similar and not significantly different decreases in Medical Outcomes Study cognitive, pain, and role functioning scales; and improved mood as assessed with the Beck Depression Inventory and Minnesota Multiphasic Personality Inventory.
Krupp et al. 2003 (15)	55	Patients with persistent severe fatigue of duration 6 months or longer	IV ceftriaxone daily for 28 days	IV placebo	Ceftriaxone treatment arm showed no significant improvement in primary outcome of laboratory measure of persistent infection. Significant improvement in the secondary outcome of disabling fatigue; no significant treatment effect on cognitive function; no difference in change in SF-36 scores.

 

Oksi et al 2007 (16)	152	Consecutive patients treated with standard antibiotic regimen for 21 days	Amoxicillin twice daily for 100 days starting immediately after standard regimen	Placebo twice daily for 100 days starting immediately after standard regimen	Patients in ceftriaxone group were significantly more likely to correctly identify their treatment assignment. Both treatment and control arms showed similar and not significantly different decreases in patient and investigator VAS outcomes (VAS evaluation of symptoms, range 0-100, 0=no symptoms) at 12 mos. B. burgdorferi-specific antibodies declined similarly in both groups over 12 mos.
Fallon et al. 2008 (17)	37	Patients with documented objective memory impairment	IV ceftriaxone daily for 70 days	IV placebo daily for 70 days	Primary outcome of cognitive function across 6 domains was similarly improved in both groups at week 24, and was not significantly different between groups; improvement between groups was marginally significantly different at week 12 (p=0.05). Exploratory subgroup analyses suggested significantly better improvement in ceftriaxone-treated patients with more severe baseline pain and physical functioning.
Cameron 2008 (18)	86	Patients with symptoms of arthralgia, cardiac of neurologic involvement with or without fatigue after previous successful antibiotic treatment of LD; study conducted in a primary care internal medicine practice	Oral amoxicillian 3 gm daily for 3 months (34 assigne, 17 evaluable)	Oral placebo daily for 3 months	--44% of enrolled patients not evaluable at 6 months; 17 of these had poorer baseline quality of life and were lost due to treatment failure --SF-36 improvements for antibiotic vs. placebo arm were significant (46% vs. 18%, p = 0.007) but text not clear if analysis of all or only evaluable patients; --SF-36 physical component improvement not significantly different between treatment arms for evaluable patients (8.5 vs 7); --SF-36 mental component significantly improved in antibiotic arm for evaluable patients (14.4 vs. 6.2, p = 0.04)

References:

 

 

Index

Policy History

Date	Action	Reason
01/30/98	Add to Prescription Drug section	New policy
10/08/02	Replace policy	Policy reviewed with literature search; policy statement unchanged regarding IV antibiotic therapy; additional policy statements on PCR technique, updated rationale and references
12/17/03	Replace policy	Policy updated with literature search; policy statement unchanged
04/1/05	Replace policy	Policy updated with literature search; policy statement unchanged
12/14/05	Replace policy – coding update only	CPT coding updated
4/25/06	Replace policy – error correction only	Corrected last sentence in Policy Guidelines to delete 87476 and read: “Therefore, codes 87475 and 87477 would be considered investigational.”
06/14/07	Replace policy	Policy updated with literature search. Policy statements updated; uncomplicated cranial nerve palsy (e.g., Bell’s palsy) not considered a medically necessary indication for intravenous antibiotics. Reference numbers 18-20 added.
07/10/08	Replace policy	Policy updated with literature search. Rationale and references reorganized and revised extensively. Policy statements unchanged
07/09/09	Replace policy	Policy updated with literature search; no new references added; minor wording changes to text. Policy statements unchanged