Lyme Disease Executive Summary


Widespread variation in testing, diagnosis and treatment still exists despite current literature-based evidence.

Scope and focus of guideline

To aid the clinician to appropriately diagnose and treat Lyme disease. The guideline offers the practitioner solid evidence to explain the rationale for treatment plans to patients.

Target audience

Pediatric and adult primary care practitioners.

Key clinical points

Key clinical strategies

Level of evidence

Expert opinion and consensus.

Committee members

Guideline initially authored by Allen Smiley, MD, William Teubl, MD, Richard Brown, MD, Anna Simpson, MD, John Donhowe, MD, Jean Collins, Betty Diorio, Carol DeLaMarter and revised by Alan Gross, MD, Jeffrey Stein, MD, Paul Lemanski, MD, Jeffrey Palmer, MD, Bonnie Richardson, MD, Steven Luger, MD (consultant).
Date: 3/98

Guideline For Diagnosis Of Lyme Disease


Annotations for Lyme Diagnostic Guideline

  1. Any set of signs or symptoms the practitioner feels is consistent with Lyme such as rash, viral syndrome, arthritis, arthralgias, myalgias, neuropathy, etc. Symptoms presenting in isolation need a careful consideration of the differential diagnosis.
  2. Symptoms suggesting need for immediate treatment include: (1) an Erythema Chronicum Migrans (ECM) rash, (2) viral syndrome after a tick bite (especially between June and September), (3) Bells Palsy in an endemic area, or (4) a tick bite in a pregnant woman. There is no benefit to testing in this setting. An ECM rash is defined as a slowly enlarging rash that develops 3 to 30 days after a tick bite. Circling the rash with a pen mark may be helpful. A viral syndrome is defined as flushing, documented increased temperature, chills, etc. with arthralgias or myalgias in the absence of gastroenteritis or a URI.
  3. Current recommendations for treatment are:

    All regimes are given for a total of three weeks. Summer viral syndrome may be Ehrlichiosis and practitioners may choose to use Doxycycline 100 mg BID for three weeks with sunscreen as first line treatment in this setting for either adults or children. At least two deaths have been reported from Ehrlichiosis treated with Amoxicillin alone.

  4. The knee is the most common site, but any large joint arthritis needs to include Lyme in a differential diagnosis.
  5. Arthrocentesis may be indicated to rule out a septic joint.
  6. An inflammatory effusion is an effusion with a white blood cell count of >2000 that had a sterile culture.
  7. Differential diagnosis would include either a septic joint or some other joint problem that has a low grade inflammatory component (e.g., mechanical).
  8. Bells Palsy could be a presentation of Lyme disease even in nonendemic areas. However many cases of Bells Palsy under these circumstances will not be due to Lyme. Hence serologic testing is recommended to help sort out etiology. There is no literature to support approaching endemic and nonendemic Bells Palsy differently; this recommendation is based on physician consensus. If Lyme serology is positive, a lumbar puncture should be performed.
  9. See testing guideline for further details.
  10. The literature uses 48 hours fairly consistently as a cutoff when evaluating the likelihood of infection being transmitted. If no bite was ever noted by the patient, the answer to this question is no and the user should proceed to box 13. If a bite is remembered by the patient but the duration is unknown, the practitioner will likely want to assume it had been on for > 48 hours, especially if the bite occurred in an endemic area.
  11. Arthralgias, myalgias, headaches or fatigue are recognized as common minor symptoms that raise the question of Lyme disease. These symptoms are termed minor not because of low morbidity but because they are not highly predictive of Lyme disease.
  12. A tick bite can be observed for one month for an ECM rash or a viral syndrome. The development of symptoms would then allow for reentry into the guideline for reevaluation.
  13. Same as #11.
  14. Endemic areas in the East include Fire Island, Block Island, Cape Cod, Nantucket, Martha's Vineyard, the Jersey Shore to Maryland, and parts of Connecticut, Massachusetts, New York State, Rhode Island, and Maine. The Pacific coast and upper midwest states are also considered endemic. Practitioners should check with local public health departments for more current information in their areas.
  15. See testing guideline for further details.
  16. This point in the guideline represents the differential diagnosis of a patient with non-specific symptoms that caused the consideration of Lyme disease but without the following: ECM, viral syndrome, Bells Palsy, large joint effusion, tick bite > 48 hours, arthralgias, myalgias or headache. The likelihood of Lyme in this setting is very low, and the chief complaint should be scrutinized and the differential diagnosis generated around that complaint.
  17. Either treating or testing is justifiable at this juncture. If empiric treatment is chosen, only oral therapy is warranted. If the practitioner feels IV treatment needs consideration, testing and/or referral is strongly encouraged. If empiric testing is chosen, follow-up Lyme testing is of unclear benefit, so clinical endpoints should be defined by the practitioner in advance of treatment. Lastly, the literature suggests that the placebo response to antibiotics in the setting of possible Lyme disease is as high as 35%.



Annotations for Lyme Testing and Treatment Guideline

  1. The Elisa test currently being used has a sensitivity of 94% and a specificity of 97%. Lyme titers may be falsely positive in patients with mononucleosis, periodontal disease, connective tissue disease and other less common conditions.
  2. Positive Elisa? No comment.
  3. It is estimated that the likelihood of Lyme disease at this point is about 1%. The main differential to consider is fibromyalgia. A referral to rheumatology or neurology is considered reasonable at this point. Frankly asking the patient if he or she is willing to accept a diagnosis other than Lyme disease may be of value. This will help identify patients who are convinced they have Lyme disease. Discussing why a patient feels that way and their fears might be more appropriate than focusing on the technical aspects of the disease alone.
  4. If a non-Lyme diagnosis reasonable, treating the patient for the diagnosis is the obvious next step. If treatment has failed, the practitioner could re-enter the guideline at this point and proceed to box 5.
  5. An early test could be falsely negative. Although literature is limited, 6 weeks is a rational interval for retesting to allow for seroconversion.
  6. Positive Elisa? No comment.
  7. At this point the likelihood of Lyme is quite low. However, the limitations of testing and the high anxiety of some patients will lead some practitioners to opt to treat. Empiric treatment is considered far more reasonable than recurrent testing because of the increasing risk for a false positive result (especially high at this point because of the low pretest probability). Should the practitioner opt for treatment, the patient should understand the risks (mainly reaction to antibiotics) and benefits (hard to define; placebo effects may be as high as 35%). Empiric treatment should consist of a single course of oral medication. If symptoms persist after treatment, consultation should be considered.
  8. The Western blot is very specific if a positive test is defined as 5 out of 10 reactive IgG bands. This step will eliminate most false positives. The IgM Western Blot is in a state of flux; review by a national consensus panel is needed before making any recommendations concerning interpretation. Practitioners are encouraged to rely on IgG values and to seek phone
    consultation with Dr. Allen Smiley, rheumatologist with expertise in
    Lyme Disease in the Hudson Valley Region (Phone # 914/471-2800 or
    Fax # 914/471-2847).
  9. Positive Western Blot? No comment.
  10. Lyme disease can cause heart block of any grade. The treatment of 3rd degree AV block requires IV antibiotics and cardiac consultation to evaluate the need for temporary pacing. Lower degrees of block can be treated with an oral course of medication along with careful follow-up.
  11. See #11.
  12. The key issue at this point is to determine whether to treat the patient with PO or IV medication. It is generally agreed among our expert consultants, and in the most recent literature, that CNS Lyme and Lyme arthritis that has failed PO therapy are indications for IV treatment. Lyme arthritis is relatively easy to identify, but most Lyme arthritis will respond to PO medication so failures will be rare.
    The major dilemma is determining the presence or absence of CNS disease. The presence of the following in CSF would be adequate to establish the diagnosis of CNS Lyme:

1. Pleocytosis plus or minus increased protein
2. Positive CNS Elisa or PCR to Lyme

It is recognized that many practitioners do not perform spinal taps and that spinal taps can be falsely negative. Furthermore, memory loss and fatigue, possible manifestations of CNS disease, are nonspecific and difficult to objectify. A guideline to help the practitioner at this juncture is too
complex and uncertain.
Should practitioners opt to treat, it should be remembered that follow-up testing after treatment with antibiotics is generally uninterpretable. Careful attention to the clinical symptoms marked for improvement with treatment is important and will likely be the only basis for determining efficacy.

  1. The most recent practice among centers treating Lyme disease is to use four weeks of Ceftriaxone instead of two weeks (2 gm OD in adults and 50-75 mg/kg OD in children).



The following articles were used as a reference by the Lyme Disease Committee:

1. "Empiric Parenteral Antibiotic Treatment of Patients with Fibromyalgia and Fatigue and a Positive Serologic Result for Lyme Disease: A Cost Effective Analysis," Robert W. Lightfoot, Jr. et al (1993, American College of Physicians, p. 503)
Comment: Even in endemic areas, the incidence of false positive serologic tests in patients with myalgia or fatigue exceeds true positives by a ratio of 4:1.

2. "The Use of Serologic Tests for Lyme Disease in a Prepaid Health Plan in California," Catherine Lay, MS et al (JAMA 2/9/94-Vol. 271, No. 6)
Comment: Over a three month period, 117 patients out of Kaiser Permanente HMO site had Lyme blood tests done. Only one was positive. Only 19% of the tests were performed because physicians suspected Lyme Disease in the patients.

3. "Appropriateness of Parenteral Antibiotic Treatment for Patients with Presumed Lyme Disease," Benjamin J. Lefft, MD et al (1993 American College of Physicians p. 518)
Comment: In patients whose only evidence for Lyme Disease is a positive immunologic test, the risk for empiric antibiotic treatment outweighs the benefits.

4. "Lyme Disease: Clinical Update for Physicians," Prepared by American Lyme Disease Foundation, Inc. (Fall 1993)

5. "Management of Lyme Disease Refractory to Antibiotic Therapy," Leonard H. Siegel, MD (Rheumatology Clinics of North America, Vol. 21, No. 1, 2/95, p. 217)
Comment: This review article by Dr. Siegel discusses the use and misuse of lab tests and other everyday issues. This is a key review article if one was going to limit their reading material.

6. "Lyme Disease," R.F. Meenan, MD (1994 Yearbook of Rheumatology published by Mosby p. 217)
Comment: This introduces the article by Dr. Dressler in the journal of Infectious Disease that sets the framework for the criteria for the interpretation of the Western blot test.

7. "The Overdiagnosis of Lyme Disease," Allen C. Steere, MD et al (JAMA, 4/14/93-Vol. 269, No. 14, p. 1812)
Comment: In this study, the most common reason for lack of response to antibiotics was misdiagnosis.

8. "Summary of the First 100 Patients Seen at a Lyme Disease Referral Center," Leonard H. Siegel, MD (American Journal of Medicine, 6/1990, Vol. 88, p. 577)
Comment: Out of the first hundred patients referred to the Robert Woods Johnson Lyme referral clinics, only 37 of the 100 patients met the criteria for Lyme Disease; 25 out of these 100 are diagnosed as a fibromyalgia. This was the first known association between the two entities.