Appropriateness of Lyme Disease Serologic Testing

We appreciate the opportunity to respond to the concerns raised by Dr. Stricker and Ms. Mervine.

Dr. Stricker asserts that we committed two major errors in our analysis. The first error, in his opinion, was the narrow definition of appropriate indications for Lyme disease testing, which “guarantees that valid screening tests would be labeled as inappropriate.” Our criteria for appropriate, inappropriate, and discretionary tests were clearly stated in the paper, and the criteria for appropriate and inappropriate tests were consistent with published guidelines (1-5). We added a third category (discretionary) because we recognized that evidence-based guidelines are lacking for many clinical presentations where Lyme disease may be considered in the differential diagnosis, including neurocognitive and musculoskeletal symptoms. Lyme disease serologic tests ordered for patients presenting with non-specific fibromyalgia-like pain syndromes and chronic fatigue were therefore classified as discretionary rather than inappropriate as Dr. Stricker implies. In fact, more than 50% of the tests evaluated in this study were ultimately classified as discretionary.

Dr. Stricker suspects that the “asymptomatic” patients did in fact have symptoms of chronic Lyme disease, citing the unlikely scenario (in his opinion) that truly asymptomatic patients would be tested for Lyme disease, especially in an emergency department. For this survey, we specifically asked clinicians “Was the patient symptomatic?” If they were symptomatic we then asked if the patient had any of the following signs or symptoms: fatigue/malaise, fever, erythema migrans, other rash illness, myalgia, headache, paresthesia, lymphadenopathy, or any other signs or symptoms for which the Lyme serologic test was ordered. While we cannot exclude the possibility of erroneous responses, it is highly unlikely that symptomatic patients were misclassified as asymptomatic based on these questions. One of us (A.R.) is a practicing emergency physician, and we can confirm that asymptomatic patients often visit emergency departments with a variety of concerns, including a desire for “a Lyme disease test” because of actual or suspected tick exposure. Indeed, we found that a known or suspected tick bite was strongly associated with inappropriate Lyme disease testing.

Dr. Stricker’s second major concern relates to the 27% rate of inappropriate Lyme disease testing, and he compares this rate to the low proportion of positive tests from HIV or syphilis screening in an effort to justify the higher rate of inappropriate Lyme disease testing. In this comparison, Dr. Stricker confuses the indication for testing (based on published evidence or guidelines) with the test result (positive or negative). A positive test is not necessarily an appropriate test, and vice-versa. That we found 27% of the tests were inappropriate certainly does not mean that 73% of the tests were positive, as Dr. Stricker’s argument would suggest.

Dr. Stricker argues that an inappropriate testing rate of 27% is justified based on the serious nature of chronic Lyme disease and the difficulty in diagnosis. However, inappropriate tests were narrowly defined for this study, and they did not apply to patients with chronic symptoms. Tests were classified as inappropriate only for patients who met at least one of the following criteria: 1) asymptomatic, 2) diagnosed erythema migrans, 3) empiric antibiotic therapy given, or 4) serology ordered as test-of-cure. The main issue is not whether the test result is positive or negative, but rather does the test contribute any clinically useful information. In the latter three circumstances, Lyme disease had already been diagnosed and the patient had received treatment. The additional benefit of serologic testing is minimal, particularly given the high rate of false-negative serologic tests among patients with early Lyme disease.

We object to Dr. Stricker’s assertion that we have trivialized Lyme disease. In particular, he uses quotation marks to imply that we said Lyme disease is “hard to catch and easy to cure.” We did not make this statement, and we believe this type of discourse is contradictory to the shared goal of accurate detection and effective treatment of Lyme disease.

Finally, we agree with Ms. Mervine and Dr. Stricker that Lyme disease serologic tests “remain flawed.” Indeed, the relatively poor accuracy of current Lyme disease serologic tests forms the foundation of our study. It is our hope that future diagnostic tests for Lyme disease will have higher sensitivity and specificity during all stages of Lyme disease to facilitate prompt and accurate diagnosis.

Sincerely,

Alan H. Ramsey, MD, MPH&TM

Edward A. Belongia, MD

Jeffrey P. Davis, MD

1. Steere AC. Lyme Disease. N Engl J Med. 2001;345:115–125.

2. Tugwell P, Dennis DT, Weinstein A, et al. Laboratory evaluation in the diagnosis of Lyme disease. Ann Intern Med. 1997;127:1009–1023.

3. Bunikis J, Barbour AG. Laboratory testing for suspected Lyme disease. Med Clin North Am. 2002;86:311–340.

4. Seltzer EG, Shapiro ED. Misdiagnosis of Lyme disease: when not to order serologic tests. Pediatr Infect Dis J. 1996;15:762–763.

5. American College of Physicians. Guidelines for laboratory evaluation in the diagnosis of Lyme disease. Ann Intern Med. 1997;127:1106–1108.

Competing interests:   None declared

The article by Ramsey et al. (1) concludes that inappropriate Lyme disease serologic testing is common in Wisconsin. We believe that this conclusion is inappropriate.

Lyme disease is a controversial illness (2-5). The controversy stems from the fact that the medical literature focuses on the limited early symptoms of the disease, such as the “bullseye” rash and joint swelling, while ignoring the serious sequelae of chronic Lyme disease such as neurocognitive dysfunction, fibromyalgia-like pain syndromes and chronic fatigue (2-5). These symptoms appear to be caused by persistent infection with the spirochete Borrelia burgdorferi and/or coinfecting organisms such as Babesia, Anaplasma, Ehrlichia and Bartonella (3-5). Chronic Lyme disease often occurs because of the missed diagnosis and/or inadequate treatment of early B. burgdorferi infection. The Centers for Disease Control and Prevention (CDC) estimate that Lyme disease is underdiagnosed by a factor of at least ten, and serologic tests for the disease remain flawed (3-5).

With this background, Ramsey et al. commit two major errors in their analysis. First, they use a very narrow definition of Lyme disease symptomatology to define “appropriate” indications for testing while ignoring the protean features of tickborne illness outlined above. This narrow definition of appropriateness guarantees that valid screening tests would be labelled as inappropriate. Conversely, truly “inappropriate” testing would only be performed on asymptomatic patients; however, the authors’ narrow symptom definition makes one suspect that “asymptomatic” patients did in fact have symptoms of chronic Lyme disease, but that these symptoms were either not recognized or not acknowledged by the authors. This probability is underscored by the fact that emergency physicians had the highest rate of ordering “inappropriate” Lyme tests. If patients were indeed “asymptomatic”, why were they being evaluated in an emergency setting? Thus the outcome measures in the article are suspect and suggest that the data is flawed.

The second major error by Ramsey et al. is based on their apparent view that Lyme disease is a trivial illness that is “hard to catch and easy to cure”. Thus an “inappropriate” testing rate of 27% would be unacceptable for this benign disease. If one compares the prevalence of Lyme disease with that of other infectious diseases such as syphilis or AIDS, however, the rate of “inappropriate” testing for Lyme disease (even by the authors’ inaccurate standards) is relatively low. For example, the yield from voluntary screening for HIV disease is generally about 2-5% (6), while the yield from screening for syphilis may be less than 0.004%, as shown in an article from the same issue of the Annals of Family Medicine (7). Thus it appears that 95% or more of serologic testing for these diseases may be “inappropriate”, but (the argument goes) since syphilis and AIDS are such devastating communicable illnesses, the “inappropriate” screening is justified. If one recognizes that chronic Lyme disease is also a serious illness that may be difficult to diagnose and treat, an “inappropriate” testing rate of only 27% seems equally justified.

Family practitioners are often the first medical professionals to encounter a patient with Lyme disease and to be faced with the challenging symptomatology of untreated or under-treated victims of this illness. Consequently we feel that it is especially important for your readers to have a better understanding of Lyme disease and tickborne coinfections. We encourage family practitioners to examine the evidence-based guidelines for the management of Lyme disease recently published by the International Lyme and Associated Diseases Society (ILADS) (8). Then test your patients in an appropriate manner.

References

1. Ramsey AH, Belongia EA, Chyou PH, Davis JP. Appropriateness of Lyme disease serologic testing. Ann Fam Med 2004;2:341-4.

2. Lautin A, McNeil EL, Liegner KB, Stricker RB; Sigal LH. Lyme disease controversy: Use and misuse of language. Ann Intern Med 2002;137:775-7.

3. Stricker RB, Lautin A. The Lyme Wars: time to listen. Expert Opin Investig Drugs 2003;12:1609-14.

4. Phillips SE, Bransfield R, Sherr VT, Brand S, Smith HA, Dickson K, Stricker RB. Evaluation of antibiotic treatment in patients with persistent symptoms of Lyme disease: An ILADS position paper. Accessed at http://www.ilads.org/ on August 1, 2004.

5. Harvey WT, Salvato P. 'Lyme disease': ancient engine of an unrecognized borreliosis pandemic? Med Hypotheses 2003;60:742-59.

6. Centers for Disease Control and Prevention (CDC). Voluntary HIV testing as part of routine medical care--Massachusetts, 2002. MMWR Morb Mortal Wkly Rep 2004;53:523-6.

7. U.S. Preventive Services Task Force. Screening for syphilis infection: Recommendation statement. Ann Fam Med 2004;2:362-5.

8. The ILADS Working Group. Evidence-based guidelines for the management of Lyme disease. Expert Rev Anti-Infect Ther 2004;2(Suppl):S1- S13.

Competing interests:   None declared