Guidelines for the Clinical Diagnosis of Lyme Disease
In den Annals of Internal Medicine 127:1106-1123, 1997 erschien ein Positionspapier "Guidelines for the Clinical Diagnosis of Lyme Disease".
Am 8. September 1998 veröffentlichten die Annals Stellungnahmen von
- Kenneth B. Liegner, MD und Janice Kochevar, FNP (Armonk, NY 10504),
- Mark E. McCaulley, MD, Steamboat Medical Group, Steamboat Springs, CO 80487,
- Angelina A.M. Blaauw, MD, University Hospital Utrecht, 3508 GA Utrecht, The Netherlands und Sjef van der Linden, MD, University Hospital Maastricht, 6202 AZ Maastricht, The Netherlands,
- Peter Tugwell, MD, MSc, Ottawa General Hospital, Ottawa, Ontario K1H 8L6, Canada, Allen Steere, MD, New England Medical Center, Boston, MA 02111 und Arthur Weinstein, MD, New York Medical College, Valhalla, NY 10595.
Das Folgende ist die Wiedergabe eines weiteren Briefs an den Herausgeber der Annals of Internal Medicine in einer html-Fassung von J. Gruber.
Dieser Brief ist im Internet u.a. vom Lyme Disease Network of New Jersey im LymeNet Newsletter Volume 6 Issue 11 abgedruckt worden.
Letter to the Editor
American College of Physicians
Independence Mall West
6th St. At Race
Philadelphia, PA
To Whom It May Concern,
We have several concerns with the report entitled Guidelines for
Laboratory Evaluation in the Diagnosis of Lyme Disease (Annals Int
Med 127, 1106-1123. 1997)
- The Centers for Disease Control have developed a set of clinical and
diagnostic criteria for surveillance purposes. The authors of these
"Guidelines" state, with no substitution, that this criteria is also
applicable to the clinical diagnosis of Lyme disease. To our
knowledge, no such evidence exists. It would appear that the published
"Guidelines" have as a basis a clinical criteria for Lyme disease
diagnosis which has never been tested except for clinical studies
published by the authors themselves. There is no external validation to
support the claim of equivalence between clinical diagnostic criteria
and the CDC surveillance definition.
- The authors state that "testing with ELISA is the cornerstone of
laboratory diagnosis for Lyme disease". In fact, it is not. The
commercially available ELISA assays for Lyme Disease do not meet
acceptable criteria according to the group that is responsible for
much of the United States Laboratory proficiency testing program in
Lyme Disease. The data, from a recent study by Bakken et.al. (J Clin
Microbiol 35:537-543,1997), indicated that "the sensitivity and
specificity of the currently used tests for Lyme Disease are not
adequate to meet the two-tier test approach being recommended by the
CDC/ASTPHLD group". Bakken et. al. (1997) also stated that a screening
test must have >95% sensitivity to adequately screen for Lyme disease
and that the currently available ELISA tests do not perform at that
level.
- The authors have not followed the CDC/ASTPHLD recommendation for
2-tiered testing. That is, all indeterminate and reactive ELISAs
should be reflexed to Western blot (WB), not just indeterminate ELISAs
as the authors of the "Guidelines" suggest. Certainly the authors
realize that reactive Lyme ELISA results may be non-specific because
of a number of cross reacting antibodies (e.g. antibody to the 41 kDa
flagellin protein).
- The authors have missed some important studies of Western blotting,
especially those that may be critical of the recommendations of
Dressler et. al. (J. Infect. Dis. 167, 392-400,1993) and the
CDC/ASTPHLD.
- The report by Engstrom et al (J Clin Microbiol 33:419-427, 1995) found, for example, that 20% of their Lyme patients
remained seronegative throughout the study and that fewer bands on the
IgG WB could be appropriately used for interpretation. They noted that
the WB was more specific and more sensitive than the ELISA. Their
study also showed that only 19% of patients treated with antibiotics
for 20 days still had a positive ELISA antibody response after one
year, yet almost 60% of their patients continued to be WB positive at
the end of the first year.
- The study by Aguero- Rosenfeld et al
(J Clin Microbiol 34:1-9, 1996) reported that 89% of patients, with
culture-confirmed erythema migrans (EM), developed specific IgG
antibodies by WB, but only 22% of these patients were positive by the
interpretive criteria proposed by the CDC/ASTPHLD. They further
reported that the duration of the antibody response was related to the
duration of the EM.
- Tilton et. al. (1997) stated that a highly
sensitive and specific Western blot is desirable for a 2 tiered test
approach or as a primary test. Despite the CDC/ASTPHLD
recommendations, many physicians who treat patients for L.D. do not
believe that an ELISA is an appropriate screening test and consequently
use the Western blot as a primary test.
The authors state that patients not be tested for Lyme Disease unless
the pretest probability of disease is between 0.20 and 0.80. These
recommendations will:
- rule out any laboratory detection of B. burgdorferi antibodies,
antigens, and/or DNA in non-endemic areas
- require physicians to screen patients based on epidemiological data
which may not be available to them outside their own local area
- require physicians to know the performance characteristics of a wide
variety of Lyme disease tests.
The authors in their comment on non-antibody based tests have chosen to
overlook a number of publications on the utility of direct detection
tests for Lyme disease.
- A comprehensive review of molecular
techniques for diagnosis of Lyme Disease has recently been published
by Schmidt (Clin Micro Rev 10, 185-201, 1997). They state "evidence
is growing that a positive PCR test can be associated with active
disease ... after adequate therapy, PCR results are usually negative"
- Manak et. al. (J Spirochet Tick Borne Dis 4., 11-20. 1997) in a
well controlled study using the CDC criteria for selection of patients
indicated that PCR on serum, plasma, or buffy coat could be effectively
used to monitor the efficacy of therapy.
- Similarly, Harris et. al.
(J Spirochet Tick Borne Dis 237, 1995) have validated the Lyme
urinary antigen test (LUAT) in more than 700 LD patients and controls.
The LUAT has a specificity of >95%.
These "Guidelines" only complicate
an already complex disease diagnostic process.
Richard C. Tilton, Ph.D.
Diplomate American Board Medical Microbiology
BBI Clinical Laboratories, Inc.
75 North Mountain Rd.
New Britain, CT 06053
rtilton@bbii.com
Mary N. Sand, Ph.D.
BBI Clinical Laboratories, Inc.
75 North Mountain Rd.
New Britain, CT 06053
Nick S Harris, Ph.D.,
Diplomate American Board Medical Laboratory Immunology
IGeneX Inc. Reference Laboratory
797 San Antonio Road
Palo Alto, CA