State of California
Department of Health Services
M e m o r a n d u m
Date: January 18, 2000
To: The Record & Interested Parties
From: Vector-Borne Disease Section
Disease Investigations & Surveillance Branch
Division of Communicable Disease Control
601 N. 7th Street, MS-486
P.O. Box 942732
Sacramento, CA 94234-7320
(916) 324-3738
Subject: Investigation of suspected Lyme disease in Malibu, Los Angeles County
Curtis L. Fritz, DVM, PhD
Research Scientist II
Disease Investigations & Surveillance Branch
[ (510) 540-2566 or CFRITZ@dhs.ca.gov ]
Vicki L. Kramer, PhD, Chief
Vector-Borne Disease Section
Disease Investigations & Surveillance Branch
[ (916) 324-3738 or VKRAMER@dhs.ca.gov ]
Duc J. Vugia, MD, MPH, Chief
Disease Investigations & Surveillance Branch
[ (510) 540-2566 or DVUGIA@dhs.ca.gov ]
Summary
Fifteen residents of the Malibu area of Los Angeles County who had previously been diagnosed with Lyme disease were evaluated for objective evidence of this and other tick-transmitted diseases. Participants completed a personal questionnaire and provided a blood sample for serologic testing; additional clinical information on their suspected LD illness was obtained by review of their medical records. Blood samples were split between two public health laboratories, one academic laboratory, and one private laboratory. The private laboratory reported serologic evidence of exposure to Borrelia burgdorferi in three specimens and to Babesia microti in 10 specimens; the national public health laboratory identified serologic evidence of B. burgdorferi exposure in no specimens, and the academic laboratory reported two reactive and two borderline titres to Babesia sp. WA1, but none reactive to Babesia microti. The state public health laboratory detected serologic evidence of exposure to Ehrlichia spp. in none of the specimens; the private lab reported detectable antibody in five patients, but not at a concentration it considered clinically significant. Clinical data on participants did not objectively support a diagnosis of Lyme disease in any of the participants. The results of this study do not indicate that residents of the Malibu area are at significantly increased risk of tick-transmitted infectious diseases, including Lyme disease.
Background
Beginning in the early 1990s, some residents of the Malibu area of Los Angeles County expressed concern to local and state public health officials that they may be at increased risk of contracting Lyme disease (LD). Anecdotal reports of numerous residents with an illness diagnosed and treated as LD were forwarded to the Los Angeles County Department of Health Services. However, few of these cases were formally reported through the standard reportable disease channels and therefore the number and character of these cases were not available for evaluation. In January 1998, the California Department of Health Services (DHS), in cooperation with the Los Angeles County Department of Health Services (LADHS), initiated a study to assess whether there was evidence of increased risk of LD and other tick-borne diseases in the Malibu area. The Centers for Disease Control and Prevention (CDC), the University of California at Davis (UCD), and IGeneX Laboratories were invited to collaborate on this investigation and provide laboratory serologic support.
Methods
Study population
Two physicians-Physicians A and B-who were identified as having diagnosed Malibu residents with LD were contacted. Lists of patients diagnosed as having LD were requested and obtained from each physician. Criteria for inclusion in the study were 1) a specific diagnosis of LD, 2) diagnosis within the preceding three years, and 3) being a resident of Malibu or surrounding community at the time of diagnosis. At least three attempts were made to contact each eligible participant by telephone. Individuals agreeing to participate were sent an information packet that included a cover letter describing the study, an informed consent form, and a personal questionnaire (Appendices A-C). Participants were requested to complete the consent form and questionnaire and return it to the DHS in an enclosed post-paid envelope. Participants who did not return the completed materials within one month of mailing were followed-up by telephone.
In addition to the case-patients, Physician A was asked to prospectively recruit patients who did not have clinical evidence of LD and who were willing to provide information and specimens as part of a comparison group. Twenty packets of questionnaires, letters of consent, and post-paid envelopes were provided to Physician A for this purpose.
Exposure/environmental information
Study participants were mailed a five-page questionnaire that requested demographic information; size and character of residential property; frequency, nature, and duration of outdoor activities; travel history; and tick bite history and current prevention measures. Each participant was requested to complete his/her own questionnaire, independently of any other participants within the same household. Parents were asked to complete separate questionnaires for each child that was enrolled in the study.
Clinical information
Following receipt of signed consent letters, medical records of patients were reviewed for clinical information relevant to the time period when they were originally diagnosed with LD. When available, particular note was made of presenting history, physical exam findings, laboratory serodiagnostic reports, and treatment regimens prescribed.
Serology
All patients were asked to provide a serum specimen for testing. Patients were invited to attend a special collection session coordinated by the LADHS and held at Physician A's office in June 1998. Patients who were unable to attend were asked to have their blood drawn by Physician A at a mutually convenient time and forwarded to the LADHS.
The LADHS forwarded all serum specimens to the DHS Viral and Rickettsial Diseases Laboratory (VRDL). Specimens were divided and aliquots distributed for testing to the VRDL, the Division of Vector-Borne Infectious Diseases at CDC, the School of Veterinary Medicine at UCD, and IGeneX Laboratories. Each specimen was examined for serologic evidence of Borrelia burgdorferi (CDC), rickettsial agents of human monocytic (HME) and granulocytic (HGE) ehrlichiosis (VRDL), and Babesia microti and Babesia sp. WA1 (UCD). IGeneX Laboratories tested each specimen for evidence of B. burgdorferi, monocytic and granulocytic Ehrlichia, and Babesia microti.
B. burgdorferi serology at CDC was conducted by an initial screening of all specimens with an enzyme immunoassay (EIA) assay based on the flagellar protein. All specimens that were positive (OD >3 standard deviations above mean of negative controls) or equivocal (3>OD>1 standard deviations) were tested by IgM and IgG Western immunoblotting (WB). Western immunoblots were interpreted in accordance with standard criteria requiring a minimum of five of 10 diagnostic bands (18, 21-24, 28, 30, 39, 41, 45, 58, 66, 93 kDa) for IgG and two of three diagnostic bands (21-24, 39, 41 kDa) for IgM [Centers for Disease Control 1995]. All specimens were tested for evidence of B. burgdorferi also by IGeneX Laboratories using a whole-cell sonicate EIA, IgM and IgG WB, and polymerase chain reaction (PCR). EIA results were considered positive if the "Lyme Index Value" was >1.2 and equivocal if between 1.0 and 1.2. IgM and IgG immunoblots were condsidered positive if at least two of five (23-25, 31, 34, 39, 41 kDa) or two of six (23-25, 31, 34, 39, 41, 93 kDa) bands, respectively, were present; specimens with only one of the indicated bands were reported as equivocal. PCR techniques were performed to detect both genomic and plasmid DNA fragments of B. burgdorferi.
Serology for monocytic and granulocytic ehrlichiosis were conducted at VRDL by immunofluorescent antibody assay (IFA), using antigens derived from Ehrlichia chaffeensis and the HGE agent. Reciprocal titres of greater than 64 were considered positive. Serology at IGeneX for HME and HGE also employed an IFA procedure, using E. chaffeensis and E. equi, respectively, as antigens. Reciprocal titres of 40 or more were considered positive; however, IGeneX regards only titres >160 as "compatible with disease states".
Serology for Babesia spp. was conducted at UCD by IFA, using Babesia WA1 and B. microti antigens derived from patients in Washington and Minnesota, respectively. Reciprocal titres of 640 or greater were reported as reactive; titres of 320 were considered "borderline". Attempts to identify piroplasm DNA were conducted by PCR on all specimens. Finally, smears of all blood specimens were examined for presence of intraerythrocytic inclusions. Babesia serology at IGeneX was conducted by separate IgG and IgM IFA using a B. microti antigen. Reciprocal titres of 16 or greater for IgM and 40 or greater for IgG were considered positive. IGeneX also conducted PCR for B. microti on all specimens.
Patients were notified of their personal serologic results by mail in July 1999. In addition, a summary of serologic results was provided to all participants and Physicians A and B.
Results
Study participants
Physician A provided a list of seven patient names; Physician B provided 24 patient names. One patient was duplicated on the two lists. Eight patients were deemed ineligible to participate because they were either diagnosed more than three years prior to commencement of the study or were not residents of the Malibu area at the time of their diagnosis. Attempts were made to contact by telephone the remaining 22 eligible patients. Three patients declined to participate. Three patients could not be reached after multiple attempts. Sixteen patients agreed to participate and were sent letters of consent and copies of the exposure/environmental questionnaire. Fifteen patients completed and returned the questionnaire and letters of consent.
A total of nine women and six men, representing 10 different households were enrolled in the study. Age of patients ranged from 8 to 71 years, with a median of 43 years. Four patients were children under 18. All 15 were white, non-Hispanic.
No potential comparison group patients were forwarded from Physician A.
Exposure/environmental questionnaire
Fifteen patients completed and returned the personal questionnaire. Six patients reported finding a tick on themselves during the 12 months preceding their LD illness: patients described the ticks as attached to skin (1), on skin but unattached (4), and on clothing unattached (1). Patients reported routinely using the following tick bite preventive measures: wearing light colored clothing (2), wearing long pants and tucking pant legs in socks (3), using repellants on skin and/or clothing (4), examining one's clothing and/or body for ticks following outdoor activity (6).
Patients' residential communities were described as urban (1), suburban (3), semi-rural (9), and rural (2). Of 12 patients who provided information, median property size was approximately 3700 square meters (range, approximately 600 to 16,000 square meters). Patients' properties ranged from 0% to 75% natural trees, shrubs, or grasses. Of the 11 patients living in rural or semi-rural areas, seven reported living within 400 meters of a naturally wooded area. Two patients reported regularly spending time in wooded areas near their residence. Five patients reported regularly spending time on trails in uncultivated areas; two of these also reported spending time off trails in these same areas.
Nine patients reported travel outside California (Arizona [3], the Midwest [2], Nevada [2], Montana [1], and Oregon [1]) during the 12 months preceding their LD illness. One patient reported camping or hiking in naturally wooded areas in the 12 months preceding onset of LD illness.
Clinical information
Medical records for all 15 patients were reviewed. Date of LD diagnosis was recorded for 11 patients and ranged from April 1995 to November 1997. Date of illness onset was recorded for 10 patients and ranged from September 1994 to August 1997. The median interval between date of onset and date of diagnosis for the eight patients for whom both dates were recorded was eight months (range, 3 to 28 months).
The symptoms most commonly recorded on patients' medical records were pain and stiffness in muscles and joints (13). The sites involved included the back (6), knee (5), hand/fingers/wrist (5), neck (4), shoulder (3), hip (2), and elbow (2). Joint swelling or arthritis was reported in two patients. One of these patient [sic] had waxing and waning swelling of a single knee. The second patient had polyarthritis, mostly of the hands and feet; orthopedic consult described the condition as compatible with rheumatoid arthritis.
Other symptoms recorded for patients included fatigue (8), headache (3), tinnitis (3), and lymphadenopathy (2). Three patients were diagnosed with peripheral nerve deficits described as extremital weakness or numbness. Four patients reported central nervous deficits described as dizziness or memory changes. Two patients reported experiencing "palpitations"; cardiologic consultation of these patients failed to identify abnormalities on electrocardiogram.
Five patients reported having a rash, two of whom also reported an antecedent tick or "insect" bite. The rash was decribed as "puritic" in two additional patients, and "papular" in the fifth patient. Duration of the rash-48 hours-was recorded for only one patient.
Results of a total of 82 serologic tests for LD were recorded for 14 patients; one patient had no laboratory serology results recorded. The median number of separate LD serologic tests performed per patient was four (range, zero to 17). For each patient, the median number of separate specimens documented in the medical record as submitted for LD testing was two, (range, one to six). The interval between onset of symptoms and date of specimen testing ranged from <1 to 44 months. LD serologic tests results recorded included EIA (14), IFA (12), and WB (51); type of test was not available for five results. Twelve patients had at least one positive result recorded. The number of patients with at least one positive result for each test type were: 2 of 10 (EIA), 4 of 7 (IFA), 4 of 11 (IgG WB), and 8 of 12 (IgM WB).
Eight patients had results recorded for serologic tests run concurrently for other tick-borne pathogens including E. chaffeensis (4), E. equi (4), and B. microti (6). Seven of these patients had at least one positive serologic result to one of these pathogens.
Fourteen patients received at least one course of antibiotics for their suspected LD. Antibiotics prescribed included clarithromycin (10), doxycycline (9), amoxicillin (4), and minocycline (2). Total duration of antibiotic therapy was not recorded for any patient; however, subsequent notes in eight patients' records permitted calculation of a median minimum treatment periods of 5.5 months (range, 1 to 18 months).
Serology
Fourteen patients provided a serum specimen for testing. Twelve patients provided specimens at the special collection session in June 1998; two others provide [sic] specimens later in June and in July. One patient elected not to provide a serum specimen.
Of 14 specimens tested at CDC, one specimen was positive and two were equivocal for B. burgdorferi antibody by EIA. Upon subsequent WB, all three of these specimens were negative for IgM and IgG antibody. Eleven specimens were negative by EIA and were not tested further. IGeneX reported one specimen positive by EIA and two equivocal. The EIA-positive specimen was the same as that detected by CDC, however the two equivocal specimens differed from those reported by CDC. Three specimens were reported positive and three equivocal by IgM WB. One specimen was positive and one equivocal by IgG WB. One specimen was positive for both IgM and IgG WB, but negative by EIA.
The VRDL tested 13 specimens for antibody to E. chaffeensis and the HGE agent. (One specimen was scant and insufficient for testing.) All 13 specimens demonstrated titres of <1:64 to each rickettsial agent. IGeneX reported all 14 specimens as negative for both IgM and IgG to E. chaffeensis. Three specimens were reactive for IgM (all 1:40) and three specimens reactive for IgG (1:40, 1:40, 1:160) to E. equi. One of these was reactive for both IgM and IgG to E. equi.
Babesia serology conducted at UCD identified two of 14 specimens reactive (1:1280, 1:640) and two borderline (1:320) for antibody to Babesia WA1. All 14 specimens were negative for intraerythrocytic inclusions upon examination of blood smears and negative for piroplasm DNA by PCR. All 14 specimens were seronegative for IgM and IgG antibody to B. microti. IGeneX reported three specimens reactive for IgM and eight reactive for IgG antibody to B. microti. Two of these were reactive for both IgM and IgG. Seven specimens were reported positive for B. microti by PCR, including two specimens that were seronegative.
Discussion
The clinical and serologic results of this study did not suggest that Malibu residents were at increased risk of LD. Using nationally recognized criteria for interpretation of serologic results, none of the patients had a detectable antibody level that would support current or past exposure to and [sic] infection with B. burgdorferi. Even using the less restrictive criteria of the private laboratory, only one patient had serologic evidence compatible with late or previous LD (WB IgG positive), further suggesting that the apparent risk of LD, as measured by seroprevalence, was low. Residual IgG antibody to B. burgdorferi has been demonstrated to remain detectable in treated LD patients up to three years after recovery [Feder et al. 1992]. Eligibility for enrollment in this study was purposefully restricted to persons with a diagnosis of LD within the preceding three years to maximize the opportunity for detecting residual antibody in any persons who truly had LD. Several patients believed, and were regarded by their physician, to have active late-stage LD at the time of the study. However, none of these had serologic evidence of either current or past infection. Furthermore, some patients reported having LD for several months or years. While the WB may be relatively insensitive in early disease (60-75%), its sensitivity improves dramatically with late disease (90-100%) as the immunologic response expands [Craven et al. 1996]. Persons with long-term LD should experience an immunologic response across numberous WB bands [Dressler et al. 1993] and thereby easily meet or exceed most interpretation criteria; however, all 14 specimens tested as part of this study had three or fewer bands present on IgG WBs conducted at either laboratory.
Based on clinical information available in their medical records, most of the patients did not have consistent objective symptoms typically associated with LD. The most common clinical presentation was of non-specific long-term myalgia and arthralgias; frank arthritis was rare and when present was usually attributable to another cause (i.e., rheumatoid arthritis). One patient had a clinical history compatible with possible LD, including a rash following an "insect bite" and pain and swelling of the knee; however, in addition to negative serology as part of the present study, all LD serology run on this patient at the time of illness (four specimens, nine tests) were also negative.
Because of differences in antigen preparation, laboratory methodology, and interpretation and reporting criteria, the discrepancies observed in the present study between serologic results from the collaborating institutional reference laboratories and the private laboratory were not unexpected. The more liberal interpretation criteria for WBs used by the private lab would be predicted to result in a greater number of "positives", as observed in the present study. Similar inconsistencies between laboratories were noted in some of the patients' medical records. Among the 14 patients who had LD serology performed, results were recorded from five different labs; all 14 patients had a least one LD serology result from IGeneX. Nine patients had LD serology results recorded from two different labs; one patient had results from three labs. It is difficult to quantify the rate of discordance since there were few instances noted where different labs performed comparable tests on contemporaneous samples. However, the problem is clearly illustrated in one patient's record where a single sample tested at Lab #1 was positive for IgG by IFA, but negative by EIA; that same sample was negative for IgM by both IFA and EIA at Lab #1, but positive for IgM at Lab #2. Other inconsistencies were evident with different specimens collected from the same patient over the course of his/her illness. One patient had a positive result for IgG WB from Lab #1. Four months later Lab #1 reported IgG negative by EIA. A third specimen collected six months later was reported as IgG positive by WB by Lab #1. These kinds of contradictory serologic results can present a conundrum for the physician who is uninformed regarding the immunologic basis of LD serologic techniques and their lack of national standardization.
The most striking discrepancy between the private laboratory and a reference laboratory in this study concerned the Babesia results: the private laboratory identified nine specimens seroreactive and seven PCR-positive to B. microti, whereas the reference lab found none reactive by either method. It is possible that the positive PCR results may represent misidentification of a prominent mammalian band as a piroplasm band that is very close in molecular weight. It is arguable whether B. microti is an appropriate antigen to use for serodiagnostic evaluation of California residents-while this species of piroplasm is prevalent in the northeastern U.S., it has to date not been isolated from ticks or mammals, nor been associated with any human disease, on the West Coast. Rather, another piroplasm, provisionally named Babesia WA1, has been associated with at least five human cases of disease, four in California, since 1991 [Persing et al. 1995]. Previous work elsewhere in California suggests that risk of exposure to this organism may be greater than that for B. burgdorferi in some ares [Fritz et al. 1997]. Whether the seroreactive specimens in the present study represent evidence of exposure to Babesia WA1 or non-specific cross-reactivity awaits further investigation; as such, it remains unclear what information these data provide relative to risk of babesiosis for residents of the Malibu area.
Subsequent to initiation of this investigation, objective evidence of B. burgdorferi-infected ticks in the Malibu area was discovered. In February 1998, a tick removed from a hiker in the Malibu hills was submitted to IGeneX Laboratory where it was reported as positive by PCR for B. burgdorferi. Unfortunately, there was insufficient remaining tick material available for confirmatory testing by an independent laboratory. In March 1999, the National Institute of Health's Rocky Mountain Laboratory reported culturing spirochetes from two of seven pools of I. pacificus collected the preceding month in Tapia County Park in the Malibu hills. These spirochetes were later confirmed as B. burgdorferi.
Recommendations
While the present study revealed little evidence to support the hypothesis that residents of the Malibu area are at elevated risk for Lyme disease, the risk of tick bite is real and the potential for acquiring tick-tranmitted disease should not be dismissed outright. Residents should take the necessary precautions when active in naturally wooded or vegetated areas to reduce the risk of tick bite, including wearing long pants, using tick repellants, and conducting tick checks at the conclusion of outdoor activities.
Lyme disease can often present a diagnostic challenge and it is incumbent upon physicians to base a diagnosis of LD on solid objective evidence and good clinical judgement. Lyme disease ought not to be a diagnosis of exclusion for patients with non-specific systemic symptoms and positive serology. Physicians should educate themselves regarding the clinical manifestations of LD and the judicious use of appropriate serodiagnostic assays. Even in a clinically compatible illness, physicians should not automatically regard a "positive" serologic result as confirmation of the diagnosis and need for treatment. Serologic and other tests for LD and other tick-borne diseases should be requested only when these diseases are suspected based on clinical symptomatology and not as part of a routine "tick panel".
References
Centers for Disease Control and Prevention. Recommendations for test performance and interpretation from the Second National Conference on Serologic Diagnosis of Lyme Disease. MMWR 1995; 44:590-1.
[See complete text at:
http://www.cdc.gov/epo/mmwr/preview/mmwrhtml/00038469.htm ]
Craven RB, Quan TJ, Bailey RE, et al. Improved serodiagnostic testing for Lyme disease: Results of a multicenter serologic evaluation. Emerg Infect Dis 1996; 2:136-40.
[See complete text at:
http://www.cdc.gov/ncidod/EID/vol2no2/craven.htm ]
Dressler F, Whalen JA, Reinhardt BN, Steere AC. Western blotting in the serodiagnosis of Lyme Disease. J Infect Dis 1993; 167:392-400.
[See abstract at:
http://www.ncbi.nlm.nih.gov/htbin-post/Entrez/query?uid=8380611&form=6&db=m&Dopt=b ]
Feder HM, Gerber MA, Luger SW, Ryan RW. Persistence of serum antibodies to Borrelia burgdorferi in patients treated for Lyme disease. Clin Infect Dis 1992; 15:788-93.
[See abstract at:
http://www.ncbi.nlm.nih.gov/htbin-post/Entrez/query?uid=1445977&form=6&db=m&Dopt=b ]
Fritz CL, Kjemtrup AM, Conrad PA, et al. Seroepidemiology of emerging tickborne infectious diseases in a northern California community. J Infect Dis 1997; 175:1432-9.
[See abstract at:
http://www.ncbi.nlm.nih.gov/htbin-post/Entrez/query?uid=9180183&form=6&db=m&Dopt=b ]
Persing DH, Herwaldt BL, Glaser C, et al. Infection with a Babesia-like organism in northern California. New Eng J Med 1995; 332:298-303.
[See abstract at:
http://www.ncbi.nlm.nih.gov/htbin-post/Entrez/query?uid=7816065&form=6&db=m&Dopt=b ]
Since 24 February 2000
This page hosted by
Get your own Free Home Page