Asymptomatic Infection with Borrelia burgdorferi
- 1Division of Rheumatology/Immunology, Tufts University School of Medicine, New England Medical Center, Boston, Massachusetts
- 2Department of Medicine, Yale University School of Medicine, New Haven, Connecticut
- 3Division of Infectious Diseases, New York Medical College, Valhalla, New York
- Reprints or correspondence: Dr. Allen C. Steere, Div. of Rheumatology, Massachusetts General Hospital, 149 13th St., Rm. 8301, Boston, MA 02129 (asteere{at}partners.org).
Abstract
The natural history of asymptomatic seroconversion to Borrelia burgdorferi has been unclear. We report here, on the basis of a post hoc assessment, the frequency and outcome of asymptomatic seroconversion to B. burgdorferi in participants of a large Lyme disease vaccine trial. We show that infection with B. burgdorferi may be asymptomatic but that asymptomatic infection is unusual in the United States.
Lyme disease, which is caused by the tickborne spirochete Borrelia burgdorferi, usually begins with an expanding skin lesion, erythema migrans [1]. Less commonly, patients may present during summer with nonspecific systemic symptoms, such as headache, arthralgias, or fever, without erythema migrans [2]. In untreated patients, these early signs or symptoms are often followed by neurologic, heart, or joint involvement [3]. However, in epidemiologic studies, patients have been identified who have antibody reactivity to B. burgdorferi but who do not remember having any signs or symptoms of the infection [4–9]. Because the methodology of these studies has been quite variable, the natural history of asymptomatic infection with B. burgdorferi has been unclear [10].
We report here, on the basis of a post hoc assessment, the frequency and outcome of asymptomatic IgG seroconversion to B. burgdorferi in participants of a large Lyme disease vaccine trial.
Patients and Methods
During the winter of 1995, 10,936 people (ages 15–70 years) were enrolled in a double-blind, placebo-controlled Lyme disease vaccine trial by investigators in 31 centers in 10 states in the United States where the infection is endemic [11]. Informed consent was obtained from all patients, the human experimentation guidelines of the US Department of Health and Human Services were followed in conducting this research, and the study was approved by the human investigations committee at each center. Participants received injections of recombinant outer-surface protein A (OspA) of B. burgdorferi in adjuvant or of placebo at study entry and at 1 and 12 months after study entry. The protocol was designed to identify all cases of definite, possible, or asymptomatic Lyme disease that occurred in the study population during the 20-month study period. Suspected cases were evaluated when symptoms were present.
Blood samples were obtained from all patients at study entry and 2, 12, and 20 months later. Asymptomatic seroconversion was defined as the development of an IgG antibody response to B. burgdorferi, as determined by whole-cell sonicate Western blotting (Mardex), between 2 and 12 months or between 12 and 20 months, the intervals encompassing the 2 summer transmission seasons. The blots were interpreted according to the criteria of the Centers for Disease Control and Prevention and the Association of State and Territorial Public Health Laboratory Directors [12]. Standard testing by sonicate ELISA was not done, because this assay would be expected to yield positive results for samples from patients who had been vaccinated with OspA.
When the results of Western blotting became available, ∼4–6 months after the summer transmission season, the investigators were asked to contact affected study participants to determine whether they had experienced possible symptoms of Lyme disease during the period of seroconversion. Because there were no official recommendations with regard to antibiotic treatment of asymptomatic infection in Lyme disease, decisions about treatment were left to the discretion of the individual investigator and participant.
For this post hoc analysis, pre- and postsummer serum samples from participants classified as having asymptomatic seroconversion were retested using a new ELISA for Lyme disease, which uses a 26-mer peptide of the sixth invariant region (IR6) of the VlsE lipoprotein of B. burgdorferi [13]. This test, which was not yet available during the vaccine study, is not affected by vaccination with OspA. The cutoff optical density value for a positive response was defined as 3 SD above the mean absorbance of the 8 negative-control samples included on the same plate, which were representative of values obtained previously from 50 healthy control subjects. For each sample, an antibody titer was also calculated by adjusting the optical density value of the sample for which the titer was unknown with a standard curve made from dilutions of a serum sample known to be positive that was tested on the same plate.
Results
Of the 269 study participants in the vaccine trial who met the criteria for Lyme disease, 30 (11%) were classified as having asymptomatic IgG seroconversion to B. burgdorferi [11]. These 30 patients did not present to their study physician for the evaluation of symptoms. Instead, they were identified in serosurveys of the entire study population. Although it was subsequently learned that some of these patients were not completely asymptomatic, they were still included in the group with asymptomatic seroconversion in the vaccine study, because their signs and symptoms were not evaluated according to protocol.
Fifteen of the 30 participants (13 in the placebo group and 2 in the vaccine group) seroconverted after 2 injections of vaccine or placebo, during the first year of the study, and an additional 15 participants seroconverted (all in the placebo group) after the third injection, 12 months later, during the second year. Because the 30 participants were similar, regardless of whether they were in the vaccine group or placebo group or in the first or second year of the study, the data from all 30 patients are presented together.
Serologic data. By definition, the 30 patients had negative IgG antibody responses to B. burgdorferi, as determined by whole-cell sonicate Western blot, in the serum samples obtained before summer (table 1). The blots showed a median of 1 of the 10 IgG bands (range, 1–4 bands) used in the Centers for Disease Control and Prevention diagnostic criteria [7]. In most instances, reactivity was directed against only the 41-kDa flagellar protein of the spirochete, a common cross-reactive response. By definition, the 30 patients had positive results of Western blotting in the samples obtained ∼4–6 months after the summer tick transmission season. These blots showed reactivity with a median of 8 of the 10 IgG bands (range, 5–10 bands) used in the diagnostic criteria.
Table 1
IgG results of Western blots of paired serum samples obtained before and after summer from 30 patients who met prospectively defined criteria for asymptomatic seroconversion to Borrelia burgdorferi.
Similarly, in the 30 samples obtained before summer, the IgG VlsE peptide ELISA showed a median absorbance of 0.02 (range, 0.001–0.44) (the cutoff value for a positive response was 0.04, which was 3 SD above the median value for healthy control subjects) (figure 1). When a standard curve made from dilutions of a highly positive control sample was used to convert these values to a titer scale, the median titer was <1 : 100 U (range, <1 : 100–1 : 800 U). After summer, the VlsE test in these patients showed a median absorbance of 1.40 (range, 0.004–2.92) and a median titer of 1 : 1600 U (range, <1 : 100–1 : 12,800 U).
Figure 1
IgG results of VlsE peptide ELISA of paired serum samples obtained before and after summer in 30 patients who met prospectively defined criteria for asymptomatic seroconversion to Borrelia burgdorferi. The range of values in 8 healthy control subjects, whose samples were included on each plate, is shown in the cross-hatched area.
Of the 30 patients, 2 had negative responses on the VlsE peptide ELISA for both the pre- and the postsummer sample. In both of these patients (patients 29 and 30 in table 1), Western blots showed 3 bands in presummer samples and 5 bands in postsummer samples. All other patients had an increase of ⩾3 bands between pre- and postsummer samples, and most had an increase of ⩾7 bands. Conversely, 6 of the remaining 28 patients (patients 23–28 in table 1) had values in presummer samples that were above the cutoff value (range, 0.09–0.44), and Western blots showed a median of 2 bands (range, 0–3 bands). However, these 6 patients had higher values in postsummer samples (range, 0.28–2.59), and Western blots showed a median of 8 bands (range, 7–10 bands). Thus, 28 of the 30 patients developed IgG antibody responses to B. burgdorferi during the tick transmission seasons, as determined by an increase of the number of bands found on Western blots and by ⩾4-fold increases in antibody levels to the VlsE peptide on ELISA.
Clinical outcome. For 25 of the 28 patients, clinical information was available that had been obtained when Western blotting determinations were completed, ∼4–6 months after the summer transmission season. Two patients reported having an expanding skin lesion, compatible with erythema migrans, during the summer in which they seroconverted. They sought antibiotic treatment from physicians who were not involved in the study. Five additional patients said that they had migratory joint and muscle pain, sometimes accompanied by fatigue, and 2 other patients reported fatigue alone during the summer in which they seroconverted. However, they did not seek medical attention for these symptoms. The remaining 16 patients said that they were completely asymptomatic during the period of seroconversion.
Approximately 6 months after seroconversion, 16 of the 25 patients received doxycycline, 100 mg twice daily for 14–30 days, and 1 was given amoxicillin, 250 mg 3 times/day for 30 days. The other 8 patients were not treated with antibiotic therapy. Approximately 12 months after seroconversion, Lyme arthritis developed in 1 of the 8 untreated patients, documented by a positive result of PCR for B. burgdorferi DNA in joint fluid and a positive result of serologic testing for the spirochete. This patient had been completely asymptomatic until knee swelling and pain developed. Thus, only 1 patient with asymptomatic seroconversion was known to have developed a late manifestation of Lyme disease.
Altogether, of the 25 patients who seroconverted for whom clinical information was available, only 15 (60%) had completely asymptomatic seroconversion and remained asymptomatic throughout the study. The serologic results for the symptomatic patients (patients 1–8, 23, and 24 in table 1) were similar to those for the asymptomatic patients.
Discussion
In the vaccine study [11], 30 (11%) of the 269 participants who acquired Lyme disease were classified as having asymptomatic IgG seroconversion to B. burgdorferi. In the present study, 28 of these 30 individuals were shown to have not only an increase in the number of bands seen on Western blots but also ⩾4-fold increases in antibody levels to the VlsE (IR6) peptide of the spirochete between pre- and postsummer serum samples. The VlsE peptide ELISA, which is not affected by vaccination, has been shown to have a high sensitivity and specificity for the diagnosis of Lyme disease, similar to that achieved with the 2-test approach that combines sonicate ELISA and Western blotting [13–15]. Although all 28 patients had negative results of Western blots of samples obtained before summer, 6 had values on VlsE ELISA that were above the cutoff value, which raises the question of whether these patients previously had Lyme disease. However, these 6 patients not only had increases in antibody responses seen on Western blot but also had ⩾4-fold increases in antibody levels on VlsE ELISA of samples obtained after summer. This suggests that they experienced reinfection, if not primary infection, with B. burgdorferi during the study.
Of the 25 patients who seroconverted for whom clinical information was available, only 15 (60%) had completely asymptomatic seroconversion and remained asymptomatic throughout the 20-month study period. If the percentage of asymptomatic cases in the vaccine study is adjusted by this number, 7% of the participants who met the criteria for Lyme disease, rather than 11%, had asymptomatic seroconversion to B. burgdorferi. In 3 previous epidemiologic studies of Lyme disease in the United States, the frequency of asymptomatic B. burgdorferi infection has ranged from 0% to 50% [4–6]. Of 129 persons from Fire Island, New York, from whom serum samples were obtained before and after summer, 4 demonstrated >4-fold increases in antibody titers to B. burgdorferi, as determined by indirect immunofluorescence, and 2 lacked symptoms of Lyme disease [4]. In serosurveys of 67 residents of Great Island, Massachusetts, 3 developed erythema migrans, and 2 had symptomatic seroconversion, as determined by whole-cell sonicate ELISA [5]. In contrast, in an outbreak of Lyme disease in Ipswich, Massachusetts, in which serologic testing was done using sonicate ELISA and Western blot, no cases of asymptomatic seroconversion were identified [6]. In Europe, where Borrelia garinii and Borrelia afzelii cause infection more often than does B. burgdorferi, 3 serosurveys performed using sonicate ELISA reported that the majority of seropositive individuals had no symptoms of the infection [7–9].
The present study had unique features. The vaccine study population included nearly 11,000 individuals in the 10 major states where Lyme disease is endemic. Patients were usually motivated to report to their study physicians for the evaluation of symptoms, and all participants were tested serologically to identify those who seroconverted during the summer tick transmission season. The major limitation of the study was that most patients with asymptomatic seroconversion were treated with antibiotic therapy ∼4–6 months after the tick transmission season, and the duration of follow-up was only 4–14 months. If patients had not been treated and if the follow-up period had been longer, other individuals might have been identified who had late manifestations of Lyme disease. Thus, the true frequency of asymptomatic infection with B. burgdorferi in the United States is probably <7% of cases of infection.
In a previous study of 55 untreated patients with erythema migrans who were followed up for 4–8 years, 6 (11%) developed neuroborreliosis and 2 (4%) had carditis within weeks after the skin lesion appeared, and 34 (62%) subsequently had Lyme arthritis [3]. Joint involvement developed within 6 months after disease onset in one-half of the cases and within 2 years in all cases. Therefore, in the vaccine study, the period without treatment and the duration of follow-up were probably long enough for identification of most of the patients who would have developed later manifestations of the infection. Despite this limitation, we believe that the study provides a more accurate determination of the frequency of asymptomatic infection with B. burgdorferi in the United States than has previously been available. In summary, this study demonstrates conclusively that infection with B. burgdorferi may be asymptomatic but that asymptomatic infection is unusual in the United States.
Acknowledgments
We thank the members of the Lyme Disease Vaccine Study Group, for providing patient information and observations; Peter Fisher, Jesus Hernandez, and Elise Taylor, for help in obtaining and analyzing patient information; Gail McHugh and Monika Rothemich, for assistance with serologic evaluation; Colleen Fitzpatrick, for help with preparation of the manuscript; and David S. Krause at GlaxoSmithKline Pharmaceuticals and Dennis L. Parenti at American Home Products, for guidance and support.
Footnotes
-
Financial support: SmithKline Beecham Pharmaceuticals; Centers for Disease Control and Prevention (cooperative agreement CCU110291); Eshe Fund.
- Received March 19, 2003.
- Accepted April 25, 2003.
- © 2003 by the Infectious Diseases Society of America
References
This Article
- AbstractFree
- » Full Text (HTML)Free
- Full Text (PDF)Free
Classifications
Services
Citing Articles
Google Scholar
PubMed
Related Content
Share
- Email this article
-
CiteULike
Connotea
Delicious
Facebook
Google+
Mendeley
Twitter
Published on behalf of
IDSA Members: For your free access to this journal, log in via the IDSA members area.
The Journal
Impact Factor: 8.186
5-Yr impact factor: 7.898
For The Media
For Authors
- Services for authors
- Self-archiving Policy
- IDSA Conflict of Interest policy
- Instructions for Authors
- ICMJE Form
- Submit Now!
Open access options for authors visit Oxford Open
This journal enables compliance with the NIH Public Access Policy
Corporate Services
Alerting Services
Related IDSA Publications
