| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
 |
ABSTRACT |
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
INTRODUCTION |
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Antibiotics administered early in the course of the clinical presentation of Lyme disease generally eliminate infection,7 and doxycycline, amoxicillin, and ceftriaxone have become "standard" therapeutic agents for this condition.8 A set of evidence-based treatment guidelines formalized these therapeutic procedures.9,10 Certain patients who receive such standard courses of therapy may experience persistent or recurrent symptoms, especially if treatment is delayed.1,4 It thus becomes apparent that host immunity, even after standard antibiotic therapy, fails to prevent subsequent episodes of Lyme disease. Although EM rashes may recur, it remains uncertain whether such an event represents relapse or reinfection.
It may be that prompt antibiotic therapy effectively eliminates the Lyme disease pathogen and thereby prevents relapse. To evaluate this possibility, we determined whether recurrent episodes of EM lesions were caused by relapse or reinfection in a longitudinal study of the residents of an endemic site who received a standard course of antibiotic therapy delivered in a timely manner. In particular, we ascertained the site on the body, antibody response, and the season in which the EM lesion appeared. In addition, we obtained detailed medical histories from these people to identify risk factors for recurrent episodes of Lyme disease.
|
MATERIALS AND METHODS |
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Residents experiencing symptoms suggestive of any of these three tick-borne infections were asked to provide a history and a blood sample and submit to a physical examination. A questionnaire was applied at 2, 4, and 8 weeks after the initial visit and until subjects became asymptomatic. Blood samples obtained during the acute and convalescent stages of each clinical episode were subjected to specific serologic and polymerase chain reaction (PCR) analyses.
We offered long-term follow-up of our subjects by means of a serosurvey that was conducted on Block Island during the spring and autumn of each year of the study. Some subjects were enrolled in the serosurvey portion of the study when they first experienced Lyme disease, whereas others had already been enrolled in the serosurvey when they developed their first episode of Lyme disease. All Block Island residents were recruited for the survey through announcements in the local newspaper, over a cable television network, and through notices posted at the Block Island Medical Center, as previously described.11 All participants were asked to respond annually to a questionnaire and to provide a blood sample for serologic and PCR analyses of these infections. Information on tick exposure included questions about recent tick bites. As a result, serum samples were available to us before and after any illness recorded by many of our study subjects. All procedures were approved by the Connecticut Children’s Medical Center and Harvard School of Public Health Institutional Review Boards, and written informed consent was obtained from all patients or their parents or guardians.
Case definitions.
A case of Lyme disease was defined as a physician diagnosis of EM (expanding, ring-like erythematous rash 
5 cm in diameter) or the presence of two or more influenza-like symptoms (fever, headache, neck stiffness, myalgia, fatigue, or joint pain) and laboratory confirmation of acute infection.12 An episode of Lyme disease was considered to be "primary" when the subject developed characteristic symptoms with no previous history of Lyme disease and no previous anti-Bo. burgdoreferi antibody detectable in serum. The episode was considered to be "recurrent" if Lyme disease had previously been diagnosed in that person. "Reinfection" was defined as the development of an EM lesion at a site remote from a previous EM lesion or the development of a subsequent episode of Lyme disease without detectable anti-Bo. burgdorferi antibody between episodes. "Relapse" was defined as the development of a second EM lesion at the same site as a previous erythema migrans lesion accompanied by persistence of Bo. burgdorferi antibody between episodes. Clinical episodes of babesiosis or HGA were defined as illnesses consisting of two or more acute influenza-like symptoms accompanied by specific laboratory evidence of these infections. "Coinfection" by the agent of Lyme disease with that of any one or a combination of these infections was defined as an illness that met the case definition of Lyme disease and one or another of these diseases.
Clinical manifestations and antibiotic therapy. We described the severity of acute illness by recording the frequency of each of 18 standardized symptoms and signs that occurred during each episode of infection. Duration of symptoms was ascertained at the 2- to 8-week follow-up visit(s), supplemented by the annual serosurvey interview. Duration of symptoms after the acute illness was categorized in terms of one or more symptoms lasting < 2 weeks, between 2 weeks and 1 month, between 1 and 2 months, between 2 and 3 months, and > 3 months. These time intervals were standardized to the following number of days: 10.5, 22, 45, 75, or 120, respectively. Specific questions about joint pain and swelling and neurologic and cardiac complications were assessed by descriptions of such complications on questionnaires that were completed during acute illness, follow-up visits, and survey follow-up. A subject was considered to have received a standard course of antibiotic therapy when given 10–28 days of an antibiotic recommended in evidenced-based treatment guidelines.9,10
Laboratory assays.
All laboratory assays were performed at the University of Connecticut Health Center. Serological evidence of exposure to the Lyme disease spirochete was derived by ELISA and Western blot.13,14 Bo. burgdorferi strain 2591 was used as antigen.15 A reactive serum using ELISA was defined as one reacting at a dilution 1:320 using IgG or IgM conjugate. For Western blot, specimens were considered positive if the IgG immunoblot contained 5 of the 10 most common Bo. burgdorferi–specific bands or the IgM immunoblot contained 2 of 3 bands as specified by the Centers for Disease Control and Prevention.14 Evidence of babesial infection was assessed serologically by an immunofluorescent assay (IFA) test.16 A reactive serum was defined as one reacting at a dilution of 1:64 or greater. HGA infection was diagnosed serologically by ELISA using a standardized recombinant HGA-44 suspension as the antigen.17 A reactive serum was defined as one reacting at a dilution of 1:64 or greater. The DNA of the agents of Lyme disease, babesiosis, and HGA was sought in whole blood samples by means of pathogen-specific PCR assays.18–20 The gene targets included a 294-bp portion of the Bo. burgdorferi osp A gene, a 238-bp portion of the Ba. microti nuclear small subunit ribosomal gene, and a 247-bp portion of the A. phagocytophilum nuclear small subunit ribosomal gene.
Statistical design. Descriptive statistics (proportions and means with associated confidence limits) were calculated within each patient group, according to disease episode. Information on tick exposure collected at the serosurvey visits was calculated as percent of visits reporting tick bite. The Wilcoxon test was used to compare tick exposure between single and multiple Lyme disease (LD) episode groups. P values are two tailed. Statistical analyses were performed using JMP version 5.0 (SAS Institute, Cary, NC).
|
RESULTS |
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
). About 90% of the episodes included an EM rash among both the single episode and multiple episode groups, whereas the remaining episodes consisted solely of a flu-like illness. These diagnoses were made by physicians who had treated numerous patients experiencing Lyme disease. In addition, serologic confirmation was available for three fourths of the cases, and borrelial DNA was identified in blood in four of those experiencing recurrent Lyme disease (Table 2). All study subjects who experienced recurrent episodes of Lyme disease met our case definition of reinfection. Although two of the subjects developed a recurrent EM rash at the same body site, they did not meet the case definition of recrudescent infection because both had no detectable anti-Bo. burgdorferi antibody between episodes. In addition, both of these subjects experienced recurrent Lyme disease 1 or more years after the previous episode, and both were exposed frequently to vector ticks; they noted a tick bite before four of their combined six disease episodes. The clinical characteristics of all recurrent episodes of Lyme disease are consistent with reinfection rather than persistent infection and relapse.
|
|
We identified several epidemiologic features of recurrent Lyme disease that might help differentiate reinfection from relapse. The mean intervals between the first and second and second and third episodes were 43 and 38 months, respectively, and never < 12 months. A shorter time interval between infections would be expected in the case of persistent infection and relapse. Among those who experienced recurrent Lyme disease, the seasonal distribution of the initial cases of Lyme disease did not differ from that of recurrent cases (Figure 1). Tick bites were reported more frequently by subjects who experienced repeated episodes of Lyme disease than by subjects who experienced only a single episode. Of 121 subjects who were permanent Island residents and who experienced recurrent Lyme disease, tick bite was reported at 61% of visits compared with 47% (P < 0.05) for those experiencing only a single episode. Time spent outdoors near vegetation, use of personal protection measures, ownership of pets, and interventions against ticks did not differ between these groups of people. Recurrent clinical episodes of Lyme disease are associated with frequent tick bites, suggesting that the mechanism of recurrence is reinfection rather than persistent infection and relapse.
|
The severity of symptoms during the initial episode of recurrent Lyme disease was compared with that associated with a second and third episode of disease. We confined this analysis to those subjects who returned for at least one follow-up visit, who had no pre-existing arthritis, and who were not co-infected. Among subjects who experienced recurrent Lyme disease, the number and mean duration of symptoms experienced during the initial episode of Lyme disease was similar to that experienced during the second episode. The number and duration of symptoms decreased during the third episode, but the difference was not statistically significant (Table 3). Although the clinical severity of the first two episodes of recurrent Lyme disease was similar, the number and duration of symptoms tended to decrease during the third episode.
|
|
DISCUSSION |
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Various elements of clinical evidence serve to distinguish persistence and relapse of infection from reinfection caused by the agent of Lyme disease. Relapse is characterized by the appearance of sequential EM lesions in the same vicinity as the original lesion, whereas EM lesions caused by reinfection generally occur at a body site remote from the original lesion.2,5,6 Based on experience with other soft tissue infections, relapse of EM would be expected to occur within a few weeks of the appearance of the original EM lesion. The appearance of a new EM lesion a year or more after the original EM would be more consistent with reinfection. Relapse of infection also is associated with persistence of specific antibody between EM episodes caused by persistent antigenic stimulation. The absence of anti-Bo. burgdorferi antibody indicates that the pathogen has been cleared, although demonstrable antibody may persist for a year or more after resolution of infection.26,27 The clinical features that characterized each of our recurrent episodes of Lyme disease were consistent with reinfection rather than relapse.
Epidemiologic data provided additional support for the concept that recurrent Lyme disease symptoms in our subjects were caused by reinfection. All episodes occurred during late spring and early summer. Tick bites were reported more often among people experiencing recurrent episodes of Lyme disease than people who experienced only a single episode of disease. Both of these epidemiologic criteria support our three clinical points of evidence, indicating that recurrent Lyme disease in our study population was caused by reinfection rather than relapse.
Immune impairment might be expected to increase the frequency and severity of recurrent Lyme disease; none of our recurrently infected subjects had evidence of immunodeficiency. Coinfection by the agents of babesiosis or HGA may immunosuppress and thereby might increase the likelihood of relapse.28–32 We found no such association, however. Although our subjects seemed to be immunologically intact, specific immunity resulting from a previous episode of Lyme disease did not protect against subsequent infection, nor did it seem to reduce the severity of symptoms during the course of recurrent infection. The acute clinical presentation in people reinfected by the agent of Lyme disease was much like that during the initial episode, although the number and duration of symptoms tended to be less during the third episode. Lyme disease appeared not to relapse in our apparently immune-intact study population.
The use of standard antibiotic therapy early in the course of Lyme disease generally provides an excellent outcome.8–10 Few, if any, patients given such treatment suffer relapsing symptoms of disease. Even in such an intensely endemic site as Block Island, relatively few residents become re-infected. Risk of reinfection can further be minimized by avoiding exposure to vector ticks. Prompt administration of standard antibiotic therapy for early Lyme disease reliably eliminates persistent infection and prevents relapse.
Received August 7, 2006. Accepted for publication August 18, 2006.
Acknowledgments: We are grateful to Dorothy Dahl and the late Norman Dahl who made possible our studies on Block Island.
Financial support: This study was supported by grants from the National Institutes of Health: AI 32403 (Drs Krause, Persing, Spielman, and Telford and Ms Christianson), AR 41497 (Drs Krause and Persing and Ms Christianson), AI 19693 (Drs Pollack, Telford, and Spielman), and the University of Connecticut Health Center General Clinical Research Center (MO1RR06192).
* Address correspondence to Peter J. Krause, 282 Washington Street, Hartford, CT 06106. E-mail: Pkrause{at}ccmckids.org 
Ivo M. Foppa, Raymond Ryan, Peter Baute, Janice Miller, Stephen Wikel, Patricio Tomas, Feliciano Dias, Theresa George, Richard Pollack, and Sam R. Telford III.
Authors’ addresses: Peter J. Krause, Georgine S. Burke, and Diane Christianson, Connecticut Children’s Medical Center, 282 Washington Street, Hartford, CT 06106, Telephone: 860-545-9490, Fax: 860-545-9371. Daniel T. Foley, Yale University School of Public Health, New Haven, CT, Telephone: 860-208-6061, Fax: 860-530-1339. Linda Closter, Peter Baute, and Janice Miller, Block Island Medical Center, Block Island, RI 02807, Telephone: 401-466-2974, Fax: 401-466-5476. Andrew Spielman, Richard Pollack, and Sam R. Telford III, Harvard School of Public Health, 665 Huntington Avenue, Boston, MA 02115, Telephone: 617-432-2058, Fax: 617-432-1796. Ivo Foppa, University of South Carolina, Columbia, SC 29208, Telephone: 803-777-5056, Fax: 803-777-2524. Raymond Ryan, Patricio Tomas, Feliciano Dias, and Theresa George, University of Connecticut Health Center, Farmington, CT, Telephone: 860-679-3132, Fax: 860-679-1401. Stephen Wikel, University of Connecticut Health Center, Farmington, CT, Telephone: 860-679-8129, Fax: 860-679-8130.
Reprint requests: Peter Krause, 282 Washington Street, Hartford, CT 06106. E-mail: Pkrause{at}ccmckids.org.
|
REFERENCES |
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  Y.-F. Hsieh, H.-W. Liu, T.-C. Hsu, J. C.-C. Wei, C.-M. Shih, P. J. Krause, and G. J. Tsay Serum Reactivity against Borrelia burgdorferi OspA in Patients with Rheumatoid Arthritis Clin. Vaccine Immunol., November 1, 2007; 14(11): 1437 - 1441. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
