Answers are grounded in this paper and cite their sources.
Lyme disease is one of the major emerging infections of the past decade. This global vector-transmitted disorder is due to a tick-borne spirochete named Borrelia burgdorferi, after its discoverer [1]. Lyme disease is geographically restricted to regions inhabited by the critical tick vector. Organisms are transmitted through the bite of infected Ixodid (hard body) ticks, particularly those belonging to the Ixodes persulcatus complex. They include I. scapularis in the northeast and midwest United States, I. pacificus in the western United States, I. ricinus in central and western Europe, and I. persulcatus in Asia. B. burgdorferi belongs to the genospecies complex B. burgdorferi sensu lato. This complex is genetically divided into multiple genospecies, including B. burgdorferi sensu stricto, B. garinii, B. afzelii, B. japonica, B. andersoni, and others [2]. Considerable genetic diversity exists among B. burgdorferi in both Europe and the United States, and different genospecies may predominate within different geographic areas. There are preliminary associations between specific strains and clinical disease patterns. A single tick may be infected with several different B. burgdorferi strains. It is likely that spirochete strain heterogeneity, as well as host immune and genetic factors, play a role in the development of Lyme disease. Although Lyme disease is a systemic infection, it has specific target organs (skin, nervous system, joint, heart, eye). Characteristic clinical syndromes are associated with early stage local infection, early stage disseminated infection, and late stage disseminated infection (Table 1). Cutaneous disease, followed by neurologic disease, is the most common feature [3]. Clinical syndromes associated with Lyme disease Clinical syndromes associated with Lyme disease At the current time, the diagnosis of Lyme disease is made on clinical and epidemiologic grounds. It is based on endemic area exposure, tick exposure, presence of abnormal symptoms and signs consistent with B. burgdorferi infection, failure to establish an alternative diagnosis, and appropriate response to antibiotic treatment. Ideally, clinical diagnosis is supported by consistent laboratory data [4]. The most convincing laboratory support would be direct detection of organisms through culture or stain. Unfortunately, with rare exceptions, these direct detection techniques have low yields and are not justified as routine tests. Culture of B. burgdorferi is most likely to be positive (30–90% yield) when skin punch biopsy is obtained from the erythema migrans (EM) lesion, which characterizes early stage local infection. Culture positive samples are reported on occasion from a variety of body tissues and body fluids, but are distinctly unusual. Culture requires the use of special (Barbour-Stoenner-Kelly) medium, which must be maintained weeks for organisms to be detected. The low yield is not surprising, given the limited number of organisms present in most infected samples. Although tissue and body fluids can be stained for spirochetes using nonspecific techniques such as silver staining, these techniques are subject to artifact and should be combined with immunologic or molecular identification techniques. Besides culture and stain, other direct diagnosis techniques include polymerase chain reaction (PCR) to detect B. burgdorferi nucleic acid, and antigen detection. These assays remain experimental at the present time, since they are not standardized for routine use on clinical samples and are often not widely available outside research facilities. Among indirect laboratory tests, the most widely used is serology to detect specific antibodies to the organism. This is not an active infection assay and can only document prior exposure, although seroconversion with an IgM response indicates recent exposure. Lymphocyte proliferation assays, which detect cellular immune responses, are not widely used due to their inherent variability as well as lack of specificity. Other nonspecific laboratory tests are helpful to document compatible synovial or cerebrospinal fluid (CSF) abnormalities, objective cardiac disturbances, or objective neurologic involvement (such as abnormalities on neurophysiologic tests, neurocognitive tests, or neuroimaging). The diagnosis of Lyme disease is not always straightforward, and there is overdiagnosis as well as underdiagnosis of cases [5]. Particularly troublesome in the United States has been the issue of patients with persistent post treatment problems, the so-called chronic Lyme disease/post Lyme disease syndrome [6]. This review will begin with a discussion of recent advances in the diagnosis of Lyme disease, and then discuss common pitfalls leading to misdiagnosis. There are recent diagnostic advances in Lyme disease in the areas of serologic testing and molecular diagnostics. Positive serology documents detectable levels of anti-B. burgdorferi antibodies. This is the single most common laboratory test used to support the diagnosis of Lyme disease. Serologic tests include immunofluorescence (IFA) and solid phase fluorescence assays, enzyme immunoassays (ELISA), and Western immunoblot assays. Antigen preparations for these tests involve whole cell sonicated spirochetes or partially purified proteins. Sensitivity and specificity of these assays vary based on when testing is done, methods used, and reagents used (Table 2) [7–14]. Much less common are indirect hemagglutination and complement fixation assays [15, 16]. In a proficiency program of 846 New York State laboratories, overall sensitivity for screening serologies averaged 95.4% and overall specificity averaged 98.8%[10]. In large part ELISA assays have replaced IFA assays. Because B. burgdorferi contains both specific and nonspecific immunodominant proteins, both IFA and ELISA assays are subject to false positive results. False positive rates as high as 20–40% are reported [14]. Western immunoblot assays improve on specificity by allowing detection of antibodies to identifiable proteins. Western blot confirms only 35–50% of positive ELISAs however, depending on the population screened [17, 18]. Serologic tests to measure anti-B. burgdorferi antibodies The wide range reflects low sensitivity in early stage infection. Serologic tests to measure anti-B. burgdorferi antibodies The wide range reflects low sensitivity in early stage infection. None of these serologic assays are standardized with regard to antigen target, assay parameters, and interpretation of negative results. Recently, in a joint effort, the Centers for Disease Control and Prevention and the Association of State and Territorial Public Health Laboratory Directors standardized interpretation of IgM and IgG Western blots in the United States (Table 3) [19]. A two step protocol was recommended, with equivocal or positive ELISA or IFA results confirmed by immunoblot. IgM and IgG Western blot testing was recommended for clinical syndromes of 4 weeks or less. Positive IgM Western blot alone was not recommended to diagnose patients with a longer clinical syndrome. These recommendations are currently under review, and are likely to be modified. One problem has been the lack of a standardized monoclonal panel for use in Western blot. In Europe, there are as yet no defined criteria for performance, analysis, or interpretation of Western blots. Recommendations for serologic testing for Lyme disease Recommendations for serologic testing for Lyme disease Several promising serologic advances are under evaluation. They include the use of antibody assays employing purified recombinant antigens (such as the p23 OspC, the p31 OspA, the p41 flagellin, and p83) rather than whole cell B. burgdorferi preparations [20–22]. Sensitivity of recombinant protein based assays also varies depending on the chosen antigen, duration of infection, and spirochete strain. In certain cases, such as early infection, recombinant based ELISA is more sensitive than either whole cell ELISA or immunoblot [20]. More recently recombinant chimeric proteins have been created, combining critical epitopes from several immunodominant antigens [23]. Another experimental assay uses flow cytometry to detect borreliacidal antibodies which may be more clinically relevant [24]. However, a recent study found quite low sensitivity (11%) for this assay in active Lyme disease [14]. Finally, anti-B. burgdorferi antibodies can be detected within isolated immune complexes using both ELISA and immunoblot techniques [25]. Preliminary studies suggest that some seronegative Lyme disease cases have complexed antibodies, and that complexed antibodies may be a marker of active infection. The inability to use culture to document infection has led to a search for an alternative technique to document presence of detectable spirochete components. To date, detection of B. burgdorferi nucleic acid has had the most widespread use. A number of PCR assays for Lyme disease are offered commercially. These assays are still considered experimental, because they are not standardized with regard to primers, DNA extraction techniques, assay parameters, or optimal sample use. Although PCR is extremely sensitive, and in theory can detect a single spirochete within a sample, it is susceptible to both false positive and false negative results. PCR has been applied to a number of clinical samples in Lyme disease, including tissue such as skin, and body fluids such as blood, urine, synovial fluid, vitreous humor, and CSF. PCR applied to skin punch biopsies of EM or the acrodermatitis chronica atrophicans lesion, which occurs with late stage disseminated infection in Europe, has an 80–90% yield [26]. Both plasmid encoded genes, and genomic DNA targets, have been used. There are reports of greater analytic sensitivity with plasmid as opposed to genomic targets [27]. Different primers have been used, including primers to identify the genes which code for OspA, OspB, flagellin, 16S rRNA, 23S rRNA, circular plasmid segments, and chromosomal segments. When proper controls are utilized along with appropriate techniques to minimize contamination, positive PCR provides evidence of organism DNA and thus recent infection. PCR appears to be highly sensitive and specific when applied to synovial fluid of Lyme arthritis patients (85% positivity rates), and punch biopsies of skin lesions. There is less experience with using PCR on blood and urine samples to document infection reliably. A recent study of EM patients reported that 18% of plasma samples were PCR positive [28]. PCR applied to CSF to evaluate neurologic Lyme disease has been quite disappointing (Table 4) [29–41]. Although earlier reports using small numbers of patients had high rates of positive results, more recent series have found that sensitivity is not much greater than 40% in clearcut neurologic patients. It is likely that the next few years will see improvements in PCR assays for Lyme disease, particularly those applied to CSF. PCR applied to CSF for neurologic Lyme disease PCR applied to CSF for neurologic Lyme disease Antigen assays are reported as helpful direct diagnosis techniques distinct from nucleic acid detection. It is postulated that these assays detect shed or blebbed outer membrane material, rather than whole organism. There is a research assay which detects OspA in CSF by ELISA and immunoblot [42]. There is also a commercial assay that has been applied to urine in particular. At the present time antigen assays in Lyme disease remain experimental. Pitfalls in the diagnosis of Lyme disease can be divided into those which lead to failure to diagnose patients, and those which lead to inappropriate diagnosis (Table 5). EM is the only pathognomonic clinical marker for Lyme disease. However, in 20–40% of cases EM either is not noted or does not occur. Although Lyme disease can cause a wide array of clinical syndromes, it particularly causes skin, joint, heart, eye, and nervous system problems with very suggestive syndromes associated with both early dissemination of spirochetes as well as with late stage disseminated infection (Table 1). These syndromes, occurring in a patient with potential exposure to B. burgdorferi infected ticks, should trigger an evaluation for Lyme disease. At the same time, the complete clinical spectrum due to Lyme disease has not been defined. For each of the characteristic body organs involved in this infection, there have been limited case reports that appear to be fairly well documented of unusual syndromes due to Lyme disease. Therefore the clinician needs to have a reasonable index of suspicion for possible Lyme disease when patients exposed to endemic areas develop otherwise unexplained disease. Finally, there are seronegative Lyme disease patients. Approximately 7% of infections are associated with seronegativity [43, 44], although some practitioners within endemic areas have claimed rates as high as 40%. Early antibiotics, as the humoral immune response is developing, can blunt antibody production. Up to 27% of EM patients who receive early treatment never seroconvert [13]. It is not known whether all seronegative Lyme disease patients can be attributed to antibiotic exposure, or whether there are some patients who do not form detectable antibody levels due to host immune factors or organism factors. Pitfalls in the diagnosis of Lyme disease Pitfalls in the diagnosis of Lyme disease Patients are also incorrectly diagnosed as having Lyme disease. This infection is geographically restricted to areas inhabited by Ixodid hard body ticks. Even in areas with strains of this tick, enzootic cycles cannot always be demonstrated. Although I. scapularis is widely distributed in the southern United States, cryptic enzootic sites have only been identified in a few states, and no tick transmitted B. burgdorferi infections have been documented. An antibiotic responsive illness that duplicates EM is reported following tick bite in the state of Missouri. This may reflect infection by a unique tick borne spirochete or other organism, distinct from B. burgdorferi. Not every red, circular rash is EM [45]. Conditions such as cellulitis, contact dermatitis, drug reaction, erythema multiforme, tinea, hypersensitivity reaction to arthropod bite, urticaria, serum sickness and granuloma annulare can mimic EM and lead to misdiagnosis. Although Lyme disease often produces fatigue/malaise, arthralgias, myalgias, headache, stiff neck, back pain, and cognitive complaints, these are all very nonspecific symptoms that can be due to a myriad of other conditions. Subjective multisymptom syndromes cannot be considered as characteristic of Lyme disease. Often patients with these complaints are presumptively diagnosed as having Lyme disease, despite lack of supporting laboratory data. It has been increasingly appreciated that ticks can carry multiple different organisms. Ixodid ticks can be infected with parasites (Babesia), rickettsiae (the agent of human granulocytic Ehrlichiosis, HGE), viruses (tick borne Flavivirus) and bacteria (a Bartonella species). Any patient exposed to ticks may be infected, or even coinfected, with a pathogen other than B. burgdorferi. There is evidence for coinfection with Babesia or the HGE agent in 4–20% of Lyme disease patients. Coinfection with a second agent can lead to a more pronounced spirochetemia, and more severe illness [46]. Not all inoculations with B. burgdorferi lead to clinical illness and Lyme disease. In fact, studies suggest that the asymptomatic infection rate may be as high as 50%[47, 48]. This may relate to infection with nonpathogenic strains, a small inoculum, or an effective early host immune response. The clinical consequences of this fact mean that within endemic areas a certain proportion of individuals have true seropositivity, but are otherwise healthy and do not have Lyme disease. These people can develop other conditions which bring them to medical attention. They may have nonspecific complaints (such as headache) that are not due to Lyme disease despite their seropositive status. Serologic testing for Lyme disease is less than optimal, with inter- and intralaboratory variability. Commercial laboratories using prepackaged kits are particularly susceptible to inaccuracies. Studies indicate there is overuse of serologic testing, with indiscriminate ordering of the test under inappropriate circumstances [49]. Such overuse markedly lowers the positive predictive value of the assay. Patients may have a false positive serology. This is most likely to occur with IFA or ELISA assays, where false positive reactions may be due to other spirochetal infections (including oral treponemes, as well as the pathogenic spirochetes which produce syphilis, relapsing fever, and leptospirosis), other infections including bacteria (up to 25% of endocarditis cases) and viruses such as Epstein-Barr virus, high autoantibody titers, or hypergammaglobulinemia. In one study more than 90% of positive ELISAs were found to be false positives [50]. Adsorption with Escherichia coli antigens removed these cross reactive antibodies. A number of B. burgdorferi proteins are immunodominant yet not unique to B. burgdorferi. They include a series of heat shock proteins as well as the p41 flagellin. A small proportion (<4%) of non Lyme disease subjects may even show reactivity to presumably specific B. burgdorferi proteins, such as the 31 kDa OspA. Even Western blot assays are susceptible to interpretation problems. Multiple antigen cross reactivity on Western blot can be seen in rheumatoid arthritis patients [51]. Attempts to apply densitometry to analyze Western blot results have had problems with regard to failure to detect faint bands, reading single broad bands as two bands, and failure to distinguish bands of similar molecular mass [52]. Finally, inappropriate diagnoses of Lyme disease have been made based on a false positive PCR or urine antigen assay. When the only supporting evidence for the diagnosis of Lyme disease is provided by a single assay, it raises the possibility that the single assay is a false positive result.
No takes yet. Share an insight, caveat, or question.
Patricia K. Coyle (2006) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: