Journal of Autoimmunity xxx (2014) 1e34
Contents lists available at ScienceDirect
Journal of Autoimmunity journal homepage: www.elsevier.com/locate/jautimm
Review
Lyme disease: A rigorous review of diagnostic criteria and treatment Andrea T. Borchers a, Carl L. Keen b, Arthur C. Huntley c, M. Eric Gershwin a, * a
Division of Rheumatology, Allergy and Clinical Immunology, University of California at Davis, Davis, CA 95616, USA Department of Nutrition, University of California at Davis, Davis, CA 95616, USA c Department of Dermatology, University of California at Davis, Davis, CA 95616, USA b
a r t i c l e i n f o
a b s t r a c t
Article history: Received 8 September 2014 Received in revised form 15 September 2014 Accepted 16 September 2014 Available online xxx
Lyme disease was originally identified in Lyme, Connecticut, based upon an unusual cluster of what appeared to be patients with juvenile rheumatoid arthritis. It was subsequently identified as a new clinical entity originally called Lyme arthritis based on the observation that arthritis was a major clinical feature. However, Lyme arthritis is now called Lyme disease based upon the understanding that the clinical features include not only arthritis, but also potential cardiac, dermatologic and neurologic findings. Lyme disease typically begins with an erythematous rash called erythema migrans (EM). Approximately 4e8% of patients develop cardiac, 11% develop neurologic and 45e60% of patients manifest arthritis. The disease is transmitted following exposure to a tick bite containing a spirochete in a genetically susceptible host. There is considerable data on spirochetes, including Borrelia burgdorferi (Bb), the original bacteria identified in this disease. Lyme disease, if an organism had not been identified, would be considered as a classic autoimmune disease and indeed the effector mechanisms are similar to many human diseases manifest as loss of tolerance. The clinical diagnosis is highly likely based upon appropriate serology and clinical manifestations. However, the serologic features are often misinterpreted and may have false positives if confirmatory laboratory testing is not performed. Antibiotics are routinely and typically used to treat patients with Lyme disease, but there is no evidence that prolonged or recurrent treatment with antibiotics change the natural history of Lyme disease. Although there are animal models of Lyme disease, there is no system that faithfully recapitulates the human disease. Further research on the effector mechanisms that lead to pathology in some individuals should be further explored to develop more specific therapy. © 2014 Elsevier Ltd. All rights reserved.
Keywords: Lyme disease Spirochetes Autoimmunity Diagnostic criteria
1. Introduction Lyme disease takes its name from the town of Lyme in Connecticut, U.S.A., where an unusual cluster of cases with an initial diagnosis of juvenile rheumatoid arthritis occurred in the mid 1970s [1]. Closer examination of this community revealed that the recurrent attacks of arthritis in these patients appeared to be a new clinical entity, originally called Lyme arthritis (LA). After it became
Abbreviations: ACA, acrodermatitis chronica atrophicans; AI, antibody index; AV, atrioventricular; Bb, Borrelia burgdorferi; CDC, Centers for Disease Control and Prevention; CNS, central nervous system; CSF, cerebral spinal fluid; EIA, enzyme immunoassay; DbpA, decorin-binding protein A; EM, erythema migrans; LA, Lyme arthritis; LNB, Lyme neuroborreliosis; PCR, polymerase chain reaction. * Corresponding author. Division of Rheumatology, Allergy and Clinical Immunology, University of California at Davis, School of Medicine, 451 Health Sciences Drive, Suite 6510, Davis, CA 95616, USA. Tel.: þ1 530 752 2884; fax: þ1 530 752 4669. E-mail address: [email protected] (M.E. Gershwin).
clear that arthritis constituted only one of the late manifestations of a multisystem disease that included dermatological, neurological and cardiac manifestations [2], the name was changed to Lyme disease. Early prospective studies in the US revealed the natural history of this disease in untreated patients. Lyme often begins with an erythematous rash, consistent with an entity called erythema migrans (EM) or erythema chronicum migrans that had first been described in 1909 in Europe [3]. Some patients (20%) experience no further signs and symptoms, the remainder may develop neurological (11%), cardiac (4e8%) or rarely ocular abnormalities, and 45e60% eventually develop arthritis [4,5]. Not only EM, but also a meningoradiculoneuritis had long been known to occur after tick bites in Europe and to be associated with each other [6]. In 1982 Burgdorfer et al. [7] isolated a previously unidentified spirochete from deer ticks (Ixodes dammini, now Ixodes scapularis) (Fig. 1). Their and other data demonstrated that this was the etiological agent of Lyme disease [7e9], which was named Borrelia burgdorferi (Bb) in honor of one of its original discoverers [10]. These bacteria
http://dx.doi.org/10.1016/j.jaut.2014.09.004 0896-8411/© 2014 Elsevier Ltd. All rights reserved.
Please cite this article in press as: Borchers AT, et al., Lyme disease: A rigorous review of diagnostic criteria and treatment, Journal of Autoimmunity (2014), http://dx.doi.org/10.1016/j.jaut.2014.09.004
2
A.T. Borchers et al. / Journal of Autoimmunity xxx (2014) 1e34
Fig. 1. Ixodes scapularis, the primary vector for Lyme disease in eastern North America. Photo credit: A new view on lyme disease: rodents hold the key to annual risk. Gross L, PLoS Biology Vol. 4/6/2006, e182. http://dx.doi.org/10.1371/journal.pbio.0040182 (Creative Commons Attribution 2.5 Generic license).
belong to the phylum Spirochetes, have an outer membrane surrounding a protoplasmic cylinder, are long (10e30 mm), but thin (0.18e0.25 mm in diameter), irregularly coiled, and highly motile due to a 7e11 flagella inserted at each end of the cell and contained in the periplasmic space. They have a small linear chromosome and a variable number of linear and circular plasmids (up to 21 plasmids, 12 of which are linear, while 9 are circular) [11]. The genome is highly complex, with almost 8% of all open reading frames encoding lipoproteins and another 6% devoted to motility and chemotaxis. Since open reading frames in the Borrelia genome often lack significant homology with previously annotated genes, the function of a considerable portion of their products remains to be elucidated. Interestingly, if an organism had not been identified, Lyme would be considered as a classic autoimmune disease with similarities to other human diseases with loss of tolerance and seemingly common effector mechanisms [12e15]. Spirochetes with similar morphology, protein profile and antigenic determinants were detected in Ixodes ricinus ticks from Switzerland and Ixodes pacificus ticks from Oregon [7,16] and subsequently in Ixodes persulcatus ticks in Russia [17]. Genotyping subsequently established that Lyme disease in Europe was caused not only by Bb sensu stricto, but also by two genotypically distinct species, which were named Borrelia afzelii and Borrelia garinii and subsumed under the B. burgdorferi sensu lato complex (here Borrelia species abbreviated as Borrelia spp.), which now contains at least 20 confirmed or proposed genomic species (see Table 1). Of these, Borrelia spielmanii, Borrelia bavariensis, Borrelia bissettii, Borrelia lusitaniae, Borrelia americana, Borrelia andersonii and Borrelia valaisiana also have pathogenic potential (see Table 1 for the available evidence). Of particular note, it had long been held that Bb is the only species causing Lyme disease in the US, but PCR-based detection of both B. americana and B. andersonii has been reported in blood or skin of a number of US patients, some of them with EM and/or other manifestations of Lyme disease [18,19]. In addition, B. bissettii, which has been isolated from the cerebral spinal fluid (CSF) of a European patient with Lyme disease [20], has also been detected by PCR in serum of three Californian subjects, but details on whether this was associated with any disease manifestations are not available [21]. It should be kept in mind, however, that detection by PCR or even isolation by culture of a
particular species of Borrelia does not prove its pathogenicity since it is not uncommon for patients to simultaneously harbor two or more species [22e25]. While direct PCR of tissue samples appears to be more sensitive than culture-based typing, it does not reliably amplify all strains identified by culture [26]. Borrelia spp. are maintained in a complex enzootic cycle involving one or more vertebrate host reservoir species and one or more tick species. The major vectors transmitting pathogenic Borrelia spp. all belong to the hard-bodied ticks of the genus Ixodes, with I. scapularis transmitting Bb in the East, mid-Atlantic and upper Midwest of the US and I. pacificus transmitting Bb and B. bissettii in the Western parts of the US, whereas I. ricinus is the major European vector for Bb, B. afzelii, and B. garinii along with other potentially pathogenic and non-pathogenic Borrelia spp., and I. persulcatus, whose geographical distribution partially overlaps with I. ricinus, is the main vector of B. afzelii and B. garinii in eastern regions of Europe and in Asia. Note, however, that there have been isolated reports implicating other tick genera in transmitting Bb and other Borrelia spp., including Amblyomma americanum in Florida and Georgia [18] and ticks of the genus Dermacentor in Spain [27]. The biology of Borrelia spp. is closely tied to the lifecycle of the tick vector. Ticks take one blood meal during each of the three stages of their life cycle, i.e., as larvae, nymphs, and adults. There is little to no transovarial transmission of Borrelia spp. infection, therefore ticks become infected when larvae or nymphs feed on an infected host, infection is then maintained transstadially and can be passed on to the host providing the next blood meal, including humans. 2. Epidemiology of Lyme disease Almost all confirmed cases of Lyme disease have occurred in the Northern hemisphere, the majority coming from the United States and Europe (including the European part of Russia), far fewer from Asia, and some from Northern Africa [28]. In the US, Lyme disease has been a nationally notifiable disease since 1991 and the number of reported cases rose steadily from 9908 in 1992 to a peak of almost 30,000 confirmed cases in 2009 [29]. Since then, the incidence appears to have stabilized below 25,000 confirmed cases annually. There are numerous indications that underreporting represents a considerable problem [30e32], and scientists from the US Centers for Disease Control and Prevention (CDC) actually estimate that the true number of cases may be closer to 300,000 each year [33]. On the other hand, it is equally clear that there is a considerable over-diagnosis of Lyme disease based upon misinterpretation of the serology and/or over-zealous interpretation of internet information [34]. The Northeast, mid-Atlantic region and upper Midwest are the primary areas of endemicity, and just 10 states (Connecticut, Delaware, Massachusetts, Maryland, Minnesota, New Jersey, New York, Pennsylvania, Rhode Island and Wisconsin) account for 93% of annual cases. But even within these endemic areas, the incidence varies considerably, exceeding 500/105 in certain counties [35]. Among the cases reported to the CDC, AfricaneAmericans are significantly underrepresented [36,37]. This may be due to a lower risk of exposure since African Americans more rarely live in highly endemic rural areas. However, the results of a detailed analysis of cases in an endemic area of Maryland revealed that area of residence alone could not fully account for the different incidence rates of African Americans compared to European Americans [36]. Instead, a significantly lower rate of EM and an excess risk of arthritis among African Americans suggested that underrecognition of EM e or unawareness of its clinical significance e resulted in a disproportionate number of late manifestations in this population. Unlike many other immune-mediated diseases, there is
Please cite this article in press as: Borchers AT, et al., Lyme disease: A rigorous review of diagnostic criteria and treatment, Journal of Autoimmunity (2014), http://dx.doi.org/10.1016/j.jaut.2014.09.004
Borrelia species
Vector
Geographical distribution
Cultured from blood
Cultured from EM
Isolated from other t issue
Positive PCR
Pathogenicity in mice
B. afzelii
I. ricinus, I. persulcatus, I. hexagonus
Asia, Europe
Rarely [104,105]
[20,111,167,168,391]
Synovial fluid or tissue, CSF [20,24,110,113,385] ACA [395,396] BL [184]
CSF, blood [104,388]; synovial fluid [602,603]; heart valve [366]
[604]
B. americana B. andersonii B. bavariensis (formerly OspA serotype 4 of B. garinii) B. bissettiia
I. pacificus, I. minor I. dentatus I. ricinus
USA USA Europe
[391]
I. I. I. I. I. I. I. I. I. I. I. I. I. I.
Europe, USA
[605]
CSF [20,605]; BL [184]
[101,103,405]
[20,80,164,166]
Synovial fluid, CSF [20,24,112,385] Heart [20] ACA [395,396]
Rarely [105,110]
[20,111,166,168]
Synovial fluid or tissue, CSF [20,24,111,385] ACA [110,395,396] BL [394]
B. burgdorferi sensu stricto
B. californiensis B. carolinensis B. garinii
ricinus, I. scapularis, pacificus, I. minor, affinis ricinus, I. scapularis, pacificus, spinipalpis, I. hexagonis, affinis, I. minor, muris pacificus, I. jellisonii, spinipalpis minor ricinus, I. persulcatus, nipponensis, uriae
Europe, USA
USA Asia, Europe
I. ricinus I. ovatus
Japan
B. kurtenbachiia B. lusitaniae
I. scapularis, (I. ricinus) I. ricinus
Europe, USA Europe, North Africa
B. B. B. B. B.
I. I. I. I. I. I. I.
China Europe Japan Japan Asia, Europe
B. yangtze Genomospecies 2
ovatus ricinus tanuki turdus ricinus, I. granulatus, nipponensis, columnae
I. granulatus, I. nipponensis, Haemaphysalis longicornis I. pacificus, I. spinipalpis
Serum [21]; blood [606]; cardiac tissue [607] CSF, blood [104,220,388]; synovial fluid or tissue [118,268,384,578,602,603]
[608]
[379,380,609,610]
USA
B. finlandensis B. japonica
sinica spielmanii tanukii turdi valaisiana (VS116)
Blood or skin [18,19] Blood or skin [18,19] EM [412]
CSF, blood [104,388]; synovial fluid [602,603]
[369,611,612]
In inbred mice [613], but not in outbred mice [614] Weak [615], absent [616]; Yes, but not arthritogenic [611,618]
CSF or skin [605] [617]
A.T. Borchers et al. / Journal of Autoimmunity xxx (2014) 1e34
Please cite this article in press as: Borchers AT, et al., Lyme disease: A rigorous review of diagnostic criteria and treatment, Journal of Autoimmunity (2014), http://dx.doi.org/10.1016/j.jaut.2014.09.004
Table 1 Borrelia spp., their vectors, and evidence for their pathogenicity (culture from human tissue, demonstration of their DNA in human tissue, pathogenicity in mice). (List of species and vectors based on [600,601]).
[20,145,382,619,620]
Skin [23]; blood [621]; CSF [622]; B. valaisiana-related species in blood [234]
[611]
China USA
a
B. bissettii and B. kurtenbachii (initially called Bb strain 25015) were originally both included in the B. bissettii genospecies. Therefore, it is not always possible in older studies to determine which patient isolates belong to which of these novel species.
3
4
A.T. Borchers et al. / Journal of Autoimmunity xxx (2014) 1e34
no obvious sex bias (see below) nor is there any significant data on concordance in twins or within large family pedigrees [38e40]. In most of Europe, Lyme disease is not a reportable disease, and until recently there were no standardized case definitions [41]. Therefore, the available data are less reliable and difficult to compare because a variety of active and passive surveillance methods were used to arrive at the available estimates. Nonetheless, it is clear that Lyme borreliosis is highly endemic in much of Europe, with the highest incidence reported from southern Sweden, Lithuania, Germany, Austria, and Slovenia. The total number of annual cases in Europe is estimated to be ~3-fold higher than the number of the cases reported to the CDC [28]. A rise in the incidence of Lyme borreliosis similar to that observed in the USA has been reported from some European regions [42e44]. Increased awareness of the disease almost certainly accounts for some of this rise, but climate change with a subsequent expansion of the territory of vector ticks, along with changes in land use, resulting in increases in reservoir hosts for ticks, and changes in residential distribution and human recreational behavior also contribute; there is also the potential error of a false positive diagnosis [42,45e49]. While the CDC data in the USA suggest that males constitute slightly more than half of the reported cases (54% overall and 61% of pediatric cases) [37], data from Europe often suggest a slight to marked female preponderance [28,42,50], but an even distribution has been reported from some areas [51e53]. The age distribution among cases with Lyme disease is generally bimodal, with the highest incidence rates seen in children aged 5e9 years and in adults >50 years of age in both the US and Europe [28,35,37,42,51,54]. 3. Risk of infection after a tick bite and factors influencing it 3.1. Risk of infection In areas that are endemic for Lyme disease, the risk of Borrelia infection after the bite of an infected tick is only 1% and 3% in the US [55e57], and 3e12% in Europe [58e62]. However, at least onefourth, and sometimes more than half, of European subjects seroconvert without developing clinical manifestations of Lyme disease [58e61]. In a cohort of highly exposed individuals, almost all of the seroconverters remained asymptomatic [63]. This is the more remarkable since a substantial proportion (30%) of European patients with EM never develop seropositivity [64e66]. In contrast, asymptomatic seroconversion is quite rare in the US [67,68]; the mechanism for this is unclear and further studies are suggested. 3.2. Environmental and behavioral factors Risk factors for Borrelia spp. infection are time spent in endemic areas, time spent outdoors, and factors associated with tick density [69e71]. Therefore, the most effective way to minimize the risk of Borrelia infection is to avoid tick habitats, such as wooded areas, shrubs, tall grass and particularly the edges where these different types of vegetation meet. Wearing protective clothing, using DEET insect repellents, conducting a visual inspection every night to check for attached ticks, and prompt removal of attached ticks are other important preventive measures. Of note, a previous diagnosis of Lyme disease was identified, along with age, as an independent predictor of Bb infection in a study that investigated both environmental and behavioral risk factors [70]. This probably reflects that the same population subgroups stays at risk of repeated infections because of the same behavioral or environmental factors that increased their original risk. The role of environment versus heritability is well established in human immune diseases [72]. The
high rates of reinfection observed in several cohort studies confirm that a previous history of Lyme disease constitutes a risk factor for reinfection [73e75]. This also means that previous infection does not protect from reinfection. Mice do develop immunity, but it is strain-specific [76], and there are recent empirical data suggesting that humans also develop strain-specific immunity that lasts for 6 years [77]. This could explain the extraordinarily low rate of symptomatic infection (~2%) in a Swiss cohort with frequent exposure to ticks most likely containing a wide variety of strains [63]. It is important to note that Lyme disease follows the pattern of a classic infection, much like that of post-streptococcal glomerulonephritis or rheumatic heart disease. It is not related to trauma, exercise, automobile accidents, or weather. It is strictly a host pathogen relationship. Indeed, both in the case of Lyme as well as any other human immunologically mediated disease, the role for environmental associated factors has reasonably defined criteria [78]. 3.3. Season and tick developmental stage In both the Eastern US and most parts of Europe, peak onset of Lyme disease occurs during the summer months, when nymphal abundance is highest and nymphs are actively questing, whereas few cases occur during the late fall and the first few months of the year when adult ticks actively seek a host [51,54,64,79,80]. However, in some European regions, a secondary peak in the EM incidence is seen late in the fall (October, November), suggesting that actively questing adults also contribute to human Borrelia infections [64,81], and in Russia, Borrelia spp. are transmitted almost exclusively by adult I. persulcatus ticks [62]. Infection rates are higher in adult than in nymphal ticks [20,79,82], but nymphs are more abundant and are much smaller and more difficult to detect and, consequently, are more likely to stay attached longer [59]. Duration of tick attachment has been identified as an important risk factor for infection, at least in the US. 3.4. Duration of tick attachment Spirochetes reside in the midgut of unfed ticks. Once the blood meal reaches the midgut, the spirochetes begin to replicate and then disseminate via the hemocoel to the salivary glands [83]. There are experimental data demonstrating that Bb-infected I. scapularis nymphs transfer considerable numbers of spirochetes as early as 24 h after attachment [84]. Nonetheless, in experimental animals, infection rarely occurs within 24 h of tick attachment, and peak infection rates are not reached until 48e72 h of attachment [85e90]. The time required for infection to occur appears to depend on the target host species [85,86] and on the infecting strain of Bb [88]. In accordance with these findings, the risk of infection for residents of endemic areas of the US was found to be minimal if ticks stayed attached for
Contents lists available at ScienceDirect
Journal of Autoimmunity journal homepage: www.elsevier.com/locate/jautimm
Review
Lyme disease: A rigorous review of diagnostic criteria and treatment Andrea T. Borchers a, Carl L. Keen b, Arthur C. Huntley c, M. Eric Gershwin a, * a
Division of Rheumatology, Allergy and Clinical Immunology, University of California at Davis, Davis, CA 95616, USA Department of Nutrition, University of California at Davis, Davis, CA 95616, USA c Department of Dermatology, University of California at Davis, Davis, CA 95616, USA b
a r t i c l e i n f o
a b s t r a c t
Article history: Received 8 September 2014 Received in revised form 15 September 2014 Accepted 16 September 2014 Available online xxx
Lyme disease was originally identified in Lyme, Connecticut, based upon an unusual cluster of what appeared to be patients with juvenile rheumatoid arthritis. It was subsequently identified as a new clinical entity originally called Lyme arthritis based on the observation that arthritis was a major clinical feature. However, Lyme arthritis is now called Lyme disease based upon the understanding that the clinical features include not only arthritis, but also potential cardiac, dermatologic and neurologic findings. Lyme disease typically begins with an erythematous rash called erythema migrans (EM). Approximately 4e8% of patients develop cardiac, 11% develop neurologic and 45e60% of patients manifest arthritis. The disease is transmitted following exposure to a tick bite containing a spirochete in a genetically susceptible host. There is considerable data on spirochetes, including Borrelia burgdorferi (Bb), the original bacteria identified in this disease. Lyme disease, if an organism had not been identified, would be considered as a classic autoimmune disease and indeed the effector mechanisms are similar to many human diseases manifest as loss of tolerance. The clinical diagnosis is highly likely based upon appropriate serology and clinical manifestations. However, the serologic features are often misinterpreted and may have false positives if confirmatory laboratory testing is not performed. Antibiotics are routinely and typically used to treat patients with Lyme disease, but there is no evidence that prolonged or recurrent treatment with antibiotics change the natural history of Lyme disease. Although there are animal models of Lyme disease, there is no system that faithfully recapitulates the human disease. Further research on the effector mechanisms that lead to pathology in some individuals should be further explored to develop more specific therapy. © 2014 Elsevier Ltd. All rights reserved.
Keywords: Lyme disease Spirochetes Autoimmunity Diagnostic criteria
1. Introduction Lyme disease takes its name from the town of Lyme in Connecticut, U.S.A., where an unusual cluster of cases with an initial diagnosis of juvenile rheumatoid arthritis occurred in the mid 1970s [1]. Closer examination of this community revealed that the recurrent attacks of arthritis in these patients appeared to be a new clinical entity, originally called Lyme arthritis (LA). After it became
Abbreviations: ACA, acrodermatitis chronica atrophicans; AI, antibody index; AV, atrioventricular; Bb, Borrelia burgdorferi; CDC, Centers for Disease Control and Prevention; CNS, central nervous system; CSF, cerebral spinal fluid; EIA, enzyme immunoassay; DbpA, decorin-binding protein A; EM, erythema migrans; LA, Lyme arthritis; LNB, Lyme neuroborreliosis; PCR, polymerase chain reaction. * Corresponding author. Division of Rheumatology, Allergy and Clinical Immunology, University of California at Davis, School of Medicine, 451 Health Sciences Drive, Suite 6510, Davis, CA 95616, USA. Tel.: þ1 530 752 2884; fax: þ1 530 752 4669. E-mail address: [email protected] (M.E. Gershwin).
clear that arthritis constituted only one of the late manifestations of a multisystem disease that included dermatological, neurological and cardiac manifestations [2], the name was changed to Lyme disease. Early prospective studies in the US revealed the natural history of this disease in untreated patients. Lyme often begins with an erythematous rash, consistent with an entity called erythema migrans (EM) or erythema chronicum migrans that had first been described in 1909 in Europe [3]. Some patients (20%) experience no further signs and symptoms, the remainder may develop neurological (11%), cardiac (4e8%) or rarely ocular abnormalities, and 45e60% eventually develop arthritis [4,5]. Not only EM, but also a meningoradiculoneuritis had long been known to occur after tick bites in Europe and to be associated with each other [6]. In 1982 Burgdorfer et al. [7] isolated a previously unidentified spirochete from deer ticks (Ixodes dammini, now Ixodes scapularis) (Fig. 1). Their and other data demonstrated that this was the etiological agent of Lyme disease [7e9], which was named Borrelia burgdorferi (Bb) in honor of one of its original discoverers [10]. These bacteria
http://dx.doi.org/10.1016/j.jaut.2014.09.004 0896-8411/© 2014 Elsevier Ltd. All rights reserved.
Please cite this article in press as: Borchers AT, et al., Lyme disease: A rigorous review of diagnostic criteria and treatment, Journal of Autoimmunity (2014), http://dx.doi.org/10.1016/j.jaut.2014.09.004
2
A.T. Borchers et al. / Journal of Autoimmunity xxx (2014) 1e34
Fig. 1. Ixodes scapularis, the primary vector for Lyme disease in eastern North America. Photo credit: A new view on lyme disease: rodents hold the key to annual risk. Gross L, PLoS Biology Vol. 4/6/2006, e182. http://dx.doi.org/10.1371/journal.pbio.0040182 (Creative Commons Attribution 2.5 Generic license).
belong to the phylum Spirochetes, have an outer membrane surrounding a protoplasmic cylinder, are long (10e30 mm), but thin (0.18e0.25 mm in diameter), irregularly coiled, and highly motile due to a 7e11 flagella inserted at each end of the cell and contained in the periplasmic space. They have a small linear chromosome and a variable number of linear and circular plasmids (up to 21 plasmids, 12 of which are linear, while 9 are circular) [11]. The genome is highly complex, with almost 8% of all open reading frames encoding lipoproteins and another 6% devoted to motility and chemotaxis. Since open reading frames in the Borrelia genome often lack significant homology with previously annotated genes, the function of a considerable portion of their products remains to be elucidated. Interestingly, if an organism had not been identified, Lyme would be considered as a classic autoimmune disease with similarities to other human diseases with loss of tolerance and seemingly common effector mechanisms [12e15]. Spirochetes with similar morphology, protein profile and antigenic determinants were detected in Ixodes ricinus ticks from Switzerland and Ixodes pacificus ticks from Oregon [7,16] and subsequently in Ixodes persulcatus ticks in Russia [17]. Genotyping subsequently established that Lyme disease in Europe was caused not only by Bb sensu stricto, but also by two genotypically distinct species, which were named Borrelia afzelii and Borrelia garinii and subsumed under the B. burgdorferi sensu lato complex (here Borrelia species abbreviated as Borrelia spp.), which now contains at least 20 confirmed or proposed genomic species (see Table 1). Of these, Borrelia spielmanii, Borrelia bavariensis, Borrelia bissettii, Borrelia lusitaniae, Borrelia americana, Borrelia andersonii and Borrelia valaisiana also have pathogenic potential (see Table 1 for the available evidence). Of particular note, it had long been held that Bb is the only species causing Lyme disease in the US, but PCR-based detection of both B. americana and B. andersonii has been reported in blood or skin of a number of US patients, some of them with EM and/or other manifestations of Lyme disease [18,19]. In addition, B. bissettii, which has been isolated from the cerebral spinal fluid (CSF) of a European patient with Lyme disease [20], has also been detected by PCR in serum of three Californian subjects, but details on whether this was associated with any disease manifestations are not available [21]. It should be kept in mind, however, that detection by PCR or even isolation by culture of a
particular species of Borrelia does not prove its pathogenicity since it is not uncommon for patients to simultaneously harbor two or more species [22e25]. While direct PCR of tissue samples appears to be more sensitive than culture-based typing, it does not reliably amplify all strains identified by culture [26]. Borrelia spp. are maintained in a complex enzootic cycle involving one or more vertebrate host reservoir species and one or more tick species. The major vectors transmitting pathogenic Borrelia spp. all belong to the hard-bodied ticks of the genus Ixodes, with I. scapularis transmitting Bb in the East, mid-Atlantic and upper Midwest of the US and I. pacificus transmitting Bb and B. bissettii in the Western parts of the US, whereas I. ricinus is the major European vector for Bb, B. afzelii, and B. garinii along with other potentially pathogenic and non-pathogenic Borrelia spp., and I. persulcatus, whose geographical distribution partially overlaps with I. ricinus, is the main vector of B. afzelii and B. garinii in eastern regions of Europe and in Asia. Note, however, that there have been isolated reports implicating other tick genera in transmitting Bb and other Borrelia spp., including Amblyomma americanum in Florida and Georgia [18] and ticks of the genus Dermacentor in Spain [27]. The biology of Borrelia spp. is closely tied to the lifecycle of the tick vector. Ticks take one blood meal during each of the three stages of their life cycle, i.e., as larvae, nymphs, and adults. There is little to no transovarial transmission of Borrelia spp. infection, therefore ticks become infected when larvae or nymphs feed on an infected host, infection is then maintained transstadially and can be passed on to the host providing the next blood meal, including humans. 2. Epidemiology of Lyme disease Almost all confirmed cases of Lyme disease have occurred in the Northern hemisphere, the majority coming from the United States and Europe (including the European part of Russia), far fewer from Asia, and some from Northern Africa [28]. In the US, Lyme disease has been a nationally notifiable disease since 1991 and the number of reported cases rose steadily from 9908 in 1992 to a peak of almost 30,000 confirmed cases in 2009 [29]. Since then, the incidence appears to have stabilized below 25,000 confirmed cases annually. There are numerous indications that underreporting represents a considerable problem [30e32], and scientists from the US Centers for Disease Control and Prevention (CDC) actually estimate that the true number of cases may be closer to 300,000 each year [33]. On the other hand, it is equally clear that there is a considerable over-diagnosis of Lyme disease based upon misinterpretation of the serology and/or over-zealous interpretation of internet information [34]. The Northeast, mid-Atlantic region and upper Midwest are the primary areas of endemicity, and just 10 states (Connecticut, Delaware, Massachusetts, Maryland, Minnesota, New Jersey, New York, Pennsylvania, Rhode Island and Wisconsin) account for 93% of annual cases. But even within these endemic areas, the incidence varies considerably, exceeding 500/105 in certain counties [35]. Among the cases reported to the CDC, AfricaneAmericans are significantly underrepresented [36,37]. This may be due to a lower risk of exposure since African Americans more rarely live in highly endemic rural areas. However, the results of a detailed analysis of cases in an endemic area of Maryland revealed that area of residence alone could not fully account for the different incidence rates of African Americans compared to European Americans [36]. Instead, a significantly lower rate of EM and an excess risk of arthritis among African Americans suggested that underrecognition of EM e or unawareness of its clinical significance e resulted in a disproportionate number of late manifestations in this population. Unlike many other immune-mediated diseases, there is
Please cite this article in press as: Borchers AT, et al., Lyme disease: A rigorous review of diagnostic criteria and treatment, Journal of Autoimmunity (2014), http://dx.doi.org/10.1016/j.jaut.2014.09.004
Borrelia species
Vector
Geographical distribution
Cultured from blood
Cultured from EM
Isolated from other t issue
Positive PCR
Pathogenicity in mice
B. afzelii
I. ricinus, I. persulcatus, I. hexagonus
Asia, Europe
Rarely [104,105]
[20,111,167,168,391]
Synovial fluid or tissue, CSF [20,24,110,113,385] ACA [395,396] BL [184]
CSF, blood [104,388]; synovial fluid [602,603]; heart valve [366]
[604]
B. americana B. andersonii B. bavariensis (formerly OspA serotype 4 of B. garinii) B. bissettiia
I. pacificus, I. minor I. dentatus I. ricinus
USA USA Europe
[391]
I. I. I. I. I. I. I. I. I. I. I. I. I. I.
Europe, USA
[605]
CSF [20,605]; BL [184]
[101,103,405]
[20,80,164,166]
Synovial fluid, CSF [20,24,112,385] Heart [20] ACA [395,396]
Rarely [105,110]
[20,111,166,168]
Synovial fluid or tissue, CSF [20,24,111,385] ACA [110,395,396] BL [394]
B. burgdorferi sensu stricto
B. californiensis B. carolinensis B. garinii
ricinus, I. scapularis, pacificus, I. minor, affinis ricinus, I. scapularis, pacificus, spinipalpis, I. hexagonis, affinis, I. minor, muris pacificus, I. jellisonii, spinipalpis minor ricinus, I. persulcatus, nipponensis, uriae
Europe, USA
USA Asia, Europe
I. ricinus I. ovatus
Japan
B. kurtenbachiia B. lusitaniae
I. scapularis, (I. ricinus) I. ricinus
Europe, USA Europe, North Africa
B. B. B. B. B.
I. I. I. I. I. I. I.
China Europe Japan Japan Asia, Europe
B. yangtze Genomospecies 2
ovatus ricinus tanuki turdus ricinus, I. granulatus, nipponensis, columnae
I. granulatus, I. nipponensis, Haemaphysalis longicornis I. pacificus, I. spinipalpis
Serum [21]; blood [606]; cardiac tissue [607] CSF, blood [104,220,388]; synovial fluid or tissue [118,268,384,578,602,603]
[608]
[379,380,609,610]
USA
B. finlandensis B. japonica
sinica spielmanii tanukii turdi valaisiana (VS116)
Blood or skin [18,19] Blood or skin [18,19] EM [412]
CSF, blood [104,388]; synovial fluid [602,603]
[369,611,612]
In inbred mice [613], but not in outbred mice [614] Weak [615], absent [616]; Yes, but not arthritogenic [611,618]
CSF or skin [605] [617]
A.T. Borchers et al. / Journal of Autoimmunity xxx (2014) 1e34
Please cite this article in press as: Borchers AT, et al., Lyme disease: A rigorous review of diagnostic criteria and treatment, Journal of Autoimmunity (2014), http://dx.doi.org/10.1016/j.jaut.2014.09.004
Table 1 Borrelia spp., their vectors, and evidence for their pathogenicity (culture from human tissue, demonstration of their DNA in human tissue, pathogenicity in mice). (List of species and vectors based on [600,601]).
[20,145,382,619,620]
Skin [23]; blood [621]; CSF [622]; B. valaisiana-related species in blood [234]
[611]
China USA
a
B. bissettii and B. kurtenbachii (initially called Bb strain 25015) were originally both included in the B. bissettii genospecies. Therefore, it is not always possible in older studies to determine which patient isolates belong to which of these novel species.
3
4
A.T. Borchers et al. / Journal of Autoimmunity xxx (2014) 1e34
no obvious sex bias (see below) nor is there any significant data on concordance in twins or within large family pedigrees [38e40]. In most of Europe, Lyme disease is not a reportable disease, and until recently there were no standardized case definitions [41]. Therefore, the available data are less reliable and difficult to compare because a variety of active and passive surveillance methods were used to arrive at the available estimates. Nonetheless, it is clear that Lyme borreliosis is highly endemic in much of Europe, with the highest incidence reported from southern Sweden, Lithuania, Germany, Austria, and Slovenia. The total number of annual cases in Europe is estimated to be ~3-fold higher than the number of the cases reported to the CDC [28]. A rise in the incidence of Lyme borreliosis similar to that observed in the USA has been reported from some European regions [42e44]. Increased awareness of the disease almost certainly accounts for some of this rise, but climate change with a subsequent expansion of the territory of vector ticks, along with changes in land use, resulting in increases in reservoir hosts for ticks, and changes in residential distribution and human recreational behavior also contribute; there is also the potential error of a false positive diagnosis [42,45e49]. While the CDC data in the USA suggest that males constitute slightly more than half of the reported cases (54% overall and 61% of pediatric cases) [37], data from Europe often suggest a slight to marked female preponderance [28,42,50], but an even distribution has been reported from some areas [51e53]. The age distribution among cases with Lyme disease is generally bimodal, with the highest incidence rates seen in children aged 5e9 years and in adults >50 years of age in both the US and Europe [28,35,37,42,51,54]. 3. Risk of infection after a tick bite and factors influencing it 3.1. Risk of infection In areas that are endemic for Lyme disease, the risk of Borrelia infection after the bite of an infected tick is only 1% and 3% in the US [55e57], and 3e12% in Europe [58e62]. However, at least onefourth, and sometimes more than half, of European subjects seroconvert without developing clinical manifestations of Lyme disease [58e61]. In a cohort of highly exposed individuals, almost all of the seroconverters remained asymptomatic [63]. This is the more remarkable since a substantial proportion (30%) of European patients with EM never develop seropositivity [64e66]. In contrast, asymptomatic seroconversion is quite rare in the US [67,68]; the mechanism for this is unclear and further studies are suggested. 3.2. Environmental and behavioral factors Risk factors for Borrelia spp. infection are time spent in endemic areas, time spent outdoors, and factors associated with tick density [69e71]. Therefore, the most effective way to minimize the risk of Borrelia infection is to avoid tick habitats, such as wooded areas, shrubs, tall grass and particularly the edges where these different types of vegetation meet. Wearing protective clothing, using DEET insect repellents, conducting a visual inspection every night to check for attached ticks, and prompt removal of attached ticks are other important preventive measures. Of note, a previous diagnosis of Lyme disease was identified, along with age, as an independent predictor of Bb infection in a study that investigated both environmental and behavioral risk factors [70]. This probably reflects that the same population subgroups stays at risk of repeated infections because of the same behavioral or environmental factors that increased their original risk. The role of environment versus heritability is well established in human immune diseases [72]. The
high rates of reinfection observed in several cohort studies confirm that a previous history of Lyme disease constitutes a risk factor for reinfection [73e75]. This also means that previous infection does not protect from reinfection. Mice do develop immunity, but it is strain-specific [76], and there are recent empirical data suggesting that humans also develop strain-specific immunity that lasts for 6 years [77]. This could explain the extraordinarily low rate of symptomatic infection (~2%) in a Swiss cohort with frequent exposure to ticks most likely containing a wide variety of strains [63]. It is important to note that Lyme disease follows the pattern of a classic infection, much like that of post-streptococcal glomerulonephritis or rheumatic heart disease. It is not related to trauma, exercise, automobile accidents, or weather. It is strictly a host pathogen relationship. Indeed, both in the case of Lyme as well as any other human immunologically mediated disease, the role for environmental associated factors has reasonably defined criteria [78]. 3.3. Season and tick developmental stage In both the Eastern US and most parts of Europe, peak onset of Lyme disease occurs during the summer months, when nymphal abundance is highest and nymphs are actively questing, whereas few cases occur during the late fall and the first few months of the year when adult ticks actively seek a host [51,54,64,79,80]. However, in some European regions, a secondary peak in the EM incidence is seen late in the fall (October, November), suggesting that actively questing adults also contribute to human Borrelia infections [64,81], and in Russia, Borrelia spp. are transmitted almost exclusively by adult I. persulcatus ticks [62]. Infection rates are higher in adult than in nymphal ticks [20,79,82], but nymphs are more abundant and are much smaller and more difficult to detect and, consequently, are more likely to stay attached longer [59]. Duration of tick attachment has been identified as an important risk factor for infection, at least in the US. 3.4. Duration of tick attachment Spirochetes reside in the midgut of unfed ticks. Once the blood meal reaches the midgut, the spirochetes begin to replicate and then disseminate via the hemocoel to the salivary glands [83]. There are experimental data demonstrating that Bb-infected I. scapularis nymphs transfer considerable numbers of spirochetes as early as 24 h after attachment [84]. Nonetheless, in experimental animals, infection rarely occurs within 24 h of tick attachment, and peak infection rates are not reached until 48e72 h of attachment [85e90]. The time required for infection to occur appears to depend on the target host species [85,86] and on the infecting strain of Bb [88]. In accordance with these findings, the risk of infection for residents of endemic areas of the US was found to be minimal if ticks stayed attached for
