Advertisement
Browse Subject Areas
?

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

A systematic review on the impact of gestational Lyme disease in humans on the fetus and newborn

  • Lisa A. Waddell ,

    Roles Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Project administration, Resources, Software, Writing – original draft

    lisa.waddell@canada.ca

    Affiliation National Microbiology Laboratory, Public Health Agency of Canada, Guelph, Ontario, Canada

  • Judy Greig,

    Roles Conceptualization, Data curation, Investigation, Methodology, Writing – review & editing

    Affiliation National Microbiology Laboratory, Public Health Agency of Canada, Guelph, Ontario, Canada

  • L. Robbin Lindsay,

    Roles Conceptualization, Methodology, Validation, Writing – review & editing

    Affiliation National Microbiology Laboratory, Public Health Agency of Canada, Winnipeg, Manitoba, Canada

  • Alison F. Hinckley,

    Roles Conceptualization, Methodology, Validation, Writing – review & editing

    Affiliation Centers for Disease Control and Prevention, National Center for Emerging and Zoonotic Infectious Diseases, Division of Vector-Borne Diseases, Fort Collins, Colorado, United States of America

  • Nicholas H. Ogden

    Roles Conceptualization, Formal analysis, Methodology, Project administration, Resources, Supervision, Validation, Writing – review & editing

    Affiliation National Microbiology Laboratory, Public Health Agency of Canada, Saint-Hyacinthe, Quebec, Canada

A systematic review on the impact of gestational Lyme disease in humans on the fetus and newborn

  • Lisa A. Waddell, 
  • Judy Greig, 
  • L. Robbin Lindsay, 
  • Alison F. Hinckley, 
  • Nicholas H. Ogden
PLOS
x

Abstract

Lyme disease (LD), caused by bacteria of the Borrelia burgdorferi sensu lato species complex, is the most common vector-borne disease in North America and Europe. A systematic review (SR) was conducted to summarize the global literature on adverse birth outcomes associated with gestational LD in humans. The SR followed an a priori protocol of pretested screening, risk of bias, and data extraction forms. Data were summarized descriptively and random effects meta-analysis (MA) was used where appropriate. The SR identified 45 relevant studies, 29 describing 59 cases reported as gestational LD in the United States, Europe, and Asia (1969–2017). Adverse birth outcomes included spontaneous miscarriage or fetal death (n = 12), newborn death (n = 8), and newborns with an abnormal outcome (e.g. hyperbilirubinemia, respiratory distress and syndactyly) at birth (n = 16). Only one report provided a full case description (clinical manifestations in the mother, negative outcome for the child, and laboratory detection of B. burgdorferi in the child) that provides some evidence for vertical transmission of B. burgdorferi that has negative consequences for the fetus. The results of 17 epidemiological studies are included in this SR. Prevalence of adverse birth outcomes in an exposed population (defined by the authors as: gestational LD, history of LD, tick bites or residence in an endemic area) was compared to that in an unexposed population in eight studies and no difference was reported. A meta-analysis of nine studies showed significantly fewer adverse birth outcomes in women reported to have been treated for gestational LD (11%, 95%CI 7–16) compared to those who were not treated during pregnancy (50%, 95%CI 30–70) providing indirect evidence of an association between gestational LD and adverse birth outcomes. Other risk factors investigated; trimester of exposure, length of LD during pregnancy, acute vs. disseminated LD at diagnosis, and symptomatic LD vs. seropositive women with no LD symptoms during pregnancy were not significantly associated with adverse birth outcomes. This SR summarizes evidence from case studies that provide some limited evidence for transplacental transmission of B. burgdorferi. There was inconsistent evidence for adverse birth outcomes of gestational LD in the epidemiological research, and uncommon adverse outcomes for the fetus may occur as a consequence of gestational LD. The global evidence does not fully characterize the potential impact of gestational LD, and future research that addresses the knowledge gaps may change the findings in this SR. Given the current evidence; prompt diagnosis and treatment of LD during pregnancy is recommended.

Introduction

Lyme disease (LD), the most common tick-borne disease in North America and Europe is caused by spirochetal bacteria of the Borrelia burgdorferi sensu lato species complex (also called Borreliella, but referred to herein as B. burgdorferi) [1]. The most commonly implicated B. burgdorferi species in human infections include B. burgdorferi sensu stricto in both North America and Europe and, B. afzelii and B. garinii in Europe and Asia [2]. Lyme disease was first recognized in North America in 1975 in the area of Lyme, Connecticut, as a result of an investigation into 51 cases (39 of which were children) that presented with a similar form of arthritis [3]. Early symptoms of infection include a characteristic rash (erythema-migrans, EM), fever, headache, and lethargy. If untreated, the disease may affect the heart, nervous system or manifest as arthritis.

Shortly after its discovery in 1975, the possible effects of gestational LD became an area of research interest given that transplacental infections by other species of spirochetes (e.g. Treponema pallidum; relapsing fever Borrelia species and Leptospira interrogans) are known to occur in several animal species (e.g. dogs, mice, cattle) and in humans [48]. The literature on transplacental transmission of B. burgdorferi in animals is outside the scope of this systematic review (SR). However, some adverse birth outcomes have been recorded for white-footed mice, dogs, cattle, horses, and a coyote. The most common outcomes were reproductive failure (inability to conceive) and fetal loss during pregnancy [916]. Animal model experiments identified B. burgdorferi infection in newborn beagles, indicating that transplacental transmission may occur; however, experiments involving rats, hamsters, and mice have not demonstrated this route of transmission for these species [13,1719]. Overall, there is some evidence that B. burgdorferi infection in pregnant animals can result in infection of the newborn, fetal death, and fertility issues [10,12,13].

Given the public health importance of LD and our understanding of other spirochetal diseases, a SR was conducted to identify and summarize the global evidence on “What is the evidence that gestational Lyme disease in humans causes adverse birth outcomes including congenital abnormalities?”

Methods

Review protocol, team and expertise

This SR was conducted using an a priori developed protocol that followed standard SR guidelines [20,21] and the review is reported in accordance with the preferred reporting items for systematic reviews and meta-analysis (PRISMA) statement (S1 Table) [22]. The protocol includes a list of definitions, search algorithms, title/abstract screening form, Risk of Bias tool, and data extraction forms (QA-DE). The protocol (S1 Text), list of relevant included articles (S2 Text) and dataset (S2 Table) are available in the supplementary material.

The review team comprised of individuals with multi-disciplinary expertise in epidemiology, microbiology, entomology, vector-borne diseases, veterinary public health, knowledge synthesis, and information science.

Search strategy

A pretested search algorithm, found below, was implemented in three bibliographic databases on October 16, 2017: Scopus, PubMed/MEDLINE, and Embase. No limits were placed on the search. The search terms were:

((lyme or borrelia or borreliosis) and (pregnancy or pregnant or maternal or fetus or foetus or newborn or congenital))

The capacity of the electronic search to identify all relevant primary research was verified by hand searching reference lists of three book chapters published between 1995 and 2011 [6,23,24] and three randomly chosen review articles from a list of topic relevant reviews identified during title/abstract screening [2527]. This process netted 13 citations, conference proceedings, and non-indexed papers that were added to the SR. When omitted citations were no longer being identified the process was stopped. Hand searching of the following websites did not yield additional references:

Relevance screening and inclusion criteria

References identified by the search were screened for relevance to the review question using a structured and pre-tested form (S1 Text). Those considered relevant to the review question were procured and relevance was confirmed using another pre-tested form implemented prior to proceeding to the risk of bias evaluation. Primary research on humans with gestational LD and any birth outcome (e.g. healthy infants, pregnancy loss, fetal and newborn abnormalities, adverse outcomes or death) were considered relevant to the research question. Chronic Lyme disease was considered to be outside the scope of this review [28]. Global research in any language was included in this SR to minimise language bias. Primary research was defined as original research where authors generated and reported their own data.

Risk of bias, GRADE and data extraction

Assessment of the risk of bias (RoB) of research relevant to the review question was executed using the whole publication and applying a direct modification of the RoB and Grading of Recommendations Assessment, Development and Evaluation (GRADE) criteria endorsed by the Cochrane collaboration [20,29,30]. The RoB assessment evaluates the internal validity of each study using eight criteria from which the reviewers determine an overall RoB (low, unclear, or high) for each outcome. This informs one of the five GRADE criteria [31]. The data extraction form captures pertinent information and results required for summarization and meta-analysis. Two reviewers (LW and JG) independently assessed the RoB and extracted data on each article.

A GRADE assessment was conducted for each relevant outcome where an outcome is a result from a study; the same outcome may be measured in different studies and as part of the systematic review, outcomes that are alike are grouped together and evaluated as follows. In addition to RoB the other criteria include study design, agreement between studies, precision of results, and evidence of a biological gradient for each outcome. This review included research from any study design; however the less controlled the study, the higher the risk of systematic biases that can result in the study findings deviating from the truth. The GRADE framework prescribes a gradually lower GRADE as the risk of bias increases [32]. This means that randomized controlled trials and well-designed cohort studies could be graded moderate/high (*** or ****), whereas case control studies and cross sectional studies are likely to be graded as low (**) and case reports and expert opinion receive a very low grade (*). Across each outcome/study design pair, groups of similar studies would be evaluated for overall RoB, agreement, precision and evidence of a biological gradient, which could result in up-grading (or down-grading) the level of confidence in the evidence for that outcome [32,33].

The final GRADE is assessed considering all five GRADE criteria for each unique outcome to indicate the level of confidence in the evidence [30]. The one to four star grading system indicates: **** high confidence that the effect estimate is close to the true effect; *** moderate confidence in the effect estimate, but future studies may be substantially different; ** limited confidence in the effect estimate, the true effect may be substantially different; * very little confidence in the effect estimate, the true effect is likely to be substantially different [3234].

Systematic review management and analysis

Search results were imported into reference management software (Endnote X7, Thomson Reuters, USA), duplicates were removed and the list of unique citations was imported into a web-based electronic SR management platform (DistillerSR, Evidence Partners, Ottawa, Canada). All stages of the SR were conducted within this software and collected data were exported into Excel spreadsheets (Microsoft Corporation, Redmond, WA), organised (sub grouped by common exposure and outcome) and summarized (frequencies and percentages).

Post-hoc calculations such as computing unadjusted odds ratios (OR) from contingency table data were done, where necessary, to ensure comparability of the data. Whenever studies reported both unadjusted and adjusted OR measures, the adjusted measure was selected for inclusion in the tables and/or meta-analysis model [20].

Random-effects meta-analysis using the Der Simonian and Laird method was conducted for each unique group of studies if sufficient data were available (ie: if there were ≥2 studies, and the studies were comparable) [35]. For meta-analysis of proportions, the Freeman-Tukey double arcsine transformation was used to stabilize the variances [36]. Heterogeneity was measured using I2, which indicates the proportion of variation in the measures of association across studies due to heterogeneity rather than sampling error [37]. It was not possible to test for publication bias due to a limited number of studies (<10) in any subgroup [38].

Results

Forty-five relevant primary research studies were identified after screening 746 unique citations and 67 full papers (Fig 1). Thirty-five of these papers were published in English, two in Czech, two in Serbian and one each in French, Italian, Dutch, German, Russian, and Polish. Despite efforts to minimise bias by including all available research, we were unable to procure two potentially relevant articles [39,40].

thumbnail
Fig 1. The flow of citations and research papers through the systematic review process.

https://doi.org/10.1371/journal.pone.0207067.g001

The literature on adverse birth outcomes associated with gestational LD is based on case reports and case series (n = 26), case series with epidemiological data (n = 4), and epidemiological studies (n = 15) (Fig 1). The case report and case series studies reported detailed information on one or more cases of gestational LD and the pregnancy and/or birth outcomes of the case. The case series with epidemiological data provided information on a group of gestational LD cases and associations between possible risk factors (e.g. untreated vs. treated LD) and the risk of an adverse birth outcome. The cohort, cross-sectional, and case control studies investigated possible differences in the frequency of adverse outcomes between an exposed and unexposed control group. Case studies captured in this review were published between 1985 and 2017, whereas the epidemiological studies were published between 1986 and 2011 (Fig 2). Across studies, exposure was defined as: evidence of clinical manifestations of LD during pregnancy, serological evidence of LD during pregnancy or surrogate measures of exposure to LD (e.g. history of tick bites or living in a geographic area considered endemic for LD).

thumbnail
Fig 2. The distribution of publication dates of 45 primary research publications relevant to the impact of gestational Lyme disease included in this systematic review grouped by studies that had epidemiological data or case report data.

https://doi.org/10.1371/journal.pone.0207067.g002

Each study was evaluated for its RoB, which evaluates how well the study was conducted. Many criteria for RoB assessment do not apply to case reports, thus case report assessment focused on complete reporting. There were 26 case reports or case series included in this SR and three had an “unclear” RoB designation because the diagnostic tests performed or outcomes were not reported in sufficient detail (Table 1). The RoB for 19 epidemiological studies was 42% low, 42% unclear and 16% high (Table 1). Studies with an “unclear” RoB had one or more criteria that could not be assessed because the required information was not reported including: i) missing information on the blinding of patients and/or outcome assessors, or ii) unexplained loss to follow-up or loss of observations. “High” RoB studies had several flaws in the research process that may bias the results including failure to account for, or examine, important confounders or other biases. Insufficient information to assess RoB criteria is likely a reporting issue in many papers but results in an “unclear” or “high” RoB classification depending on the cumulative deficiencies of the study. The RoB evaluation for each study is available in S2 Table.

thumbnail
Table 1. General characteristics of 45 included primary research publications.

https://doi.org/10.1371/journal.pone.0207067.t001

Case reports of gestational Lyme disease

Details of 59 cases were summarized in 29 publications from the USA, Europe, and Asia describing gestational LD and pregnancy outcomes between 1969 and 2017, Table 2. These case reports and case series received a GRADE of *, indicating that future evidence may be inconsistent with the conclusions of these studies. Across 59 cases, negative outcomes for the fetus or newborn occurred in 36 (61%) pregnancies. Negative outcomes ranged from spontaneous miscarriage (termination of pregnancy prior to when the fetus is considered viable, approximately 28weeks) (n = 10), fetal death and stillbirth after 28 weeks (n = 2) and death shortly following birth (n = 8, four were premature, born before 36 weeks gestation), to a range of congenital abnormalities and health issues (n = 16) including hyperbilirubinemia, respiratory distress, syndactyly, and ureter and heart abnormalities, Table 2. For six infants, a wide range of long term conditions were reported [4145], Table 2. Healthy children (including one set of twins) were born in 23 pregnancies from mothers who had clinical manifestations (n = 10), serological evidence (n = 3) or both (n = 10) consistent with LD.

thumbnail
Table 2. Summary of pregnancy outcomes from 59 case reports diagnosed with gestational Lyme disease and test results for direct detection of B. burgdorferi or spirochetes or antibodies against the agent of Lyme disease.

https://doi.org/10.1371/journal.pone.0207067.t002

Laboratory testing of the newborn or fetus was not reported in 28 cases, Table 3, and for these the possible role of LD was determined by evidence in the mother (clinical manifestations [n = 13], diagnostic test results [n = 3], or both [n = 10]), while for 2 cases the only evidence of infection was identification of spirochetes in the placenta reported to be B. burgdorferi, Table 2. The newborn and fetus cases with laboratory evidence of infection (n = 31), included serological test results (n = 13) of which only two were considered positive and one was borderline, Table 3. Fetal or newborn tissue samples (n = 19) were examined and B. burgdorferi was identified in 17 cases using staining, indirect immunofluorescence (IF), or PCR to confirm the presence of B. burgdorferi. Not all tests used in the studies are considered reliable; those considered reliable include PCR with specific primers, IF using specific antibodies, and culture when it is confirmed by IF or PCR (n = 12) [6971]. Direct microscopic detection (n = 7) of B. burgdorferi (using bacteria staining and dark field microscopy) is generally considered of limited value in diagnosis due to both the small number of spirochetes seen and the risk of false positive results for a range of reasons reviewed elsewhere [72]. Across the case reports the specificity of the primers used for PCR and methods of culture confirmation were often not mentioned, but we included these reports nevertheless. Evidence of infection in the placenta (presence of spirochetes) and/or cord blood (presence of antibody) was sought in 12/59 cases, Table 3. Of these 5/11 placentas and 1/5 cord blood samples were positive (for IgG antibodies only), two of the placenta positive results were based on the results of indirect immunofluorescence. In the positive cord blood case and one placenta-positive case the child was healthy at birth.

thumbnail
Table 3. An overview of the features of 59 case reports diagnosed with gestational Lyme disease.

https://doi.org/10.1371/journal.pone.0207067.t003

Amongst the 59 pregnancies identified in the case studies in this SR, 33 (56%) pregnant women were tested for LD; but diagnostic methods currently considered reliable (direct detection methods as described above or the two-tier EIA followed by the Western Blot) were used in only four cases, Table 3 [27,41,46,64,69,72,73]. In five other cases, the mother was not diagnosed with LD by clinical symptoms or a diagnostic test, but instead was considered retrospectively to have suffered from LD when the cause of fetal or newborn demise was investigated and possible B. burgdorferi spirochetes were identified in the placenta (n = 3) and/or fetal tissue (n = 3) [49,51].

There was treatment information for 19/23 pregnancies that resulted in healthy newborns; 18/19 (95%) were treated for LD during pregnancy and subsequently had no adverse birth outcomes. Healthy newborns were born to mothers that had a wide range of gestational LD symptoms from EM to neuroborreliosis and LD was acquired in all trimesters (Table 2). In contrast, 34/36 pregnancies have treatment information and a negative birth outcome, of these only 14/34 (41%) were treated during pregnancy. Among the 20 cases that were not treated and had negative birth outcomes, 10 occurred in mothers with no clinical history of symptoms consistent with LD according to current guidelines [70,71]. The other 10 untreated cases were mainly diagnosed retrospectively or after parturition, so there was no opportunity for treatment during pregnancy.

Across cases, evidence that transplacental transmission of B. burgdorferi can occur was shown by testing the placenta (n = 11) and deceased fetal/newborn tissue (n = 18), Table 3. Adverse birth outcomes occurred in 4/5 placenta positive cases (2 stillbirths and 2 cases of respiratory distress that recovered), in 2/6 placenta-negative cases (one premature birth and one case reported as relapsing LD beginning at 3 months of age, and spirochetes were identified in one or more fetal tissues in 15/18 autopsies (Table 2). Only one case (in Germany) described the full range of expected observations (clinical manifestations in the mother, negative outcome for the child, and laboratory detection of B. burgdorferi in the child) that would give confidence that vertical transmission of B. burgdorferi, with negative consequences for the fetus, occurs [57,58]. The reports from the autopsies (n = 18) did not provide an explanation for how the presence of B. burgdorferi was associated with the pathology seen in the fetus [49,51,52,57,58]. A common autopsy observation was the lack of inflammation or immune response against B. burgdorferi infection in the fetus. Across all cases there were no consistent clinical outcomes resulting from gestational LD, and a linkage between fetal loss and gestational LD remains unclear from the case reports.

Epidemiological studies

There are 19 epidemiological studies identified in this review, which include nine cohort studies, four cross sectional studies, two case control studies, and four case series. The studies were conducted in the USA (n = 10) and Europe (n = 9) and published between 1986 and 2011. One cohort did not have extractable epidemiological outcomes, but the data from the pregnant LD cases are included in the previous section [66]. A second study is not included in this review because there were no extractable outcomes [74]. This study also received a very high RoB evaluation due to incomplete reporting of methods and outcomes, lack of blinding, failure to account for or examine important confounders, and other biases including the potential of funding bias [74].

Adverse outcomes in LD exposed vs unexposed populations.

Eight studies reported differences in prevalence of one or more types of adverse birth outcomes in exposed compared to unexposed populations, Table 4. The definition of ‘exposed’ in these studies included women diagnosed with gestational LD during the study, a history of LD (based on clinical chart review), positive LD serology during pregnancy, or those considered to be at higher risk of LD (measured indirectly by having a history of tick bites or living in a known endemic area for LD risk). Adverse birth outcomes amongst and within the epidemiological studies varied widely and included very common outcomes such as preterm birth and hyperbilirubinemia [75,76], as well as less frequent and more serious major congenital malformations. Some studies reported all adverse outcomes together without additional details by type of outcome (Table 4). Meta-analysis was considered inappropriate for all outcomes due to the variability in study design, definition of LD, and range of adverse outcomes.

thumbnail
Table 4. Measures of association extracted from eight studies on adverse birth outcomes and LD during or before pregnancy, positive LD serology during pregnancy, and surrogate measures of possible exposure to LD (e.g. tick bites or living in an endemic area).

(Significant odds ratios are bolded in the results.).

https://doi.org/10.1371/journal.pone.0207067.t004

The results included six studies where no significant association was reported between adverse birth outcomes (e.g. spontaneous miscarriage and congenital abnormalities) and the mother’s LD status, determined by serology or clinical diagnosis [7782] (Table 4). The RoB was low (n = 3) and unclear (n = 3) across these six studies, study designs and diagnosis of LD varied, but the conclusions were consistent. This gives this group of studies a *** GRADE indicating some confidence the overall conclusions of this research will not change with future research.

Other exposure measures included seropositive cord blood (n = 3 studies), tick bites during pregnancy (n = 3), history of tick bites (n = 1), and residing in an endemic area (n = 2) (Table 4). No association was shown from cord blood serology (IgG or IgM antibodies) results and adverse birth outcomes [8284]. In addition, congenital abnormalities overall were not associated with surrogate measures of LD exposure including exposure to ticks or expected exposure to ticks by virtue of living in an endemic area (Table 4) [77,8183]. Among the results from the LD endemic area of Westchester, New York, USA, a significantly higher odds of cardiac abnormalities and lower odds of minor abnormalities was observed compared to a population in a non-endemic area [82](Table 4). In this study, these associations were shown to be unrelated to a clinical history of LD in the mother and are assumed to be independent of LD [82]. Therefore there was no association between gestational LD or surrogate measures of exposure and adverse birth outcomes across the eight studies in Table 4.

Risk factors for adverse outcomes in LD exposed populations

There were ten studies that examined various risk factors for adverse birth outcomes among women diagnosed with gestational LD. In nine of these studies there were data on the proportion of adverse birth outcomes in treated and untreated women with gestational LD (Fig 3). LD status was determined in several ways across these studies; active gestational LD diagnosed by a physician (n = 5) [43,44,8486], retrospective identification based on medical records (n = 1) [44], or a positive serology result on a screening test (IgG and/or IgM) during pregnancy (n = 3) [87,88]. A random effects meta-analysis was conducted to examine the proportion of adverse outcomes across subgroups based on LD and treatment status: i) diagnosed with gestational LD and treated during pregnancy, ii) diagnosed with gestational LD, but not treated during pregnancy and iii) seropositive on a screening test (single immunoassay [80,87] or two tier test [88]) for LD during pregnancy, but did not have a clinical history of illness and consequently was not treated (Fig 3). There was a significantly higher proportion of adverse birth outcomes in the untreated subgroup (50%, 95%CI 30–70, I2 = 0%) compared to the group that received treatment (11%, 95%CI 7–16, I2 = 0%) (Fig 3). For the subgroup that had a seropositive screening test, but were considered healthy, the results from two studies were quite different from each other and in the meta-analysis this sub-group was not significantly different from the frequency of adverse birth outcomes in the treated and untreated subgroups (19%, 95%CI 6–35, I2 = 0%) (Fig 3).

thumbnail
Fig 3. Random effects meta-analysis of nine studies that reported the proportion of women with gestational Lyme disease that experienced an adverse birth outcome.

Studies were sub-grouped by treatment status: treated active LD, untreated LD that had a clinical history of LD symptoms, and seropositive with no history of LD. LD status was determined by retrospective medical record review, clinical diagnosis with and without serology or culture, or positive IgG and/or IgM serology. (NR = not reported, AO = Adverse outcome).

https://doi.org/10.1371/journal.pone.0207067.g003

Among these ten studies, two reported that the country level frequency of adverse birth outcomes was the same as the frequency among women in the study. Specifically, treated patients diagnosed with gestational LD had a lower frequency of spontaneous miscarriages and premature births and similar frequency of congenital malformations compared to the country level frequency [43,84]. This suggests that there was no increased risk of adverse birth outcomes among women with treated gestational LD compared to the country birth statistics. The overall GRADE of the studies in Fig 3 is ** meaning that we have limited confidence the results and estimates presented here will not change with future research. This is mainly due to the limited number of studies and observations in each sub-group despite homogeneity across studies and these limitations prevent exploration of potentially important confounding factors such as geographic region, year study was conducted and methods of diagnosis. Therefore, we caution that the summary results are not generalizable beyond the populations studied.

The odds of an adverse birth outcome in gestational LD cases that were treated compared to those that were untreated were described in four of the studies (Table 5), two involving symptomatic women and two involving asymptomatic women. The largest study was from Hungary and reported significantly increased odds of adverse birth outcomes (OR 7.61, 95% CI 1.90–30.51) [84]. The other studies, two from the USA and one from Italy, found no difference between the treated and untreated groups although the results were in the same direction as the study from Hungary [44,80]. Thus, the data suggest there is some evidence that adverse birth outcomes may occur more frequently if gestational LD is not treated.

thumbnail
Table 5. Measures of association extracted from six studies examining the odds of an adverse birth outcome in patients with gestational LD by treatment status, timing of exposure, severity and progression of LD.

(Significant odds ratios are bolded in the results.).

https://doi.org/10.1371/journal.pone.0207067.t005

Possible risk factors other than LD treatment status were investigated in three studies including trimester of B. burgdorferi exposure, length of LD during pregnancy, early vs. disseminated gestational LD, and women presenting with an EM only compared to those with an EM and other symptoms of LD [44,84,89](Table 5). None of these studies found a significant association between adverse birth outcomes and these possible risk factors, although most of these studies were small and may have had limited power to detect a difference. For example, in one study with only 19 observations the association was not significant, but there was a higher proportion of adverse birth outcomes in women with disseminated LD (cardiac manifestations and neuroborreliosis; 43% had adverse birth outcomes) compared to early LD (EM only; 17% had adverse birth outcomes) [44]. There were no significant associations between trimester of LD infection and adverse birth outcomes [84] Table 5, or with the frequency of spontaneous miscarriages [77,79]. Birth weight, a surrogate measurement for newborn health, was unrelated to gestational LD in two studies [77,82]. Spirochetes found in the placenta were not associated with adverse birth outcomes in a case series where 3/60 placentas were spirochete positive and all infants were healthy [85]. Coinfection with Anaplasma phagocytophilum based on PCR of the blood and/or placenta was reported in one study for 37.5% (3/8) LD positive women that had adverse birth outcomes (n = 2) and healthy twins (n = 1). The small sample size prohibited investigation of the association between coinfection and adverse birth outcomes [90,91].

Discussion

The literature included in this SR was published between 1985 and 2017, 58% of which are case studies. As evidence, case studies are helpful to generate hypotheses for future research, but cannot be used to further our understanding of a causal relationship, if one exists, between gestational LD and adverse birth outcomes. There were a number of reporting issues in the case studies included in this SR, mostly related to missing or limited information on the mother’s clinical symptoms and the use of diagnostic methods and laboratory tests currently considered unreliable [6972]. The latter issue also applies to many of the epidemiology studies. This is not to indicate that the results from these studies are false, but that they are questionable, which is a feature of the age of the majority of studies identified for inclusion in this SR.

Diagnosis of LD relies on clinical evaluation, plausible exposure history to infected ticks, and if needed, supplemental diagnostic laboratory tests [70,71]. Reliable test methods would include direct demonstration of B. burgdorferi in tissues or in culture by IF or PCR using, respectively, specific antibodies or primers, or results of a two tier serological test interpreted by current guidelines, the latter being the most common type of testing for diagnosis of LD [9294]. Typically, two tier serological testing includes an enzyme immunoassay (EIA) to detect IgM or IgG serum antibodies to B. burdorferi; positive or equivocal tests are followed by an immunoblot assay (IB, e.g. Western blot) to confirm the positive screening test result [9294]. Although these guidelines improved the performance of LD testing, all currently available testing options are imperfect [73,95]. Thus, the inadequate sensitivity of serological tests in early LD necessitates physician awareness of LD and careful clinical assessment supported by laboratory testing results when appropriate. Pregnant women who have acquired LD should be treated according to current guidelines [9294] as the meta-analysis in this SR suggested that treatment of LD during pregnancy was associated with a decrease in the risk of adverse birth outcomes.

The issue of misdiagnosis or misclassification of LD across studies in this SR is important to carefully consider. The use of unreliable tests or test protocols for LD could contribute to misclassification of cases, since 58% of the studies in this SR were conducted prior to the current serological testing guidelines and LD serological tests were used to screen healthy pregnant women in some of the epidemiology studies [7780,8284,87,88,92]; there is a measurable risk of false positives and thus misclassification of observations in the sample population [96]. Among several case reports there is little information on the diagnosis of LD and given the age of these articles there appears to be a reasonable risk that some cases were misclassified as gestational LD. Unfortunately, both false positives and false negatives could have occurred in these studies and had an impact on the results in an unknown direction and magnitude possibly resulting in the distortion of detected associations or failure to detect associations, which undermines our confidence in the research results. Considering the potential for misclassification of LD in the included studies, additional research using currently accepted methods of LD diagnosis, an improved understanding of LD, and larger sample sizes (e.g. via large multi-center observational studies) is needed to more adequately explore possible effects of gestational LD and further investigate potential risk factors suggested in this SR.

It is biologically plausible that transplacental transmission of B. burgdorferi occurs given our understanding of transplacental spirochete transmission for other species of spirochetes (T. pallidum) in humans [6,7]. There are examples among the 59 case reports included in this SR that suggested transplacental transmission occurs including 4 cases of infection in the fetus or newborn determined using relatively reliable laboratory diagnostic methods. Of these only one case reported clinical LD in the mother, an adverse birth outcome and potential demonstration of B. burgdorferi in the child; that would provide some confidence that vertical transmission of B. burgdorferi occurred and may have resulted in a negative outcome for the fetus [57,58]. Examination of the pathological findings from case studies where B. burgdorferi was identified in various fetal tissues does not provide evidence that the presence of B. burgdorferi was linked to the pathological findings and there was a lack of inflammatory response noted in several cases [43,44,47,4952,57,87,89,97]. These findings are in alignment with literature reviews by medical practitioners on this topic [6,98]. Therefore, it is possible that vertical transmission with negative outcomes can occur, but there are knowledge gaps in terms of the pathology and frequency of occurrence.

Common adverse birth outcomes reported across studies in this SR included preterm birth and hyperbilirubinemia, which are also common outcomes in the general population [75,76]. The potential for increased risk of an adverse outcome in women with gestational LD that is not treated was shown in this SR; however an explanation for this was not addressed in the available research and it is very possible that this was due to many factors such as sub-optimal maternal health as opposed to a single specific pathology caused by B. burgdorferi infection. Congenital malformations of the cardiac or genitourinary system are also among the most common malformations reported [75,99] and were frequently reported in case reports and epidemiological studies in this SR. Hypotheses that there may be higher rates of cardiac malformations as a result of gestational LD were investigated in the early epidemiology studies and case reports included in this SR, but these studies were unable to clarify a relationship with gestational LD [51,52,81]. Given recent research characterising the impact of B. burgdorferi on the cardiac system, additional work on the teratogenic potential of B. burgdorferi particularly on the cardiac system may be warranted [100]. However, the evidence in this SR on congenital malformations does not provide sufficient evidence to exclude or confirm a role for B. burgdorferi in congenital malformations. Future research is needed to address knowledge gaps such as the pathogenesis of B. burgdorferi infection in the developing fetus and its relationship to adverse birth outcomes.

Several risk factors were investigated based on the pathology observed in the early case reports and our biological understanding of LD. These studies failed to find an association between the mother’s LD status and cardiac, minor or major malformation, spontaneous miscarriages, and fetal death [77,8082]. Adverse birth outcomes were also not associated with the severity of gestational LD (early vs. disseminated), length of LD during pregnancy, or trimester of infection [44,84,89]. However, there was some evidence of increased risk of adverse outcomes in symptomatic women who were not treated with antibiotics, and it is possible that associations with rare or infrequent outcomes were not detected because the sample sizes (or number of LD cases) in most of the epidemiological studies was small and the range of reported outcomes was quite large.

There are several limitations to the evidence included in this SR. This includes limited generalisability of the results to populations other than those studied as there is not enough research to determine whether population differences exist and how they could have impacted the findings. Country level or regional rates for adverse birth outcomes are influenced by many factors related to socio economic factors, healthcare, and genetic predispositions that should be considered when weighing the generalizability of the data [77,82]. Other possible sources of variation in the frequency and type of outcomes include differences among genospecies of B. burgdorferi (e.g. B. burgdorferi sensu stricto, B. afzelii, and B. garinii) that can cause different manifestations of LD. The data on Borrelia species was scarce among the studies in this SR and should be considered in the design of future research to clarify if there are different outcomes or impacts of gestational LD depending on the pathogen [6].

Conclusion

This SR summarizes the research and anecdotal evidence on the potential impact of gestational LD on adverse birth outcomes. Overall there is a limited amount of evidence; with 29 case report articles and 17 epidemiological studies on this topic, and the results highlight a number of knowledge gaps and significant uncertainty about the impact of LD during pregnancy. Due to the variability in the study size and study design, the lack of evidence in the epidemiological research does not rule out uncommon consequences of LD during pregnancy. There is some evidence to suggest that it is biologically plausible for B. burgdorferi to be vertically transmitted to the fetus, however these studies have been unable to define a characteristic pathological effect of B. burgdorferi infection in the fetus, thus there are significant knowledge gaps about the relationship of B. burgdorferi infection and adverse birth outcomes [32]. Given the uncertainty around the impact of B. burgdorferi on the fetus and the consistent evidence suggesting fewer adverse birth outcomes if LD is promptly treated, it is recommended that physicians continue to remain thorough in their diagnosis and treatment of LD in pregnant women and that new research address the knowledge gaps identified in this review.

Acknowledgments

Disclaimer: The findings and conclusions in this paper are those of the authors and do not necessarily represent the official position of the Centers for Disease Control and Prevention.

The authors thank the Public Health Agency of Canada’s library staff for their assistance in the procurement of articles.

References

  1. 1. Steere AC (2006) Lyme borreliosis in 2005, 30 years after initial observations in Lyme Connecticut. Wiener Klinische Wochenschrift 118: 625–633. pmid:17160599
  2. 2. van den Wijngaard CC, Hofhuis A, Harms MG, Haagsma JA, Wong A, de Wit GA, et al. (2015) The burden of Lyme borreliosis expressed in disability-adjusted life years. European journal of public health 25: 1071–1078. pmid:26082446
  3. 3. Steere AC, Malawista SE, Snydman DR, Shope RE, Andiman WA, Ross MR, et al. (1977) Lyme arthritis: an epidemic of oligoarticular arthritis in children and adults in three connecticut communities. Arthritis Rheum 20: 7–17. pmid:836338
  4. 4. Larsson C, Andersson M, Guo BP, Nordstrand A, Hagerstrand I, Carlsson S, et al. (2006) Complications of pregnancy and transplacental transmission of relapsing-fever borreliosis. Journal of Infectious Diseases 194: 1367–1374. pmid:17054065
  5. 5. Yagupsky P, Moses S (1985) Neonatal Borrelia species infection (relapsing fever). American Journal of Diseases of Children 139: 74–76. pmid:3969988
  6. 6. Shapiro ED, Gerber MA (2011) Chapter 17: Borrelia Infections: Lyme Disease and Relapsing Fever. In: Britt W, editor. Infectious Diseases of the Fetus and Newborn, 7th ed: Elsevier.
  7. 7. Kollmann D, Dobson S (2011) Treponema pallidum. In: Remington JK, J., editor. Infectious Diseases of the Fetus and Newborn Infant, seventh ed,. Philadelphia: Saunders.
  8. 8. Madjunkov M, Chaudhry S, Ito S (2017) Listeriosis during pregnancy. Arch Gynecol Obstet 296: 143–152. pmid:28536811
  9. 9. Anderson JF, Johnson RC, Magnarelli LA (1987) Seasonal prevalence of Borrelia burgdorferi in natural populations of white-footed mice, Peromyscus leucopus. Journal of Clinical Microbiology 25: 1564–1566. pmid:3624451
  10. 10. Burgess EC (1988) Borrelia burgdorferi infection in Wisconsin horses and cows. Ann N Y Acad Sci 539: 235–243. pmid:3190095
  11. 11. Burgess EC, Wachal MD, Cleven TD (1993) Borrelia burgdorferi infection in dairy cows, rodents, and birds from four Wisconsin dairy farms. Veterinary Microbiology 35: 61–77. pmid:8362496
  12. 12. Burgess EC, Windberg LA (1989) Borrelia sp. infection in coyotes, black-tailed jack rabbits and desert cottontails in southern Texas. Journal of Wildlife Diseases 25: 47–51. pmid:2644452
  13. 13. Gustafson JM, Burgess EC, Wachal MD, Steinberg H (1993) Intrauterine transmission of Borrelia burgdorferi in dogs. American Journal of Veterinary Research 54: 882–890. pmid:8323057
  14. 14. Silver RM, Yang L, Daynes RA, Branch DW, Salafia CM, Weis JJ (1995) Fetal outcome in murine Lyme disease. Infection & Immunity 63: 66–72.
  15. 15. Sorensen K (1990) Lyme disease antibodies in thoroughbred broodmares. correlation to early pregnancy failure. Journal of Equine Veterinary Science 10: 166–168.
  16. 16. Eisner RJ, Meirs DA, Meirs DA, III, Ralston SL (1994) Lack of correlation between exposure to lyme disease (Borrelia burgdorferi) and pregnancy loss in mares. Journal of Equine Veterinary Science 14: 102–105.
  17. 17. Mather TN, Telford SR, Adler GH (1991) Absence of transplacental transmission of Lyme disease spirochetes from reservoir mice (Peromyscus leucopus) to their offspring. Journal of Infectious Diseases 164: 564–567. pmid:1869842
  18. 18. Moody KD, Barthold SW (1991) Relative infectivity of Borrelia burgdorferi in Lewis rats by various routes of inoculation. American Journal of Tropical Medicine & Hygiene 44: 135–139.
  19. 19. Woodrum JE, Oliver JH (1999) Investigation of venereal, transplacental, and contact transmission of the Lyme disease spirochete, Borrelia burgdorferi, in Syrian hamsters. Journal of Parasitology 85: 426–430. pmid:10386432
  20. 20. Higgins J, Green S (2011) Cochrane Handbook for Systematic Reviews of Interventions. Cochrane collaboration.
  21. 21. Young I, Waddell L, Sanchez J, Wilhelm B, McEwen SA, Rajic A (2014) The application of knowledge synthesis methods in agri-food public health: recent advancements, challenges and opportunities. Prev Vet Med 113: 339–355. pmid:24485274
  22. 22. Moher D, Altman DG, Liberati A, Tetzlaff J (2011) PRISMA statement. Epidemiology 22: 128; author reply 128. pmid:21150360
  23. 23. Gardner T (1995) Lyme Disease. In: Remington JK, J., editor. Infectious Diseases of the fetus and newborn infant. 4th ed: W. B. Saunders company. pp. 447–529.
  24. 24. Gardner T (2001) Chapter 11, Lyme Disease. In: Remington JK, J., editor. Infectious Diseases of the Fetus and Newborn, 5th ed: Saunders.
  25. 25. McClure EM, Goldenberg RL (2009) Infection and stillbirth. Seminars In Fetal & Neonatal Medicine 14: 182–189.
  26. 26. Mylonas I (2011) Borreliosis during pregnancy: a risk for the unborn child? Vector Borne & Zoonotic Diseases 11: 891–898.
  27. 27. Walsh CA, Mayer EW, Baxi LV (2007) Lyme disease in pregnancy: case report and review of the literature. Obstetrical & Gynecological Survey 62: 41–50.
  28. 28. Lantos PM (2015) Chronic Lyme disease. Infect Dis Clin North Am 29: 325–340. pmid:25999227
  29. 29. Guyatt G, Oxman AD, Akl EA, Kunz R, Vist G, Brozek J, et al. (2011) GRADE guidelines: 1. Introduction-GRADE evidence profiles and summary of findings tables. J Clin Epidemiol 64: 383–394. pmid:21195583
  30. 30. Group GW (2013) GRADE guidelines—best practices using the GRADE framework.
  31. 31. Higgins JPT, Altman DG (2008) Chapter 8: Assessing risk of bias in included studies. In: In: Higgins Jpt G Se, editor. Cochrane Handbook for Systematic Reviews of Interventions Version 501 (updated September 2008): The Cochrane Collaboration.
  32. 32. Schunemann H, Hill S, Guyatt G, Akl EA, Ahmed F (2011) The GRADE approach and Bradford Hill's criteria for causation. J Epidemiol Community Health 65: 392–395. pmid:20947872
  33. 33. Guyatt GH, Oxman AD, Sultan S, Glasziou P, Akl EA, Alonso-Coello P, et al. (2011) GRADE guidelines: 9. Rating up the quality of evidence. Journal of clinical epidemiology 64: 1311–1316. pmid:21802902
  34. 34. Balshem H, Helfand M, Schunemann HJ, Oxman AD, Kunz R, Brozek J, et al. (2011) GRADE guidelines: 3. Rating the quality of evidence. J Clin Epidemiol 64: 401–406. pmid:21208779
  35. 35. DerSimonian R, Laird N (1986) Meta-analysis in clinical trials. Control Clin Trials 7: 177–188. pmid:3802833
  36. 36. Freeman MF, Tukey JW Transformations Related to the Angular and the Square Root.
  37. 37. Borenstein M, Hedges LV, Higgins JPT, Rothstein HR (2009) Introduction to Meta-Analysis. Chichester; Hoboken.: John Wiley & Sons.
  38. 38. Sterne JA, Sutton AJ, Ioannidis JP, Terrin N, Jones DR, Lau J, et al. (2011) Recommendations for examining and interpreting funnel plot asymmetry in meta-analyses of randomised controlled trials. Bmj 343: d4002. pmid:21784880
  39. 39. Hercogova J, Moidlova M, Zirny J (1994) Could borrelia found in the placenta influence the fetus? Study of 19 women with erythema migrans during pregnancy. In Program and Abstracts of the 6th Interantioanl Conference on Lyme Borreliosis Bologna, Italy: Societa Editrice Esculapio. pp. p76.
  40. 40. Sigal LH (2005) Pregnancy complicated by Lyme disease.
  41. 41. Troyano-Luque J, Padilla-Perez A, Martinez-Wallin I, Alvarez de la Rosa M, Mastrolia SA, Trujillo JL, et al. (2014) Short and long term outcomes associated with fetal cholelithiasis: a report of two cases with antenatal diagnosis and postnatal follow-up. Case Reports in Obstetrics and Gynecology 2014: 714271. pmid:25349754
  42. 42. Gasser R, Dusleag J, Reisinger E, Stauber R, Grisold M, Pongratz S, et al. (1994) A most unusual case of a whole family suffering from late Lyme borreliosis for over 20 years. Angiology 45: 85–86. pmid:8285392
  43. 43. Maraspin V, Cimperman J, Lotric-Furlan S, Pleterski-Rigler D, Strle F (1999) Erythema migrans in pregnancy. Wiener Klinische Wochenschrift 111: 933–940. pmid:10666804
  44. 44. Markowitz LE, Steere AC, Benach JL, Slade JD, Broome CV (1986) Lyme disease during pregnancy. JAMA 255: 3394–3396. pmid:2423719
  45. 45. Lampert F (1986) Infantile multisyslem inflammatory disease: another case of a new syndrome European Journal of Pediatrics 144: 593–596. pmid:3709576
  46. 46. Schaumann R, Fingerle V, Buchholz K, Spencker FB, Rodloff AC (1999) Facial palsy caused by Borrelia infection in a twin pregnancy in an area of nonendemicity. Clinical Infectious Diseases 29: 955–956. pmid:10589933
  47. 47. Lavoie PE, Lattner BP, Duray PH, Malawista SE, Barbour AG, Johnson RC (1987) Culture positive, seronegative transplacental Lyme borreliosis infant mortality. Arthritis Rheum. pp. S50.
  48. 48. Schutzer SE, Janniger CK, Schwartz RA (1991) Lyme disease during pregnancy. Cutis 47: 267–268. pmid:2070648
  49. 49. MacDonald AB (1989) Gestational Lyme borreliosis. Implications for the fetus. Rheumatic Diseases Clinics of North America 15: 657–677. pmid:2685924
  50. 50. MacDonald AB, Benach JL, Burgdorfer W (1987) Stillbirth following maternal Lyme disease. New York State Journal of Medicine 87: 615–616. pmid:3480464
  51. 51. MacDonald AB (1986) Human fetal borreliosis, toxemia of pregnancy, and fetal death. Zentralblatt fur Bakteriologie, Mikrobiologie, und Hygiene—Series A, Medical Microbiology, Infectious Diseases, Virology, Parasitology 263: 189–200. pmid:3554838
  52. 52. Schlesinger PA, Duray PH, Burke BA, Steere AC, Stillman MT (1985) Maternal-fetal transmission of the Lyme disease spirochete, Borrelia burgdorferi. Annals of Internal Medicine 103: 67–68. pmid:4003991
  53. 53. O’Brien JM, Baum JD (2017) Low-grade fever, erythematous rash in pregnant woman • Dx? Journal of Family Practice 66: E9–E10.
  54. 54. Andrasova V, Svarovsky J, Matousek B (1988) [Lyme disease in pregnancy]. Ceskoslovenska Gynekologie 53: 39–41. pmid:3370692
  55. 55. Mikkelsen AL, Palle C (1987) Lyme disease during pregnancy. Acta Obstetricia et Gynecologica Scandinavica 66: 477–478. pmid:3425250
  56. 56. Remy JM, Chevrant-Breton O, Logeais B, Patoux-Pibouin M, Chevrier S, Chevrant-Breton J (1994) Traitement de la maladie de lyme pendant la grossesse, a propos d'un cas Nouvelles Dermatologiques 13: 682.
  57. 57. Weber K, Bratzke HJ, Neubert U, Wilske B, Duray PH (1988) Borrelia burgdorferi in a newborn despite oral penicillin for Lyme borreliosis during pregnancy. Pediatric Infectious Disease Journal 7: 286–289. pmid:3130607
  58. 58. Weber K, Neubert U (1986) Clinical features of early erythema migrans disease and related disorders. Zentralblatt fur Bakteriologie, Mikrobiologie, und Hygiene—Series A, Medical Microbiology, Infectious Diseases, Virology, Parasitology 263: 209–228. pmid:3577481
  59. 59. Bussen S, Steck T (1994) [Manifestation of Lyme arthritis in the puerperal period]. Zeitschrift fur Geburtshilfe und Perinatologie 198: 150–152. pmid:7975802
  60. 60. Trevisan G, Stinco G, Cinco M (1997) Neonatal skin lesions due to a spirochetal infection: a case of congenital Lyme borreliosis? International Journal of Dermatology 36: 677–680. pmid:9352409
  61. 61. Brzostek T (2004) Ludzka erlichioza granulocytarna współistniejaca z borelioza z Lyme u kobiety ciezarnej Przeglad epidemiologiczny 58: 289–294.
  62. 62. Moniuszko A, Czupryna P, Pancewicz S, Kondrusik M, Penza P, Zajkowska J (2012) Borrelial lymphocytoma-A case report of a pregnant woman. Ticks and Tick-borne Diseases 3: 257–258. pmid:22906496
  63. 63. Grandsaerd MJ, Meulenbroeks AA (2000) Lyme borreliosis as a cause of facial palsy during pregnancy. European Journal of Obstetrics, Gynecology, & Reproductive Biology 91: 99–101.
  64. 64. Hemels MAC, Gerards LJ, Kwee A, Wolfs TFW (2002) Lyme-borreliose in de zwangerschap: Gevolgen voor de pasgeborene?. Tijdschrift voor Kindergeneeskunde 70: 65–67.
  65. 65. Isailović G, Veljković M, Soć N, Krstić B, Bjekić M (1993) Erythema migrans posle uvoda krpelja kod trudnice. Glas Srpska akademija nauka i umetnosti Odeljenje medicinskih nauka: 173–175.
  66. 66. Jovanovic R, Hajric A, Cirkovic A, Mikovic Z, Dmitrovic R (1993) Lajmska bolest i trudnoca. Glas—Srpska Akademija Nauka i Umetnosti, Odeljenje Medicinskih Nauka: 169–172.
  67. 67. Önk G, Acun C, Kalayci M, Çaǧavi F, Açikgöz B, Tanriverdi HA (2005) Gestational lyme disease as a rare cause of congenital hydrocephalus. Journal of the Turkish German Gynecology Association 6: 156–157.
  68. 68. Nafeev AA (2001) Sluchai bolezni Laima pri beremennosti. Zhurnal Mikrobiologii Epidemiologii i Immunobiologii: 121–122.
  69. 69. Aguero-Rosenfeld ME, Wang G, Schwartz I, Wormser GP (2005) Diagnosis of lyme borreliosis. Clin Microbiol Rev 18: 484–509. pmid:16020686
  70. 70. Wormser GP, Dattwyler RJ, Shapiro ED, Halperin JJ, Steere AC, Klempner MS, et al. (2006) The clinical assessment, treatment, and prevention of lyme disease, human granulocytic anaplasmosis, and babesiosis: clinical practice guidelines by the Infectious Diseases Society of America. Clin Infect Dis 43: 1089–1134. pmid:17029130
  71. 71. Stanek G, Fingerle V, Hunfeld KP, Jaulhac B, Kaiser R, Krause A, et al. (2011) Lyme borreliosis: clinical case definitions for diagnosis and management in Europe. Clin Microbiol Infect 17: 69–79. pmid:20132258
  72. 72. Wang G (2002) Direct detection methods for Lyme Borrelia, including the use of quantitative assays. Vector Borne Zoonotic Dis 2: 223–231. pmid:12804163
  73. 73. Waddell LA, Greig J, Mascarenhas M, Harding S, Lindsay R, Ogden N (2016) The Accuracy of Diagnostic Tests for Lyme Disease in Humans, A Systematic Review and Meta-Analysis of North American Research. PLoS One 11: e0168613. pmid:28002488
  74. 74. Jones CR, Smith H, Gibb E, Johnson L (2005) Gestational Lyme Disease Case Studies of 102 Live Births. Lyme Times: 36–38.
  75. 75. (2017) Perinatal health indicators for Canada 2017, a report from the Canadian perinatal surveillance system. Ottawa.
  76. 76. Agency EP (2013) America's Children and the Environment, 3rd Ed. United States: EPA. 264 p.
  77. 77. Strobino BA, Williams CL, Abid S, Chalson R, Spierling P (1993) Lyme disease and pregnancy outcome: a prospective study of two thousand prenatal patients. American Journal of Obstetrics & Gynecology 169: 367–374.
  78. 78. Dlesk A, Broste SK, Harkins PG, McCarty PA, Mitchell PD (1989) Lyme seropositivity and pregnancy outcome in the absence of symptoms of Lyme disease. Arthritis Rheum pp. S46.
  79. 79. Carlomagno G, Luksa V, Candussi G, Magaton Rizzi G, Trevisan G (1988) Lyme Borrelia positive serology associated with spontaneous abortion in an endemic Italian area. Acta Europaea Fertilitatis 19: 279–281. pmid:3252658
  80. 80. Bracero LA, Wormser GP, Leikin E, Tejani N (1992) Prevalence of seropositivity to the Lyme disease spirochete during pregnancy in an epidemic area. A preliminary report. journal of maternal and fetal investigations 2: 265–268.
  81. 81. Strobino B, Abid S, Gewitz M (1999) Maternal Lyme disease and congenital heart disease: A case-control study in an endemic area. American Journal of Obstetrics & Gynecology 180: 711–716.
  82. 82. Williams CL, Strobino B, Weinstein A, Spierling P, Medici F (1995) Maternal Lyme disease and congenital malformations: a cord blood serosurvey in endemic and control areas. Paediatric and Perinatal Epidemiology 9: 320–330. pmid:7479280
  83. 83. Williams CL, Benach JL, Curran AS, Spierling P, Medici F (1988) Lyme Disease during Pregnancy A Cord Blood Serosurvey. Annals of the New York Academy of Sciences 539: 504–506.
  84. 84. Lakos A, Solymosi N (2010) Maternal Lyme borreliosis and pregnancy outcome. International Journal of Infectious Diseases 14: e494–498. pmid:19926325
  85. 85. Figueroa R, Bracero LA, Aguero-Rosenfeld M, Beneck D, Coleman J, Schwartz I (1996) Confirmation of Borrelia burgdorferi spirochetes by polymerase chain reaction in placentas of women with reactive serology for Lyme antibodies. Gynecologic & Obstetric Investigation 41: 240–243.
  86. 86. Maraspin V, Ruzic-Sabljic E, Pleterski-Rigler D, Strle F (2011) Pregnant women with erythema migrans and isolation of borreliae from blood: course and outcome after treatment with ceftriaxone. Diagnostic Microbiology & Infectious Disease 71: 446–448.
  87. 87. Nadal D, Hunziker UA, Bucher HU, Hitzig WH, Duc G (1989) Infants born to mothers with antibodies against Borrelia burgdorferi at delivery. European Journal of Pediatrics 148: 426–427. pmid:2920747
  88. 88. Londero F, San Marco L, Silvestri D, Ruscio M, Bassini D (1998) Enfermedad de Lyme y embarazo. Progresos en Obstetricia y Ginecologia 41: 146–148.
  89. 89. Hercogova J, Tomankova M, Frosslova D, Janovska D (1993) Early stage of Lyme borreliosis during pregnancy: Treatment of 15 women with erythema migrans. Cesko-Slovenska Gynekologie 58: 229–232.
  90. 90. Hulinska D, Votypka J, Vanousova D, Hercogova J, Hulinsky V, Drevova H, et al. (2009) Identification of Anaplasma phagocytophilum and Borrelia burgdorferi sensu lato in patients with erythema migrans. Folia Microbiologica 54: 246–256. pmid:19649743
  91. 91. Knox KK, Thomm AM, Harrington YA, Ketter E, Patitucci JM, Carrigan DR (2017) Powassan/Deer Tick Virus and Borrelia Burgdorferi Infection in Wisconsin Tick Populations. Vector Borne Zoonotic Dis 17: 463–466. pmid:28488932
  92. 92. CDC (1995) Recommendations for test performance and interpretation from the Second National Conference on Serologic Diagnosis of Lyme Disease. MMWR Morb Mortal Wkly Rep 44: 590–591. pmid:7623762
  93. 93. CDC (2017) Lyme Disease Diagnosis and Testing.
  94. 94. Lindsay LR, Bernat K, Dibernardo A (2014) Laboratory diagnostics for Lyme disease. Canada Communicable Disease Report 40.
  95. 95. Leeflang MM, Ang CW, Berkhout J, Bijlmer HA, Van Bortel W, Brandenburg AH, et al. (2016) The diagnostic accuracy of serological tests for Lyme borreliosis in Europe: a systematic review and meta-analysis. BMC Infect Dis 16: 140. pmid:27013465
  96. 96. Canadian Public Health Laboratory N (2007) The laboratory diagnosis of Lyme borreliosis: Guidelines from the Canadian Public Health Laboratory Network. Can J Infect Dis Med Microbiol 18: 145–148. pmid:18923770
  97. 97. Ciesielski CAR, H.; Johnson, S.; (1987) Prospective study of pregnancy outcome in women with Lyme disease (abstract). 27th ICAAC.
  98. 98. Elliott DJ, Eppes SC, Klein JD (2001) Teratogen update: Lyme disease. Teratology 64: 276–281. pmid:11745834
  99. 99. Egbe A, Uppu S, Lee S, Stroustrup A, Ho D, Srivastava S (2015) Congenital malformations in the newborn population: a population study and analysis of the effect of sex and prematurity. Pediatr Neonatol 56: 25–30. pmid:25267275
  100. 100. Muehlenbachs A, Bollweg BC, Schulz TJ, Forrester JD, DeLeon Carnes M, Molins C, et al. (2016) Cardiac Tropism of Borrelia burgdorferi: An Autopsy Study of Sudden Cardiac Death Associated with Lyme Carditis. The American journal of pathology 186: 1195–1205. pmid:26968341