EM indicates erythema migrans; MEM, multiple erythema migrans.
eTable. Factors Associated With Incomplete Response According to Dissemination (Multiple Erythema Migrans Versus Solitary Erythema Migrans)
Customize your JAMA Network experience by selecting one or more topics from the list below.
Stupica D, Maraspin V, Bogovič P, et al. Comparison of Clinical Course and Treatment Outcome for Patients With Early Disseminated or Early Localized Lyme Borreliosis. JAMA Dermatol. 2018;154(9):1050–1056. doi:10.1001/jamadermatol.2018.2306
In what way does clinical course and treatment outcome for patients with early disseminated Lyme borreliosis differ from those with early localized Lyme borreliosis?
In this cohort study of 603 patients with multiple erythema migrans or erythema migrans, constitutional symptoms accompanied early disseminated disease more often than early localized disease. The long-term outcome at 12 months after treatment, however, was comparable regardless of dissemination.
Follow-up of at least 12 months after treatment is recommended in future studies assessing post–Lyme borreliosis symptoms.
Multiple erythema migrans (MEM) has been suggested as a risk factor for unfavorable antibiotic treatment outcome compared with solitary erythema migrans (EM). However, no direct comparison of early Lyme borreliosis manifested as MEM with solitary EM has been undertaken.
To investigate the potential differences in clinical course and treatment outcome between MEM and solitary EM.
Design, Setting, and Participants
This prospective cohort study was conducted from June 1, 2010, to October 31, 2015, at the University Medical Center Ljubljana, Slovenia. Data were analyzed from June 1, 2017, to January 3, 2018. Of the 778 consecutive adult patients with early Lyme borreliosis evaluated, 200 patients with MEM and 403 patients with solitary EM were enrolled. Patients were asked to refer a family member or a friend of similar age (±5 years) without a history of Lyme borreliosis to serve as a control participant. Clinical course and posttreatment outcome of MEM were compared with those of solitary EM. Outcome was assessed at 14 days and at 2, 6, and 12 months after enrollment. At each visit, patients completed a written questionnaire about their symptoms; controls completed the same questionnaire. Nonspecific symptoms reported by patients and controls without a history of Lyme borreliosis were compared.
Main Outcomes and Measures
The proportion of patients with incomplete response at 12 months after enrollment and the associated 2-sided 95% CI for the difference between MEM and solitary EM were estimated using the normal approximation with continuity correction.
A total of 200 patients with MEM and 403 patients with solitary EM were included. Among the 200 patients with MEM, 94 (47.0%) were males and 106 (53.0%) were females, with a median (interquartile range [IQR]) age of 47 (35-58) years. Among the 403 patients with solitary EM, 182 (45.2%) were males and 221 (54.8%) were females, with a median (IQR) age of 55 (42-62) years. Patients with MEM reported Lyme borreliosis–associated constitutional symptoms at enrollment more often than those with solitary EM (93 [46.5%]; 95% CI, 39.4-53.7 vs 96 [23.8%]; 95% CI, 19.7-28.3; P < .001). During the initial 6 months after treatment, the proportion of patients with incomplete response was higher in the MEM group than in the solitary EM group (14 days: 62 of 193 [32.1%] vs 72 of 391 [18.4%]; P < .001; 2 months: 38 of 193 [19.7%] vs 55 of 394 [14.0%]; P = .28; 6 months: 29 of 182 [15.9%] vs 31 of 359 [8.6%]; P = .02). However, at the 12-month visit, the outcome was comparable: 10 of 170 (5.9%) patients with MEM vs 20 of 308 (6.5%) patients with solitary EM showed incomplete response (−0.6; 95% CI, –5.5 to 4.3; P = .95). The frequency of nonspecific symptoms in patients was similar to that in controls.
Conclusions and Relevance
The long-term outcome at 12 months after treatment was comparable, regardless of dissemination. Follow-up of at least 12 months after treatment is thus recommended for future studies that investigate post–Lyme borreliosis symptoms.
Multiple erythema migrans (MEM) is the most frequent manifestation of early disseminated Lyme borreliosis1 and is interpreted as the consequence of hematogenous dissemination of borreliae.2 It is defined as the presence of 2 or more erythema, 1 of which should fulfill the criteria for solitary EM.3 Guidelines by the Infectious Diseases Society of America do not make distinct treatment recommendations for solitary EM and for MEM. The guidelines recommend oral cefuroxime axetil or amoxicillin for 14 to 21 days or oral doxycycline for 10 to 21 days and intravenous ceftriaxone sodium for those with neurological involvement or advanced atrioventricular heart block4; however, some data suggest that MEM might predispose patients to less favorable treatment outcome compared with solitary EM.5-7 Treatment of MEM has been specifically addressed in 1 study performed in the United States8 and in 1 European report9; other treatment studies included predominantly or exclusively patients with solitary EM.5-7,10-15
The aim of the present study was to further investigate the potential differences in clinical course and treatment outcome between patients with MEM and patients with solitary EM.
The study was approved by the Medical Ethics Committee of the Ministry of Health of the Republic of Slovenia. Written informed consent was obtained from all study participants.
Patients aged 18 years or older evaluated between June 1, 2010, and October 31, 2015, at the University Medical Center Ljubljana, Slovenia, were prospectively enrolled in the study. Data were analyzed from June 1, 2017, to January 3, 2018. These patients had either MEM or solitary EM, defined according to European criteria, as assessed by 4 of us (D.S., V.M., P B., and K.O.).3 Solitary EM was defined as an expanding red or bluish-red plaque, with or without central clearing, that develops days to weeks after the bite of a tick or exposure to ticks in a region in which Lyme borreliosis is endemic. For a reliable diagnosis, the erythema must reach 5 cm or greater in diameter. If it is less than 5 cm in diameter, a history of tick bite, a delay in appearance of at least 2 days, and an expanding erythema at the site of the tick bite are required.2 Multiple erythema migrans was defined as the presence of 2 or more erythema, at least 1 of which must fulfill the size criteria for solitary EM.1
Patients were excluded if they were pregnant or lactating; had received an antibiotic with known anti-borrelial activity within the previous 10 days; had an extracutaneous manifestation of Lyme borreliosis; had received the study antibiotic for fewer than 10 days; or had a new episode of the disease during follow-up. On the basis of open-label alternate treatment observational study protocol, patients with MEM were assigned to 14-day treatment consisting of either oral doxycycline 100 mg twice daily or intravenous ceftriaxone 2 g once daily according to their arrival order at the clinic so that every second patient received 1 of the 2 treatment options. Data on comparison of pretreatment characteristics, adverse effects of therapy, and treatment outcomes according to antibiotic regimen in patients with MEM are published elsewhere.9 Because the doxycycline treatment was not inferior to the ceftriaxone treatment,9 the 2 treatment groups of patients with MEM were combined for further comparison with patients with solitary EM. For each patient with MEM, 2 consecutive patients with solitary EM were evaluated for enrollment. Patients with solitary EM were treated with oral doxycycline 100 mg twice daily for 14 days.
Each patient was asked to refer a family member or a friend of similar age (±5 years) without a history of Lyme borreliosis to serve as a control participant.
At baseline and follow-up at 14 days and 2, 6, and 12 months, patients were given a physical examination and asked open-ended questions about their health-related symptoms and consumption of any extra antibiotics since the previous visit. Symptoms that had newly developed or worsened since the onset of EM that had no other known medical explanation were regarded as Lyme borreliosis–associated constitutional symptoms at enrollment or as post–Lyme borreliosis symptoms at follow-up.
At each visit, patients completed a written questionnaire asking whether they had had any of 8 nonspecific symptoms (ie, fatigue, arthralgias, headache, myalgias, paresthesias, memory difficulties, concentration difficulties, and irritability) within the preceding week.
At enrollment, lumbar puncture was proposed for patients with MEM and was performed in those who consented to the procedure.
Complete response to treatment was defined as a return to pre–Lyme borreliosis health status. Partial response was defined as the presence of post–Lyme borreliosis symptoms, whereas the occurrence of new objective signs of Lyme borreliosis, persistence of borreliae as detected by culture of skin re-biopsy sample, and/or persistence of EM 2 or more months after treatment were interpreted as failure. Persistence of EM was defined as EM that could still be seen in daylight and at room temperature. We grouped failures and partial responses together as incomplete response.
Control participants completed the same 8-symptom questionnaire within 14 days of the examination date of the corresponding patient at enrollment and again at 6 and 12 months. They were also asked to list their comorbidities.
Patients and controls graded the severity of each individual symptom on a 10-cm visual analog scale (score range: 0-10, with 10 indicating the most severe).16
Cerebrospinal fluid samples collected at enrollment were analyzed for cell counts as well as protein and glucose levels. A leukocyte count greater than 5 × 106 cells/L (pleocytosis) was considered abnormal. IgM, IgG, and albumin levels were determined in serum and cerebrospinal fluid samples. Serum and cerebrospinal fluid IgM antibodies to OspC and VlsE and IgG antibodies to VlsE borrelial antigens were determined with indirect chemiluminescence immunoassay (LIAISON; DiaSorin), and the results were interpreted according to the manufacturer’s instructions. Intrathecal synthesis of borrelial antibodies was determined, as described by Reiber and Peter,17 as antibody index values greater than 1.4, indicating intrathecal production of borrelial antibodies.
At baseline, a 3-mm punch biopsy skin specimen was obtained from the expanding edge of the primary EM and placed in a 6-mL modified Kelly-Pettenkofer culture medium. If the first skin specimen was culture positive for borreliae, a second specimen was collected from the same site 2 to 3 months after the first biopsy. Cultures were examined weekly by dark-field microscopy for the presence of spirochetes and were interpreted as negative if no growth was established by 9 weeks. Baseline blood samples were cultured for borreliae as previously described.18 Isolates were identified to species level using large restriction fragment pattern analysis after MluI restriction18 or real-time reverse transcriptase–polymerase chain reaction that targets the HBB (U48676.1) gene.19
Categorical data were summarized as frequencies (%) and numerical data as medians (interquartile range [IQR]).
The proportion of patients with incomplete response at 12 months after enrollment and the associated 2-sided 95% CIs for the difference (MEM vs solitary EM) were estimated using the normal approximation with continuity correction. The same analysis was used to estimate the proportion of incomplete response at the other time points and at the final evaluable visit.
The median duration of the EM was calculated using the Kaplan-Meier method. Log-rank test was used to test the difference in duration curves between the 2 treatment groups.
The association between incomplete response and other covariates (eg, sex, age, and presence of Lyme borreliosis–associated constitutional symptoms at enrollment) was estimated using multiple logistic regression analysis, adjusting the model for dissemination and time from enrollment (used as a categorical variable). To account for multiple measurements in each patient, random intercept by patient ID was included in the model. Results were presented as odds ratios (ORs) with 95% CI. A 2-sided P < .05 was considered significant. R software, version 3.0.1 (R Foundation for Statistical Computing) was used for the analyses.
Of the 234 adult patients with MEM and 544 patients with solitary EM assessed for eligibility between June 1, 2010, and October 31, 2015, a total of 200 patients with MEM and 403 patients with solitary EM were included in the present study (Figure). Among the 200 patients with MEM, 94 (47.0%) were males and 106 (53.0%) were females, with a median (IQR) age of 47 (35-58) years. Among the 403 patients with solitary EM, 182 (45.2%) were males and 221 (54.8%) were females, with a median (IQR) age of 55 (42-62) years (Table 1).
At enrollment, patients with MEM compared with patients with solitary EM were younger (median [IQR] age, 47 [35-58] years vs 55 [42-62] years; P < .001); remembered a tick bite less often (66 [33%] vs 204 [50.6%]; P < .001); had smaller primary EM (median [IQR] diameter, 13 [9-20] cm vs 16 [10-26] cm; P < .001); and were more likely to report Lyme borreliosis–associated constitutional symptoms, particularly fatigue (51 [54.8%] vs 35 [36.5%]; P = .02) (Table 1).
After starting antibiotic treatment, the median (IQR) duration of the erythema was the same for patients with MEM (10 [5-18] days) and solitary EM (10 [5-21] days; P = .34). Clinical outcomes in the 2 treatment groups at follow-up visits are summarized in the eTable in the Supplement. Failure was documented in 2 of 200 (1.0%) patients with MEM and in 8 of 403 (1.9%) patients with solitary EM because a residual erythema could still be seen at the 2-month visit. All 10 patients were re-treated with oral antibiotics and had an uneventful further course. None of the patients developed extracutaneous manifestation of Lyme borreliosis.
The proportion of patients with incomplete response, represented predominantly by the presence of post–Lyme borreliosis symptoms, steadily decreased over the study period (eTable in the Supplement). At enrollment, patients with MEM reported Lyme borreliosis–associated constitutional symptoms more frequently than those with solitary EM (93 [46.5%]; 95% CI, 39.4-53.7 vs 96 [23.8%]; 95% CI, 19.7-28.3; P < .001; Table 1). During the initial 6 months after treatment, the proportion of patients with incomplete response was higher in patients with MEM than in those with solitary EM (14 days: 62 of 193 [32.1%] vs 72 of 391 [18.4%]; P < .001; 2 months: 38 of 193 [19.7%] vs 55 of 394 [14.0%]; P = .28; 6 months: 29 of 182 [15.9%] vs 31 of 359 [8.6%]; P = .02). At 12 months, 10 of 170 (5.9%) patients with MEM and 20 of 308 (6.5%) with solitary EM showed an incomplete response (−0.6; 95% CI, –5.5 to 4.3; P = .95). Findings were similar also for the last evaluable visit (14 of 200 [7.0%] vs 30 of 403 [7.4%]). None of the patients with post–Lyme borreliosis symptoms at 12 months qualified as having “post–Lyme disease syndrome” because the reported symptoms were not severe enough to reduce previous activity levels.4
The multiple logistic regression model for repeated measurements of incomplete response (with dissemination of borreliae as the explanatory variable, adjusted for sex, age, time from enrollment, and presence of Lyme borreliosis–associated constitutional symptoms at enrollment) indicated that patients with MEM had higher odds for incomplete response compared with patients with solitary EM, although the difference was not statistically significant (OR, 1.47; 95% CI, 0.91-2.37; P = .11; eTable in the Supplement).
A small proportion of patients received 1 or more courses of antibiotics with anti-borrelial activity for unrelated conditions during the follow-up period: 11 (5.5%) patients in the MEM group and 34 (8.4%) in the solitary EM group (95% CI, –7.4% to 1.6%; P = .26). No statistically significant association was found between patients who received additional antibiotics but had incomplete response at the last evaluable visit and patients who did not take extra antibiotics and also had incomplete response (1 of 45 [2.2%] vs 44 of 556 [7.9%]; P = .27).
Patients with MEM were more often seropositive for borreliae at enrollment compared with patients with solitary EM, but the difference was not statistically significant (137 of 194 [70.6%] vs 237 of 393 [60.3%]; P = .02; Table 1). The same trend continued during follow-up at the 2-month visit (76.9% vs 63.6%; P < .001), at the 6-month visit (76.5% vs 55.4%; P < .001) and at the 12-month visit (66.3% vs 48.6%; P < .001).
Compared with patients with solitary EM, in patients with MEM, Borrelia burgdorferi sensu lato was isolated less often from pretreatment skin biopsy specimens (88 of 161 [54.7%] vs 253 of 382 [66.2%]; P = .01) but more often from blood (5 of 194 [2.6%] vs 2 of 393 [0.5%]; P = .04); however, the differences were not statistically significant (Table 1). Among the skin isolates, 308 of 341 (90.3%) were Borrelia afzelii. Different causative Borrelia species were not associated with higher probability of disseminated disease (Table 1) or with unfavorable outcome. Thus, 28 of 308 (9.1%) patients infected with B afzelii and 3 of 28 (10.7%) patients infected with other identified Borrelia species (Borrelia garinii and B burgdorferi sensu stricto) showed incomplete response at the last evaluable visit (P = .73). A re-biopsy was performed in 74 of 88 (84.1%) patients with MEM who were initially culture positive and in 246 of 253 (97.2%) patients with solitary EM. All the re-biopsies were culture negative.
Basic demographic variables did not differ substantially between the controls and the patients. At enrollment, controls reported nonspecific symptoms more frequently than did patients, but no substantial differences in the frequency were observed at 6 and 12 months after enrollment. Detailed information is provided in Table 2.
To date, no direct comparison of the course and outcome in patients with MEM and patients with solitary EM has been reported. Our comparison has revealed some similarities and several distinctions, for some of which we have potential explanations. After starting antibiotic treatment, the duration of EM was comparable (median, 10 days) between patients with MEM and solitary EM. In both groups, none of the patients qualified as having post–Lyme borreliosis syndrome and none developed extracutaneous manifestations of Lyme borreliosis, the microbiological efficacy was excellent (all skin re-biopsy specimens were culture negative for borreliae), and the incidence of incomplete response 12 months after treatment was comparably low (5.9% in MEM vs 6.5% in solitary EM).
Multiple erythema migrans is more common in children than in adults.20-22 In our study, MEM was more common in younger patients, suggesting that a similar age distribution may be a factor in the adult age group. The smaller primary EM observed in patients with MEM might have been associated with a shorter duration of illness before diagnosis compared with patients with solitary EM (10 vs 14 days), possibly because constitutional symptoms were present more often in patients with MEM, prompting them to look for medical help sooner. However, our data do not support this assumption, as patients with constitutional symptoms attended the first visit even later than those without constitutional symptoms (median [IQR], 14 [6-30] days vs 12 [5-24.8] days). Thus, the multiple erythema themselves might represent a stronger signal to visit a physician.
The greater frequency of Lyme borreliosis–associated constitutional symptoms in patients with MEM could be due to a stronger inflammatory response associated with disseminated infection. The finding that a relatively low proportion of patients with MEM had Lyme borreliosis–associated constitutional symptoms at enrollment contrasts with the findings from the US study, in which patients with solitary EM reported constitutional symptoms more often and experienced hematogenous dissemination, clinically manifested as MEM, more frequently than did patients in the European study.2,12-14,16 These differences have been ascribed to the distinct origin of Lyme borreliosis—nearly exclusively from B burgdorferi sensu stricto in North America and predominantly from B afzelii and B garinii in Europe12-14,16—and to the different inflammatory potential of American and European Borrelia species.23,24 At the 14-day and 6-month visits, patients with MEM more often showed incomplete response to treatment, represented predominantly by the presence of post–Lyme borreliosis symptoms. In addition to the inherent association of disseminated disease with the presence of post–Lyme borreliosis symptoms, the higher proportion of these symptoms at the 14-day visit could also be partly the result of interference from post–lumbar puncture symptoms in some patients.
The Infectious Diseases Society of America guidelines recommends the same management for MEM as for solitary EM.4 However, some data suggest that the presence of MEM might predispose patients to less favorable treatment outcome.5-7 Our direct comparison of patients with MEM with patients with solitary EM clearly showed that disseminated disease presents with Lyme borreliosis–associated constitutional symptoms and more often leads to post–Lyme borreliosis symptoms. Our multiple logistic regression model indicated that patients with MEM had higher odds for incomplete response than those with solitary EM, although the difference was not significant.
Until now, we assumed that no differences existed among Borrelia species that cause early localized and early disseminated skin Lyme borreliosis in Europe, but data to support such an assumption was limited.25 The finding that B burgdorferi sensu lato was isolated less often from skin but more often from blood in patients with MEM than in patients with solitary EM might suggest the existence of variations in dissemination capacity among different strains in individual Borrelia species in Europe.
This study has several limitations. First, some of the post–Lyme borreliosis symptoms in patients could have been misattributed to Lyme borreliosis. Second, the comparable proportion of nonspecific symptoms among patients and controls at 6 and 12 months after enrollment does not exclude that some of these symptoms in patients were triggered by borreliae infection. Third, we did not actively search for borrelial infection in control participants by laboratory means, and some of the nonspecific symptoms they reported might have been due to Lyme borreliosis. Fourth, our results are likely applicable to European regions with similar ratios of Borrelia genospecies that cause EM, but these results may not entirely apply to North America, where Lyme borreliosis is caused by B burgdorferi sensu stricto.3
To our knowledge, this prospective cohort study is the first to report a direct comparison between early disseminated and early localized Lyme borreliosis manifested as MEM and solitary EM. Our results show that the proportion of patients who reported Lyme borreliosis–associated constitutional symptoms at enrollment and post–Lyme borreliosis symptoms at follow-up was higher in those with MEM than in those with solitary EM up until 6 months after treatment; however, the long-term outcome at 12 months after treatment was comparable and excellent, regardless of dissemination. Different causative Borrelia species were not associated with higher probability of disseminated disease. The frequency of nonspecific constitutional symptoms in patients after treatment did not exceed the frequency of these symptoms in controls without a history of Lyme borreliosis; therefore, these symptoms should not be automatically associated with past borrelial infection. We advocate for follow-up of at least 12 months after treatment in future studies in which post–Lyme borreliosis symptoms are assessed.
Accepted for Publication: May 20, 2018.
Corresponding Author: Daša Stupica, MD, PhD, Department of Infectious Diseases, University Medical Center Ljubljana, Japljeva 2, 1525 Ljubljana, Slovenia (email@example.com).
Published Online: August 1, 2018. doi:10.1001/jamadermatol.2018.2306
Author Contributions: Drs Stupica and Strle had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.
Study concept and design: Stupica, Strle.
Acquisition, analysis, or interpretation of data: All authors.
Drafting of the manuscript: Stupica, Strle.
Critical revision of the manuscript for important intellectual content: All authors.
Statistical analysis: Blagus.
Obtained funding: Strle, Stupica.
Administrative, technical, or material support: Maraspin, Bogovič, Ogrinc, Cerar.
Study supervision: Strle.
Conflict of Interests: Dr Strle reported being an unpaid member of the steering committee of the ESCMID (European Society of Clinical Microbiology and Infectious Diseases) Study Group on Lyme Borreliosis (ESGBOR). No other disclosures were reported.
Funding/Support: This study was supported by grants P3-0296 and J3-6788 from the Slovenian Research Agency.
Role of the Funder/Sponsor: The funder had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.
Create a personal account or sign in to: