Search for a command to run...
Background and objective Diagnosis of early Lyme disease (LD) often relies on clinical recognition of the skin lesion, erythema migrans (EM), a diagnostic sign of disease when laboratory testing is insensitive. Because EM can present in morphologically distinct forms, its recognition by clinicians can be challenging. This study aimed to characterize the clinical spectrum of lesions in patients presenting with suspected early LD in an ambulatory care setting to identify features that might help clinicians to be better prepared to recognize EM lesions. Methods Images of lesions from 69 participants suspected to have early LD were retrospectively evaluated by a dermatologist and a family practitioner with expertise in early LD. Reviewers made determinations on the diagnoses and morphological features of lesions. Agreement between reviewers and associations among lesion types and participant demographics, symptomology, and laboratory evidence of infection were examined using the kappa statistic and contingency tables, respectively. Results Challenges in diagnosing EM were evident in our study: initial concordance between reviewers was moderate [kappa statistic (95% CI): 0.45 (0.245 - 0.657)]. The final classification included 35 lesions (51%) that were agreed to be EM; 23 lesions (30%) were considered to be possible early EM or tick bite reactions, and 11 (16%) were thought not to be EM, but rather other diagnoses, including ringworm, allergic contact dermatitis, and mosquito bites. Only two lesions (6%) were observed with a classic bull's eye or ring-within-a-ring pattern. Most EM lesions were uniform (51%), pink (74%), oval lesions (63%), with well-demarcated borders (92%). Early EM or tick bite reactions were typically <5 cm in size (74%), red (52%), round lesions (61%), with a punctum present (100%). Lesions thought not to be EM also tended to be pink or red (64%), round (55%), or uniform (45%) lesions, but also had raised (25%) or irregular borders (33%), which were not commonly observed in the reviewer-classified EM or tick bite reaction lesions. Participants with lesions classified as EM reported that they had the lesions for more days (p = 0.043) and reported more symptoms (p = 0.017) than participants with other lesions. Only 14 (20%) participants overall had positive laboratory evidence for LD; these included 13 (37%) of the participants with EM-classified lesions. Conclusions EM commonly occurs in forms that are not the classic bull's eye. Patients often present with lesions that may represent the very early stage of EM or tick bite reactions, and most patients will test negative on currently available laboratory tests, challenging clinicians in making an LD diagnosis or treatment decisions. Additional studies to further characterize the morphological features of EM and how variation in skin lesions may be perceived among clinicians would be helpful for developing guidelines on improving clinician recognition of EM.