Case Report

Top Case Report Methods Results Discussion Author & Article Info References

A 60-year-old woman presented on 15 July 1991 with a single 11 x 12-cm erythema migrans lesion on her right upper thigh that she had noticed 5 days before presentation. She was otherwise asymptomatic. Culture of a skin biopsy sample grew B. burgdorferi. The rash resolved after 14 days of oral amoxicillin therapy (500 mg three times daily). Two years later (on 12 July 1993), the patient presented with one 19 x 24-cm erythema migrans lesion on the right popliteal fossa that had developed 12 days before presentation. Culture of a skin biopsy specimen was negative for B. burgdorferi, and the rash resolved after 10 days of oral doxycycline therapy (100 mg twice daily). On 24 June 1994, the patient returned with a single 12.5 x 9.5-cm erythema migrans lesion on the left axilla that had been present for 1 day. The patient was asymptomatic at presentation; 12 days before presentation, however, the patient had had a body temperature of 38.3 °C and developed a headache that lasted 1 day. Borrelia burgdorferi was recovered from culture of a biopsy specimen obtained from the erythema migrans lesion. The rash resolved after 10 days of oral doxycycline therapy (100 mg twice daily). For each episode of erythema migrans, therapy began on the first day of presentation.

Methods

Top Case Report Methods Results Discussion Author & Article Info References

Serum samples that had been frozen at –70°C were simultaneously tested for antibodies to B. burgdorferi by using a polyvalent ELISA (Whittaker Bioproducts, Inc., Walkersville, Maryland) and separate IgG and IgM immunoblots that were done with Marblot Strip Test System (MarDx Diagnostics, Inc., Carlsbad, California). Both tests were performed according to the manufacturer's instructions.

Punch biopsy specimens (2 mm in diameter) obtained from the leading edge of the skin lesions were cultured in modified Barbour-Stoenner-Kelly medium. A culture was considered positive if spirochetes were seen on microscopy and were identified by PCR analysis as B. burgdorferi [6].

Culture aliquots of 0.5 mL were processed for PCR amplification by using primers P_A (5'-GGTATGTTTAGTGAGGG) and P95 (5'-GGTTAGAGCGCAGGTCTG) and were subjected to restriction fragment length polymorphism analysis as described elsewhere [7].

Results

Top Case Report Methods Results Discussion Author & Article Info References

Results on polyvalent ELISA and separate IgG and IgM immunoblot for antibodies to B. burgdorferi are shown in the (Table 1). Polymerase chain reaction-restriction fragment length polymorphism analysis was performed on the B. burgdorferi isolates obtained from the 15 July 1991 and 20 June 1994 lesions and on reference strains B31 and 297. Restriction fragments obtained after digestion with Hinf I or Mse I showed distinct patterns indicating that the two clinical isolates were different strains of B. burgdorferi.

Discussion

Top Case Report Methods Results Discussion Author & Article Info References

Although reinfection with B. burgdorferi has been suspected clinically [2-5], ours is the first reported case in which B. burgdorferi was isolated during separate episodes of erythema migrans. Polymerase chain reaction-restriction fragment length polymorphism analysis indicated that two distinct molecular types of B. burgdorferi were involved. The different molecular types isolated, the occurrences of erythema migrans during the summer (when cases of early Lyme disease would be expected), the history of appropriate antibiotic therapy with resolution of illness, and the presence of only one lesion at different cutaneous sites all support the contention that the patient was reinfected with B. burgdorferi. A more remote possibility is that the patient had been infected with two different strains of B. burgdorferi simultaneously or separately during a previous episode of erythema migrans and then had relapse with the second strain. Although unlikely, this scenario is at least theoretically possible [8].

Serologic testing showed that polyvalent ELISA may be better than immunoblot as an indicator of reinfection with B. burgdorferi (Table 1). Polyvalent ELISA documented seroconversion during reinfection, whereas immunoblot primarily revealed an intensification of existing bands rather than expansion. On immunoblot testing, reinfection did not elicit IgM antibodies to additional borrelial antigens. This finding is important because the IgM immunoblot has been considered a confirmatory serologic test for early infection [9]. Previous studies in which immunofluorescence testing was done on serum samples from patients believed to have a second episode of erythema migrans have also noted the absence of a significant IgM antibody response to B. burgdorferi [3-5]. However, too few cases have been studied serologically to conclude that lack of an IgM response is characteristic of reinfection with B. burgdorferi.

Our findings suggest that reinfection with B. burgdorferi may occur and may lead to recurrent episodes of erythema migrans. A previous episode of early Lyme disease should not be interpreted by the patient or physician as providing immunity to subsequent episodes; all patients living in or traveling to areas endemic for Lyme disease should take measures to prevent tick bites.

Drs. Schwartz and Liveris: Department of Biochemistry and Molecular Biology, Room 111, Basic Science Building, New York Medical College, Valhalla, NY 10595.

Dr. Aguero-Rosenfeld: Clinical Laboratories, Room 1J-11a, Westchester County Medical Center, Valhalla, NY 10595.