Erythema migransfollowing ACA Case report ERYTHEMA MIGRANS FOLLOWING ACRODERMATITIS CHRONICA ATROPHICANS - FURTHER EVIDENCE FOR POSSIBLE REINFECTION WITH BORRELIA BURGDORFERI R.R. Miillegger, H.P. Soyer and S. Hodl ABSTRACT A 57-year-old woman, living in an area endemic for Lyme boITeliosis (LB), presented with erythema migrans (EM) that developed four weeks after a tick bite on the left thigh. The serum ELISA IgG antibody tiu·e against Borrelia burgdorferi (Bb) was highly positive. The patient was treated with minocyclin (200 mg/day, 14 days). Upon completion of therapy, the EM had cleared completely. Two years earlier, however, a diagnosis of acrodermatitis chronica atrophicans (ACA) on the left forearm and hand had been made on clinical as well as histopathological grounds. The serum ELISA IgG antibody titre against Bb was highl y positive at that tirne also. A course of treatrnent with peroral amoxicillin (500 mg) and clavulanic acid ( 125 mg) three times a day (three weeks) led to marked improvement of the ACA skin changes. No more signs of this inflarnmation were found when the patient presented with EM. Many questions conceming molecular and immunologic aspects ofLB remain open, especially the extent to which Bb antibodies protect against reinfection. The occurrence ofEM in a patient who has previously had ACA may be explained by reinfection with Bb. Thus, the presented case substantiates the hitherto held suspicion that infection with Bb does not necessarily lead to protective immunity. KEYWORDS Lyme boneliosis, acrodermatitis chronica au·ophicans, erythema migrans, reinfection, immunity acta dermatovenerologica A.P A. Vo/ 2 93 No 4 119 Erythema migransfo/lowing ACA INTRODUCTION Lyme borreliosis (LB) is caused by the spirochete Borrelia burgdorferi (Bb), of which at least three subtypes have been described (Bb sensu stricto, Borrelia garinii, Bb group VS 461) (1 ,2). LB can be subdividedinto three stages, andnearly ali organs may be affected. Skinmanifestations are seen most frequently, and may occur during ali stages of the disease . Erythema migrans (EM) is the hallmark ofthe early stage of LB, developing at the site of inoculationofBb. Acrodermatitis chronica atrophicans (ACA) is a late, chronic skin manifestation ofLB, arising months to years after infection with Bb. According to current knowledge, it appears that the human immune response to Bb infection does not necessaril y lead to protection against renewed Bb infection (3) . Sirnultaneous or subsequent OCCU1Tence of EM and ACA, due either to superinfection orreinfection, have been reported (4,5,6,7). Superinfection is possible by transmission of two different subtypes of Bb (5). Reinfection may be explained bythelackofprotectiveimrnunityresultingfromBpinfection (3) . This report considers a 57-year-old woman, living in an area endemic for LB, who presented with EM on her left thigh, and had a history of ACA two years previously. This situation may be explained by reinfection with Bp. CASEREPORT In December 1991, a then 55-year-old woman was seen at the Departrneilt of Dermatology in Graz, Austlia. She presented with an one year history of a diffuse red swelling together with signs of skin atrophy on the ulnar aspect of her distal left forearm and hand, consistent with the clinical diagnosis of ACA (Fig. 1). The semm ELISA IgG antibody titre against Bb was highly positive; no IgM antibodies were found. Histopathologic examination of a punch biopsy, taken from the left forearm, confirmed the clinical diagnosis . Said examination revealed a perivascular and interstitial in- flamrnatory infiltrate within the dermis, cytomorphologicall y characterized by a predominance oflymphocytes and plasma cells (Fig. 2). Peroral treatment with amoxicillin (500 mg) and clavulanic acid (125 mg) three tirnes a day was given for three weeks, which led to marked improvement of the skin changes. Two years later, the patient was seen again, this tirne with a ring-shaped, sharply demarcated erythematous skin lesi on of30 cm diameter on the left thigh, clinically consistent withEM (Fig. 3). The erythemafirst occurred 6 weeks before admission, and expanded centrifugally during this tirne . Four weeks before the onset of EM, the patient was bitten on the left thigh by a tick in a geographic region endemic for LB. Inguinal lymph nodes were swollen, and the patient complained of malaise. A punch biopsy from the border of the erythematous lesion was taken. Histopathologic examination revealed a moderately dense, superficial and deep, perivascular infiltrate of lymphocytes, histiocytes and plasma cells, thus suppmting the clinical diagnosis of EM (Fig. 4). The left foreaim and hand, previously affected by ACA. displayed skin which was wrinkled, but no sign of inflammation. The sernm ELISA IgG antibody titre against Bb was highly positive, the IgM tiu·e was negative. Blood chemistry was within normal limits, only the BSR was slightly elevated. An ECG and echogram of the heart, a neurologic exa.Il1ination, and elecu-omyography (upper and lowerextremities) clisclosedno pathologicfindings. Treatment Figure 2: Histopathology of acrodermatitis chronica au·ophicans. Epidem1al au·ophy, bandlike inflan1matory infilu·ate composed ofly~phocytes_and ~lasmacells, and telangiectatic blood vessels in the upperde1mis; reduction ofthe breadth of the dermis. H&E, scanmng magnificat10n. 120 acta dermatovenerologica A.P.A. Vol 2. 93 No 4 Erythema migransfollowing ACA Figure 4: The histopathological features from a biopsy specimen of the left thigh are consistent with erythema migrans. H&E, scanning magnification. withminocycline (2 x 100mg daily, 14 days) was instituted. By the end of therapy, the EM had cleared completely. The IgG antibody titre against Bb remained highly positive, the IgM titre was stili negative. DISCUSSION EM is the typical manifestation of early LB, arising at the inoculation site of Bb, days to months after a tick or insect bite. Initially, a homogeneous red or bluish-red lesi on occurs. Later, a bright red, expanding ring with a fading cenu·e develops in most cases. Serology is not very helpful in diagnosis, because positive antibody titres against Bb are only found in 20 - 50 % of the cases (8) . Histopathology is nonspecific, but some features - as found in the EM specimen of the patient described - may support the diagnosis, namely a superficial and deep perivascular infilu·ate, composed of lymphocytes and plasma cells. EM usually disappears within a few months, even without therapy. Nevertheless, early treatment with oral antibiotics is necessa.ry to prevent complications. ACA is a late skin manifesta.ti on of LB with a chronic, and usually progressive course. It begins months to years after Bb infection with an acute inflarnmatory stage. The predilection sites are the acral body parts. It predominates in women and in thelater decades oflife. The skin lesions gradually develop into the chronic au·ophic stage. The diagnosis of ACA is acta dermatovenerologica A.P A. \lol 2, 93, No 4 primarily ba.sed on clinica.l criteria.. The histopathologic features, as observed in our patient, are characteristic, and confinn the diagnosis of ACA. Elevated IgG antibody titres against Bb are found in nearly all patients with ACA (9). On the other hand, seroprevalence in Europe ranks between 3 % and more than 40 % (10, 11). In our patient, IgG antibodies against Bb were highly positive in 1991, a well known finding in patients with ACA (9). It is not possible to determine the extent to which EM has influenced the stil! highl y positive IgG antibody ti tre in 1993. Many immunologic issues in LB remain unclear. An essential point concerns the question, whether or not Bb antibodies protect against a repeat infection, and if so, for how long. Animal experiments ha.ve demonstrated the capability of Bb antibodies in defence against infection. However, spirochetes alreadyresiding in the tissue cannot be effectively attacked by these antibodies (12, 13). Moreover, the protective effect of these antibodies seems to be limited (14) . Fram human medicine, it is known that specific antibodies and T-cells develop in LB (15 , 16). However, it is unlikely that these mechanisms provide uuly a protective immunity (3). Reports of the subsequent occurrence of the same or different manifestations ofLB support the possibility ofreinfection due to a lack of such protective immunity, As early as 1955, Hauser desci·ibed EM preceding ACA in a few patients (17). In the recent literature, a preceding EM ha.s been found in about one fifth of a Swedish study group of 50 121 Erythema migransfollowing ACA patients withACA (6). InaGermanstudy, iout of21 patients with ACA had a history ofEM (7). Furthermore, Weber and co-workershave observedreappearence ofEM in two patients. These secondary EM usually developed more than one year after the first EM in another part of the body ( 18). Possible reinfection has also been described by Pfister and co-workers in 1986 (19). There are also rep01ts about EM arising in patients already suffering from ACA, possibly due to superinfection. ( 4, 20). Recently, Wienecke and co-workers published the case of a patient presenting with ACA following a tick bite 4 years before he first came to medical attention (5). Three weeks later, the patient was bitten by another tick, and developed EM at this site. Biopsies were taken, and a Bb-specific gene segment was successfully amplified by PCR from both ACA and EM. Molecular subtyping revealed the VS 461 group of Bb in the ACA lesion. Bb sensu stricto and Borrelia garinii were identified within the EM. Thus, the case of Wienecke and co-workers may be interpreted asa superinfection by two borrelia subtypes during persistent infection with a third one. The authors conclude that an infection with one Bb subtype does not lead to irnmunity against infection with other subtypes. Considering the presented literature, various interpretations are possible of the subsequent or simultaneous appearence of the same or different manifestations of LB. Reinfection ( 6, 7, 17, 18, 19) might be due to an insufficient immune response to one ofthe three subtypes ofBb (Bb sensu stricto, Borrelia garinii, Bb group VS 461). On the other hand, reinfection could be caused by a Bb subtype different from the one in primary infection, against which no specific antibodies have been produced. Due to the permanent change in surface proteins of Bb after it has persisted in tissue for a long tirne (21), "reinfection" could also be caused by activation of Bb (e.g. during another infection (22)) following antigenic shift. Occurrence ofEM simultaneously with ACA (4, 5, 20) may be explained by a superinfection with another Bb subtype (5). Furthermore, heterogeneity of Bb strains has been demonstrated, even in patients from areas geographically very close to one another (23). In conclusion, this case report of a patient who developed EM two years after having had ACA lends further weight to the concept of reinfection with Bb. Our observation fmther emphasizes that protective irnmunity does not necessarily develop in patients with LB. REFERENCES l. Baranton G, Postic D, Girons IS. Delineation of Borrelia burgdorferi sensu stricto, Borrelia garinii sp nov, and group VS 461 associated with Lyme borreliosis. Int J Syst Bacteriol 1992; 42: 378-83 2. Wienecke R, Koch OM, Neubert U, Gobel U, Volkenandt M. Detection of subtype-specific nucleotide sequence differences in a Borrelia burgd01feri specific gene segment by analysis of conformational polymorphisms of cRNA molecules. Med Microbiol Lett 1993; 2: 239-46 3. Schaible UE, Wallich R, Kramer MD, Museteanu C, Ritting M, Moter S, Simon MM. Role of the immune response in Lyme disease: lessons from the mouse model. In: Schutzer SE (ed): Lyme disease - Molecular and Immunologic approaches. 1992; Cold Spring Harbor Laboratory Press. pp 243-262 (Current communications in cell and molecular biology) 4. Asbrink E, Hovmark A, Hederstedt B. The spirochetal etiology of acrodermatitis chronica atrophicans Herxheirner. ActaDerm Venereol (Stockh) 1984; 64: 506-512 122 5. Wienecke R, Neubert U, Volkenandt M. Cross immunity among types of Borrelia burgdorferi. Lancet 1993; ii: 435 6. Asbrink E, Hovmark A, Olsson l. Clinical manifestations of acrodermatitis chronica atrophicans in 50 Swedish patients. Zbl BaktHyg (A) 1986; 263: 253-61 7. Weber K, Neubert U. Clinical features of early erythema migrans disease and related disorders. Zbl Bakt Hyg (A) 1986; 263: 209-28 8. Asbrink E, Hederstedt B, Hovmark A. The spirochetal etiology of erythema chronicum migrans Afzelius. Acta De1m Venereol (Stockh) 1984; 64: 291-95 9. AsbrinkE, Hovmark A, Hederstedt B. Serologic studies of erythema chronicum migrans Afzelius and acrodennatitis chronica atrophicans with indirect immunofluorescence and enzyme-linked immunosorbent assays. Acta De1m Venereol (Stockh) 1985; 65: 509-14. 10. Schmutzhard E, Stanek G, Pletschette M, Hirschl AM, acta dermatovenerologica A.P.A. Vol 2 93 No 4 Erytlzema migrans following ACA Pallua A, Schmitzberger R, Schlogl R. Infections after tickbites. Tick-bome encephalitis and Lyme borreliosis - a prospective epidemiological study from Tyrol. Infection 1988; 16: 269-72 11. Gem L, Frossard E, Walter A, Aeschlimann A. Presence of antibodies against Borrelia burgdorferi in a population of the Swiss plateau. Zbl Bakt (Suppl) 1989; 18: 321-28 12. Johnson RC, Kodner C, Russell M. Passive immunization of hamsters against experimental infection with the Lyme disease spirochete. Infect Immun 1986; 53: 713-14 13. Schmitz JL, Schell RF, Hejka AG, England DM. Pas sive immunization prevents induction of Lyme arthritis in LSH Hamsters. Infect Immun 1990; 58: 144-48 14. Schmitz JL, Schell RF, Lovrich D, Callister SM, Coe JE. Characterization of the protective antibody response to Borrelia burgdorferi in experimentally infected LSH hamsters. Infect Immun 1991; 59: 1916-21 15. Barbour AG, Burgdorfer W, Grunwaldt E, Steere AC. Antibodies of patients with Lyme disease to components of the Ixodes darnmini spirochete. J Ciin Invest 1983; 72: 504- 15 16.DattwylerRJ, VolkmanDJ,Luft BJ,HalperinJJ, Thomas J, Golightly MG. Specific immune response in Lyme borreliosis: Characterization of T cell and B cell response to Borrelia burgdorferi. Ann NY Acad Sci 1988; 539: 93-102 17. Hauser W. Zur Klinik, Atiologie und Pathogenese der Acrodermatitis chronica atrophicans. Hautarzt 1955; 6: 77- 80 18. Weber K, Schierz G, Wilske B, Neubert U, Krampitz HE, Barbour AG, Burgdorfer W. Reinfection in erythemamigrans disease. Infection, 1986; 14: 32-5 19. Pfister HW, Neube1t U, Wilske B, Preac-Mursic V, Einhaupl KM, Borasio GD. Reinfection with Borrelia burgdorferi. Lancet 1986; ii: 984-5 20. Hauser W. Wahrscheinliche Infektionskrankheiten der Haut. In: Marchionini A, Gotz H ( eds ): Infektionskrankheiten der Haut I. 1965; Springer, Berlin-Heidelberg-New York. pp 555-629 (Handbuch der Haut-und Geschlechtskrankheiten, IV, 1 A) 21. Wilske B, Preac-Mursic V, Schierz G, Gueye W, Herzer P, Weber K. Immunchemische Analyse der Immunantwort bei Spatrnanifestationen der Lyme Borreliose. Zbl Bakt Hyg (A) 1988;267:549-58 22. Millner MM, Schimek MG, Miillegger RR, Stanek G. Borrelia burgdorferi ELISA titres in children with recent mumpsmeningitis. Lancet 1990; ii: 125-6 · 23. Khanakah G, Millner MM, Miillegger RR, Stanek G. Preliminary characterization of B01relia burgdorferi CSF isolates. Infection 1991; 19: 287-88 AUTHORS' ADDRESSES Robert R. Miillegger M.D., Departrnent ofDermatology, Auenbruggerplatz 8, University of Graz, A-8036 Graz, Austria H. Peter Soyer M.D., associate professor, same address Stefan Hodi M .D., professor, same address acta dermatovenerologica AP A. Vol 2, 93, No 4 123