- Email: [email protected]
Chemoprophylaxis for Lyme borreliosis?
Zent.bl. Bakteriol. 289, 655-665 (1999) © Urban & Fischer Verlag http://www.urbanfischer.de/j oumals! zblbakteriol
Zentralblatt fUr
Forum paper
Chemoprophylaxis for Lyme Borreliosis? Gerold Stanek 1 and Olaf KahIl 1
2
Hygiene Institute, University of Vienna, Vienna, Austria and Institute of Microbiology and Hygiene, Charite, Humboldt-University, Berlin, Ger many
Summary Chemoprophylaxis is a term which describes treatment with an antimicrobial chemo therapeutic before, during or shortly after an actual or suspected exposure to an infec tious agent in order to prevent clinical disease, which may be severe or even fatal. Lyme borreliosis is considered the most frequent ixodid-tick-transmitted human bacterial infection in the northern hemisphere. For several years there has been a debate on the prophylactic application of antimicrobial chemotherapeutics after an attached Ixodes tick was removed. Would this measure prevent a subsequent borrelia infection and would it be practical? People are exposed to tick-bites mostly during lei sure spent in recreational areas which are often tick infested. The proportion of I. ric inus ticks infected with B. burgdorferi s.l. varies from area to area and in a given area also from year to year (infection rate up to a maximum of 55 %). The transmission rate strongly depends on the duration of feeding, but it could be shown that the criti cal time of feeding is much shorter for European I. ricinus than for the North American I. scapularis or I. pacificus ticks. Nevertheless, even the low risk of complications de spite the very good chance of treating erythema migrans successfully seems to justify prophylactic treatment for some investigators whilst others do not see an argument for this. Double blinded studies in the USA showed a relatively low frequency of illness after vector tick-bite and absence of disseminated disease manifestations. The efficacy of prophylactic antibiotic treatment after tick-bites is not established. Suggestions to examine removed ticks for borrelia in order to obtain indication for prophylactic anti biotic treatment will fail in practice because of high costs and uncertainty in verifying the transmission. Do we need blinded studies in central Europe on a representative number of cases, although it is known that Lyme borreliosis can be treated effectively even in its second and third stage and has never caused a fatal outcome? We conclude that only a reliable diagnosis of symptoms is the basis for a rational antibiotic treat ment, and that instead of chemoprophylaxis for Lyme borreliosis after a vector tick bite the wait and watch policy is recommended. Key words: Lyme borreliosis, chemoprophylaxis, tick attachment, incidence
0934-8840/99/289/5-7-655 $12.00/0
656
G. Stanek and
o. Kahl
Introduction Chemoprophylaxis is a term which describes treatment with an antimicrobi al chemotherapeutic before, during or shortly after an actual or suspected ex posure to an infectious agent in order to prevent clinical disease, which may be severe or even fatal. In our geographical area of moderate climate chemo prophylaxis is used selectively or moderately selectively. Chemoprophylaxis of bacterial infection differentiates between prevention of infection (e. g. expo sure to pertussis, scarlet fever, meningococcal meningitis, diphtheria, syphilis, and tuberculosis), prevention of recurrences (e. g. absolute indication to avoid re-infection with A streptococci after rheumatic fever), and precautions against complications (in order to prevent infection after trauma or during surgical interventions e. g. in organs which are naturally colonised with bac teria, such as the intestine or simply tooth extraction etc.). Currently, there is no generally accepted recommendation for anti-bacterial chemoprophylaxis of tick-bites. Lyme borreliosis is caused by spirochaetal bacteria of the genus Borrelia (B. burgdorferi sensu lato) and transmitted by Ixodes ticks (Fam. Ixodidae, hard ticks). Although haemato phagous insects have occasionally been found to be carriers of borreliae their vector status is unproven. Discussion about the chemoprophylaxis of Lyme borreliosis started with the suggestion to take antibiotics immediately after removal of an attached tick to kill any transmitted borreliae as early as possible in order to prevent any further complication. Is it reasonable to administer antibacterial chemo therapeutics after each vector tick-bite in order to prevent the entire clinical spectrum of Lyme borreliosis? Is the infection rate of ticks and the transmis sion rate of borrelia that high and the manifestation rate of Lyme borreliosis that frequent that such a measure can generally be recommended?
Infection rate of ticks The infection rate of unfed, questing vector ticks with B. burgdorferi is a prin cipal factor for the human infection risk and has been studied in different parts of Europe with different methods since the eighties. Results of selected stud ies (emphasis on central Europe) show a broad range of I. ricinus infection rates with B. burgdorferi s.l. with a maximum rate of up to about 55 % (1,4, 7, 8, 9, 10, 15, 17, 18, 21, 33, 34, 35). According to these findings one may deduce that the risk of becoming infected with borreliae by a vector tick-bite ranges from to 55 % depending on the developmental stage of the tick and the area where the tick was encountered (7).
°
Lyme borreliosis chemoprophylaxis
657
Attachment period and transmission rate It seems helpful to remove an attached tick as soon as possible since we know from the results of several studies that transmission of B. burgdorferi s.l. is closely related to the feeding period. Piesman and colleagues (19,20) were the first to study the duration of attachment required for the transmission of bor reliae in North America. Briefly, Piesman et al. found in their animal experi ments that only 7 % of the rodents (white-footed mice) became infected with B. burgdorferi after an exposure of 24 hours, and only 36 % after 48 hours of exposure to feeding I. scapularis nymphs but 93 % after 72 hours (Table 1). Unfortunately, it is unknown whether the results obtained in an animal experiment can be exactly applied to the Borrelia transmission in humans. Surprisingly long attachment periods were found in humans of the north eastern USA, a highly endemic area of Lyme borreliosis. The calculated feed ing time for nymphs was about 35 hours mean, and the attachment time in creased with the age of the victim (6). The duration of nymphal and female I. scapularis tick attachment was determined by the scutal index (body length/scutal width). Whether persons with a prolonged tick attachment con stitute a special high-risk group for infection was studied by Sood et al. (24). The answer was YES; the incidence of B. burgdorferi infection was signifi cantly elevated for attachment periods of;:::: 72 hours with 3 out of 15 subjects becoming seropositive. One should emphasize, however, that altogether only 4 out of 105 subjects subsequently developed erythema migrans in that pros pective study. This means that the overall manifestation risk was only 3.8 %. This figure is almost identical with that found in a study on recruits in Aus tria who were exposed to vector tick-bites during outdoor activities. Only 2 Table 1. Animal experiments undertaken to determine the transmission of borreliae in relation to the duration of tick attachment Author
Experimental animal, tick species
Duration of attachment
Transmission
Piesman et al. 1987 USA Piesman et al. 1991 USA Peavey and Lane 1995 USA
hamster, white-footed mouse Ixodes scapularis rabbit Ixodes scapularis deer-mouse Ixodes pacificus
Kahl et al. 1998 Germany
gerbil Ixodes ricinus
24 h 48 h 72h 48 h >120 h 48 h 72h >96h 17 h 29 h 47 h 65 h
7% 36% 93% >60% 100% 11 % 25% 80% ca 50% ca 50% 100% 100%
658
G. Stanek and
o. Kahl
out of 50 soldiers (4 %) developed an erythema migrans, and no further man ifestation was recorded after antibiotic treatment (22). In contrast to observations in North America, anecdotal reports of patients and physicians in central Europe suggest that a tick-feeding period of only a few hours can be sufficient to become infected with borreliae and to develop disease. In Austria and in Slovenia records of patients with erythema migrans were evaluated independently and ended up with almost identical results (31). It became obvious that most of the patients discovered the attached tick on the day after the putative tick attachment. In the majority of patients the feeding period of ticks until detection and removal was more than 12 hours and in about a third more than 24 hours in the two countries (Fig. 1). How ever, there was a certain· proportion of exposed persons (14-18 %) who removed the tick within 6 hours and subsequently developed erythema mi grans. Whether this is due to a quicker passage of spirochaetes from the mid gut of I. ricinus to its salivary glands or due to spirochaetes which are already in the salivary glands of unfed I. ricinus awaits clarification. The latter has been shown by the results of a study conducted in Switzerland (13); 11 % of unfed nymphal ticks carried borreliae of the species B. afzelii in their salivary glands. Additionally, in animal experiments with I. ricinus nymphs it was demon strated that gerbils develop eventual spirochaetal infection in about 50 % after a feeding period of less than 17 hours (11). These results underline the fact that the attachment period required to transmit borreliae is significantly short er with I. ricinus nymphal ticks than with I. scapularis (Table 2).
40 DADSL
35
~
~
30 ~
25
~c-
20
1ft
15
o
~
i
~,---
,---
10 5 0
I
, to 6
to 12
to 24
> 24
Hours of feeding
Fig. 1. Time period of attachment before a vector tick was removed from the skin. Re ports about the estimated duration of tick attachment when patients presented with erythema migrans. Results from studies in Austria (A, n = 254) and in Slovenia (SL, n = 212) (ref. 31).
Lyme borreliosis chemoprophylaxis
659
Table 2. Results of studies on the efficacy of prophylactic administration of antibacterial chemotherapeutics after vector tick-bite in order to avoid subsequent Lyme borreliosis (LB) Authors
Costello et al. 1989 Shapiro et al. 1992 Agre and Schwartz 1993 Korenberg et al. 1996
Country Tick species USA USA USA Russia
Ixodes scapularis Ixodes scapularis Ixodes scapularis Ixodes persulcatus
Treatment Group
Placebo Group
LB/total
(%)
LBltotal
(%)
0/27
(0.0)
1/29
(3.4)
0/192
(0.0)
2/173
(1.2)
0/89
(0.0)
1/90
(1.1)
3/261
(1.1)
12/97
(12.4)
Manifestation rate of Lyme borreliosis in humans How many "tick-bite-victims" will develop clinical disease eventually? Re sults of previous studies from Europe and North America show that about 3 % of people bitten by a vector tick will develop erythema migrans subse quently (16, 22, 24). How many people will be bitten by a tick per year? Most recent reports tell us that very many persons may be bitten by ticks in certain geographical areas of Europe. Figures from southern Germany amount to 16 % (von Sonnenburg, personal communication) and figures from Slovenia up to 40 % (Strle, personal communication). Given an entire population of 8 millions in Austria, the calculated tick attack rate may vary from 1.3 millions to 3.2 millions of inhabitants per year. This would mean, that between about 39000 and 96000 persons would develop Lyme borre liosis in one year, which corresponds to a calculated incidence rate of 0.5 %-1.2 %. However, the incidence rates reported in various studies range from 0.07% to 0.12 % (3,29).
Clinical diagnosis and antibiotic treatment What is also of great interest is the number of patients who were treated with antibiotics because of the clinical diagnosis Lyme borreliosis. This has been evaluated in Austria since 1986. It showed, that each general practitioner treated a constantly increasing number of patients, up to 14 on average in the year 1994. After that year figures dropped down to 12 and 11 and stayed at this level since 1997 (Fig. 2). If we consider these figures as accurate - since they are based on the clinical diagnosis of Lyme borreliosis - and representa-
660
G. Stanek and O. Kahl
16 14 .--
14
12
-
12
11 11
r:- 10
99
.--
.-"
8
7 r-
6
4.9 ...-
4 2.1
2
o
2.9
333.6 ~
r-
r-
n
1986 1988
""
1990 1992 1994
1996
Fig. 2. Average number of cases of Lyme borreliosis treated per year and per practi tioner in Austria. The data were obtained by continuous marketing analysis including 200 physicians and conducted on behalf of Biochemie Austria. (We are indebted to Biochemie Austria that provided these data and permission for publication).
tive we will find that the number of patients and thus the annual incidence of Lyme borreliosis in Austria is 1.5 %. This compares well with a calculated manifestation rate of 3 % after a vector tick-bite and with the estimated num ber of tick-bites (over 16 % of the entire population). These figures show that a considerable number of people were treated for Lyme borreliosis each year, and one may ask the question whether a prophy lactic administration of antibiotics after tick-bite would help to avoid any manifestation.
Studies on the efficacy of prophylactic treatment How many cases of Lyme borreliosis will be avoided by prophylactic treat ment? There are several studies available that sought answers for this questi on (Table 2). Costello et al. (5) treated with phenoxymethyl penicillin versus placebo for 10 days and found one patient in the placebo group. The authors
Lyme borreliosis chemoprophylaxis
661
concluded that their study group was too small to definitely answer the que stion whether prophylactic antibiotics for vector tick-bites is an advisable po licy, and the decision of whether to treat remains a clinical judgement in the individual case. And further" ... the risk of acquiring Lyme disease from a re cognised tick-bite is less than one would expect from the infection rate in ticks and may not be much greater than the risk of an adverse reaction to antibio tic therapy." Shapiro et al. (23) conducted a more extensive double-blind, placebo controlled study. Amoxicillin or placebo was administered for 10 days. Fol low up time was 1 year. Also the ticks submitted were analysed for borreliae by peR and found positive in 15 %. Two of the placebo group developed ery thema migrans. No seroconversion and no late manifestations of Lyme borre liosis were observed. Agre and Schwartz (2) prescribed a three days treatment (penicillin or tetra cyclines) following the tick-bite for patients 3 to 19 years of age. They were fol lowed for stage I and II Lyme borreliosis. The main result was that only one per son out of 90 victims in the placebo group developed clinical Lyme borreliosis. In a Russian study adult ticks of the species I. persulcatus were searched by darkfield microscopy for the presence of spirochaetes (12). Only victims with positive ticks were treated with doxycycline, 200 mg for 3-5 days. Of the treatment group 1.1 % and of the defined control group 12.4 % developed Lyme borreliosis. The authors consider these results as an argument for pro phylactic antibiotic treatment.
Prophylactic treatment in case of a borrelia positive tick The controversial topic of prophylactic antibiotic treatment after tick-bite was affirmatively answered by Maiwald et al. (14). The authors examined ticks that were collected from 730 persons from southwestern Germany. The poly merase chain reaction (peR) was used to detect borrelial DNA in the ticks. 11.3 % of ticks were peR positive. By clinical and serological follow up of the persons bitten the authors found a total transmission rate of 2.6 %. Additionally, they found true sero-conversion in about 27% of persons who were initially sero-negative. This sero-conversion rate is similar to that obtai ned previously (22). The authors draw the conclusion from their results that examination of ticks and antibiotic prophylaxis in case of a positive vector tick appears to be indicated. Sood et al. (24) prospectively studied persons with tick-bites (1. scapularis) and found that those with a duration of attachment;::: 72 h versus < 72 h had a significantly higher incidence of B. burgdorferi infection (20 % to 1.1 %). The authors conclude that identification and measurement of engorgement (determined from the scutal index) rather than peR examination for borrelia infection can be used to identify a small, high-risk subset of persons who may benefit from antibiotic prophylaxis.
662
G. Stanek and O. Kahl
Although the high value of information that comes out of these studies is very much acknowledged we are not in favour of prophylactic antibiotic treat ment after a vector tick-bite. Certainly, people who present with an attached tick will benefit from it because they will get proper tick removal and infor mation about the risks and will be invited to return as soon as any sign or symptom will occur within the next few weeks. But there are several aspects which should be borne in mind. Firstly, most ticks are not removed by physi cians or scientific investigators. Thus the origin of a tick submitted may be doubtful. Secondly, a certain proportion of victims may not recognise the tick bite. Thirdly, the examination of the tick should be done immediately after removal. This raises the question of the availability of suitably trained person al. Besides this, the PCR procedure itself is expensive and is usually not cov ered by insurance. It is uncertain who will pay for this. Although determina tion of the scuta I index is comparatively simple and less expensive it would require special training for practitioners. Fourthly, it is not guaranteed that a positive tick has transmitted the spirochaetes to the individual from whom it was removed. And finally, one must consider the costs of a therapeutic enter prise which is not based on evidence. If we do a calculation based on the figures of the study undertaken in Ger many and select for example, amoxicillin (a 10 days course) as the substance for prophylactic treatment of a person with a positive tick this would cost DM 40.- (EURO 21.-) plus the approximate costs for the tick examination by PCR (about DM 100.-/EURO 51.-); together DM 140.-/EURO 72.-. Given a pop ulation of about 80 million people in Germany the number of vector tick-bites a year may be an estimated number of 12.8 millions. Of these 11.3 % may be considered as having acquired a borrelia infection because of a PCR positive tick. In terms of total costs, including tick examination and prophylactic treat ment, they would amount to DM ca. 1338 million/EURO ca. 690 million. Treatment of resulting erythema migrans would be applied to only ca. 0.4 mil lions and would not need tick examination nor serology and would cost DM ca. 16 millions/EURO ca. 8 millions. Although calculation of costs has an al truistic aspect since there are expenses for one part but income for another part of the population, it is a doubtful approach to encourage people to use anti biotics without a sound basis for it. We therefore support the statement that only a reliable diagnosis is the basis for a rational treatment (27, 32).
Conclusions and recommendations Routine prophylactic treatment with antibiotics after a tick-bite appears not justified because of a manifestation rate of only about 3 %. About 97 % of people bitten by vector ticks would not need this treatment but every attempt to select any special high risk group would be expensive. Then, ticks can trans mit - although very rarely - causative agents of several other bacterial disea ses. Because of this, a broad spectrum antibiotic would be needed and would,
Lyme borreliosis chemoprophylaxis
663
on a larger scale, consequently lead to a selection of antibiotic resistant bac terial strains; this, of course, does not hold for Ixodes-transmitted bacteria. Long-term studies would be needed to assess which substance and which treatment period will be the best. This gives rise to the question whether the complexity of a large scale double blinded study is justified for an infectious disease which can be treated effectively even in its second and third stage and has never caused a fatality. If a tick is found on the skin it should be removed as soon as possible. The tick should not be handled with blunt curved forceps or tweezers. It should be grasped as close to the surface of skin as possible and pulled out gently without jerking or twisting. However, contrary to previous statements neither squeezing nor the application of non-poisonous compounds such as petrole um jelly, salad oil or fingernail polish might increase the risk of infection (11). Thermal methods such as burning the tick with the naked flame of a lighter or a match or another heat producing device must be avoided because serious damage of the skin may result. But even proper and early tick removal gives no absolute guarantee that subsequent disease is prevented. Thus, the conse quence of this is the wait and watch policy (30). In our opinion this concept is both ethically and economically convincing. However, patients, tick-bite victims, and practitioners show an increasing interest in measures to avoid Lyme borreliosis. In Austria, for example, peo ple are familiar with protection against tick-borne encephalitis by vaccination. The same is desired for Lyme borreliosis, since treatment with antibiotics over several weeks burdens the patients. Since two vaccine trials were completed in the USA (25, 26, 28) and one of these Lyme disease vaccines is already in use in North America, there is an expectancy for the introduction of immuno prophylactic measures for Lyme borreliosis also in Eurasia.
References 1. Ackermann, R., J. Kabatzki, H. P. Boisten, A. C. Steere, R. L. Grodzicki, S. Hartung, and U. Runne: Ixodes ricinus spirochete and European erythema chronicum mi grans disease. Yale]. BioI. Med. 57 (1984) 573-580 2. Agre, F. and R. Schwartz: The value of early treatment of deer tick bites for the pre vention of Lyme disease. Am.]' Dis. Child. 147 (1993) 945-947 3. Berglund, j., R. Eitrem, K. Ornstein, A. Lindberg, A. Ringer, H. Elmrud, M. Carls son, A. Runehagen, C. Svanborg, and R. Norrby: An epidemiologic study of Lyme disease in southern Sweden. N. Eng!.]. Med. 333 (1995) 1319-1327 4. Burger, 1., A. Radda, G. Wewalka und G. Stanek: Untersuchungen zur Okologie von Borrelia burgdorferi. Mitt. Osterr. Ges. Tropenmed. Parasito!' 7 (1985) 51-54 5. Costello, C. M., A. C. Steere, R. E. Pinkerton, and H. M. Feder Jr.: A prospective study of tick bites in an endemic area for Lyme disease.]. Infect. Dis. 159 (1989) 136-139 6. Falco, R. c., D. Fish, and J. Piesman: Duration of tick bites in a Lyme disease endemic area. Am.]. Epidemio!. 143 (1996) 187-192
664
G.Stanek and O.Kahl
7. Gray, j. S., o. Kahl, j. N. Robertson, M. Daniel, A. Estrada-Peiia, G. Gettinby, T. G.Jaenson, P.Jensen, F.Jongejan, E. Korenberg, K. Kurtenbach, and P. Zeman: Lyme borreliosis habitat assessment. Zent.bl. Bakteriol. 287 (1998) 211-228 8. Jenek, j. and A. Glazaczow: The evaluation of Borrelia burgdorferi sensu lato spir ochaetes distribution in Ixodes ricinus ticks collected in selected regions of Wiel kopolskai - by polymerase chain reaction (PCR) method. Przeg!. Epidemio!. 50 (1996) 383-386 9. Junttila, J., M. Peitomaa, Soini, M. Marjamaki, and M. K. Viljanen: Prevalence of Borrelia burgdorferi in Ixodes ricinus ticks in urban recreational areas of Helsin ki. J. Clin. Microbiol. 37 (1999) 1361-1365 10. Kahl, 0., K. Schmidt, A. Schonberg, U. Laukamm-Josten, W. Kniille, and U. Bienz Ie: Prevalence of Borrelia burgdorferi in Ixodes ricinus ticks in Berlin (West). Zen tralbl. Bakteriol. Mikrobio!. Hyg. A 270 (1989) 434-440 11. Kahl, 0., c.Janetzki-Mittmann, j. S. Gray, R.Jonas, J. Stein, and R. de Boer: Risk of infection with Borrelia burgdorferi sensu lato for a host in relation to the dura tion of nymphal Ixodes ricinus feeding and the method of tick removal. Zentralbl. Bakteriol. 287 (1998) 41-52 12. Korenberg, E. I., N. N. Vorobyeva, H. G. Moskvitina, and L. Ya Gorban': Preven tion of borreliosis in persons bitten by infected ticks. Infection 24 (1996) 187-189 13. Lebet, N. and L. Gern: Histological examination of Borrelia burgdorferi infections in unfed Ixodes ricinus nymphs. Exp. Appl. Acarol. 18 (1994) 177-183 14. Maiwald, M., R. Oehme, O. March, T. N. Petney, P. Kimmig, K. Naser, H. A. Zap pe, D. Hassler, and M. von Knebel Doeberitz: Transmission risk of Borrelia burg dorferi sensu lato from Ixodes ricinus ticks to humans in southwest Germany. Epi demiol. Infect. 121 (1998) 103-108 15. Nohlmans, M. K., R. de Boer, A. E. van den Bogaard, A. A. Blaauw, and C. P. van Boven: Occurrence of Borrelia burgdorferi in Ixodes ricinus in The Netherlands. Ned. Tijdschr. Geneeskd. 134 (1990) 1300-1303 16. Paul, H., H.J. Gerth, and R. Ackermann: Infectiousness for humans of Ixodes ric inus containing Borrelia burgdorferi. Zbl. Bakt. Hyg. A 263 (1986) 473-476 17. Peavey, C. A. and R. S. Lane: Transmission of Borrelia burgdorferi by Ixodes pa cificus nymphs and reservoir competence of deer mice (Peromyscus maniculatus) infected by tick-bite. J. Paras ito I. 81 (1995) 175-178 18. Peter, 0., A. G. Bretz, and D. Bee: Occurrence of different genospecies of Borrelia burgdorferi sensu lato in ixodid ticks of Valais, Switzerland. Eur. J. Epidemiol. 11 (1995) 463-467 19. Piesman, j., T. N. Mather, R.J. Sinsky, and A. Spielman: Duration of tick attach ment and Borrelia burgdorferi transmission. J. Clin. Microbiol. 25 (1987) 557-558 20. Piesman, j., G. O. Maupin, E. G. Campos, and C. M. Happ: Duration of adult fe male Ixodes dammini attachment and transmission of Borrelia burgdorferi, with de scription of a needle aspiration isolation method. J. Infect. Dis. 163 (1991) 895-897 21. Rijpkema, S., j. Nieuwenhuijs, F. F. Franssen, and F.Jongejan: Infection rates of Borrelia burgdorferi in different instars of Ixodes ricinus ticks from the Dutch North Sea Island of Ameland. Exp. App!. Acarol. 18 (1994) 531-42 22. Schmutzhard, E., G. Stanek, M. Pletschette, A. M. Hirschi, A. Pallua, R. Schmitz berger, and R. Schlogl: Infections after tick-bite: a prospective epidemiological study from Tyrol. Infection 16 (1988) 269-272 23. Shapiro, E. D., M. A. Gerber, N. B. Holabird, A. T. Berg, H. M. Feder Jr., G. L. Bell, P. N. Bell, and D. H. Persing: A controlled trial of antimicrobial prophylaxis for Lyme disease after deer-tick bites. N. Engl.J. Med. 327 (1992) 1769-1773
Lyme borreliosis chemoprophylaxis
665
24. Sood, S. K., M. B. Salzmann, B.].]ohnson, C. M. Happ, K. Feig, L. Carmody, L. G. Rubin, E. Hilton, and]. Pies man: Duration of tick attachment as a predictor of the risk of Lyme disease in an area in which Lyme disease is endemic. J. Infect. Dis. 175 (1997) 996-999 25. Sigal, L. H.,]. M. Zahranik, P. Lavin, S.]. Patella, G. Bryant, R. Haselby, E. Hilton, M. Kunkel, D. Adler-Klein, T. Doherty,]. Evans, P.]. Molloy, A. L. Seidner,]. R. Sa betta, H.]. Simon, M. S. Klempner,]. Mays, D. Marks, and S. E. Malawista: A vac cine consisting of recombinant Borrelia burgdorferi outer-surface protein A to pre vent Lyme disease. Recombinant Outer-Surface Protein A Lyme Disease Vaccine Study Consortium. N. Eng!. J. Med. 23 (1998) 216-222 26. Simon, M. M., Y. Bauer, W. Zhong, H. Hofmann, and R. Wallich: Lyme disease: pathogenesis and vaccine development. Zent.b!. Bakterio!' 289 (1999) 690-695 27. Stanek, G., S. O'Connell, M. Cimmino, E. Aberer, W. Kristoferitsch, M. Gran strom, E. Guy, and]. Gray: European Union concerted action on risk assessment in Lyme borreliosis: clinical case definitions for Lyme borreliosis. Wien. Klin. Wo chenschr. 108 (1996) 741-747 28. Steere, A. c., V. K. Sikand, F. Meurice, D. L. Parenti, E. Fikrig, R. T. Schoen, ]. Nowakowski, C. H. Schmid, S. Laukamp, C. Buscarino, and D. S. Krause: Vacci nation against Lyme disease with recombinant Borrelia burgdorferi outer-surface lipoprotein A with adjuvant. N. Eng!. J. Med. 23 (1998) 209-215 29. Strle, F. and M. Stantic-Pavlinic: Lyme disease in Europe. N. Eng!. J. Med. 334 (1996) 803 30. Strle F.: Recommended procedures for tick bites in a Lyme borreliosis endemic area. Acta dermatovenerologica A.P.A. 5 (1996) 173-177 31. Strle, F., V. Maraspin, S. Furlan-Lotric, and]. Cimperman: Epidemiologic study of a cohort of adult patients with erythema migrans registered in Slovenia in 1993. Eur. J. Epidemio!. 12 (1996) 503-507 32. Strle, F.: Principles of diagnosis and treatment of Lyme borreliosis. Wien. Klin. Wo chenschr. 112 (1999) in press 33. Wegner, Z., M. Racewicz, B. Kubica-Biernat, W. Kruminis-Lozowska, and]. Stanc zak: The prevalence of Ixodes ricinus ticks (Acari, Ixodidae) in the forested areas of Gdansk, Sopot, and Gdynia and their infection rate with Borrelia burgdorferi spirochetes. Przeg!. Epidemio!. 51 (1997) 11-20 34. Wilske, B., R. Steinhuber, H. Bergmeister, V. Fingerle, G. Schierz, V. Preac-Mursic, E. Vanek, and B. Lorbeer: Lyme borreliosis in South Germany. Epidemiologic data on the incidence of cases and on the epidemiology of ticks (Ixodes ricinus) carry ing Borrelia burgdorferi. Dtsch. Med. Wochenschr. 112 (1987) 1730-1736 35. Zeman, P: Borrelia-infection rates in tick and insect vectors accompanying human risk of acquiring Lyme borreliosis in a highly endemic region in Central Europe. Folia Parasito!' (Praha) 45 (1998) 319-325
Corresponding author: Gerold Stanek, MD, Hygiene Institute of the University of Vienna, A-I095 Wien, Kinderspitalgasse 15, Austria, E-mail gerold.stanek@uni vie.ac.at
44
Zent.bl. Bakteriol. 28915-7
Zentralblatt fUr
Forum paper
Chemoprophylaxis for Lyme Borreliosis? Gerold Stanek 1 and Olaf KahIl 1
2
Hygiene Institute, University of Vienna, Vienna, Austria and Institute of Microbiology and Hygiene, Charite, Humboldt-University, Berlin, Ger many
Summary Chemoprophylaxis is a term which describes treatment with an antimicrobial chemo therapeutic before, during or shortly after an actual or suspected exposure to an infec tious agent in order to prevent clinical disease, which may be severe or even fatal. Lyme borreliosis is considered the most frequent ixodid-tick-transmitted human bacterial infection in the northern hemisphere. For several years there has been a debate on the prophylactic application of antimicrobial chemotherapeutics after an attached Ixodes tick was removed. Would this measure prevent a subsequent borrelia infection and would it be practical? People are exposed to tick-bites mostly during lei sure spent in recreational areas which are often tick infested. The proportion of I. ric inus ticks infected with B. burgdorferi s.l. varies from area to area and in a given area also from year to year (infection rate up to a maximum of 55 %). The transmission rate strongly depends on the duration of feeding, but it could be shown that the criti cal time of feeding is much shorter for European I. ricinus than for the North American I. scapularis or I. pacificus ticks. Nevertheless, even the low risk of complications de spite the very good chance of treating erythema migrans successfully seems to justify prophylactic treatment for some investigators whilst others do not see an argument for this. Double blinded studies in the USA showed a relatively low frequency of illness after vector tick-bite and absence of disseminated disease manifestations. The efficacy of prophylactic antibiotic treatment after tick-bites is not established. Suggestions to examine removed ticks for borrelia in order to obtain indication for prophylactic anti biotic treatment will fail in practice because of high costs and uncertainty in verifying the transmission. Do we need blinded studies in central Europe on a representative number of cases, although it is known that Lyme borreliosis can be treated effectively even in its second and third stage and has never caused a fatal outcome? We conclude that only a reliable diagnosis of symptoms is the basis for a rational antibiotic treat ment, and that instead of chemoprophylaxis for Lyme borreliosis after a vector tick bite the wait and watch policy is recommended. Key words: Lyme borreliosis, chemoprophylaxis, tick attachment, incidence
0934-8840/99/289/5-7-655 $12.00/0
656
G. Stanek and
o. Kahl
Introduction Chemoprophylaxis is a term which describes treatment with an antimicrobi al chemotherapeutic before, during or shortly after an actual or suspected ex posure to an infectious agent in order to prevent clinical disease, which may be severe or even fatal. In our geographical area of moderate climate chemo prophylaxis is used selectively or moderately selectively. Chemoprophylaxis of bacterial infection differentiates between prevention of infection (e. g. expo sure to pertussis, scarlet fever, meningococcal meningitis, diphtheria, syphilis, and tuberculosis), prevention of recurrences (e. g. absolute indication to avoid re-infection with A streptococci after rheumatic fever), and precautions against complications (in order to prevent infection after trauma or during surgical interventions e. g. in organs which are naturally colonised with bac teria, such as the intestine or simply tooth extraction etc.). Currently, there is no generally accepted recommendation for anti-bacterial chemoprophylaxis of tick-bites. Lyme borreliosis is caused by spirochaetal bacteria of the genus Borrelia (B. burgdorferi sensu lato) and transmitted by Ixodes ticks (Fam. Ixodidae, hard ticks). Although haemato phagous insects have occasionally been found to be carriers of borreliae their vector status is unproven. Discussion about the chemoprophylaxis of Lyme borreliosis started with the suggestion to take antibiotics immediately after removal of an attached tick to kill any transmitted borreliae as early as possible in order to prevent any further complication. Is it reasonable to administer antibacterial chemo therapeutics after each vector tick-bite in order to prevent the entire clinical spectrum of Lyme borreliosis? Is the infection rate of ticks and the transmis sion rate of borrelia that high and the manifestation rate of Lyme borreliosis that frequent that such a measure can generally be recommended?
Infection rate of ticks The infection rate of unfed, questing vector ticks with B. burgdorferi is a prin cipal factor for the human infection risk and has been studied in different parts of Europe with different methods since the eighties. Results of selected stud ies (emphasis on central Europe) show a broad range of I. ricinus infection rates with B. burgdorferi s.l. with a maximum rate of up to about 55 % (1,4, 7, 8, 9, 10, 15, 17, 18, 21, 33, 34, 35). According to these findings one may deduce that the risk of becoming infected with borreliae by a vector tick-bite ranges from to 55 % depending on the developmental stage of the tick and the area where the tick was encountered (7).
°
Lyme borreliosis chemoprophylaxis
657
Attachment period and transmission rate It seems helpful to remove an attached tick as soon as possible since we know from the results of several studies that transmission of B. burgdorferi s.l. is closely related to the feeding period. Piesman and colleagues (19,20) were the first to study the duration of attachment required for the transmission of bor reliae in North America. Briefly, Piesman et al. found in their animal experi ments that only 7 % of the rodents (white-footed mice) became infected with B. burgdorferi after an exposure of 24 hours, and only 36 % after 48 hours of exposure to feeding I. scapularis nymphs but 93 % after 72 hours (Table 1). Unfortunately, it is unknown whether the results obtained in an animal experiment can be exactly applied to the Borrelia transmission in humans. Surprisingly long attachment periods were found in humans of the north eastern USA, a highly endemic area of Lyme borreliosis. The calculated feed ing time for nymphs was about 35 hours mean, and the attachment time in creased with the age of the victim (6). The duration of nymphal and female I. scapularis tick attachment was determined by the scutal index (body length/scutal width). Whether persons with a prolonged tick attachment con stitute a special high-risk group for infection was studied by Sood et al. (24). The answer was YES; the incidence of B. burgdorferi infection was signifi cantly elevated for attachment periods of;:::: 72 hours with 3 out of 15 subjects becoming seropositive. One should emphasize, however, that altogether only 4 out of 105 subjects subsequently developed erythema migrans in that pros pective study. This means that the overall manifestation risk was only 3.8 %. This figure is almost identical with that found in a study on recruits in Aus tria who were exposed to vector tick-bites during outdoor activities. Only 2 Table 1. Animal experiments undertaken to determine the transmission of borreliae in relation to the duration of tick attachment Author
Experimental animal, tick species
Duration of attachment
Transmission
Piesman et al. 1987 USA Piesman et al. 1991 USA Peavey and Lane 1995 USA
hamster, white-footed mouse Ixodes scapularis rabbit Ixodes scapularis deer-mouse Ixodes pacificus
Kahl et al. 1998 Germany
gerbil Ixodes ricinus
24 h 48 h 72h 48 h >120 h 48 h 72h >96h 17 h 29 h 47 h 65 h
7% 36% 93% >60% 100% 11 % 25% 80% ca 50% ca 50% 100% 100%
658
G. Stanek and
o. Kahl
out of 50 soldiers (4 %) developed an erythema migrans, and no further man ifestation was recorded after antibiotic treatment (22). In contrast to observations in North America, anecdotal reports of patients and physicians in central Europe suggest that a tick-feeding period of only a few hours can be sufficient to become infected with borreliae and to develop disease. In Austria and in Slovenia records of patients with erythema migrans were evaluated independently and ended up with almost identical results (31). It became obvious that most of the patients discovered the attached tick on the day after the putative tick attachment. In the majority of patients the feeding period of ticks until detection and removal was more than 12 hours and in about a third more than 24 hours in the two countries (Fig. 1). How ever, there was a certain· proportion of exposed persons (14-18 %) who removed the tick within 6 hours and subsequently developed erythema mi grans. Whether this is due to a quicker passage of spirochaetes from the mid gut of I. ricinus to its salivary glands or due to spirochaetes which are already in the salivary glands of unfed I. ricinus awaits clarification. The latter has been shown by the results of a study conducted in Switzerland (13); 11 % of unfed nymphal ticks carried borreliae of the species B. afzelii in their salivary glands. Additionally, in animal experiments with I. ricinus nymphs it was demon strated that gerbils develop eventual spirochaetal infection in about 50 % after a feeding period of less than 17 hours (11). These results underline the fact that the attachment period required to transmit borreliae is significantly short er with I. ricinus nymphal ticks than with I. scapularis (Table 2).
40 DADSL
35
~
~
30 ~
25
~c-
20
1ft
15
o
~
i
~,---
,---
10 5 0
I
, to 6
to 12
to 24
> 24
Hours of feeding
Fig. 1. Time period of attachment before a vector tick was removed from the skin. Re ports about the estimated duration of tick attachment when patients presented with erythema migrans. Results from studies in Austria (A, n = 254) and in Slovenia (SL, n = 212) (ref. 31).
Lyme borreliosis chemoprophylaxis
659
Table 2. Results of studies on the efficacy of prophylactic administration of antibacterial chemotherapeutics after vector tick-bite in order to avoid subsequent Lyme borreliosis (LB) Authors
Costello et al. 1989 Shapiro et al. 1992 Agre and Schwartz 1993 Korenberg et al. 1996
Country Tick species USA USA USA Russia
Ixodes scapularis Ixodes scapularis Ixodes scapularis Ixodes persulcatus
Treatment Group
Placebo Group
LB/total
(%)
LBltotal
(%)
0/27
(0.0)
1/29
(3.4)
0/192
(0.0)
2/173
(1.2)
0/89
(0.0)
1/90
(1.1)
3/261
(1.1)
12/97
(12.4)
Manifestation rate of Lyme borreliosis in humans How many "tick-bite-victims" will develop clinical disease eventually? Re sults of previous studies from Europe and North America show that about 3 % of people bitten by a vector tick will develop erythema migrans subse quently (16, 22, 24). How many people will be bitten by a tick per year? Most recent reports tell us that very many persons may be bitten by ticks in certain geographical areas of Europe. Figures from southern Germany amount to 16 % (von Sonnenburg, personal communication) and figures from Slovenia up to 40 % (Strle, personal communication). Given an entire population of 8 millions in Austria, the calculated tick attack rate may vary from 1.3 millions to 3.2 millions of inhabitants per year. This would mean, that between about 39000 and 96000 persons would develop Lyme borre liosis in one year, which corresponds to a calculated incidence rate of 0.5 %-1.2 %. However, the incidence rates reported in various studies range from 0.07% to 0.12 % (3,29).
Clinical diagnosis and antibiotic treatment What is also of great interest is the number of patients who were treated with antibiotics because of the clinical diagnosis Lyme borreliosis. This has been evaluated in Austria since 1986. It showed, that each general practitioner treated a constantly increasing number of patients, up to 14 on average in the year 1994. After that year figures dropped down to 12 and 11 and stayed at this level since 1997 (Fig. 2). If we consider these figures as accurate - since they are based on the clinical diagnosis of Lyme borreliosis - and representa-
660
G. Stanek and O. Kahl
16 14 .--
14
12
-
12
11 11
r:- 10
99
.--
.-"
8
7 r-
6
4.9 ...-
4 2.1
2
o
2.9
333.6 ~
r-
r-
n
1986 1988
""
1990 1992 1994
1996
Fig. 2. Average number of cases of Lyme borreliosis treated per year and per practi tioner in Austria. The data were obtained by continuous marketing analysis including 200 physicians and conducted on behalf of Biochemie Austria. (We are indebted to Biochemie Austria that provided these data and permission for publication).
tive we will find that the number of patients and thus the annual incidence of Lyme borreliosis in Austria is 1.5 %. This compares well with a calculated manifestation rate of 3 % after a vector tick-bite and with the estimated num ber of tick-bites (over 16 % of the entire population). These figures show that a considerable number of people were treated for Lyme borreliosis each year, and one may ask the question whether a prophy lactic administration of antibiotics after tick-bite would help to avoid any manifestation.
Studies on the efficacy of prophylactic treatment How many cases of Lyme borreliosis will be avoided by prophylactic treat ment? There are several studies available that sought answers for this questi on (Table 2). Costello et al. (5) treated with phenoxymethyl penicillin versus placebo for 10 days and found one patient in the placebo group. The authors
Lyme borreliosis chemoprophylaxis
661
concluded that their study group was too small to definitely answer the que stion whether prophylactic antibiotics for vector tick-bites is an advisable po licy, and the decision of whether to treat remains a clinical judgement in the individual case. And further" ... the risk of acquiring Lyme disease from a re cognised tick-bite is less than one would expect from the infection rate in ticks and may not be much greater than the risk of an adverse reaction to antibio tic therapy." Shapiro et al. (23) conducted a more extensive double-blind, placebo controlled study. Amoxicillin or placebo was administered for 10 days. Fol low up time was 1 year. Also the ticks submitted were analysed for borreliae by peR and found positive in 15 %. Two of the placebo group developed ery thema migrans. No seroconversion and no late manifestations of Lyme borre liosis were observed. Agre and Schwartz (2) prescribed a three days treatment (penicillin or tetra cyclines) following the tick-bite for patients 3 to 19 years of age. They were fol lowed for stage I and II Lyme borreliosis. The main result was that only one per son out of 90 victims in the placebo group developed clinical Lyme borreliosis. In a Russian study adult ticks of the species I. persulcatus were searched by darkfield microscopy for the presence of spirochaetes (12). Only victims with positive ticks were treated with doxycycline, 200 mg for 3-5 days. Of the treatment group 1.1 % and of the defined control group 12.4 % developed Lyme borreliosis. The authors consider these results as an argument for pro phylactic antibiotic treatment.
Prophylactic treatment in case of a borrelia positive tick The controversial topic of prophylactic antibiotic treatment after tick-bite was affirmatively answered by Maiwald et al. (14). The authors examined ticks that were collected from 730 persons from southwestern Germany. The poly merase chain reaction (peR) was used to detect borrelial DNA in the ticks. 11.3 % of ticks were peR positive. By clinical and serological follow up of the persons bitten the authors found a total transmission rate of 2.6 %. Additionally, they found true sero-conversion in about 27% of persons who were initially sero-negative. This sero-conversion rate is similar to that obtai ned previously (22). The authors draw the conclusion from their results that examination of ticks and antibiotic prophylaxis in case of a positive vector tick appears to be indicated. Sood et al. (24) prospectively studied persons with tick-bites (1. scapularis) and found that those with a duration of attachment;::: 72 h versus < 72 h had a significantly higher incidence of B. burgdorferi infection (20 % to 1.1 %). The authors conclude that identification and measurement of engorgement (determined from the scutal index) rather than peR examination for borrelia infection can be used to identify a small, high-risk subset of persons who may benefit from antibiotic prophylaxis.
662
G. Stanek and O. Kahl
Although the high value of information that comes out of these studies is very much acknowledged we are not in favour of prophylactic antibiotic treat ment after a vector tick-bite. Certainly, people who present with an attached tick will benefit from it because they will get proper tick removal and infor mation about the risks and will be invited to return as soon as any sign or symptom will occur within the next few weeks. But there are several aspects which should be borne in mind. Firstly, most ticks are not removed by physi cians or scientific investigators. Thus the origin of a tick submitted may be doubtful. Secondly, a certain proportion of victims may not recognise the tick bite. Thirdly, the examination of the tick should be done immediately after removal. This raises the question of the availability of suitably trained person al. Besides this, the PCR procedure itself is expensive and is usually not cov ered by insurance. It is uncertain who will pay for this. Although determina tion of the scuta I index is comparatively simple and less expensive it would require special training for practitioners. Fourthly, it is not guaranteed that a positive tick has transmitted the spirochaetes to the individual from whom it was removed. And finally, one must consider the costs of a therapeutic enter prise which is not based on evidence. If we do a calculation based on the figures of the study undertaken in Ger many and select for example, amoxicillin (a 10 days course) as the substance for prophylactic treatment of a person with a positive tick this would cost DM 40.- (EURO 21.-) plus the approximate costs for the tick examination by PCR (about DM 100.-/EURO 51.-); together DM 140.-/EURO 72.-. Given a pop ulation of about 80 million people in Germany the number of vector tick-bites a year may be an estimated number of 12.8 millions. Of these 11.3 % may be considered as having acquired a borrelia infection because of a PCR positive tick. In terms of total costs, including tick examination and prophylactic treat ment, they would amount to DM ca. 1338 million/EURO ca. 690 million. Treatment of resulting erythema migrans would be applied to only ca. 0.4 mil lions and would not need tick examination nor serology and would cost DM ca. 16 millions/EURO ca. 8 millions. Although calculation of costs has an al truistic aspect since there are expenses for one part but income for another part of the population, it is a doubtful approach to encourage people to use anti biotics without a sound basis for it. We therefore support the statement that only a reliable diagnosis is the basis for a rational treatment (27, 32).
Conclusions and recommendations Routine prophylactic treatment with antibiotics after a tick-bite appears not justified because of a manifestation rate of only about 3 %. About 97 % of people bitten by vector ticks would not need this treatment but every attempt to select any special high risk group would be expensive. Then, ticks can trans mit - although very rarely - causative agents of several other bacterial disea ses. Because of this, a broad spectrum antibiotic would be needed and would,
Lyme borreliosis chemoprophylaxis
663
on a larger scale, consequently lead to a selection of antibiotic resistant bac terial strains; this, of course, does not hold for Ixodes-transmitted bacteria. Long-term studies would be needed to assess which substance and which treatment period will be the best. This gives rise to the question whether the complexity of a large scale double blinded study is justified for an infectious disease which can be treated effectively even in its second and third stage and has never caused a fatality. If a tick is found on the skin it should be removed as soon as possible. The tick should not be handled with blunt curved forceps or tweezers. It should be grasped as close to the surface of skin as possible and pulled out gently without jerking or twisting. However, contrary to previous statements neither squeezing nor the application of non-poisonous compounds such as petrole um jelly, salad oil or fingernail polish might increase the risk of infection (11). Thermal methods such as burning the tick with the naked flame of a lighter or a match or another heat producing device must be avoided because serious damage of the skin may result. But even proper and early tick removal gives no absolute guarantee that subsequent disease is prevented. Thus, the conse quence of this is the wait and watch policy (30). In our opinion this concept is both ethically and economically convincing. However, patients, tick-bite victims, and practitioners show an increasing interest in measures to avoid Lyme borreliosis. In Austria, for example, peo ple are familiar with protection against tick-borne encephalitis by vaccination. The same is desired for Lyme borreliosis, since treatment with antibiotics over several weeks burdens the patients. Since two vaccine trials were completed in the USA (25, 26, 28) and one of these Lyme disease vaccines is already in use in North America, there is an expectancy for the introduction of immuno prophylactic measures for Lyme borreliosis also in Eurasia.
References 1. Ackermann, R., J. Kabatzki, H. P. Boisten, A. C. Steere, R. L. Grodzicki, S. Hartung, and U. Runne: Ixodes ricinus spirochete and European erythema chronicum mi grans disease. Yale]. BioI. Med. 57 (1984) 573-580 2. Agre, F. and R. Schwartz: The value of early treatment of deer tick bites for the pre vention of Lyme disease. Am.]' Dis. Child. 147 (1993) 945-947 3. Berglund, j., R. Eitrem, K. Ornstein, A. Lindberg, A. Ringer, H. Elmrud, M. Carls son, A. Runehagen, C. Svanborg, and R. Norrby: An epidemiologic study of Lyme disease in southern Sweden. N. Eng!.]. Med. 333 (1995) 1319-1327 4. Burger, 1., A. Radda, G. Wewalka und G. Stanek: Untersuchungen zur Okologie von Borrelia burgdorferi. Mitt. Osterr. Ges. Tropenmed. Parasito!' 7 (1985) 51-54 5. Costello, C. M., A. C. Steere, R. E. Pinkerton, and H. M. Feder Jr.: A prospective study of tick bites in an endemic area for Lyme disease.]. Infect. Dis. 159 (1989) 136-139 6. Falco, R. c., D. Fish, and J. Piesman: Duration of tick bites in a Lyme disease endemic area. Am.]. Epidemio!. 143 (1996) 187-192
664
G.Stanek and O.Kahl
7. Gray, j. S., o. Kahl, j. N. Robertson, M. Daniel, A. Estrada-Peiia, G. Gettinby, T. G.Jaenson, P.Jensen, F.Jongejan, E. Korenberg, K. Kurtenbach, and P. Zeman: Lyme borreliosis habitat assessment. Zent.bl. Bakteriol. 287 (1998) 211-228 8. Jenek, j. and A. Glazaczow: The evaluation of Borrelia burgdorferi sensu lato spir ochaetes distribution in Ixodes ricinus ticks collected in selected regions of Wiel kopolskai - by polymerase chain reaction (PCR) method. Przeg!. Epidemio!. 50 (1996) 383-386 9. Junttila, J., M. Peitomaa, Soini, M. Marjamaki, and M. K. Viljanen: Prevalence of Borrelia burgdorferi in Ixodes ricinus ticks in urban recreational areas of Helsin ki. J. Clin. Microbiol. 37 (1999) 1361-1365 10. Kahl, 0., K. Schmidt, A. Schonberg, U. Laukamm-Josten, W. Kniille, and U. Bienz Ie: Prevalence of Borrelia burgdorferi in Ixodes ricinus ticks in Berlin (West). Zen tralbl. Bakteriol. Mikrobio!. Hyg. A 270 (1989) 434-440 11. Kahl, 0., c.Janetzki-Mittmann, j. S. Gray, R.Jonas, J. Stein, and R. de Boer: Risk of infection with Borrelia burgdorferi sensu lato for a host in relation to the dura tion of nymphal Ixodes ricinus feeding and the method of tick removal. Zentralbl. Bakteriol. 287 (1998) 41-52 12. Korenberg, E. I., N. N. Vorobyeva, H. G. Moskvitina, and L. Ya Gorban': Preven tion of borreliosis in persons bitten by infected ticks. Infection 24 (1996) 187-189 13. Lebet, N. and L. Gern: Histological examination of Borrelia burgdorferi infections in unfed Ixodes ricinus nymphs. Exp. Appl. Acarol. 18 (1994) 177-183 14. Maiwald, M., R. Oehme, O. March, T. N. Petney, P. Kimmig, K. Naser, H. A. Zap pe, D. Hassler, and M. von Knebel Doeberitz: Transmission risk of Borrelia burg dorferi sensu lato from Ixodes ricinus ticks to humans in southwest Germany. Epi demiol. Infect. 121 (1998) 103-108 15. Nohlmans, M. K., R. de Boer, A. E. van den Bogaard, A. A. Blaauw, and C. P. van Boven: Occurrence of Borrelia burgdorferi in Ixodes ricinus in The Netherlands. Ned. Tijdschr. Geneeskd. 134 (1990) 1300-1303 16. Paul, H., H.J. Gerth, and R. Ackermann: Infectiousness for humans of Ixodes ric inus containing Borrelia burgdorferi. Zbl. Bakt. Hyg. A 263 (1986) 473-476 17. Peavey, C. A. and R. S. Lane: Transmission of Borrelia burgdorferi by Ixodes pa cificus nymphs and reservoir competence of deer mice (Peromyscus maniculatus) infected by tick-bite. J. Paras ito I. 81 (1995) 175-178 18. Peter, 0., A. G. Bretz, and D. Bee: Occurrence of different genospecies of Borrelia burgdorferi sensu lato in ixodid ticks of Valais, Switzerland. Eur. J. Epidemiol. 11 (1995) 463-467 19. Piesman, j., T. N. Mather, R.J. Sinsky, and A. Spielman: Duration of tick attach ment and Borrelia burgdorferi transmission. J. Clin. Microbiol. 25 (1987) 557-558 20. Piesman, j., G. O. Maupin, E. G. Campos, and C. M. Happ: Duration of adult fe male Ixodes dammini attachment and transmission of Borrelia burgdorferi, with de scription of a needle aspiration isolation method. J. Infect. Dis. 163 (1991) 895-897 21. Rijpkema, S., j. Nieuwenhuijs, F. F. Franssen, and F.Jongejan: Infection rates of Borrelia burgdorferi in different instars of Ixodes ricinus ticks from the Dutch North Sea Island of Ameland. Exp. App!. Acarol. 18 (1994) 531-42 22. Schmutzhard, E., G. Stanek, M. Pletschette, A. M. Hirschi, A. Pallua, R. Schmitz berger, and R. Schlogl: Infections after tick-bite: a prospective epidemiological study from Tyrol. Infection 16 (1988) 269-272 23. Shapiro, E. D., M. A. Gerber, N. B. Holabird, A. T. Berg, H. M. Feder Jr., G. L. Bell, P. N. Bell, and D. H. Persing: A controlled trial of antimicrobial prophylaxis for Lyme disease after deer-tick bites. N. Engl.J. Med. 327 (1992) 1769-1773
Lyme borreliosis chemoprophylaxis
665
24. Sood, S. K., M. B. Salzmann, B.].]ohnson, C. M. Happ, K. Feig, L. Carmody, L. G. Rubin, E. Hilton, and]. Pies man: Duration of tick attachment as a predictor of the risk of Lyme disease in an area in which Lyme disease is endemic. J. Infect. Dis. 175 (1997) 996-999 25. Sigal, L. H.,]. M. Zahranik, P. Lavin, S.]. Patella, G. Bryant, R. Haselby, E. Hilton, M. Kunkel, D. Adler-Klein, T. Doherty,]. Evans, P.]. Molloy, A. L. Seidner,]. R. Sa betta, H.]. Simon, M. S. Klempner,]. Mays, D. Marks, and S. E. Malawista: A vac cine consisting of recombinant Borrelia burgdorferi outer-surface protein A to pre vent Lyme disease. Recombinant Outer-Surface Protein A Lyme Disease Vaccine Study Consortium. N. Eng!. J. Med. 23 (1998) 216-222 26. Simon, M. M., Y. Bauer, W. Zhong, H. Hofmann, and R. Wallich: Lyme disease: pathogenesis and vaccine development. Zent.b!. Bakterio!' 289 (1999) 690-695 27. Stanek, G., S. O'Connell, M. Cimmino, E. Aberer, W. Kristoferitsch, M. Gran strom, E. Guy, and]. Gray: European Union concerted action on risk assessment in Lyme borreliosis: clinical case definitions for Lyme borreliosis. Wien. Klin. Wo chenschr. 108 (1996) 741-747 28. Steere, A. c., V. K. Sikand, F. Meurice, D. L. Parenti, E. Fikrig, R. T. Schoen, ]. Nowakowski, C. H. Schmid, S. Laukamp, C. Buscarino, and D. S. Krause: Vacci nation against Lyme disease with recombinant Borrelia burgdorferi outer-surface lipoprotein A with adjuvant. N. Eng!. J. Med. 23 (1998) 209-215 29. Strle, F. and M. Stantic-Pavlinic: Lyme disease in Europe. N. Eng!. J. Med. 334 (1996) 803 30. Strle F.: Recommended procedures for tick bites in a Lyme borreliosis endemic area. Acta dermatovenerologica A.P.A. 5 (1996) 173-177 31. Strle, F., V. Maraspin, S. Furlan-Lotric, and]. Cimperman: Epidemiologic study of a cohort of adult patients with erythema migrans registered in Slovenia in 1993. Eur. J. Epidemio!. 12 (1996) 503-507 32. Strle, F.: Principles of diagnosis and treatment of Lyme borreliosis. Wien. Klin. Wo chenschr. 112 (1999) in press 33. Wegner, Z., M. Racewicz, B. Kubica-Biernat, W. Kruminis-Lozowska, and]. Stanc zak: The prevalence of Ixodes ricinus ticks (Acari, Ixodidae) in the forested areas of Gdansk, Sopot, and Gdynia and their infection rate with Borrelia burgdorferi spirochetes. Przeg!. Epidemio!. 51 (1997) 11-20 34. Wilske, B., R. Steinhuber, H. Bergmeister, V. Fingerle, G. Schierz, V. Preac-Mursic, E. Vanek, and B. Lorbeer: Lyme borreliosis in South Germany. Epidemiologic data on the incidence of cases and on the epidemiology of ticks (Ixodes ricinus) carry ing Borrelia burgdorferi. Dtsch. Med. Wochenschr. 112 (1987) 1730-1736 35. Zeman, P: Borrelia-infection rates in tick and insect vectors accompanying human risk of acquiring Lyme borreliosis in a highly endemic region in Central Europe. Folia Parasito!' (Praha) 45 (1998) 319-325
Corresponding author: Gerold Stanek, MD, Hygiene Institute of the University of Vienna, A-I095 Wien, Kinderspitalgasse 15, Austria, E-mail gerold.stanek@uni vie.ac.at
44
Zent.bl. Bakteriol. 28915-7