Sydney University Department of Medical Entomology Westmead Hospital
Lyme Disease
Introduction History - Australia
Ecology Clinical Investigations
Clinical Features Serological Investigations
Diagnosis - Culture/PCR Vector and Reservoir Host Investigations
Diagnosis - Serology Conclusions - Australia
Treatment Summary
Prevention Further Reading/Links

bottom of page



Lyme disease (LD) is a tick-borne zoonosis caused by the spirochaete bacterium, Borrelia burgdorferi. Since the disease was first recognised in 1975 it has become the most frequently reported human tick-borne infection worldwide. It has been reported from every continent (except Antarctica) although doubt remains as to whether it occurs in the southern hemisphere in general, and in Australia in particular.


LD is transmitted to humans by ticks. Larval and nymphal stages feed on infected reservoir hosts, acquire the organism and then, after moulting to the next life stage (nymphs and adults respectively), pass on the infection to humans and other animals. In the northern hemisphere, small placental mammals are reservoir hosts. The only species of ticks shown to be competent vectors of B.burgdorferi to humans belong to the Ixodes persulcatus complex, including I. scapularis and I. pacificus in the United States, I. ricinus in western Europe, and I. persulcatus in eastern Europe and Asia. No species of this complex exist in Australia.

Clinical Features

The epithet "The Great Imitator" is now used for LD as it was once for another major spirochaetal disease, syphilis. Like syphilis, LD causes a wide range of (mostly) non-specific symptoms and signs; there are 3 clinical stages:

Stage I:

Manifestations include fever, fatigue, headaches, myalgia, arthralgia (but not arthritis) and lymphadenopathy, usually within 2-3 weeks of infection. A characteristic skin lesion, erythema migrans (EM), appears 3-30 days after the bite of an infected tick, usually at the site of inoculation. The initial lesion is a red maculo-papular lesion greater than 5cm in diameter, rarely painful and expands and may reach more than 50cm in diameter, with central clearing and a well defined, circinate border ("bulls-eye"). Multiple lesions at different stages of evolution may be present. Allergic reactions, associated with tick bites, may be confused with early EM, but occur within a few hours of the bite and resolve within a few days. In North America, EM occurs in 60-80% of serologically confirmed cases of LD.

Stage II:

Symptoms are non-specific and occur weeks or months after the tick bite. A causative association with LD is often uncertain. The lesions resemble those of secondary syphilis: carditis, chronic meningitis, mononeuritis (eg Bell's palsy) and conjunctivitis. Arthralgia and myalgia are often prominent.

Stage III:

Symptoms occur months or years after exposure. The most typical feature, North American, is an erosive arthritis of large joints, particularly the knees. A chronic skin manifestation, acrodermatitis chronica atrophicans (ACA) occurs, mainly in European patients.

Recently, four genospecies of B.burgdorferi have been described; B. burgdorferi sensu stricto, B. garinii, B. afzelii and B. japonica. They are associated with different patterns of disease, any of which can mimic other diseases. With the possible exceptions of typical EM and ACA associated with appropriate exposure, clinical features alone are insufficient for diagnosis of LD and laboratory tests are required for confirmation.



Isolation of the causative organism from a punch biopsy taken at the edge of the EM lesion is successful in up to 80% of cases but it may be up to 8 weeks before spirochaetes can be detected. Polymerase chain reaction (PCR) has the advantage of greater sensitivity and speed; a result is available within 24 hours. Isolation and/or PCR should be attempted upon presentation of a patient with an EM lesion.


Laboratory diagnosis of late LD (Stage II & III) is less reliable and depends on serological tests including the indirect fluorescence antibody test (IFAT), enzyme linked immunosorbent assay (ELISA) and Western immunoblot (WB). False positive results occur due to cross reactions with other bacteria, especially other spirochaetes, viruses and in unrelated syndromes such as autoimmune diseases. IFAT and ELISA are used as screening tests but there has been little standardisation of methods.

Any borderline or positive results should be confirmed by WB which can detect protein bands specific for B. burgdorferi. However, nonspecific bands occur, particularly with the highly cross reactive 41kDa (flagellin) protein. At the Sixth International Congress on Lyme Borreliosis a standardised interpretation of WB results was accepted: an IgG immunoblot is considered positive if 5 of the following 10 bands are present: 18, 21, 28, 30, 39, 41, 45, 58, 66 and 93kDa. Patients with late stage LD will show 10 or more bands on a WB.


The antibiotic therapy of early LD generally results in complete recovery. A 2 week course of oral doxycycline or amoxycillin for Stage I and a third generation cephalosporin for Stage II are the most commonly used regimens. Treatment of late stage LD is less successful and a chronic or relapsing course is common. A third generation cephalosporin for 3 weeks is recommended.


The prevention of LD is mainly through avoidance of tick infested areas and of tick bites by the use of repellents (particularly those containing DEET), wearing of light coloured clothing so that ticks are more easily seen and prompt removal of attached ticks. Transmission of spirochaetes generally does not occur until after 24 hours attachment of the tick. Antibiotic prophylaxis is not recommended. Trials of LD vaccines are in progress with varying degrees of success.



History - Australia

The first Australian cases of a syndrome consistent with Lyme disease were reported from the Hunter Valley region of New South Wales in 1982. Serology was initially negative on one of the 6 patients, but later reported as positive in low titre. Cases of EM with febrile illness were reported in 1986 from the south and central coasts of New South Wales. All had negative serology. In Queensland, from 1986 to 1989, the State Health Laboratories tested 1,247 patients for B.burgdorferi antibody using an IFAT and reported 186 (15%) positive (titre 64) titres. In none of these cases was confirmatory serology (WB) undertaken.

In 1988 at Westmead Hospital, a multidisciplinary investigation of putative LD in coastal New South Wales began, encompassing clinical, serological, vector and reservoir host studies.

Clinical investigations - Australia

Over the past 6 years, due principally to local publicity, there has been an increase in serological testing for LD. This is often initiated by patients, who believe that LD may be an explanation for an undiagnosed health problem. Thus, most patients seen by infectious diseases specialists are self selected and referred for assessment on the basis of tick exposure and reported positive screening serology.

Patients frequently have long-standing symptoms for which no other diagnosis has been established including myalgia, arthralgia without objective evidence of joint disease, neurological symptoms such as frequent headaches, inability to concentrate and impairment of memory, and syndromes resembling chronic fatigue syndrome. The late LD dermatological manifestation, ACA, has not been reported in Australia.

A few cases of EM have been reported from South-Eastern Australia. However, diagnosis can be confounded by a spectacular erythematous hypersensitivity reaction to the bite of I. holocyclus, the most common tick biting humans in New South Wales. Of eight skin biopsies submitted to Westmead Hospital for spirochaete isolation, one, from a patient returning from a LD endemic area in Europe, was culture positive for B.burgdorferi. There has been no isolation from local patients. 

Serological Investigations - Australia

No significant difference was found in seroprevalence rates for B.burgdorferi infection in humans between high (rural residents) and low (urban residents) tick exposure groups, using an IgG ELISA. The overall seropositive rate was 2.2% (9/400). The seroprevalence in New South Wales is comparable with that in non-LD areas, where 1-3% of human sera are seropositive due to cross reacting antibodies and contrasts with reports from known endemic areas, outside Australia, where rural populations have considerably higher seropositive rates. A serosurvey of dogs in New South Wales showed a similar result with 2.5% (6/239) seropositive and another from Brisbane also showed no evidence of B.burgdorferi infection. These suggest that southeastern Australia is a non-endemic area.

From 1988 to 1994 at Westmead Hospital, 78 (1.8%) of 4,372 from local patients with suspected LD were positive for IgG by ELISA and IFAT. All 78 were tested by WB, using North American and European strains of Borrelia; 46 sera showed one or more bands. None, including those with putative late stage disease, showed more than 4 specific bands and thus were all negative by international criteria. Twenty-four patients with various bacterial, viral or autoimmune syndromes unrelated to LD were tested in parallel and 11/24 showed one or 2 indicative bands. Thus a high degree of cross reactivity was demonstrated with non-LD patients.

Recently, there have been reports from eastern Australia of LD-like illness associated with WB serology yielding bands at 31kDa (OspA) and the highly-cross reactive 41kDa band. None of these results conforms with internationally accepted criteria for a positive WB. Concomitant with this are results of WB analysis of sera from patients with syndromes unrelated to LD, >30% of which reacted with a 41kDa band and >10% with the OspA band.

The sensitivity of serological testing for LD sometimes depends on the strain of Borrelia used and could confound interpretation of results in Australia, where no local spirochaete has been isolated for use as a reference antigen.

Vector and Reservoir Host Investigations - Australia

To detect a possible causative agent, ticks were collected from areas associated with putative infections and examined for spirochaetes by dark field microscopy, culture of gut contents, and direct testing of ticks with PCR for the Borrelia-specific flagellin gene.

In total, over 12,000 ticks were tested including >1,000 by PCR. Spirochaete-like objects (SLOs), were observed in 92 cultures from bloodfed ticks but were not typical of Borrelia spp. They were found only found in cultures with bacterial contaminants, presumably from the bloodmeal. Electron micrographs were similar to those of SLOs recovered from contaminated cultures from ticks in Missouri, USA and were composed of aggregations of bacterial flagella, thought to originate from the contaminants. Molecular characterisation indicated that the SLOs shared some antigens with B.burgdorferi, but were not genetically related. Similar objects found in cultures from dissected bloodfed ticks taken from animals on the mid-north coast of NSW were purported to be related to B.burgdorferi and the probable cause of LD in Australia.

A small number (17) of native vertebrate animals were sampled by ear punch biopsy for culture and PCR investigation but there was no evidence of borreliae.

It is possible that the PCR primers used were unable to identify Australian spirochaetes. However, the tick gut contents were also negative by culturing and dark field microscopy.

Conclusions - Australia

There are some major differences between Australia and the endemic areas of the northern hemisphere with respect to the natural history of LD:

No ticks of the I. persulcatus complex, the principal vectors to humans in the northern hemisphere, occur in Australia. In eastern Australia, the logical candidate vector would be I. holocyclus which has a wide host range and is the most common tick biting humans. It was unable to transmit a North American strain of B.burgdorferi but an association with a so far undiscovered Australian spirochaete can not be excluded.

None of the mammal species identified as reservoir hosts in the northern hemisphere are present in Australia. There are reports of spirochaetes in Australian native animals, and a local mammal could be a reservoir host for an indigenous spirochaete that occasionally infects humans through a tick vector and produces a clinical syndrome similar to LD; however, no spirochaete was detected in the 12,000 ticks or animals processed.


The diagnosis of LD outside known endemic areas cannot be based solely on serological tests especially if they fail to conform with internationally accepted criteria, because of the high incidence of false positive results.

A clinical diagnosis in a non-endemic disease area (especially of Stage II or III disease), is difficult to support without isolation of the causative agent from the patient, from other patients with similar illness or from a known vector in the region.

The existence of LD in Australia will remain controversial until an organism is isolated from a local patient and fully characterised, or until a tick-borne organism can be shown to be responsible for the human infection. If it exists it shares few of the epidemiological or clinical characteristics of US or European patterns of LD.

Further Reading

Baldock, F.C., Yamane, I., and Gardner, I. (1993). Pilot survey for Lyme disease antibodies in Brisbane dogs. Australian Veterinary Journal, 70:356-7.

Barbour, A.G. and Fish, D. (1993). The biological and social phenomenon of Lyme disease. Science, 260: 1610-1616.

Dickeson, D. and Gilbert, G.L. (1994). Lyme Disease in Australia? Western immunoblots not the final answer. Annual Scientific Meeting Australian Society of Microbiology, A125.

DOGGETT, S.L., RUSSELL, R.C., Munro, R., Dickeson, D., Ellis, J. Avery, D., Hunt, C.L., Simmonds, J. and Trivett, N. (1994). Lyme disease - the search for the causative agent in southeastern Australia. Arbovirus Research in Australia. Arbovirus Research in Australia, 6: 313-315.

Hudson, B.J., Barry, R.D., Shafren, D.R., Wills, M.C., Caves, S.F. and Lennox, V.A. (1994). Does Lyme borreliosis exist in Australia? Journal of Spirochaetal and Tick-Borne Disease; 1: 46-51.

Piesman, J. and Stone B.F. (1991). Vector competence of the Australian paralysis tick, Ixodes holocyclus, for the Lyme disease spirochaete Borrelia burgdorferi. International Journal Parasitology, 21: 109-11.

RUSSELL, R.C., DOGGETT, S.L., Munro, R., Ellis, J., Avery, D., Hunt, C., and Dickeson, D. (1994). Lyme disease: A search for a causative agent in ticks in southeastern Australia. Epidemiology and Infection, 112: 375-384.

RUSSELL, R.C. (1995). ?Lyme disease in Australia - still to be proven! Emerging Infectious Diseases, 1: 29-31.

Wills, M.C. and Barry, R.D. (1991). Detecting the cause of Lyme disease in Australia [letter]. Medical Journal of Australia, 155: 275.

Stephen L. Doggett, Richard C. Russell, Richard Lawrence and David Dickeson

Originally appeared in: Inoculum, 4: 1-4. Modified and updated; November, 1997.

Related Links

<> (Information from America for ticks, Lyme disease, Ehrlichiosis, Human Babesiosis).

<> (Lyme disease from Europe, tick biology, control and images).

<> (Lyme disease network from America).

top of page