Epidemiological and Clinical Features of Rickettsial Infection Among Australian Individuals from 2015 to 2018

Abstract

Objective:

In this study, the clinical, laboratory and epidemiological features of rickettsial infection in Australian patients were examined.

Methods:

Patients with Rickettsia spp. seropositivity were selected from the database of the Australian Rickettsial Reference Laboratory. Questionnaires were sent to each referring physician with a request for information regarding the patient’s illness.

Results:

A retrospective analysis of 178 patients revealed the following common features in Australian rickettsial infection: fever, headache, myalgia, rash and a history of recent exposure to the Australian bush. The history included tick bites, eschar presence, respiratory abnormalities and acute fatigue, accompanied by abnormal haematology and liver function tests.

Conclusions:

A differential diagnosis including rickettsial infection should be confirmed (or refuted) by rickettsial serology.

Main article text

INTRODUCTION

Rickettsial infections occur worldwide, including in Australia, but are often unrecognised, because their clinical features are unfortunately similar to those of many other bacterial or viral acute infections [1]. However, recognition and treatment with doxycycline usually result in rapid resolution of rickettsial infection.

The true incidence of rickettsial infections in Australia remains unclear, because these diseases do not require notification in most states and territories of Australia. Only Western Australia and Tasmania require notification of rickettsial infections. Ad hoc reports from other region of Australia are available, but reliable incidence data are lacking. Most urban-based physicians in Australia might be unaware of rickettsial infections.

If a patient has a severe infection, bacterial pathogens other than rickettsiae are often considered (e.g., staphylococcal, streptococcal or meningococcal). If a patient has a mild infection, viral pathogens are usually considered (e.g., enterovirus, cytomegalovirus, Epstein-Barr virus or Ross River virus). In severe cases in which bacterial infections are considered, beta-lactam antibiotics are likely to be used initially (to which rickettsiae are not susceptible). In mild and suspected viral cases, antibacterial agents are ideally not used. However, rickettsial infections respond much more rapidly when patients are given doxycycline rather than no antibiotics.

The main rickettsial infections in temperate Australia are the spotted fever group (SFG) rickettsiae, including Queensland tick typhus (Rickettsia australis) and Flinders Island spotted fever (R. honei), both of which are tick transmitted, and cat flea typhus (R. felis) and the typhus group (TG) rickettsia (R. typhi) causing murine typhus, both of which are flea transmitted. In tropical Australia, a mite-transmitted rickettsial infection, scrub typhus due to Orientia tsutsugamushi infection, also occurs but was not included in this study, because no cases were detected.

In this retrospective study, seropositive patients (n=178) diagnosed in the Australian Rickettsial Reference Laboratory (ARRL) [2] between 2015 and 2018 were included. This study was performed to alert Australian physicians to the presence of rickettsial infections in the country, with the hope that describing the features of these illnesses might help physicians make correct diagnoses and provide appropriate treatment (doxycycline).

MATERIALS AND METHODS

Study design

This study was approved by the Barwon Health Human Research Ethics Committee (19/186).

A dataset of all patients with seropositivity to Rickettsia spp. during 2015–2018 from samples submitted to the ARRL in Victoria, Australia was collated. The referring physician for each patient was contacted to complete a questionnaire on their patient’s illness, according to their clinical notes. Contacting the patient was not required, and the information was coded to preserve patient anonymity. The overall diagnostic procedures are summarised in Fig 1.

FIGURE 1 |

Outline of Australian rickettsial disease analysis.

Serological assay for rickettsial infection

The serology assay used in this study was a microimmunofluoresence antibody assay performed by the ARRL as a routine, NATA-accredited, diagnostic assay [2]. Briefly, patient sera were screened at a dilution of 1:128. If positive, a titration was set up with four doubling dilutions of the positive serum at 1:128, 1:256, 1:512 and 1:1024 dilutions. The highest positive serum dilution (showing fluorescence similar to that of the positive control) was recorded as the antibody titre of that serum. Higher antibody titres indicated greater confidence that the patient had recently been exposed to Rickettsiae. In the routine IF assay, six SFG rickettsial species were used as antigens (to cover infections in both Australian patients and returned travellers): R. australis (Queensland tick typhus), R. honei (Flinders Island spotted fever), R. felis (cat flea typhus), R. africae (African tick bite fever), R. conorii (Mediterranean spotted fever) and R. rickettsii (Rocky Mountain spotted fever). Only the first three species are endemic to Australia. In addition, two TG rickettsial species, R. typhi (murine typhus) and R. prowazekii (epidemic typhus), were routinely used as antigens; only the former is endemic to Australia.

Seropositive status and patient grouping

The seropositive patients were divided into three groups according to rickettsial antibody titre. Group A (n=13) consisted of patients who had seroconverted (serology changing from seronegative to seropositive). This finding was considered definite evidence of recent rickettsial infection. Group B (n=76) consisted of patients with an antibody titre of 1:512 or greater, which was considered very good evidence of recent rickettsial infection. Group C (n=89) consisted of patients with an antibody titre of 1:256, which was considered likely evidence of recent rickettsial infection.

RESULTS

Survey responses

Approximately half the referring physicians who received survey returned the completed questionnaire (n=178). Results for each question on the questionnaire were collated and separated into group A, group B and group C patients.

Demographic data

A statistically significant difference was observed between male and female patients, and more patients were female than male overall (60%; 107/178). Whereas females predominated in group C, males (40%; 71/178) predominated in group A (Table 1). Both sexes showed a broad range of patient ages.

TABLE 1 |

Comparative demographic and epidemiological features of patients with rickettsial seropositivity.

Patient details		Group A (n=13)	Group B (n=76)	Group C (n= 89)	P value (chi-squared)
1	Sex and age group
	Male (23–89 years)	11 (85%)	32 (42%)	28 (31%)	<0.01
	Female (14–85 years)	2 (15%)	44 (58%)	61 (69%)
2	Occupation/hobby
	Agricultural worker/bush walking (high risk)	8 (62%)	33 (45%)	22 (25%)	<0.08
	Others (low risk)	5 (38%)	35 (46%)	57 (64%)
3	Suspected tick bites	8 (62%)	33 (45%)	43 (48%)	0.76
4	Animal contact	4 (31%)	15 (21%)	22 (25%)	0.72
	-Cattle	1 (8%)	2 (3%)	1 (1%)	-
	-Cat	0 (0%)	2 (3%)	3 (4%)	-
	-Dog	1 (8%)	5 (7%)	4 (4%)	-
	-Kangaroo	2 (15%)	1 (1%)	0 (0%)	-
	-Rhinoceros	0 (0%)	2 (3%)	0 (0%)	-
	-Mixed	0 (0%)	3 (4%)	14 (16%)	-
5	History of presentation of disease
	Australia	12 (92%)	48 (63%)	63 (71%)	0.09
	Overseas	1 (8%)	28 (37%)	26 (29%)

Group A: seroconverted (definite recent rickettsial infection).

Group B: antibody titre >1:512 (probable recent rickettsial infection).

Group C: antibody titre of 1:256 (likely recent rickettsial infection).

Epidemiological data

Various epidemiological features were included in the questionnaire (Table 1). A tick bite was recognised in approximately half the patients (47%; 84/178), and a substantial number of patients (35%; 63/178) reported occupations or hobbies that took them into the Australian bush, such as working in the agricultural sector and/or bushwalking. The latter feature showed a gradation among groups A, B and C, and was most common in group A and least common in group C. Exposure to the Australian bush and being bitten by a tick appeared to be clear risk factors for rickettsial infection.

Clinical features

A summary of clinical features is presented in Table 2. Each feature was recorded as present, absent or not commented on by the doctor. For each feature, the proportion of patients with the respective feature recorded as present was determined.

TABLE 2 |

Comparative (group A, B and C) clinical features of patients with rickettsial seropositivity, numbers of cases (%).

Presenting signs/symptoms	Group A (n=13)	Group B (n=76)	Group C (n=89)	P value (chi-squared)
Fever	10 (77%)	35 (46%)	27 (30%)	0.10
Headache	7 (54%)	33 (43%)	29 (33%)	0.48
Myalgia	5 (38%)	38 (50%)	56 (63%)	0.51
Arthralgia	5 (38%)	24 (32%)	37 (42%)	0.66
Acute fatigue	10 (77%)	43 (57%)	58 (65%)	0.74
Respiratory sign	1 (8%)	14 (19%)	4 (5%)	0.03
Abdominal pain/GI disorder	2 (15%)	5 (7%)	3 (4%)	0.24
Cognitive dysfunction	0 (0%)	4 (5%)	2 (3%)	0.37
Lymphadenopathy	0 (0%)	3 (4%)	0 (0%)	-
Myocarditis	0 (0%)	4 (5%)	0 (0%)	-
Photophobia	0 (0%)	3 (4%)	1 (1%)	-
Disorder thinking	2 (15%)	20 (27%)	25 (28%)	0.74
Eschar	4 (31%)	5 (7%)	3 (4%)	<0.06
Rash:	8 (62%)	14 (19%)	14 (16%)	<0.02
-Macular	4 (31%)	3 (4%)	0 (0%)	-
-Papular	2 (15%)	1 (1%)	0 (0%)	-
-Maculo-papular	2 (15%)	3 (4%)	3 (4%)	-
-Petechial	0 (0%)	7 (10%)	2 (2%)	-
-Erythema migrans	0 (0%)	0 (0%)	1 (1%)	-
-Diffuse	0 (0%)	0 (0%)	1 (1%)	-
-Erythematous	0 (0%)	0 (0%)	1 (1%)	-
-Licheniform rash	0 (0%)	0 (0%)	1 (1%)	-
-Unclassified	0 (0%)	0 (0%)	5 (6%)	-

Groups A, B and C: see footnote in Table 1.

Fever, headache, myalgia and acute fatigue were seen in most patients with rickettsial infection. A rash was seen less often, in only 20% (36/178) of patients, and an eschar was reported in only 7% (12/178) of patients. Features causally associated with rickettsial infection were assumed to be most often seen in group A, followed by group B patients and group C. A gradation was indeed observed for fever, headache, myalgia, respiratory signs, rash and eschar. These clinical features were all statistically more common in group A, less common in group B and even less so in group C (p<0.05 by chi-square test).

Pathology results

The pathology results (Table 3) showed abnormalities in haematology and biochemistry in a subset of patients. Low haemoglobin, an abnormal white blood cell count and low platelets all showed trends towards being most common in group A and least common in group C, thus suggesting that these markers may be causally related to the recent rickettsial infection. However, their sensitivity was low: only 6% (6/105, low haemoglobin), 8% (9/115, abnormal white cell count) and 7% (8/108, low platelets) of patients had these abnormalities. Elevated liver function tests (total bilirubin, ALP, GGT, ALT and AST) also showed trends, with the most abnormal results in group A and the least abnormal results in group C. Likewise, overall sensitivities were not high: 6% (6/102) for total bilirubin, 10% (10/99) for ALP, 14% (16/111) for GGT, 21% (24/117) for ALT and 14% (15/110) for AST in patients in whom liver function was measured.

TABLE 3 |

Significant pathology features of patients with rickettsial seropositivity.

Clinical chemistry		Variables	Group A	Group B	Group C	P value (chi-squared)
1	Haematology values
	Low	Haemoglobin (g/L) R 135–180	3/10 (30%)	2/44 (5%)	1/51 (2%)	<0.01
	Low or raised	WBC (×109/L) R 4–11	3/10 (30%)	4/51 (8%)	2/54 (4%)	0.05
	Low	Lymphocytes (×109/L) R 1–4	3/9 (33%)	8/42 (19%)	6/50 (12%)	0.40
	Low	Platelets (×109/L) R 140–400	3/8 (38%)	3/47 (6%)	2/53 (4%)	<0.01
2	Liver function test
	Raised	Total bilirubin (μmol/L) R <20	4/9 (44%)	1/38 (3%)	1/55 (2%)	<0.01
	Raised	ALP (U/L) R 30–110	4/9 (44%)	4/37 (11%)	2/53 (4%)	<0.01
	Raised	GGT (U/L) R <55	7/11 (64%)	5/48 (10%)	4/52 (8%)	<0.01
	Raised	ALT (U/L) R <45	8/12 (67%)	11/53 (21%)	5/52 (10%)	<0.01
	Raised	AST (u/L) R >35	7/11 (64%)	7/47 (15%)	1/52 (2%)	<0.01
3	Inflammatory markers
	Raised	ESR R 0–22	3/6 (50%)	4/38 (11%)	5/48 (10%)	0.10

Groups A, B and C: see footnote in Table 1.

Longitudinal serological profile

One group A patient had five sequential serology tests conducted over a 12-month period. The data in Table 4 illustrate the antibody titre changes during that time. This patient showed seroconversion (the first serum sample was negative for rickettsial antibodies). By week 4 (second serum sample collected), the patient seroconverted, with a maximum antibody titre of 1:512. If the negative first serum had not been available, this patient would have been classified into group B. A decline in antibody titre occurred over 12 months. Because of antigenic cross-reaction among different rickettsial species, the rickettsial species that infected this patient was unclear, and all five different antigens showed the same antibody titre (1:512). The actual causative rickettsial species shows the highest antibody titre often, but not always, as in this case.

TABLE 4 |

Serological profile over 50 weeks of a patient who seroconverted to rickettsial infection.

Rickettsial antigens	Patient sera (S1–5) assessed with IF assays at different time points
	S1 (week 0)	S2 (week 4)	S3 (week 6)	S4 (week 40)	S5 (week 50)
Spotted fever group rickettsia (SPG)
R. australis	NEG	512	256	256	128
R. honei	NEG	512	256	128	NEG
R. conori	NEG	256	256	256	128
R. africae	NEG	512	128	128	128
R. rickettsii	NEG	512	128	128	128
R. felis	NEG	NEG	NEG	NEG	128
Typhus group rickettsia (TG)
R. prowazekii	NEG	256	NEG	128	128
R. typhi	NEG	512	128	128	128

DISCUSSION

Well-recognised inherent problems are associated with retrospective studies such as this study. However, the physicians’ clinical notes were made at the time of presentation of the unwell patient. Overall, however, most rickettsial infections in Australia (according to referrals to the ARRL) were clinically mild, with fever, headache, myalgia and acute fatigue. They were sometimes associated with haematological and biochemical abnormalities, in patients who had recently been in the Australian bush and/or bitten by a tick. However, rickettsial infections are not always mild [3].

Rickettsial infections initially came to Australia in convicts and early European migrants, in the form of epidemic typhus (R. prowazekii) carried by the human body louse [4]. However, this infection never became established in Australia, presumably because the warm, sunny climate enabled even very poor people to wash themselves and their clothes regularly. Nevertheless, the reactivation form of this disease (Brill-Zinsser disease) has been reported in Australia [5].

Endemic rickettsial bacteria were already present in Australia, associated with native animals and their ectoparasites. Gradually, these endemic rickettsiae demonstrated an ability to infect humans. In 1922, murine typhus was recognised in South Australia [6] (followed by an unspecified endemic typhus [7] and R. honei in 2005 [8]) and then in Western Australia (WA) [9] (and later R. honei in 2016 [10]). Later, scrub typhus became a problem for settlers in northern tropical Queensland [11], along with many other causes of tropical fever [12], and continues to this day [13] in areas including the Torres Strait islands [14]. Murine typhus was detected in Queensland [15,16], New South Wales [17] and Victoria [18]. During World War II, Queensland tick typhus was recognised, initially in Queensland [19,20] and then further south in NSW [21] and Victoria [22]. Many decades later, another SFG rickettsial infection, Flinders Island spotted fever, was detected on Flinders Island in Bass Strait [23–26] and later in Tasmania [27,28] and southern Australia [8]. The most recently recognised rickettsial infection in Australia is cat flea typhus [29]. Laboratory diagnosis of rickettsial infections has recently been reviewed [30].

Outdoor recreationalists and bushwalkers in WA have a high SFG rickettsiae seroprevalence [31] suggesting exposure to rickettsiae in the Australian bush. Many people in Australia have companion dogs and cats, both of which are susceptible to the ubiquitous cat flea. This invertebrate carries R. felis; therefore, urban Australians are at risk of cat flea typhus from their pets if they do not keep them free of fleas. In SE Australia, 12% of dogs are seropositive to SFG rickettsiae [32].

Although Australians infected with rickettsiae might not always present with the typical triad of fever, rash and eschar, they do show clinical, epidemiological and pathological features, as identified herein. Unfortunately, because these features are not pathognomonic, diagnosis of these infections requires physicians’ awareness of Australian rickettsiae and related bacteria species as part of the differential diagnosis for unwell patients. The prospective usefulness of these data remains to be determined. Notification of rickettsial infections in Australia should be required in all states and territories, to establish the true incidence of rickettsial infections.

ACKNOWLEDGEMENTS

The many physicians who responded to the questionnaire are gratefully acknowledged for their contribution to this study.

Dr. John Stenos and Ms. Chelsea Nguyen of the Australian Rickettsial Reference Laboratory (ARRL) are thanked for their contributions to the study.

This study was funded by the Australian Rickettsial Reference Laboratory (ARRL) Foundation Ltd., Geelong University Hospital, Geelong, VIC 3220, Australia.

CONFLICTS OF INTEREST

Both authors contributed equally to all aspects of the research, and there are no conflicts of interest.

REFERENCES

1. Graves SR. Management of rickettsial diseases and Q fever. Med Today. 2013. Vol. 14(7):65–69

2. Stenos J, Graves SR. Laboratory diagnosis of human infections transmitted by ticks, fleas, mites and lice in Australia. Microbiol Aust. 2018. Vol. 39(4):182–184

3. McBride WJ, Hanson JP, Miller R, Wenck D. Severe spotted fever group rickettsiosis in Australia. Emerg Infect Dis. 2007. Vol. 13(11):1742–1744

4. Cumpston JHL, McCallum F; Australia Department of Health. The History of the Intestinal Infections (And Typhus Fever) in Australia 1788-1923: Commonwealth of Australia, Department of Health. Service Publication No. 36. Melbourne: Green. 1927. p. 1–523

5. Tonge JI. Brill’s disease (recrudescent epidemic typhus) in Australia. Med J Aust. 1959. Vol. 46(2):919–921

6. Hone FS. A series of cases closely resembling typhus fever. Med J Aust. 1922. Vol. 1:1–13

7. Dwyer JM, Atkinson N. Some observations on endemic typhus in South Australia. Med J Aust. 1940. Vol. 2(22):573–576

8. Dyer JR, Einsiedel L, Ferguson PE, Lee AS, Gordon DL, Unsworth NB, et al.. A new focus of Rickettsia honei spotted fever in South Australia. Med J Aust. 2005. Vol. 182(5):231–234

9. Saint EG, Drummond AF, Thorburn IO. Murine typhus in Western Australia. Med J Aust. 1954. Nov 6;731–737

10. Raby E, Pearn T, Marangou A, Merritt AJ, Murray RJ, Dyer JR, et al.. New foci of spotted fever group rickettsiae including Rickettsia honei in Western Australia. Trop Med Infect Dis. 2016. Vol. 1(1):5

11. Mathew RY. Endemic typhus in north Queensland. Med J Aust. 1938. Vol. 2:371–377

12. Derrick EH. The challenge of north Queensland fevers. Aust Ann Med. 1957. Vol. 6(31):173–188

13. Stewart AGA, Smith S, Binotto E, McBride WJH, Hanson J. The epidemiology and clinical features of rickettsial diseases in North Queensland, Australia: implications for patient identification and management. PLoS Neglect Trop Dis. 2019. Vol. 13(7):e0007583

14. Unsworth NB, Stenos J, Faa A, Graves SR. Three rickettsioses, Darnley Island, Australia. Emerg Inf Dis. 2007. Vol. 13(7):1105–1107

15. Wheatland FT. A fever resembling a mild form of typhus fever. Med J Aust. 1926. Vol. 1(10):261–266

16. Derrick EH, Pope JH. Murine typhus, mice rats and fleas on the Darling Down. Med J Aust. 1960. Vol. 24:924–928

17. Hughes JC, Diethelm OA, Tebbutt AH. Australian typhus with a report of a fatal case. Med J Aust. 1936. Vol. 7:327–329

18. Jones SL, Athan E, O’Brien D, Graves SR, Nguyen C, Stenos J. Murine typhus: the first reported cases from Victoria. Med J Aust. 2004. Vol. 180:482

19. Andrew R, Bonnin J, Williams S. Tick typhus in North Queensland. Med J Aust. 1946. Vol. 2:253–258

20. Brody J. A case of tick typhus in North Queensland. Med J Aust. 1946. Vol. 1:511–512

21. Wang JM, Hudson BJ, Watt MR, Karagiannis T, Fisher NJ, Anderson C, et al.. Diagnosis of Queensland tick typhus and African tick bite fever by PCR of lesion swabs. Emerg Infect Dis. 2009. Vol. 15(6):963–965

22. Dwyer BW, Graves SR, McDonald M, Yung AP, Doherty RR, McDonald JK. Spotted fever in East Gippsland: a previously unrecognised focus of rickettsial infection. Med J Aust. 1991. Vol. 154:121–125

23. Stewart RS. Flinders Island Spotted Fever: a newly recognised endemic focus of tick typhus in Bass Strait. Part 1. Clinical and epidemiological features. Med J Aust. 1991. Vol. 154:64–69

24. Graves SR, Stewart L, Stenos J, Stewart RS, Schmidt E, Hudson S, et al.. Spotted fever group rickettsial infection in South-Eastern Australia: isolation of rickettsiae. Comp Immunol Microbiol Infect Dis. 1993. Vol. 16(3):223–233

25. Baird R, Stenos J, Stewart R, Hudson B, Lloyd M, Aiuto S, et al.. Genetic variation in Australian spotted fever group rickettsiae. J Clin Microbiol. 1996. Vol. 34(6):1526–1530

26. Stenos J, Roux V, Walker D. Rickettsia honei sp. nov., the aetiological agent of Flinders Island spotted fever in Australia. Int J Syst Bacteriol. 1998. Vol. 48:1399–1404

27. Chin RH, Jennens I. Rickettsial spotted fever in Tasmania. Med J Aust. 1995. Vol. 162(2):669

28. Willis G, Lodo K, McGregor A, Howes F, Williams S, Veitch M, et al.. New and old hotspots for rickettsial spotted fever acquired in Tasmania, 2012-2017. Aust NZ J Pub Hlth. 2019. Vol. 43(4):389–394

29. Williams M, Izzard L, Graves SR, Stenos J, Kelly JJ. First probable Australian cases of human infection with Rickettsia felis (cat-flea typhus). Med J Aust. 2011. Vol. 194(1):41–43

30. Stewart AG, Stewart AGA. An update on the laboratory diagnosis of Rickettsia spp infection. Pathogens. 2021. Vol. 10:1319

31. Abdad MY, Cook A, Dyer J, Stenos J, Fenwick SG. Seroepidemiological study of outdoor recreationists’ exposure to spotted fever group Rickettsia in Western Australia. Am J Trop Med Hyg. 2014. Vol. 91(3):584–588

32. Sexton DJ, Banks J, Graves S, Hughes K, Dwyer B. Prevalence of antibodies to spotted fever group rickettsiae in dogs from southeastern Australia. Am J Trop Med Hyg. 1991. Vol. 45(2):243–248