Introduction

Australian bat lyssavirus (ABLV) is a member of the Rhabdoviridae virus family, genus Lyssavirus. There are 12 recognised species (genotypes) within the genus Lyssavirus, including classic rabies virus and other closely related lyssaviruses (e.g. European bat lyssaviruses) [1, 2]. Lyssaviruses are usually transmitted to humans via bites or scratches from an infected animal, which inoculate virus-laden animal saliva through human skin into muscle and subcutaneous tissues [3].

ABLV was first identified in Australia in 1996 from the brain of a black flying fox (Pteropus alecto) in New South Wales [4]. There has been virological evidence of ABLV infection in four common species of megachiroptera (megabat, including flying fox) in mainland Australia, and one species of microchiroptera (microbat) [1] [5]. These five ABLV reservoir bat species are widely distributed along the seaboard and inland Australia (Fig 1). Further serological evidence of ABLV infection has been reported in five of the six families of microbats in Australia [5]. Therefore, it is considered that all Australian bats have the potential to transmit ABLV.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 1. Distribution of ABLV host reservoir bat species in mainland Australia, reproduced from [5] under a CC BY license.

https://doi.org/10.1371/journal.pntd.0005227.g001

To date, the only three known human cases of ABLV, all fatal, have occurred in Queensland, Australia. Cases occurred in 1996 [6], 1998 [7], and 2013 [8], with each case having a history of bat bite/s and/or scratch/es within Queensland.

In Australia, a bite or scratch by a bat, or mucous membrane/broken skin exposure to bat saliva or neural tissue, constitutes a potential exposure to ABLV. Under Queensland’s Public Health Act 2005 [9], potential exposure to ABLV is a notifiable condition requiring immediate reporting by telephone call or facsimile to the local public health unit.

In this paper, we report on 1) notifications of potential exposures to ABLV in Queensland between 2009 and 2014; 2) the nature of and reasons for potential exposures; 3) provision of post-exposure prophylaxis; and 4) ABLV detection among bats involved in potential human exposure and available for testing.

Methods

Results

Notifications of potential exposures to ABLV, 2009–2014

There were 1,515 potential exposures to ABLV notified in Queensland residents over the period 2009–2014 (Table 1), with 1,500 occurring in Queensland and 15 occurring in other Australian states/territories. This gives an average annual notification rate of 5.6 potential exposures per 100,000 population per year. The highest notification rate was reported in 2013 (10.9 per 100,000 per year).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Notifications of potential exposures to ABLV in Queensland, 2009–2014.

https://doi.org/10.1371/journal.pntd.0005227.t001

Potential exposures occurred at all times of the year, with relatively fewer potential exposures in the winter months (June–August, 16% of total notifications) (Fig 2).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 2. Potential exposures to ABLV notified in Queensland by month of potential exposure, 2009–2014.

https://doi.org/10.1371/journal.pntd.0005227.g002

Time lags between potential exposure dates and notification dates to PHUs are shown in S1 Fig. Notifications to PHUs were made on the same day for 41% of potential exposures; 76% were notified within 2 days following potential exposure; 86% were notified within 7 days following potential exposure; and 93% were notified by 21 days. The median time from potential exposure to notification was 1 day. The longest period from potential exposure to notification was 3 years and 8 months (1,344 days). There was no statistically significant difference in median times from potential exposure to notification by gender, age group (under 40 year olds versus 40 years or older), and occupational group (those with occupational contact with bats versus the general public). Time lag patterns were similar for notifications in 2013 and those in other years (S2 Fig).

Demographic details of people potentially exposed

The median age of people notified as having a potential exposure to ABLV was 42 years. The highest number of notifications occurs in the 40–49 year age group and the highest notification rate was in the 50–69 year age group (Table 2). Males had generally higher notification counts than females, but notification rates were not statistically different for the two groups.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 2. Notifications of potential exposures to ABLV by sex and age group, 2009–2014, Queensland.

https://doi.org/10.1371/journal.pntd.0005227.t002

There was variation in average annual notification rates across geographical areas of residence (Fig 3). Notification rates ranged from 3.7 per 100,000 per year in Metro South to 12.8 per 100,000 per year in Cairns and Hinterland (S1 Table). Rates were generally higher in north Queensland (the tropical area) than in south Queensland (the temperate area).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 3. Average annual notification rates of potential exposures to ABLV by Hospital and Health Service of residence in Queensland, 2009–2014.

https://doi.org/10.1371/journal.pntd.0005227.g003

Nature of potential exposures to ABLV

Animals.

Of 1,515 cases of potential exposures to ABLV, 1,461 (96.4%) were exposed via contact with bats, 8 (0.5%) via contact with two ABLV infected horses, 20 (1.3%) via contact with an ABLV infected human, and 26 (1.7%) via contact with animals thought to be bats. Of these 1,461 cases of potential exposures to ABLV via contact with bats, 1,001 reported contact with megabats (including 985 flying foxes), 254 contact with microbats, and 206 contact with unspecified bats.

Routes of potential exposure.

The most common potential exposures were through bat scratches (709, 46.8%) and bites (565, 37.3%), with less common routes being mucous membrane/broken skin exposure to bat saliva/brain tissue (34, 2.2%), mucous membrane/broken skin exposure to horse/human saliva or brain tissue (28, 1.8%), and other close contact with bats, including situations during which infectious contact could not be excluded (179, 11.8%).

Circumstances of potential exposure.

The most common circumstance of potential exposure was intentional contact with a bat by the general public (56.3%), followed by accidental contact with a bat by the general public (32.7%), occupational contact with bats (3.5%), and contact with bats by volunteer bat carers and their families (1.2%). Contact with two ABLV infected horses or an ABLV infected human occurred in 1.8% of notifications (Table 3).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 3. Circumstances of potential exposures to ABLV in Queensland, 2009–2014.

https://doi.org/10.1371/journal.pntd.0005227.t003

The most common activities for intentional contact with a bat by the general public included: the rescue of a bat caught in a fence (mostly described as barbed wire fences), in fruit tree netting, or following attack by dogs/cats; intentional removal of a bat from a room; and contact when a bat was touched or fed.

Accidental contact with a bat by the general public most commonly occurred when people encountered bats inside their houses, or during outdoor exercise.

Occupational contact with bats was reported in 53 cases, including staff of animal clinics (18), veterinarians (8), bat collectors/handlers (8), and bat researchers (6).

Post-exposure prophylaxis

Of 1,515 notified cases of potential exposures to ABLV, 1,399 (92.3%) received PEP consistent with national guidelines at the time of potential exposure; the remaining 116 (7.7%) had PEP use varying from national guidelines (Table 4).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 4. Use of post-exposure prophylaxis (PEP) among notified cases of potential exposures to ABLV in Queensland, 2009–2014.

https://doi.org/10.1371/journal.pntd.0005227.t004

ABLV testing in bats

Over the period 2009–2014, 393 bats involved in ABLV potential exposure incidents were available for testing (Table 5). Most of these bats were retrieved in circumstances where people intentionally rescued sick/injured bats or had occupational exposure to bats. Of those bats, 20 had ABLV detected, with an overall ABLV detection rate of 5.1% in bats tested. In addition, fourteen bats tested equivocal for ABLV by Immunofluorescent Antibody (IFA) testing but negative for ABLV by PCR method.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 5. Detection of ABLV in bats available for testing, 2009–2014.

https://doi.org/10.1371/journal.pntd.0005227.t005

Details on bat species and ABLV testing results are shown in Table 6. Bats classified as Pteropus genus (megabats) accounted for 80% (314/393) of bats tested, and 19 of 20 (95%) bats testing positive for ABLV belonged to this genus. The ABLV detection rate was 6.0% (19/316) among megabats tested and 1.6% (1/64) among microbats tested. For individual bat species, the ABLV detection rate was 14.9% among Pteropus scapulatus, followed by Pteropus conspicillatus (7.7%), Pteropus poliocephalus (7.7%), and Pteropus alecto (2.6%). Of 3 Saccolaimus flaviventris microbats tested, one (33.3%) was ABLV positive.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 6. Bat species and ABLV testing results, 2009–2014.

https://doi.org/10.1371/journal.pntd.0005227.t006

Discussion

There are two key findings that emerge from our review of six years of potential human exposures to ABLV in Queensland. Firstly, potential exposure most commonly occurred through intentional contact with bats. Secondly, up to a quarter of notifications occurred after two days from the potential exposure, but with some delays being more than three weeks. Both of these findings highlight avoidable risks in terms of potential exposure and preventing disease progression, and should continue to be part of the key messages in future health promotion and public education campaigns on this topic.

Timely and appropriate PEP remains the only effective measure in preventing human ABLV infection following scratches/bites by bats or mucosal contact with bats’ saliva/neural tissue [2, 15]. The incubation period of classic rabies is usually 3–8 weeks (but can be highly variable). The incubation period for the three known human ABLV cases has ranged from 4 weeks to 27 months [6–8]. Immediate management of any potential exposure is recommended [2], making any delay in notification potentially very serious. It is important that individuals potentially exposed to ABLV seek medical care immediately to receive timely post-exposure prophylaxis. Our data show that when a potential exposure was notified, there was a high level of appropriate PEP administration, with 92% of notified cases with PEP use consistent with national guidelines. This included approximately 16% of notified cases who did not require PEP or had their PEP discontinued due to confirmation of negative ABLV testing results in bats. The remainder all had either refusal of PEP or partial PEP documented. Where partial PEP was documented, the majority had appropriate doses of vaccine but with no documentation of HRIG. Importantly, retrieval of bats for laboratory testing should be performed by trained, vaccinated, and well equipped professional bat handlers, to avoid further exposure from bat bites/scratches.

Statewide notification data demonstrate that 56% of potential exposures to ABLV were from members of the public who intentionally handled bats, most commonly in circumstances such as rescue of trapped bats from fences or fruit tree netting, or removal of bats from a home. Fewer than 5% of reported potential exposures were in professional groups (e.g. veterinarians, bat handlers) and volunteer bat carers. Previous data from the Brisbane South PHU [16–18] showed that professional bat handlers and volunteer bat carers accounted for 50% of potential exposure cases in 1996–1999 (the first two fatal human ABLV cases were reported in 1996 [6] and 1998 [7]), but the proportion reduced to around 20% in 2000–2008. In contrast, the proportion of potential exposures reported by the general public who intentionally handled bats increased from approximately 40% to 60% over the same period [18]. Public health information campaigns to raise awareness of ABLV risks targeted professional and volunteer groups handling bats, which may have partially contributed to the substantial decline in the proportion of potential exposures from these groups [16]. There may also be potential for under-reporting among volunteer bat carer groups due to requirement of bat euthanasia for testing [17]. However, a sustained high proportion of potential exposures from the general public who continue to handle bats intentionally highlights a continuing need to address the strong motivation of some members of the public to help bats in distress.

Recent surveys of community members in Queensland [19] and New South Wales [20] shed light on perception of bat risk and intention to handle bats from the general population. In 2014 Young et al [19] reported a survey of residents in South East Queensland and found that 20% (140/700) of participants considered bats as a high risk to human health, 31% considered a moderate risk, and 42% considered a low risk. Approximately 25–30% of participants from the survey indicated they would handle a bat trapped in wire mesh fencing or a net over a fruit tree, and 56% would handle a bat coming into their houses. Similarly, a population-based survey in New South Wales [20] reported that a quarter of respondents would handle an injured or trapped bat. Key reasons for people reporting a willingness to handle bats included [19]: 1) protection of their families, pets, and themselves from bats, in situations such as a dead bat in the yard or a live bat in the house; 2) protection of injured or trapped bats; and 3) perception of minimal or no risk from injury if implements, gloves, or towels were used to handle bats.

The largest number of potential exposures to ABLV (508 individuals) was notified in 2013. This was probably attributable to increased community awareness of ABLV in the context of bat exposures following the reporting (both through Queensland Health and the media) of a human death from ABLV in February 2013 [8], and of the first confirmed ABLV infections in two horses in May 2013 [10].

Given the considerable proportion of people (approximately a quarter) in the community who reported an intention to handle injured or trapped bats [20], public health interventions should adopt multiple strategies, including: 1) communicating the risk messages to the general public that they may harm the bat and put themselves at risk of contracting the fatal ABLV infection if they attempt to rescue/handle the animal [21], because an untrained person is highly likely to be injured when handling a bat, and the bat involved in human injury will be euthanised for ABLV testing. 2) advising the public to take an alternative action when encountering injured/trapped bats; for example, contacting animal rescue groups (e.g. RSPCA) [22] to handle bats, as these professionals are vaccinated, experienced and well equipped to deal with bats; 3) collaboration with industry for better design of fences and fruit tree netting to minimise bats being trapped; and 4) education on the need for immediate cleansing of wound (with soap and water for at least 5 minutes and application of povidone-iodine or alcohol) when indicated, and to immediately seek medical attention following bat scratches/bites or mucosal membrane/broken skin contact with bat saliva/neural tissue.

Our bat testing data confirmed that ABLV was detected in four species of megabats (all flying foxes) and one species of microbats over the reporting period: Pteropus scapulatus (common name—little red flying fox), Pteropus alecto (black flying fox), Pteropus conspicillatus (spectacled flying fox), Pteropus poliocephalus (grey-headed flying fox), and Saccolaimus flaviventris (yellow-bellied sheathtail bat). Currently flying fox camp locations and populations have been monitored in Australia through the National Flying Fox Monitoring Programme [23]. There are two distinct variants of ABLV: ABLVp found in Pteropus bats and ABLVs found in Saccolaimus bats [5].

Based on our data, 5.1% of bats tested were ABLV infected (6.0% for megabats and 1.6% for microbats). Recent data from Wildlife Health Australia showed the prevalence of ABLV in bats submitted for testing was 6.7% [24]. For individual species, the little red flying fox had relatively high prevalence of ABLV infection (14.9% in our data, 17.5% reported by Field [25], and 16.9% by Barrett [26]). It was noted that of 3 yellow-bellied sheathtail microbats tested, one (33%) was ABLV positive. A previous study found 5 of 7 (71.4%) yellow-bellied sheathtail bats tested ABLV positive [26], an indication of extremely high prevalence of ABLV among such bat species. The ABLV infection rate in sick, injured, or orphaned bats with central nervous symptoms could be as high as 21% [26]. Sick or injured bats are more likely to be involved in human-bat interactions [20]. These findings further reinforce the message of avoiding any bat handling by the general public.

Eight people were exposed to two ABLV infected horses which were first confirmed in 2013 [10]. Spillover of ABLV from natural reservoir bats to horses raises the question of whether ABLV can transmit to other animals, such as dogs. Contact between domestic dogs and ABLV infected bats has been reported previously [27], and our data indicate occasional pet animal-bat interactions (dogs/cats attacking bats on 62 occasions). Molecular studies on lyssavirus host cell entry mechanisms have found that ABLV host cell receptor is broadly conserved among mammals [28], indicating these animals are likely to be susceptible to ABLV infection. Future research is needed to better understand the potential transmission of ABLV from bats to domestic animals and implications for public health.

Some caution is needed when interpreting our notification data. Firstly, the number of potential exposures to ABLV in Queensland was likely to be under-reported. The trivial nature of the injury, coupled with lack of awareness of serious risk associated with bat-related injury, and subsequent failure to seek medical attention by some individuals potentially exposed to ABLV are likely to contribute to under-reporting. It is important that any person with bat-related injury sees a doctor immediately—no matter how small the injury is [15]. Secondly, applying the case definition on what constitutes a potential exposure to ABLV in public health practice can be challenging in some situations and careful consideration of the use of HRIG as a scarce resource is required. For instance, providing PEP to a person who: is scratched by a flying creature at night when that creature has not been identified as a bat; or awakens to a bat in the bedroom but is not aware of any exposure. Consideration for PEP on a case by case basis is warranted in public health practice to prevent rare but fatal disease. Thirdly, there was insufficient documentation of reasons to explain why PEP use in some individuals varied from guidelines. Our data showed 95 people had appropriate doses of rabies vaccine, but had no information on why HRIG was not documented. HRIG rationing measures applied during periods of HRIG shortage might contribute to restriction of HRIG use in some of these cases, however there is a need to clearly document the role of these restrictions in the PEP recommended.

While human disease from ABLV is uncommon, historically, identified clinical disease has been universally fatal. Further efforts are required to minimise potentially infectious contact between humans and bats, and to ensure that timely notification and appropriate post-exposure prophylaxis is administered following potential exposures to ABLV.

Supporting Information

Acknowledgments

We thank staff at the Public Health Units for management of notified cases of potential exposures to ABLV, provision of surveillance data, and comments and feedback to this report.

We thank staff at Queensland Health Forensic and Scientific Services for undertaking laboratory testing of ABLV in bats.

We appreciate comments and feedback to this report provided by Biosecurity Queensland, and Workplace Health and Safety Queensland.

We appreciate comments from Dr Margaret Young, Wide Bay Public Health Unit, Wide Bay Hospital and Health Service, Queensland Health.