A print friendly PDF version is available from this Communicable Diseases Intelligence issue's table of contents.

Vectorborne diseases

Introduction

Vectorborne diseases reported to the NNDSS include arbovirus infections and malaria. Arboviruses (arthropod-borne viruses) belong to two families, the alphaviruses, which include Barmah Forest virus infection (BF) and Ross River virus infection (RR) and the flaviviruses, which include dengue and the Murray Valley encephalitis virus (MVE), Kunjin virus and Japanese encephalitis virus (JE). Malaria cases recorded by NNDSS include infections caused by four Plasmodium species.

Arboviruses and malaria are transmitted to humans through the bite of infected mosquitoes. The human population acts as the host species for dengue and malaria. The remaining arboviruses discussed in this chapter have a complex life cycle involving vertebrate hosts, mosquitoes and humans. The vertebrate reservoirs include marsupials, introduced placental mammals and avian species. During epidemics it is possible that newly infected humans further spread these viruses.

Malaria and dengue cases have been reported to NNDSS since 1991. At this time, all other arboviruses were reported to NNDSS as 'Arbovirus - NEC'. Infection with RR became separately notifiable in 1993 and BF in 1995. Infection with MVE, Kunjin virus and JE were separately reported in 2001.

Alphaviruses

Barmah Forest virus infection and Ross River virus infection

Clinical infections with BF and RR are characterised by arthritis, myalgia, fever, headaches and lethargy. A rash is also usually present. The spectrum of illness ranges from the sub-clinical, through to illness that may last for months or even years.⁶⁵ Recent results, however, suggest that persistence for long periods may be overestimated.⁶⁶

Infections with BF and RR are diagnosed by serological tests. As exposure to the viruses may have occurred in the past (resulting in persistent antibodies), it is important to obtain both acute and convalescent sera, approximately two weeks apart, in order to demonstrate seroconversion or the increase in antibody levels that characterises new infections.

In Australia the primary arthropod vectors for both BF and RR are mosquitoes of the Ochlerotatus (previously Aedes) and Culex genera. The primary hosts for RR are believed to be macropods (kangaroos and wallabies), but horses and fruit bats have shown serological evidence of infection. The range of primary hosts for BF is less understood.

The first human infection with BF was identified in 1986. The first recognised outbreak occurred in Nhulunbuy, on the Gove Peninsula in the Northern Territory in 1992.⁶⁷

RR is the most commonly reported arboviral disease in Australia. During 2001, 3,219 notifications of RR infection and 1,141 notifications of BF infection were received. A comparison of the annual trends in the notification rates of RR infection and BF infection since 1995 is shown in Figure 48. The wide fluctuation in the rates of RR infection probably reflects the occurrence of outbreaks against a background of sporadic cases.

Top of pageFigure 48. Trends in notification rates of Barmah Forest virus infection and Ross River virus infection, Australia, 1995 to 2001, by year of onset



The number of cases of BF notified in 2001 was the largest recorded since it became separately notifiable in 1995. The national notification rate for 2001 was 5.9 cases per 100,000 population. Prior to this rates have ranged between 4.8 cases per 100,000 population (1996) and 2.8 cases per 100,000 population (1998). In 2001 most notifications of BF infection came from Queensland (n=603) and New South Wales (n=398). The highest rate at the state and territory level occurred in the Northern Territory (18.5 cases per 100,000 population).

By contrast, the number of notifications for RR infection in 2001 was lower than in 2000 (n=3,219). While this represents a decrease of 25 per cent compared to the previous year, Figure 48 indicates that there is variation in reports over years. The overall rate for 2001 was 16.5 cases per 100,000 population. Most notifications came from Queensland (n=1,569) and the highest rate was observed in the Northern Territory (111.5 cases per 100,000 population). An outbreak of 18 cases of RR was reported in the Morgan Council area, which lies on the lower Murray River in South Australia.

The co-occurrence of BF and RR infection between 1995 and 2001 by month of onset, measured as the ratio of RR to BF notifications, is shown in Figure 49. The seasonality of the curve possibly represents a more episodic characteristic of BF compared to RR. Further exploration of the epidemiology of the two viruses at smaller geographical areas and over longer periods will allow the interaction between them to be better understood.

Figure 49. Trends in ratio of Ross River virus infection to Barmah Forest virus infection notification, Australia, 1995 to 2001, by month of onset



Both BF and RR infections show similar age group and sex distributions (Figures 50 and 51). For both, there is a relative paucity of notifications in younger children and the early teenage years. The highest rates of notification were observed in 35-60 year old males and females, with the highest rates being about 10 cases per 100,000 population for infection with BF, and between 25 and 30 cases per 100,000 population for RR infection.

Figure 50. Notification rates of Barmah Forest virus infection, Australia, 2001, by age group and sex



Top of pageFigure 51. Notification rates of Ross River virus infection, Australia, 2001, by age group and sex



In 2001 infection with RR and BF were notified relatively equally in males and females. For BF infection, the male to female ratio was 1.2:1, and for RR infection the male to female ratio was 1.1:1.

Notification rates by Statistical Division are shown for the two diseases in Maps 8 and 9. The areas of highest rates of occurrence for BF infection were the New South Wales Mid-North Coast Statistical Division (82.2 case per 100,000 population) and the south-west areas of Queensland (44.6 cases per 100,000 population). For notifications of RR infection the highest rates occurred in north-west Queensland (186.4 cases per 100,000 population) and the Kimberley Statistical Division in Western Australia (159.0 cases per 100,000 population).

Map 8. Notification rates of Barmah Forest virus infection, Australia 2001, by Statistical Division of residence



Top of pageMap 9. Notification rates of Ross River virus infection, Australia 2001, by Statistical Division of residence



There were geographically isolated outbreaks of BF, in the south-west of Western Australia (the South West and Lower Great Southern Statistical Divisions), and outbreaks of RR along the Murray River region.

There is debate as to whether there is any overall increase in the activity of RR, and whether disease activity is expanding into areas where it has not occurred before.^68,69,70,71 Variations in notifications of RR can reflect the natural ecological variability of virus activity, while the availability of susceptible populations and human geographical and environmental change, also affect rates of RR in humans. At the local level, the numbers of infections have been shown to correlate with the abundance of various mosquito species.⁷² Mosquito control programs may affect vector abundance and virus transmission to humans. Possible artefacts should be considered when assessing surveillance data. Increased awareness and testing for RR may account for increases in notifications. The change in reporting methods from clinical diagnosis to laboratory notification together with the adoption of standardised clinical and laboratory definitions may also have contributed to changes in surveillance data.

Change in the rate of notification of BF may also be affected by surveillance artefacts or may be the result of changes in the epidemiology of this infection. A combination of unusual environmental conditions, together with a highly susceptible (that is, previously unexposed) human population, were the most probable factors contributing to the magnitude of the BF epidemic on the New South Wales south coast in 1995.⁷³ This outbreak is possibly reflected in the lower than usual ratio of RR to BF notifications for that year in Figure 49.

The extensive outbreak of RR in the Western Pacific in 1979-80 demonstrates the potential of the virus to spread throughout the region.⁶⁹ Serological surveys have shown the virus to exist in PNG, areas of Indonesia and in the Solomon Islands,^74,75 and the virus has recently been isolated for the first time from PNG.⁷⁶ A competent vector is generally required for transmission to humans. The introduction of Aedes camptorhynchus to New Zealand during 1998, highlights the risk to neighbouring countries.⁷⁷ During the late 1970s, at the time of the Pacific outbreak, three cases of RR infection were notified in New Zealand but all were acquired outside the country, and none have been notified since.

Top of page

Murray Valley encephalitis and Kunjin

MVE is enzootic in the Kimberley region of Western Australia and the top two-thirds of the Northern Territory, where it is active principally in the wet season. The virus is epizootic in the Pilbara of Western Australia and the southern part of the Northern Territory, where it is associated with high summer rainfalls and flooding. Human cases occur sporadically in northern Queensland. Since 1974, however, nearly all cases of arboviral encephalitis due to MVE have been reported from Western Australia and the Northern Territory, with MVE activity (as indicated by seroconversion in chicken flocks) and human disease occurring in most years.^75,78 The major vector for the virus is Cx. annulirostris, and the main vertebrate hosts are water birds of the ardeid (heron) family.^75,78

Although most cases are asymptomatic, infection with MVE can lead to serious illness or death. Symptoms include headache, neck stiffness, fever, tremor, weakness, confusion, fitting and sometimes coma.⁷⁹ The case fatality rate following symptomatic infection with MVE is 20 per cent and approximately 25 per cent of survivors will be left with significant residual neurological damage.⁸⁰ Infection with Kunjin virus generally causes milder disease,⁷⁹ but recent cases of Kunjin-associated encephalitis have been reported from Central Australia.^81,82 The number of MVE infections that are asymptomatic have been estimated to be between 500 and 1,500 for each one that is clinically identified.^75,78

Infections with MVE and Kunjin virus have been separately notifiable since 2001. Prior to this, information on the location of cases and deaths was recorded in individual states and territories. A summary of these cases has previously been reported.⁷⁸ In addition, a summary of cases, and of the ecology and epidemiology of Kunjin virus, has been extensively reviewed because of its reclassification as a member of the West Nile virus lineage 1.⁸³

In 2001, five cases of MVE and four cases of Kunjin virus infection were notified. Table 16 describes the eight cases that occurred in the first half of the year (corresponding to the 2000-01 arbovirus season). From January to June 2001, four notifications of MVE infection and four cases of infection with Kunjin virus were notified. In the second half of the year (the 2001-02 arbovirus season) only one case of MVE was notified. This was reported in the Northern Territory in July. The case was a two-year-old female. There were no cases of Kunjin notified in the second half of the year.

Table 16. Notifications of infection with Murray Valley encephalitis and Kunjin viruses, Australia, 2001

Notifying jurisdiction	Murray Valley encephalitis virus infection: Jan - June 2001
	Onset month	Gender	Age (yrs)	Follow-up
Qld	Feb	M	3	Alive, severe neurological sequelae
NT	Feb	F	49	Died
SA	Feb	M	59	Alive
WA*	Mar	M	60	Unknown
	Kunjin virus infection: Jan - June 2001
	Onset month	Gender	Age (yrs)	Follow-up
NSW	March	Unknown	58	Unknown
NT	March	M	11	Alive
WA	March	F	27	Unknown
NT	May	F	23	Alive

* Possibly acquired in Queensland or the Northern Territory.

Although cases of MVE and Kunjin were both notified from Western Australia and the Northern Territory, there was some uncertainty as to where the infections were acquired as they occurred in tourists travelling through northern Australia. An additional case of MVE was diagnosed in Germany in a tourist returning from travel in Northern Australia.⁸⁴

Viral infection of mosquitoes and seroconversion in sentinel animals may be used to monitor the potential for human infections with MVE and Kunjin virus. The locality of sentinel chicken flocks is shown in Map 10, and Figures 52a and 52b chart the number of seroconversions to MVE and Kunjin respectively during 2001. Although seroconversion counts for the different states and the Northern Territory are given they are not directly comparable as the number of flocks and birds within them vary across jurisdictions. Seroconversion in sentinel chicken flocks occurs mostly during the arbovirus season, but also throughout the year.

Top of pageFigure 52a. Numbers of seroconversions to Murray Valley encephalitis virus in sentinel chickens, New South Wales, Northern Territory and Western Australia, 2001



Figure 52b. Numbers of seroconversions to Kunjin virus in sentinel chickens, New South Wales, Northern Territory, Western Australia, and Victoria, 2001



Map 10. Geographical distribution of sentinel chicken flocks for the surveillance of arboviruses, Australia, 2001



Surveillance of MVE and its control have been recently reviewed and an integrated system of surveillance based on sentinel chicken and mosquito monitoring was recommended.⁷⁸ As viral activity is demonstrated in these systems, public health control measures can then be implemented to prevent human infections.

Top of page

Japanese encephalitis

Most infections with JE are asymptomatic. Between 1 in 30 and 1 in 300 infections result in clinical disease. ⁵ The fatality rate in symptomatic cases can be as high as 30 per cent, and neurological sequelae are reported in 50 per cent of survivors. A higher case-fatality rate is reported in the elderly, but serious sequelae (neurological) are more frequent in the very young.

Surveillance efforts in Australia have focussed on the detection of JE virus activity in Torres Strait and are based on the detection of virus carriage in the mosquito vector (Cx. annulirostris) and infection of domestic pigs, the amplifying host. In 1995 a localised outbreak of three cases occurred in the Torres Strait region.⁸⁵ Two additional sporadic cases were detected in 1998 at Badu Island and at Mitchell River, on the Gulf in Far North Queensland.⁸⁶ There were no notifications of JE in Australia in 2001. The absence of human infections in Australia since 1998 can be attributed to a human vaccination program and changes in pig husbandry following the earlier outbreaks.⁸⁶

Feral pigs may be important in the ecology of JE in the Far North Queensland region. Birds, abundant during the wet season in tropical regions, may also play a role in virus transmission cycles. Native Australian macropods though, are thought to be unlikely hosts for JE.⁸⁷

JE is a widespread and emerging disease in South-East Asia, and is probably now endemic in New Guinea.^75,88 The practice of rice-paddy farming throughout much of South-East Asia provides favourable conditions for the mosquito vector, and intensive pig husbandry provides an ideal reservoir species for the virus. These factors, combined with large and growing human populations, provide a fertile environment for the spread of the disease.

Despite an effective vaccine, outbreaks of the disease continue to occur and extend its known range. In Indonesia, the general absence of pig farming is associated with a relative lack of JE. The island of Bali is an exception because of its different religion and attitudes towards pork. The 1995 and 1996 cases of JE in the Torres Strait and West Papua marked a sudden unexplained range extension to the east. The isolation of the virus at three locations since then, and serological detection of more cases suggest that the virus has become endemic on the island of New Guinea.⁸⁸

Top of page

Dengue

Infection with any of four dengue viruses (serotypes 1- 4) is characterised by fever, muscle and joint pain, lymphadenopathy and rash. Secondary infection with a heterologous serotype of dengue virus can result in dengue shock syndrome or dengue haemorrhagic fever.^89,90

Dengue was reintroduced to Australia in 1981 after an absence of more than 25 years.⁹¹ Dengue virus activity occurs only in Far North Queensland, where the host mosquito Aedes aegypti has become established. Because of the absence of the mosquito vector elsewhere in Australia, dengue cases reported from other states and territories are infected either in Queensland or overseas.

During 2001, 176 notifications of dengue were received, corresponding to a national rate of 0.9 cases per 100,000 population. The largest number were from New South Wales (n=50, 0.8 cases per 100,000 population), and 43 notifications came from both the Northern Territory and Queensland (21.5 and 1.2 cases per 100,000 population respectively). Most notifications were for adults between the ages of 20 and 50 years, with a male to female ratio of 1.1:1. Rates according to age group and sex are shown in Figure 53. Relatively low numbers of notifications (n=176) most probably account for gaps in the age distribution.

Figure 53. Notification rates of dengue, Australia, 2001, by age group and sex



In Queensland eight of the 43 cases notified were locally acquired, 21 were acquired overseas, while the place of infection was unknown for the remaining 14 cases. All of the remaining cases of dengue notified from other jurisdictions in 2001 were known to have been acquired overseas, except for three cases where the place of acquisition was not known. Information on dengue serotype was available for six cases notified from Victoria. Five were of serotype 1, and one was serotype 3.

The number of dengue notifications by month between 1991 and 2001 is shown in Figure 54. Two extensive outbreaks are represented in these data. The first was in Townsville and Charters Towers, in 1992-93, and the second in Cairns, Mossman and Port Douglas from 1997 to 1999. In these epidemics, 900 and 498 cases, respectively, were notified. The first outbreak was of dengue type 2, and the second outbreak was of dengue type 3. For 2001, the number of notifications are similar to the number notified in the preceding year (n=216), which included an outbreak of 49 cases in Cairns.

Figure 54. Trends in notifications of dengue, Australia, 1991 to 2001, by month of onset



The five-year Dengue Fever Management Plan for North Queensland 2000-2005 aims to improve disease surveillance in humans, enhance mosquito control and surveillance, and to educate the community and professional groups on mosquito control and the prevention of infection (Queensland Government Health Department: Dengue fever for North Queensland, 2000-2005, http://www.health.qld.gov.au/phs/Documents/tphun/9168dmp.htm). The effectiveness of control measures when undertaken without delay was demonstrated in 2001, when an outbreak in Townsville was limited to only nine cases.⁹²

Dengue affects large numbers of people throughout the Pacific region and remains a key public health concern. In 2001 the WHO reported 132,949 cases and 586 deaths from dengue in the WPR (www.wpro.who.int/public/regstatistics/reg_spec.asp). French Polynesia reported 30,000 cases of dengue serotype 1. The epidemic spread to other states in the Pacific, including Western Samoa, New Caledonia, American Samoa, Tokelau and the Cook Islands. New Zealand reported 60 imported dengue cases.

Factors contributing to increases in global dengue incidence include the creation of mosquito breeding sites by poorly planned urbanisation, lapses in mosquito eradication programs in the Americas and large-scale movements of people and cargo around the world. In the 1950s an annual average of 900 cases of dengue haemorrhagic fever were reported worldwide. This increased to over 514,000 cases during the 1990s, and in 1998 1.2 million cases of dengue and dengue haemorrhagic fever infections were reported. Modelling suggests that over 51 million infections in total occur each year.⁹³

Top of page

Arbovirus - not elsewhere classified

Thirty-six notifications were categorised as 'Arbovirus - not elsewhere classified' in 2001. New South Wales and Victoria reported 15 and 16 notifications respectively. Cases may include unspecified flavivirus infections (where serology is inconclusive and Kunjin or MVE cannot be distinguished) or infections caused by other arboviruses.

Top of page

Malaria

Indigenous transmission of malaria in Australia occurs infrequently. Australia has maintained its WHO malaria-free status since 1983, despite the continued presence of competent Anopheles vectors, principally An. farauti. This mosquito is found in coastal regions of the Northern Territory and Queensland north of approximately the 17 or 18 degree south of latitude.

All 705 malaria cases reported to NNDSS in 2001 were people who had returned to Australia from malaria endemic regions (Table 17). Most notifications were reported from Queensland (n=300), New South Wales (n=153) and Victoria (n=88). The incidence of malaria has been stable over the last decade, with about 700 notifications occurring each year, giving an annual rate ranging between three and five cases per 100,000 population (Figure 55). Most reported cases were in the 20-24 and 25-29 year age groups. Overall, the male to female ratio was 2.4:1. Rates according to age group and sex are shown in Figure 56.

Table 17. Notifications of malaria, Australia, 2001, by country of infection

Country of infection	State or territory notifying* (% of total)
	ACT	NSW	NT	Qld	SA	Tas	Vic
PNG	63.1	42.2	19.3	24.7	19.0	11.1	40.2
Solomon Islands	5.3	3.5	1.8	0.0	3.0	0.0	0.0
Vanuatu	5.3	1.7	0.0	0.0	0.0	0.0	0.0
Indonesia	5.3	8.1	33.3	3.0	6.0	11.1	8.1
Vietnam	0.0	0.6	0.0	0.3	0.0	0.0	0.0
East Timor	0.0	13.9	43.9	3.7	3.0	0.0	14.9
India/Pakistan	10.5	9.8	1.7	0.0	9.0	0.0	0.0
Afghanistan	0.0	0.6	0.0	0.0	0.0	11.1	0.0
Africa	10.5	17.3	0.0	0.7	39.0	55.6	13.8
South America	0.0	0.0	0.0	0.3	0.0	0.0	2.3
Other	0.0	2.3	0.0	0.0	9.0	0.0	20.7
Unknown	0.0	0.0	0.0	67.3	12.0	11.1	0.0

* Data unavailable from Western Australia

Figure 55. Trends in notification rates of malaria, Australia, 1991 to 2001, by year of onset



Top of pageFigure 56. Notification rates of malaria, Australia, 2001, by age group and sex



The species of Plasmodium the causing of infection was identified for 274 notifications (38%) (Table 18). Of these most were P. vivax (67%) and P. falciparum (28%).

Table 18. Notifications of malaria, Australia, 2001, by Plasmodium species

Plasmodium sp.	Number of notifications by species	Per cent species where known
P. vivax	184	67
P. falciparum	76	28
P. vivax AND falciparum	2	1
P. malariae	8	3
P. ovale	4	1
Unknown	431
Total	705

A large number of malaria infections in Australians have occurred in defence personnel during their peace-keeping operations overseas (LtCdr P Corrigan, personal communication). The number of malaria cases reported by area of operation (East Timor, Bougainville, Papua New Guinea and 'other') and against Plasmodium species by year is shown in Table 19.

Table 19. Number of malaria cases reported to the Army Malaria Institute, 1998 to 2001, by area of operation and Plasmodium species*

Year	East Timor					Bougainville/Papua New Guinea					Other				Total
	Pf	Pv	Pf/ Pv	Po	Unk	Pf	Pv	Pf/ Pv	Po	Unk	Pf	Pv	Po	Unk
1998	0	0	0	0	0	0	9	0	0	0	0	2	0	0	11
1999	23	7	2	0	0	3	22	1	0	0	0	2	1	0	61
2000	26	290	6	1	14	3	27	1	1	1	2	23	0	1	396
2001	1	46	0	0	2	3	5	0	0	2	0	11	0	2	72

* Pf - Plasmodium falciparum; Pv - P. vivax; Pf/Pv - infected with both; Po - P. ovale, Unk - unknown

These data represent clinical occurrences of malaria, and so include relapses of previously acquired infections. Cases are diagnosed either in the area of operation, or on return to Australia. Infections with P. falciparum are commonly diagnosed in East Timor, whereas diagnoses of P. vivax are more commonly detected on return to Australia. It should be noted that infections with P. falciparum can mask dual infection with P. vivax. After treatment for the P. falciparum infection, the P. vivax infection can then become patent.

The endemicity of malaria in South-East Asia and the Pacific and the presence of competent mosquito vectors in Australia underscores the potential for the re-introduction of malaria to Australia.


This article was published in Communicable Diseases Intelligence Volume 27, No 1, March 2003.