The random opportunistic bleeding of chickens in the northeastern inland suggested that MVE did not spread into that area. However, the lack of positive birds may have been a reflection on the fact that these were not situated in areas of likely high virus activity. The recent review of the sentinel chickens ensured that the flocks were positioned in potential virus ‘hot spots’ and should be more sensitive in detecting the presence of the virus. What this testing may have shown was that the MVE did not spread far from these areas of high virus activity, but this requires confirmation.

The opportunistic testing of animals conducted by the NSW Agriculture Department indicated a similar pattern of MVE activity, with no positives in the northeast inland and most positives coming from the northwest. The additional seropositive animals from both Pooncarie and Wentworth (where no sentinel flocks are currently located), demonstrated that MVE had been active in those areas at some time. All positive animals were born prior to the start of the recent season and it could be possible that the infections occurred in a previous year, but as there has been no other recent evidence of MVE activity, this is unlikely. The positive Nankeen Night Herons adults are not unexpected, these are a nomadic bird which disperse in response to rainfall patterns and probably acquired the infection elsewhere, and past studies have demonstrated high levels of MVE antibodies in these birds (Marshall, 1988). The positive juvenile birds may well be due to maternally acquired antibodies, rather than becoming infected through local transmission, as maternal antibodies are known to persist for at least 15 weeks in other waterbird species (Anderson, 1954). Figure 3 illustrates those locations across southeastern Australia from where both MVE and KUN were detected.

No isolates of MVE were made from the mosquitoes trapped in the north, but unfortunately, collections were either absent or very irregular from those sites that had the chicken seroconversions. For example, there were no mosquito collections from Menindee. Fords Bridge had positive chickens in late February, but there were no mosquito collections from late December to Mid March. For Bourke, mosquito collections started after the positive chickens. For Wanaaring, there were positive chickens in mid-January, but no mosquitoes were collected from mid-December to mid-January. Likewise for the Macquarie Marshes site, where there were several missed mosquito collections over the Christmas period prior to the positive chickens. It is critical that these sites undertake routine mosquito trapping from early on in the next season, in the event that MVE has over-wintered in the local mosquito populations, as has been demonstrated in Western Australia (Broom et al. 1989, 1995).

For the southern NSW localities, MVE was not detected. Rainfall was less intense than in the north, especially during the early months of 2000. The heavy rainfall in October and November across the state did lead to a dramatic rise in mosquito numbers in the south but by the time the MVE activity was detected in the north, conditions had become hot and dry, and southern mosquito populations had declined. This combination of lower overall rainfall, with the reduced mosquito numbers probably prevented the spread of the virus to the more populated regions in the south of the state. However, there may have been other as of yet unidentified reasons for the lack of MVE spread to the south.

In response to the MVE activity, NSW Health undertook a series of health measures. This included widespread media releases with MVE Fact Sheets. A 24-hour telephone hot line was established (1800 064 400) [click here to hear this message. Note this is a 16MB wav file], which reported on the activity, advised the public on the symptoms of MVE and who to contact for further information. Weekly teleconferences were initiated between NSW Health, ICPMR and the various Public Health Units to ensure that all stakeholders were up to date on the latest virus activity. In response to advice from ICPMR, the chicken bleeding season was extended to the end of May, further flocks were established at Moree and Wee Waa, and opportunistic testing of chickens from the north eastern inland began. In the end, there were no human cases of MVE and it is impossible to know if these health measures contributed to the lack of infections. The most fortunate aspect of the MVE activity this season was that the virus failed to spread to the more populated areas in the south and north east of the inland region, and that mosquito numbers declined dramatically during the main part of the season. These were probably the defining reasons for there being no human MVE cases.

Following the detection of MVE antibodies in sentinel chickens in the 2000-2001 summer, potential ongoing MVE activity for the 2001-2002 is of major concern. It is recognised that the human population in NSW is highly susceptible to MVE (Hawkes et al., 1993) and the recent activity most likely (although unproven) resulted in little exposure and thus may have not reduced susceptibility. Both the Forbes’ and Nicholls’ hypotheses are again suggesting an MVE outbreak, emphasising the need for continued monitoring of weather patterns. In Western Australia, for example, a season of high MVE activity is often preceded by a less active season (Annette Broome, University of Western Australia pers. comm.). This must be a concern for NSW, as it is not known whether the virus has disappeared, or has remained in cryptic enzootic cycles or over-wintered in resident mosquito populations. Clearly, a high level of vigilance is required, especially towards the start of the season. It is important that all forms of surveillance are operating in the northern locations, in the event that the virus is reintroduced via viraemic migratory birds from endemic regions of the north (Marshall, 1988). The mosquito monitoring and virus isolation from the mosquitoes will be especially imperative for detecting virus that may have over-wintered. Prior to the start of the 2001-2002 season, increased public awareness will be needed to raise the profile of mosquitoes and the potential risks of mosquito borne diseases. A series of health measures should be implemented prior to the start of the upcoming season.

The early months of 2001 also saw unprecedented KUN activity. Past and ongoing human serosurveys in NSW have indicated that KUN is endemic within the state (Hawkes et al.,1993, L. Hueston et al., unpubl. data), however never before has activity been recorded to such a degree. Prior to 2001, KUN had not been detected in the sentinel chicken flocks since 1991, when there were seroconversions in four flocks (and several mosquito isolates from different locations), mostly from southern NSW. For the recent season, KUN positive birds occurred in nine of the original ten flocks, the exception being Forbes. The epidemic was widespread, intense (with some 38 seroconversions), and prolonged, with positive birds occurring from 18/Jan to 13/Mar. KUN activity was greater in the north of the state, with more seroconversions and longer activity than the south. Record levels of KUN activity were also recorded in the sentinel flocks along the Murray Valley by the Victorian Arbovirus Surveillance Program, with positive birds from Barooga down to Mildura (Wishart et al., 2001).

The opportunistic bleeding of the chickens from the north resulted in very few KUN positive birds. Again, as with the lack of MVE seropositives, this could have been due to the less than ideal locations of the opportunistic birds, rather then there being no activity in these regions. There were no positive birds at the new flocks of Moree and Wee Waa, but these flocks were implemented in early March after the majority of the KUN (and MVE) activity had already occurred. The four KUN positive sera from the opportunistic testing at Macquarie Marshes were from birds older than one year and may have been infected in a previous season. The KUN positive animals from the bleeding conducted by NSW Agriculture showed a parallelled distribution of activity to that observed by the NSW & Victorian Arbovirus Surveillance Programs.

As stated above, the poor mosquito collections from the northern localities were the main reason for the lack of viral isolates. In the south, collections were more consistent and KUN was isolated from mosquitoes trapped at Leeton, which heralded the start of the activity in the southern locations. KUN, like MVE and SIN, is thought to have birds as the main reservoir host and it is interesting to note that there were several isolates of SIN prior to the KUN activity in the south, much as occurred prior to the MVE in the north. No KUN isolate occurred from the traps located within the towns; at each location there are generally two traps, one at a potential ‘hotspot’ and the other within the township close to populated areas for detecting which mosquitoes and virus enter the town. This suggests that there may have been a low level of contact between KUN infected mosquitoes and the majority of the rural population. The low contact, along with the below average mosquito numbers from January onwards, may explain the apparent lack of human clinical KUN cases.

Despite the heavy early seasonal rainfall with the big mosquito collections, overall numbers from the inland were slightly down upon the previous year, although close to the long-term average. There were two new mosquito records from the inland for the program; Aedes imperfectus, which was trapped from the Sewage Treatment Works at Albury, and Aedes Marks sp. no. 51 from Boggabilla. Both of these species are normally collected from coastal areas, are rare and little is known about their ecology or their capability to transmit arboviruses.

In last year’s Annual Report (Doggett et al., 2000), it was noted that a mosquito similar in appearance to Culex gelidus, was collected from the Sewage Treatment Works at Condobolin in March 2000. This is an exotic species and a known vector of Japanese Encephalitis, which has become established in several parts of northern Australia (Muller et al., 2001). No specimens were trapped this season, although collections were sporadic from Condobolin. It is imperative that continual monitoring is undertaken for this important species.

Coastal trapping locations were again reduced this year, from 19 in 1999, to five in the recent season, with no locations along the south or lower north coast undertaking monitoring. Thus, the stated trends in mosquito numbers are based on a very limited data set. Generally, numbers were average up until late summer when a series of high tides coupled with rainfall led to some large collections. Conditions were again favourable for mosquitoes in April, resulting in big collections around the Port Stephens area, and the local public health unit was notified of the higher than average numbers. New mosquito records for the north coast included Aedes sp. nr. monocellatus trapped at Ballina.

The Sydney collections showed a similar trend, with mosquito numbers peaking in late summer/early autumn. The Penrith sites consistently produced some of the highest mosquito collections from the Sydney region, with the noted arbovirus vector Culex annulirostris being the dominant species. The only Sydney location that produced any viral isolate was Homebush with one RR, although it is not known if there was any local transmission or human cases. Previously, RR virus was isolated at Homebush in February 1999 from Aedes vigilax, and at that time, represented the most ’urban’ isolate of RR virus. The isolate this season was from Aedes notoscriptus, which is a domestic breeding mosquito and has yielded numerous arboviruses from the Sydney region. Until recently, it was assumed that there was little to no risk to the community of arbovirus activity within the urbanised areas of Sydney. These isolates from Homebush, as well as others (e.g. BF & EH from Ryde, STR from Parramatta, STR from Pennant Hills, EH from Pittwater, RR & STR from Penrith, etc.), demonstrate that there are pockets of activity within urban areas and that there may be a risk of arbovirus transmission to some sections of the Sydney population.

Human disease notifications (Fig 3) for 2000-2001 across NSW were slightly above average. The drop in cases from the inland was likely due to the low mosquito numbers from January onwards, which is when peak notifications normally occur.

Figure 4. Reports of human cases of arbovirus infections by month in NSW, July 1993 - June 2001.

Notifications from the mid-north coast have continued to climb over the last four seasons, from 69 in 1997-1998, to 158 in 1998-1999, to 186 in 1999-2000, and to 319 for the recent season. The mid-north coast again yielded the highest number of cases for any area health service in NSW. This region was the major contributor of notifications for the entire north coast, which had some 522 reports, around twice the average. The increase was due to a major BF epidemic, with 307 cases from the north coast over the period July 2000 to June 2001, and most (214) being reported from April to June 2001. This outbreak is now the largest BF epidemic in Australia and well surpasses the 1994-1995 south coast epidemic where there were 135 clinical cases (Sam and Crerar, 1996). The activity was spread along the north coast, but concentrated in the Kempsey region, which experienced heavy flooding following record levels of rainfall during March. As this coincided with spring tides, conditions would have been highly favourable for mosquito breeding. This combination of high tides/extreme rainfall has been a classic indicator of BF outbreaks (see Doggett et al., 1999, for a review of environmental conditions preceding past BF epidemics). No trapping occurred in this area which was unfortunate - in the 1994-1995 epidemic, BF virus was detected in the mosquitoes many weeks in advance of the epidemic peak (Doggett et al., 1999), ensuring that adequate health warnings could be put in place. A more flexible approach to trapping and virus isolation is warranted for the future, particularly when environmental conditions occur that are conducive for high mosquito populations.

Following the much above average mosquito populations from Port Stephens and Gosford, there were a series of RR cases reported from both the Hunter and Central Coast AHSs. The 126 and 39 notifications respectively from January to June 2001 were the highest totals since the outbreak in 1997.

Human arboviral infections from the south coast were again below normal, probably due to the continuing dry conditions. Rainfall patterns have been mostly low since the BF outbreak in 1994-1995, with nominal subsequent arbovirus activity, suggesting that both animal reservoirs and the human population would be becoming increasingly susceptible to infection. In the event of favourable conditions for mosquito production during the summer months, the residents of the south coast will be at risk of major activity. Re-establishment of monitoring in this region is highly advisable.

The Western Sydney region had a notable increase in reports this year, due almost solely to infection with RR virus. These cases followed closely on the unusually high mosquito populations. Several human infections were from the urban area in and around Werrington, near Penrith (Kris Hort, Wentworth PHU, pers. comm.), and in the general vicinity of an outbreak from early 1999 (Brokenshire et al., 2000). In light of this recent activity, the Wentworth Population Health Unit in collaboration with ICPMR is undertaking a serosurvey of local residents to determine the seroprevalence of RR virus.

The recent season saw many technical improvements in the laboratory aspects of the program, especially in time savings for virus identification. Seven years ago, a result would have taken four weeks with neutralisations. This was improved with ELISA technology to 2-3 weeks, and during the recent season, some of the positive RR and KUN viruses took only a little over a week to identify. Clearly, the mosquito processing system is improving and the time for identification of viruses from the mosquitoes will be shortened still further with some planned developments in 2001-2002.

Improvements in the techniques for blood collection from chickens significantly reduced the number of repeat testing of specimens, and preliminary results using filter paper methods for rapid processing of chicken samples show promise.

Back to Contents

top of page