 |
NSW Arbovirus Surveillance & Vector Monitoring Program |  |
|
NSW Arbovirus Surveillance & Vector Monitoring Program | |
 1998 - 1999
Annual Report  |
The drought
conditions that had dominated NSW through the 1990's ended with extremely heavy rainfall
across much of the state during the winter and spring months of 1998. This abundant
rainfall was reflected in the mosquito numbers and arbovirus activity in the mosquitoes;
the total of 277,197 mosquitoes trapped was the highest ever collected during one season
for the Program, and 126 virus isolates was the second highest number detected.
The heavy rains
in inland regions resulted in some extraordinary collections. Over 28,000 mosquitoes were
trapped over two consecutive weeks from one site at Leeton and Cx.
annulirostris numbers were 34 times the average. One trap from Griffith had over 10,000
mosquitoes but, more significantly, Cx. annulirostris numbers in the first week of
November were 10 times higher than the average. From this week, there was one isolate of RR, and never before had a virus been
isolated during the first week of the Program. Over the subsequent two weeks there were
another 3 RR isolates, 1 from Menindee and 2 from Griffith. Eventually, there were
104 viruses isolated from inland locations, 101 of these were from the south, and of
these, 58 were from Leeton and 40
from Griffith. This is the
most number of isolates recorded from the inland for the history of the Program. As the
summer was dry in inland regions, the large mosquito abundances declined through February
and were around normal. By March, numbers were well down. The heavy rains during this
month did not affect the mosquito populations, as by this time the temperatures were not
conducive for high production of mosquitoes.
Culex
annulirostris dominated
collections from the inland regions, comprising around 80% of all mosquitoes trapped. This
species also yielded the majority (94) of isolates. Most of the remaining 20% of
mosquitoes were An. annulipes, and the rest of the isolates
were from this species.
The early and
extensive mosquito and virus activity in southern inland NSW (as documented by the
Program) was associated with higher than normal numbers of human infections, principally RR. Within the Greater Murray AHS, there
was a record 77 and 81 laboratory notifications from November and December, respectively,
and this was around 50% of all notifications for the state during these months.
Additionally, Macquarie, Mid Western and the Far West AHSs had very high notification
rates for the same period. The dry summer, which saw the decline in mosquito numbers, also
halted the rise in notifications in the inland, with February figures comparable to the
previous two months.
In addition to
the unusual RR activity in southern
inland NSW, the 1998-99 season had a re-emergence of Kunjin (KUN) virus. Kunjin is a flavivirus, which
has been reported as causing encephalitis (Marshall, 1988). The virus had not been seen
since 1990-91 when there were mosquito isolates, followed by chicken seroconversions and
eventually human cases (Russell et al., 1992).
This year there were three isolates from mosquitoes, one from Leeton and two from Griffith. In response to this
year's KUN isolate from Leeton in January, public health
warnings were intensified (McAnulty, 1999), an extra chicken flock was added to Leeton and the chicken surveillance
extended for an extra month, however there were no chicken seroconversions.
Sindbis (SIN) activity has also not occurred for
several years in mosquitoes. The 68 isolates for this season is an historical high for the
program and it has been five seasons since there has been any major activity. The reason
for the recent low activity of SIN is unknown,
although the large mosquito numbers for this season may have allowed the virus to amplify
and spread. There were three human SIN
notifications, one each from the Illawarra AHS in September 1998, the South Eastern Sydney
AHS in October 1998 and the Mid Western AHS in April 1999.
As noted above,
both the Forbes' and Nicholls' models are suggesting possible MVE activity during the next
mosquito season [this is no longer the case for Nicholls' model - SD]. An additional
concern is that there was extensive MVE activity in
northern Australia during the first five months of 1999. In Western Australia, there have
been substantial numbers of chicken seroconversions and the most extensive MVE activity since 1974, the year of the last major outbreak, although no
human cases (Annette Broom, Arbovirus Laboratory, University of WA, personal communication). The Northern Territory
also had several chicken seroconversions, and health protection warnings of the dangers of
MVE were released
to the public.
Evidence
would suggest that MVE is not endemic to
southeastern Australia, rather that it is introduced via viraemic migratory birds from
endemic regions of the north (Marshall, 1988). Naturally, with this year's high MVE
activity in the north, there is concern that MVE
may again be introduced to southeastern Australia in the summer of 1999-2000, particularly
if rainfall patterns continue to be wetter than normal. The environmental conditions will
be monitored over the upcoming months, and appear on the NSW Arbovirus Surveillance web
site.
In contrast to
the inland regions, weather patterns and mosquito abundances were quite varied along the
coastal regions. Generally, mosquito numbers on the north coast were below average during
November to January, however some very localised and intense rainfall led to some large
numbers in February and March, and there were several isolates of RR from mosquitoes collected during this
time. Of note was the unusually high abundance of freshwater species such as Cx.
annulirostris, Cx. orbostiensis and Coquillettidia spp. and, atypically, more isolates
were obtained from freshwater than saltwater species. The south coast had moderately low
rainfall and mosquito numbers were correspondingly well below average.
Peak
notifications of human arbovirus infections on the coast also contrasted strongly with the
inland. Along the coast, notifications peaked in March-April and followed the rise in
mosquito numbers. Over 70% of all arbovirus infections from NSW were reported from the
coast during this period. Although RR
was responsible for the vast majority of arbovirus infections, BF was also common this year with the
second highest numbers of cases ever reported for any season. The mid-north coast and far
north coast of the state had the highest numbers of BF cases yet reported with 130
notifications. Despite the south coast being relatively dry and having below average
mosquito numbers, two RR were isolated
from mosquitoes and arbovirus notifications in humans were quite high with 154 RR and 62 BF infections. This was the highest number
of BF notifications on the south coast
since the 1994-95 epidemic when there were 135 reports (Sam and Crerar, 1996).
While the
numbers of arbovirus notifications did not peak as high as they have during the summer
months in previous seasons, the overall number of notifications (around 1,450) during
1998-99 was the highest ever recorded for NSW. This was due to the wide time spread of
infections during this season, with many notifications prior to January. A comparison of
notifications across the state with recent years is provided in the accompanying graph.
The Sydney
sites trapped few mosquitoes, however a relatively high number of isolates were made, with
two RR and four Stratford. There were human cases of RR and a horse seroconversion to RR in Western Sydney. The majority of the
mosquito isolates from Sydney were from Ae. notoscriptus, which also yielded two RR isolates from other coastal localities. Aedes
notoscriptus is an important domestic mosquito and has been shown capable of
transmitting both RR and BF (Doggett and Russell, 1997). This
species would now appear to be a potentially important vector of arboviruses and must be
considered in any urban mosquito control program.
