|
Summary of West Nile Virus in
the United States, 1999
United States
Department of Agriculture Animal and Plant Health
Inspection Service
West Nile virus (WNV) was identified in a limited
area of the northeastern United States in wild
birds, mosquitoes, humans, and horses in 1999. The
affected area consisted of parts of Connecticut, New
York, New Jersey, and one county in Maryland.
The map in Figure 1 indicates the areas where
evidence of the virus was found. Clinical illness in
humans and horses occurred
during a period from early August through late
October (Figure 2). WNV activity in the United
States has ended because of various factors,
including climate and vector control activities.
The scientific literature about WNV indicates that
transmission is primarily through a mosquito-bird
cycle, with occasional incursions into other
vertebrates as terminal hosts only. The literature
supports the conclusion that horses are terminal
hosts for WNV and do not maintain a sufficient
viremia to infect either other mammals (including
humans) or mosquitoes.
An outbreak of human encephalitis of then unknown
etiology began in New York City in early August
1999. On September 14, 1999, a virus was isolated at
the National Veterinary Services Laboratories (NVSL)
in Ames, Iowa, from tissues of a crow from the New
York City area. This virus was later identified as
WNV and confirmed as the cause of the human
encephalitis outbreak. WNV was also isolated
from central nervous system tissue of a horse on
Long Island, New York, that exhibited neurologic
signs; the isolate was confirmed to be WNV by the
Centers for Disease Control and Prevention (CDC) on
October 18, 1999. A total of 25 cases of WNV
have been identified in horses, all in Suffolk and
Nassau Counties on Long Island, New York.
Because horses are not known to play a role in
transmission of WNV, quarantines were never placed
on any non-clinically ill horses in the outbreak
area.
Surveillance has continued in the affected area and
in additional States on the eastern coast of the
United States to monitor the possible spread of WNV.
This surveillance consists of investigating suspect
cases in horses and testing wild birds, along with
the seasonal collection and testing of mosquitoes
and the testing of sentinel chickens. No spread of
WNV activity beyond Connecticut, New York, New
Jersey, and Maryland has been detected. No cases of
WNV infection have ever been
found in any commercial poultry in the United
States.
The deaths of crows and other wild birds coincided
with cases of illness in humans and horses.
Anecdotal reports suggest that significant wild bird
mortality very likely preceded cases of human and
equine illness. Based on the high mortality of crows
in the United States, it appears that corvids are
highly susceptible to WNV. The lack of unusual
mortality in crows or other corvids in areas outside
of the known affected States, therefore, is a good
indicator that WNV activity is unlikely to be
present in other regions of the country.
To obtain additional information about the virus,
inoculation studies were carried out in a small
number of horses, chickens, and turkeys. In each of
the species, WNV was reisolated from the blood after
inoculation. Chickens showed amplification of the
virus, up to 5 logs (105) per milliliter of plasma
in one bird, while turkeys had a maximum viremia of
4 logs. Titers in inoculated horses never exceeded
2.5 logs of virus; this is consistent with previous
findings in the scientific literature
that indicate horses are not involved in the
transmission cycle of WNV. Each species developed
detectable antibodies after being inoculated.
In genetic sequencing studies, WNV isolates from the
New York outbreak showed strong similarities to
isolates from the Middle East, suggesting that this
region may be the origin of the WNV that caused the
U.S. outbreak. How WNV was introduced into the
United States is unknown, but speculation has
centered on infected humans, mosquitoes, or birds
being transported
by aircraft (see Lancet, Research Letters, Volume
354, Number 9194, 4 December 1999).
WNV activity in the United States has ceased because
of various factors including climate and previous
vector control activities. The information below
includes specifics on the outbreak, on surveillance
and vector control, and on evidence that virus
activity has ended.
Humans
62 cases, including 7 deaths (all in New York)
First human onset - August 2, 1999
Last human onset - September 22, 1999
[Dates given are for clinical onset.]
Enhanced surveillance for human encephalitis cases
in Connecticut and New Jersey did not detect any WNV
cases.
Horses
25 clinical cases (all in New York: 22 in Suffolk
County, 3 in Nassau County)
First horse onset - August 26, 1999
Last horse onset - October 18, 1999
[Dates given are for clinical onset, an earlier
point in the course of disease progression than the
dates given for wild birds (see below), for which
only a date of collection is known.]
Twenty-five cases of WNV were identified in horses
from New York: 22 from Suffolk County (eastern Long
Island) and 3 from Nassau County (western Long
Island), which is closer to New York City, where
humans, mosquitoes, and wild birds tested positive
for WNV. Virus was isolated from tissues of 3 of the
22 Suffolk County cases. The genders and breeds of
horses infected with WNV appear to reflect the
general population of horses on Long Island. Females
(mares and fillies) and males (stallions and
geldings) of several breeds were infected; most were
quarterhorses and standardbreds.
Initial Investigation of Riverhead Area Cases
Beginning on August 26, 1999, a private veterinary
practitioner observed multiple horses with
neurologic signs in eastern Long Island, which
prompted investigations to determine the cause. At
the request of the New York State Department of
Agriculture and Markets, USDA sent an Early Response
Team (ERT) to assist in investigating the cases.
An epidemiological investigation was conducted on
the cluster of equine cases, all of which were
within a 5-mile radius of a site near the city of
Riverhead in Suffolk County, New York. Of the
eventual total of 25 cases found on Long Island, 20
were identified in the initial investigation on a
total of 15 premises in this Riverhead cluster. All
affected premises shared a similar ecological
environment, with standing water in large pools,
barrels, or watering areas.
The map in Figure 3 documents the locations of the
initially identified premises near Riverhead and
indicates the number of serologically positive
horses on each premises. The map shows that the
positive horses were clustered around an epicenter,
with apparently decreasing levels of exposure
farther out.
Samples were obtained from 83 of the 146 horses that
resided on the affected premises at the time of the
investigation. Samples were taken from both
convalescing horses and contact horses on these
premises. Efforts were made to obtain samples from
contact horses that either were commingled with a
case or were on a premises that was co-owned or
managed by an owner
of a case animal at the time the animal was showing
clinical signs.
Of the 83 horses sampled in the Riverhead area, 15
had a recent clinical illness compatible with WNV
infection and were also serologically positive for
WNV antibodies. The other 68 horses did not have any
recent history of clinical illness, but 21 of those
68 (31 percent) were serologically positive for WNV
antibodies.
Samples were not available from five horses in the
Riverhead area that had clinical signs compatible
with WNV and died or were euthanatized early in the
outbreak. Based on their clinical signs and their
proximity to other WNV-positive horses, these horses
were considered to be WNV cases.
The age was known for 19 of the 20 clinically ill
horses. Those ages ranged from 3 to 30 years, with a
mean of 15.6 and a median of 17.0. Of those 19 ill
horses, the ages for the 8 that died or were
euthanatized [range = 4-30; mean = 16.1; median =
16.5] were similar to the ages for the 11 that
recovered [range = 3-28; mean = 15.3; median =
17.0].
Additional Suffolk County Cases
In addition to the 20 cases initially investigated,
2 more horses later were found to have been
clinically affected by WNV in Suffolk County. One
was a 19-year-old that was ill in early September on
another premises located in the Riverhead area.
In addition to the ill horse, nine other horses on
that premises were found to be serologically
positive for WNV antibodies. The second case
found after the initial investigation in Riverhead
was a 20-year-old horse that became ill in early
October
on a premises near the town of Centereach in
west-central Suffolk County. A summary of the ages
of all known WNV-exposed horses in Suffolk County is
given in Table 1.
Table 1.
Ages of Horses with Known Exposure
to WNV,
Suffolk County, New York, 1999
Status of Horses Range (years) Mean (years) Median
(years)
Clinically ill (n = 21*) 3-30 15.5 17.0
Died/euthanatized (n = 8) 4-30 16.1 16.5
Recovered (n = 13) 3-28 15.1 17.0
Seropositive only (n = 31**) 1-30 11.2 9.0
* Age known for only 21 of 22 ill horses.
** Age known for only 31 of 36 seropositive horses.
Nassau County Cases
The three WNV-affected horses in Nassau County were
2, 2, and 23 years old, with a mean of 9.0 years.
Two of the three WNV cases in Nassau County were
horses stabled at Belmont Park race track. Both of
these horses had clinical illness compatible with
WNV infection, and both were found to have high WNV
antibody titers. Both horses have recovered and no
other clinically ill horses have been reported from
Belmont Park. The third case in Nassau County, a
23-year-old horse that exhibited neurological signs
consistent with WNV, recovered within a few days and
remains healthy. The horse was stabled in an area
near the north shore of Nassau County, close to
areas with WNV-positive human cases and positive
mosquito pools.
WNV Attack Rates in Horses
Attack rates of WNV, per 10,000 horses, are shown in
Table 2 for each of the four States in which at
least one WNV-positive wild bird was found. Table 3
shows attack rates for counties in the State of New
York where at least one WNV-positive wild bird was
found. The horse population data for Table 3 is from
the last census for which data is currently
available, but the population is not expected to
have changed significantly. Even in Suffolk County,
New York, which had 88 percent of the equine cases,
less than 0.5 percent of all horses had a clinical
illness caused by WNV.
Table 2. Attack Rates of WNV in Horses by State,
1999
States with WNV Activity Horse Population as of
January 1, 1999* Cases of
WNV in Horses Attack Rate
(per 10,000)
Connecticut 26,000 0 0.0
Maryland 45,000 0 0.0
New Jersey 45,000 0 0.0
New York 155,000 25 1.6
Total 271,000 25 0.9
* Includes horses, ponies, mules, burros, and
donkeys for farm and non-farm premises.
Table 3. Attack Rates of WNV in Horses by County,
New York, 1999
Counties with WNV Activity Horse Population as of 31
December 1988* Cases of
WNV in Horses Attack Rate
(per 10,000)
Columbia 4,400 0 0.0
Nassau 4,800 3 6.3
New York City** 3,200 0 0.0
Rockland 1,000 0 0.0
Suffolk 5,200 22 42.3
Ulster 4,900 0 0.0
Westchester 8,500 0 0.0
Total 32,000 25 7.8
* Includes horses, ponies, mules, burros, and
donkeys for farm and non-farm premises.
** Includes 5 counties: Bronx, Kings, New York,
Queens, Richmond.
Wild Birds
First positive wild bird - August 8, 1999 (New York)
Last positive wild bird - November 5, 1999 (New
York)
[Dates given are for wild bird collection, i.e.,
date of death or later.]
This surveillance has focused on American crows and
other wild birds. As of December 31, 1999, more than
550 dead birds had been tested for WNV. Of those,
194 birds were positive; confirmatory testing of
positives was done at CDC. The following orders of
native wild birds were represented among the WNV-positive
cases: Anseriformes (mallard), Falconiformes (eagle,
hawks, and kestrel), Ciconiiformes (black-crowned
night heron), Gruiformes (sandhill crane),
Charadriiformes (gulls),
Columbiformes (rock dove), Coraciiformes
(kingfisher), and Passeriformes (crows, jays, and
robin). Surveillance in wild birds will continue
through the fall of 2000 in at least 20 eastern and
southern States. No commercial poultry have been
affected by WNV.
The last positive wild bird associated with the 1999
outbreak was found on November 5; however, a
red-tailed hawk found dead in Westchester County,
NY, on February 6, 2000, was reported as positive
for WNV by the Connecticut Agricultural Experiment
Station. This finding is still being evaluated to
determine the most likely date of infection. One
possibility is that the hawk had been persistently
infected since the previous mosquito season, as
Westchester County is known to have had WNV
activity, including human cases, in 1999. In any
case, the discovery reinforces the need to continue
surveillance for WNV.
Mosquitoes
Last positive mosquito pool - October 14, 1999 (New
York)
Mosquitoes were collected from light and gravid
traps in various locations. Pools of mosquitoes from
each location were then tested for WNV. The
locations and positive results are documented in the
map in Figure 4. The mosquitoes most often found
positive for WNV were Culex spp., including Culex
pipiens. Culex pipiens is a species that prefers
breeding sites generally associated with urban
settings.
With the onset of freezing weather in the WNV-affected
areas, routine mosquito collection stopped because
of the low numbers of insects being collected. After
the last positive collection on October 14, 1999,
additional pools were collected for 2 more weeks;
those pools were negative for WNV.
More recently, overwintering Culex mosquitoes were
collected to determine whether WNV might persist
throughout the winter. Mosquitoes were
collected during January-February 2000 from various
structures around New York City. WNV was isolated
from one of 67 mosquito pools containing Culex spp.
The positive pool was from Queens County where WNV-positive
mosquitoes and human cases were found during the
1999 outbreak. This finding indicates that live
virus is present in overwintering mosquitoes in New
York City. It emphasizes the need to maintain
surveillance for WNV and to implement control of
larval mosquitoes before adults emerge. More
information can be found in the March 10 and March
17 issues of CDC's Morbidity and Mortality
Weekly Report (http://www2.cdc.gov/mmwr/weekcvol.html).
Vector Control
The State Departments of Public Health in New York,
New Jersey, and Connecticut conducted intensive
mosquito control activities beginning in the latter
part of August 1999 because of suspect cases of St.
Louis encephalitis. These efforts intensified once
WNV was confirmed as the actual etiologic agent.
Mosquito control activities included area-wide
spraying for control of adult mosquitoes and source
reduction, i.e., eliminating mosquito breeding
grounds. Spraying for adults
continued until the first frost or until
surveillance indicated that adults were no longer
active.
Climate/Weather Factors
The map in Figure 5 documents the lowest
temperatures observed in the relevant States through
December 12, 1999. All locations known to have had
WNV activity had a hard freeze (temperatures below
-3C) and, for all practical purposes, mosquito
activity had ended by that date. With the cold
temperatures seen in WNV-affected areas, and the
concomitant reduction of mosquito activity,
pesticide spraying activities ended.
Summary
An outbreak of West Nile virus occurred in a limited
area of the northeastern United States from August
through October of 1999. The affected area consisted
of parts of Connecticut, New York, New Jersey, and
one county in Maryland. WNV was identified in wild
birds, mosquitoes, humans, and horses. WNV activity
in the United States has ceased because of various
factors, including climate and vector control
activities.
~top
|
|
|
|
