ARCHIVED - ASSESSMENT OF SURVEILLANCE OF HUMAN WEST NILE VIRUS INFECTION IN QUEBEC, 2003

 

Organization of Human West Nile Virus Infection Surveillance in 2003

In the course of the 2003 season, surveillance of human West Nile virus (WNV) infection was undertaken throughout the province of Quebec. Physicians and laboratories were asked to report cases of WNV to their regional boards of public health.

Of the various potential clinical presentations of WNV infection, priority was given to the surveillance of cases with serious symptoms, including neurological manifestations. The regional boards of public health were expected to conduct an epidemiologic survey whenever a serious case of WNV infection was reported and subsequently confirmed through appropriate laboratory tests. In order to facilitate laboratory investigations by physicians, a provincial working group developed a case definition. This was disseminated to all regional boards of public health, which were required to convey the definition to their networks of physicians. The definition is consistent with that used throughout Canada. The surveillance period for WNV infection in humans was from 1 July to 31 October, 2003.

Clinically suspect case definition for laboratory investigation purposes

A clinically suspect case definition was disseminated to Quebec physicians to ensure that cases of neurological infection were investigated by laboratories on a high priority basis.

Clinically suspect case of serious WNV infection with neurological manifestations

History of exposure in an area where WNV activity is occurring*; OR History of exposure to a mode of transmission other than a mosquito bite**; AND fever of >= 38°C; AND AT LEAST ONE of the following neurological syndromes: encephalitis or meningoencephalitis; viral meningitis; acute flaccid paralysis***.

The case definition took into consideration the fact that the clinical presentation of neurological infection by WNV can vary considerably. WNV should be considered in the differential diagnosis of all suspected cases with muscle weakness or acute flaccid paralysis - more frequently unilateral but possibly bilateral - with or without sensory deficit. Several emerging clinical manifestations associated with neurological infection by WNV were identified in 2002. These include movement disorders (e.g. tremors, myoclonus); parkinsonism (e.g. cogwheel rigidity, bradykinesia, postural instability); peripheral neuropathy, polyradiculopathy; and acute demyelinating encephalitis. Other clinical manifestations have also been reported, including rhabdomyolysis (acute destruction of muscle cells) and ocular involvement (optic neuritis and chorioretinitis).

At the end of June 2003, HÉMA-QUÉBEC began to use a gene amplification test (RT-PCR) to track the presence of WNV in all donated blood. Blood donors with positive tests were to be reported to the regional branch for investigation. Moreover, HÉMA-QUÉBEC instituted a variety of measures to reduce the risk of WNV transmission through transfusions.

As part of ongoing monitoring and surveillance activities, physicians were urged to look for WNV in patients presenting signs of serious infection with compatible neurological involvement. Although public health officials did not encourage physicians to look for WNV in patients who were merely feverish, physicians could order a WNV test for patients with fever but no neurological involvement. All sera were analyzed at the Laboratoire de santé publique du Québec (LSPQ), and positive results were reported to the appropriate regional board of public health. The regional boards were required to conduct an epidemiologic investigation of all cases with a positive WNV result.

Case definition for public health monitoring purposes

Laboratory investigations were undertaken by clinicians whenever compatible clinical symptoms were present. Subsequently, the cases were reported and classified according to their laboratory results, including cases identified as part of blood donation procedures. Cases were defined according to the results of the laboratory test (Table 1).

Table 1
Case definition of West Nile virus (WNV) infection by laboratory tests, Quebec 2003

Case definition

Laboratory tests

Cases reported by HÉMA-QUÉBEC

RT-PCR positive blood donations

Serologically suspect case

IgM MAC-EIA initially reactive or indeterminate
or
WNV antibody (Ab) positive for the hemagglutination inhibition (HI) test >= 1/10 and < 1/320

Probable case

IgM MAC-EIA positive when additional step is carried out
or
4-fold increase in Ab HI or EIA IgG titre
or
a single titre of Ab >= 1/320 (HI)

Confirmed case

4-fold increase in Ab HI or EIA IgG titre and a positive neutralization assay (PRNT) as confirmation
or
a single titre of Ab HI >= 1/320 and a positive neutralization assay (PRNT) as confirmation
or
IgM MAC-EIA positive and positive PRNT confirmation assay
or
isolation of the virus in blood, tissue, or cerebrospinal fluid (CSF)
or
RT-PCR positive CSF or autopsy tissue


Available laboratory tests

The LSPQ used an anti-WNV IgM immunocapture technique involving an enzyme immunoassay (MAC-EIA, PANBIO) and an anti-flavivirus IgG test (EIA PANBIO). During the course of the season, an additional step was added to the MAC-EIA IgM test in order to increase its specificity. This additional step was incorporated into the case definition applied for the remainder of the season.

Hemagglutination inhibition (HI) tests were also performed as in past years. Neutralization assays (PRNT) were carried out at the National Microbiology Laboratory in Winnipeg at the start of the season and at the LSPQ as of 22 October, 2003. Gene amplification tests (RT-PCR) were conducted on cerebrospinal fluid to detect the presence of West Nile virus RNA.

Integrated health monitoring data system

Cases with positive laboratory results for WNV infection were reported to the appropriate public health branches by the LSPQ. As soon as a case was identified, it was immediately fed into a computerized system designed to integrate case data on human infections with entomologic and avian surveillance data. This integrated system enabled the real-time analysis of all available information on WNV infection. All of this information was rapidly geo-referenced using the map of Quebec; a spatial analysis of the collected data was also performed. All the regions were given access to this information from a secure site.

Assessment of Human Surveillance in 2003

Surveillance of human cases

1. Characteristics of human cases

From 1 July to 31 October, 2003, 750 serum samples from 509 individuals were sent to the LSPQ for WNV screening.

In the course of 2003, the province of Quebec recorded 17 cases of WNV infection. Of these, 14 people were given a diagnosis of serious infection with neurological manifestations, and three presented with fever. All cases were confirmed with the aid of the neutralization assay.

The average age of these cases was 58, with an age distribution from 23 to 79 years; the median was 63 years. The disease affected more men (59%) than women in 2003, and the cases were distributed over five Quebec regions (Figure 1).


Figure 1
Distribution of cases of West Nile virus infection by area of residence, province of Quebec, 2002 and 2003

Figure 1 Distribution of cases of West Nile virus infection by area of residence, province of Quebec, 2002 and 2003


We know that the severe forms of the disease generally appear in people aged >= 60. In 2002, three out of 20 people with a serious form of the disease were <= 40 years of age. In 2003, it was observed that five out of 14 with a serious form of the disease were <= 40 years of age.

Of those investigated in 2003, one person living in the Saguenay/Lac St-Jean region was infected. This region is about 500 km from the epicentre of WNV activity (Metropolitan Montreal). However, it would seem that the person in question was infected in Montérégie, a region with more intense WNV activity.

The temporal distribution of WNV cases during the 2003 season was largely the same as in 2002 (Figure 2), except that cases occurred 2 weeks later in 2003 than in the previous year.


Figure 2
Distribution of cases of West Nile virus infection according to date of onset of symptoms*, province of Quebec, 2002 and 2003

Figure 2 Distribution of cases of West Nile virus infection according to date of onset of symptoms*, province of Quebec, 2002 and 2003

*Week numbering adopted from the U.S. Centers for Disease Control and Prevention



In Figure 3, which illustrates the evolution of human, insect and avian positive cases, it can be seen that the first human cases occurred 11 weeks after the first positive bird was identified in Quebec. A similar time frame was observed in 2002.



Figure 3
Evolution of confirmed cases of West Nile virus in birds,mosquitoes, horses and humans, province of Quebec, summer 2003*

Figure 3 Evolution of confirmed cases of West Nile virus in birds,mosquitoes, horses and humans, province of Quebec, summer 2003*

*Week numbering adopted from the U.S. Centers for Disease Control and Prevention



As shown in Table 2, the number of people hospitalized was approximately the same in 2002 as in 2003. However, half of those hospitalized in 2003 spent time in intensive care as compared with slightly more than a third in 2002. The average age of people hospitalized in intensive care was slightly higher than the age of those not admitted to intensive care. No deaths occurred in 2003, whereas there were three in 2002.

Table 2
Outcome of West Nile virus cases, province of Quebec, 2002 and 2003

Outcome

2002 (20 cases)

2003 (17 cases)

Hospitalization

13

12

Intensive care

5

6

Average age of patients:

hospitalized

60 years

61 years

intensive care

69 years

66 years

Case fatality

3

0

Complete recovery

12

11

Sequelae

?

?


It is difficult to properly evaluate long-term sequelae since there is no systematic monitoring of patients who have had WNV infection. Some regional boards recontacted those who had been infected, and 65% of these are known to have made a full recovery. Others had residual fatigue, and one person was still in hospital in November 2003.

2. Risk factors

a) Medical history

Half of the people infected also had an underlying chronic condition such as diabetes, heart disease, or pulmonary disease. One person had been vaccinated against yellow fever. In 2002, three people out of 20 had reported some form of medical history (e.g. diabetes, heart disease).

b) Place infection was acquired

It is very difficult to qualify this variable since people are very mobile. Several questions had to be asked in order to specify the probable site of transmission and thereby geographically locate cases and obtain the best possible picture of WNV activity throughout Quebec.

Among those who became ill in 2003, four had remained in their area of residence during the period preceding the onset of illness, and two were, beyond any doubt, infected in another municipality. Eleven people had spent time in several municipalities, which made it difficult to determine the place where the infection was acquired. As a result, these cases were mapped according to each individual's place of residence.

c) Risk behaviours

Certain risk behaviours were identified in infected people in 2003. In particular, two had slept on their balconies on warm nights without using any kind of insect protection. Moreover, the majority of those infected (15 of 17) had not used insect repellent during outdoor activities. Half of them had no recollection of being bitten by a mosquito, and most (15 of 17) did not recall having mosquitoes in their homes.

There were no cases of infection through blood transfusions or through other modes of transmission (percutaneous, transplacental, or breast milk).

Laboratory aspects

1. Faster results

This year, the use of the IgM test greatly reduced the time required to identify cases of human WNV infection. Test results were available after 3.8 days on average, and the information was directly incorporated into the integrated surveillance system. The regional branches had rapid access to this information. Furthermore, the LSPQ called the regional branches to inform them of positive test results in their respective regions.

2. Development of the neutralization assay at the LSPQ

By the end of October 2003, the LSPQ had developed a neutralization assay to confirm probable cases of WNV infection.

Discussion

Surveillance was conducted in real time using an integrated information system. The simultaneous analysis of human, avian, and entomologic surveillance data facilitated the detection of active zones or focal areas of infection. Laboratory times were reduced by developing the capacity to perform all tests and by the use of more rapid detection tests.

By comparing Quebec data with the data found in the literature on human cases, one can observe that serious cases of infection occurred, for the most part, in people aged >= 60(1-5), although a certain number were <= 40. It will be important to establish comparisons with Canadian and U.S. data when these are available, in order to determine the severity of disease in people aged <= 40.

In 2002, the case-fatality rate among those infected with WNV was 15% (3 of 20). This rate is higher than that reported in the literature(1-3). In 2003, no deaths were recorded as a result of this disease.

WNV infection can result in major sequelae for approximately 60% of those who experience a severe form of the disease(2,4,5). In Quebec, the data collected in the course of non-systematic follow-up activities show that approximately 60% to 65% of those with a serious form of WNV infection recovered completely, a rate higher than that reported in the literature. However, the number of people infected was limited, and this may have resulted in an overestimation of the true picture. Moreover, there is no standardized mechanism to collect this information once a public health survey has been completed. As a result, the information on sequelae is, at best, fragmentary.

Since we are dealing with an emerging disease, it would be desirable to have a standardized mechanism to document sequelae at 3 months, 6 months and 1 year, particularly in those who contract a serious form of the disease.

A summary analysis of risk factors and risk behaviours failed to yield any conclusions. Other types of study will be needed in order to characterize the risk behaviours of our populations.

On the basis of a seroprevalence study conducted in Queens, New York, it is estimated that for each case of severe WNV infection 150 people are infected in the community(6,7). If this proportion were to be applied to the surveillance data gathered in Quebec, it would mean that approximately 2000 people were infected in the regions where WNV activity was reported in 2003. However, little information is available, and the laboratory data collected by the LSPQ suggest that the rate of WNV infection is, in fact, quite low in the exposed population. Nonetheless, an anti-WNV antibody study of individuals exposed at the epicentre of the probable zone of transmission should be undertaken in order to document the distribution of all human cases and to more accurately characterize the populations at risk of contracting WNV.

One of the primary difficulties associated with the surveillance of human cases is determining the area in which the infection was likely acquired. People are very mobile, particularly in summer when they vacation outside their area of residence. Except when the area in which the infection was acquired could be established beyond a doubt, cases were mapped by area of residence. It is difficult to be more specific. This lack of precision with respect to the area where infection occurred tends to complicate data analysis and decision-making with respect to control measures, such as the use of larvicides. This is why the integrated analysis of all surveillance data concerning WNV infection (be it entomologic, avian, or human) can facilitate decision-making with respect to the implementation of seasonal control and prevention measures.

Acknowledgements

The authors wish to thank all public health branch personnel who participated in the WNV surveys as well as all LSPQ personnel who carried out WNV testing.

References

  1. Nash D, Mostashari F, Fine A et al. The outbreak of West Nile virus infection in the New York City area in 1999. N Engl J Med 2001;344(24):1807-14.

  2. Pepperell C, Rau N, Krajden S et al. West Nile virus infection in 2002: morbidity and mortality among patients admitted to hospital in southcentral Ontario. Can Med Assoc J 2003;168(11): 1399-1405.

  3. Petersen LR, Marfin AA, Gubler DJ. West Nile virus. JAMA 2003;290(4):524-28.

  4. Sejvar JJ, Haddad MB, Tierney BC et al. Neurologic manifestations and outcome of West Nile virus infection. JAMA 2003; 290(4):511-15.

  5. Sejvar JJ, Leis AA, Stokic DS et al. Acute flaccid paralysis and West Nile virus infection. Emerg Infect Dis 2003;9(7):788-93.

  6. Mostashari F, Bunning ML, Kitsutani PT et al. Epidemic West Nile encephalitis. New York, 1999: results of a household-based seroepidemiological survey. Lancet 2001;358(9278):261-64.

  7. CDC. Serosurveys for West Nile virus infection - New York and Connecticut counties, 2000. MMWR 2001;50(3):37-9.

Source: C Gaulin, MD, Direction génerale de la protection, ministère de la Santé et des Services sociaux (MSSS), Quebec City, Quebec; M Couillard, PhD, Laboratoire de santé publique du Québec, Institut national de santé publique du Québec; PA Pilon, MD, M Tremblay, MD, Direction de santé publique de Montréal- Centre; L Lambert, MD, Direction de santé publique de la Montérégie; M Douville Fradet, MD, Institut national de santé publique du Québec; L Deschênes, MD, Service de microbiologie, Quebec City, Quebec; A Fortin, MD, C Poulin, MSc, Secrétariat du sang, MSSS.


Footnotes

* Infection acquired in Quebec, Canada or abroad.

**Other modes of transmission identified to date include blood transfusion, organ transplantation or the percutaneous route. Transmission through breast milk and transplacental transmission are also possible.

*** A person with West Nile virus-associated acute flaccid paralysis may present with or without fever or mental status changes. Altered mental status can range from confusion to coma, with or without additional signs of brain dysfunction.


Report a problem or mistake on this page
Please select all that apply:

Thank you for your help!

You will not receive a reply. For enquiries, contact us.

Date modified: