|
|
||||||||
| ABSTRACT |
|
|
|---|
| INTRODUCTION |
|
|
|---|
| METHODS |
|
|
|---|
We included confirmed and probable cases in this analysis.5 According to the national case definitions, a confirmed case has laboratory evidence of recent arboviral infection based on one or more of the following tests performed on serum and/or cerebrospinal fluid: 1) virus isolation, 2) nucleic acid amplification test, 3) virus–specific immunoglobulin (Ig) M antibody by enzyme immunoassay (EIA) confirmed by virus–specific IgG antibody by EIA or neutralization test, or 4) a 4-fold or greater change in virus–specific serum antibody. A probable case has laboratory evidence of disease based on: 1) virus–specific immunoglobulin IgM antibody by EIA without IgG confirmation, or 2) stable but elevated virus–specific serum antibody.
Data collection From 1990–1995, state health departments reported cases of "arboviral encephalitis" to the Nationally Notifiable Diseases Surveillance System (NNDSS). The CDC Division of Vector-Borne Infectious Diseases subsequently obtained information regarding the specific arboviral etiologies for these cases. However, no additional demographic in formation was reported to CDC. From 1996–2002, state health departments reported cases of neuroinvasive disease resulting from SLEV, EEEV, WEEV, CAL serogroup viruses, and POWV to NNDSS; data reported included age, sex, race, ethnicity, county and state of residence, and date of symptom onset. In 2000, in response to the detection of WNV in the United States, CDC and state health departments developed ArboNET, a real-time internet-based surveillance system.8,9 From 2000–2002, only WNV cases were reported to ArboNET; starting in 2003, other arboviral diseases were also reported. Variables routinely collected in ArboNET for human arboviral disease cases include age, sex, race, ethnicity, county and state of residence, date of symptom onset, and outcome (died or survived). Cases are classified by clinical syndrome as neuroinvasive (i.e., meningitis, encephalitis, or acute flaccid paralysis) or non-neuroinvasive (i.e., uncomplicated fever or other/unknown) disease. For this analysis, data were drawn from NNDSS reports for 1990–2002 and from ArboNET for 2003–2007.
Data analysis We described the reported number of cases and incidence of neuroinvasive domestic arboviral disease in the United States from 1999–2007. Reported incidence was calculated as cases per 100,000 persons per year using annual population estimates from the U.S. Census Bureau.10 For the most common causes of neuroinvasive arboviral disease, additional epidemiologic characteristics were evaluated, including age, sex, race, month of onset, and county of residence. For 2003–2007, fatal cases were also enumerated and case fatality rates calculated. Finally, we compared the incidence of neuroinvasive disease resulting from CAL serogroup viruses, SLEV, and EEEV before (1990–1998) and after (1999–2007) the detection of WNV in the United States. Categorical variables were described using proportions and rates; continuous variables were described by mean and range. Geographic Information System software (ESRI ArcMap version 9.0, Redlands, CA) was used to map the reported incidence of CAL serogroup virus, SLEV, and EEEV neuroinvasive disease by county.
| RESULTS |
|
|
|---|
|
|
|
|
40 years of age (0.013 per 100,000), and 75% of all cases occurred in this age group (Table 2
The average annual reported incidence of neuroinvasive SLEV disease from 1999–2007 was 0.007 cases per 100,000 population per year (Table 2
). This incidence was approximately one-third of the average annual incidence from 1990–1998 (0.019 per 100,000 per year) (Figure 2
). However, the higher incidence of SLEV disease during the pre-WNV period was primarily because of a large outbreak that occurred in 1990 (0.099 per 100,000 in 1990 compared with 0.009 per 100,000 per year for 1991–1998). Between 1999 and 2007, the incidence of neuroinvasive SLEV disease peaked at 0.027 per 100,000 in 2001, and then declined to an average low of 0.001 per 100,000 for 2005–2007.
Eastern equine encephalitis virus
From 1999–2007, 80 EEEV disease cases were reported in the United States from 13 states. In the northeast, cases occurred in Massachusetts (N = 15) and New Hampshire (N = 9), including 11 cases that occurred during an outbreak in 2005.13 In the southeast, cases most often occurred in Florida (N = 15), Georgia (N = 6), Louisiana (N = 6), South Carolina (N = 6), North Carolina (N = 5), and Alabama (N = 5) (Figure 1
). The highest cumulative incidence occurred in counties along the Atlantic and Gulf coasts. Average annual incidence was similar among all age groups and among whites and blacks (Table 2
). However, incidence was two times higher among males than among females; 63 (79%) cases occurred from July through September. Among the 53 cases reported between 2003 and 2007, 22 (42%) were fatal.
The average annual incidence of neuroinvasive EEEV disease from 1999–2007 was 0.003 cases per 100,000 population per year (Table 2
). This incidence was similar to the average annual rate from 1990–1998 (0.002 per 100,000 per year), prior to the detection of WNV in the United States (Figure 2
). Between 1999 and 2007, the incidence of neuroinvasive EEEV disease remained relatively stable from year to year with the exception of 2005 when it peaked at 0.007 per 100,000 because of the outbreaks in the northeast.
| DISCUSSION |
|
|
|---|
Seasonality, patient demographics, and case fatality rates were also similar to previous reports for each of the most common arboviruses.2,18–25 The vast majority of all neuroinvasive domestic arboviral disease occurred from July through September, corresponding to the peak feeding times of the vectors, and emphasizing the importance of targeting public health interventions during these months. The CAL sero-group virus disease occurred primarily among children, whereas SLEV disease occurred mostly among adults
40 years of age. These demographic patterns may reflect the relative abundance and likelihood of exposure to the vectors and the virus, differences in viral pathogenesis and virulence, or host-related susceptibility that varies with age.3,4,14,15 The EEEV disease remained the most severe of the domestic arboviral diseases, creating a public health challenge of appropriately targeting and prioritizing prevention efforts for this relatively rare and sporadic but devastating disease.13,17
An unanticipated finding in our analysis was that the race–specific incidence for CAL serogroup virus disease was higher among whites compared with blacks. Although this difference has not been previously reported, race data were not collected for arboviral diseases in the United States until 1996. The most likely explanation for this finding is that the racial demographics of the population vary in the geographic areas where these diseases occur. Indeed, for SLEV disease, although comparison on the basis of the national racial denominators showed a difference in incidence, there was a minimal difference in incidence by race using denominators for counties where the disease actually occurred. However, the higher incidence of CAL serogroup virus disease among whites persisted at a county level. The CAL serogroup virus disease cases occurred in a limited number of areas within the affected counties, and the remaining difference may reflect the racial distribution of the local populations. The relatively high proportion of CAL serogroup virus cases with unknown race also could have impacted this finding. Further evaluation is warranted.
During the period from 1999–2007, the incidence of both CAL serogroup virus and SLEV disease peaked between 2001 and 2003, and subsequently declined from 2004–2007. Although this finding likely represents normal fluctuations in the incidence of these diseases, several other factors may have contributed to these trends. The case definition of arboviral neuroinvasive disease changed several times during the time period of this analysis. The most significant changes occurred in 2001 when "arboviral meningitis" was added to the nationally notifiable diseases case definition, and in 2003 when non-neuroinvasive arboviral diseases became nationally notifiable and all domestic arboviral diseases were first reported to ArboNET. These changes, along with increased awareness of WNV among physicians, patients, and public health officials, may have led to an increase in testing and reporting of all arboviral diseases. In response to the spread of WNV, between 2000 and 2003, diagnostic testing capacity for arboviruses was markedly enhanced at state health laboratories likely leading to improved ability to detect disease resulting from CAL serogroup viruses, SLEV, and WNV at a state level. In more recent years, WNV serologic assays have been frequently performed at commercial laboratories. Although this has improved the availability of WNV testing, commercial laboratories may not routinely perform tests for other arboviral diseases, resulting in fewer alternative diagnoses on WNV-negative specimens. Furthermore, as WNV spread across the United States, increased funding for prevention and control became available at the state and local levels. Mosquito control programs and public education campaigns regarding use of mosquito repellent were implemented in areas affected by WNV disease. These prevention programs may have led to a decrease in overall incidence of arboviral neuroinvasive disease, which was seen beginning in 2003.
Despite fluctuations in the number of annual cases, the average annual incidence of CAL serogroup virus and EEEV neuroinvasive disease remained stable before (1990–1998) and after (1999–2007) the detection of WNV in the United States. By contrast, the incidence of SLEV disease declined 3-fold between the pre- and post-WNV eras. Although WNV and SLEV share many of the same vectors and animal hosts, the highest incidence of WNV neuroinvasive disease has occurred in the Great Plains and Mountain states,26 areas with relatively few cases of SLEV disease since 1999. The most likely explanation for the apparent decrease in SLEV disease is simply one of timing; SLEV disease historically has occurred in sporadic epidemics every 10 to 15 years, and not enough time has yet elapsed to see the first expected large SLEV epidemic in the post-WNV era. The higher incidence of SLEV disease during the pre-WNV period (1990–1998) was primarily the result of a large outbreak that occurred in 1990. Some Midwest and western states experienced explosive outbreaks in 1975 and 1976, and this could presumably happen again. However, some researchers have postulated that the introduction of WNV may result in the competitive displacement of SLEV in parts of the United States.27 There is evidence that birds infected with WNV develop immunity to both WNV and SLEV, but birds that have been infected with SLEV can still become viremic when infected with WNV.28 The rapid spread of WNV may result in decreased SLEV viremia among birds, and therefore decreased SLEV transmission to humans. Given the sporadic occurrence of SLEV epidemics, it may be some time before we can confidently distinguish actual changes in disease patterns from natural fluctuations in incidence.
There are several limitations to this analysis. The NNDSS and ArboNET are passive surveillance systems that depend on clinicians to consider the diagnosis of an arboviral disease, obtain the appropriate diagnostic test, and report any positive results. Therefore, these reports likely underestimate the true burden of domestic neuroinvasive arboviral diseases. Furthermore, physician awareness, testing capacity, and reporting may vary by jurisdiction, and such reporting bias may distort surveillance data so that it does not reflect the true national or regional incidence of disease. Moreover, although the incidence of neuroinvasive disease and the case-fatality rates have been discussed, this analysis does not address the significant morbidity resulting from these viruses. Finally, although arboviral IgM enzyme immunoassays are relatively specific, cross-reaction does occur among viruses within the same family. Neutralizing antibodies can be used to confirm the diagnosis and differentiate potential cross-reactivity between related viruses. However, ArboNET does not collect information regarding the specific laboratory methods used to confirm each case. Therefore, we cannot estimate what proportion of SLEV disease cases had confirmatory testing to further differentiate them from WNV disease.
In addition to WNV disease, CAL serogroup virus, SLEV, and EEEV arboviral disease are persistent public health concerns in the United States warranting continued prevention efforts to promote control of vector mosquitoes and personal protection against mosquito bites. On the basis of the epidemiology described here and in previous reports, efforts to prevent CAL serogroup virus disease should be targeted to children living in Appalachia and the upper Midwest, public health education for SLEV disease should be primarily directed to older adults in southern states, and EEEV disease prevention should focus on susceptible communities along the Atlantic and Gulf coasts.
Received May 16, 2008. Accepted for publication July 27, 2008.
Acknowledgments: The authors thank Peggy Collins, Erin Staples, and all the state and local health department staff who perform surveillance for arboviral diseases for their contributions to this manuscript.
Financial support: This work was funded by the Centers for Disease Control and Prevention.
Disclosure: Carolyn Reimann is a Lieutenant Commander in the United States Navy. The views expressed in this article are those of the authors and do not necessarily reflect the official policy or views of the U.S. Department of the Navy, U.S. Department of Defense, or U.S. Department of Health and Human Services.
* Address correspondence to Marc Fischer, Arboviral Diseases Branch, CDC, 3150 Rampart Road, Fort Collins, CO 80521. E-mail: mfischer{at}cdc.gov ![]()
Authors addresses: Carolyn A. Reimann, Carolyn DiGuiseppi, and Richard Hoffman, Department of Epidemiology, Colorado School of Public Health, 4200 East Ninth Avenue, B-119, Denver, CO 80262, Tel: 303-315-7605, Fax: 303-315-1010, E-mail: carolyn.diguiseppi{at}uchsc.edu. Edward B. Hayes, Jennifer A. Lehman, Nicole P. Lindsey, Grant L. Campbell, and Marc Fischer, Arboviral Diseases Branch, Division of Vector-Borne Infectious Diseases, Centers for Disease Control and Prevention, 3150 Rampart Road, Fort Collins, CO 80521, Tel: 970-221-6400, Fax: 970-266-3568, E-mail: mfischer{at}cdc.gov.
| REFERENCES |
|
|
|---|
This article has been cited by other articles:
![]() |
B. K. Schweitzer, N. M. Chapman, and P. C. Iwen Overview of the Flaviviridae With an Emphasis on the Japanese Encephalitis Group Viruses Lab Med, August 1, 2009; 40(8): 493 - 499. [Abstract] [Full Text] [PDF] |
||||
![]() |
N. P. Lindsey, E. B. Hayes, J. E. Staples, and M. Fischer West Nile Virus Disease in Children, United States, 1999-2007 Pediatrics, June 1, 2009; 123(6): e1084 - e1089. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |