Feature Articles

Published: April 1, 2009
Find more content on:
Diagnostic approaches to managing West Nile virus

In diagnosing WNV, rapid strip-based lateral-flow tests offer results comparable to the more prevalent ELISA-based devices.

By: Nisar A. Shaikh








Figure 1. (click to enlarge) West Nile virus transmission cycle. (Taken from the Public Health Agency of Canada's Web site and reproduced with the permission of the Minister of Public Works and Government Services.)
Ever since West Nile virus (WNV)-related cases of encephalitis were first reported in New York City in 1999, which resulted in seven deaths, the virus has become endemic to North America. Clinical management of WNV relies on timely and accurate diagnostic test results. The problem is that WNV infection is difficult to diagnose since the initial symptoms resemble the common flu, and in many patients, it can resolve itself without any complications. However, some WNV patients can develop serious neuroinvasive complications such as encephalitis and meningitis, which are similar to symptoms manifested by other viruses in the Flaviviridae family.


This article describes clinical aspects of WNV and some of the approaches that hospitals and public health laboratories are currently using to diagnose the infection. This article also includes a brief description of the enzyme-linked immunosorbent assay (ELISA)-based devices that are commonly employed and a summary of their clinical efficacy data. While a detailed description of their principle performance characteristics, cross reactivity data, and limitations is beyond the scope of this article, such information can be found in their respective package inserts. A brief study describing the comparative data of some of these devices is also available.1


WNV Transmission


WNV is an arbovirus within the Japanese encephalitis antigenic complex. The natural transmission cycle for WNV is enzootic. This cycle involves several species of mosquitoes as the vectors and different types of birds (primarily the Corvidae family) as the hosts or reservoirs, before humans and other mammals get infected as the incidental hosts. Mosquitoes belonging to the Culex genus (e.g., C. tarsalis, C. pipiens, C. restuans, C. quinquefasciatus) that feed on birds appear to be the most important WNV vectors. Among the Culex species, C. tarsalis is a species that is abundant in western North America and is the most efficient vectors of WNV among those tested. This species is also responsible for transmitting, maintaining, and amplifying Western equine encephalitis and St. Louis encephalitic virus, and it appears to have the greatest potential to amplify and maintain WNV.


Culex pipiens is another species that is abundant in urban and suburban areas in eastern North America and British Columbia, and is considered to be the most likely vector for WNV in those areas. Mosquitoes become infected with WNV when they feed on infected birds that can carry the virus in their blood for several days. Infected mosquitoes can transmit WNV to humans and other animals when they get bitten for their blood (see Figure 1). In rare cases, human-to-human transmission of WNV may occur through organ donations and blood transfusions; the virus may also be transmitted from pregnant women either directly to their fetuses or through their breast milk.2 However, there is no evidence suggesting that humans can get infected with WNV by birds, ticks, or any insects other than mosquitoes. Even though household pets can get infected with WNV by mosquitoes, they cannot transmit the virus to their owners.


Diagnosing WNV


Since the initial WNV outbreak in New York City, IVD manufacturers have developed various laboratory tests to detect WNV infection. Most of these tests are based on detecting either immunoglobulins (i.e., IgM, IgG) in the blood or spinal fluid that are produced in response to the infection, or viral nucleic acid fragments in the blood by using nucleic acid amplification techniques.


Other WNV tests include hemagglutination inhibition (HI) tests and plaque reduction neutralization tests (PRNT). HI tests measure the highest dilution of serum that inhibits hemagglutination (HI titer) of erythrocytes by adding the virus to specific antibodies in the test sera. PRNT measures the presence of specific neutralization antibodies (IgM and IgG) in the test serum when they are added to the growing virus culture. This test is performed in biosafety level-three facilities and is used to confirm the presumptive positive ELISA results. But conducting both of these tests is time-consuming.


Although useful for screening blood bank samples, the nucleic acid amplification test (NAAT) method is not readily applicable to a clinical diagnosis of WNV infection. While the method is specific for detecting acute infections during viremia, a small amount of viral nucleic acid present in the blood may not be clinically significant or sensitive in developing WNV since the disease may resolve itself in many cases. However, a transfusion of WNV-infected blood in immunocompromised patients may produce serious outcomes. The NAAT method for detecting WNV has received clearance in the United States and Canada, and is commercially available to detect the presence of all viruses in the Japanese encephalitis serocomplex.


Detecting the presence of IgM and IgG over a certain concentration threshold appears to be the most efficient tool for diagnosing WNV infection. Typically, WNV-infected patients have detectable amounts of WNV-specific serum IgM by the eighth day postinfection, while WNV-specific serum IgG is detectable by three weeks postinfection. The virus itself usually cannot be detected when WNV-specific serum IgM appear, although both IgM and IgG may remain for more than one year. Based on this fact, CDC developed an immunoglobulin M antibody capture ELISA (MAC-ELISA) as a sensitive early marker for WNV infections and a corresponding IgG ELISA using monoclonal antibodies as the antigen capture vehicles.3,4 Many public health laboratories adopted these tests to diagnose WNV antibodies in blood and spinal fluid.


Panbio Ltd. (Queensland, Australia) developed the first commercial indirect fluorescent antibody test that provided results comparable to the CDC MAC-ELISA by using WNV recombinant protein E as an antigen. A number of IgM- and IgG-based ELISA tests were also produced by Focus Technologies Inc. (Cypress, CA), InBios International Inc. (Seattle), and Panbio by using recombinant envelop protein E or inactivated purified WNV as an antigen (NY 99 strain). These tests have been cleared by FDA and Health Canada, and are commercially available.


Other WNV tests include an ELISA format using a combination of a purified recombinant envelope and nonstructural protein 5 of WNV as antigens to increase specificity. Another test is a fluorescent microsphere immunoassay format that couples an rWNV-E antigen to polystyrene microspheres and can provide the basis for multiplex microsphere immunoassays to measure simultaneously antibody reactivity with several recombinant flavivirus antigens.5,6 Studies have described a duplex microsphere immunoassay for detecting the anti-WNV and anti-St. Louis encephalitis viruses, which couple selective antibodies to microspheres.7


Previously, all WNV IgM or IgG tests on the market were primarily ELISA-based devices (e.g., Focus Technologies' WNV IgM Capture ELISA, Panbio's WNV IgM Capture ELISA). These assays are multistep procedures that require instrumentation to read the tests, need subsequent calculations to obtain the index values for interpreting the test results, and take 6-10 hours to complete. Spectral Diagnostics Inc. (Toronto) recently developed RapidWN, a rapid format immunochromatographic strip-based WNV IgM assay that produces results in less than 30 minutes and is substantially equivalent to not only the CDC MAC-ELISA but also Focus' and Panbio's IgM capture ELISAs. RapidWN has been cleared by FDA and licensed by Health Canada. Unlike other ELISA-based devices that use serum specimens, the RapidWN test can be performed with both serum and plasma samples.


In all commercially available IgM capture devices, since they do not differentiate between past and current WNV infections, additional IgG avidity testing is required. Avidity is usually estimated based on the ability of IgG to dissociate from the antigen after incubation with a urea solution. The presence of both IgM and low avidity IgG in patient serum indicates a recent infection. As the infection proceeds, newly generated IgG antibodies showed higher binding affinity to the virus antigen, and the IgG avidity increases. The presence of high avidity IgG reveals a previous infection.


Comparative Assessment


Table I. (click to enlarge) Performance characteristics of some commercially available WNV IgM ELISA devices.*
Table I summarizes the diagnostic efficacy of the ELISA-based commercial devices for WNV. All three devices were comparable in clinical sensitivity when confirmed-positive WNV encephalitis or meningitis patients' sera were tested. The apparent lower sensitivity in Focus' WNV IgM ELISA was due to one sample that produced indeterminate results. The clinical sensitivity of Panbio's WNV IgM ELISA varied from 80.4% in one report to 100% in another study. Thus, the apparent variations in percentage sensitivities among the devices must be taken within the context of each study in which underlying explanations are provided.


The serological sensitivities with PRNT-confirmed specimens or the percentage agreement data with CDC's WNV IgM ELISA were also comparable among the devices. The devices' serological specificities with endemic nonflavivirus specimens were also similar. However, the Panbio test reported a serological specificity of 85.5% with PRNT negative specimens, and separate data for nonflavivirus specimens were not reported.


Table II. (click to enlarge) Serological specificity and clinical sensitivity of the RapidWN WNV IgM test compared with other devices.
The diagnostic sensitivity of RapidWN, a strip-based lateral-flow device, is also similar to the ELISA-based devices.8 Table II shows the comparative data generated from blind, randomized retrospective sample studies at three sites. Compared with Focus' WNV IgM Capture ELISA, RapidWN produced 99% agreement when nonflavivirus IgM serum specimens were tested, 98% agreement in negative samples, and 95% agreement in positive samples. RapidWN also had 100% agreement in the PRNT-confirmed samples with neuroinvasive symptoms. In addition, the device showed 97% and 96% agreement with CDC's MAC-ELISA in positive and negative samples, respectively, and 100% and 98% agreement with Panbio's WNV IgM Capture ELISA in positive and negatives specimens.


An independent study recently conducted at five public health laboratories in the United States evaluated the performance of the RapidWN device versus some of the predicate commercial assays and produced similar results. RapidWN showed 98.8% sensitivity and 95.3% specificity with positive and negative predictive values of 96.3 and 98.4%, respectively.9


The small differences in efficacy between the ELISA-based and strip-based devices are found in their test configurations and the use of antigens, both native and recombinant. Some devices contain sections of WNV envelope glycoprotein E, while others may also contain preM fragments of the envelope protein. But all the devices use monoclonal antiflavivirus antibodies (e.g., 6B6C-1 developed by CDC) as a detector molecule and antihuman IgM derived from rabbit, sheep, or goat as a capture entity.


Table III. (click to enlarge) Test configurations of different WNV IgM ELISA devices.*
Table III shows the differences in the test configurations and other parameters. Since the glycoprotein E structure is similar for all flaviviruses, some degree of cross-reactivity with other species of the genus is expected. Although, attempts have been made to synthesize different sections of this protein, use different test configurations, and add blocking reagents to minimize cross-reactivity. Because of the differences in the assay constituents and configurations among the devices, different cutoff levels or index values were assigned and clinically validated (see Table III). While the ELISA devices provided positive, negative, and equivocal (indeterminate) results based on an assigned index value, RapidWN produced visible bands indicating positive, negative, or invalid results based on WNV IgM levels in the sample (see Figure 2).


Figure 2. (click to enlarge) The RapidWN WNV IgM Device by Spectral Diagnostics Inc. (Toronto).
For the RapidWN device, the reactants' concentrations have been adjusted and optimized by using calibrators so that the test produces a visible band or positive signal at WNV IgM index values greater than or equal to 1.1 of the Focus WNV IgM ELISA device, and negative results below that number. Visible pink-purple horizontal bands will appear in the test area if the level of the WNV IgM antibodies in the human serum sample is above the cutoff level in relation to the Focus WNV IgM ELISA test. A pink-purple band in the control area indicates that the test is working properly, and regardless of the WNV IgM levels, such a band must always appear in order for the test to be valid (see Figure 3). Besides being a device that is easy to use with minimal training, RapidWN also offers time savings (e.g., results in 15 minutes after sample application) and does not require instrumentation to read the test results.


Figure 3. (click to enlarge) The results interpretation of the RapidWN device. The figure shows the position of the test and control bands and examples of positive (both bands present), negative (test bands absent), and invalid (control bands absent) test results.
Cross-Reactivity with Other Viruses


While all the commercially available WNV test devices described above are IgM selective, they show some degree of cross-reactivity with other flaviviruses. The Panbio and Focus ELISA-based devices cross reacted with dengue virus specimens (12% and 40%, respectively), while the InBios test reacted with both dengue and systemic lupus erythematosus specimens (29% and 53%, respectiely). In a limited samples study, RapidWN also showed some cross-reactivity with dengue virus specimens (10%), while no cross-reactivity was observed with St. Louis, Japanese, California, or Eastern equine encephalitis viruses.


Similarly, like other ELISA devices, no significant interferences were observed with common serum analytes such as albumin, hemoglobin, bilirubin, or triglycerides at 2-3 times the normal concentrations found in human blood. All the devices exhibited some degree of cross-reactivity with the RF samples, and the extent of cross-reactivity can be either reduced by using blocking solutions or eliminated by background subtraction.


Nisar A Shaikh, MD, PhD, is senior vice president of corporate quality, scientific, and regulatory affairs at Spectral Diagnostics Inc. (Toronto). He can be reached at nshaikh@spectraldx.com.


IVD devices have proven to be indispensable in diagnosing WNV and providing timely information for disease management. All currently available WNV IgM capture ELISA devices efficiently screen patients for the presumptive diagnosis of WNV infection and require confirmation by PRNT. Since IgM antibodies appear in a few days postinfection, serum samples should be taken multiple times to confirm the presence of IgM in a patient's blood. With ELISA-based devices, a number of samples can be processed simultaneously in an active laboratory, although the test is time-consuming, laborious, and requires instrumentation to read and interpret the results. A strip-based device, such as the RapidWN test, offers ease of use and rapid results that can be visibly read. The test can be run on single or multiple samples, provides positive results in minutes, and has a good cost-benefit ratio and room temperature stability of reagents. If warranted, test results can be quantified with a strip reader. Microsphere immunoassay-based tests can also provide enhanced specificity by multiplexing with additional antigens and discrimination between current and past infections.



1. AK Malan, TB Martins, HR Hill, et al., “Evaluations of Commercial West Nile Virus Immunoglobulin G (IgG) and IgM Enzyme Immunoassays Show the Value of Continuous Validation,” Journal of Clinical Microbiology 42 (2004): 727-733.


2. JM Conly and BL Johnston, “Why the West in West Nile Infections?” Canadian Journal of Infectious Disease and Medical Microbiology 18 (2007): 285-288.


3. AJ Johnson, DA Martin, N Karabatsos, et al., “Detection of Anti-Arboviral Immunoglobulin G by Using a Monoclonal Antibody-Based Capture Enzyme-Linked Immunosorbent Assay,” Journal of Clinical Microbiology 38 (2000): 1827-1831.


4. DA Martin, DA Muth, T Brown, et al., “Standardization of Immunoglobulin M Capture Enzyme-Linked Immunosorbent Assays for Routine Diagnosis of Arboviral Infections,” Journal of Clinical Microbiology 38 (2000): 1823-1826.


5. SJ Wong, RH Boyle, VL Demarest, et al., “Immunoassay Targeting Nonstructural Protein 5 to Differentiate West Nile Virus Infection from Dengue and St. Louis Encephalitis Virus Infections from Flavivirus Vaccination,” Journal of Clinical Microbiology 41 (2003): 4217-4223.


6. SJ Wong, VL Demarest, RH Boyle, et al., “Detection of Human Anti-Flavivirus Antibodies with a West Nile Virus Recombinant Antigen Microsphere Immunoassay,” Journal of Clinical Microbiology 42 (2004): 65-72.


7. AJ Johnson, AJ Noga, O Kosoy, et al., “Duplex Microsphere-Based Immunoassay for Detection of Anti-West Nile Virus and Anti-St. Louis Encephalitis Virus Immunoglobulin M Antibodies,” Clinical and Diagnostic Laboratory Immunology 12 (2005): 566-574.


8. NA Shaikh, J Ge, YX Zhao et al., “Development of a Novel, Rapid and Sensitive Immunochromatographic Strip Assay Specific for West Nile Virus (WNV) IgM and Testing of Its Diagnostic Accuracy in Patients Suspected of WNV Infection,” Clinical Chemistry 53 (2007): 2031-2034.


9. AR Sambol and SH Hinrichs, “Evaluation of a New West Nile Virus Lateral-Flow Rapid IgM Assay,” Journal of Virological Methods (2009), in press.



Copyright ©2009 IVD Technology


Your rating: None Average: 5 (2 votes)