JOURNAL OF CLINICAL MICROBIOLOGY, Apr. 2004, p. 1782–1784 Vol. 42, No.

4
0095-1137/04/$08.00⫹0 DOI: 10.1128/JCM.42.4.1782–1784.2004
Copyright © 2004, American Society for Microbiology. All Rights Reserved.

Use of an Immunoglobulin G Avidity Test To Discriminate between

Primary and Secondary Dengue Virus Infections
Vanda Akico Ueda Fick de Souza,1 Silvana Fernandes,1 Evaldo Stanislau Araújo,2
Adriana Fumie Tateno,1 Olímpia M. N. P. F. Oliveira,2 Renato dos Reis Oliveira,1
and Cláudio Sérgio Pannuti1*
Laboratório de Virologia (LIMHC-52), Instituto de Medicina Tropical de São Paulo, Departamento de Moléstias
Infecciosas e Parasitárias da Faculdade de Medicina, Universidade de São Paulo,1 and Hospital Ana
Costa, Santos,2 São Paulo, Brazil
Received 15 August 2003/Returned for modification 26 November 2003/Accepted 25 December 2003

An enzyme-linked immunosorbent assay-based immunoglobulin G (IgG) antibody avidity test was evaluated
by using sera from 57 patients with acute dengue infection. Overall, 55 of 57 patients were correctly classified
(27 of 27 with primary dengue and 28 of 30 with secondary dengue). We conclude that the IgG avidity test can
be useful for differentiating between acute, primary, and secondary dengue infections.

An estimated 2.5 billion people in over 100 countries are at
risk of acquiring dengue infection. Up to 50 million infections,
resulting in 500,000 cases of dengue hemorrhagic fever (DHF)
and 22,000 deaths, occur annually (World Health Organization
website [http://www.who.int/ctd/dengue/burdens.htm]). The
pathogenesis of DHF is still unclear (6, 8), although there is
good evidence that secondary infection by a different serotype
in patients harboring preexisting heterologous dengue antibod-
ies increases the risk of DHF (3, 7, 9, 15, 20). The mechanism
postulated to explain this epidemiological correlation is known
as immune enhancement (9, 10, 16, 25, 29).
The immunoglobulin M (IgM) capture enzyme immunoas-
say is the most widely used serologic test for dengue diagnosis
(6). In addition, the hemagglutination inhibition test has tra-
ditionally been used to classify dengue infections as primary or
secondary (29). However, when the interval between the acute-
and convalescent-phase samples is less than 7 days, in the
absence of a change in antibody titers in acute- and convales-
cent-phase samples or in single specimens, interpretation of
the dengue hemagglutination inhibition antibody response be-
comes difficult (6, 29). The hemagglutination inhibition test
also requires serum pretreatment with acetone or kaolin to
remove nonspecific inhibitors of hemagglutination and further
absorption with gander or human type O red blood cells (29)
and is not commercially available. Methods based on IgM/IgG
ratios have also been used (13, 28). The assessment of IgG
antibody avidity is a very useful tool for differentiating between
primary and secondary immune responses (12). The IgG anti-
body avidity test makes use of the fact that in the primary
infection, the specific IgG antibody response begins with low-
avidity IgG antibodies which gradually evolve to high-avidity
antibodies. In the secondary infection, the rapid antibody re-
sponse is characterized by the production of high-avidity anti-
bodies (23). The IgG antibody avidity test has been used to
diagnose acute viral infections (5, 12, 17, 21, 27), to discrimi-
* Corresponding author. Mailing address: Laboratório de Virologia,
Instituto de Medicina Tropical de São Paulo, Av. Eneas de Carvalho
Aguiar, 470, CEP 05043-000 São Paulo (SP), Brazil.
nate recent primary infection from reinfection or reactivation
(1, 2, 4), to estimate the efficacy of vaccines (22), and to
identify vaccine failures (18, 19). The aim of the present study
was to evaluate the IgG avidity test for differentiating primary
from secondary dengue infections.
Sera were obtained between March and July 2002 during a
dengue outbreak in Santos, São Paulo, Brazil. During the study
period, the Health Department of the State of São Paulo
reported the occurrence of three dengue virus serotypes (Den
1, Den 2, and Den 3) in the city of Santos (Centro de Vigilância
Epidemiológica website [http://www.cve.saude.gov.br/htm/z00
/if_sem_den02.htm]). Blood samples were collected during the
acute phase of infection, and a convalescent-phase sample was
obtained at least 7 days later. The study was approved by the
Ethics and Research Committee of the University of São Paulo
Medical School. Confirmation of acute dengue was obtained
by the detection of IgM antibodies. IgG and IgM antibodies
were detected by using Dengue ELISA (enzyme-linked immu-
nosorbent assay) IgG and Dengue ELISA IgM (Focus Tech-
nologies, Cypress, Calif.). To standardize the IgG avidity test,
primary and secondary dengue infections were defined by us-
ing strict diagnostic criteria. Primary dengue infection was de-
fined by a negative IgG test for a serum sample from the acute
phase collected at least 4 days after disease onset, followed by
seroconversion in the convalescent-phase serum sample. Sec-
ondary dengue infection was defined by a positive IgG test for
an acute-phase sample obtained within 4 days of disease onset
and a laboratory-confirmed diagnosis of acute dengue infec-
tion. The IgG avidity test was performed by using a modified
Dengue ELISA IgG kit (Focus Technologies) to which a urea
incubation step was introduced. Briefly, serum samples diluted
1:100 were dispensed in duplicate into wells coated with den-
gue antigen. After 1 h of incubation at room temperature, the
plates were rinsed once with washing buffer; half of the wells
were then rinsed with phosphate-buffered saline (pH 7.2) con-
taining urea, and the other half were rinsed with phosphate-
buffered saline without urea. After two further rinses, the test
was performed according to the manufacturer’s instructions.
The avidity index, expressed as a percentage, was calculated as

1782
VOL. 42, 2004
FIG. 1. IgG antibody avidity in sera from patients with primary and
secondary dengue infections with washing schedules of 7 M urea for 10
min and 8 M urea for 5 min.
the ratio of the optical density with urea to the optical density
without urea times 100. Samples whose absorbances were
above the limit of the ELISA reader were retested after a
1:1,000 or 1:10,000 dilution to allow better calculation of the
avidity index.
Variability in results when ELISA is used to test avidity has
been reported when different commercial plates and different
sources of antigen or coating techniques have been used (14,
26). Different urea concentrations and elution times also have
been employed for different antigens (2, 11). For this reason,
we tested different schedules of urea washing to standardize
the procedure for the commercial dengue ELISA used here.
Sera from 12 convalescent patients with confirmed primary
infections and sera from 12 patients with confirmed secondary
dengue infections were tested by using washing schedules of 6
M urea for 10 min, 7 M urea for 10 min, and 8 M urea for 5
min. Similar results were obtained when schedules of 7 M urea
for 10 min and 8 M urea for 5 min were used to discriminate
a primary from a secondary immune response. The mean avid-
ity index for primary infection was 11 with urea at 7 M for 10
min as well as with urea at 8 M for 5 min. The mean avidity
indices for secondary infection were 59 ⫾ 17 and 57 ⫾ 17 for
urea at 7 M for 10 min and for urea at 8 M for 5 min,
respectively (Fig. 1). The schedule of 7 M urea for 10 min was
chosen, since it has been used to evaluate IgG avidity in mea-
sles virus infections (22).
The performance of the IgG avidity test with serum samples
selected from convalescent patients with laboratory-confirmed
acute dengue infections was evaluated. Serum samples from 57
patients were collected within 30 days of disease onset. Of
these samples, 27 were obtained from patients with confirmed
primary infection and 30 were obtained from patients with
confirmed secondary infection according to the definition cri-
teria adopted. A receiver operating characteristic (ROC) curve
analysis employing Analyze-it software was used to evaluate
the ability of the avidity test to distinguish between primary
and secondary dengue infections. The cutoff point was defined
by the highest sum of the sensitivity and specificity estimates.
The area under the ROC curve was used as a measure of test
accuracy (24). Quantitative data (avidity indices and times of
NOTES 1783
TABLE 1. Performance of IgG avidity test with convalescent-phase
sera from patients with primary and secondary dengue infectionsa
Data for patients with:
Primary infection Secondary infection
AI range
No. (%) No. (%)
Mean AI (SD) Mean AI (SD)
of patients of patients
ⱖ24 0 (0.0) 28 (93.3) 52 (16.7)
⬍24 27 (100.0) 12 (4.6) 2 (6.7) 21 (2.8)
Total 27 (100.0) 12 (4.6) 30 (100.0) 50 (18)
a
AI, avidity index.
sample collection) were compared by one-way analysis of vari-
ance (data are presented here as means ⫾ standard devia-
tions).
The highest sum of the sensitivity and specificity estimates
used to define secondary infection was obtained at a cutoff
point of ⱖ24% for the IgG avidity index. At this cutoff point,
the IgG avidity test provided correct classifications of 55 of 57
patients (27 of 27 patients with primary dengue [100%] and 28
of 30 patients with secondary dengue [93%]). The test was
highly accurate, as the area under the ROC curve was almost
perfect (0.995). The mean avidity index for secondary infection
was 50 ⫾ 18.0, significantly higher than that for the primary-
infection group (12 ⫾ 4.6) (Table 1). It is known that the
avidity of IgG antibodies increases according to the time
elapsed after infection (23). However, the higher avidity indi-
ces found for secondary infection in the present study were not
related to the duration of infection, since all samples were
collected within 30 days of disease onset and the mean times
elapsed for sample collection were similar for the primary (20
⫾ 3.4 days) and secondary (19 ⫾ 2.7 days) infection groups (P
⫽ 0.71).
We conclude that the IgG avidity test, which uses a single
blood sample and a commercially available kit, has the poten-
tial to be a useful tool for differentiating between acute pri-
mary and secondary dengue infections.
REFERENCES
1. Baccard-Longere, M., F. Freymuth, D. Cointe, J. M. Seigneurin, and L.
Grangeot-Keros. 2001. Multicenter evaluation of a rapid and convenient
method for determination of cytomegalovirus immunoglobulin G avidity.
Clin. Diagn. Lab. Immunol. 8:429–431.
2. Blackburn, N. K., T. G. Besselaar, B. D. Schoub, and K. F. O’Connell. 1991.
Differentiation of primary cytomegalovirus infection from reactivation using
the urea denaturation test for measuring antibody avidity. J. Med. Virol.
33:6–9.
3. Burke, D. S., A. Nisalak, D. E. Johnson, and R. M. Scott. 1988. A prospective
study of dengue infections in Bangkok. Am. J. Trop. Med. Hyg. 38:172–180.
4. Enders, G., and F. Knotel. 1989. Rubella IgG total antibody avidity and IgG
subclass-specific antibody avidity assay and their role in the differentiation
between primary rubella and rubella reinfection. Infection 4:218–226.
5. Gray, J. J. 1995. Avidity of EBV VCA-specific IgG antibodies: distinction
between recent primary infection, past infection and reactivation. J. Virol.
Methods 52:95–104.
6. Gubler, D. J. 1998. Dengue and dengue hemorrhagic fever. Clin. Microbiol.
Rev. 11:480–496.
7. Guzmán, M. G., G. P. Kouri, J. Bravo, M. Soler, S. Vazquez, and L. Morier.
1990. Dengue hemorrhagic fever in Cuba. Am. J. Trop. Med. Hyg. 42:179–
184.
8. Guzmán, M. G., and G. Kouri. 2002. Dengue: an update. Lancet Infect. Dis.
2:33–42.
9. Halstead, S. B. 1988. Pathogenesis of dengue: challenges to molecular biol-
ogy. Science 239:476–481.
10. Halstead, S. B., and E. J. OⴕRourke. 1977. Dengue viruses and mononuclear
1784 NOTES
phagocytes. I. Infection enhancement by non-neutralizing antibody. J. Exp.
Med. 146:201–217.
11. Hedman, K., and S. A. Rosseau. 1989. Measurement of avidity of specific
IgG for verification of recent primary rubella. J. Med. Virol. 27:288–292.
12. Hedman, K., M. Lappalainen, M. Söderlund, and L. Hedman. 1993. Avidity
of IgG in serodiagnosis of infectious diseases. Rev. Med. Microbiol. 4:123–
129.
13. Innis, B. L., A. Nisalak, S. Nimmannitya, S. Kusalerdchariya, V. Chongs-
wasdi, S. Suntayaborn, P. Puttisri, and C. H. Hoke. 1989. An enzyme-linked
immunosorbent assay to characterize dengue infections where dengue and
Japanese encephalitis co-circulate. Am. J. Trop. Med. Hyg. 40:418–427.
14. Inouye, S., A. Hasegawa, S. Matsuno, and S. Katow. 1984. Changes in
antibody avidity after virus infections: detection by an immunosorbent assay
in which a mild protein-denaturing agent is employed. J. Clin. Microbiol.
20:525–529.
15. Kliks, S. C., S. Nimmannitya, A. Nisalak, and D. S. Burke. 1988. Evidence
that maternal dengue antibodies are important in the development of den-
gue hemorrhagic fever in infants. Am. J. Trop. Med. Hyg. 38:411–419.
16. Kliks, S. C., A. Nisalak, W. E. Brandt, L. Wahl, and D. S. Burke. 1989.
Antibody-dependent enhancement of dengue virus growth in human mono-
cytes as risk factor for dengue hemorrhagic fever. Am. J. Trop. Med. Hyg.
40:444–451.
17. Meurman, O., M. Waris, and K. Hedman. 1992. Immunoglobulin G antibody
avidity in patients with respiratory syncytial virus infection. J. Clin. Micro-
biol. 30:1479–1484.
18. Paunio, M., K. Hedman, I. Davidkin, M. Valle, O. P. Heinonen, P. Leinikki,
and H. Peltola. 2000. Secondary measles vaccine failures identified by mea-
surement of IgG avidity: high occurrence among teenagers vaccinated at a
young age. Epidemiol. Infect. 124:263–271.
19. Paunio, M., K. Hedman, I. Davidkin, and H. Peltola. 2003. IgG avidity to
distinguish secondary from primary measles vaccination failures: prospects
J. CLIN. MICROBIOL.
for a more effective global measles elimination strategy. Expert Opin. Phar-
macother. 4:1215–1225.
20. Rigau-Perez, J. G., G. G. Clark, D. J. Gubler, P. Reiter, E. J. Sanders, and
A. V. Vorndam. 1998. Dengue and dengue haemorrhagic fever. Lancet 352:
971–977.
21. Söderlund, M., C. S. Brown, B. J. Cohen, and K. Hedman. 1995. Accurate
serodiagnosis of B19 parvovirus infections by measurement of IgG avidity.
J. Infect. Dis. 171:710–713.
22. Souza, V. A. U. F., C. S. Pannuti, L. M. Sumita, and H. F. Andrade, Jr. 1997.
Enzyme-linked immunosorbent assay-IgG antibody avidity test for single
sample serologic evaluation of measles vaccines. J. Med. Virol. 52:275–279.
23. Steiner, L. A., and H. N. Eisen. 1967. Sequential changes in the relative
affinity of antibodies synthesized during the immune response. J. Exp. Med.
126:1161–1183.
24. Swetz, J. A. 1988. Measuring the accuracy of diagnostic systems. Science
240:1285–1293.
25. Thein, S., M. M. Aung, T. N. Shwe, M. Aye, A. Zaw, K. Aye, K. M. Aye, and
J. Aaskov. 1997. Risk factors in dengue shock syndrome. Am. J. Trop. Med.
Hyg. 56:566–572.
26. Thomas, H. I. J., and P. Morgan-Capner. 1991. The use of antibody avidity
measurements for the diagnosis of rubella. Rev. Med. Virol. 1:41–50.
27. Tuokko, H. 1995. Detection of acute measles infections by indirect and
␮-capture enzyme immunoassays for immunoglobulin M antibodies and
measles immunoglobulin G antibody avidity enzyme immunoassay. J. Med.
Virol. 45:306–311.
28. Vaughn, D. W., A. Nisalak, T. Solomon, S. Kalayanarooj, N. M. Dung, R.
Kneen, A. Cuzzubbo, and P. L. Devine. 1999. Rapid serologic diagnosis of
dengue virus infection using a commercial capture ELISA that distinguishes
primary and secondary infections. Am. J. Trop. Med. Hyg. 60:693–698.
29. World Health Organization. 1997. Dengue haemorrhagic fever: diagnosis,
treatment, prevention and control. World Health Organization, Geneva,
Switzerland.