Background & objectives: Human parvovirus B 19 (PVB 19) causes aplastic crisis in children with congenital haemolytic anaemia, erythema infectiosum, abortion and stillbirth. Since data on PVB 19 prevalence is lacking in India, a pilot study was undertaken to estimate the prevalence of IgG antibody in children and adults.
Methods: The samples were obtained from children attending our hospital and from volunteer blood donors, majority of whom were from south India. They included 45 children aged 1-5 yr, 39 aged 6-10 yr, 42 aged 11-15 yr and 100 healthy blood donors > 15 yr of age. Sera were tested for the presence of antibody to PVB 19 using a commercial enzyme immuno assay (EIA).
Results: Of 226 samples tested, 113 (50%) were positive for PVB 19 IgG. The prevalence of antibody increased from 8.9 per cent at 1-5 yr to 70 per cent in those >15 yr : the median age of infection was between 6 and 15 yr. Sex and domiciliary status did not have significant effect on the prevalence of antibody. The IgG antibody index increased significantly with age, suggesting repeated exposure to the virus.
Interpretation & conclusion: This seroprevalence study indicates that large numbers of individuals show exposure to PVB 19 virus. The exposure as indicated by IgG positivity is seen to increase with age. The IgG negative individuals may be considered to be at risk of developing infections due to PVB 19.
Key words IgG antibodies - parvovirus B 19 - pilot study - seroprevalence
Parvovirus B 19 (PVB 19) was discovered serendipitously by Cossart et all while conducting a hepatitis B virus laboratory assay on serum samples. They found parvovirus-like particles by electron microscopy in the sera that tested falsely positive for hepatitis B virus antigen. Since most of the PVB 19 viraemic blood donors in this study were healthy, it was concluded that the virus caused asymptomatic infection. The etiological role of PVB 19 in the causation of aplastic crisis in children with haemolytic anaemia and in erythema infectiosum (EI) was documented in later studies2.3. Subsequently, the role of intrauterine infection by PVB 19 as a cause of foetal hydrops and stillbirth was demonstrated4. Studies from the United Kingdom had shown a clear age-specific increase and widespread exposure to this virus in different population groups5. In the Asia-Pacific region and Africa also, the infection has been shown to be high in the general population6,7. Information on the prevalence of PVB 19 infection in India is lacking. Hence, a pilot study was done to estimate the prevalence of IgG antibody to the virus in a population attending a hospital in southern India.
Material & Methods
Subjects: The study was conducted from January 1999 to September 2000. The samples tested included, those obtained by convenient sampling. Part of the sera submitted for pre-operative screening from children aged 1-15 yr were used in this study. These children had no clinical manifestations related to PVB 19 infection such as haematological disorders (anaemias, hereditary spherocytosis, thalassaemias, pyruvate kinase deficiency), rashes, congenital deformities and miscarriages, arthritis and arthralgias, neurological manifestations and myocarditis. For the above 15 yr age group samples were collected from healthy volunteer blood donors (adults, age range of 18 to 55 yr) at the Blood Bank at the Christian Medical College & Hospital (CMCH), Vellore, India. These individuals came predominantly from the southern Indian states, though there were some from other parts of India. The medical records of all the children whose sera were used were reviewed and those with any acute infection, haematological disorder, bone or joint infection. acute abdomen and those who had received transfusions of blood or blood products were excluded. These children between I and 15 yr of age were admitted for one of the following reasons. systemic lupus erythromatosus, inguinal hernias, gastroenteritis, lymphangina/lymphadenopathy, hydronephrosis, hydrocele, pyogenic abscess, sebaceous cyst and other cysts, stricture urethra and, other non-specific miscellaneous conditions. They included 45 in the 1-5 yr age group, 39 in the 6-10 yr group, 42 in the 11-15 yr group and 100 adults >15 yr of age. All samples had been stored at -20 deg C before testing, for periods not longer than 3 months (range 2 wk to 3 months).
Enzyme immuno assay: The PVB 19 IgG (3rd generation) EIA, Lot Nos. 35KTO53 & 35KTO84 (Biotrin International, Dublin, Ireland) used in this study is a sandwich EIA that detects the IgG class of antibodies to PVB 19 virus in human serum. Three kits were used to complete the study. The PVB 19 specific IgG present in the serum (1:100 dilution) binds to purified B 19 recombinant VP 2 protein coated on the microtiter well. Following a wash step, peroxidase labelled rabbit anti-human IgG is added which binds to the IgG present. The washes were carried out using the ELx 50 (Bio-Tek Instruments Inc. Vermont, USA) automated ELISA washer. The whole complex detected by the addition of substrate, tetra methyl benzidine (TMB), turns blue in the presence of peroxidase (a stable yellow end product is achieved by addition of a stopping reagent). The optical density (OD) of the product was read colorimetrically in an EIA reader, (dual wave length 450/630nm filter) ELx 800 (Bio-Tek Instruments Inc. Vermont, USA). The positive and negative control sera were supplied by the manufacturer with the kit.
The presence or absence of anti-PVB19 IgG antibody was determined in relation to a calculated cut-off value (COV). The COV was calculated by multiplying the mean absorbance of the positive control (PC) with the lot specific constant (LSC), i.e., COV = PC x LSC. The LSC is a correction value to compensate for variability between lots of reagents. Samples with a mean absorbance reading greater than or equal to the COV x 1.1 are reactive (positive) for anti-PVB 19 IgG antibody. Calculation of the IgG index: This was estimated by using the principle of Signal: Noise ratio (SIN). The OD of the test sample divided by the mean OD of the negative controls was expressed as the IgG index.
Statistical analysis: The Chi square test for difference/trend in proportions was carried out, using the EPI info ver. 6.04 software. Linear regression analysis was employed to study the association between IgG antibody index and age.
Results & Discussion
The mean OD value of the kit positive control ranged from 1.602-2.041 in three different runs. The LSC for the two lots of the kits used for this study were 0.16 and 0.14 respectively. The COV for the three assays were 0.256, 0.245 and 0.286 respectively. Of the 226 sera tested, 113 (50%) were positive. The range of OD for the 113 negative samples was 0.004-0.27 in the three kits used and was below the respective cut-off values. The seroprevalence according to age, sex and domiciliary status is shown in Table I. The seroprevalence of IgG antibody to PVB 19 virus increased significantly with age, from 8.9 per cent at 1-5 yr to 70 per cent in those >15 yr (P15 yr age group.
The Fig. shows a histogram of the IgG index value stratified and according to age groups. The percentage of subjects with IgG index of 11 in the different age groups increased steadily from 4.4 per cent at 1-5 yr of age to 67 per cent in those >15 yr (a significant trend, P 15 yr, respectively.
Our data showed that half the children and adults studied had IgG antibody to PVB 19. The seroprevalence increased steadily with age, with 70 per cent of adults having prior exposure to the virus. This finding is similar to the observations in a few other countries like England and Wales5 and Japan6. On the other hand, in Eritrea 56 per cent of children aged 1-5 yr had antibodies with a further increase in seroprevalence to 76 per cent in the children above 5 yr of age, suggesting an earlier age of infection. In contrast, seroprevalence among those
We did not document any statistically significant differences in seroprevalence between males and females or between those with urban and rural residence. A study in rural Brazil documented very low rates in the tribal populations. This is likely to be due to the fact that this was a closed population with little exposure to those from outside the community, decreasing their risk of exposure to the virus8.
Among the males in our study, the seroprevalence increased steadily with age. In contrast, among females there was an increase in seroprevalence from 6.3 per cent in the 1-5 yr age group to 71.4 per cent in the 11-15 yr age group, but subsequently there was a drop in the prevalence in the > 15 yr age group to 50 per cent. The difference was not statistically significant between the age groups 11-15 and >15 yr.
We noted an increase in the antibody index which is a reflection of the specific IgG in a given serum sample increasing with the age of the individual. Hence, it can be surmised that the high IgG levels in adults do not reflect persistence of antibodies following childhood infection but subsequent booster of the immune response due to repeat exposure to the virus. We could postulate that PVB 19 circulation in the population results in infections starting early in life with repeated exposure at a later age, resulting in a boost of antibody levels.
In conclusion, the data from this pilot study carried out in a part of south India show high rates of exposure to PVB 19 during childhood. Further, the sample size and the age intervals of subjects may not reflect the true age related factors of infection in different areas of a large country like India. More detailed population-based studies in different parts of the country would be required to generate robust information. The clinical manifestations of infection in the Indian children also remain to be studied. In a previous study, we had documented the occurrence of foetal PVB 19 infection in India but further studies are needed to delineate the magnitude of the problem .
References
1 Cossart YE, Field AM, Cant B, Widdows D. Parvoviruslike particles in human sera. Lancet 1975; 1: 72-3
2. Anderson MJ, Davis LR, Hodgson J, Jones SE, Murtaza L, Pattison JR, et al. Occurrence of infection with a parvovirus-like agent in children with sickle cell anemia during a two-year period. J Clin Pathol 1982; 35 : 744-9.
3. Anderson MJ, Jones SE, Fisher-Hoch SP, Lewis E, Hall SM, Bartlett CL, et al. Human parvovirus, the cause of erythema infectiosum (fifth disease)? Lancet 1983; /:13 78.
4. Knott PD, Welply GAC, Anderson MJ. Serologically proved intrauterine infection with parvovirus. Br Med J 1984; 289 : 1660.
5. Cohen BJ, Buckley MM. The prevalence of antibody to human parvovirus 819 in England and Wales. J Med Microbiol 1988; 25 151-3.
6. Kelly HA, Siebert D, Hammond R, Leydon J, Kiely P, Maskill W. The age-specific prevalence of human parvovirus immunity in Victoria, Australia compared with other parts of the world. Epidemiol Infect 2000, 124 : 449-57.
7. Tolfvenstam T, Enbom M, Ghebrekidan H, Ruden U, Linde A, Grandien M, et al. Seroprevalence of viral childhood infections in Eritrea. J Clin Virol 2000; 16 : 49-54.
8. de Freitas RB, Wong D, Boswell F, de Miranda MFR, Linhares AC, Shirley J, et al. Prevalence of human parvovirus (B19) and rubellavirus infections in urban and
remote rural areas in northern Brazil. J Med Virol 1990; 32: 203-8.
Abraham M. Abraham P. Jana AK, Kuruvilla KA, Cherian T, Moses PD, ef al. Serology in congenital infections : experience in selected symptomatic infants. Indian Pediatr 1999; 36 : 697-700,
Abraham, Rajeev Rudrariju, Rajesh Kannangai, Koshy George*, Thomas Cherian* Dolly Daniel**, Sandeep Ramalingam & Gopalan Sridharan
Departnents of Clinical Virology, *Child Health Unit I & **Blood Bank, Christian Medical College & Hospital, Vellore, India
Received July 31, 2001
Reprint requests:
Dr G. Sridharan, Professor & Head, Department of Clinical, Virology Christian Medical College & Hospital, Vellore 632004, India
Copyright Indian Council of Medical Research Apr 2002
Provided by ProQuest Information and Learning Company. All rights Reserved
Contact
Home
A
Arthritis