European Journal of Clinical Microbiology & Infectious Diseases

, Volume 28, Issue 8, pp 997–999

Adenovirus gastroenteritis in Hungary, 2003–2006

Authors

    • Veterinary Medical Research InstituteHungarian Academy of Sciences
    • Regional Laboratory of VirologyBaranya County Institute of State Public Health Service
  • P. Kisfali
    • Department of Medical Genetics and Child Development, Faculty of MedicineUniversity of Pécs
  • Á. Bogdán
    • Regional Laboratory of VirologyBaranya County Institute of State Public Health Service
  • V. Martella
    • Department of Animal Health and Well-BeingUniversity of Bari
  • B. Melegh
    • Department of Medical Genetics and Child Development, Faculty of MedicineUniversity of Pécs
  • D. Erdman
    • Division of Viral DiseasesNational Center for Immunization and Respiratory Diseases, Centers for Disease Control and Prevention
  • G. Szűcs
    • Regional Laboratory of VirologyBaranya County Institute of State Public Health Service
Brief Report

DOI: 10.1007/s10096-009-0722-8

Cite this article as:
Bányai, K., Kisfali, P., Bogdán, Á. et al. Eur J Clin Microbiol Infect Dis (2009) 28: 997. doi:10.1007/s10096-009-0722-8

Abstract

The incidence and type distribution of enteric human adenoviruses (HAds) among diarrheic children in south-western Hungary was investigated from 2003 through 2006. Laboratory studies were conducted using commercial antigen detection tests (latex agglutination or immunochromatography), polymerase chain reaction (PCR) amplification, single-strand conformation polymorphism, and sequencing and phylogenetic analysis of a conservative region of the HAd hexon gene. The overall rate of HAd infection in childhood gastroenteritis cases during the 4-year study was 8.1%, with a gradual decrease in detection rates from 11.7% in 2003 to 5.7% in 2006. Molecular studies of a subset of HAd-positive samples found that enteric HAd type 40 strains were identified only in 2003 and 2004, while HAd type 41 strains were identified throughout the 4-year study. Higher detection rates of non-enteric HAds was documented during the first half of the study period when latex agglutination was used in our laboratory for detection. Our study suggests that the choice of diagnostic method may profoundly influence the epidemiologic picture and disease burden attributed to enteric HAd infections.

The disease spectrum of the 52 currently recognized human adenovirus (HAd) types include, but is not limited to, respiratory, ocular and urinary tract infections, and gastroenteritis [1]. Of these, only two serotypes, HAd-40 and HAd-41, are considered to be true enteric pathogens. HAd gastroenteritis most commonly occurs in early childhood. Differential diagnosis for HAd gastroenteritis is impossible based on the clinical picture alone; therefore, adequate case management requires laboratory-based diagnosis. The number of fecally shed viral particles of enteric HAds often exceeds a magnitude of 109 per gram of sample, allowing the use of diagnostic procedures of low to moderate sensitivity, such as electron microscopy, latex agglutination, enzyme immunoassay, or immunochromatography tests [24]. Consequently, inexpensive commercial diagnostic methods that require no specialized equipment or personnel skill are popular and widely used in diagnostic laboratories.

The epidemiology of the enteric HAds has been described in many parts of the world. However, the prevalence of enteric HAds in Central and Eastern Europe has not been well described. Our aims were, therefore, to: (i) provide corresponding information on the prevalence of enteric adenovirus infections in Hungary and (ii) determine the relative importance of different adenovirus types with emphasis on the two enteric HAd types, 40 and 41.

Stool specimens were collected between 2003 and 2006 from diarrheic patients using commercially available antigen detection tests (2003–2004, Slidex Rota-Adeno test, bioMérieux; 2004–2005, VIKIA Rota-Adeno test, bioMérieux; 2005–2006, Rotavirus-Adenovirus Blister test, CerTest BIOTEC). Molecular confirmation and further analysis of a subset of HAd-positive samples included the polymerase chain reaction (PCR) amplification of a conservative region of the hexon gene using primer pair hexAA1913/hexAA1885 [5]. Amplicons (301 bp) were subjected to single-strand conformational polymorphism (SSCP) analysis, as described previously [6]. SSCP patterns were categorized and then selected samples were subjected to direct sequencing. A neighbor-joining tree based on the nucleotide sequences was constructed with MEGA4 software [7] using the p-distance algorithm; bootstrap replicates of 500 were implemented.

Stool specimens from 2,862 gastroenteritis patients (<15 years of age) were obtained between January 2003 and December 2006 from the service area of our laboratory located in south-western Hungary. Of these, 2,258 were screened for enteric HAd and 184 samples (8.1%; annual detection rates: 2003, 11.7%; 2004, 9.9%; 2005, 7.8%; 2006, 5.7%) were tested positive. The median age of patients with HAd gastroenteritis was 1.8 years (range, 0.1–14.9 years). Cases were identified throughout the year, without remarkable seasonal variation. No significant difference in the incidence was seen for those patients who required hospital treatment versus those who sought outpatient clinics (8.1% vs. 6.1%; P = 0.138, chi-square test).

Of the positive samples, 124 (67.4%) were available for molecular confirmation. Of these, 84 were PCR-positive, whereas 40 gave no visible bands on ethidium bromide-stained gels. A total of six different SSCP patterns were identified among the 64 strongly positive samples; representatives of each of these six patterns were sequenced (Fig. 1). Phylogenetic analysis revealed five major genetic clusters, one each representing an individual hexon type (Fig. 1). Based on SSCP patterns and sequence data, we categorized the 51 pattern I samples to be HAd-41, with the five pattern II samples to be HAd-40, showing the predominance of HAdV-41 strains during the 4-year period. In addition to enteric HAds, one HAd-1 (SSCP pattern IV), five HAd-2 (patterns V and VI), and two HAd-31 (pattern III) strains were also identified. In general, good agreement was obtained between the SSCP patterns and the sequence data, suggesting that SSCP analysis may offer an inexpensive and rapid screening tool for typing large numbers of strains. Of note, our SSCP assay did not resolve 3 nt mismatches for HAdV-41 and 1 nt mismatch for HAdV-2 strains.
https://static-content.springer.com/image/art%3A10.1007%2Fs10096-009-0722-8/MediaObjects/10096_2009_722_Fig1_HTML.gif
Fig. 1

Panel A Single-strand conformational polymorphism (SSCP) patterns and their relative annual distribution determined for a subset (n = 64) of HAd-positive specimens collected between 2003 and 2006. Panel B Partial hexon gene (280 nt) sequencing and phylogenetic analysis of 20 selected strains revealed that the six distinguishable patterns corresponded to HAd types 41 (I), 40 (II), 31 (III), 1 (IV), and 2 (V and VI), respectively

Our study had several limitations. First, we found a high prevalence of non-enteric HAds in the period when latex agglutination was used for the laboratory diagnosis of enteric HAd infections. Significant improvement was seen in the specificity of enteric HAd diagnosis when latex agglutination was replaced with immunochromatography (enteric HAd vs. non-enteric HAd detection rates were 71% vs. 29% for latex agglutination and 95% vs. 5% for immunochromatography; P = 0.012, Fisher’s exact test). Second, the cross-reactivity of our primers with non-enteric HAd types was not unexpected given the sequence conservation of the hexon gene region to which they were designed. Since the very first description of these primers [5], sequence modifications have been introduced to facilitate the detection of other Ad types [8, 9] and we cannot exclude the possibility that, by using these advanced primer pairs, we would have been able to detect more positives among the 40 PCR-negative samples. Finally, it is of note that our conclusions on type distribution were based on data available for only a limited number of adenovirus-positive samples.

In conclusion, our findings suggest that enteric HAds are important pathogens in pediatric gastroenteritis in Hungary. Studies with larger sample sets are needed to assess the reliability of individual HAd antigen detection tests and to provide a better assessment of the public health importance of enteric HAds.

Acknowledgments

The study was supported by the Hungarian Scientific Research Fund (OTKA, T049020). K.B. is a recipient of the Bolyai János fellowship.

Copyright information

© Springer-Verlag 2009