Human parechovirus type 3 central nervous system infections in Israeli infants

Abstract

Introduction

Human parechoviruses (HPeV) have been recognized as the causative agents of central nervous system (CNS) infection of infants and young children in different parts of the world. The role of HPeV in CNS infection of Israeli infants and children is unknown.

Objectives

To assess the detection rate of HPeV in enterovirus RT-PCR-negative cerebrospinal fluid (CSF) samples obtained during the years 2007–2009 from children 0–5 years old with suspected CNS infection or from very young infants with unexplained fever in four medical centers in Israel.

Study design

A total of 367 CSF samples were retrospectively tested for the presence of HPeV RNA using nested RT-PCR assay. Positive samples were further typed on the basis of molecular sequencing. Retrospective analysis of the medical charts was performed.

Results

HPeV3 RNA was detected in CSF obtained between May and September 2008 in 13 patients, all of whom were <3 months old (3.5% of all CSFs; 11.3% of all infants <3 months in 2008). The HPeV-positive CSF samples were without pleocytosis. All HPeV3-positive patients recovered without obvious short term sequelae.

Conclusion

HPeV infection could play an important role in summertime febrile/CNS illness in young infants during specific years with high HPeV activity. PCR detection of parechoviral RNA in CSF should be included in the diagnostic evaluation of fever or CNS infection of neonates and very young infants. The rapid identification of HPeV in CSF could curtail unnecessary empirical antibiotic treatment and shorten hospital stay in selected patients.

Introduction

Human parechoviruses (HPeV) belong to the parechovirus genus in the Picornaviridae family [1]. Since its first description [2], HPeV3 has been reported from different areas of the world as a causative agent of a neonatal sepsis-like illness [3], [4] as well as central nervous system (CNS) infection in young children [5], [6], [7], [8], [9], [10], [11]. HPeV3 has been associated with significant morbidity [11], [12], [13], sequelae [2], [13] and even mortality [11], [14], [15]. In comparison with other genotypes of human parechovirus, HPeV3 seems to infect younger children, especially neonates [3], [4].

The goal of this study was to assess the prevalence of HPeV in CNS infection in Israeli infants and children in a multicenter retrospective study.

CSF specimens from four medical centers in three different regions of Israel – Jerusalem (east), Haifa (north, 2 medical centers) and metropolitan Tel Aviv (central-coastal) were selected. All CSF specimens were originally tested at the participating medical centers’ laboratories for enterovirus by RT-PCR and were negative. If PCR for herpes simplex virus and/or varicella zoster virus were performed these results were also necessarily negative in order to be included. CSF specimens were chosen from patients younger than 5 years old, during the years 2007–2009, except for one center (central-coastal) which only provided samples from 2008 to 2009. If the CSF specimen met the above criteria, clinical criteria for inclusion were checked. For neonates ≤1 month of age, all patients with fever were included. For infants older than one month and younger than one year, patients having suspected meningitis or encephalitis, or those with CSF pleocytosis, or fever without a well-documented alternative diagnosis, were included. For children between 1 and 5 years of age, patients with clinically suspected meningitis or encephalitis or with CSF pleocytosis were included. The study was approved by the participating hospitals’ ethical review committees.

RNA was extracted from stored CSF samples using the automated extractor NucliSENS^® easyMAG^® (Biomérieux). cDNA was generated from 10 μl of extracted or stored RNA using reverse transcriptase (MMLV, Roche Diagnostics, Mannheim, Germany) and a Random Hexamers primer (GE Helathcare UK LTD.). The RT reaction was carried out at 37 °C for 60 min, and was followed by 3 min incubation at 95 °C. cDNA (5 μl) was subjected to nested PCR using the following primer pairs: outer pair (sense) 5′-GGG TGG CAG ATG GCG TGC CAT AA-3′ and (antisense) 5′-CTR CGG GTA CCT TCT GGG CAT CC-3′; inner pair: (sense): 5′-YCA CAC AGC CAT CCT CTA GTA AG-3′ and (antisense) 5′-GTG GGC CTT ACA ACT AGT GTT TG-3′. Amplification conditions for the 1st PCR were 30 cycles of denaturation at 94 °C for 20 s, annealing at 50 °C for 25 s, and extension at 75 °C for 90 s. Amplification conditions for the 2nd PCR were 30 cycles of denaturation at 94 °C for 20 s, annealing at 55 °C for 25 s and extension at 75 °C for 90 s using 2 μl of the 1st PCR product. The 2nd PCR product (244 base pairs) was visualized after electrophoresis on a 3% agarose gel in Tris borate/EDTA buffer by staining with 1 μg/ml ethidium bromide and documented on Bio Imaging System (Renium).

cDNA (5 μl) was subjected to nested PCR using the following primer pairs: HPeV-out-s: GAY AAT GCY ATM TAY ACD ATD TGT GA and HPeV-out-as: ACH GTR AAR ATR TCH ACA TTS ATD G. HPeV-inner-s: TTY TCD ACH TGG ATG MGG AAR AC and HPeV-inner-as: DGG YCC ATC ATC YTG WGC TGA described in Harvala et al. [16]. Amplification conditions for the 1st PCR were 30 cycles of denaturation at 94 °C for 20 s, annealing at 50 °C for 25 s, and extension at 75 °C for 90 s. Amplification conditions for the 2nd PCR were 30 cycles of denaturation at 94 °C for 20 s, annealing at 55 °C for 25 s and extension at 75 °C for 90 s using 2 μl of the 1st PCR product. The 2nd PCR product was visualized after electrophoresis on a 3% agarose gel in Tris borate/EDTA buffer by staining with 1 μg/ml ethidium bromide and documented on Bio Imaging System (Renium).

An ABI PRISM BigDye Terminator Cycle Sequencing kit (Applied Biosystems, Foster City, CA, USA) was used to sequence 1 μl of amplified cDNA. Labeled products were purified using an ABI XTerminator Purification kit (Applied Biosystems) and the sequence determined on an ABI 3500 Genetic Analyzer (Applied Biosystems). The 240 nt sequence (amplicon with primer sequences removed) was genotyped by phylogenetic analysis using Sequencher (Gencodes, Anne Arbor MI, USA) to align the sequence with equivalent regions of parechovirus prototype genotypes, the Clustal X [17] program to prepare neighbor joining trees after bootstrapping data 1000 times, and the NJPlot program [18] to visualize the resultant phylogenetic trees. The new sequences reported in this paper have been assigned GenBank/EMBL/DDBJ accession numbers KC559740 to KC559752 (Fig. 1).

Statistical analysis was performed using PASW Statistics, Version 18, SPSS, Inc., Chicago. Continuous variables are presented as medians and interquartile ranges (IQR) while categorical variables are presented in percentages. Chi square and Mann Whitney tests, as appropriate, were used to compare parechovirus PCR positives and negatives. P < 0.05 was considered statistically significant.