Enterovirus D68 detection in respiratory specimens: Association with severe disease

Abstract

Molecular techniques increased the number of documented respiratory infections. In a substantial number of cases the causative agent remains undetected. Since August 2014, an increase in Enterovirus(EV)-D68 infections was reported. We aimed to investigate epidemiology and clinical relevance of EV-D68. From June to December 2014 and from September to December 2015, 803 and 847 respiratory specimens, respectively, were tested for respiratory viruses with a multiplex RT-PCR. This multiplex RT-PCR does not detect EV-D68. Therefore, 457 (2014) and 343 (2015) specimens with negative results were submitted to an EV-specific-RT-PCR. EV-positive specimens were tested with an EV-D68-specific-RT-PCR and genotyped. Eleven specimens of 2014 tested positive in the EV-specific-RT-PCR and of these seven were positive in the EV-D68-specific-RT-PCR. Typing confirmed these as EV-D68. Median age of EV-D68-positive patients was 3 years (1 month-91 years). Common symptoms included fever (n = 6, 86%), respiratory distress (n = 5, 71%), and cough (n = 4, 57%). All EV-D68-positive patients were admitted to hospital, 4 (57%) were admitted to intensive care units and 6 (86%) received oxygen. One patient suffered from acute flaccid paralysis. Seven specimens of 2015 were positive in the EV-specific-RT-PCR but negative in the EV-D68-specific-RT-PCR. In conclusion, use of an EV-specific-RT-PCR allowed us to detect EV-D68 circulation in autumn 2014 that was not detected by the multiplex RT-PCR and was associated with severe disease.

Keywords: acute flaccid paralysis; human enterovirus 68; myelitis; respiratory tract infection; sequencing.

Figure 1

Flow diagram for patients and specimens included in the study. EV: enterovirus. *one specimen was untypeable

Figure 2

Seasonal distribution of EV‐D68 and HRV positive specimens. The histogram represents the number of specimens tested in the EV‐specific‐RT‐PCR and that gave positive (black and hatched bars) or negative results (white bars). The result of molecular typing of the positive specimens is indicated (HRV, black bars; EV‐D68, hatched bars)

Figure 3

Phylogenetic analysis of the nucleotide sequences of the partial VP2/VP4 region of Enterovirus strains detected in this study. A neighbor‐joining tree of VP2/VP4 sequences was constructed by using MEGA6. Coxsackievirus A21 was used as outgroup. Sequences from this study are indicated with a black circle and numbered as in Table 1. The percentage of bootstraps (out of 1000) that supports the corresponding clade is shown if higher than 70%. HRV‐A, ‐B, and ‐C, Coxsackievirus A21, human echoviruses E6 and E9, and several EV‐D68 reference sequences were included. The sequence that could be obtained for the specimen of patient 4 was significantly shorter and therefore not included in the phylogenetic analysis

Figure 4

Phylogenetic analysis of the nucleotide sequences of the partial VP1 region of Enterovirus strains detected in this study. A neighbor‐joining tree of partial VP1 sequences was constructed by using MEGA6. Coxsackievirus A21 was used as outgroup. Sequences from this study are indicated with a black circle and numbered as in Table 1. The percentage of bootstraps (out of 1000) that supports the corresponding clade is shown if higher than 70%. HRV‐A, ‐B and, ‐C, Coxsackievirus A21 and several EV‐D68 reference sequences were included. The sequence that could be obtained for the specimen of patient 2 was significantly shorter and therefore, not included in the phylogenetic analysis