• viral respiratory infections

Respiratory viruses cause many diseases, from mild to severe illnesses, and contribute significantly to morbidity and mortality worldwide. Different viruses can establish respiratory tract infections, they belong to the Orthomyxoviridae, Coronaviridae, Picornaviridae, Paramyxoviridae, Adenoviridae, Parvoviridae and in immunocompromised patients Herpesviridae families. Altogether, the etiologic diagnosis of respiratory viral infections has been underestimated so far. In patients with predisposing conditions the outcome of these infections can be more severe, sometimes requiring hospitalization, even in intensive care units (ICU), because of the development of pneumonia and acute respiratory distress syndrome (ARDS). For these patients, in particular, rapid diagnosis is essential. In addition, in ICU patients the significance of the detection of some members of Herpesviridae family, like HSV1, CMV and EBV, is controversial. The aim of this study was to clarify the prevalence of respiratory viruses and herpesviruses, and their role in ICU patients. Large part of this study was devoted to the development of diagnostic assays able to accurately characterise respiratory viruses quickly and at the lowest costs. Viral detection was performed in both upper and lower respiratory samples in order to compare the viral populations in these two compartments and, possibly, to draw informations concerning the role of the infection in severe cases.

Four duplex RT real-time PCRs, using EvaGreen fluorescent dye, were developed to identify and characterize the main respiratory RNA viruses directly from clinical samples. A duplex was performed to detect influenza A and influenza B viruses; a second duplex was performed to detect PIVs belonging to Respirovirus genera (PIV1 and PIV3) and RSV; the viruses target for the third duplex PCR were PIV type 2 and hMPV; and a fourth duplex was performed to detect CoV I and EV/RV.
The results of the duplex real-time PCRs were confirmed by sequencing positive samples and by comparison with other assays, including commercial, validated, assays, which gave similar results. A total of 156 clinical samples from upper and lower respiratory tract of 58 adult patients hospitalized in ICU were analysed. In 80% of positive adult patients influenza A viruses were detected, in 8% influenza B viruses and rhinovirus/enterovirus and in 4% metapneumoviruses. In particular in all ICU adult patients positive for influenza A or B viruses, the virus was demonstrated in both upper or lower respiratory tract samples. For an adult patient positive for RV/EV, the virus was detected in both samples; while for another only the upper respiratory tract sample was positive. Only one upper respiratory tract sample from one patient was positive for hMPV. The results obtained in this study were in agreement with other published studies that showed influenza virus as the most common virus detected in ICU patients, followed by rhinoviruses.

Furthermore, three nested PCRs were developed to detect herpesviruses (CMV, EBV and HSV1). To understand better the role of these viruses in ICU patients all herpesviruses positive clinical samples obtained by nested-PCRs were further analyzed by quantitative real-time PCRs.
Altogether, 37 of 58 (about 64%) patients were positive for one or more herpesviruses.
EBV was detected in 25 patients of 58, either as single or mixed infection; CMV and HSV1 were detected in 15 patients either as single or mixed infection. Mixed infections were not rare.
In general, for EBV viral load in TS samples was higher than in BAL samples. These results could suggest a possible viral contamination of the lower respiratory tract from mouth or throat or both. In one case EBV DNA was detected in the BAL only in two successive specimens at a low viral load (103-104). Further monitoring should be performed to better understand these data.
In this study CMV was detected as single infection only in two patients. Regarding CMV association with pneumonia, the average values of viral load reported in literature vary; however a viral load in BAL samples, between 4,6x104 and 5x105 copies number/ml, has been proposed as a threshold for the diagnosis of pneumonia. None of the patients analysed in this study had a viral load within this range only in BAL sample.
HSV1 was detected in 15 patients of 58 as single or mixed infection. In general, all patients HSV1 positive had high viral load in TS and in BAL samples (average value 106). In one patient only the detection of a high viral load in the BAL in absence of viral DNA in the TS could suggest an involvement of this virus in the lower respiratory tract disease.

The assays described could be particularly useful to screen a large number of patients for epidemiological studies and to assess the prevalence in the lower and upper respiratory tract of ICU patients with, regarding CMV, EBV and HSV1, the ultimate goal to understand the clinical significance of this phenomenon.

The possible contribute of the use of the NGS to the knowledges of the viruses involved in upper and lower respiratory tract infections was also studied preliminary in a small number of ICU patients. Nextera-XT protocol to MiSeq platform has been used.