Detection of respiratory syncytial virus, parainfluenzavirus 3, adenovirus and rhinovirus sequences in respiratory tract of infants by polymerase chain reaction and hybridization

Background: Immunofluorescence assay (IFA) of viral antigens in nasal aspirates is largely used for the diagnosis of respiratory syncytial virus (RSV), parainfluenzavirus (PIV) type 3 and adenovirus (AdV) infections, whilst rhinovirus (RV) are detected by virus isolation technique (VIT) only. Using the two techniques, IFA and VIT, a significant number of specimens remain negative in spite of clinical and epidemiological presumptions of viral infection.

Objectives and study design: The polymerase chain reaction (PCR) should improve the sensitivity of viral detection in clinical specimens. From October 1995 to March 1996, 277 nasal aspirates from hospitalized infants were tested simultaneously by IFA, VIT, polymerase chain reaction and hybridization with a DNA enzyme immunoassay (PCR-EIA) for RSV, PIV-3, AdV and RV.

Results: RSV were detected in 177 (64%) samples, PIV-3 in 23 (8%), RV in 40 (14%), and AdV in 30 (10%). PCR-EIA detected RSV in more samples 173 (62%) than $\text{IFA}\text{VIT}$: 109 (39%) (P < 10⁻⁷). In most cases (79%), RSV-infected infants had lower respiratory tract disease, and routine and PCR techniques were positive. Out of the 23 PIV-3 infections, 12 were $\text{IFA}\text{VIT}\text{-}$ and PCR-EIA-positive, and 11 $\text{IFA}\text{VIT}\text{-}\text{negative}$ and PCR-EIA-positive. For RV, 35 (87%) specimens were PCR EIA-positive and 11 (27%) culture-positive; for AdV 30 samples were PCR-EIA-positive and four were culture-positive. Simultaneous viral infections were revealed in a significantly higher proportion than in conventional techniques: 18% ($\text{50}\text{277}$) versus 2.5% ($\text{7}\text{277}$); P < 10⁻⁷. One RSV infection in four was associated with the presence of another virus, mainly PIV-3 (16 cases) and AdV (13 cases).

Conclusions: PCR-EIA detects more positive-specimens than $\text{IFA}\text{VIT}$, 1.5 times more for RSV, 1.9 for PIV-3, 4 for RV and 10 for AdV, respectively. This increased sensitivity of viral detection by PCR-EIA compared to the $\text{IFA}\text{VIT}$ could suggest that samples containing low levels of virus are missed by routine methods $\text{IFA}\text{VIT}$, and consequently, RSV or PIV-3, and above all RV or AdV are overlooked as agents of respiratory diseases. However, apart from the fact that the economic and convenient aspects of virus diagnostic cannot be missed, it is difficult to answer the following questions: what is the meaning of the detection of a viral sequences in nasal aspirates of infants. or may PCR have detected virus in patients who would not developed disease?