Lowered Sample Rejection Rate for Rapid Direct Detection of Respiratory Viruses in Children by Use of a Flocked Swab and Nasopharyngeal Sampling

ABSTRACT

Background: Prior to the 2002-3 respiratory season, the rejection rate for DFA on polyester nasopharyngeal swabs (NPS) for respiratory virus detection in children in our institution was 11-15% per year. Based on a study showing equivalence of NPS and nasal swabs (NS) for the detection of respiratory viruses in children in an outpatient setting, and following staff education, NS became the recommended specimen type. Soon after institution of NS sampling, DFA rejection rate rose to 22-27% per year, with rates as high as 38% on some wards. High rates of rejection of NS were unaffected by repeated in-service education efforts to improve the yield by addressing technical aspects of obtaining these specimens. Objective: To study the effect of swab type (polyester vs. flocked) and specimen type (NS vs. NPS) on yield of respiratory viruses by direct fluorescent antibody (DFA) testing and viral isolation.

Methods: A pilot study (n=121) of flocked swabs (Copan) vs. polyester swabs (Puritan) for detection of respiratory viruses from NS of children with suspected viral infection was performed 11/05–01/06 on 3 wards with the highest rate of rejection of specimens for DFA due to insufficient cellular sampling. This was followed by a repeat study (n=146) on the same wards (09/06–02/07), using the same two swab types, but sampling the nasopharynx. Quantitation of cells by swab type (insuff (No cells few cells), 1+, 2+, 3+, 4+), DFA rejection rates, DFA positivity rates and yield of viral isolation was compared between the two groups. Each child was tested with both swab types, one per nare, in each study. All specimens were tested by DFA/culture for influenza A/B, parainfluenza 1,2,3, adenovirus, RSV and human metapneumovirus.

Results: The 2005/6 pilot study of flocked vs. polyester NS showed rejection rates of 32.8% (42/128) and 22.3% (27/121), respectively. The 06/07 study of flocked vs. polyester NPS revealed a dramatically reduced rejection rate of 11.0% (16/146) for flocked swabs vs. 28.1% (41/146) for polyester. For those specimens with adequate cells for analysis, 60% had 3-4+ cells using flocked swabs vs. 43% for polyester swabs. 32 NPS were DFA positive by both flocked and polyester swabs, 3 by flocked only and 2 by polyester only. 5/112 (4.5%) were culture pos/DFA neg, all RSV with 2+ cells. 8/144 (5.6%) were culture neg/DFA pos, 5 RSV, one each of influenza A, parainfluenza 1, 3.

Conclusions: In our cohort of children presenting/admitted to hospital, two factors appear to be necessary for optimal sampling of the upper airway for respiratory virus detection: nasopharyngeal sampling rather than nasal, and the use of flocked swabs rather than polyester swabs. Staff education and re-education is also critical but was not as important as swab type or sampling site.