Background: Few data exist to assess the sensitivity of different specimen types for viral detection during the course of influenza virus infection.

Objectives: This study assessed the relationships between quantitative influenza A virus replication and antigen detectability by the enzyme immunosorbent assay (EIA) Directigen Flu A in different type of samples during experimental human infection.

Study design: Fourteen volunteers were inoculated with influenza A virus A/Texas/36/91 (H1N1). Four specimens types were collected in sequence for quantitative isolation in cell culture and antigen testing from days 1 to 8 after inoculation.

Results: Seventy-one (63%) of nasopharyngeal wash specimens were culture positive, compared to 51 (46%) of throat gargles, 51 (46%) of nasal swabs, and 27 (24%) of throat swabs. All subjects shed virus in their nasopharyngeal wash at least one day and 86% of subjects had a positive nasopharyngeal wash culture on day 2 after inoculation. The mean viral titers were highest on day 2 post inoculation for all specimen types and averaged 3.6 log10 TCID50/ml for nasal washes, 1.2 log10 TCID50/ml for throat gargles, 1.8 log10 TCID50/ml for the nasopharyngeal swabs, and 0.6 log10 TCID50/ml for the throat swabs. Mean viral titers in the nasal washes were significantly different (P<0.05) compared to other specimen types. The peak of sensitivity of EIA (compared to culture) was the second day after inoculation. Nasopharyngeal and throat swab results were combined for this analysis and considered positive by culture if positive in either or both samples. Thus, on day 2 the number of EIA positive samples relative to the number culture positive was 9/12 (75%) for nasopharyngeal wash specimens, 2/9 (22%) for throat gargles, and 7/11 (64%) for the combined throat and nasal swabs specimens.

Conclusions: Nasopharyngeal washes are the most sensitive sample type detecting influenza A virus in adults. For rapid diagnosis the Directigen Flu A is an alternative with a sensitivity compared to culture ranging between 64 and 78% if performed on nasopharyngeal specimens on day two or three after experimental infection in adults. However, if performed on other specimens or later in the course of infection the sensitivity is lower.