Of the 99 boiled samples, 82were positive in the ImmuView® assay,
and 61 were positive in the BinaxNOW® assay, giving sensitivities of
82.8% and 61.6% respectively. Of the 71 samples tested directly, 60
were positive in the ImmuView® test, and 55 were positive in the
BinaxNOW® assay, giving sensitivities of 84.5% and 77.5% respectively.
The difference in sensitivity observed for the two different kits used
for the boiled samples was statistically significantly different, with the
ImmuView® kit showing the best sensitivity. For the un-boiled samples
the observed differencewas not significant. The difference in sensitivity
between the two kits in the two groups might indicate that boiling decreases the sensitivity of the BinaxNOW® assay but not of the
ImmuView® assay, since BinaxNOW® performs relatively worse on
the boiled samples compared to the ImmuView® assay. Another explanation
can be that the 71 culture positive patients might have a higher
overall antigen level, thus the lower sensitivity observed for the boiled
samples is solely a consequence of lower sensitivity of the BinaxNOW®
assay. Ten of the positive L. pneumophila samples were tested in the
ImmuView® assay both before and after boiling with similar results in
each case, indicating that boiling did not interfere with the Legionella results
of the ImmuView® assay. Unfortunately, no material was left for
the same investigation of the S. pneumoniae samples.