A solution to small samples: A method validation study for sputum assay

Background: Induced sputum differential leukocyte counts as a relatively non-invasive and repeatable measure of airway inflammation have been used to phenotype patients with asthma. However, the current gold standard processing method (1) is not suitable for sputum plugs Aim: To compare our modified (small sample) sputum processing method (Method B) with the current gold standard (Method A). Methods: Sputum samples of sufficient quantity from 13 asthmatic patients (age-range 18-68years) were processed simultaneously using both methods. The median sample size used for Method B was 52mg (range 19-82) of sputum plugs. The difference in eosinophil %age between the methods was plotted against the mean result as proposed by Bland et al. (2). Results: The median %age eosinophils obtained using Method A was 4.75% (range 1.0-37.0) and 3.5% (range 0.5-32.6) using Method B. This difference was not statistically significant. There was no evidence of a systematic error. The bias and precision are shown in Figure 1 . Conclusions: Our modified method for small samples gives similar results to the gold standard method described. This technique is particularly useful for paediatric studies. References: 1) Pavord, I. et al . Thorax 1997; 52:498 2) Bland, J. M. et al . Stat Methods Med Res 1999; 8:135.