N. Okazaki et al., Deconvolution of chemiluminescent emission curve associated with phagocytosis into three logarithmic normal distributions, TOH J EX ME, 185(4), 1998, pp. 271-280
The chemiluminescent emission reaction dependence on the activity of phagoc
ytosis is well known. However, this method is not used to diagnostically in
clinical assessment because the relationship between phagocytizing activit
y and chemiluminescent intensity has not been clearly established. Therefor
e, we attempted to analyze quantitatively the chemiluminescent emission cur
ve by the phagocytosis of leukocytes. Mathematical assessment of the emissi
on curve with respect to time was performed by fitting the curve to several
regression models using the unweighed non-linear least squares method. A t
riple logarithmic normal distribution model provided a reasonable goodness
of fit to the measured emission curve. The first component, about 5% of the
calculated total counts, was assumed to arise from monocytes activity, the
second component, about 20% from eosinocytes activity and the third compon
ent, up to 75%, from neutrophils activity. This method seems promising as a
means for assaying whole blood without the need for pretreatment and for t
he providing a valid index that is independent of the technical differences
between laboratories.