Resampling model of the complement component functional assay: Are we measuring what we think we are measuring?

Abstract

Introduction: Complement component functional assay employs mix of the patient serum and serum depleted by the component of interest. To verify that such assay does actually measure functional activity of that component, authors introduced the simple resampling model of the functional test.

Methods: Virtual experiment of the functional test model consists of random sampling of component concentrations for depleted and patient sera. It was assumed that all complement components have Gaussian distribution, and the lowest component concentration determines the outcome of the assay. This outcome was compared with the chosen concentration of the target component. The goal was to evaluate how often and under which conditions our virtual experiment results were not determined by the component of interest.

Results: We could only underestimate functional activity of the complement component of interest. Underestimation could happen only when functional activity of the component of interest is above the population average multiplied by the ratio of depleted to patient serum volumes.

Conclusions: The chance to underestimate functional activity of the component is low for the real-life component distribution for the assay volume ratio 2:1, and practically absent for higher assay ratios.