How it works
The approach published by Matuszewski and colleagues compares three sets of samples: analyte in neat solvent (A), blank matrix extract spiked after extraction (B), and matrix spiked before extraction (C). B/A isolates the effect of co-eluting matrix on ionisation, C/B isolates extraction recovery, and C/A combines them.
A stable-isotope-labelled internal standard that co-elutes with the analyte usually corrects most of the matrix effect, which shows up as an IS-normalised matrix factor close to 1. ICH M10 evaluates matrix effect with at least six sources of matrix, with accuracy within ±15% and precision ≤ 15%.