We conduct a lot of LC-MS-MS small molecule drug analysis of bioanalytical samples. I have observed that sometimes when protein precipitate extracts from whole blood are diluted (with blank matrix) that the calculated concentration of the drug in the diluted samples can be 3-fold greater than the same sample analyzed with no dilution. Both the non-diluted and diluted samples (e.g. 30-fold dilution) fall within the range of the calibration curve. Has anyone else observed this and do you know the cause?

Thanks!

The only unique thing here is that you will still have protein in your extract which may "hold on" to part of the analyte and prevent a normal chromatography. The equilibrium between analyte in the solvent part and that attached to the protein will be shifted toward the solvent part, hence the increase. I would think this to be the more unlikely cause, though. Maybe you inject different volumes for the two types of samples? How do you dilute? Also, sometimes a strong dilution like this results in a very small peak (if it can not be compensated by injection volume) whith accompanying high determination uncertainty. I have often seen this uncertainty to "prefer" a positive deviation, especially if the integration is done by hand.

Thanks for your response! I agree that it might be the result of the dilution resulting in a small peak because the resulting peak is near the lower limit of quantitation. Even though this small diluted peak is within the calibration range, I think it might be biased upwards by carryover that gets magnified by the dilution multiple. The injection volumes are the same.