It will depend on your assay- if you are doing TBARS, for example, and your measurement tool is a spectrophotometer, you need to be aware of the absorption peaks of heme.

If that is your assay, you might want to consult the following paper:

Proc Soc Exp Biol Med. 1993 Apr;202(4):407-19.

Spectrophotometric measurement of plasma 2-thiobarbituric acid-reactive substances in the presence of hemoglobin and bilirubin interference.

Pyles LA, Stejskal EJ, Einzig S.

as a general rule if you release hemoglobin or any other hemoprotein from their natural cellular environment, you increase the risk of artifactually generating oxidation products by allowing them to redox cycle. This is especially true if you are measuring lipid per oxidation. I would make sure to add either an inhibitor of redox cycling(a very good one is paracetamol) and /or antioxidants (we typically use butylhydroxytoluene and triphenylphosphine in our lab). Another good additive if it does not interfere with your assay is to add EDTA to chelate any free iron that could be released by Hb denaturation.

To avoid the LPO of erythrocyte, it is always better to measure blood serum instead of plasma by TBARS or other methods. Or, you can follow that Heidi O'Neill or Alaaeldin Hamza said.

Potentiometric method gives the possibility to investigate red cells. See: New Electrochemical Method of Determining Blood and Blood Fractions Antioxidant Activity

Kh. Z. Brainina,a,b* L. V. Alyoshina,a E. L. Gerasimova,a,b Ya. E. Kazakov,b A. V. Ivanova,a Ya. B. Beykin,c S. V. Belyaeva, c T. I. Usatova, M. Ya. Khodos. Electroanalysis, 2009, 21, No.3-5, 618-624.

Thank you everyone for your answers. Apparently there is more than one problem in order to assess oxidative stress in erythrocytes. First of all, there is the problem linked to spectrophotometric absorbance interference, but calculations should resolve this problem. i'm more concerned by the oxidant power of hemoproteins, but thanks to the answer of Olivier Boutaud, this shouldn't be a problem for MDA determination. However antioxidant defences (total capacity, like TAC measurements) couldn't be assessed if we add antioxidant in our samples... Someone has an idea?

I don't know if it can help:

http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1448694/

In comet assay, 10% DMSO is added to reduce the oxidative stress associated with red cells leackage of hb.

Please, see a paper Electroanalysis 2009, issue 3-5, p. 618-624

New Electrochemical Method of Blood and Blood Fractions Antioxidant Activity Determination

Kh. Z. Brainina,a,b* L. V. Alyoshina,a E. L. Gerasimova,a,b Ya. E. Kazakov,b A. V. Ivanova,a Ya. B. Beykin,c S. V. Belyaeva, c T. I. Usatova,c M. Ya. Khodosa