Thiobarbituric acid reactive substances as a biomarker for coronary heart disease.

to the TBARS assay. In particular, different aldehydes can be formed in the lipid peroxidation process and aldehydes other than MDA may form chromogenic complexes with TBA, which exhibit similar absorbance or emission wavelengths. Several other biological compounds, including sugars, amino acids and bilirubin, are also reactive towards TBA and form TBA reactive materials which confound the measurement by the conventional TBARS assay. More importantly, MDA is also a by-product of cyclooxygenase activity in platelets; therefore, the measurement of MDA in serum may lead to an over-estimation due to the formation of MDA formation ex-vivo, as platelet activation is required for the formation of serum. When using serum as the sample matrix for analysis of oxidative stress, it is worth considering the issue of specificity for this particular reason. For example, in the PREVENT trial, the authors attempted to address the issue of specificity of TBARS measurement in serum by subjecting the samples to a two-stage extraction process. Serum samples were first treated in the classic thermoreaction TBARS assay, followed by separation of TBARS using reverse phase high performance liquid chromatography (HPLC) to exclude other TBA reactive materials which may have the same absorbance/emission wavelengths as the MDA-TBA2 adduct. This was achieved by coupling a spectrophotometric detector at 532 nm and a fluorescence detector (excitation=515 nm, emission=553 nm) on a 150 mm×4.6 mm C18 column with 5 μm particle size. The mobile phase consisted of 80% phosphate buffer (10 nm, pH 5.8) with 20% methanol and the flow rate set to 1 mL/min. However, the additional HPLC separation step is time consuming and it is logistically challenging to apply it to clinical studies involving a large number of patients. One alternative to improve the specificity of the TBARS assay in clinical samples may be to use plasma instead of serum samples, as this can minimise the interference of other aldehydes produced ex-vivo by platelets via the cyclooxygenase pathway, as discussed earlier. Further, the use of a chelating agent (such as EDTA) as an anticoagulant in the preparation of plasma may further reduce the variability of the TBARS assay due to the iron content of reagents used in the analysis. To the Editor