Increased plasma malondialdehyde levels in glomerular disease as determined by a fully validated HPLC method

Background. Reactive oxygen species and particularly free radical induced l ipid peroxidative tissue damage have been implicated in the pathogenesis of various renal diseases. Lipid peroxidation is assessed indirectly by the m easurement of secondary products, such as malondialdehyde (MDA), using the widely employed thiobarbituric acid reactive substances (TBARS) method. How ever, this method lacks sensitivity and specificity. We have therefore deve loped and validated an HPLC thigh-performance liquid chromatography) method for measurement of MDA and applied this to a variety of plasma samples in renal patients. Methods. The optimized method involves antioxidant treatment of the plasma sample, followed by a protein precipitation step using trichloroacetic acid , acid hydrolysis and formation of an MDA thiobarbituric acid complex. The MDA-(TBA)(2) adduct is separated from other interfering compounds by C-18 r everse-phase HPLC techniques, with visible detection at 532 nm. Results, The assay was linear over the ranges 0.25-1.0 mu M MDA and the det ection limit was 0.06 mu M MDA. Within-run precision was <4.5% and between- run precision was <10.0%. MDA plasma concentrations (mean +/- SD) were high er in ESRF diabetic patients (0.32 +/- 0.14 mu M, n = 20), non-diabetic ESR F patients (0.32 +/- 0.09 mu M, n = 20), and CRF patients (0.14 +/- 0.06 mu M, n = 40) compared to healthy controls (0.11 +/- 0.03 mu M, n = 40, (P < 0.001, P < 0.001 and P = 0.008). Levels were similar in healthy controls wi th normal renal function and transplanted patients (0.12 +/- 0.03 mu M MDA, n = 40), (P = NS). No correlation was observed between MDA and creatinine levels (r(2) = 0.05, n = 80), which suggests that MDA does not correlate wi th the degree of renal impairment. We matched CRF patients with glomerular and nonglomerular causes of renal failure for creatinine levels and found t hat MDA levels were higher in patients with glomerulonephritis (0.16 +/- 0. 06 mu M) than in those with CRF from non-glomerular causes (0.12 +/- 0.04 m u M, P = 0.002). Conclusions. We have introduced a reliable and sensitive HPLC technique to enhance the specificity of MDA-(TBA)(2) measurement, with a significant imp rovement in HPLC column life. Using this method, picomole quantities of MDA can be detected in plasma. We have shown that MDA levels are significantly raised in patients with CRF due to glomerulonephritis, regardless of serum creatinine, which suggests that there is oxidative injury independent of a ny possible MDA retention due to renal impairment.