[Quantifiable method of polyunsaturated fatty acids in human plasma by optimized gas chromatography].

ABSTRACT

OBJECTIVE: To optimize the quantifiable method for determining polyunsaturated fatty acids in human plasma by gas chromatography(GC). METHODS: Plasma was added with heptadecanoic acid methyl ester internal standard, and the total plasma lipids was extracted with methanol and n-hexane under the condition of ultrasonic water bath. The sulfuric acid methanol was used for methyl esterification. After centrifugation, the supernatant was filtered with a membrane and dried by nitrogen, and then re-dissolved by n-hexane. HP-88 GC column(30 m×0.25 mm, 0.2 µm) was used to separate linoleic acid(LA), α-linolenic acid(α-ALA), arachidonic acid(AA) and docosahexaenoic acid(DHA) with programmed temperature, and internal standard method was used to draw standard curve for quantitative analysis. RESULTS: Ultrasonic water bath at 40 ℃ for 20 min could realize a higher total lipid extraction rate in a shorter time. The experimental concentration of four fatty acids range showed a good linear relation with the peak area ratio, correlation coefficient(r>0.995). The rate of recovery of four fatty acids were between 84.05%-101.23% with relative standard deviation(RSD) of 1.21%-6.77%, and the RSD of precision, within-day stability and day to day stability were 0.01%-0.04% and 0.02%-0.07%, respectively. No significant differences in the concentration of α-ALA, AA and DHA were observed between plasma and serum. The concentration(M(P25, P75)) of LA was higher in serum [93.38(79.18, 116.41) µg/mL]compared with plasma [72.12(50.93, 101.13) µg/mL], and the difference were significant(P<0.05). CONCLUSION: This method has a higher rate of total plasma lipids with shorter time, and the internal standard-standard curve method could eliminate the influence of the sample amount on the result. The determination of four polyunsaturated fatty acids was accurate, which could be used for the determination of plasma polyunsaturated fatty acids content in large sample populations.