Comments concerning ‘Meta-analysis of the association between vitiligo and the level of superoxide dismutase or malondialdehyde’
Abolghasem Jouyban, Maryam Khoubnasabjafari · Clinical and Experimental Dermatology · 2018
We read with interest the recent article published in Clinical and Experimental Dermatology by Shi et al.,1 who reported the results of a meta‐analysis on the association between vitiligo and levels of superoxide dismutase (SOD) and malondialdehyde (MDA) in biological samples. They summarized the levels of SOD and MDA collected from 16 different research groups, and used metaregression analyses of ethnicity, country and sample sources for exploring the possible cause of heterogenicity. Shi et al. mentioned different laboratory analyses as one of the possible source of variations, and also reported MDA as a specific indicator of oxidative stress. Their work provides a brief review on the present status of variations of SOD and MDA in vitiligo, and we do not aim to criticize their valuable work, but rather to focus on one of the most important sources of variations, which was not mentioned, namely, the reliability of the analytical method employed in determination of the analyte of interest. Our earlier observations on MDA levels in serum2 and exhaled breath condensate3 revealed that there are some problems with the validation data of analytical methods, and consequently the obtained MDA results using such methods are not reliable. A comprehensive literature review4 confirmed our observations on healthy controls, in whom the MDA levels varied from 320 to 53 797nmol/mL, which is a very wide range for plasma samples. Detailed investigations revealed that most of research groups used a simple spectrophotometric or spectrofluorimetric analysis for MDA after derivatization with thiobarbitoric acid without revalidation tests, which was also the case for data used in the analysis of Shi et al. (Table 1). Simplicity and ease of operation is the main reason for the very wide application of spectroscopic methods in clinical investigations. However, it is possible that this simplicity results in invalid and unreliable data to be used in clinical practice, and ignoring this point may mislead research groups and cause some discrepancies in clinical findings. In addition, the spectroscopic methods possess poor selectivity, especially in a complex matrix such as serum or plasma. In a number of research works, the mixture of derivatization samples was separated using a liquid chromatography system and detected by an ultraviolet or fluorescence detector. The use of a separation system resolves some of the validation problems associated with the selectivity of spectroscopic methods; however, these methods still have low reproducibility and repeatability. The reasons for this poor validation of data originate from the high chemical reactivity of MDA, the strong acidic conditions and high temperature of the derivatization solutions, cross‐reactivity of MDA with other biochemicals in the biological samples, and the low stability of standard solutions of MDA and derivatization reagents, which were comprehensively discussed in a recent publication.4 Analytical methods used for malondialdehyde determination in biological samples Hb, haemoglobin; MDA, malondialdehyde; UV, ultraviolet. *Reference numbers are those in the original paper by Shi et al.,1 in which further details can be found; †original report was not accessible. Analytical methods used for malondialdehyde determination in biological samples Hb, haemoglobin; MDA, malondialdehyde; UV, ultraviolet. *Reference numbers are those in the original paper by Shi et al.,1 in which further details can be found; †original report was not accessible. We strongly recommend that researchers in the field should test the validity of the analytical techniques used in their determination before reporting the data. Full details of validation procedures for analytical methods can be found in the literature.5 The validity and reliability of biochemical data should be double‐checked before any meta‐analysis. Finally, MDA should not be considered as a specific indicator of oxidative stress, as there is some doubts about its reliability. Conflict of interest: the authors declare that they have no conflicts of interest.