Reply to comments on “Increased the circulating levels of malondialdehyde in patients with obstructive sleep apnea: a systematic review and meta-analysis”

Abstract

We read the comment by Dr. Jouyban on our published article entitled: “Increased the circulating levels of malondialdehyde in patients with obstructive sleep apnea: a systematic review and meta-analysis” [1]. Several issues have been raised and this letter was written to address them. There is concern about the heterogeneity and possible source of heterogeneity. As we stated in the results, there was a significant heterogeneity, and accordingly, standardized mean difference (SMD) was used, and subgroup analyses with possible confounding factors were performed. There are several factors that were used for subgroup analysis such as age, body mass index (BMI), method for measuring malondialdehyde (MDA), and disease severity. Jouyban believed that detection method should be used for subgroup analysis; however, it seems likely that the reaction has a higher importance than the detection method such colorimetric, UV, spectroflourimetric, and/or chromatographic. Therefore, we used a classification for subgroup analysis according to the measuring method of MDA. On the other hand, in the table provided by Jouyban, UV was mentioned as a measurement method in two studies, while in those studies, the colorimetric measurement was performed, and the product has an absorbance in 532–535 nm which is not in UV area. Furthermore, Jouyban categorized methods of three kits for measuring MDA, while, kit 1 and 3 used the TBARS protocol, and kit 2 used MDPI to measure MDA. Chen et al. used the thiobarbituric acid reactive substances (TBARS) method, and the product was detected using flourimetry. According to the protocol of kit 1 and 3, TBARS is the method of measurement, and we categorized all studies that used TBARS in one specific class. Indeed, kit 1 and 3, colorimetric, flourimetric, and UV categories that provided by Jouyban used TBARS. In addition, there are two different methods (HPLC and MDPI) that we used in the subgroup analysis. It is also worthy to note that each study used the same method for measuring MDA in control and obstructive sleep apnea (OSA) groups, and the SMD was compared between the studies; therefore, methodological differences between the studies could not be the main source of heterogeneity. It would be more reasonable to search some differences between cases and controls in each study as the source of heterogeneity, such as age and BMI, as we considered them as confounding factors. In addition, there were not enough data to perform subgroup analysis according to storage time. Jouyban mentioned that the study by Yardim-Akaydin has more subgroups; there are 117 patients with OSA categorized into three subgroups: mild (n = 28), moderate (n = 30), and severe (n = 58). These subgroups were divided into other subgroups, and if we considered all the subgroups, the sample size of the patients needed to reach statistical significance would be unsupportable in a meta-analysis. In Table 2 of the systematic review and meta-analysis, there are typographical errors in unit presentation, as mentioned by Dr. Jouyban. The units for MDA in Table 1 of the systematic review and meta-analysis needed to be changed for studies by Jurado-Ga ́mez et al., from mM to μM; Wysocka et al., from mMol/L to μMol/L; Li et al., from nMol/L to μM/L; and Asker et al., form Hmol/L to μM. This reply refers to the comment available online at https:// doi. org/ 10. 1007/ s1132502102467-0.