Introduction

Selenium (Se) is an essential micronutrient for human health. It plays a key role in regulation of selenoprotein synthesis, selenoproteins being the means by which Se exerts its biological functions throughout the body [1, 2]. The selenoproteins, including glutathione peroxidase (GPX), iodothyronine deiodinases (IDD), and thioredoxin reductase (TrxR), participate in antioxidant defense and prevent oxidative stress damage [3, 4]. Se is hence a critical factor for optimal functioning of the immune system, as well as of cellular processes and metabolic cycling [5,6,7].

Oxidative stress, which is a disturbance in the balance between antioxidant defense and reactive oxidant species can cause toxic effects and damage cellular function and biomolecules (e.g., lipids), lipoproteins, DNA, and proteins [5, 8, 9]. Oxidative stress has thus been implicated in initiation and development of such diseases as insulin resistance, diabetes mellitus, cardiometabolic diseases, cancer, and neurodegenerative diseases [8,9,10,11].

A number of studies have shown that low Se status, and hence selenoprotein insufficiency, is associated with suboptimal health while also causing several diseases [5, 12]. More specifically, Se deficiency impairs immune function and selenoprotein synthesis, which results in immune cells being unable to protect against oxidative stress [13, 14]. Furthermore, animal studies have demonstrated that low levels of selenoproteins are related to hyperglycemia and insulin resistance [15].

Previous studies have demonstrated that Se supplementation may have beneficial effects in reducing the risk of chronic metabolic disorders, hyperglycemia, hyperlipidemia, and cancer [16,17,18]. Meanwhile, there is scientific evidence that dietary Se intake is inversely associated with all-cause mortality [18]. There is also an inverse relationship between the onset of cardiovascular diseases and blood Se levels [19, 20]. An increase of Se intake in Beijing, China, led to a reduction in the cardiovascular disease mortality rate from 1984 to 1990 [1]. Given the critical role of selenoproteins in antioxidant defense and regulation of circulating lipoproteins, vascular endothelial cells, and cardiac function, it is hypothesized that Se supplementation may be protective against cardiometabolic disease [5, 21].

Due to the numerous controversies surrounding Se and its impact on both health and disease, it is clear that further studies are required to provide solid evidence which may be used by both clinicians and policymakers for the setting up of relevant health programs and nutritional interventions [22, 23].

Although several controlled trials have investigated the effects of Se supplementation on antioxidant markers, its pooled effect on antioxidant markers is as yet unclear and controversial. The aim of this systematic review and meta-analysis was therefore to assess the effects of Se supplementation on antioxidant markers.

Methods

This systematic review followed the PRISMA guidelines [24], while the meta-analysis was performed according to a previously published protocol [23].

Data sources and search strategy

The PubMed, ISI/WOS, and Scopus databases were analyzed meticulously for relevant data, without time or language restriction, up to June 1, 2017 (Fig. 1). The reference lists of the retrieved reviews and meta-analyses were also screened to identify pertinent data. We excluded non-relevant articles and those with duplicate citation.

Fig. 1.
figure 1

Flowchart of the number of studies selected for the meta-analysis

Full size image

Study selection and data extraction

We included randomized controlled trials (RCTs) and quasi-RCT crossover studies, in which the control group received a placebo. Studies were considered regardless of dosage or duration of administration of antioxidants. Regarding the crossover studies, data from the first period were recorded and analyzed.

With regard to the RCTs, all studies among the different target groups of diabetic patients, polycystic ovarian syndrome (PCOS) patients, obese participants, and healthy subjects were included. Studies in which Se was applied as a single therapy or even combination therapy were considered as acceptable.

All clinical trials assessing the effect of Se supplementation on antioxidant markers, including oxidative stress index (OSI), antioxidant potency composite (APC) index, plasma malonaldehyde (MDA), total antioxidant capacity (TAC), antioxidant enzymes (superoxide dismutase (SOD), GPX, thiobarbituric acid-reactive substances (TBARS), catalase (CAT)), and total antioxidant plasma (TAP), were included.

Studies were excluded if they were (1) non-relevant; (2) a duplicate citation; (3) more than one paper from a specific study; (4) case reports, reviews, editorials, and studies performed on children (age < 18); and (5) concerned specific sub-group populations.

To establish inter-rater reliability, titles and abstracts were screened by two experts independently. Possible discrepancies were resolved by the main researcher. After assessment of relevancy, based on the inclusion/exclusion criteria, the full-text review determined the eligibility of papers.

The comprehensive recommended guidelines of the Consolidated Standards of Reporting Trials (CONSORT) 2010 25-item checklist [25], quality assessment, and data extraction and analysis were used, followed by two independent experts.

The main fields of relevant data were citation, type of study, study subjects, publication year, sample size, dose of supplementation, intervention group, control group, mean age of participants, outcome, intervention duration, follow-up duration, measurement interval, results, and effect size.

Data analysis

A meta-analysis was conducted for relevant antioxidant markers on which data were available from more than two individual studies. The mean change from baseline in antioxidant marker concentrations and standard deviation for both intervention and control groups was used to calculate the effect size. The effect size is presented as standardized mean difference (SMD). To pool the data, we used the random-effect model based on the result of heterogeneity among studies. A random-effect model was used if the Q-statistic for heterogeneity was significant at the level of 0.1 [26]. In other cases, the fixed-effect model was used [27]. The degree of heterogeneity was quantified, using I2 statistics, which is the estimation of the total variation across studies due to heterogeneity [28]. Possible sources of heterogeneity (such as quality assessment score, duration of intervention, study subjects, mean age of participants, dose of Se supplementation, type of study subjects, subjects with metabolic diseases, and healthy subjects) and sex ratio were explored by random-effect meta-regression analysis. Subgroup analyses were performed according to type of study population (type of disease) to identify between-study heterogeneity. Subgroup heterogeneity was evaluated using the fixed-effect model. A forest plot was used to present the results of the meta-analysis schematically. Publication bias was evaluated by visual inspection of the funnel plot and the Egger’s regression test. In addition, sensitivity analysis was conducted according to the study quality to test the robustness of the results.

The statistical analysis was carried out using Stata software, version 10 [25]. P value ≤ 0.05 was considered statistically significant.

Results

Search results and characteristics of included studies

From 3320 searched documents, following the refinement processing according to the inclusion/exclusion criteria, in total, 13 RCTs were included. These publications were dated from 2004 to 2016. Figure 1 shows the selection process of the articles included.

The meta-analysis included 13 studies (12 randomized controlled trials and 1 crossover). All studies had assigned a total of 2790 participants randomly to intervention and control groups. The age of the patients was from 10 to 85 years. Nine trials had recruited both men and women, while, in four other studies, only female subjects were enrolled [29,30,31,32]. Four trials used Se combined with other vitamins or minerals [10, 30, 33, 34] and nine trials used Se only as an oral supplement. The daily Se dose was 200 m/day in seven trials [29, 30, 32, 34,35,36,37]; one trial used 300 mg/day [33]; one RCT used 50 mg/day [10], and one study used 60 and 960 mg/day [31, 38]. In two studies, three different groups were given three different doses (100, 200, 300 mg/day); hence, they were considered as separate studies [15, 39]. Six trials were conducted in Iran, six in Europe, and one in the USA. All trials were placebo-controlled and all were double-blinded. The length of the intervention periods ranged from 42 days [25] to 180 days [15, 39] (Table 1).

Table 1. Characteristics of the included studies in the meta-analysis
Full size table

Meta-analysis

Four trials involving 228 participants in the Se or placebo groups reported the effect of Se supplementation on TAC. Three studies reported GSH (glutathione), five studies MDA, 10 Se, three GPX, two TBARS, and four adiponectin as outcome at baseline and follow-up (Table 2).

Table 2 Meta-analysis of effect of selenium supplementation on antioxidant profile
Full size table

The effect of Se supplementation on antioxidant levels is shown in Table 2. The pooled results indicated that Se supplementation significantly increased GPX levels (SMD = 0.54; 95% CI = 0.21–0.87). There was no significant heterogeneity between the three included trials (Q = 5.74, I2 = 65.2%, P = 0.057).

There was a significant improvement of antioxidant profile such as TAC [(SMD) = 0.39, 95% CI = (0.13, 0.66)] with obvious heterogeneity (Q = 10.27; P = 0,016; I2 = 70.8%), Se [(SMD) = 3.24, 95% CI = (3.1, 3.4)] with obvious heterogeneity (Q = 358.82; P = 0.0; I2 = 97.5%), and MDA [(SMD) = − 0.54, 95% CI = (− 0.78, − 0.30)] without significant heterogeneity (Q = 7.50; P = 0.11; I2 = 46.6%) through Se supplementation.

According to the meta-analysis, the intake of Se compared with placebo resulted in no significant improvement in GSH [(SMD) = 0.16, 95% CI = (− 0.12, 0.45)] and significant heterogeneity (Q = 10.94; P = 0.004; I2 = 81.7%), TBARS [(SMD) = − 0.12, 95% CI = (− 0.50, 0.25)], and adiponectin [(SMD) = 0.04, 95% CI = (− 0.09, 0.18)]. There was no significant evidence of heterogeneity for TBARS (Q = 0.14; P = 0.71; I2 = 0%) and adiponectin (Q = 0.40; P = 0.94; I2 = 0%), respectively. The pooled effect of Se supplementation on each antioxidant marker is depicted in Fig. 2.

Fig. 2.
figure 2

Forest plot of the effect of Se supplementation on antioxidant markers

Full size image

Quality assessment

Table 3 shows the quality of the included studies. Six studies were classified as high quality, with the CONSORT score higher than 30 [15, 29, 30, 36, 37, 39]; four studies were classified as medium quality, with the CONSORT score in the range of 25–29 [10, 31, 32, 35]; and three studies were classified as low quality, with the CONSORT score lower than 25 [33, 34, 38]. Randomization as a prerequisite for inclusion in this meta-analysis was conducted in 13 studies.

Table 3. Quality assessment of included studies according to the CONSORT checklist
Full size table

All 12 RTCs were double-blind, but only four studies were described as blinding [15, 32, 37, 39].

Meta-regression and subgroup analysis

The effect of influencing factors was analyzed using a random-effect meta-regression. There was no effect of influencing factors, such as duration, mean age, study subjects, dose, antioxidant profile, and female ration on heterogeneity (P > 0.05). The results of meta-regression showed that the quality score was also non-significant (coefficient = 0.12, P = 0.2). To determine the source of heterogeneity, the subgroup analyses were divided according to study subjects (subjects with metabolic diseases and healthy subjects). Subgroup analysis showed that the effect of Se supplementation on antioxidant markers did not change according to study subjects.

Publication bias

Publication bias was estimated by Egger’s test. The results of Egger’s test did not support the existence of publication bias by Se supplementation on antioxidant profile (coefficient = 2.2, P = 0.5) (Fig. 3).

Fig. 3.
figure 3

Funnel plot detailing publication bias in the studies reporting the impact of Se supplementation on Overall antioxidant profile.

Full size image

Sensitivity analysis

To test the robustness of the results, we conducted sensitivity analysis according to the study quality. After the exclusion of three low-quality studies, a total of 10 trials were analyzed. There was no significant difference in the results of the effect of Se supplementation on the level of plasma Se [(SMD) = 3.36, 95% CI = (3.2, 3.5)] and overall effect [(SMD) = 1.09, 95% CI = (0.5, 1.6)] by sensitivity analysis.

Discussion

The meta-analysis of this study was carried out on the basis of pooled data from 13 RCTs to assess the effect of Se supplementation on antioxidant factors. The results of analysis showed that Se supplementation increased GPX and Se levels and the total antioxidant capacity decreased MDA levels, whereas it had no significant effect on GSH, TBARS, and adiponectin levels.

According to the study analysis, Se intake increased Se and GPX levels increased considerably. Epidemiological studies have shown that Se intervention increases GSH-PX activity in plasma and erythrocytes [40, 41].

Various published reports support the fact that Se supplementation increases TAC and reduces MDA levels. A randomized, double-blind, placebo-controlled trial in 2016 reported that high Se-yeast supplement is effective in increasing TAC and decreasing MDA levels [42]. Several trials have shown that Se supplementation can reduce MDA levels [43, 44] and improve TAC and GSH levels [44]. Mahmoodpoor et al. [45] conducted a randomized, placebo-controlled trial in 2018 to investigate the effect of high-dose Se on antioxidant reserve of the lungs and ventilator-associated pneumonia (VAP) in critically ill patients. They demonstrated that serum Se and GPX-3 activity levels increased steadily in the treatment group within 10 days (P b 0.025), while they remained unchanged in the placebo group. They also stated that despite increasing antioxidant activity, Se supplementation did not affect the incidence of VAP in critically ill patients.

The results of one meta-analysis in 2017 on the effect of Se supplementation on coronary heart disease demonstrated that Se supplementation decreased serum CRP and increased GSH-PX levels, suggesting a positive effect on reducing oxidative stress and inflammation in CHD [46].

In an animal study carried out in 2017 by Mansour on the effect of Se yeast supplementation on growth, antioxidant status in meagre, it was shown that catalase and superoxide dismutase activity, as well as total antioxidant status, significantly increased, while thiobarbituric-reactive substances in liver homogenate significantly decreased with Se supplementation [47].

In another study, Se supplementation in 1-day-old male broilers resulted in amelioration of GPX activity, total antioxidant capacity, and malondialdehyde formation (P < 0.05) [48].

Significant heterogeneity was observed regarding the impact of Se supplementation on TAC, Se, and GPX levels. A random-effect model was subsequently applied. Meta-regression was performed to identify the source of heterogeneity. According to the meta-regression analysis, influencing factors, such as duration, mean age, dose, and female ratio had no effect on heterogeneity. However, there was a significant association between the type of antioxidant factors and heterogeneity (P = 0.008). We therefore conducted subgroup analyses based on antioxidant factors to determine the effect of Se supplementation on each factor.

The meta-analysis of the study has some limitations. In many investigations, Se has been used in combination with other vitamins or minerals, which makes it impossible to isolate the specific effects of Se or of different Se forms in these trials. In addition, the included studies were heterogeneous in type of patients, species of Se, and duration of treatment, as well as Se content of the soil in the various environments. These factors could well have affected the results of the studies. As other challenges assessed the complementary required data through the time-consuming process, we contacted the corresponding authors.

In the future, carefully designed double-blind clinical trials with large sample sizes with stratified participants based on the initial Se levels are recommended.

Conclusion

The study findings indicated that Se supplementation might be effective in reducing oxidative stress and inflammation through increasing TAC and GPX levels and decreasing serum MDA, both of which are crucial influencing factors for reduction of oxidative stress. However, the results also showed that Se supplementation is not sufficient to reduce TBARS or affect adiponectin and GSH levels. There was no evidence of adverse events. To the best of our knowledge, this study is the first published meta-analysis assessing the effect of Se supplementation on oxidative stress markers.