Red Ginseng as an Ergogenic Aid: A Systematic Review of Clinical Trials

[Purpose] This systematic review was performed to summarize clinical trials assessing the effect of Red Ginseng (RG) supplementation on exercise performance and fatigue recovery. [Methods] Two English databases (PUBMED, MEDLINE) and two Korean databases (KISS, RISS) were used as systematic searching engines. We included only articles written in the English and Korean languages. Clinical trials, which evaluated exercise performance and recovery variables with RG supplementation, were included in this review. The methodological quality of all studies was assessed using the Cochrane Risk of Bias tool. Analysis was conducted with Comprehensive Meta-Analysis Version 3. [Results] In total, 135 potentially relevant studies were identified, and 14 studies were included. Overall, the aerobic capacity (VO2max, heart rate, time to exhaustion, shuttle run, and anaerobic threshold) exhibited no improvement with RG supplementation. In anaerobic capacity (peak power, mean power, and 30 m dash run), no significant improvements with RG supplementation was described in most of the studies. The antioxidant function predominantly measured by levels of superoxide dismutase (SOD) and malondialdehyde (MDA) showed mixed results. Red Ginseng’s effects on fatigue recovery were evaluated using lactate as a main outcome. Two studies observed significant effects while other 5 studies showed no significant effects. [Conclusion] The clinical effects of RG have been assessed in various conditions. Although the number of studies presented in this review is small and results of studies are mixed, it is hypothesized that this review article may provide useful guideline to design and conduct future studies investigating efficacy of RG supplementation on exercise performance and fatigue recovery in human trials.


INTRODUCTION
[Results] In total, 135 potentially relevant studies were identified, and 14 studies were included. Overall, the aerobic capacity (VO2max, heart rate, time to exhaustion, shuttle run, and anaerobic threshold) exhibited no improvement with RG supplementation. In anaerobic capacity (peak power, mean power, and 30 m dash run), no significant improvements with RG supplementation was described in most of the studies. The antioxidant function predominantly measured by levels of superoxide dismutase (SOD) and malondialdehyde (MDA) showed mixed results. Red Ginseng's effects on fatigue recovery were evaluated using lactate as a main outcome. Two studies observed significant effects while other 5 studies showed no significant effects.
[Conclusion] The clinical effects of RG have been assessed in various conditions. Although the number of studies presented in this review is small and results of studies are mixed, it is hypothesized that this review article may provide useful guideline to design and conduct future studies investigating efficacy of RG supplementation on exercise performance and fatigue recovery in human trials.

Red Ginseng as an Ergogenic Aid
Journal of Exercise Nutrition & Biochemistry performance 1 . In general, ginseng supplements are safe, although individual variability exists and potentiation with stimulants may occur 7, 8 .
Cultivated ginseng can be classified into four types: 1) fresh ginseng, 2) white ginseng, 3) red ginseng (steamed and dried process), and 4) wild ginseng 9, 10 . RG contains bioactive compounds such as ginsenosides and phenolic compounds, and the compounds can vary according to the conditions of the heat process 11 . Extracts of red and white ginseng contain different ginsenosides 12 . Previous studies demonstrated that RG contain a large dose of ginsenosides and saponin through specific manufacturing processes such as steaming and drying ginseng 13 . Together with ginsenoside Rg3, a nitrogen-containing component and ginsenoside 20(R)-Rh1 were detected as characteristic components of RG, whereas malonyl ginsenoside Rb1/ isomer and malonyl ginsenoside Rg1/isomer were found to be characteristic components of WG 14 .
In particular, Korean Red Ginseng has ginsenoside such as Rg3, Rg 5, Rh1, and Rk1 15, 16 . For these reasons, the identification of bioactive components in RG has been regarded as an objective of scientific study 17, 18 . Recently, several studies have reported that RG has biological activities, such as vasodilation, antihypertension, antioxidant, anti-inflammatory, and improving cognitive function [19][20][21][22] .
RG study for exercise performance has been conducted predominantly with respect to aerobic capacity. RG supplementation demonstrated increase in the concentration of hemoglobin in the blood 23 . It also showed stimulation of mitochondrial metabolism in the muscles 24 . RG has been reported to improve the blood and organ lipid profile when combined with exercise 13 . It contains various biological and psychological activities and may also alleviate fatigue-related disorders. RG supplementation showed positive effect on fatigue resistance capacity, glycogen sparing, and oxidation of free fatty acid during prolonged sub maximal exercise 25 .
In contrast, effect of RG supplementation on anaerobic capacity does not look promising as compared to aerobic capacity 26 . A study by Engels, et al (2001) 27 reported no significant changes in anaerobic capacity after 8 weeks of RG supplementation. Similarly, Yoon, et al (2001) 28 also showed that there were no improvements in maximal and mean powers in Wingate power test. However, Hwang, et al (2004) 29 reported that RG supplementation can reduce generation of oxidative free radical resulting in possible positive effect on exercise performance. However, the number of studies and quality of RG studies are very limited as compared to the studies investigating regular ginseng. Also, there are only a few studies written in the English language available to scientific community.
Therefore, we conducted systematic review to summarize the current evidence of RG supplementation on exercise performance and fatigue recovery both written in the English and Korean languages. We anticipate that this review will provide valuable information about current ongoing research on RG to our colleagues and help them to design and conduct future research projects.

Eligibility of studies
The inclusion criteria in this study were as follows: 1) RG as sole supplementation, 2) Studies involving a control or a placebo group, 3) Evaluated acute and chronic effects of RG supplementation, 4) Oral administration of RG in the form of capsules, powders or liquids, and 5) Assessed the efficacy of RG treatment on exercise performance and fatigue recovery. On the other hands 1) RG supplementation combined with other products, 2) Nonoral administration of compounds, and 3) animal studies, case studies, and uncontrolled trials were excluded in this study.

Data Sources
The following four electronic English and Korean databases were searched with restriction of language (English and Korean) from their respective inceptions up to June 2016: the Korean Studies Information Service System (KISS), Korea Institute of Science and Technology Information, Research Information Service System (RISS), US National Library of Medicine National Institutes of Health (PUBMED), and Advancing the Health of Healthcare (MEDLINE). The search terms used were ''Red Ginseng'' and ''exercise'' in the English and Korean languages. Unpublished articles were excluded. The references in all located articles were also searched. Hard copies of all articles were obtained. Figure 1 illustrates process of study selection.

Types of interventions
Trials that included extract of RG, regardless of age, gender, or dose, were included. According to the processing status, we excluded studies targeted on animals. We compared placebo or no treatment to RG supplementation treatments on exercise performance, antioxidant function, and fatigue recovery.

Data collection
The data (author, publication year, country, sample size, conditions of the participants, intervention, dosage, treatment duration, outcome measures, main results, and language) were extracted by a standard form. All the articles were read by two independent reviewers, who extracted data from the articles according to predefined criterion.

Assessment of risk of bias
The risk of bias for each study was evaluated by the 'Risk of Bias' assessment with Review manager 5.1 (Copenhagen: The Nordic Cochrane Centre, The Cochrane Collaboration, 2011) in this study. The following sources were assessed: 1) sequences generation, 2) allocation concealment, 3) subjects and personnel blinding, 4) assessor

Red Ginseng as an Ergogenic Aid
Journal of Exercise Nutrition & Biochemistry blinding, 5) incomplete outcome data, and 6) selective outcome report. Our review used unclear (U), low (L) and high (H) as keys for judgments. Differences in opinions between the reviewers were settled through discussion.

Analysis
The studies were organized according to their primary outcomes. Estimated effect size for each outcome of included studies was calculated by comparative analysis with each control intervention individually. The out-

Red Ginseng as an Ergogenic Aid
Journal of Exercise Nutrition & Biochemistry comes were presented as mean difference (MD) and 95% Confidence Interval (CI). Analysis was conducted with Comprehensive Meta Analysis (CMA) 3.0 version. All statistical analyses were independently cross-checked by a statistician.

Study Selection and Description
We considered 135 articles screened from 4,283 potentially relevant articles (Figure 1). After assessing full-text articles for eligibility, a total of 82 articles were evaluated. Subsequently, 68 were excluded because they used animal models (16 articles), did not meet the eligibility criteria (18 articles), or because of other reasons (31 articles: irrelevant articles with our research topic). The three studies excluded for intervention conditions were as follows: a mixture of RG plus peonia radix (PR) 30 , while two studies employed vitamin E or electrolyte 31, 32 .
Finally, 14 studies met our inclusion criteria 33-46 . Thirteen studies 33-38,40-46 were published in the Korean language, and one study 39 was written in the English language.
Two studies were described as 'randomized'; one study was specified as 'double-blind', two studies were specified as 'single-blind', and thirteen studies did not state blinding. Thirteen studies were designed as parallel, and one study was cross-over.

Intervention
The trials used RG as an intervention for exercise performance, antioxidant, and fatigue recovery after intense exercise. Eight studies compared the efficacy of RG with placebo 35

Risk of bias
Risks of bias from all 14 studies are summarized in Figure 2. Four studies 35,37,38,45 used methods of random sequence generation. The risk of bias in sequence genera-

Red Ginseng as an Ergogenic Aid
Journal of Exercise Nutrition & Biochemistry tion was considered as high risk in one study 36 . However, none of the studies reported the methods of randomization such as computerized. Most of the studies did not clearly report the manner in which allocation concealment was generated, while five studies 35-38,45 employed allocation concealment. One study 44 was conducted with double blinded design (participants/researcher), while other two studies 40,41 just reported 'blinding (participants)' design. None of the studies mentioned 'adopted assessor blinding'. The risk of bias in terms of the incomplete outcome data and the selective reporting were unclear in the present review.
In case of aerobic capacity, VO2max was the main variable evaluated by 6 studies, with HR measured in 5 studies, and time to exhaustion measured in 3 studies. Other variables like shuttle run test, AT, and biological variation (red blood cell; RBC, hemoglobin; Hb, and Hematocrit; Hct) were measured in one study each. A total 22 outcomes from 10 variables, all related to VO2max out- In case of anaerobic capacity, peak power was the main outcome evaluated by 4 studies. Mean power outcomes were measured in 2 studies, and 30 m dash run was measured in one study. A total of 8 outcomes from 3 variables, most of the outcomes such as peak power 39,45, 46 and mean power 39,45,46 showed no significant differences. Other variable, GPX was measured in one study 37 . A total of 11 outcomes were evaluated from 4 variables. In four SOD outcomes 37,40,43,44 , two 37,40 had positive effects and two 43, 44 were not significant. Most of the outcomes involving MDA 40,43, 44 were not significant excluding the study reported by   37 . In terms of CAT, one of those 43 showed significant effect, while another study 40 was not significant. In the study reported by  GPX showed significant mean differences 37 .
The effect of RG on fatigue recovery was evaluated in  7 studies 33,35,36,38,39,44,46 . In all the studies related with fatigue variables, lactate was evaluated as a main outcome. In a total of 9 outcomes, 7 were lactate 33

DISCUSSION
This study reviewed 14 published articles investigating the effect of RG on exercise performance and fatigue recovery in human trials. This review has differentiated significance compared to previous review articles in terms of summarizing only RG supplementation related with exercise. Furthermore, the present review represents a systematic evaluation of clinical trials published in the Korean and English literatures. Finally, 14 studies that met the inclusion criteria were categorized based on aerobic, anaerobic, antioxidant function, and fatigue recovery. We observed that previous studies primarily focused on aerobic capacity and fatigue recovery as variables of ergogenic ability. To the best of our knowledge, only 10 outcomes have shown that RG supplementation can help in exercise performance. Specifically, only 2 of 22 outcomes were in aerobic category, 1 of 8 outcomes was in anaerobic category, 5 of 11 outcomes were in antioxidant category, and 2 of 7 outcomes were in fatigue recovery category. VO2max and HR were commonly evaluated as an aerobic capacity outcome. There was some evidence that RG may have an effect on aerobic capacity, with 2 of 22 outcomes partially demonstrating improvement in Hb, and shuttle run test score in RG groups compared with control groups. However, most of the studies showed that RG supplementation had no effects on aerobic outcomes including VO2max, HR, AT, VE, RBC, and time to exhaustion test. Most of the results clearly state that RG supplementation had no effects on anaerobic capacity and fatigue recovery. In terms of antioxidant function, 5 of 11 outcomes had positive significant effects. However, the outcomes observed heterogeneities between each variable, which means that the data in each category were inconsistent and statistically insufficient. Therefore, more studies are obliged to judge the effects clearly. Many studies conducted using animals have usually shown that RG may improve exercise performance; whereas, data from human subjects reveal contradictory results. We believe that the inference could be due to various methodological problems including small sample size, different dosage, and lack of placebo group. The actual composition of ginseng preparations has been a persistent issue, and differences in

Red Ginseng as an Ergogenic Aid
Journal of Exercise Nutrition & Biochemistry the effects of ginseng may be due to the dosage and type (powder, extract, etc) of RG consumed. There may be interactions with diet, lifestyle, exercise, and other drugs.
We wish to highlight some of the difficulties faced while conducting research on RG and offer some suggestions for future research. Most of the reported studies did not used standardized RG preparations. Moreover, the majority of studies did not describe a ginsenoside profile except for two articles 43, 44 . None of the studies reported independent testing of the preparation to confirm purity, consistency, and stability. In addition, dose ranges varied among the studies (900 mg to 5 g/day for capsules, 100 to 200 ml/day for liquid, and 1500 mg/day for powder). Non-randomization may lead to a substantial overestimation of the effect size. It is also suggested that post-harvest handling processing for RG should be standardized using the identified characteristic components as chemical markers to ensure the quality and efficacy of RG.
In the methodological aspect, the quality of most of the studies was poor, as they did not clearly describe adequate randomization, blinding, and/or description of withdrawals and dropouts. In addition, none of the studies conducted statistical calculation to determine reasonable subject numbers. Moreover, we observed that interpretation of the results was incorrect based on statistical analysis. For instance, a study concluded that RG had effects on exercise performance despite no significant interaction by repeated measures ANOVA. Therefore, the limitations in methodological quality could induce the different interpretation of RG supplementation on exercise performance. As such, it is necessary to conduct further studies that are of high quality and with larger sample sizes to contribute towards formation of a definitive conclusion. Specifically, it is required to describe the accurate information of ginsenoside dosage, and ratios to evaluate the efficacy of RG as an ergogenic aid. In addition, future studies need to fulfill accurate research procedures.
This review may serve as a foundation for future systematic reviews and further studies, but it also has some limitations. First of all, data synthesis for this study was limited by heterogeneity between outcomes. Pooling results were confounding, which was inappropriate. This phenomenon may be caused by the small sample size and varied supplementation dosage. Another limitation was that this review article included studies written in the Korean and English languages.

CONCLUSION
This review explores the scientific evidence for use of RG extracts as ergogenic aid for exercise performance. The results of this systematic review reveal that RG is not effective on aerobic and anaerobic capacity, antioxidant function, and fatigue recovery. However, we conclude that there is absence of strong scientific research evidence regarding the efficacy of RG on improving exercise performance in humans. In future studies, it is necessary to conduct further systematic review of high quality and involving large sample sizes with an aim to form a definitive conclusion.