The Effects of Grape/Grape Products on Inflammatory and Oxidative Stress Markers: Systematic Review and Meta-Analysis of Randomized Controlled Trials

,


Introduction
Reactive oxygen species (ROS) are produced in the cells in small quantities [1] as a consequence of physiological oxidative metabolism, mitochondrial bioenergetics, and immune function [2].Tey act as a regulatory agent in the signaling pathways including gene expression, receptor activation, and signal transduction as well as cell diferentiation, cell growth, metabolic adaptation, and immune responses [3].Te excess production of ROS in the cells which exceeds the ability of the antioxidant system to neutralize them is known as oxidative stress [4], which can damage cellular molecules including lipids, protein, and DNA [5].Oxidative stress also activates various infammatory processes, such as nuclear factor κB (NF-κB) and activating protein-1 (AP-1), leading to the synthesis and secretion of pro-infammatory cytokines [6].Oxidative stress and infammation have been shown to be associated with, and implicated in, various diseases including, diabetes, cardiovascular diseases [5], cancer, neurodegenerative diseases, and arthritis [7].
Te cumulative evidence shows that dietary polyphenols are able to exert antioxidant activity, ROS scavenging properties, and anti-infammatory functions which can alter the expression of pro-infammatory cytokines [8].Terefore, a possible approach to reduce oxidative stress [9] and infammation is to add polyphenol-rich foods, such as grape or its products, to the diet [6].Grapes are the major source of phenolic acids, proanthocyanidins, anthocyanins, favonoids, and stilbenes [10].Te most common grape-derived products include raisins, juices, and wines [6], which are also reported to have a high nutraceutical level and are on the market in the form of powders, dried, and concentrated extracts [6,9].It has been shown that grape and its product consumption can reduce the risk factors for cardiovascular and neurodegenerative disease, diabetes, cancer, and cognitive decline [11].
A serious criticism of prior studies is that they did not examine the efects of whole grape fruit or its products.Moreover, grape derivatives such as extracted polyphenols or grape seed extract were included in the same analysis along with fresh grape.Terefore, ascertaining whether whole grape fruit and its products can be efective on infammatory factors is still unanswered.
Terefore, we sought to conduct a systematic review and meta-analysis of randomized controlled trials (RCTs) to assess the efect of whole grape/grape products on infammatory and oxidative markers in adults.

Methods
Te present systematic review and meta-analysis was conducted in accordance with the 2020 Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines [22].Te protocol of the present study was registered at the International Prospective Register for Systematic Reviews (PROSPERO) database (ID of CRD42022349904).

Search Strategy and Study Selection.
To identify relevant studies, a systematic search was conducted in PubMed, Scopus, ISI Web of Science, and the Cochrane Library, from database inception to January 2022, without any restrictions for language or publication date.A combination of MeSH and non-MeSH terms related to "grape" and "study design" was applied.Further details on the search strategy and PICO components are provided in Tables S1 and S2, respectively.In addition, reference lists of all included studies were screened manually in order to fnd additional eligible trials.

Eligibility Criteria.
Original randomized clinical trials with parallel or crossover design investigating the efects of grape or its products in comparison with a control/placebo group were eligible for inclusion in this study if they were conducted on an adult population (age >18 years), lasted for a 2-week intervention or more, and reported at least one of our key outcomes (including serum levels of oxidative stress (ox-LDL; MDA, SOD, ORAC, GPX, nitric oxide (NO), advanced oxidation protein products (AOPPs), thiobarbituric acid reactive substances (TBARSs), catalase (CAT), and Trolox equivalent antioxidant capacity (TEAC)) and infammatory markers (TNF-α, hs/CRP, IL-6, IL-8, IL-10, monocyte chemoattractant protein-1 (MCP-1), and C3 protein) as mean ± standard deviation (SD) (or other statistics that can be converted to mean ± SD)), before/after or change during the study.
Te exclusion criteria were as follows: (1) trials that administered grape seed, resveratrol, or any individual ingredient of grape; (2) trials conducted on pregnant or lactating women; and (3) trials that administered multiple intervention components that independent efect of grape/ grape products could not be determined.
Te initial abstract and full-text screening of all retrieved studies were executed independently by two authors (SS and RKM) according to inclusion and exclusion criteria, and disagreements were resolved by consensus with the third author (SA).

Data Extraction.
Two authors (FD and SS) extracted the following data from eligible studies: (1) study characteristics including frst author's name, publication year, study location, study design, sample size, follow-up duration, type of intervention (grape, raisin, grape juice or powder), placebo, and dosage of intervention; (2) participant characteristics including sex, age, and health status; and (3) mean/SD or other convertible data of antioxidants and infammatory markers before/after intervention or mean changes/SD during the follow-up period.Disagreements were resolved by consensus with the third author (SA).Te dose of grape products was converted to cup equivalent unit using Food Patterns Equivalents Database (FPED) [23], to make the dose data of grape products comparable between trials.
PlotDigitizer software (https://plotdigitizer.sourceforge.net/) was used for extracting data only obtainable in graphical formats.Te units of antioxidants and infammatory markers were unifed before analysis.For studies that administered multiple dosages or follow-up durations, the higher dosage or duration was included in the analysis.

2
Journal of Food Biochemistry 2.4.Quality Assessment.Te Cochrane risk of the bias tool for RCTs was used to assess the risk of bias based on the following six parameters: selection bias, performance bias, detection bias, attrition bias, reporting bias, and bias due to problems not covered elsewhere [24].Te eligible studies were classifed as low risk, high risk, and unclear.Studies with low risk of bias for all criteria were judged as good quality, studies with one criterion not met or two criteria unclear were judged as fair quality, and studies with high risk of bias for two or more criteria were judged as poor quality.

Statistical Analysis.
Te efect of grape products on the change in the outcomes of infammation (hs-CRP, TNF-α, IL-6, IL-10, and monocyte chemoattractant protein-1 (MCP-1)) and oxidative stress (ox-LDL), ORAC, thiobarbituric acid reactive substances (TBARSs), TAC, carbonyl, MDA, catalase (CAT), and SOD were examined.We applied analysis for the outcomes, when three studies or more reported data.Te mean diferences (MDs) and their corresponding 95% confdence interval (CI) between intervention and control groups were used to calculate the efect size.For studies that did not report change values, SDs were calculated according to the formulas suggested by the Cochrane Handbook of Systematic Review [27].For all infammatory and oxidative stress markers, we used a correlation coefcient of 0.5, with the exception of MDA, which was 0.48.Weighted mean diferences (WMDs) and their SDs were calculated by using the DerSimonian and Laird random-efects model [28].Due to the small number of included studies for each outcome, a modifed method for random-efects meta-analysis (the Hartung-Knapp-Sidik-Jonkman method) was used to reduce the error rate [29].
To evaluate the statistical heterogeneity between the studies, Cochran's Q test and I 2 statistic (I 2 ) were used [30].In order to detect the sources of heterogeneity, predefned subgroup analyses were conducted for potential factors including sex (male, female, both), participants' health status (healthy, metabolic syndrome, cardiovascular disease, and hemodialysis), study type (parallel, crossover), duration of intervention (≤12 wk, >12 wk), intervention type (fresh grape, dried grape (including pomace, raisin, powder), and grape extract (including juice, concentrate, extract)), dosage (<150 gr, ≥150 gr), study location (Asia, Europe, USA, and south USA), and quality of studies (unclear, low, and high).Sensitivity analysis was conducted by the leave-one-out method to determine whether a single trial afects the overall efect size [31].For outcomes with at least 10 trials, publication bias was evaluated by the visual inspection of the funnel plot and assessed statistically by Begg's adjusted rank correlation and Egger's regression asymmetry tests [32].In addition, meta-regression analysis was conducted to explain the between-study variation based on the dose and duration of the intervention.Te efect of other variables could not be assessed, due to the incomplete data.Statistical analyses were performed using STATA version 17 (StataCorp).Two-sided P values <0.05 were considered statistically signifcant.
Excluded studies, along with the reasons for exclusion, are shown in Supplemental Table 3.

Quality Assessment.
Te methodological quality of the included studies was assessed using the Cochrane collaboration tool.We found eight studies to have a good quality design [37,40,41,44,45,50,52,54]: one study had fair quality [38] and the remaining had poor quality [12-17, 33-36, 39, 42, 43, 46-49, 51, 53, 55].Te most Journal of Food Biochemistry contributing factors were random sequence generation and allocation concealment, details of the process of which were not explained (Table S4).
Te GRADE assessment system showed that the quality of evidence was very low for the efect of grape products on carbonyl, ox-LDL, MDA, SOD, total antioxidant capacity (TAC), TNF-α, and MCP-1 and low for the remaining outcomes (catalase, ORAC, TBARS, hs-CRP, and IL-6) (Table S5).

Advanced Oxidation Protein Products (AOPPs).
Tree studies reported the efect of raisin [44], black current [56], and freeze-dried grape [39] on AOPP levels.Te data from the articles could not be analyzed because the unit reported in one of the articles could not be converted and equated.No signifcant efect was observed in all three studies.

Paraoxonase (PON1).
Te efect of the freeze-dried grape extract on the PON1 activity has been investigated in adults with metabolic syndrome.Te results showed that grape supplementation did not signifcantly change the PON1-arylesterase activity and PON1-lactonase activity compared with placebo [57].

GPx.
Te grape extract and powder supplementation (400 mg/day) could not signifcantly change GPx levels in elite athletes [45] and hemodialysis patients [58], respectively.Grape juice consumption (400 ml/day) in patients with Parkinson's disease for 4 weeks signifcantly reduced GPx levels compared to the placebo group [15].Te data of the articles could not be analyzed because the unit reported in one of the articles could not be converted and equated.

Nitric Oxide.
In one study, raisin consumption showed no signifcant efect on nitric oxide in healthy smokers [44].

Complement Component 3-Protein (C3).
One study reported the efect of dietary supplementation with concentrated red grape juice on C3 in hemodialysis patients, which had no signifcant diferences between the intervention and control groups [35].

IL-1β.
Two trials examined the efects of dietary grape powder and grape on IL-1β in obese and healthy subjects, respectively.Zunino et al. reported no signifcant diference in IL-1β concentrations between groups [17].In Ammollo's study, grape intake did not show any signifcant change in the plasma levels of IL-1β in intervention and control groups, while a signifcant reduction in the release of IL-1β by lipopolysaccharide (LPS-) stimulated blood cells was observed in the intervention group compared with the control group [59].

IL-18.
One study showed a signifcant reduction in IL-18 after grape consumption for one year in patients receiving primary CVD prevention, but it was not signifcant in comparison with the control group [60].
4.3.9.IL-6/IL-10.Two studies explored the efects of grape extract supplementation in patients with stable coronary artery disease [52] and those on primary CVD prevention [60].In both studies, grape extract supplementation did not change the IL-6/L-10 ratio signifcantly.and men with metabolic syndrome [34].Grape powder supplementation did not signifcantly alter the IL-8 levels between the intervention and control groups in both studies.

IL-10.
Te efect of the grape extract and grape powder on IL-10 was investigated in three trials.However, a contradictory result was reached.A study showed that grape consumption increased IL-10 levels in patients with metabolic syndrome [34], while another study revealed no signifcant change in patients with stable coronary artery disease [52] and patients undergoing primary prevention of cardiovascular disease [60].
Publication bias was assessed for the efect of grape products on levels of hs-CRP, TNF-α, IL-6, and ox-LDL.Te visual inspection of the funnel plots revealed no sign of asymmetry, which was also confrmed by Egger's and Begg's tests.

Meta-Regression.
No signifcant associations were found between LFDs compared with HFDs on CRP, TNF, LDL-ox, and IL-6 and dose or duration of the intervention in meta-regression analysis (Supplementary Table 11).

Discussion
In the present meta-analysis, we pooled data from RCTs examining the efect of grape products on infammation and oxidative stress markers in adults.Our main results showed that grape products signifcantly increased the catalase activity, ORAC, TAC, and decreased TBARS levels.We also found the grape extract may have more benefcial efects on infammatory markers, including CRP, ox-LDL, and TNF-α, as compared to the other types of intervention (fresh grape or dried grape).
Tere are bidirectional communications between oxidative stress and infammation; indeed, both are able to activate and aggravate each other in a direct or indirect manner [61].Frequent exposure to ROS leads to cell damage and consequently induces proinfammatory pathways.Oxidative damage can induce the release of TNF-α from the cells, which binds to its cell surface receptors, leading to NF-?B infammasome activation [61].As a result of NF-?B signaling activation, other proinfammatory cytokines, such as IL-1?, IL-6, and TNF-α, are produced [61][62][63].Grapes are known as one of the main sources of antioxidants due to their high polyphenol content [64].Polyphenols are capable of enhancing antioxidant enzyme activities directly or indirectly through the activation of Keap1/Nrf2/ARE and sirtuin 1 (Sirt1) signaling pathways, which can enhance the expression of antioxidant genes [65].Moreover, polyphenols can suppress the NF-κB pathway, toll-like receptor (TLR), and expression of infammatory genes, consequently [66].
Although we found no signifcant efect of grape and its products on infammatory markers in the overall analysis, it seems grape extracts can reduce infammatory markers.It may be associated with the higher dose of antioxidant in the grape extract compared to fresh grape.On the other hand, none of the included studies investigated patients with preexisting acute infammation, so the low level of infammatory factors was not afected by the short-term intervention.In line with our results, a meta-analysis of eight studies regarding grape polyphenols did not show any signifcant efect on hs-CRP, IL-6, and TNF-α [21].In another meta-analysis, grape products had no signifcant efect on IL-6 and TNF-α, while they showed a signifcant efect on CRP levels [18].However, two meta-analyses showed grape seed extract supplementation [20] and grape products containing polyphenols signifcantly reduced the CRP levels [19].Te inconclusive fndings in the extant literature may be related to the diferent eligibility criteria used by the studies.For example, in the meta-analysis of grape products containing polyphenols, red wine and resveratrol interventions were also included in the analysis, which might distort our understanding of the actual efect of grape and its derivatives.
We also found that grape supplementation signifcantly increased catalase, ORAC, and TAC and decreased the TBARS levels.However, no signifcant changes were observed regarding carbonyl, ox-LDL, MDA, and SOD levels.Te results of recent meta-analyses are also equivocal.In one study, grape products had no signifcant efect on TAC and MDA levels [18].In another study, grape products containing polyphenols showed a signifcant increase in TAC, whereas they did not have any signifcant efect on SOD, MDA, ORAC, and GPx levels [19].Moreover, grape seed extract supplementation signifcantly decreased MDA and ox-LDL along with eliciting an increase in TAC levels [20].
In the subgroup analysis, grape and its products signifcantly reduced the ox-LDL levels in hemodialysis patients.Hemodialysis patients exhibit oxidative stress throughout dialysis due to the accumulation of oxidative products (such as ox-LDL) and insufcient antioxidant ability [67].Grape polyphenols are able to reduce lipid levels through pancreatic lipase inhibition, lipoprotein synthesis reduction in hepatocytes, and increasing fatty acid metabolism [68].
Te present study has several strengths that should be acknowledged.To the best of our knowledge, this is the largest and most comprehensive meta-analysis to have examined the efects of grape/whole grape products on circulating levels of infammatory and oxidative markers in adults.In addition, only RCTs were included in our study, permitting causal statements to be made.Previous metaanalyses regarding the efect of grape/grape products on oxidative and infammatory markers have some limitations such as including grape seed and grape products containing polyphenols, which failed to show the real efect of grape and whole grape products on these factors.Furthermore, our search was more comprehensive, and we included all the studies with grape intervention without any missing publications.Moreover, it has been reported that I 2 has not enough power to show heterogeneity in meta-analysis with small number of studies [69], so the CI for I 2 has been reported in our study to avoid misconceptions.
However, despite the strengths outlined above, there are some potential limitations worth noting.First, the number of included studies for most outcomes (except for hs-CRP, IL-6, TNF, and ox-LDL) was small, which could afect the results' validity.Second, several factors including sex, age, environmental factors, diet, and genetics might afect the bioavailability of polyphenols and physiological responses to oxidative stress [64], and we were unable to include all these variables in our subgroup analysis.Tird, although all the eligible studies administered grape and its products, there was a wide variety of the interventions or grapes, which might afect the products' antioxidant content due to different processing methods.Tis is probably the reason for the observed signifcant result regarding the geographical location for TNF-α and ox-LDL.It is suggested that future studies focus on the composition of nutrients of grapes depending on the geographical location of plant growth and their health efects.Fourth, based on the CI of I 2 , there was a considerable heterogeneity between studies for all outcomes, which we could not fnd the source of it.Fifth, some of the included studies used untreated group and some others used placebo as the control group.Although in dietary intervention studies, the criteria of a placebo group may not be possible to achieve, untreated groups are accepted as a control group in these studies.However, blinding in these studies may not be possible, but it may not matter for objective outcomes.Nevertheless, a subgroup analysis was performed and no signifcant diferences between placebo-and untreated control studies were found (data not shown).Sixth, the methodological quality of most of the included studies was poor, and the certainty of evidence was low and very low for all the outcomes, mainly due to the small sample size for each outcome that did not reach the optimal information size, which means the observed results may change when further studies are added to the present analyses.Lastly, the compliance rate for most included studies was not mentioned, which might afect the results.

Conclusion
In conclusion, grape products can signifcantly improve some biomarkers of oxidative stress, although appear to have no signifcant efect on reducing infammation in an adult population.Low certainty of evidence emphasizes the need to further good quality of randomized clinical trials to clarify the actual efect of grape and its derivatives on oxidative stress and infammation.

Table 1 :
Characteristics of eligible randomized controlled trials that investigated the efect of the grape and its products on infammatory and oxidative stress biomarkers in adults.