Effect of Resveratrol Supplementation on Inflammatory Markers: A Systematic Review and Meta-analysis of Randomized Controlled Trials

ARTICLE IN PRESS Clinical Therapeutics/Volume &, Number &, 2018

Review

Effect of Resveratrol Supplementation on Inflammatory Markers: A Systematic Review and Meta-Analysis of Randomized Controlled Trials D1X XMehdi Koushki, D2X XPhD1; D3X XNasrin Amiri Dashatan, D4X XPhD2; and D5X XReza Meshkani, D6X XPhD1 1

Department of Clinical Biochemistry, Faculty of Medicine, Tehran University of Medical Sciences, Tehran, I.R Iran; and 2Proteomics Research Center, Shahid Beheshti University of Medical Sciences, Tehran, I.R. Iran

TAGEDPABSTRACT Purpose: The evidence has suggested that resveratrol has anti-inflammatory effect; however, the results are inconsistent and inconclusive. The aim of this study was to assess the effect of resveratrol supplementation on the levels of inflammatory markers through a systematic review and meta-analysis of available randomized controlled trials (RCTs). Methods: A search strategy was completed using Medline, ISI Web of Science, Directory of Open Access Journal, SID, ProQuest, Cochrane Library, Scopus, and EMBASE up to May 2017, to identify placebocontrolled RCTs that assessed resveratrol effects on circulating (serum and plasma) inflammatory markers (interleukin [IL]-6, tumor necrosis factor a [TNF-a], and high-sensitivity C-reactive protein [hs-CRP]) among adult participants aged 17 years and older in 17 RCTs with a total of 736 subjects. The evaluation of study quality was performed using the Jadad scale. Weighted mean difference (WMD) was calculated for evaluating the changes in the inflammatory markers using fixed-effects or random-effects models. We performed subgroup and sensitivity analyses to evaluate the heterogeneity of the studies. Findings: Seventeen RCTs, including 736 subjects, fulfilled the eligibility criteria and were selected for analyses. The results of meta-analysis found significant reductions in the level of TNF-a (WMD, 0.44; 95% CI, 0.71 to 0.164; P = 0.002; Q statistic = 21.60; I2 = 49.1%; P = 0.02) and hs-CRP (WMD, 0.27; 95% CI, 0.5 to 0.02; P = 0.033; Q statistic = 26.95; I2 = 51.8%; P = 0.013) after supplementation with resveratrol. Resveratrol supplementation had no significant effect on the level of IL-6 (WMD, 0.16; 95% CI, 0.53 to 0.20; P = 0.38; Q statistic = 36.0; I2 = 72.3%; P = 0.001). Statistically significant

& 2018

heterogeneity was observed for the type of sample in IL-6 and study duration in inflammatory markers IL-6, TNF-a, and hs-CRP. Implications: Available evidence from RCTs suggests that resveratrol supplementation significantly reduced TNF-a and hs-CRP levels. Significant improvement in inflammatory markers support resveratrol as an adjunct to pharmacologic management of metabolic diseases. (Clin Ther. 2018;&:1 18) © 2018 Elsevier Inc. All rights reserved. Key words: hs-CRP, inflammation, meta-analysis, resveratrol, systematic review, TNF-a.

TAGEDH1INTRODUCTIONTAGEDN Inflammation is a series of cellular and molecular events that help to defend the body from infection.1 The long-term consequence of a prolonged inflammation has detrimental effects and it is now well understood that its dysregulation has as major role in the pathogenesis of several diseases such as type 2 diabetes (T2D), rheumatoid arthritis, atherosclerosis, asthma and other autoimmune diseases.1 The chronic lowgrade inflammation is usually characterized by an increased abundance and activation of innate and adaptive immunity cells in tissues along with an enhanced release of inflammatory factors and chemokines locally and systemically.2 Among various factors, tumor necrosis factor a (TNF-a) and interleukin 6

Accepted for publication May 25, 2018. https://doi.org/10.1016/j.clinthera.2018.05.015 0149-2918/$ - see front matter © 2018 Elsevier Inc. All rights reserved.

1

ARTICLE IN PRESS Clinical Therapeutics (IL-6) are two cytokines released in large amount during inflammation and are important mediators in T2D and cardiovascular diseases.3 These cytokines are potent inducers of C-reactive protein (CRP), an acute phase reactant mainly synthesized in the liver, which has been shown to be a strong risk factor for metabolic diseases.4,5 According to the literature, strategies to restrain chronic inflammation may be effective in treating chronic diseases.6 In this context, natural products have been documented to reduce systemic low-grade inflammation and are acknowledged as anti-inflammatory interventions. Resveratrol (3,5,40 -trihydroxy-trans-stilbene), a natural polyphenol with potential health benefits, is generally contained in grapes, peanuts, berries and polygonum cuspidatum.7 Resveratrol was shown to exert anti-aging activity mimicking some of the molecular and functional effects of caloric restriction.8 Resveratrol was demonstrated to prevent and treat chronic conditions including cardiovascular disease (CVD), T2D and neurodegenerative disorders.9,10 The evidence demonstrate that the beneficial effect of resveratrol on health is mediated through its antioxidant, antiinflammatory, cardioprotective, and neuroprotective activities.9 Studies using cultured cells, and laboratory animals have also suggested that resveratrol has an anti-inflammatory property.8 Resveratrol downregulates inflammatory reactions via inhibition of the synthesis of pro-inflammatory mediators, and modification of immune cells through the inhibitory effect on activator protein-1 (AP-1) and NF-kB and regulation of mitogen-activated protein kinases/hemeoxygenase-1 pathway, and the mediation of pro-inflammatory cytokines and reactive oxygen species formation.7,11 In addition to in vitro and animal studies, randomized clinical trial (RCT) studies have also provided the evidence that resveratrol might reduce inflammatory markers in different clinical conditions.12,13 However, the evidence for the effects of resveratrol on the levels of inflammatory markers is inconsistent and inconclusive. Therefore, here we performed a systematic review and meta-analysis study to assess the effectiveness of resveratrol consumption on the levels of inflammatory markers (hsCRP, TNF-a, and IL-6) by searching different bibliographic databases for published RCTs. The meta-analysis was conducted in accordance with the guidelines of the 2009 Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) statement.14

2

TAGEDH1MATERIAL AND METHODSTAGEDN Search strategy We searched PubMed-Medline, Scopus, ScienceDirect, ISI, google scholar, DOAJ (Directory of Open Access Journal), SID ProQuest, Cochrane Library, and EMBASE up to May 2017. We used Mesh databases and the following search queries: ((resveratrol [supplementary concept] OR resveratrol [Title/Abstract]) AND (inflammatory markers [Mesh terms] OR inflammation [Title/Abstract] OR cytokines [Title/abstract].

Study selection: Inclusion/Exclusion criteria Original studies were included if they met the following criteria:1) randomized clinical trials with or without concurrent medications, 2) English language, 3) represented the relationship between the consumption of purified resveratrol or resveratrol-enriched extracts with changes of the circulating inflammatory markers, 4) presented sufficient information [standard deviations (SDs), standard error (SE), or 95% CIs] on inflammatory marker levels at the baseline and at the end of the study in both resveratrol and control groups, 5) randomized clinical trials (parallel or cross-over), 6) the studies with an appropriate controlled design, 7) subjected ingested resveratrol supplementation for at least one month and maximum 12 months and 8) participants with the age 17 years. The following studies were excluded: 1) non-clinical studies, 2) studies without control or placebo groups, 3) lack of sufficient information on inflammatory markers concentrations at the baseline or at the end of the follow-up, and 4) inability to obtain the adequate details of study methodology.

Data extraction Two authors extracted the data independently using a standardized electronic form. The extracted data included the following: 1) first author's name; 2) origin country; 3) year of publication; 4) study population; 5) study design; 6) dose of resveratrol; 7) treatment duration; 8) age 17 and gender (male/female) of the participants; 9) type of resveratrol; 10) total number of subjects in intervention and placebo groups; 11) mean and standard deviation of circulating (serum/plasma) inflammatory markers (IL-6, TNF-a and hsCRP) in both the intervention and placebo groups at the baseline and at end of study (Tables I and II). The rationale for selecting IL-6, TNF-a and hsCRP as the inflammatory markers in this study is based on considerable

Volume & Number &

ARTICLE IN PRESS M. Koushki et al. evidence showing that these factors are important mediators of the metabolic diseases.2 4

Quality assessment The quality of the studies was assessed by two authors using the Jadad scale.31 The items included were 1) randomization, 2) double blinding, 3) reporting of the number of dropouts and reasons for withdrawal, 4) allocation concealment, and 5) generation of random numbers (Supplemental Table I).

Statistical analyses We combined the data using Mantel-Haenszel fixed and random effects models in meta-analysis. A resveratrol and inflammatory markers were treated as continuous variables. Weighted averages were reported as Standardizes Mean Difference (SMD) (Cohen’s d) or Weighted Mean Difference (WMD in Cochrane database) with 95% confidence intervals (CIS) calculated using STATA statistical software version 12.0 (STATA corporation college station, TX USA). The statistical heterogeneity was measured using the Higgins I-square (p<0.1, I2> 50%, was considered representative of significant statistical heterogeneity). To establish the effect of clinical heterogeneity between the studies on metaanalysis outcome, subgroup analysis was performed according to the type of sample, type of disease, study duration and dose of resveratrol. We then performed sensitivity analysis by estimating the overall WMD in the absence of each study. Publication bias is a major concern for all types of meta-analysis; therefore, several methods were used to assess publication bias. The funnel plots as visual were calculated. The Begg rank correlation and the Egger weighted regression methods were also used to assess publication bias. p<0.05 was considered representative of statistically significant publication bias.

TAGEDH1RESULTSTAGEDN Results of the literature search A summary of the study selection process is shown in Figure 1. The initial search resulted in 150 articles. After screening the titles and abstracts, 131 articles were excluded because they were not clinical trials or had no available data. After a careful review of the full-texts, 2 studies were excluded due to the incomplete reported data. Thus, 17 studies were included in the current meta-analysis.

& 2018

Study characteristics We identified 17 RCTs that reported the changes in the inflammatory markers as an outcome after resveratrol supplementation. Tables I and II lists the RCTs used in the meta-analysis along with the baseline study characteristics. Of the 17 RCTs, 6 studies used plasma and 11 studies used serum samples. In 17 RCTS evaluated in meta-analysis, four trials used T2D patients, four trials reported CVDs, coronary artery disease and atherosclerosis, two used non-alcoholic fatty liver disease (NAFLD) patients, two used normal individuals, two trials performed on obese participants and three trials reported data on stable angina pectoris, nonobese postmenopausal and ulcerative colitis. A resveratrol dose of  500 mg/day was used in some trials, but a higher proportion of the trials used a dose of 500 mg/day. In all trails, the inflammatory markers in intervention group were compared to the placebo group. The mean patient age of intervention and placebo groups ranged between 36 and 70 years; and duration of patient follow-up for resveratrol supplementation ranged from 1 month to 12 months. A total of 736 individuals participated in 17 RCTs. The number of participants in 16 trials ranged from 19 to 20 patients, whereas in one of the trials, the number of participants was 127 and participants were mostly male (66.03%). Of the 17 RCTss, sixteen trials were parallel group and one trial was crossover.

Data quality assessment In order to achieve the research objective and ensure that the data are collected using the highest methodological quality, we carried out the data quality assessment using Jadad scale (Table 1, supplementary file). Scores graded as 0 to +5 that the zero demonstrates a low score, +5 means a strong score and +3 considered as the moderate score. The quality of the data was moderate to strong in all trials but two of the studies had a high risk of bias with a low score (+2 score). Scores constructed may lack validity and the results may not be associated with the quality, therefore, in addition to the jadad scale, we performed subgroup analysis for some of the data extracted from included studies in meta-analysis.

The effects of resveratrol on inflammatory markers Figure 2 presents the results of the overall WMD meta-analysis testing in the effects of resveratrol on the inflammatory markers. The overall WMD were

3

Studies characteristics in systematic review and meta-analysis

First author of the study

Country

Total sample Age (y) Publication P/RSV year Sample size 2013

Serum 22

Jao-Tome carniero 15 Spain Jao-Tome carniero 16 Spain

2013 2012

Serum 50 Serum 50

Agarwal B 17 Brasnyo P 18 Faghihzadeh F 19 Zahedi H 20 Ghanim H 21 Militaru C 22

USA Hungary Iran Iran USA Romania

2012 2011 2014 2013 2010 2013

Plasma 40 Serum 19 Serum 50 Serum 20 Plasma 20 Serum 58

Bo.S23 Bo. S 24 Samsamikor M 25 Chen S 26 Yoshino J 27

Italy Italy Iran China USA

2013 2016 2015 2014 2012

Serum 50 Serum 127 Serum 49 Serum 60 Plasma 30

Volume & Number &

Denmark 2013 Poulsen M 28 Trimmers S 29 Netherland2014 Khodabandehloo H 30 Iran 2017

Plasma 24 Plasma 22 Plasma 45

Dose of RSV

350 mg/ day 6 months 12 and for the next 6 months (700mg/day) 58/60 46/4 CAD 8.1 mg/day 12 63/62 20/30CVD 8 mg for 6 months and 12 6mg/day for the next 6 months 48/50 13/27Atherosclerosis 400 mg/day 1 52.5/57.9 19/0 T2D 10mg/ day 1 46.28/44/0435/15NAFLD 500 mg/day 3 17-35/17-35 20/0 Normal 40 mg/day 1.5 36/36 10/10Normal 40 mg/day 1.5 64.2/64.9 35/23Stable angina 20 mg/day 2 pectoris 35.4/34.7 15/35COPD 500 mg/day 1 65.4/65 86/41T2D 500 mg/day 6 39/37.7 25/24Ulcerative Colitis500 mg/day 1.5 20-60/20-60 42/18NAFLD 300 mg/day 3 59.8/58.2 0/30 postmenopausal 75 mg/day 3 women 31.9/44.7 24/0 Obese 500 mg/day 1 52.5/52.5 22/0 Obese 150 mg/day 1 61.10/56.48 23/22T2D 800 mg/day 2 57/63

22/0 T2D CAD

PG Changed

PG Changed PG Changed

PG PG PG PG PG PG

Unchanged Unchanged Changed Changed Changed Changed

PG PG PG PG PG

Changed Changed Changed Changed Unchanged

PG Unchanged CO Changed PG Unchanged

Y: year, WK: weeks, (P/RSV): (Placebo / Resveratrol), (M/F): (Male/Female), PLA: Polylactide, PG: (Parallel group), CO: (Cross Over), T2D: type 2 diabetes, COPD: Chronic Obstructive Pulmonary Disease, CAD: coronary artery disease, CVD: cardiovascular disease, NAFLD: non-alcoholic fatty liver disease

ARTICLE IN PRESS

Jao-Tome carniero13 Spain

Sex M/F Disease

Primary Outcome Study length Trial (inflammatory (months) Type makers)

Clinical Therapeutics

4 Table 1.

& 2018 Table 2.

The characteristics of type of resveratrol and levels of inflammatory markers in this meta-analysis

First author of the study

grape extract containing resveratrol grape extract containing resveratrol grape extract containing resveratrol trans-resveratrol trans-resveratrol trans-resveratrol trans-resveratrol trans-resveratrol trans-resveratrol trans-resveratrol trans-resveratrol trans-resveratrol trans-resveratrol trans-resveratrol consuming food containing resveratrol trans-resveratrol trans-resveratrol

TNF-a (pg/ml)

IL-6 (pg/ml)

hsCRP (mg/l)

P

RSV

P

RSV

P

RSV

13.9§6.9 15.2§7 12.4§8.9 2.72§1.06 NR 20.6§10.8 9.83§0.25 0.72§0.2 NR NR NR 23.5§14.8 -0.16§1 NR 5§1.7 16.15§2.27 -0.39§1.7

10.7§5.8 13.8§8.2 13§12.9 2.64§1.23 NR 16.2§5.47 9.31§0.2 0.46§0.2 NR NR NR 17.2§10.0 -0.53§1.30 NR 4.14§1 15.14§2.03 -0.13§2.08

2.4§1 2.6§1.8 1.8§1.5 0.91§05 NR 6.65§4.2 77.7§7.79 33§5% NR NR NR NR 1.84§1.68 3.59§0.72 3.13§0.67 6.21§5.23

2.6§1.9 2.1§1.7 1.8§1.4 1.20§0.6 NR 3.96§1.9 70.8§7.2 22§9% NR NR NR NR 2.01§1.9 4.5§0.77 2.42§0.38 7.05§3.98

4.5§1.8 4§1.8 4.8§4.3 NR 4.05§1.72 3.40§2.13 NR 0.77§0.18 5.9§2.6 1.39§2.6 3.53§2.3 NR 1.51§1.72 4.04§1.01 1.52§0.35 -0.73§3.1

3.7§2.5 3.2§2.1 3.7§2.8 NR 3.35§2.3 2.11§0.96 NR 0.46§0.1 5.2§1.7 1.27§2.1 2.58§1.79 NR 2.75§3.0 4.58§0.91 1.33§0.31 -0.74§1.86

ARTICLE IN PRESS

Jao-Tome carniero13 Jao-Tome carniero15 Jao-Tome carniero16 Agarwal B 17 Brasnyo P 18 Faghihzadeh F 19 Zahedi H 20 Ghanim H 21 Militaru C 22 Bo. S 23 Bo. S 24 Samsamikor M 25 Chen S 26 Yoshino J 27 Poulsen M 28 Timmers S 29 Khodabandehloo H 30

Type of RSV

Data are shown as the mean§SD. P (Placebo); RSV (Resveratrol); NS (Non-significant); NR (Non report)

M. Koushki et al.

5

ARTICLE IN PRESS Clinical Therapeutics

Figure 1. The flowchart of the randomized controlled trials (RCTs) selection process in this systematic review and meta-analysis.

evaluated for IL-6 [WMD (-0.16), (95% CI -0.53 to 0.20), p=0.38], TNF-a [WMD (-0.44), (95% CI -0.71 to 0.164), p=0.003], and hsCRP [WMD (-0.27), (95% CI -0.50 to -0.02), p= 0.033]. Resveratrol supplementation resulted in a statistically significant reduction in TNF-a and hsCRP levels. Albeit IL-6 [WMD (-0.16), (95% CI -0.53 to 0.20), p=0.38] was reduced but it was not statistically significant. There was statistically significant heterogeneity between the trials for IL-6 (p=0.00, I2=72.3%), TNF-a (p=0.02, I2=49.1%), and hsCRP (p=0.013, I2=51.8%). Visual inspection of the funnel plot showed no evidence of publication bias for IL-6, and hsCRP, with the exception of TNF-a (Figure 3). Publication bias was not evident when the Begg rank correlation method (p= 0.69 for IL-6, and p= 0.41 for hsCRP) and the Egger weighted regression

6

method (p= 0.36 for IL-6, and p= 0.29 for hsCRP) were used, with the exception of TNF-a with the Begg rank correlation method (p= 0.02) and the Egger weighted regression method (p= 0.004) (Figure 4). To explore the reasons for the heterogeneity, we conducted subgroup analysis. We found no heterogeneity in the effects of “dose of resveratrol” (data not shown) and “type of disease” (data not shown), except for “type of sample” and “study duration (Supplemental Figures 1 3) on outcomes analysis. A statistically significant heterogeneity was observed for “type of sample” for IL-6 and “study duration” in inflammatory markers of IL-6, TNF-a and hs-CRP (Supplemental Table II and Supplemental Figure 1). We also performed sensitivity analysis by removing the trials with the low quality. In the sensitivity analysis,

Volume & Number &

ARTICLE IN PRESS M. Koushki et al.

Figure 2. The results of the overall weighted mean difference (WMD) test for the effects of resveratrol supplementation on the inflammatory markers. WMD was calculated as mean difference intervention versus the placebo difference for each inflammatory marker. (A) Interleukin-6, (B) tumor necrosis factor a, (C) high-sensitivity C-reactive protein.

omission of the low quality studies (Ghanim H, Millitaru C) did not change the overall results, thus, supporting the robustness of our findings. When two RCTs were removed, there was no significant difference in overall meta-analysis conclusions for TNF-a (WMD -0.41) (95% CI -0.71 to 0.16) and hs-CRP (WMD -0.21) (95% CI -0.55 to -0.02). Taken together, these findings suggest that there is an association between the use of resveratrol and reduction of inflammation.

TAGEDH1DISCUSSIONTAGEDN

The findings of this systematic review and meta-analysis suggest a protective effect of resveratrol supplementation on inflammation. Meta-analysis of 17 RCTS showed that resveratrol use significantly reduced TNFa and hs-CRP levels. No statistically significant

& 2018

difference was observed for the effects of resveratrol supplementation on IL-6 in the intervention group in comparison with the placebo group. There is a significant heterogeneity between the trials. Publication bias was not evident when the Begg rank correlation and the Egger weighted regression methods were used. In included trials, Faghihzadeh and colleague reported that trans-resveratrol supplementation with a dose of 500 mg/day for two weeks significantly decreased inflammatory markers (hs-CRP, and IL-6) as compared with the placebo group in NAFLD patients.19 In another study, Hoda Sadat et al. reported that 200 mg/day Polygonum Caspidatum extract containing 40 mg trans-resveratrol for 6 weeks resulted in a reduction of plasma level inflammatory factors (IL-6 and TNF-a).20 The results of the study by TomeCarneiro et al. suggested that one-year consumption of resveratrol-rich grape supplement significantly

7

ARTICLE IN PRESS Clinical Therapeutics

Figure 3. The results of publication bias with funnel plot for (A) interleukin (IL)-6, (B) tumor necrosis factor a (TNF-a), and (C) high-sensitivity C-reactive protein (hs-CRP). Visual inspection of the funnel plot showed no evidence of publication bias for IL-6 and hs-CRP with the exception of TNF-a. Due to the small number of studies for nuclear factor kB, funnel plot was not drawn. SMD = standardized mean difference.

decreased hs-CRP, TNF-a levels, and IL-6/IL-10 ratio in seventy-five patients undergoing primary prevention of CVD.16 A randomized, double-blind, clinical trial in subjects with stable angina pectoris with oral consumption of 20 mg/day resveratrol for 2 months demonstrated a significant decrease of hs-CRP in resveratrol group compared to the control group.22 Chen et al. demonstrated that using 300mg/day resveratrol could reduce the levels of TNF-a in NAFLD patients.26 Furthermore, the anti-inflammatory effect of resveratrol was reported in healthy humans. Consumption of polygonum cuspidatum extract containing 40 mg resveratrol daily for 6 weeks led to a significant reduction in the levels of TNF-a, IL-6, and hs-CRP levels in normal-weight healthy subjects.21 Taken together, in agreement with the most of the studies included in the meta-analysis, our pooled analysis showed a significant

8

reduction in TNF-a and hs-CRP levels in overall WMD of the intervention group compared with the placebo group. In contrast to these findings, Poulsen et al. was found that consumption of 500 mg trans-resveratrol thrice daily for 4 weeks in obese human subjects had no effect on the levels of hs-CRP, TNF-a, and IL-6.28 In addition, it was reported that the markers of inflammation (hs-CRP and IL-6), did not change after 12 weeks of resveratrol supplementation in non-obese women with normal glucose tolerance.27 Inflammation has been shown to have an important role in the pathogenesis of several metabolic disorders. T2D-associated inflammation is characterized by an increased abundance of immune cells in different tissues along with production of inflammatory cytokines. Cytokines and chemokines produced mostly by macrophages generate local and systemic inflammation and

Volume & Number &

ARTICLE IN PRESS M. Koushki et al.

Figure 4. The results of publication bias with the Begg rank correlation and the Egger weighted regression methods. (A) Interleukin-6 (IL-6), (B) tumor necrosis factor a (TNF-a), (C) high-sensitivity C-reactive protein (hs-CRP), and (D) nuclear factor kB (NF-kB). Publication bias was not evident when the Begg rank correlation method (P = 0.69 for IL-6, P = 0.41 for hs-CRP, and P = 0.60 for NF-kB) and the Egger weighted regression method (P = 0.36 for IL-6, P = 0.29 for hs-CRP, and P = 0.43 for NF-kB) were used, with the exception of TNF-a with the Begg rank correlation (P = 0.02) and the Egger weighted regression methods (P = 0.004). SMD = standardized mean difference.

this condition leads to pancreatic b-cell dysfunction and insulin resistance in liver, adipose and skeletal muscle tissues.1 Data from human and animal studies also suggest that inflammation contribute to T2D complications such as nephropathy, neuropathy, retinopathy and cardiovascular diseases through cell cell interactions and the release of pro-inflammatory cytokines, chemokines, and proteases to induce inflammatory cell recruitment, cell apoptosis, angiogenesis, and matrix protein remodeling.1,2 Therefore, it appears that strategies based on the inhibition of the inflammatory responses potentially can be of interest in the field of treatment of T2D. There is increasing evidence showing the beneficial effects of resveratrol on alleviating insulin resistance, and T2D.32 Clinical trials in T2D patients raise a point of anti-diabetic effects of resveratrol.33,34 More recently, a meta-analysis of 11 randomized controlled trials demonstrated that resveratrol consumption significantly improves insulin

& 2018

sensitivity and glycemic control in patients with diabetes.35 Several mechanisms have been suggested to explain the anti-diabetic effects of resveratrol including decrease of oxidative stress, reduction of inflammation, down-regulation of protein-tyrosine phosphatase 1B (PTP1B), prevention of pancreatic b cell loss and decrease of lipid accumulation in muscle and liver.9,36 Regarding the anti-inflammatory effect of resveratrol in T2D, several studies have reported that resveratrol reduces macrophage infiltration into adipose, skeletal muscle, pancreas, liver and kidney tissues.37 Jeon et al. demonstrated that resveratrol reduces macrophage migration to adipose tissue in mice fed a high-fat diet.38 Chronic administration of resveratrol also improved insulin sensitivity and inflammatory responses in visceral white adipose tissue of dietinduced obese monkeys and 3T3-L1 adipocytes possibly through upregulating SIRT1 expression, decreasing adipocyte size, and suppressing decreased NF-kB

9

ARTICLE IN PRESS Clinical Therapeutics activation.39 Importantly, resveratrol suppressed the up-regulation of inflammatory cytokines such as TNFa, Interferon a (IFNa) and IFNb, and of their upstream signaling molecules, including Toll-like receptor 2 (TLR2), TLR4, MyD88, TNF receptor associated factor 6 (TRAF6), and NF-kB in epididymal adipose tissues of obese mice.40 Furthermore, resveratrol administration significantly suppressed chemokine (C-C motif) receptor (CCR) 6 production in immune cells and down-regulated CCR6 expression in Th17 and CD11b+F4/80hi macrophages in non-obese mouse model of type 1 diabetes.41 Resveratrol treatment also inhibited renal lipotoxicity, inflammation and oxidative stress by enhancing AMPK-SIRT1-PGC1a signaling.42 Taken together, this evidence emphasizes the key role of resveratrol in prevention of T2D by decreasing the inflammatory responses. Studies also suggested that an aberrant cytokine production from liver innate immune cells could be the common pathogenic mechanisms for hepatic insulin resistance and NAFLD.43 It was reported that pathological stimuli such as lipid accumulation could lead to over-activation of the inflammatory pathways (TNF-a, IL-12, and IL-6) in the liver tissue.43 Several human and animal studies have demonstrated that TNF-a is a key factor in the development of hepatic insulin resistance and NAFLD. In this regard, Hotamisligil et al. demonstrated the relationship between TNF-a expression and insulin resistance in nonalcoholic steatohepatitis (NASH).44 In addition, a positive correlation was reported between the degree of liver fibrosis and circulating TNF-a levels in patients with NASH.45 Moreover, Li et al. reported that treatment with antiTNF-a antibodies can improve hepatic insulin resistance and NAFLD induced by a high-fat diet in mice.46 Based on these reports, alleviating hepatic inflammation through the diet, exercise, lifestyle modifications and pharmacological intervention may be helpful in treatment of NAFLD and insulin resistance.47 In this regard, chronic supplementation of resveratrol in NAFLD patients was associated with a significant reduction in inflammatory markers (hs-CRP, IL-6, and NF-kB) and cytokeratin-18 (as a biomarker of hepatocellular apoptosis).19 In another study, resveratrol significantly reduced the levels of TNF-a, and fibroblast growth factor 21 (FGF-21) and increased adiponectin levels in patients with NAFLD.26 At the molecular level, several targets including SIRT1, NF-kB, cyclooxygenase 2 and NOD-like receptor pyrin domain-

10

containing-3 (NLRP3) inflammasome have been proposed for the anti-inflammatory effect of resveratrol in liver tissue. The role of inflammation in the process of CVDs has been increasingly well recognized over the past decade. Inflammation has a significant role at all stages of atherosclerosis including initiation, progression and plaque formation.48 Inflammation is also a key player in myocardial ischemia. TNF-a can trigger the inflammatory reaction caused by myocardial ischemia-reperfusion.49 Resveratrol was reported to have the beneficial and protective effects against most degenerative and cardiovascular diseases, including ischemia/reperfusion injury, atherosclerosis, hypertension, and heart failure. Resveratrol treatment has been shown to inhibit IL-6 and IL-8 production in human coronary artery smooth muscle cells,50 inhibit IL-6 release by stimulated macrophages, and reduce the serum levels of IL-1b, IL-6, and TNF-a in an atherosclerotic rabbit model.51 Resveratrol has been shown to decrease the expression of adhesion molecules (intercellular adhesion molecule-1, ICAM-1, and vascular cell adhesion molecule-1, VCAM-1) via inhibition of NF-kB pathway activation.52 The studies showed that resveratrol may also protect the myocardium against ischemia-reperfusion injury by inhibiting the expression and activation of the NLRP3 inflammasome.53 Moreover, resveratrol suppressed the inflammatory reaction in the rat heart with myocardial ischemiareperfusion injury by inhibiting TLR4/NF-kB signaling, activation of Nrf2/ARE pathway and inhibiting neutrophil infiltration,suppressing TNF-a expression and increasing NO production.49,54,55,49 Collectively, these studies demonstrated that resveratrol can ameliorate inflammatory state in cardiovascular diseases. In this study, we found no significant evidence for the effect of resveratrol on IL-6. Though the inhibitory effects of resveratrol on IL-6 are clearly demonstrated in vitro studies,56 the overall effect of resveratrol on this factor in this meta-analysis is somewhat unexpected. Due to the strong link between TNF-a and IL-6, if resveratrol supplementation is beneficial in reducing TNFa levels, we expected to observe a decreasing influence of resveratrol on IL-6 levels. It remains unclear why we did not find a preventive effect of resveratrol on IL-6 levels; however, it is likely that a relatively small number of pooled participants provide not enough statistical power to estimate the intervention effect. We enhanced statistical power to examine the source of heterogeneity using Higgins I-square and

Volume & Number &

ARTICLE IN PRESS M. Koushki et al. subgroup analysis. Our analysis suggested the differences in the effect of resveratrol use on inflammatory markers by factors such as type of sample and study duration. However, given the low quality and the limited number of trials available for the subgroup analysis, these findings might need to be investigated in further studies. Our systematic review and meta-analysis has several strengths. This review obtained the most comprehensive data on the effect of resveratrol supplementation on inflammation. Our search strategy was very detailed and spanned multiple databases. Statistical testes showed no evidence of publication bias in our analyses. A detailed assessment of the methodological quality of the included trials using Jadad scale revealed that the most of the RCTS had low risk of bias. We also conducted the subgroup analysis to explore the possible sources of the heterogeneity. Finally, our results remained robust in sensitivity analysis. Our systematic review and meta-analysis also had several limitations: First, due to the small number of the studies, we were not able to better evaluate the effect of resveratrol on the specific outcomes such as IL-6. Second, we pooled the estimates from studies having different type of sample, type of disease, study duration and also dose of resveratrol, which potentially might lead to bias. Third, there was a substantial heterogeneity between the RCTS. However, the possible sources of the heterogeneity were explored using Higgins I2 and subgroup analysis. Another limitation was that there were different sources of resveratrol that vary in the composition. Some studies used purified trans-resveratrol, while the others applied the extracts containing resveratrol. Given the limitations, the findings of the present review should be interpreted with caution. Furthermore, future intervention studies on the levels of inflammatory markers as a primary outcome are needed to establish the beneficial effect of resveratrol in prevention of inflammation.

TAGEDH1CONCLUSIONTAGEDN Available evidence from RCTS in this meta-analysis suggests a potential preventive effect of resveratrol supplementation on TNF-a and hs-CRP, but not on IL-6 levels. Treatment with resveratrol significantly reduced TNF-a and hs-CRP levels in intervention group compared with placebo group. Further studies are required to validate the effect of resveratrol supplementation on inflammation.

& 2018

TAGEDH1ACKNOWLEDGMENTSTAGEDN

This work was financially supported by grants (96-0130-33989 and -00-30-38294) from the Deputy of Research, Tehran University of Medical Sciences.

TAGEDH1AUTHORS CONTRIBUTIONTAGEDN Reza Meshkani conceptualized and designed the research, critically reviewed the manuscript and approved the final manuscript as submitted. Mehdi koushki and Nasrin Amiri Dashatan, wrote the draft of this paper, the literature searches and Data collection. All the authors contributed to the manuscript writing, statistical analyses and Data interpretation.

TAGEDH1CONFLICTS OF INTERESTTAGEDN The authors have indicated that they have no conflicts of interest regarding the content of this article.

TAGEDH1SUPPLEMENTARY MATERIALSTAGEDN Supplementary material associated with this article can be found in the online version at https://doi.org/ 10.1016/j.clinthera.2018.05.015.

TAGEDH1REFERENCESTAGEDN 1. Meshkani R, Vakili S. Tissue resident macrophages: Key players in the pathogenesis of type 2 diabetes and its complications. Clinica Chimica Acta. 2016;462:77–89. 2. Khodabandehloo H, Gorgani-Firuzjaee S, Panahi G, Meshkani R. Molecular and cellular mechanisms linking inflammation to insulin resistance and b-cell dysfunction. Translational Research. 2016;167:228–256. 3. Su H, Liu R, Chang M, Huang J, Jin Q, Wang X. Effect of dietary alpha-linolenic acid on blood inflammatory markers: a systematic review and meta-analysis of randomized controlled trials. European Journal of Nutrition 2017;1–15. 4. Collaboration ERF. C-reactive protein concentration and risk of coronary heart disease, stroke, and mortality: an individual participant meta-analysis. The Lancet. 2010;375:132–140. 5. Kasperska-Zajac A, Grzanka A, Damasiewicz-Bodzek A, et al. Elevated plasma il-8 concentration is related to severity of systemic inflammation in chronic spontaneous urticaria. Journal of biological regulators and homeostatic agents. 2017;31:957–961. 6. Lujan HL, DiCarlo SE. Physical activity, by enhancing parasympathetic tone and activating the cholinergic antiinflammatory pathway, is a therapeutic strategy to restrain chronic inflammation and prevent many chronic diseases. Medical hypotheses. 2013;80:548–552. 7. Sahebkar A, Serban C, Ursoniu S, et al. Lack of efficacy of resveratrol on C-reactive protein and selected

11

ARTICLE IN PRESS Clinical Therapeutics

8.

9.

10.

11.

12.

13.

14.

15.

cardiovascular risk factors—Results from a systematic review and metaanalysis of randomized controlled trials. International journal of cardiology. 2015;189:47–55. Csiszar A. Anti inflammatory effects of resveratrol: possible role in prevention of age related cardiovascular disease. Annals of the New York Academy of Sciences. 2011;1215:117–122. Smoliga JM, Baur JA, Hausenblas HA. Resveratrol and health a comprehensive review of human clinical trials. Molecular nutrition & food research. 2011;55:1129–1141. Leporini L, Giampietro L, Amoroso R, et al. In vitro protective effects of resveratrol and stilbene alkanoic derivatives on induced oxidative stress on C2C12 and MCF7 cells. Journal of biological regulators and homeostatic agents. 2017;31:589–601. Liu F-C, Tsai Y-F, Tsai H-I, Yu H-P. Anti-inflammatory and organ-protective effects of resveratrol in traumahemorrhagic injury. Mediators of inflammation. 2015;2015. Ghanim H, Sia CL, Korzeniewski K, et al. A resveratrol and polyphenol preparation suppresses oxidative and inflammatory stress response to a high-fat, high-carbohydrate meal. The Journal of Clinical Endocrinology & Metabolism. 2011;96:1409–1414. Tome-Carneiro J, Larrosa M, YanezGasc on MJ, et al. One-year supplementation with a grape extract containing resveratrol modulates inflammatory-related microRNAs and cytokines expression in peripheral blood mononuclear cells of type 2 diabetes and hypertensive patients with coronary artery disease. Pharmacological research. 2013;72:69–82. Moher D, Liberati A, Tetzlaff J, Altman DG, Group P. Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. PLoS medicine. 2009;6 e1000097. Tome-Carneiro J, Gonzalvez M, Larrosa M, et al. Grape resveratrol increases serum adiponectin and downregulates inflammatory genes in

12

16.

17.

18.

19.

20.

21.

22.

peripheral blood mononuclear cells: a triple-blind, placebo-controlled, oneyear clinical trial in patients with stable coronary artery disease. Cardiovascular Drugs and Therapy. 2013;27:37–48. Tome-Carneiro J, Gonz alvez M, Larrosa M, et al. One-year consumption of a grape nutraceutical containing resveratrol improves the inflammatory and fibrinolytic status of patients in primary prevention of cardiovascular disease. American Journal of Cardiology. 2012;110:356– 363. Agarwal B, Campen MJ, Channell MM, et al. Resveratrol for primary prevention of atherosclerosis: clinical trial evidence for improved gene expression in vascular endothelium. International journal of cardiology. 2013;166:246–248. Brasny o P, Moln ar GA, Moh as M, et al. Resveratrol improves insulin sensitivity, reduces oxidative stress and activates the Akt pathway in type 2 diabetic patients. British Journal of Nutrition. 2011;106:383–389. Faghihzadeh F, Adibi P, Rafiei R, Hekmatdoost A. Resveratrol supplementation improves inflammatory biomarkers in patients with nonalcoholic fatty liver disease. Nutrition research. 2014;34:837–843. Zahedi HS, Jazayeri S, Ghiasvand R, Djalali M, Eshraghian MR. Effects of polygonum cuspidatum containing resveratrol on inflammation in male professional basketball players. International journal of preventive medicine. 2013;4:S1. Ghanim H, Sia CL, Abuaysheh S, et al. An antiinflammatory and reactive oxygen species suppressive effects of an extract of Polygonum cuspidatum containing resveratrol. The Journal of Clinical Endocrinology & Metabolism. 2010;95:E1–E8. Militaru C, Donoiu I, Craciun A, Scorei ID, Bulearca AM, Scorei RI. Oral resveratrol and calcium fructoborate supplementation in subjects with stable angina pectoris: effects on lipid profiles, inflammation

23.

24.

25.

26.

27.

28.

29.

30.

markers, and quality of life. Nutrition. 2013;29:178–183. Bo S, Ciccone G, Castiglione A, et al. Anti-inflammatory and antioxidant effects of resveratrol in healthy smokers a randomized, double-blind, placebo-controlled, cross-over trial. Current medicinal chemistry. 2013;20:1323–1331. Bo S, Ponzo V, Ciccone G, et al. Six months of resveratrol supplementation has no measurable effect in type 2 diabetic patients. A randomized, double blind, placebo-controlled trial. Pharmacological research. 2016;111:896–905. Samsami-kor M, Daryani NE, Asl PR, Hekmatdoost A. Anti-inflammatory effects of resveratrol in patients with ulcerative colitis: a randomized, double-blind, placebo-controlled pilot study. Archives of medical research. 2015;46:280–285. Chen S, Zhao X, Ran L, et al. Resveratrol improves insulin resistance, glucose and lipid metabolism in patients with non-alcoholic fatty liver disease: a randomized controlled trial. Digestive and Liver Disease. 2015;47:226–232. Yoshino J, Conte C, Fontana L, et al. Resveratrol supplementation does not improve metabolic function in nonobese women with normal glucose tolerance. Cell metabolism. 2012; 16:658–664. Poulsen MM, Vestergaard PF, Clasen BF, et al. High-dose resveratrol supplementation in obese men: an investigator-initiated, randomized, placebo-controlled clinical trial of substrate metabolism, insulin sensitivity, and body composition. Diabetes. 2013;62:1186–1195. Timmers S, Konings E, Bilet L, et al. Calorie restriction-like effects of 30 days of resveratrol supplementation on energy metabolism and metabolic profile in obese humans. Cell metabolism. 2011;14:612–622. Khodabandehloo H, Seyyedebrahimi S, Esfahani EN, Razi F, Meshkani R. Resveratrol supplementation decreases blood glucose without changing the

Volume & Number &

ARTICLE IN PRESS M. Koushki et al.

31.

32.

33.

34.

35.

36.

37.

38.

39.

circulating CD14+ CD16+ monocytes and inflammatory cytokines in patients with type 2 diabetes: a randomized, double-blind, placebo-controlled study. Nutrition Research. 2018;54:40–51. Jadad AR, Moore RA, Carroll D, et al. Assessing the quality of reports of randomized clinical trials: is blinding necessary? Controlled clinical trials. 1996;17:1–12. R Neves A, Lucio M, LC Lima J, Reis S. Resveratrol in medicinal chemistry: a critical review of its pharmacokinetics, drug-delivery, and membrane interactions. Current medicinal chemistry. 2012;19:1663–1681. de Ligt M, Timmers S, Schrauwen P. Resveratrol and obesity: can resveratrol relieve metabolic disturbances? Biochimica et Biophysica Acta (BBA)Molecular Basis of Disease. 2015;1852: 1137–1144. Szkudelski T, Szkudelska K. Anti diabetic effects of resveratrol. Annals of the New York Academy of Sciences. 2011; 1215:34–39. Liu K, Zhou R, Wang B, Mi MT. Effect of resveratrol on glucose control and insulin sensitivity: a metaanalysis of 11 randomized controlled trials. The American journal of clinical nutrition. 2014;99:1510–1519. Liu K, Zhou R, Wang B, Mi M-T. Effect of resveratrol on glucose control and insulin sensitivity: a metaanalysis of 11 randomized controlled trials. The American journal of clinical nutrition. 2014;99:1510–1519. Zhao Y, Chen B, Shen J, Wan L, Zhu Y, Yi T. The Beneficial Effects of Quercetin, Curcumin, and Resveratrol in Obesity. 2017;2017:1459497. Jeon BT, Jeong EA, Shin HJ, et al. Resveratrol attenuates obesity-associated peripheral and central inflammation and improves memory deficit in mice fed a high-fat diet. Diabetes. 2012;61:1444–1454. Jimenez-Gomez Y, Mattison JA, Pearson KJ, et al. Resveratrol improves adipose insulin signaling and reduces the inflammatory response in adipose

& 2018

40.

41.

42.

43.

44.

45.

46.

47.

48.

49.

tissue of rhesus monkeys on high-fat, high-sugar diet. Cell metabolism. 2013; 18:533–545. Kim S, Jin Y, Choi Y, Park T. Resveratrol exerts anti-obesity effects via mechanisms involving down-regulation of adipogenic and inflammatory processes in mice. Biochemical pharmacology. 2011;81:1343–1351. Lee S-M, Yang H, Tartar D, et al. Prevention and treatment of diabetes with resveratrol in a non-obese mouse model of type 1 diabetes. Diabetologia. 2011;54:1136–1146. Kim M, Lim J, Youn H, et al. Resveratrol prevents renal lipotoxicity and inhibits mesangial cell glucotoxicity in a manner dependent on the AMPK SIRT1 PGC1a axis in db/db mice. Diabetologia. 2013;56:204–217. Li Z, Diehl A. Innate immunity in the liver. Current opinion in gastroenterology. 2003;19:565–571. Hotamisligil GS, Shargill NS, Spiegelman BM. Adipose expression of tumor necrosis factor-alpha: direct role in obesity-linked insulin resistance. Science. 1993;259:87–91. Lesmana CRA, Hasan I, Budihusodo U, et al. Diagnostic value of a group of biochemical markers of liver fibrosis in patients with non alcoholic steatohepatitis. Journal of digestive diseases. 2009;10:201–206. Li Z, Yang S, Lin H, et al. Probiotics and antibodies to TNF inhibit inflammatory activity and improve nonalcoholic fatty liver disease. Hepatology. 2003;37:343–350. Yang SJ, Lim Y. Resveratrol ameliorates hepatic metaflammation and inhibits NLRP3 inflammasome activation. Metabolism. 2014;63:693–701. Libby P, Ridker PM, Maseri A. Inflammation and atherosclerosis. Circulation. 2002;105:1135–1143. Li J, Xie C, Zhuang J, et al. Resveratrol attenuates inflammation in the rat

50.

51.

52.

53.

54.

55.

56.

heart subjected to ischemia-reperfusion: Role of the TLR4/NF-kB signaling pathway. Molecular medicine reports. 2015;11:1120–1126. Wakabayashi I, Takeda Y. Inhibitory effects of resveratrol on MCP-1, IL-6, and IL-8 production in human coronary artery smooth muscle cells. Naunyn-Schmiedeberg's archives of pharmacology. 2013;386:835–839. Song R, Li WQ, Dou JL, et al. [Resveratrol reduces inflammatory cytokines via inhibiting nuclear factor-kappaB and mitogen-activated protein kinase signal pathway in a rabbit atherosclerosis model]. Zhonghua xin xue guan bing za zhi. 2013;41:866–869. Deng YH, Alex D, Huang HQ, et al. Inhibition of TNF a mediated endothelial cell monocyte cell adhesion and adhesion molecules expression by the resveratrol derivative, trans208;3, 5, 40 trimethoxystilbene. Phytotherapy research. 2011;25:451–457. Dong W, Yang R, Yang J, et al. Resveratrol pretreatment protects rat hearts from ischemia/reperfusion injury partly via a NALP3 inflammasome pathway. International journal of clinical and experimental pathology. 2015;8:8731. Cong X, Li Y, Lu N, et al. Resveratrol attenuates the inflammatory reaction induced by ischemia/reperfusion in the rat heart. Molecular medicine reports. 2014;9:2528–2532. Cheng L, Jin Z, Zhao R, Ren K, Deng C, Yu S. Resveratrol attenuates inflammation and oxidative stress induced by myocardial ischemia-reperfusion injury: role of Nrf2/ARE pathway. International journal of clinical and experimental medicine. 2015;8:10420. Ren Z, Wang L, Cui J, et al. Resveratrol inhibits NF-kB signaling through suppression of p65 and IkappaB kinase activities. Die Pharmazie. 2013; 68:689–694.

Address correspondence to: Reza Meshkani, PhD, Department of Clinical Biochemistry, Faculty of Medicine, Tehran University of Medical Sciences, Tehran, I.R. Iran. E-mail: [email protected]

13

ARTICLE IN PRESS Clinical Therapeutics

Supplemental Table I.

Quality assessment of selected studies based on Jadad scale.

Authors

Q1

Q2

Q3

Q4

Q5

Q6 (Negative score)

Q7 (Negative score)

Total Score

Jao-Tome carniero Beamon A Brasnyo P Faghihzadeh F Ghanim H Zahedi H Militaru C S.Bo Jao-Tome Samsamikor M Jao-Tome C Shihui Chen Jun Yoshino Poulsen M S.Bo Trimmer S Khodabandehlo H

+1 +1 +1 +1 +1 +1 +1 +1 +1 +1 +1 +1 +1 +1 +1 +1 +1

0 +1 +1 +1 0 +1 +1 +1 +1 +1 +1 +1 +1 +1 +1 +1 +1

+1 +1 +1 +1 0 +1 +1 +1 +1 +1 +1 +1 +1 +1 +1 +1 +1

+1 +1 +1 +1 +1 +1 +1 +1 +1 +1 +1 +1 +1 +1 0 0 +1

+1 +1 +1 +1 0 +1 0 +1 +1 +1 +1 +1 +1 +1 +1 +1 +1

0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

0 -1 0 0 0 0 -1 0 0 0 0 0 0 0 0 0 0

+4 +4 +5 +5 +2 +5 +3 +5 +5 +5 +5 +5 +5 +5 +4 +4 +5

Q: questions

Supplemental Table II.

Subgroup analysis of primary outcome was performed for identification of heterogeneity in included trials based on “type of sample”, “Dose of resveratrol” and “study duration”.

Inflammatory markers

No. of trials

Total No. of trials

WMD (95% Confidence Interval)

Q Statistic P value

6 5

6 11

0.05(-0.26 to 0.35) -0.35(-0.64 to -0.06)

0.00 0.14 0.062

3 1 2 3 1 1

3 2 3 5 1 3

-0.05(-0.40 to o.31) 0.18(-0.41 to o.77) -0.47(-0.92 to -0.01) 0.26(-0.19 to o.70) -0.66(-1.01 to o.38) -0.92(-1.84 to o.01)

I-squared

IL-6 Sample type: Plasma Serum Overall Test for heterogeneity between sub-groups Study duration(months):

12 2

80.9% 41.4%

0.82 0.05 0.00 (continued)

13.e1

Volume & Number &

ARTICLE IN PRESS M. Koushki et al.

Supplemental Table II.

(Continued)

Inflammatory markers 3 1 6 1.5 TNF-a Sample type: Plasma Serum Overall Test for heterogeneity between sub-groups Study duration(months) 12 2 3 1 1.5

hs-CRP Sample type Plasma Serum

No. of trials

Total No. of trials

WMD (95% Confidence Interval)

& 2018

I-squared

-

5 7

611

-0.29(-0.62 to 0.04) -0.39(-0.62 to -0.15)

0.11 0.03 0.63

3 1 2 3 3

3 2 3 5 3

-0.01(-0.36 to 0.35) -0.41(-1.0 to 0.19) -0.40(-0.78 to -0.03) 1.27(-2.30 to -0.25) -0.32(-0.75 to 0.11)

0.44 0.62 0.01 0.5

5 8

6 11

-0.08(-0.42 to 0.26) -0.32(-0.51 to -0.13)

0.001 0.64

46% 57%

0.24

Overall Test for heterogeneity between sub-groups Study duration(months) 12 2 3 1 1.5 6

Q Statistic P value

3 2 2 4 2 1

3 2 3 5 3 1

79% 0.0% -0.11(-0.47 to 0.26) -0.26(-0.64 to 0.13) -0.29(-0.75 to 0.16) -0.20(-0.57 to -0.17) -0.80(-1.31 to -0.29) -0.05(-0.40 to 0.30)

0.39 0.71 0.007 0.21 0.009 -

13.e2

ARTICLE IN PRESS Clinical Therapeutics

Supplemental Figure 1. Subgroup analysis for primary studies. Heterogenity was performed for subgroup of type of sample, and study duration for IL-6. A: IL-6 (type of sample), B: IL-6 (study duration).

13.e3

Volume & Number &

ARTICLE IN PRESS M. Koushki et al.

Supplemental Figure 2. subgroup analysis for primary studies. Heterogenity was performed for subgroup of type of sample, and study duration for TNF-a. A: TNF-a (type of sample), B: TNF-a (study duration).

& 2018

13.e4

ARTICLE IN PRESS Clinical Therapeutics

Supplemental Figure 3. subgroup analysis for primary studies. Heterogenity was performed for subgroup of type of sample, and study duration for hs-CRP. A: hs-CRP (type of sample), B: hs-CRP (study duration).

13.e5

Volume & Number &