Profiling Inflammatory Biomarkers following Curcumin Supplementation: An Umbrella Meta-Analysis of Randomized Clinical Trials

Objective Several meta-analyses have shown that curcumin can reduce inflammatory biomarkers, but the findings are inconsistent. The objective of the present umbrella meta-analysis was to provide a more accurate estimate of the overall effects of curcumin on inflammatory biomarkers. Methods The following international databases were systematically searched until March 20, 2022: PubMed, Scopus, Embase, Web of Science, and Google Scholar. A random-effects model was applied to evaluate the effects of curcumin on inflammatory biomarkers. Meta-analysis studies investigating the effects of curcumin supplementation on inflammatory biomarkers with corresponding effect sizes (ES) and confidence intervals (CI) were included in the umbrella meta-analysis. GRADE (Grading of Recommendations Assessment, Development, and Evaluation) was used to evaluate the certainty of evidence. Results A meta-analyses of ten studies with 5,870 participants indicated a significant decrease in C-reactive protein (CRP) (ES = −0.74; 95% CI: −1.11, −0.37, p < 0.001; I2 = 62.1%, p=0.015), interleukin 6 (IL-6) (ES = −1.07; 95% CI: −1.71, −0.44, p < 0.001; I2 = 75.6%, p < 0.001), and tumour necrosis factor α (TNF-α) levels (ES: −1.92, 95% CI: −2.64, −1.19, p < 0.0; I2 = 18.1%, p=0.296) following curcumin supplementation. Greater effects on CRP and TNF-α were evident in trials with a mean age >45 years and a sample size >300 participants. Conclusion The umbrella of meta-analysis suggests curcumin as a promising agent in reducing inflammation as an adjunctive therapeutic approach in diseases whose pathogenesis is related to a higher level of inflammatory biomarkers.


Introduction
Infammation is a physiological response of the immune system induced by various harmful factors such as pathogens, injury, or damaged cells to restore body function [1,2]. Chronic infammation is typically characterized by a high number and excessive activation of innate immune cells in tissues and increased release of infammatory mediators and chemokines at local and systemic levels. In the context of infammation, interleukin 6 (IL-6) and tumor necrosis factor (TNF-α) are two cytokines that are released in signifcant amounts. Tey have been proven to be powerful inducers of C-reactive protein (CRP), which is a major acute-phase reactant produced mostly in the liver and strongly associated with metabolic diseases. It has been widely recognized that autoimmune diseases, atherosclerosis, asthma, type 2 diabetes (T2D), and other conditions can be caused by chronic infammation [3][4][5].
A wide range of pharmacotherapies can improve infammatory biomarkers, but they are also known to cause complications and side efects. Hence, nutraceutical therapies such as dietary supplements can be considered adjunctive or alternative treatments that can be utilized as antiinfammatory agents [6][7][8]. Curcumin is remarkable bioactive polyphenol extracted from the rhizome of turmeric (Curcuma longa) [9,10]. Curcumin has a wide range of medicinal efects such as hepatoprotective, antimicrobial, antiinfammatory, antioxidant, and antitumor activities [11,12].
Despite promising results, some studies have found that curcumin supplementation has no considerable efcacy on infammatory markers [18][19][20]. Taking into account of inconsistency, the current umbrella of meta-analysis was designed to reevaluate the anti-infammatory efects of curcumin.

Methods
According to the guiding principle of the Preferred Reporting Items for Systematic Reviews and Meta-analysis (PRISMA), this study was carried out and reported [21]. Te protocol of this study has been registered in the International Prospective Register of Systematic Reviews (PROSPERO) under number CRD42022323546.

Search Strategy and Study Selection.
Relevant studies published on international scientifc databases including PubMed, Scopus, Embase, Web of Science, and Google Scholar were searched until March 20, 2022. Te search strategy was developed using the following MeSH terms and keywords: ("curcumin" OR "curcuminoid" OR "turmeric") AND ("C-Reactive Protein" OR "crp" OR "hs-crp" OR "high sensitivity C-reactive protein" OR "Tumor Necrosis Factoralpha" OR "tumor necrosis factor-α" OR "tnf-alpha" OR "tnf-α" OR "Interleukin-6" OR "IL-6" OR "infammation") AND ("systematic review'' OR "meta-analysis") Te search strategy is shown in Suppl. Table 1. To increase the sensitivity of the search strategy, the wild-card " * " term was used. Also, the articles were limited to English.

Inclusion and Exclusion Criteria.
Meta-analysis studies investigating the efects of curcumin supplementation on infammatory biomarkers including CRP, IL-6, and TNF-α with corresponding efect sizes (ES) and confdence intervals (CI) were included in the umbrella meta-analysis. In addition, the following studies were excluded: in vitro, in vivo, and ex vivo studies; case reports; observational studies; quasi-experimental studies; controlled clinical trials; and studies with the lowest quality score.

Methodological Quality Assessment. A Measurement
Tool to Assess Systematic Reviews (AMSTAR) 2 questionnaire was utilized by two independent researchers (VM and ZK) to assess the methodological quality of included metaanalyses of randomized controlled trials (RCTs) [22]. Disagreements were resolved by consensus with a third researcher (AO). Te instrument (AMSTAR 2) retains 10 of the original domains and has 16 items in total. Te questionnaire consists of 16 questions asking the reviewers to answer "yes" or "partial yes" or "no" or "no meta-analysis". Te AMSTAR 2 checklist was categorized into "critically low quality," "low quality," "moderate quality," and "high quality." 2.3.1. Grading of the Evidence. We evaluated the overall strength and quality of evidence using the GRADE tool based on the Cochrane Handbook of systematic reviews of interventions. Tis tool consists of fve factors: bias risk, consistency of results, directness, precision, and publication bias. GRADE results were categorized as "high," "medium," "low," and "very low." When one of the above factors is not met, the quality of a level decreases [23].

Study Selection and Data Extraction.
Two independent reviewers (VM and AHM) screened the articles based on the eligibility criteria. In the frst step, the title and abstract of the articles were reviewed. Ten, the full-text of relevant articles was assessed to ascertain study suitability for inclusion in the umbrella meta-analysis. Any disagreements were resolved through the consensus with the senior author (AO).
Te frst authors' name, publication year, sample size, study location, dose and duration range of supplementation, ESs, and CIs for CRP, IL-6, and TNF-α were extracted from the selected meta-analyses.

Data Synthesis and Statistical Analysis.
To estimate the pooled efect size, ESs and the corresponding CIs were utilized. I 2 statistic and Cochrane's Q-test were applied to detect heterogeneity, and I 2 value > 50% or p < 0.1 for Q-test was considered as meaningful heterogeneity among studies. Te restricted maximum likelihood method was carried out to conduct statistical analysis based on random-efects model. Based on predetermined variables consisting of duration of intervention, mean age of participants, dose of curcumin, and sample size, subgroup analyses were conducted to detect potential sources of heterogeneity. Sensitivity analysis was conducted to examine the impact of one individual study removal on the pooled efect size. Publication bias was formally assessed using funnel plots and the Egger test if the number of included datasets was ten or higher; otherwise, only Begg's test was reported. STATA version 16.0 was used to carry out all statistical analyses (Stata Corporation, College Station, TX, US). P value less than 0.05 was considered signifcant.

Selected Studies and Systematic
Review. Te fow diagram of the literature search process is demonstrated in Figure 1. Based on a systematic search of electronic databases, 105 articles were found. Upon removing duplicate articles and careful screening based on titles and abstracts, 19 articles were included, among which nine articles were excluded after full-text review. Overall, 10 articles met the inclusion criteria to be eligible for the umbrella meta-analysis. Te characteristics of the included studies are reported in Table 1. Te mean age of participants was between 29 and 52 years. Te included studies were conducted from 2013 to 2021 in Iran [24][25][26][27][28][29], United States [30,31], Australia [32], and Italy [33]. Curcumin was administered in a wide range of doses from 300 to 1900 mg, lasting from 4.5 to 10.5 weeks. Cochrane risk of bias tool and Jadad scores were used for quality assessment. Te quality of the eligible RCTs in the meta-analyses is summarized in Table 1.

Assessment of Risk of Bias.
According to AMSTAR 2, out of ten meta-analyses included in the current umbrella review, three and seven articles have high-and moderatequality, respectively. Te results of quality assessment of meta-analyses are reported in Table 2.

Te Efects of Curcumin on CRP Levels.
Overall, seven meta-analyses on 3,271 participants revealed a signifcant reduction in CRP following curcumin supplementation ( Figure 2). Tere was a signifcant between-study heterogeneity (I 2 � 62.1%, p � 0.015). Tus, subgroup analysis was performed based on various variables. Te mean age, sample size, intervention duration, and dose were potential sources of between-study heterogeneity ( Table 3). Subgroup analysis showed that subjects with >45 years of age and a sample size >300 contributed to a greater decrease in the CRP level (Table 3). Sensitivity analysis revealed that the pooled results on the efects of curcumin supplementation on CRP levels Studies included in Umbrella meta-analysis ( n =10) Full-text articles excluded, with reasons (n = 9) i) did not provide information's (n=2) ii) Other irrelevant studies (1) iii) Non appropriate design (1) iv) Systematic review (5)    (small study bias) and discuss its likely impact on the results of the review? (16) Did the review authors report any potential sources of conficts of interest, including any funding they received for conducting the review? Each question was answered with "yes," "partial yes" or "no." When no meta-analysis was done, question 11, 12, and 15 were answered with "no meta-analysis conducted."

Evidence-Based Complementary and Alternative Medicine
Evidence-Based Complementary and Alternative Medicine 5 did not depend on one particular study. No signifcant publication bias was using Begg's test (p � 0.452). Te quality of evidence for CRP was rated moderate (Table 4).

Te Efects of Curcumin on IL-6 Levels.
Te efect of curcumin on IL-6 levels was reported in six studies on 2,972 individuals. Te analysis showed a signifcant decrease in IL-6 levels by curcumin supplementation (Figure 3). Te amount of heterogeneity was high (I 2 � 75.6%, p < 0.001), so that mean age, duration, intervention dose, and sample size were detected as the sources of it (Table 3). Curcumin supplementation ≤700 mg/day among subjects with age 45 years or younger, intervention duration ≤7 weeks, and sample size ≤300 contributed to a greater decrease in IL-6 level (Table 3). Sensitivity analysis demonstrated no evidence of a signifcant efect of a single study on the overall efect size. Tere was no signifcant publication bias by Begg's test (p � 0.452). Using the GRADE tool, the overall quality of evidence for TNF-α was classifed as moderate (Table 4).

Te Efects of Curcumin on TNF-α Levels.
Pooled data from six studies with 3,224 participants revealed a signifcant lowering efect of curcumin on TNF-α levels ( Figure 4). Tere was no signifcant heterogeneity between studies (I 2 � 18.1%, p � 0.296). Subgroup analysis showed a more prominent efect of curcumin supplementation on TNF-α levels in the sample size >300, mean age >45 years, intervention duration >7-week, and dose >700 mg/day ( Table 3). Sensitivity analysis revealed that no special study afected signifcantly the pooled efect size. Te results of Begg's test were not signifcant for detecting publication bias (p � 0.060). Te GRADE tool rated the overall quality of evidence for IL-6 as moderate quality (Table 4).

Summary of Study Objectives and Procedures.
In the present umbrella review, we consider the results of 10 systematic review meta-analyses of clinical trials that have studied the efects of curcumin supplementation on infammatory factors including CRP, IL-6, and TNF-α levels.
Most of the included systematic reviews 60%) have been published during three year ago (2019-2022). Most of the included studies reported the weighted mean diference (WMD) as the efect size. In the present study, the quality assessment of systematic reviews was assessed using AMSTAR 2. Most of the included studies 70%) in the fnal analysis had moderate quality, and 30% of the studies were scored as high quality (Table 2). Failure to comply with the PICO strategy in defning research questions, the possibility of conficts of interest due to not reporting of fnancial sources, failure to consider high heterogeneity in the interpretation of results that can lead to less accurate conclusion, and failure to examine the small study efect in their analysis that can lead to less reliable results were the major faws in some studies, resulting in their moderate quality.

Evaluation of Current Evidence.
Overall, curcumin supplementation had a signifcant efect on CRP, IL-6, and TNF-α levels; however, there was a high level of heterogeneity in pooled CRP and IL-6 levels. Terefore, results on TNF-α were more reliable due to low heterogeneity. However, sources of high heterogeneity were determined using subgroup analysis, with age, study duration, sample size, and administered dose being possible sources. Terefore, the detected heterogeneity was related to clinical and methodological issues, not statistical problems resulting in high reliability of the results. Tis means that these  subgroups had a diferent true efect. In terms of mean age, subgroup analysis showed that curcumin supplementation in people who are older than 45 years had a signifcant decreasing efect on infammatory biomarkers in comparison with younger participants. As oxidative imbalance and infammation increase with aging, more improving efect of curcumin on infammation in this subgroup was not a surprising fnding. According to this promising fnding, curcumin can be considered as a useful agent for longevity through the reduction of oxidative stress [34]. In terms of sample size, subgroup analysis based on sample size indicated that both sample sizes ≤300 and >300 had sufcient power to detect signifcant results. Terefore, our results can be reliable. Subgroup analysis based on dose and duration revealed that the efect of curcumin on infammation was not dose-or time-dependent. In fact, both high and/or low doses, and long-and/or short-term curcumin supplementation can have anti-infammatory efects. According to the US Food and Drug Administration (FDA) report, curcumin has been considered as "Generally Recognized as Safe" (GRAS) even at doses between 4000 and 8000 mg/day [35,36]. Te bioavailability of curcumin is poor in the body, due to low absorption, rapid metabolism, limited tissue distribution, and short half-life [37]. Terefore, some studies have used some compounds such as piperine to increase the bioavailability of curcumin [38,39]. Te insignifcant efect   MIDs used for each outcome were 3.16 mg/l for CRP, 7.9 pg/ml for TNF-α, and 2 pg/ml for IL-6. (e) Downgraded if there was evidence of publication bias using funnel plot. (f ) Since all included studies were meta-analysis of randomized controlled trials, the certainty of the evidence was graded as high for all outcomes by default, and then downgraded based on prespecifed criteria. Quality was graded as high, moderate, low, or very low. 8 Evidence-Based Complementary and Alternative Medicine of curcumin on CRP in >700 mg/day doses and ≤7-week durations may be related to the study of Sahebkar et al. [27] who focused on the efect of curcuminoids on CRP level with characteristics of the mentioned subgroups [27]. Tis focus on only the type of curcuminoid supplement could potentially lead to high heterogeneity compared to other studies that include the investigations on curcumin/curcuminoids. Curcuminoids consist of three compounds including curcumin, demethoxycurcumin, and bisdemethoxycurcumin [11]. Our study population included people with NAFLD, dyslipidemia, metabolic syndrome, obesity, osteoarthritis, and migraine. Te benefcial efect of curcumin on various health conditions in some studies shows that the anti-infammatory efect of curcumin is not dependent on the disease. However, due to the diversity of diseases, subgroup analysis was not possible. As a result, no defnite conclusions can be drawn. In terms of gender, participants in most of the included studies were of both genders. Terefore, it can be concluded that curcumin has anti-infammatory efects on both genders.
Curcumin has a polyphenol nature; therefore, its antiinfammatory mechanisms may be due to its antioxidant   Evidence-Based Complementary and Alternative Medicine properties. Many properties of curcumin such as antioxidant, anti-infammatory, antimicrobial, and antimutagenic attribute to the presence of hydroxyl and methoxy groups in the curcumin structure [40]. One of the most important factors in regulating innate immunity, cell proliferation, cell survival, and infammation is nuclear factor-kappa-B (NF-κB). Tis transcription factor exists in an inactive state in the cytoplasm; however, the activated form translocates from the cytoplasm to the nucleus [41]. It has been reported that curcumin can suppress NF-κB activation as a proinfammatory transcription factor and prevent its translocation, which contribute to an inhibition of gene expression of several infammatory cytokines such as IL-6, IL-2, and TNF-α [36,42]. Curcumin suppresses the IκB kinase (IKK) signaling complex responsible for IκB phosphorylation, thereby inhibiting NF-κB activation [43]. H 2 O 2 is one of the molecules produced during oxidative stress and directly activates NF-κB [44]. Curcumin can also inhibit NF-κB activation by reducing oxidative stress and free radicals due to its antioxidant characteristics [45]. Also, it has been reported that dietary polyphenols have a benefcial efect on nuclear factor erythroid 2-related factor 2 (Nrf2) and internal antioxidant status, through which they play the important roles in balancing oxidative stress. Nrf-2 in an inactive form binds to the Kelch-like ECH-associated protein 1 (Keap1) and is present in the cytoplasm [46]. Dietary polyphenols interact with the amino acid cysteine and cause the dissociation of Keap1 from Nrf2, thereby translocating Nrf2 to the nucleus. Subsequently, Nrf-2 binds to antioxidant response elements (AREs) that are responsible for stimulating the expression of genes involved in the antioxidant system such as superoxide dismutase (SOD), catalase (CAT), and glutathione peroxidase (GPx) [47,48]. Curcumin may also act as a TNF-α blocker by connecting directly to TNF-α [49]. Furthermore, curcumin is able to inhibit the expression and release of proinfammatory factors through interacting with receptors and signaling pathways [50]. In addition, curcumin can inhibit cyclooxygenase-2 gene expression, which is one of the main enzymes involved in infammatory pathways [51]. In addition to the anti-infammatory efects of curcumin mentioned above, which can afect CRP levels as a general indicator of infammation, curcumin also interacts with signaling pathways leading to CRP production. Zhang et al. reported that curcumin reduces CRP production in vascular smooth muscle cells by inhibiting the reactive oxygen species-ERK1/2 pathway [52]. ERK1/2 (extracellular signalregulated protein kinase) has a pivotal role in delivering extracellular signals to the nucleus [53]. Based on in silico approaches, it was found that curcumin also has a direct interaction with CRP through binding to GLN 150 and ASP 140 sites [54].
To our knowledge, the present study is the frst umbrella review of meta-analysis studies on clinical trials investigating the efects of curcumin supplementation on infammation. All included studies had a moderate to high quality score based on the AMSTAR 2 tool. Furthermore, subgroup analysis was conducted to adjust for heterogeneity and evaluate the efect of curcumin in diferent conditions. Tis study has some limitations that must be mentioned. Te frst is the repetition of some studies in diferent metaanalyses, which can afect the fnal result. However, further assessments showed that the repeated studies did not have much weight on the fnal result. Second, the interpretation of our results must be with precaution due to high betweenstudy heterogeneity. However, possible sources of heterogeneity were studied by subgroup analysis. Tird, due to the wide range of diseases that systematic reviews incorporated in their studies, subgroup analysis based on the study population was not possible. Fourth, only published systematic review and meta-analysis studies were considered for the fnal analysis in the present umbrella review. Tis approach might lead to the missing of a number of published clinical trials. Te current umbrella systematic review and meta-analysis had several strengths. First, a comprehensive subgroup analysis was conducted to fnd possible sources of heterogeneity and bias. Second, the data quality of the current study was thoroughly evaluated to present reliable results. Tird, the GRADE tool was employed to assess the certainty of the evidence.

Conclusion
Te present study shows that curcumin has reducing efects on IL-6, CRP, and TNF-α levels. Older people benefted more from curcumin supplementation. Terefore, curcumin can be considered as a useful agent for longevity through decreasing oxidative stress. Te anti-infammatory efect of curcumin was not dose-and time-dependent.

Data Availability
No data were used to support this study.

Conflicts of Interest
Te authors declare that they have no conficts of interest.