This systematic review was registered on the International Prospective Register of Systematic Reviews (PROSPERO) platform, with registration numbers CRD42020164513 and CRD42020168827 for the review with human studies and studies with animal models, respectively.

The search was conducted until January 2021 in the following databases: MEDLINE (via PubMed), Cochrane Controlled Register of Trials (CENTRAL), Literatura Latino-Americano e do Caribe em Ciências da Saúde (LILACS), Science Direct, and Clinical Trials. The following keywords were used: “inflammatory bowel disease”, “ulcerative colitis”, “colitis”, “Crohn’s disease”, “curcumin”, “curcuma”, “turmeric”, and “Indian saffron”. Boolean operators “OR” and “AND” were used. The full search strategy for all the databases is reported in the Supplementary Material. All records retrieved had their titles and abstracts evaluated. In addition, no year of publication filter was used. Then, we evaluated titles for the removal of duplicate records. A complete search strategy is shown in Suppl. Box 1.

Studies with experimental models (rats or mice) of UC or CD were included, without restriction for the inducing agent or the inflammation model (acute or chronic) and in which Curcuma longa/curcuminoids were administered orally (diet or gavage) for treatment. The results of the research should include at least one of the following aspects: macroscopic, anatomical and/or histological evaluation of the colon; body weight; disease activity index; and serological and/or tissue analysis of biomarkers of nitroxidative stress and inflammation. Studies carried out exclusively in vitro, with the use a mixture of turmeric, and/or curcuminoids with other substances as an intervention, were excluded.

Studies were divided according to Curcuma longa/curcumin and/or other curcuminoids supplementation doses: ≤0.005 mg/kg day/0.01 mmol/kg day; 5–50 mg/kg day/0.01–0.15 mmol/kg day; 60–200 mg/kg day/0.25–0.50 mmol/kg day; >200 mg/kg day/0.54 mmol/kg day; ≤2% (w/w) of the diet; and >2% (w/w) of the diet. In studies that have tested more than one dose we chose to categorize it in the subgroup of higher dosages.

Randomized clinical trials with participants of both sexes, aged 18 years or older, diagnosed with UC or DC, and treated with oral Curcuma longa/curcuminoids isolated or combined with drugs, were included. There was no restriction on the severity of the disease (mild, moderate, or severe) or the intestinal lesion location (proximal or distal). Studies were excluded if they evaluated pregnant or lactating women or participants with other associated comorbidities, such as diabetes and hepatic, kidney, and autoimmune diseases. Finally, records in languages other than English, Portuguese and Spanish were excluded.

The following data were extracted from the studies:

• Animal model research: sex, species, age (week), and/or body weight (b.w.); experimental model of IBD; supplement presentation; doses, time of supplementation; administration via; groups of treatment; nitroxidative stress and inflammation effects; other results. The studies with multiple doses of supplementation were allocated, according to the highest dose.

• RCT: IBD clinical situation; number of randomized individuals (n)/age (years)/and sex; supplement presentation/doses, and time of supplementation; treatment association; clinical, histological, and image parameters, nitroxidative stress/inflammation/serum and fecal biomarkers effects.

• Meta-analysis: RCTs included in the meta-analysis needed to present data on clinical remission (primary outcome) and endoscopic remission (secondary outcome). Outcomes that assessed disease activity based on one or more of the following parameters: clinical manifestations, histological patterns of intestinal lesions, inflammation/nitroxidative stress biomarkers, and serological or fecal markers were also included. The percentage of patients in the intervention and control groups who achieved remission and/or clinical response at the end of the experiment were statistically analyzed, using tools that assign scores based on the intensity and frequency of symptomatic manifestations. Other analyzed outcomes included the percentage of patients in the intervention and control groups who achieved remission as assessed by endoscopic examinations at the end of the test period. Trials were classified as ITT (intention-to-treat) or PP (per protocol). Missing data were requested to the authors of the studies by e-mail, when applicable.

For animal model research, the Systematic Review Centre for Laboratory Animal Experimentation (SYRCLE) was applied. This tool assesses the risk of bias, according to ten domains: bias in sequence generation, bias due to baseline characteristics, bias according to allocation concealment, bias due to random housing, bias in blinding of trial caregivers and researchers, bias in random outcome assessment, bias in blinding of outcome assessors, bias due to incomplete outcome data, bias in selective outcome reporting, and overall bias. Each category was assessed as having low, high, or uncertain risk of bias. Then, each study received an overall rating of risk of bias.

For RCTs, the risk of bias was assessed using the Cochrane Collaboration tool, which uses six domains: sequence generation, allocation concealment, blinding of participants and researchers, blinding of outcome assessors, missing data, and selective outcome report. Each category was defined as low, high risk or uncertain risk of bias, and introduced, in each study. The evidence quality was also assessed by the method proposed by the Grading of Recommendations Assessment, Developing and Evaluation (GRADE). This method classifies the evidence of each outcome in the meta-analysis into four categories: high, moderate, low, or very low. Five criteria were evaluated: study limitations (risk of bias), inconsistency of the results (heterogeneity), indirect evidence, inaccuracy, and publication of bias, which generated a score to allow the final classification.

As all the meta-analyzed variables were categorized, the relative risk (RR) between groups for each variable was calculated for each study. Studies weights were assigned, according to the inverse variance method, and calculations were based on a random-effects model. An alpha value of 0.05 was adopted.

Statistical heterogeneity among the studies was tested using the Cochran Q test, and inconsistency was assessed using I² statistics. Whenever a result showed heterogeneity, it was explored by repeating the analysis with the removal of one study at a time to assess whether a particular study explained the heterogeneity. All analysis were conducted using the RevMan 5.3 program (The Nordic Cochrane Centre, The Cochrane Collaboration, Denmark).

From the nine hundred and ninety-seven unique screened records, sixty-two were included in the qualitative synthesis. Of these, fifty-four (87.1%) studies were in animal models, and eight (12.9%) were randomized controlled trials (RCTs). Five (62.5%) RCTs were included in the meta-analysis for the following outcomes: clinical remission and clinical response, and three RCTs (37,5%) were included for endoscopic remission. Fig. 1 shows the flow diagram of study selection.

Among the fifty-four studies (Table 1) that evaluated oral Curcuma longa or curcumin (extract or modified formulations) in experimental colitis, twenty-seven used as inducing agent, the dextran sulfate sodium (DSS) (50%), thirteen used 2,4,6-trinitrobenzenesulfonic acid (TNBS) (24.1%), seven used acetic acid (13%), four used genetic modification (7.4%), and three used dinitrobenzene sulfonic acid (DNB) (5.6%). Curcumin was the main form of supplementation used in the studies (n = 52; 96.3%). Same studies used modified curcumin to improve its bioavailability (n = 20; 37%): nanoparticles/nanocarriers in seven (35%), microparticles/microspheres in six (30%), polymers complexed with other substances in two (10%), and other forms in three (15%). In the studies in which these modified forms were compared to curcumin (n = 12; 60%), all showed more beneficial results in oxidative stress/inflammation markers, general and histological parameters, or in both.

However, even in cases where pure curcumin was used, the majority of studies confirmed its antioxidant and anti-inflammatory roles as well as its effectiveness in improving clinical, metabolic, macroscopic, and histological parameters. Only two studies showed no beneficial effects or negative action of curcumin. In this latter study, a worsening of all biochemical parameters of anemia was observed, in addition to the worsening of clinical and histological signs of IBD.

There was a great variability in the included studies of the curcumin doses (0.005 mg/kg day to 500 mg/kg day), supplementation period (three days until eighteen weeks), and the moment when supplementation was performed (before, during or after colitis induction, or in two different moments of the disease).

The most used doses of curcumin (modified or not)/Curcuma longa were 100 mg/kg day, followed by 50 mg/kg day, and 15 mg/kg day. Among those who were supplemented in the diet (w/w), six had ≤0.3% (w/w) and five ≥2%. Multiple doses were used in twelve (22.2%) studies. In these cases, only one reported better results at the lowest dose (2 mg/kg day vs. 15 mg/kg day), using curcumin nanocarriers; in most of them, the highest dose led to better results. These results indicate that the action of curcumin may be dose-dependent, which would be related to its low toxicity and bioavailability.

The effect of oral curcumin/Curcuma longa supplementation on inflammatory and nitroxidative stress biomarkers was not evaluated in eight (14.8%) studies. Among those who analyzed these parameters (n = 46; 85.2%), forty-five (45, 97.8%) observed some beneficial effects. The main inflammatory biomarkers studied were the enzyme myeloperoxidase (MPO); the cytokines tumor necrosis factor alpha (TNF-α), the IL-6, IL-10, IL-1β and interferon gamma (INF-γ); the nuclear factors, NF-κB and signal transducer and activator of transcription 3 (STAT3); and the membrane receptor toll-like receptor 4 (TLR-4).

Concerning redox imbalance biomarkers, the reactive species-producing enzymes iNOS and COX-2; the enzymatic and nonenzymatic defense biomarkers catalase (CAT), SOD, glutathione peroxidase (GPx), and reduced glutathione (GSH); the cell membrane damage biomarkers thiobarbituric acid reactive substances (TBARS)/malondialdehyde (MDA); and the reactive species nitric oxide (•NO)/nitrite are highlighted.

The risk of bias assessment, according to SYRCLE, is shown in Suppl. Table 1. In general, there was a high risk of bias in the domains of “random sequence generation”, “allocation concealment”, “blinding”, and “incomplete outcome data”. Another consideration refers to the high number of studies whose main objective was to produce and test formulations containing curcumin. In such publications, the manuscript, specifically, on animal experimentation tends to be less detailed, which can result in the assessment of risks as “high” or “unclear” bias in several domains.

In the systematic review, eight RCTs (Table 2) were identified. Six (75%) included, in its supplementation protocol, only individuals with UC and two (25%), only patients with CD. Most of them (n = 6; 75%) chose to study mild or moderate IBD, while Bommelaer et al. (2020) analyzed CD surgical cases, and Kumar et al. (2020) comprised UC active forms. In all RCTs, individuals of both sexes were included.

In three RCTs (3, 37.5%), modified curcumin/curcuminoids were utilized to increase its bioavailability; in other three RCTs (37.5%), pure curcumin was used; and in two RCTs (25%), Curcuma longa extract was evaluated. Doses and supplementation period showed a high variation between RCTs. The doses used ranged from 100 mg/day to 10 g/day, while the supplementation period varied from one month to six months. All included RCTs used an association of oral supplementation with drug therapy, especially mesalamine, either in the oral or oral + rectal route.

Finally, positive results in clinical, histological, and imaging parameters in the curcumin/Curcuma longa oral supplementation group versus the placebo group were observed in five (5, 62.5%) RCTs. The better effects were observed in the clinical remission rate, clinical response rate, endoscopic remission rate, and quality of life. Bommelaer et al. (2020), that tested 3 g/day, were the only to observe negative effects related to an increase in the severe endoscopic postoperative recurrence rate (54,8% vs. 25,5%; p = 0,034), while Kumar et al. (2020)–who used Curcuma longa (10 g/day for 8 weeks)–and Kedia et al. (2017)–who tested curcumin at a dose of 240 mg/day for 4 weeks–both in patients with UC– did not find any effect caused by oral supplementation.

Unlike the animal models included in this study, no nitroxidative stress/cytokines/nuclear factors were analyzed in the RCTs included in the present systematic review. In fact, half of them studied the effect of curcumin/Curcuma longa on serological or fecal parameters, and only two (25%) found positive results: decreased fecal calprotectin, C-reactive protein (CRP), erythrocyte sedimentation, and monocytes.

The risk of bias at the primary outcome level was assessed using the Cochrane collaboration tool. It is included in the Supplementary information. In general, all studies assessed risks were classified as “low” or “unclear” bias (Suppl. Table 2).

According to Fig. 2, oral supplementation with Curcuma longa extract or curcumin (modified or not) led to higher rates of clinical remission than placebo in both intention-to-treat (ITT) analysis (n = 281, RR: 3.15; CI 95% [1.22–8.10] p = 0.0017; i² = 72.2%, p = 0.006, Fig. 2A), and per-protocol (PP) analysis (n = 239, RR: 3.35; CI 95% [1.39–8.06] p = 0.007; i² = 71.7%, p = 0.006, Fig. 2B). High heterogeneity was observed in the analysis performed. When we removed the studies by Banerjee et al. (2020) and Lang et al. (2015), the heterogeneity of clinical remission in the ITT category was reversed (n = 162, RR: 1.70 ic95% [1.16–2.49] p = 0.006; i² = 0.0%, p = 0.38.

For clinical response, PP analysis (n = 259, RR: 1.60 CI 95% [1.09–2.35] p = 0.0017; i² = 59.7%, p = 0.042, Fig. 3A), but not ITT analysis (n = 304, RR: 1.51 CI95% [0.94–2.41] p = 0.086; i² = 67.7%, p = 0.015, Fig. 3B), indicates a protective effect of the Curcuma longa extract and curcumin.

For endoscopic remission, both ITT (n = 161, RR: 2.91 CI 95% [0.58–14.58] p = 0.195; i² = 72.7%, p = 0.026, Fig. 4A), and PP analysis (n = 129, RR: 3.34 CI 95% [0.76–14.68] p = 0.11; i² = 69.7%, p = 0.033, Fig. 4B) did not demonstrate any superior efficacy of curcumin/Curcuma longa extract compared to placebo.

A summary of the findings according to the GRADE assessment is shown in Table 3. In general, the quality of evidence ranged from low to very low, especially due to inconsistency and/or imprecision.

According to the present systematic review, animal studies consistently explore the effects of turmeric, attenuating nitroxidative and inflammatory responses on the characteristics of IBD. Studies also show that the antioxidant action of supplementation occurs at the molecular level by inhibiting the formation of ERONs, increasing the endogenous antioxidant response, and reducing cell damage (Esatbeyoglu et al., 2012; Maiti and Dunbar, 2018).

Chronic intestinal inflammation is notably mediated by proinflammatory immune responses stimulated by antigens of different nature (Jian et al., 2005). In this context, immune cells such as macrophages and lymphocytes act as important agents of the expression of proinflammatory cytokines (Duque and Descoteaux, 2014). Cytokines lead to the main proinflammatory signals, characteristic of colitis. The IL-1, IL-1β, and IL-18 families can be synthesized by intestinal epithelial cells or phagocytes (Duque and Descoteaux, 2014). In response to pathogens, IL-1β can stimulate T-cell differentiation in Th17 cells and the production of the cytokine IFN-γ (Friedrich et al., 2019). In experimental models, the absence of IL-1β and IL-18, either due to genetic deficiency or signaling inhibition, is associated with colitis improvement (Dinarello et al., 2013).

TNF-α, produced by phagocytes, participates in intestinal inflammation, exerting multiple intestinal cell effects (Sands and Kaplan, 2007). This cytokine’s high levels may alter the intestinal barrier’s functionality and integrity since stimulation of apoptosis occurs in epithelial cells. IL-6 is produced by phagocytes, epithelial cells, and mesenchymal cells. In the latter two, it influences the healing process, as it participates in the recruitment of polymorphonuclear leukocytes and macrophages (Mudter and Neurath, 2007). It also acts to prevent apoptosis of type T cells. On the other hand, IL-10, one of the main cytokines, is produced by innate and adaptive immune cells such as dendritic cells, macrophages, mast cells, natural killer cells, and Treg cells, among other cell types (Moore et al., 2001).

In IBD, the reduction in IL-10 production has already been associated with more severe clinical cases (Engelhardt and Grimbacher, 2014). IL-10 may trigger different signaling pathways related to anti-inflammatory activity and has already been considered a potential therapeutic target in IBD (Katsanos and Papadakis, 2017). Thus, by associating with higher levels of anti-inflammatory cytokines and decreased pro-inflammatory mediators, curcumin seems to exert a protective effect against inflammation.

Another line of evidence identified in the results of this systematic review indicates that curcumin can inhibit or reduce NF-κB activation. NF-κB is a factor that plays a central role in regulating the transcription of proinflammatory cytokine genes. Physiologically, the NF-κB transcription factor family in mammals consists of five proteins, p65 (RelA), RelB, c-Rel, p105/p50 (NF-κB1), and p100/52 (NF-κB2) that associate with each other to form distinct transcriptionally active homo and heterodimeric complexes (Oeckinghaus and Ghosh, 2009). By inhibiting the activation of transcription factors, curcumin interferes with the start-up of the process and the subsequent cascade of characteristic reactions of the inflammatory process by negatively regulating the suppression of multiple proinflammatory genes (Katsanos and Papadakis, 2017).

Some stimuli that culminate in NF-κB activation pass through cellular receptors that recognize pathogen-associated molecular patterns (PAMPs), especially TLR-4 (Ni et al., 2015). The stimulation of the TLR-4 isoform and the polymorphism of these receptors were identified as important mechanisms in the inflammatory stimulus in IBD (De Jager et al., 2007; Fukata and Abreu, 2008). The TLR-4-MyD88-NF-κB signaling pathway is activated when TLR-4 activates myeloid differentiation factor 88 (MyD88), which stimulates other molecules that act together for NF-κB activation with consequent expression of proinflammatory genes (Lubbad et al., 2009; Luo et al., 2020). Through these results, it seems that curcumin presents a mechanism of action: inhibition of the TLR-4-MyD88-NF-κB signaling pathway.

Intestinal tissue in IBD is histologically characterized by infiltration and accumulation of immune system cells in the mucous region, such as neutrophils and monocytes (Gui et al., 2018). In a complex tissue environment and loaded with proinflammatory molecules, these immune cells are probably recruited by chemotaxis mechanisms (Harbord et al., 2006). Neutrophils and monocytes release the MPO enzyme, whose activity may be able to indirectly reflect colonic inflammation (Chami et al., 2018). MPO may also be associated with the oxidative mechanisms of the inflamed colon, since this enzyme generates reactive species of oxygen (ROS), including hypochlorous acid (HOCl), through halogenation or peroxidase cycles (Myzak and Carr, 2002). The great number of experimental studies that observed a decrease in MPO activity, independent of dose, supplementation time, or moment of colitis, seem to indicate that curcuminoids can attenuate the infiltration of immune cells and reduce changes in the mucous region generated by intestinal inflammation in animals (Zeng et al., 2013).

Redox imbalance is related to inflammation and is characterized by an imbalance between oxidants and antioxidants, with a predominance of oxidation reactions (Vasconcelos et al., 2007). The gastrointestinal tract, which commonly interacts with food metabolites, microorganisms, and its own immune system, is a region that favored pro-oxidant molecules (Moura et al., 2015). Research conducted with animals and humans demonstrated the presence of redox imbalance in the disease by increasing the levels of markers of oxidative damage and reducing the level of antioxidant systems present in the blood, saliva, or colonic tissue. Lipid peroxidation, protein denaturation, and deoxyribonucleic acid (DNA) mutation are examples of damage caused by nitroxidative stress and interfere with the integrity of the intestinal mucosal barrier (Wang et al., 2016). Additionally, the evolution of IBD symptoms such as ulcer, toxic megacolon, and colorectal cancer has been related to the action of RONs (Moura et al., 2015; Moura et al., 2020).

The process of inflammation characteristic of IBD may incur gene transcription of some enzymes implicated in the endogenous generation process of reactive species. Peroxidases such as xanthine oxidase (XO), lipoxygenases (LPO), MPO, iNOS, COX, and the NADPH oxidase complex (NOX) (Balmus et al., 2016a; Balmus et al., 2016b) COX-2 are directly involved in ulceration formation and may be associated with inflammatory processes in the colon region and precancerous changes in the gastrointestinal tract (Binion et al., 2008; Camacho-Barquero et al., 2007; Mccarty, 2012). The results of the experimental studies in this systematic review demonstrate that oral curcumin supplementation can inhibit these enzymes directly or by blocking NF-κB.

In this systematic review, damage to the lipid membrane was the most evaluated cell damage and showed the best results after curcumin supplementation. Lipid peroxidation results in the formation of products such as TBARS and MDA (Niki, 2013). In this context, it has been shown that MDA is increased in IBD in animals and humans in different biological environments, such as saliva (Jahanshahi et al., 2004; Rezaie et al., 2006; Szeląg et al., 2012) and blood (Boehm et al., 2012; Alzoghaibi et al., 2007). Another finding of the study points to the existence of a positive correlation of MDA with Crohn’s Disease Activity Index (CDAI) and C-reactive protein (CRP) and a negative correlation with antioxidant defense (Szczeklik et al., 2018), composed of enzymatic and nonenzymatic biomarkers. The positive effects of curcumin/Curcuma longa on this cell damage marker suggest that this polyphenol/extract may act, protecting membrane integrity.

Although few studies, pointed out in this review, have evaluated the antioxidant defense system, represented by enzymes SOD, CAT, and GPx, and nonenzymatic system GSH, several studies on humans confirm that these biomarkers are altered in IBD (Guan and Lan, 2018; Szczeklik et al., 2018). SOD–represented by its three main isoforms SOD1 (Cu/ZnSOD) present in the cytosol, SOD2 (MnSOD) located in mitochondria, and SOD3 (Cu/ZnSOD) present in extracellular medium–catalyzes the conversion from anion radical superoxide (O2•-) to oxygen (O2); in sequence CAT–present in peroxisomes–and GPx–found in cytosol–convert O₂ in hydrogen peroxide (H₂O₂). In turn, the GSH system (reduced glutathione–GSH–, oxidized (GSSG) and glutathione reductase–GR), a nonenzymatic system defense, also responds against oxidative damage, and their levels are altered in CD or UC (Pinto et al., 2013; Sido et al., 1998). The improvement or normalization of the levels of antioxidant defense in animals that received oral curcumin/Curcuma longa indicates that they may directly stimulate enzymatic end nonenzymatic synthesis/activity.

In the inflamed colon, the presence of macrophages and neutrophils that are stimulated by proinflammatory cytokines, as well as the recognition of bacterial lipopolysaccharides (LPS) by TRL-4, and the action of IFN-γ are associated with super expression of iNOS and consequent production of •NO (Camacho-Barquero et al., 2007; Motawi et al., 2012), which appears to be associated with a more severe inflammatory response and tissue injury, in experimental colitis (Motawi et al., 2012). On the other hand, •NO may interact with O2•- and generate peroxynitrate (-OONO) that can react with DNA (Cooke et al., 2003). In this context, the decrease in •NO levels observed in some animal studies (Altinel et al., 2020; Camacho-Barquero et al., 2007; Kao et al., 2016; Motawi et al., 2012; Venkataranganna et al., 2007; Ukil et al., 2003) may reflect the scavenging action of curcumin/Curcuma longa or their role in inhibiting iNOS activity. The alteration of the microbiome stands out as a factor involved in the pathogenesis of IBD (Guan, 2019). It is hypothesized that curcumin may interfere in the microbiome, either through the action of the compound and/or metabolites or the action of products resulting from the microbial metabolism of the substance itself. Regulatory effects can be attributed to modulation in the intestinal biome’s amount, diversity, and composition (Scazzocchio et al., 2020). This relationship between curcumin and microbiota can be confirmed in the study by Ohno et al. (2017), which when evaluating mice with DSS-induced colitis, observed that a diet consumption containing 2.0% (w/w) of curcumin for 18 days was able to stimulate the proliferation of bacteria Clostridium cluster IV and XIVa, which was accompanied by increased levels of fecal butyrate. Unfortunately, human studies have not yet tested this hypothesis, leaving this gap in the beneficial actions of curcumin in the health of patients with IBD.

Together, these beneficial actions reveal the protective role of curcuminoids/Curcuma longa against inflammation, redox imbalance, and dysbiosis and suggest the necessity of evaluating these biomarkers in human studies.

The IBD is not a disease with an established cure; therefore, the treatment adopted in current protocols is nonspecific and aims to minimize symptoms, improve quality of life, achieve remission, and decrease complications of the disease (Abraham et al., 2017). In addition, IBD pathogenic mechanisms remain unclear, and global epidemiologic data suggest an important crosslink with socioeconomic development (Molodecky et al., 2012). Furthermore, IBD has a low mortality rate and is usually associated with complications such as colorectal cancer (CRC), infections and surgical complications. Its clinical manifestations, such as diarrhea, weight loss, and low digestive bleeding, have a great impact on the quality of life of its patients (Moura et al., 2015). Together, these data confirm IBD as an important global public health problem.

Conventional IBD therapy involves, in general, the use of sulfasalazine, corticosteroids, immunosuppressive agents, such as azathioprine, mercaptopurine or methotrexate (Lamb et al., 2019). In addition, is used the biological therapy, represented especially by the anti-TNF, anti-integrin, and anti-IL 12/23 drugs (Hindryckx et al., 2018). On the other hand, the adverse effects associated with the prolonged use of these drugs and the high rate of relapses significantly limit their use (Xu et al., 2004). In this context, the estimate indicates that 10%–30% of patients with UC and 38%–70% of patients with CD, with complications, will undergo some surgical intervention (Palacio et al., 2021) and the long-term collateral effects of drug therapy, together with the high cost of surgical management, leads the scientific community to investigate alternative treatments for IBD.

In the present meta-analysis, even with the variation in the rates of assessment of disease severity among RCTs, curcumin/Curcuma longa oral supplementation combined with conventional drug therapy was able to induce clinical remission and response in adult IBD patients. These outcomes were assessed by the score of different tools, such as Crohn’s Disease Activity Index (CDAI) (Sugimoto et al., 2020), Simple Clinical Colitis Activity Index (SCCAI) (Lang et al., 2015; Masoodi et al., 2018; Sadeghi et al., 2020) and Ulcerative Colitis Disease Activity Index (Mayo/UCDAI) (Banerjee et al., 2020; Kedia et al., 2017; Kumar et al., 2020), which measure disease activity through the severity of the presented symptoms (Table 4).

The SCCAI includes assessing six questions, including the number of nighttime evacuations and fecal urgency. This index was proposed by Walmsley et al. (1998) and has been used by several authors (Lang et al., 2015; Sadeghi et al., 2020; Schroeder et al., 1987; Walmsley, 2014). The Mayo Clinic Score and UCDAI are similar tools composed of four questions, and in addition to including clinical symptoms, they also evaluate endoscopic changes. Such tools are the most adopted in the methodology for assessing disease activity in RCTs (Bewtra et al., 2014). Although the CDAI is one of the tools commonly used in CD and evaluates the severity of several signs and symptoms, such as weight loss and anemia, it is considered an insufficient form since the questions included in the tool can be influenced by subjectivity, in addition to presenting little agreement with histological altars identified by the Global Histologic Activity Score (GHAS) (Tajra et al., 2019).

Despite the different methods used, clinical activity is an important parameter to assess IBD. The intensity with which symptoms manifest in the patient and endoscopic and histological changes can predict whether the disease is active. However, few tools have been validated to categorize the severity of the disease (Peyrin-Biroulet et al., 2016). Patients with active disease have a worse quality of life, increasing between 70% and 80% chances of having a recurrent episode in one year (Sairenji et al., 2017). In addition, IBD clinical activity can be an independent risk factor for extraintestinal manifestations, such as acute arterial disease, ischemic heart, and cerebrovascular arterial disease (Le Gall et al., 2018) and anemia (Parra et al., 2020).

Endoscopic evaluation is known to be recommended for diagnosis, monitoring and therapeutic evaluation of patients with IBD. Although endoscopic remission is not statistically evidenced in this meta-analysis, it is important to highlight those other methods are emerging as an alternative for the accompaniment of these patients, such as fecal markers (such as calprotectin), because they have lower cost and impact for these individuals, facilitating their performance, including in places of difficult access to imaging methods (common in poor or developing countries). In this context, we strongly suggest that the new human trials include these biomarkers aiming at an expanded view of the effects of curcumin (and other antioxidants) on the intestinal health of patients with IBD.

In this context, identifying therapies that reduce clinical activity, and consequently, lead and keep the patient in the remission phase is one of the main objectives of clinical research in IBD.

This study presents to the scientific community two-point of view about oral supplementation of curcumin/Curcuma longa: the impact of this polyphenol/extract on animal models of UC, and a systematic review and meta-analysis of RCTs. In both, beneficial effects were observed. However, RCTs still do not carry out analysis on the nitroxidative and anti-inflammatory impacts of curcumin, even with several confirmations observed both in vitro and in animal models. Additionally, the curcumin administration led to different effects in animal experiments and in clinical trials. Probably, these differences occurred because colitis in animal models is induced by external factor. Since IBD is a multifactorial disease, its induction cannot mimic the various risk factors involved, and for that, it is not possible to have the same results in animals and humans. However, the results in animals serve as a basis for conducting human trials.

Another unanswered question is whether curcumin changes its beneficial effects in symptomatic or asymptomatic colitis. Although animal studies show different models of colitis, in randomized studies, curcumin/curcuminoids were mostly tested in the acute phase of the disease. Likewise, studies are still insufficient to determine whether curcuma/curcuminoids is more effective among patients with CD or UC.

Given this information, new clinical studies’ objectives should confirm whether curcumin shows similar results in humans, which would justify its positive result on the clinical improvement found in this meta-analysis.

Future studies should evaluate the effect of Curcuma longa/curcuminoids on nitroxidative stress and its crosslink with inflammation and disease activity and remission.

Some questions, as listed below, should direct research efforts.

Which is the better moment to initiate supplementation: the symptomatic or the asymptomatic phase?

What are the better doses and period of treatment?

Does modified curcumin (nanoparticles, microparticles, combined, and other formulations) show beneficial effects vs. pure curcumin in humans?

In addition to these issues, it is necessary to be established reproducible models that ensure the evaluation of this bioavailability in different IBD scenarios.

However, despite these still unanswered questions, the analysis of the present data, allow to suggest that the oral prescription of Curcuma longa and/or curcumin, when associated with drug therapy, is safe and effective in the treatment of patients with IBD.