Cardioprotective effects of curcumin against myocardial I/R injury: A systematic review and meta-analysis of preclinical and clinical studies

Objective: Myocardial ischemia-reperfusion (I/R) injury is a complex clinical problem that often leads to further myocardial injury. Curcumin is the main component of turmeric, which has been proved to have many cardioprotective effects. However, the cardioprotective potential of curcumin remains unclear. The present systematic review and meta-analysis aimed to evaluate the clinical and preclinical (animal model) evidence regarding the effect of curcumin on myocardial I/R injury. Methods: Eight databases and three register systems were searched from inception to 1 November 2022. Data extraction, study quality assessment, data analyses were carried out strictly. Then a fixed or random-effects model was applied to analyze the outcomes. SYRCLE’s-RoB tool and RoB-2 tool was used to assess the methodological quality of the included studies. RevMan 5.4 software and stata 15.1 software were used for statistical analysis. Results: 24 animal studies, with a total of 503 animals, and four human studies, with a total of 435 patients, were included in this study. The meta-analysis of animal studies demonstrated that compared with the control group, curcumin significantly reduced myocardial infarction size (p < 0.00001), and improved the cardiac function indexes (LVEF, LVFS, LVEDd, and LVESd) (p < 0.01). In addition, the indexes of myocardial injury markers, myocardial oxidation, myocardial apoptosis, inflammation, and other mechanism indicators also showed the beneficial effect of curcumin (p < 0.05). In terms of clinical studies, curcumin reduced the incidence of cardiac dysfunction, myocardial infarction in the hospital and MACE in the short term, which might be related to its anti-inflammatory and anti-oxidative property. Dose-response meta-analysis predicted, 200 mg/kg/d bodyweight was the optimal dose of curcumin in the range of 10–200 mg/kg/d, which was safe and non-toxic according to the existing publications. Conclusion: Our study is the first meta-analysis that includes both preclinical and clinical researches. We suggested that curcumin might play a cardioprotective role in acute myocardial infarction in animal studies, mainly through anti-oxidative, anti-inflammatory, anti-apoptosis, and anti-fibrosis effects. In addition, from the clinical studies, we found that curcumin might need a longer course of treatment and a larger dose to protect the myocardium, and its efficacy is mainly reflected on reducing the incidence of myocardial infarction and MACE. Our finding provides some meaningful advice for the further research.


Introduction
Cardiovascular disease (CVD), especially acute myocardial infarction (AMI), has become one of the main risk factors threatening human health (Mozaffarian et al., 2015). Thrombolytic, percutaneous coronary intervention (PCI), and coronary artery bypass grafting (CABG) are currently effective strategies to limit infarct size and reduce mortality in clinical practice (Ribas et al., 2017). However, there is growing evidence indicating that myocardial ischemia/reperfusion (I/R) injury is an inevitable pathological change in the process of reperfusion after revascularization (Heusch and Gersh, 2017). In addition, it will lead to myocardial dysfunction, structural damage, and myocardial electrical activity disorder, which further aggravates myocardial necrosis, severe complications such as arrhythmia, decreased ventricular function, and even sudden death (Grech et al., 1995). The mechanism is not yet fully understood, but research showed that it might be related to inflammation, oxidative stress, mitochondrial membrane permeability transition pore, cell apoptosis, et al. (Hausenloy and Yellon, 2013). Therefore, seeking novel myocardial protection strategies to ameliorate myocardial I/R injury and preserve normal heart function after revascularization is still of high priority.
Turmeric is a commonly used herb as a traditional medicine in Asian countries (China, Japan, Korea, et al.). According to the records of ancient Chinese medical books, turmeric is capable of relieving pain through promoting qi and blood circulation. Therefore, it is frequently used in the treatment of chest tightness, chest pain, shoulder and back pain, dysmenorrhea, et al. Curcumin (chemical structure shown in Figure 1) is the main component of turmeric and has a variety of biological activities, especially in the cardiovascular system (Ghosh et al., 2010). Curcumin has been proved to have anti-oxidative, antiinflammatory, and anti-apoptosis properties, and it has been observed in experiments that curcumin inhibits myocardial fibrosis (Gorabi et al., 2020) and ventricular remodeling after acute infarction (Yao et al., 2004) and prevents the progression of heart failure after myocardial infarction (Saeidinia et al., 2018). Therefore, curcumin may be a potential cardioprotective candidate to ameliorate myocardial I/R injury.
However, the efficacy and mechanisms of curcumin on myocardial I/R injury have not been systematically reviewed in animal models. Moreover, some clinical trials (Offermanns et al., 1994;Srivastava et al., 1995;Peng and Qian, 2014;Hirsch et al., 2017) showed the contradictory results. In this study, we aim to review and analyze the animal and clinical research to find out whether curcumin has the potential for clinical application and suggested how such primary studies need to be improved.

Search strategies and study selection
Eight databases, including Pubmed, EMBASE, The Cochrane Library, Web of Science, China National Knowledge Infrastructure (CNKI), Wanfang database, China Science and Technology Journal Database (VIP), and China Biology Medicine disc (CBM), were searched systematically from inception to 1 November 2022. Moreover, references of eligible studies were also manually searched to identify additional eligible studies. The search language was restricted to English and Chinese, and the search strategy was as follows. The detailed search strategies in eight databases and retrieved results were provided in Supplementary Material.

Eligibility criteria
Inclusion criteria were as follows: 1) The I/R experimental model were established by ligating the left coronary artery (LCA) such as left anterior descending (LAD) or other proper methods, or ischemic heart disease patients receiving PCI or CABG in the hospital; 2) The treatment group (animal studies and human studies) received any dose of curcumin as monotherapy, and the control group received the same amount of non-functional fluid (normal saline) or no treatment at all; 3) The primary outcome of animal studies were myocardial infarction (MI) size and echocardiogram indicators. The secondary outcomes of animal studies were biomarkers of myocardial injury and mechanisms of curcumin in the treatment of myocardial I/R injury; 4) The human studies were RCT study. Exclusion criteria were set as follows: 1) not a myocardial I/R model or patients who beyond the relevant clinical diagnostic criteria; 2) combined with other drugs; 3) no control group; 4) duplicate publication; 5) not an animal model or not RCT study (e.g., case reports, reviews, non-RCT clinical trials and cell experiments).

Data extraction
Two independent reviewers (Fenglan Pu, Ying Bai) extracted the following details from the included studies: 1) first author name and year of publication; 2) Specific information about the animals and patients in each study, including species, number, sex, and body weight; 3) Myocardial I/R model and the anesthetic method used to prepare the model; 4) Information about curcumin treatment, including dose, method of administration, course of treatment; as well as corresponding information in the control group; 5) The mean and standard deviation (SD) of the results. Suppose there were many different time point results, only the last time point results would be recorded. Since some of the data is only presented in graphical format, we tried to contact the authors for detailed information. If we did not receive any response, the digital ruler software would be used to measure the value of the graph.

Risk of bias in individual studies
The quality of the included animal studies was assessed using Systematic Review Center for Laboratory Animal Experimentation (SYRCLE) risk of bias (RoB) tool 10-item scale (Hooijmans et al., 2014) as follows (A) sequence generation; (B) baseline characteristics; (C) allocation concealment; (D) random housing; (E) blinding investigators; (F) random outcome assessment; (G) blinding outcome assessor; (H) incomplete outcome data; (I) selective outcome reporting; (J) other sources of bias. Give one point to each item.
The quality of the included human studies was assessed using RoB-2 tool (Sterne et al., 2019) as follows: Randomization process; Assignment to intervention; Adhering to intervention; Missing outcome data; Measurement of outcome; Selection of the reported result; RoB-2 overall score. Two reviewers (Jialin Jin and Fenglan Pu) assessed the study quality independently. The divergences were resolved through negotiation or in consultation with the corresponding author (Yan Li and Xian Wang).

Data synthesis
The problem of curcumin divided into different dose subgroups in the original study was handled following the strategy recommended by Cochrane Handbook for Systematic Reviews of Interventions (CHSRI) (Cumpston et al., 2019). We merged the different dose subgroups in each original study into one treatment group, and the following was the merge formula for continuous variable (Zhou et al., 2020):

Statistical analysis
RevMan software (version 5.4) was used for statistical analysis. Heterogeneity was determined by the Q-statistical test (p < 0.05 was considered statistically significant) and I 2 -statistical test. According to the Cochrane handbook on heterogeneity analysis, I 2 value ranges from 0% to 40% means heterogeneity might not be important; 30%-60% may represent moderate heterogeneity; 50%-90% may represent substantial heterogeneity; 75%-100% means considerable heterogeneity. A fixed-effects model was adopted if I 2 < 50%; otherwise, a random-effects model would be applied. All outcomes were continuous variables, so we used weighted mean difference (WMD) with 95% confidence intervals (CIs) to present, for outcomes reported in various measurement methods or different scales of measurement, we calculated with a standard mean difference (SMD). p < 0.05 was considered statistically significant.

Subgroup analysis and meta-regression
We preset four subgroups: 1) method of model establishment, 2) administration method, 3) genus of animals, 4) intervention moment. We conducted a subgroup analysis on the primary outcomes to evaluate the impact of variables and to explore the source of heterogeneity. A meta-regression analysis was further performed using Stata 15.1 software once the heterogeneity in the study was significant (I 2 > 50%) and the preset subgroups failed to find the source of heterogeneity.

Sensitivity analysis
Sensitivity analysis should be performed on the outcomes whose heterogeneities were considerable. The purpose of the sensitivity analysis was not only to indirectly find the source of heterogeneity, but also to assess the stability and reliability of the merged results.

Publication bias
Publication bias was detected by a funnel plot combined with Egger's regression test on primary outcomes. The funnel plot was produced using RevMan 5.4 software, and Egger's regression test was completed using Stata 15.1 software.

Dose-response meta-analysis
The relationship between the standard dosage of curcumin and the ratio of means (RR = means experimental /means control ) was modeled using STATA 15.1 software. According to the characteristics of point distribution, linear, exponential, logarithmic, quadratic, and cubic regression equations can be selected to try to fit. The generalized least squares method (GLS) was used to calculate each of the parameters while R 2 represents the degree of fitting (R 2 closer to one indicates a better fit of the prediction model).

Human studies
All human studies (Wongcharoen et al., 2011;Wongcharoen et al., 2012;Aslanabadi et al., 2019;Phrommintikul et al., 2019) are English literature studies, mainly distributed in Iran and Thailand. All 435 patients included in the study were ischemic heart disease patients who needed PCI or CABG treatment, and were randomly divided into curcumin treatment group and conventional treatment group. The clinical characteristics are summarized in Table 2.

Human studies
RoB-2 tool was used to evaluate the methodological quality (deviation risk) of the four included human studies. Two of the studies were judged as low risk, while the other two were judged as some concerns. The details are shown in Table 4.

Data analysis of animal studies 3.4.1 Myocardial infarction size
Myocardial infarction (MI) size was calculated by infarct area/ left ventricular area. A meta-analysis of 16 animal studies demonstrated that curcumin significantly decreased MI size compared with the control group (n = 246, WMD = −17.91%, 95% CI (−22.24%, −13.59%), p < 0.00001, I 2 = 97%) (Figure 3). Because of the obvious heterogeneity in the included animal studies, we performed a sensitivity analysis. After removing the animal studies one by one, we failed to determine the source of heterogeneity, but the sensitivity analysis indicated that the outcome was steady.

Dose-response meta-analysis of curcumin
According to previous studies, different doses of curcumin could affect the efficacy (Yang et al., 2013a;Liu et al., 2017b), which might be one of the main reasons for the high outcome heterogeneity. Therefore, we performed a dose-response meta-analysis of animal studies Wang et al., 2012;Yang et al., 2013a;Wang et al., 2014;Chen et al., 2016;Liu et al., 2017b;Jo et al., 2020) in which curcumin was administered orally. Firstly, RR was recalculated and defined as the ratio of means (means experimental /means control ). Then, scatter plot was drawn for curcumin dose and RR (Supplementary Figure S1). Through image analysis, the point distribution showed a non-linear relationship. We hypothesized that a non-linear relationship might be applicable to the relationship between dose and the protective effect of curcumin. Nonlinear regression was conducted according to the quadratic regression method of meta-analysis (Orsini et al., 2012). with the increase of curcumin dosage, the protective effect of curcumin on myocardial I/R was firstly enhanced, then weakened and then strengthened (Figure 4). In our model, 200 mg/kg BW per day was the optimal dose of curcumin (when dosage ranges from 10 to 200 mg/kg BW per day) and had the best predictive protective effect, and the prediction relationship can be expressed as LnRR = −1*10 −6 x 3 + 0.0003 x 2 − 0.0205 x − 0.2386, R 2 = 0.638.
Frontiers in Pharmacology frontiersin.org

FIGURE 3
The meta-analysis results of the effect of curcumin on MI size.
Because of the small number of animal studies (<3 studies), other mechanism indicators reflecting the cardioprotective function of curcumin only are described as follows. Unless otherwise specified, these reports indicated positive effects of the curcumin group on the mechanism of anti-myocardial I/R injury (p-value indicates comparison with the control group). These indicators included Bcl-2 (Duan et al., 2012;Deng et al., 2018) (p < 0.05), Caspase-3 (Chen et al., 2016;Liu et al., 2017b) (p < 0.05 or p < 0.01), MMP-9 (Yeh et al., 2005;Wang et al., 2012) (p < 0.05), MPO (Kim et al., 2008;Wang et al., 2014) (p < 0.05), IL-6 mRNA (Yeh et al., 2005;Wang et al., 2014) (p < 0.05). The included animal studies also reported the regulation of curcumin on other proteins. However, because the frequency of occurrence was <2, they would not be described in detail (All reported indicators can be found in Table 1).

Subgroup analysis and meta-regression
We performed the subgroup analysis on primary outcomes according to preset subgroups. The LVEF, LVFS, LVDD, LVSD in the primary outcome indicators were not included in the subgroup analysis because of not enough animal studies, so we only carried out the subgroup analysis on MI size. Analysis of the three preset subgroups (method of model establishment, administration method, and genus of experimental animals) indicated outcomes that were consistent with the overall rusults (p < 0.05) (Figures 7-9). Some outcomes of the subgroup analysis by intervention moment (before ischemia/between ischemia and reperfusion/during reperfusion) were inconsistent with the overall results ( Figure 10). The subgroup analyzed for before ischemia group and during reperfusion group were consistent with the overall results (p < 0.05), while the outcome of between ischemia and reperfusion group was inconsistent with the overall results (p > 0.05). Moreover, I2 did not decrease, suggesting that the preset subgroups did not explain the source of heterogeneity of MI size.

Publication bias
A funnel plot was adopted to assess the publication bias regarding myocardial infarction size ( Figure 11). Furthermore, we used Egger's regression test to check out the asymmetry of the graph, and the results indicated no publication bias in this study (p = 0.92 > 0.05).

.1 Methods and characteristics of curcumin treatment
Oral curcumin nanomicelle was used in four human studies. Two studies (Aslanabadi et al., 2019;Phrommintikul et al., 2019) used a shorter course of treatment (1-2 days), with a total amount of 0.48g-8 g of curcumin. The other two studies (Wongcharoen et al., 2011;Wongcharoen et al., 2012) used a longer course of treatment (8-10 days), with a total amount of 32g-40 g of curcumin. All the curcumin treatment groups in the study were combined with conventional treatment, and only one study described the conventional treatment protocol in detail.

Biomarkers of inflammation and oxidative stress injury
Serum CRP and MDA were used to evaluate the degree of inflammation and oxidative stress injury. The results of study (Phrommintikul et al., 2019) showed that there was no statistically significant difference in CRP reduction between the curcumin group and the control group (7.2 ± 18.8 vs 6.6 ± 17.5 mg/dL, p = 0.87), but the results of study (Wongcharoen et al., 2012) showed that compared with the control group, the curcumin group could reduce CRP (161.8 ± 54.1 vs 128.6 ± 60.5 mg/ dL, p = 0.031) and MDA (0.8 ± 1.4 vs 5.7 ± 1.5 mmol/mL, p = 0.001).

Cardiac function, incidence of in-hospital MI and MACE within 30 days after revascularization
For patients with ischemic heart disease, in-hospital MI and MACE after revascularization are the most important clinical outcome, and cardiac function is the most important factor affecting the above outcome. The study (Wongcharoen et al., 2012) showed that compared with the control group, the curcumin group could reduce the incidence of left ventricular dysfunction, LVEF<40% (3.3% vs 25.9% p = 0.021), and NT-proBNP (1822.1 ± 2,102.9 vs 2,542.2 ± 2,631.2 pg/mL, p = 0.015). The results of two studies (Wongcharoen et al., 2011;Wongcharoen et al., 2012) showed that the incidence of in-hospital MI in the curcumin group was significantly lower than that in the control group. The study (Wongcharoen et al., 2011) showed that compared with the control group, the incidence of MACE within 30 days after revascularization in the curcumin group was lower (13.5% vs. 34.6%, p = 0.02).

Contradiction between human studies and animal studies
This is the first systematic review and meta-analysis consisting of preclinical and clinical evidence to investigate the cardioprotective effects of curcumin on myocardial I/R injury. Twenty-four studies with 503 animals and four human studies with 435 patients were included, and the overall methodological quality of the included studies was moderate. The results of animal studies meta-analysis suggested that curcumin might diminish myocardial infarction size, improve heart function, suppress biomarkers concentration for myocardial infarction, as well as ameliorate cardiomyocyte inflammation, oxidation, apoptosis, and myocardial fibrosis in animal studies.
Meta-analysis of animal studies shown the clinically potential of curcumin in the treatment of myocardial I/R injury. Interestingly, some of results of human studies showed different opinions. The results of two studies (Aslanabadi et al., 2019;Phrommintikul et al., 2019) showed that there was no statistically significance between curcumin group and control group in serum myocardial injury biomarkers (hscTnT, cTnI, CK-MB) after PCI or CABG. However, we found that in N. Aslanabadi et al.'s study (Aslanabadi et al., 2019), the incidence of CK-MB rising above the normal value during PCI in curcumin group was 50% less than that in control group. In addition, the decrease of serum CK-MB in patients in curcumin group 24 h after PCI were more notable than that in the control group (2.6 ± 8.7 vs 0.29 ± 8.8 U/L, p = 0.75). Although p > 0.05, both indicators showed descending trend. Besides, it should be noted that the protocol of curcumin treatment in these two studies (Aslanabadi et al., 2019;Phrommintikul et al., 2019) was only 0.48 g P.O. before PCI and 4 g P.O. before and after PCI. This short course of treatment may affect the statistical significance of the results. Furthermore, the other two studies (Wongcharoen et al., 2011;Wongcharoen et al., 2012) adopted a longer course of treatment, a larger total amount of drugs, and a longer follow-up time. The results of the two latter studies showed that curcumin might has cardioprotection in term of anti-inflammation, cardiac function, incidence of in-hospital myocardial infarction and MACE within 30 days after CABG or PCI. The reason why human studies showed drastic difference

FIGURE 8
The subgroup analysis (administration method) of curcumin on MI size.

Frontiers in Pharmacology
frontiersin.org

FIGURE 9
The subgroup analysis (genus of experimental animals) of curcumin on MI size.

FIGURE 10
The subgroup analysis (intervention moment) of curcumin on MI size.
Frontiers in Pharmacology frontiersin.org 20 might be relative to the dosage of curcumin and duration of treatment.

Safety and dose-response of curcumin
According to the existing research results, we found that poor resorption of curcumin was one of the reasons why curcumin was limited clinically. For example, volunteers took a large dose of curcumin 12 g a day in human pharmacokinetic trials. However, the blood concentration was too low to be detected, indicating that its bioavailability was low (Peng and Qian, 2014). Therefore, it is meaningful to find the relationship between the dosage and effectiveness within the safe dosage range of curcumin. First, we conducted a dose-response meta-analysis of animal studies in which curcumin was administered orally. In our model (animal studies), 200 mg/kg BW per day was the optimal dose of curcumin (when dosage ranges from 10 to 200 mg/kg BW per day) and had the best predictive protective effect. This analysis result is consistent with another curcumin dose response analysis result (Lin et al., 2020).
According to other curcumin toxicity studies, 200 mg/kg BW per day is acceptable. Many animal studies reported that significant toxicity, pathological effects, genotoxicity and death were not observed when the maximum intake dose of curcumin reached 5,000 mg/kg BW per day within 14 consecutive days (Wahlström and Blennow, 1978;Perkins et al., 2002;Aggarwal et al., 2016;Damarla et al., 2018). In human studies, when treated with purified nano curcumin, the most remarkable side effects observed include gastrointestinal flatulence (Hanai et al., 2006) and changes in liver function (Na et al., 2013;Zingg et al., 2017). Overall, based on the current research results, curcumin will be safe when it reaches the concentration we predicted.

Potential mechanisms of curcumin in myocardial I/R injury
Myocardial I/R injury is associated with various pathophysiological processes, such as calcium overload, generation of oxygen free radicals, endothelial dysfunction, inflammatory response, mitochondrial dysfunction, myocardial cell apoptosis, and autophagy (Kloner et al., 1989;Hausenloy and Yellon, 2003;Yellon and Hausenloy, 2007;Oerlemans et al., 2013;Merz et al., 2019). Considerable evidence indicates that curcumin exerts cardioprotective effects in myocardial I/R injury through multiple molecular mechanisms. Multiple molecular mechanisms might be involved in the underlying protective mechanisms of curcumin against myocardial injury, and a better understanding of these protective mechanisms will provide better insight into curcumin ( Figure 12).

Anti-oxidant effect
Oxidative stress plays an essential role in myocardial I/R injury. Reducing oxidative stress is one of the strategies for dealing with I/R injury. The antioxidant activity of curcumin is mainly reflected by scavenging various reactive oxygen species (Broskova et al., 2013). Curcumin could regulate mitochondrial dysfunction after I/R through activating the silent Information regulator 1 (SIRT1) signal, up-regulating Bcl-2, and down-regulating Bax (Mokhtari-Zaer et al., 2018). Some studies also showed a direct antioxidant effect of curcumin, including increasing mitochondrial SOD activity and decreasing the generation of mitochondrial hydrogen peroxide and MDA (Yang et al., 2013a).

Anti-inflammatory effect
Inflammation is usually an active defense mechanism, but excessive inflammation will exacerbate myocardial reperfusion

FIGURE 11
Funnel plot detailing publication bias in animal studies reporting the effect of curcumin on MI size.
Frontiers in Pharmacology frontiersin.org injury. Studies have shown that curcumin exerts antiinflammatory effects by regulating NF-κB signaling pathway, an essential proinflammatory signaling pathway involved in cardiac injury (Fiorillo et al., 2008). It was shown that curcumin pretreatment over a week reduced TLR2 and MCP1 expression in cardiomyocytes, macrophage infiltration (CD68) and cardiac fibrosis . In addition, curcumin could inhibit the expression of early growth response 1 (EGR-1) in ischemic myocardial cells through lowering the levels of TNF-α and IL-6 (Wang et al., 2012). This might suggest that curcumin has a potent antiinflammatory effect and could protect the cardiomyocytes through modulating the expression or activity of proinflammatory cytokines.

Anti-apoptosis effect
Apoptosis is a physiological phenomenon in most circumstances.
However, excessive apoptosis under pathological conditions lead to unnecessary cell death. Due to myocardial I/R, STAT3 upregulates the expression of Bcl-2, Bclxl, and other genes, which have been shown to reduce cell death and lessen adverse cardiac remodeling after myocardial infarction (Yang et al., 2013b). We found that activation of Janus kinase signal sensor and transcriptional activator (JAK2-STAT3) signal pathway, upregulation of Bcl-2 and downregulation of caspase-3 could be the mechanisms of curcumin in relieving myocardial I/R injury (Duan et al., 2012).

Regulating autophagy effect
Autophagy is functional at low baseline levels in myocardium, removing unnecessary proteins and damaged organelles as a self-protective mechanism (Klionsky et al., 2021). Dysregulated or excessive autophagy could induce cell death (Zhao et al., 2021). A few studies indicated that curcumin might protect mouse cardiomyocytes from oxidative stress by stimulating autophagy (Rainey et al., 2020;Gu et al., 2021;Lin et al., 2021). Other studies claimed that curcumin protects myocardial I/R via downregulating elevated autophagy (Huang et al., 2015;Ma et al., 2021). This suggests that curcumin might be a potential bidirectional regulatory agent in myocardial I/R, either inhibiting excessive autophagy or promoting autophagy. In general, curcumin exerts cardioprotective effect through regulating autophagy.

Implications
Our study found that the animal myocardial IR model is not completely consistent with the characteristics of the clinical patients, and this is because that myocardial ischemia-reperfusion injury (MIR) is often accompanied by a variety of conditions, such as aging, hypertension, diabetes, etc. Therefore, more complex animal models such as aging animal models, should be used to study the pharmacological effects and mechanisms of curcumin in future researches.
Frontiers in Pharmacology frontiersin.org The rationality and standardization of in vivo researches directly affect the further transformation from pre-clinical studies to clinical applications. According to SYRCLE 2.0 tool, the methodology quality of included animal studies is moderate. Therefore, when designing experimental protocols, researchers should strictly follow the animal experiment guidelines, such as random allocation, blinding investigators, random outcome assessment, etc., and record this information in the paper.
Many animal studies have shown that curcumin has acceptable security, and no significant toxicity was observed even when rats or mice were given a high dose (5 g/kg BW per day). According to the dose-response model of animal studies, the cardioprotective effect of curcumin may be stronger with the further increase of the dosage. Further, researchers developed a Polylactic-co-glycolic acid (PLGA) and Plga-polyethylene (PEG) (Plga-peg) nanoparticle containing curcumin, and it was observed that the nanoparticles increased the mean half-life of curcumin in about 6 h, enhanced the maximum blood concentration of curcumin by 7.4 times, and improved the bioavailability of curcumin by 55.4 times compared with curcumin aqueous suspension (Khalil et al., 2013). Therefore, we suggest that higher doses of curcumin (more than 200 mg/kg BW per day) and nanoparticle containing curcumin could be explored in future studies to determine the best effective dose.
Conducting more clinical trials should be encouraged because animal experiments have shown excellent effectiveness. In the existing clinical trials, the results showed that curcumin significantly reduced the incidence of MACE in patients after PCI or CABG in the short term. In addition, curcumin is widely consumed as a food material in Southeast Asia, and a large number of studies supported its high biological safety.
In clinical research, researchers should not only focus on the incidence of MACE events caused by MIR in the short term after PCI or CABG, but also focus on the incidence of MACE and cardiac function within 1 year after discharge by increasing follow-up time.
Patients with myocardial injury and reperfusion often need to take antiplatelet drugs. Some studies claimed that curcumin plays an antiplatelet role by inhibiting platelet aggregation (Srivastava et al., 1995;Hirsch et al., 2017), inhibiting cyclooxygenase activity, and blocking calcium channels (Offermanns et al., 1994). However, there is no study explaining the relationship between curcumin and other antiplatelet drugs in vivo experiments.

Limitations
First of all, we only retrieved English and Chinese studies, which may lead to selection bias. Secondly, positive results are no doubt more likely to be published, so the dominance of positive studies may lead to an overestimation of the efficacy of curcumin, although the funnel plot and Egger's regression test did not show significant publication bias. Thirdly, the overall quality of the animal studies was moderate, ranging from two to six points out of 10, for example, many animal studies ignored the need for randomization. In addition, myocardial infarction usually occurs in patients with cardiovascular risk factors, such as aging, diabetes, hypertension, and hyperlipidemia (Blankstein et al., 2012). However, no animal model with comorbidities was set in the included studies, so their results may be inconsistent with the complexity of the real medical environment.

Conclusion
Our study is the first study that includes both preclinical and clinical evidence in terms of curcumin in treatment of MIR, and our results suggest that curcumin might play a cardioprotective role in acute myocardial infarction, mainly through its anti-oxidative, antiinflammatory, anti-apoptosis, and anti-fibrosis effects. In addition, we found that curcumin might need a longer course of treatment and a larger dosage to exert cardioprotection in the clinical studies, and its efficacy is mainly reflected in reducing the incidence of myocardial infarction and MACE in short term. Finally, our study summarized the main defects of the existing research and suggested how such primary studies need to be improved.

Data availability statement
The original contributions presented in the study are included in the article/Supplementary Material, further inquiries can be directed to the corresponding authors.