Impacts of Supplementation with Silymarin on Cardiovascular Risk Factors: A Systematic Review and Dose–Response Meta-Analysis

It has been suggested that silymarin (SIL) supplementation has positive effects on cardiovascular health and reduces the risk of cardiometabolic syndrome (CMS). This systematic review and dose–response meta-analysis assessed the impacts of SIL administration on cardiovascular risk factors. A systematic search of multiple databases was performed to identify eligible controlled trials published up to January 2023. The analysis used a random-effects model and included 33 trials with 1943 participants. It was revealed that SIL supplementation led to a notable reduction in serum levels of fasting blood glucose (FBG) (weighted mean difference (WMD): −21.68 mg/dL, 95% CI: −31.37, −11.99; p < 0.001), diastolic blood pressure (DBP) (WMD: −1.25 mmHg; 95% CI: −2.25, −0.26; p = 0.013), total cholesterol (TC) (WMD: −13.97 mg/dL, 95% CI: −23.09, −4.85; p = 0.003), triglycerides (TG) (WMD: −26.22 mg/dL, 95% CI: −40.32, −12.12; p < 0.001), fasting insulin (WMD: −3.76 mU/mL, 95% CI: −4.80, −2.72; p < 0.001), low-density lipoprotein (LDL) (WMD: −17.13 mg/dL, 95% CI: −25.63, −8.63; p < 0.001), and hemoglobin A1C (HbA1c) (WMD: −0.85%, 95% CI: −1.27, −0.43; p < 0.001) in the SIL-treated groups compared to their untreated counterparts. In addition, there were no substantial differences in body mass index (BMI), systolic blood pressure (SBP), C-reactive protein (CRP), body weight, and high-density lipoprotein (HDL) between the two groups. These outcomes suggest that SIL consumption reduces certain CMS risk factors and has favorable impacts on lipid and glycemic profiles with potential hypotensive effects. These findings should be supported by additional trials with larger sample sizes and longer durations.


Introduction
Cardiometabolic syndrome (CMS) is a group of risk factors or metabolic disorders that include a combination of systemic arterial hypertension, diabetes mellitus, hyperlipidemia, and central obesity [1].These metabolic conditions are linked to the onset of atherosclerotic cardiovascular disease (ASCVD) [2,3].Antidiabetic, antihypertensive, lipid-lowering, and anti-obesity medications are the main focus of routine CMS management [4].There is a growing interest in complementary and alternative medicine (CAM), such as herbal medicine, to treat metabolic dysfunction due to the persistent degenerative nature and prolonged therapy of CMS, its potential adverse effects, and its considerable economic burden [5].In addition, herbal medicine has received growing interest in developing better therapeutic strategies for secondary or primary prevention of cardiovascular diseases (CVDs) [6,7].Evidence supports the beneficial effects of silymarin (SIL) as a medicinal herb on cardiometabolic risk factors [8][9][10] with anti-atherosclerotic actions [11].
Milk thistle (Silybum marianum (L.) Gaertn) is a spiny herb that has been used as a herbal remedy and functional food ingredient for centuries [12,13].Dried milk thistle seeds were used to extract SIL [14].Silymarin is a polyphenolic flavonoid with 20-35% fatty acids, 70-80% SIL flavonolignans, and several other polyphenolic ingredients [15].Silibinin (silybin) is the most prevalent and biologically active flavonoid isomer of SIL [13].It is most commonly used for medicinal purposes to treat or protect the liver from toxic substances [12,[16][17][18].Several studies have revealed the cardioprotective and anti-obesity properties of SIL, as well as the beneficial impacts of SIL on CMS and lipid profiles [11,[19][20][21].Most of its potential therapeutic effects are related to its antioxidative characteristics [22].Dietary antioxidant supplements may exert modulatory effects on oxidative stress (OS) [23].Numerous clinical trials and reviews have been conducted to confirm the antioxidant effects of SIL and the effectiveness of its administration for the treatment of several medical disorders [22,[24][25][26][27].
Silymarin may have atheroprotective effects and reduce LDL oxidation, which is a stage in the progression of atherosclerosis [11,28].It may enhance vascular function [29,30] and have cardioprotective effects [31,32].It could increase high-density lipoprotein (HDL) levels, but its effect on total cholesterol levels is debatable [33,34].Furthermore, it has the potential to reduce levels of fasting blood glucose (FBG), triglyceride (TG), blood pressure (BP), and body weight [33][34][35][36].It may decrease overall OS levels with a specific focus on inhibiting the release of oxidants by the liver [35,37].The hepatoprotective effects of SL may positively influence various metabolic parameters [11,38].Specifically, SIL can inhibit the generation of inflammatory agents from the liver, which play a significant role in the development of CVDs [11,39].Several systematic reviews and meta-analyses have documented the positive impact of SIL on metabolic status [26], type 2 diabetes mellitus (T2DM) [24,40], and the features of CMS in adults [9,41].However, the impact of SIL supplements on cardiovascular health in various consumer or patient groups has not been thoroughly evaluated, and the results are inconclusive, contradictory, and inconsistent.Therefore, this systematic review and dose-response meta-analysis of trials assessed the impacts of supplementation with SIL on cardiovascular risk factors in adults.

Materials and Methods
The systematic review and meta-analysis protocol was registered in the international prospective register of systematic reviews (PROSPERO) with the registration number CRD42023488215.This procedure followed the guidelines outlined in the preferred reporting items for systematic reviews and meta-analyses (PRISMA) checklist [42].

Eligibility Criteria
The researchers employed EndNote software (version 20) to facilitate the export of articles.Subsequently, two researchers (V.F. and Y.J.) independently checked the abstracts with the titles of the articles and extracted the pertinent data from the eligible full-text articles.Any discrepancies were addressed through collaborative discussions with a third researcher (D.A.L.).The current systematic review and meta-analysis encompassed all trials that investigated the efficacy of supplementation with SIL on body weight, BMI, SBP, DBP, serum levels of FBG, HbA1c, TG, fasting insulin, LDL, CRP, TC, and HDL in the SIL group compared to the placebo group.
Eligible RCTs enrolled adult participants; they had control or placebo groups and presented sufficient data on predetermined outcomes in the SIL-treated and untreated groups at the study endpoint and baseline.Furthermore, SIL was not administered as a multi-component supplement.Studies that did not have the following criteria were omitted from the analysis: uncontrolled trials; observational studies utilizing case-control, cross-sectional, or cohort designs; non-peer-reviewed articles or abstracts; studies that involved pregnant women or individuals under 18 years old; articles that did not report the impact of SIL administration on the specified outcomes compared to the control group; and studies with insufficient data on baseline or follow-up measurements of body weight, BMI, SBP, DBP, serum levels of HDL, fasting insulin, FBG, HbA1c, LDL, TC, TG, and CRP.

Data Extraction
Two researchers independently extracted the data (S.M. and Y.J.).Information regarding the study's characteristics (e.g., study design, duration and location of the trial, sample size, first author's name, publication year, and dosage of SIL supplement), participants' demographic details (e.g., gender, average age, BMI, and comorbidities), initial and final measurements of specific outcomes, and mean differences in outcome changes between the study's initial and final measurements.The main outcomes were body weight, BMI, SBP, DBP, and serum levels of TG, fasting insulin, HbA1c, LDL, CRP, FBG, TC, and HDL.In addition, two researchers resolved disagreements and discrepancies through discussions.

Quality Assessment
Two separate researchers (S.M. and D.A.L.) evaluated the quality of the trials using the Cochrane risk of bias (RoB 2) tool [43].It determined potential sources of bias, including reporting bias, performance bias, attrition bias, detection bias, and allocation bias.The RoB for each domain was categorized as low, unclear, or high [43].

Statistical Analysis
This meta-analysis used Stata software (version 17).The impact of SIL was assessed through the calculation of weighted mean differences (WMDs) along with 95% confidence intervals (CIs) to measure the absolute changes in outcomes between the SIL and placebo groups from the study baseline to endpoints.The outcomes were quantified using the mean ± standard deviation (SD) measurement, and the effect size was identified by calculating the mean difference.The subsequent formula was employed to compute the SD change from the SD at the baseline of the trial to the SD at the end of the trial: square root ([SD 2  baseline + SD 2 final ] − [2 × R × SD baseline × SD final ]) [44].Furthermore, a random-effects model [45] was applied to determine the pooled WMDs.The between-study heterogeneity was assessed using the I 2 statistic [46]; I 2 values of 50%, 25%, and 75% were categorized as representing moderate, low, and high levels of heterogeneity, respectively [47].
Subgroup analyses were carried out to explore the potential factors contributing to the heterogeneity across trials.The analysis considered factors such as baseline values of serum levels of HDL, TC, FBG, TG, LDL, and CRP, as well as the dosage of SIL (≥400 mg/day vs. <400 mg/day), baseline BMI (obese (≥30 kg/m 2 ) vs. overweight (25-29.9kg/m 2 ) vs. normal (18.5-24.9kg/m 2 )), trial duration (<12 weeks vs. ≥12 weeks), and health status (healthy vs. unhealthy).A leave-one-out sensitivity analysis was applied to assess the effects of individual studies on the analysis.Statistical significance was considered at a level of p < 0.05.In addition, funnel plots and Begg's [48] and Egger's tests [49] were utilized to identify possible publication bias.Additionally, the fractional polynomial model was employed to investigate the potential non-linear effects of the duration of the trial (weeks) and the dosage of SIL (mg/day).Meta-regression analysis was implemented to assess the dose-response relationship and potential linear associations between duration trials, SIL dosage, and effect sizes [50].

GARDE Assessment
The certainty of the evidence was evaluated using the GRADE framework (Grading of Recommendations Assessment, Development, and Evaluation) categorized into four levels (moderate, high, very low, and low) [51].

Study Selection
The screening and selection processes for the included trials are illustrated in Figure 1.The first search across multiple databases yielded 896 publications.After eliminating 348 duplicate records, 548 records were reviewed, and 467 were not relevant.The remaining 81 articles were assessed for eligibility, and 33 articles were ultimately included in this meta-analysis .

Fasting Insulin
The association between serum fasting insulin levels and SIL intake was assessed in eight studies [52,53,56,67,71,75,77,80] involving 571 participants (Figure 2D).The metaanalysis revealed that serum fasting insulin concentrations decreased following SIL consumption compared to the untreated group (WMD: −3.76 mU/mL, 95% CI: −4.80, −2.72; p < 0.001); nevertheless, there was substantial heterogeneity between trials (I 2 = 98.6%, p < 0.001) (Table 2).Subgroup analyses indicated a notable reduction in serum levels of insulin in the SIL-treated group compared to controls in studies among participants with overweight or obesity, in short-term or long-term supplementation with low or high doses of SIL.

Sensitivity Analysis
Sensitivity analyses revealed that the findings were not influenced by the removal of any specific trial related to body weight, BMI, SBP, and serum levels of insulin, FBG, HbA1c, TG, TC, LDL, and CRP.However, the outcomes related to DBP values changed after excluding two studies [64,65].In addition, the outcomes related to serum HDL values changed after excluding one study [55].

GRADE Assessment
The GRADE framework was utilized to evaluate the quality of evidence for all outcomes (Supplementary Table S2).The certainty of evidence related to DBP values was deemed low due to a very serious risk of bias.For the other assessed outcomes, GRADE was rated as very low, primarily attributed to serious risks of imprecision, publication bias, very serious inconsistency, and bias.

Linear and Non-Linear Dose-Response Associations
No linear (Supplementary Figures S4 and S5) or non-linear (Supplementary Figures S2  and S3) correlation was depicted between the duration of the trial or the dosage of SIL supplement and the alterations in all assessed outcomes.

Discussion
This systematic review and dose-response meta-analysis examined the impact of supplementation with SIL on CVD risk factors.The findings suggest that SIL intake may improve certain CMD risk factors by decreasing levels of FBG, fasting insulin, HbA1c, TC, TG, LDL, and DBP.However, it did not significantly affect anthropometric measurements, SBP, or serum CRP levels.Supplementation with SIL demonstrated enhancements in glycemic control, lipid profile, and potential hypotensive effects.
The subgroup analysis revealed a substantial decrease in BMI among individuals with obesity following SIL supplementation compared to their untreated counterparts.It also indicated a notable reduction in FBG levels in trials involving unhealthy participants with obesity and baseline FBG levels >126 mg/dL during both long-and short-term supplementation (<12 and ≥12 weeks) with low or high doses of SIL (≥400 or <400 mg/d).Furthermore, a considerable decline in serum insulin levels was observed among participants with overweight or obesity following short-and long-term supplementation with low or high doses of SIL.It also suggested a significant fall in HbA1c concentration in trials with short-or long-term duration and high doses of SIL among participants with obesity.Additionally, a substantial reduction in TG levels was noted in unhealthy participants with a normal BMI or obesity and baseline serum TG levels below 150 mg/dL after long-term supplementation with high doses of SIL.It indicated a significant decrease in serum TC concentration among unhealthy obese participants during short-term supplementation with high doses of SIL.A notable fall in serum LDL levels was detected in unhealthy overweight participants after receiving high doses of SIL over a short or long period.Furthermore, a considerable reduction in serum HDL levels was depicted in participants with a normal BMI after supplementation with low doses of SIL.In addition, there was a considerable decrease in serum levels of CRP among both healthy and unhealthy participants with obesity and baseline CRP levels >3 mg/L over long-term supplementation with high doses of SIL.
The outcomes of this study regarding serum lipids were in agreement with two previous meta-analyses [41,85].Similarly, the results related to lipid profiles and glycemic parameters were consistent with those reported in another meta-analysis [26].However, there were discrepancies when compared with a meta-analysis of seven trials among patients with T2DM, which indicated that SIL did not lead to a substantial decline in serum levels of TC and TG [40].Furthermore, a meta-analysis comprising five trials demonstrated that SIL could lower FBS and HbA1c levels without affecting the lipid profile of T2DM patients [24].Although one meta-analysis indicated a positive impact of SIL on HDL levels, the clinical significance of this effect was minimal [41], which aligns with the non-significant increase observed in the present study.
In recent years, there have been extensive investigations into the potential health benefits and cardiovascular-protective properties of SIL and its supplements.Previous studies in animals have indicated that SIL intake may reduce the risk of CVDs [86,87].Silymarin has been shown to contribute to the restoration of pancreatic function [88].Animal studies have demonstrated that SIL and its components have the potential to improve glucose regulation in diabetic rats [89,90].It could promote the restoration of pancreatic function by upregulating insulin and glucagon proteins, leading to the normalization of blood glucose levels and the restoration of insulin serum levels [88].It has been reported that supplementation with SIL may improve the restoration of Langerhans islet β cells [91], stimulation of beta precursor cells to develop into insulin-producing cells, and reduction of advanced glycation end product (AGE) formation [92].Silymarin exhibits low water solubility and limited absorption following oral administration [93], potentially leading to insignificant responses in certain studies.Therefore, enhancing the bioavailability of SIL by consuming it with or after a meal may increase its effectiveness [85].Additionally, evidence suggests that SIL may promote the proliferation of insulin-producing cells [94].Silymarin was considered safe for human consumption at therapeutic levels and was well tolerated, even at a high dosage of 700 mg administered three times daily over 24 weeks [95].
The positive impacts of SIL on metabolic status are related to its antioxidant effects [28].In addition, SIL may have beneficial impacts on blood lipids through various mechanisms [9,78,[96][97][98][99][100].Hypertension is a significant factor contributing to CVDs [101].Existing evidence suggests that the mechanisms through which SIL influences the prevention and management of high BP can be categorized into various groups [102].Silymarin plays a role in regulating vascular tone and suppressing platelet aggregation, thereby contributing to the regulation of BP in individuals with hypertension [102].
It has been indicated that incorporating antioxidant-rich foods into the diet may enhance the treatment of patients with T2DM [103].Given the antioxidant properties of SIL, it is hypothesized that it may be particularly effective in improving CVD risk factors; however, further evidence is required to fully support this assertion.The present systematic review and meta-analysis demonstrated a considerable cardioprotective benefit of supplementation with SIL.
It is important to emphasize that in all clinical trials, participants received standard treatment in conjunction with SIL supplementation [41].Consequently, the specific effects of SIL remain unclear [41].However, discontinuing standard treatment for ethical reasons was not feasible, and only the synergistic effects were investigated.Furthermore, the stringent regulations governing the production of nutraceuticals and herbal and dietary supplements are less comprehensive than those for pharmaceuticals.This could potentially lead to ineffectiveness in some cases due to the limited examination of effective dosage, bioavailability, and formulation [104].Since most studies did not provide detailed information on SIL supplements, addressing this aspect was not feasible in the current systematic review and meta-analysis [41].
This first dose-response meta-analysis examined the effect of SIL consumption on CVD risk factors.The selection of trials for the systematic search was not limited by period or language.Moreover, a sufficient number of trials was evaluated in the analysis.Additionally, supplementary analyses including subgroup analyses, publication bias assessments, and sensitivity analyses were conducted.Systematic reviews and meta-analyses are widely regarded as the most reliable forms of clinical evidence [104,105].However, evidence regarding the efficacy of herbal dietary supplements remains inconclusive [106].Key concerns related to herbal nutraceuticals include the absence of standardized extracts, ambiguity regarding plant parts, and failure to disclose the type of extract [106].The absence of rigorous regulations poses challenges for nutraceutical manufacturers in verifying the quality, safety, and effectiveness of their products [106].Consequently, numerous available products may be ineffective [106].
This meta-analysis encountered various limitations, including significant methodological and clinical heterogeneity among trials.This heterogeneity was evident in the varying trial durations, doses of SIL supplements, participants' underlying health conditions, and sample sizes.The majority of trials had poor quality with a high risk of bias.Moreover, over half of the trials in this meta-analysis were performed in a single country (Iran), emphasizing the necessity for additional research in various geographical locations to validate positive results across different ethnic populations.In addition, the included trials involved diverse non-intervention or control groups.

Conclusions
In summary, the results of this meta-analysis indicated that SIL may mitigate certain cardiometabolic risk factors by reducing FBG, fasting insulin, HbA1c, TC, TG, LDL, and DBP levels.However, it did not significantly affect the anthropometric parameters, SBP values, or serum CRP levels.It demonstrated positive effects on lipid and glycemic profiles with potential hypotensive impacts but did not affect anthropometric or inflammatory parameters.SIL has the potential to be a beneficial adjunctive treatment for enhancing various aspects of chronic CMS.However, the existing data show significant variability and a lack of extensive clinical trials for specific parameters.These findings should be validated by additional well-designed RCTs with larger sample sizes and longer durations to ascertain the clinical efficacy and potential therapeutic application of SIL supplementation in cardiovascular health.

Figure 1 .
Figure 1.Flow chart of study selection.

Figure 1 .
Figure 1.Flow chart of study selection.

Table 1 .
Characteristic of included studies in meta-analysis.

Table 2 .
Subgroup analyses of the impacts of silymarin supplementation on cardiovascular risk factors.