Participants: Adults aged ≥18 years diagnosed with hyperlipidemia based on clinical criteria.

Interventions: Experimental groups received phytosterol-enriched food supplements (e.g., functional foods or nutritionally enhanced diets).

Control: The placebo group received inactive food matrices that were visually and indistinguishable in taste and appearance from the active interventions.

Outcome: The primary outcomes were lipid parameters, including Low-density lipoprotein cholesterol (LDL-C), Total cholesterol (TC), and the inflammatory marker C-reactive protein (CRP). The secondary outcomes included Triglycerides (TG) and High-density lipoprotein cholesterol (HDL-C). The Studies must provide data on at least one outcome parameter.

Study type: Randomized controlled trials (RCTs).

1. Studies on phytosterols combined with other drugs/nutrients intervention; 2. Studies that are replications of published studies; 3. Studies for which the full text or incomplete data were unavailable; 4. Reviews, conference abstracts, animal experimental studies, etc.; and 5. Studies on phytosterol pharmaceutical supplements.

Two researchers independently searched nine databases (China National Knowledge Infrastructure (CNKI), Wanfang Data, VIP, SinoMed, PubMed, Cochrane Library, Embase, Scopus, Web of Science) from inception to 15 February 2025. A hybrid search strategy combining subject headings with free terms was employed. Search terms included: Phytosterols, phytosterol*, Plant sterol*, Phytostanol*, Sitosterol*, plant stanol*, sitostanol*, Campestanol*, Stigmasterol*, Stigmastanol*, brassicasterol*, Hypercholesterolemia, hyperlipoproteinemia, Hyperlipemia, dyslipidemias, randomized controlled trial, RCT, random, stud*. The language filter included both Chinese and English literature.

Two researchers independently conducted systematic search and literature management using EndNote 20, ensuring duplicate removal. Subsequently, the titles and abstracts were carefully reviewed to exclude irrelevant studies, followed by a thorough examination of the full texts to select relevant articles based on predefined inclusion and exclusion criteria. Data extraction was performed by two researchers independently, encompassing information such as the first author’s name, publication year, publication country, participants involved, sample size, intervention measures employed, intervention duration, outcome indicators assessed, among others.

The Cochrane Risk of Bias Tool was utilized to evaluate the methodological quality in seven domains: Randomization sequence generation; Allocation concealment; Blinding of participants and personnel; Blinding of outcome assessment; Incomplete outcome data; Selective reporting and Other bias. Studies were categorized into three quality levels (low, high, or unclear risk of bias) based on the risk-of-bias diagram. In case of discrepancies during the above process, a third researcher would act as an arbitrator to facilitate consensus-building.

The certainty of the evidence was assessed using the Grading of Recommendations, Assessment, Development and Evaluations (GRADE) approach. The quality of evidence for Randomized Controlled Trials (RCTs) was initially rated as high according to the GRADE methodology. This rating may be downgraded to moderate, low, or very low if limitations are identified in any of the five domains: Risk of bias, Inconsistency, Indirectness, Imprecision, or Publication bias. Conversely, the evidence quality may be upgraded if large effect sizes or dose-response gradients are observed. In the event of disagreement during this assessment, a third researcher served as an arbitrator to reach a consensus.

Meta-analysis was performed using Review Manager 5.4, adhering to the PRISMA guidelines for eligible studies. The results were presented as forest plots. Interstudy heterogeneity was assessed by the Cochrane heterogeneity test: a fixed-effect model was applied if P ≥ 0.1 and I² ≤ 50%, otherwise a random-effects model was used for analysis. All outcome measures were standardized continuous variables, and effect sizes were reported as weighted mean differences (MD) with 95% confidence interval (CI). A P-value < 0.05 denoted statistical significance.

A total of 2,925 relevant literatures were obtained from the preliminary search database, and after excluding 1,375 duplicate literatures, 1,550 literatures remained. After preliminary screening by reading titles and abstracts, 1,436 articles that clearly did not meet the inclusion criteria were excluded, and 114 articles that might meet the inclusion criteria were obtained. After further reading the full text, 100 ineligible literatures were excluded, and 14 literatures were finally included (Dewi et al., 2024; Wang et al., 2015; Orem et al., 2017; Eady et al., 2011; Lestiani et al., 2018; Athyros et al., 2011; Vásquez-Trespalacios and Romero-Palacio, 2014; Hallikainen et al., 2013; Oliveira et al., 2020; Theuwissen et al., 2009; Kriengsinyos et al., 2015; Buyuktuncer et al., 2013; Cicero et al., 2023; Dong et al., 2016). The literature screening process and results are shown in Figure 1.

This study incorporated 14 randomized controlled trials involving 1,088 hyperlipidemic patients. All studies described the baseline characteristics of the two groups, and these groups were comparable. Furthermore, all studies reported the outcome measures. Detailed baseline information for the included studies is presented in Table 1.

The methodological quality of the 14 included studies was systematically assessed using the Cochrane Risk of Bias tool, revealing heterogeneous risk profiles across evaluation domains. Eight studies (Wang et al., 2015; Eady et al., 2011; Vásquez-Trespalacios and Romero-Palacio, 2014; Hallikainen et al., 2013; Oliveira et al., 2020; Kriengsinyos et al., 2015; Buyuktuncer et al., 2013; Cicero et al., 2023) demonstrated low risk through explicit random number table or stratified randomization methods, while six studies lacked sufficient detail on randomization processes and were classified as unclear risk. Allocation concealment was adequately described in two studies (Eady et al., 2011; Cicero et al., 2023), earning them low risk ratings, but unclear in the rest. Regarding blinding, all studies reported participant blinding with low risk; three studies (Dewi et al., 2024; Eady et al., 2011; Kriengsinyos et al., 2015) further blinded outcome assessors, reinforcing their low risk classification. Conversely, 10 studies omitted details regarding blinding, resulting in unclear risk, and one study (Oliveira et al., 2020) failed to blind statisticians, resulting in a high-risk designation. Data integrity was generally robust, as all studies documented dropout rates and reasons, yielding low risk for bias. Reporting bias remained unclear due to insufficient evidence of selective outcome reporting, and no studies described quality control protocols. These findings are comprehensively presented in Figures 2A,B.

According to the GRADE assessment, the certainty of evidence was rated as low for TG, LDL-C, HDL-C, and TC outcomes. For CRP, the evidence was deemed very low certainty. These downgrades primarily stem from serious limitations in risk of bias (due to inadequate randomization, allocation concealment, and blinding) and imprecision (attributable to small sample sizes and wide confidence intervals). The evaluation details are in Table 2.

Only one of the included studies (Lestiani et al., 2018) on smoothie drinks reported phytosterol-related gastrointestinal adverse effects. There were no differences between the study groups in the symptoms of mild and transient changes in stool characteristics, upper abdominal discomfort, abdominal distension, increased flatulence and dyspepsia.

The pooled effect sizes for TC, LDL-C, HDL-C, and CRP remained statistically significant (P < 0.05) and unchanged after excluding any single study. The confidence intervals consistently stayed within the clinical significance threshold, indicating robustness of the results. For TG, removing the study (Oliveira et al., 2020) substantially reduced heterogeneity (I² from 85% to 0%) and yielded a more precise estimate (MD = −0.18, 95% CI −0.22 to −0.14, P < 0.00001). The results remained stable. This adjustment was attributed to the extreme effect size in the excluded study, which may reflect methodological differences or data distribution variations compared to other trials.

Funnel plot analysis was performed on TC, TG, LDL-C and HDL-C of ≥10 included studies: The funnel plot of TC and LDL-C was symmetrical, and the study points were distributed symmetrically on both sides of the axis, with a small publication bias. However, the funnel plots of TG and HDL-C showed funnel asymmetry, suggesting potential publication bias and heterogeneity. See Figure 8A–B.

This meta-analysis found phytosterol consumption significantly reduced TC and LDL-C while increasing HDL-C. Effects on TG and CRP, however, were inconsistent across studies (Amir Shaghaghi et al., 2013; Jie and Kang, 2017; Gao et al., 2023; Jm et al., 2009). While two studies reported no significant impact on TG and HDL-C (Jie and Kang, 2017; Gao et al., 2023), this study confirmed significant TG reduction and HDL-C elevation in hyperlipidemic patients.

A dose-dependent TG reduction was observed. Subgroup analysis indicated that high-dose phytosterol intake (>2 g/day) significantly lowered TG levels (MD = −0.31, 95% CI −0.56 to −0.07, P = 0.01), particularly in individuals with baseline TG > 150 mg/dL (Orem et al., 2017; Theuwissen et al., 2009). In contrast, the low-dose group (≤2 g/day) and overall combined analysis showed non-significant trends (MD = −0.25, 95% CI −0.53 to 0.03, P = 0.08). A dose-effect correlation was confirmed with phytosterol-fortified milk (Wang et al., 2015). However, the Egger test, conducted using Stata17 software to explore publication bias, suggested potential publication bias (P = 0.0154), and the effect size remained stable (Hedges’s g = −0.387, 95% CI −0.794 to 0.02) after trim-and-fill analysis, possibly due to small sample sizes and methodological heterogeneity. Thus, current evidence is insufficient to conclusively support the TG-lowering effect of phytosterols, and further large-scale studies are needed.

Regarding HDL-C, studies (Sun et al., 2014; Wang et al., 2015) reported significant increases, whereas another study (Ras et al., 2015) observed no significant changes. A meta-analysis (Jie and Kang, 2017) suggested that phytosterols had no effect on HDL-C in either healthy individuals or hyperlipidemic populations. Indicating that population-specific metabolic characteristics may influence the observed outcomes. The Egger test did not detect significant publication bias (P = 0.6054).

Several studies (Gagliardi et al., 2010; Athyros et al., 2011; Devaraj et al., 2006) have shown that phytosterols significantly reduce CRP levels, suggesting potential anti-inflammatory mechanisms via antioxidant pathways. However, current study found no significant overall effect of phytosterols on inflammation (MD = −0.00, 95% CI −0.01 to 0.00, P = 0.32). Subgroup analysis revealed that intervention durations >4 weeks were associated with significantly lower CRP levels (MD = −0.03, 95% CI −0.04 to −0.02, P < 0.00001), indicating duration-dependent modulation of inflammation. Further research on mechanisms and clinical relevance is needed.

Phytosterols feature a sterane ring system with a C-24 methyl or ethyl group, a C-3 hydroxyl group, and one to two double bonds in ring B (Khallouki et al., 2024). Resembling cholesterol structurally, they reduce cholesterol via a multi-tiered “gut-liver-fat” regulatory axis. Their effectiveness hinges on the C-24 substituent and ring saturation. Hydrophobic C-24 groups boost efficacy, like in 4-desmethyl sterols (sitosterol, stigmasterol) which activate liver X receptors (LXRα) and upregulate ABCG5/G8 (He et al., 2018). Sitostanol, owing to its saturated structure, exhibits minimal absorption and persists within intestinal micelles and emulsions, thereby continuously disrupting cholesterol solubilization to exert hypocholesterolemic effects (Ikeda and Sugano, 1983).

In the intestine, phytosterols branched hydroxyl groups enhance lipid solubility and displace dietary cholesterol from bile acid micelles, reducing cholesterol absorption by 30%–50% (Nechchadi et al., 2024; Dumolt and Rideout, 2017; Xue et al., 2019). Key mechanisms involve the Niemann-Pick C1-Like1 (NPC1L1) protein: phytosterols enter intestinal cells via NPC1L1 but inhibit its cholesterol uptake function. They also suppress the acyl-CoA: cholesterol acyltransferase (ACAT) enzyme, reducing chylomicron formation needed for cholesterol transport into the bloodstream (Paalvast et al., 2017; Liang et al., 2011; Alphonse and Jones, 2016). Furthermore, ABCG5/G8 transporters pump absorbed phytosterols back into the intestine, limiting their circulation and further reducing cholesterol absorption (Ghosh et al., 2021).

Within the liver, phytosterols trigger LXRα, a transcription factor controlling cholesterol levels. Activated LXRα increases the expression of transporters like ABCA1 and ABCG5/G8, promoting cholesterol excretion into bile It also stimulates bile acid synthesis via cytochrome P450 7A1 (CYP7A1), creating a feedback loop that lessens cholesterol reabsorption (He et al., 2013; Pannu et al., 2013). Furthermore, phytosterols interfere with the activation of SREBP2, a protein crucial for making cholesterol. This reduces the activity of HMG-CoA reductase, a key enzyme in cholesterol biosynthesis (Alphonse and Jones, 2016; Batta et al., 2006).

Regarding blood fats, phytosterols lower liver TG production by blocking fat-making enzymes and help break down TG by boosting an enzyme called lipoprotein lipase (Nechchadi et al., 2024). Phytosterols also reduce cholesterol absorption and synthesis, thereby prompting an upsurge in endogenous cholesterol production and augmenting the hepatic uptake of plasma LDL-C. enhancing its clearance and lowering plasma concentration (Poli et al., 2021).

Recent studies Indicated that non-nutrient bioactive compounds, like polyphenols and phytosterols, found in plant foods, have therapeutic potential for chronic diseases. These compounds possess antioxidant, anti-inflammatory, and other health-promoting properties, acting through mechanisms distinct from conventional nutrients. (Zhu et al., 2023). Researchers have analyzed common healthy dietary patterns such as the Mediterranean and Japanese diets, highlighting the significance of non-nutrients in disease prevention and health promotion. Based on this analysis, “theoretical model of family nurse diet therapy” has been proposed, highlighting the potential benefits of non-nutrients within dietary interventions (Han et al., 2023). This theory states that non-nutrients aid in the treatment of chronic diseases through their anti-inflammatory, antioxidant, and metabolic regulatory properties.

Study (Jia et al., 2024a) confirmed that polyphenol-rich non-nutrient foods can improve metabolic abnormalities in patients with hyperlipidemia. Furthermore, a systematic review (Jia et al., 2024b) on patients with coronary heart disease demonstrated that polyphenol-rich seed foods can significantly reduce blood lipid and inflammation levels in patients with coronary heart disease. It is worth noting that as an important non-nutrient, phytosterols exemplify this theory by integrating cholesterol-lowering, anti-inflammatory, and antioxidant effects. While humans lack endogenous synthesis pathways for these sterols, dietary supplementation confers measurable health benefits (Shi et al., 2023).

Current guidelines recommend ≥2 g/day of phytosterols to achieve LDL-C reductions (Yuen et al., 2019), with some trials suggesting enhanced efficacy at doses >2.5–3 g/day (Fontané et al., 2023). However, the efficacy of phytosterols is influenced by several factors, including the food matrix, dosage, and intervention duration (Abumweis et al., 2008). For instance, oil-based carriers such as palm oil and butter may enhance the bioavailability of phytosterols due to their solubility advantages, while water-soluble substrates like soy milk and yogurt show relatively limited effects (Dewi et al., 2024). Despite these differences, one study has reported that the lipid-lowering efficacy of phytosterols is independent of the food substrate (American Heart Association Nutrition Committee et al., 2006). Long-term intake of phytosterols has not been associated with serious adverse reactions, but the bioavailability differences among various substrates should be considered. Existing evidence indicates that phytosterols offer potential health benefits in dyslipidemia populations. However, the clinical translation of their anti-inflammatory and lipid-lowering mechanisms requires further verification through large-scale, high-quality studies.

Food is increasingly recognized as a frontline therapy for chronic disease management, offering advantages in safety, accessibility, and sustainability over pharmaceuticals (Moreno-Fernández et al., 2018). With the in-depth research on nutrition and chronic diseases, “Functional food” has attracted much attention as a new strategy for disease prevention and treatment, which has health-promoting and potential therapeutic use for chronic diseases (Malaguti et al., 2014). As natural, safe, and cost-effective dietary components, phytosterols exemplify this approach. Evidence indicates that dietary adjustments increasing plant sterol intake can effectively reduce chronic disease risk and improve health (Nattagh-Eshtivani et al., 2022).

In clinical practice, plant sterol intake strategies should be individualized to patients’ lipid profiles and dietary preferences. For patients with elevated baseline lipids, fortified foods (e.g., margarine, cereals) offer standardized dosing and effortless dietary integration. Conversely, those with lower lipid levels or seeking holistic nutrition benefit more from natural sources like nuts and vegetable oils, which simultaneously provide essential fatty acids and fiber while increasing sterol intake. Healthcare professionals must therefore tailor sterol source and dosage to each patient’s metabolic status and health objectives to optimize hyperlipidemia prevention and management.

The studies included in this study were all randomized controlled trials that had been assessed by Cochrane risk of bias, covering 11 countries, and the evidence level was high. As natural dietary components, plant sterols have a significantly lower incidence of adverse events than chemical drugs, which is in line with the concept of “homologous medicine and food” and has prominent clinical safety advantages. This study provides robust evidence that phytosterols not only lower lipid levels but also modulate TG levels in a dose-dependent manner in hyperlipidemic patients, with a clinically significant reduction in TG requiring an intake >2 g/day. Secondly, it reveals that the duration of intervention >4 weeks is crucial for its anti-inflammatory effects, significantly reducing CRP levels. Furthermore, within the framework of “Functional food” and non-nutrient therapy, phytosterols have been shown to be a safe, cost-effective, and health-promoting intervention, offering a new approach to the nutritional management of chronic diseases.

This study also has some limitations. The scope of study search is limited to Chinese and English studies, and high-quality studies in non-English/non-Chinese regions may be missed. Most of the included studies were short - and medium-term trials, and lack of long-term efficacy and safety tracking of phytosterols may affect the results of phytosterols on blood lipids and inflammatory indicators. Subgroup analysis of certain outcome indicators, such as TG is limited by small sample size, and the results may be biased. In the future, multi-language, multi-center, long-term randomized controlled trials should be carried out, and personalized dosing strategies should be explored.