L. barbarum (Lycium barbarum L.) supplementation for lipid profiles in adults: A systematic review and meta-analysis of RCTs

Abstract

Background:

Dyslipidemia is a global health concern with an increasing prevalence worldwide. Lycium barbarum (L. barbarum) is widely used as a medicinal and functional food, and evidence suggests that it may be beneficial for lipid management. In this study, we performed a systematic review and meta-analysis of randomized controlled trials investigating the effects of L. barbarum supplementation on lipid profiles in adults.

Methods:

PubMed, China National Knowledge Infrastructure, The Cochrane Library, Web of Science, and Wanfang Database were searched from inception until October 2022. The random-effect model was applied, and the pooled effect sizes were expressed as mean differences (MDs) and 95% confidence intervals (CIs).

Results:

The meta-analysis of 5 randomized controlled trials involving 259 subjects indicated that L. barbarum supplementation significantly decreased the triglyceride (TG) concentration (MD: 0.14 mmol/L, 95% CI: 0.08–0.20) and increased the high-density lipoprotein cholesterol concentration (HDL-C) (MD: −0.07 mmol/L, 95% CI: −0.13 to −0.01). However, the reductions in total cholesterol (TC) concentration (MD: 0.11 mmol/L, 95% CI: −0.37 to 0.59) and low-density lipoprotein cholesterol (LDL-C) concentration (MD: 0.21 mmol/L, 95% CI: −0.46 to 0.89) were not statistically significant.

Conclusion:

The present study showed that L. barbarum supplementation might have some beneficial effects on TG and HDL-C concentrations in adults, and L. barbarum fruit has an even greater effect on TG and HDL-C concentrations. Considering the sensitivity analyses and limitations of the study included, further large-scale studies are needed to confirm these findings.

1. Introduction

Dyslipidemia is characterized by elevated low-density lipoprotein cholesterol (LDL-C), hypercholesterolemia, hypertriglyceridemia, and low high-density lipoprotein cholesterol (HDL-C). Dyslipidemia is a worldwide public health problem, and it is associated with several clinical conditions. For example, epidemiological evidence suggests that dyslipidemia plays an important role in coronary artery disease, stroke, type 2 diabetes mellitus, and metabolic syndrome.^[1–4]

Currently, lipid-lowering medications, such as stains are widely used for the clinical treatment and prevention of cardiovascular conditions. Evidence-based medical research has revealed that lowering LDL-C levels with statin therapy is associated with a significant reduction in cardiovascular events and mortality.^[5,6] However, stains also have undesirable side effects, such as myopathy and hepatotoxicity.^[7] In dyslipidemia management guidelines,^[8] in addition to lipid-lowering medications, the benefits of adopting and sustaining a healthy lifestyle are emphasized. Furthermore, dietary supplements and functional foods for the treatment of dyslipidemias have attracted substantial attention. Several meta-analyses studies and randomized controlled trials (RCTs) have demonstrated that functional foods and dietary supplements such as red yeast rice, berberine, and garlic have a positive effect on lipid profiles, suggesting that these kinds of products may reduce cardiovascular risks.^[9–11] Therefore, the use of functional foods and dietary supplements was recommended to help people to achieve lipid targets.^[12]

Lycium barbarum L., also known as wolfberry or Goji berries, is widely used for food and medicine in Asia, including China, Korea, and Japan. L. barbarum and Lycium chinense (L. chinense)are two closely related species,^[13] and they have been used for more than 2000 years with early records dating back to the Tang Dynasty.^[14] However, only L. barbarum is recorded in the Pharmacopoeia of the People’s Republic of China. The fruits of L. barbarum is used in several processed forms. The dried fruit is administrated with other herbs or consumed alone with hot water, at a dose varying between 6 and 18 g.^[15] As a good source of homology of medicine and food, L. barbarum was consumed with soups containing a lot of meat and vegetables, which is typical of the traditional Chinese diet. In recent years, L. barbarum has become very popular in Europe and North America.^[13] The fruit of L. barbarum and wolfberry juice are sold as health foods and anti-aging supplementations. L. barbarum contains many phytochemicals and nutrients, including carotenoids, flavonoids, zeaxanthin, Lycium barbarum polysaccharides (LBP), fats, proteins, vitamins, and mineral elements.^[16] LBP, the main active ingredients in the fruit of L. barbarum, have been reported to have some health benefits, such as anti-inflammatory,^[15] antioxidant,^[17,18] hypoglycemic and hypolipidemic^[19–21] properties in recent years. Because of these health properties, L. barbarum has excellent potential as a functional food^[15,22] and for the prevention of age-related diseases, including cardiovascular disease,^[23] neurodegenerative diseases,^[24] type 2 diabetes,^[25] and age-related macular degeneration.^[26]

Results from previous animal models and clinical studies have suggested that L. barbarum has the potential to lower lipid concentrations. In mouse models fed a high-fat diets,^[27–29] LBP significantly reduced LDL-C, TC, and triglyceride (TG) when compared with the high-fat diet only group. However, fewer studies have investigated the clinical efficacy of dried L. barbarum and LBP on lipid profiles. The first meta-analysis on the effect of Lycium barbarum L. on cardiovascular risk showed that L. barbarum supplementation marginally reduced the TG and TC concentrations,^[23] but the results did not reach a statistically significant difference. Moreover, this study did not address the effects of the L. barbarum supplementation on the HDL-C and LDL-C concentrations. In 2021, a meta-analysis conducted by Zhou et al^[30] summarized that the LBP supplementation may have significant effects on TG, HDL-C, and LDL-C concentrations. A recent meta-analysis assessing both L. barbarum and L. chinense showed that wolfberry as a whole fruit in dietary regimens improved TG and HDL-C concentrations.^[31] Together, these meta-analyses suggested that the effects of L. barbarum on blood lipids are inconsistent in human trials, which may be due to differences in the population samples and study methods. In recent 5 years, researchers in Singapore^[32] and Brazil^[33] have conducted RCTs to study the health benefits of dried L. barbarum fruit in combination with a healthy diet. In addition, Chinese researchers have also conducted a clinical study to investigate the effects of LBP in the healthy male population at the metabolomics level.^[34] An update on these previous meta-analyses is essential. Therefore, we conducted a systematic review and meta-analysis of RCTs investigating the effects of L. barbarum supplementation on lipid profiles in adults, which may provide a clearer view on the use of L. barbarum supplementation for dyslipidemia management.

2. Methods

This study complied with the guidelines of the Preferred Reporting Items for Systematic reviews and Meta-Analyses (PRISMA) statement guidelines.^[35] The study protocol was registered on the International Platform of Registered Systematic Review and Meta-analysis Protocols (INPLASY) with registration number INPLASY2021110043.^[36] Because this study does not involve private patient information, no ethical review is required.

2.1. Search strategy

PubMed, the China National Knowledge Infrastructure, Cochrane Library, Web of Science, and Wan-fang databases were searched from inception to August 2022 with no language restriction. The following keywords were used in the search: “lycium” OR “wolfberry” OR “goji” AND “lipid” OR “intervention study” OR “intervention” OR “controlled trial” OR “randomized” OR “random” OR “randomly” OR “placebo” OR “assignment.” Furthermore, we reviewed the references of the retrieved articles to identify possible eligible articles.

2.2. Inclusion criteria

The inclusion criteria were studies involving participants aged ≥ 18 years, regardless of their sex or disease duration; using different L. barbarum supplement forms, including L. barbarum, LBPs from L. barbarum fruit, and L. barbarum with a healthy diet; comparing with placebo or healthy diet; with the outcome measures of total cholesterol (TC), TG, low-density lipoprotein cholesterol (LDL-C), and high-density lipoprotein-cholesterol (HDL-C); and with an RCT design, either parallel or open-label.

2.3. Exclusion criteria

The exclusion criteria were studies that studies that administrated L. barbarum in combination with other herb components; studies that used L. chinense as an intervention; and used the purified polysaccharide fraction as an intervention.

2.4. Data extraction

Two reviewers strictly followed the inclusion criteria to selected RCT publications related to the topic. After reading the full texts of the selected articles, 2 reviewers independently extracted the data using a standard form. Discrepancies in the inclusion or exclusion of articles during screening were discussed until consensus was reached, and any final discrepancies were resolved by a third reviewer (J.X.R.). The extracted information was as follows: study characteristics (first author’s last name, year of publication, study location, sample size, and study design); participants’ information (gender, mean age, and health status); intervention details (treatment duration and dose); and investigated outcomes including total cholesterol (TC), triglyceride (TG), LDL-C, and HDL-C concentrations.

2.5. Risk of bias assessment

The included studies were independently assessed by 2 reviewers using the Cochrane Handbook for Systematic Reviews to determine the risk of bias. The Cochrane Collaboration’s tool was used to assigned scores for the following domains: random sequence generation (selection bias); allocation concealment (selection bias); blinding of participants and personnel (performance bias); blinding of outcome assessment (detection bias); incomplete outcome data (attrition bias); selective reporting (reporting bias); and other sources of bias. The risk of bias was evaluated at 3 levels: low risk, high risk, and unclear risk.

2.6. Statistical analysis

Baseline and endpoint mean ± standard deviations (SD) values for TC, TG, LDL-C, and HDL-C in both the intervention and control groups were extracted. If studies reported the mean change value before compared with that after the intervention, the mean change and SD were extracted. When the SD was not reported, the standard error (SE) was extracted and converted into an SD value for further analyses using the following formula: SD = SE × sqrt (n), where n is the number of subjects. The SDs of the mean differences (MDs) were calculated using the following formula: SD = square root [(SD_baseline)² + (SD_endpoint)² − (2R × SD_baseline × SD_endpoint)], assuming a correlation coefficient (R) of 0.5. If the data were only reported in graphical form, the values were extracted using with an online tool (https://apps.automeris.io/wpd/).

Heterogeneity between the included studies was analyzed using the I² statistic. The random-effects model (DerSimonian-Laird, D-L) was performed. For all included outcomes, the pooled effect sizes are expressed as MDs and 95% confidence intervals (CIs) with forest plots. Pre-planned subgroup analyses were conducted based on treatment duration (>3 month vs ≤ 3 month), health status and intervention types to explore possible sources of inter-study heterogeneity. Sensitivity analyses were performed using the one-study remove approach to evaluate the stability. Funnel plots were not included in this study, because tests for funnel plot asymmetry was not recommended when a meta-analysis contains fewer than ten studies. All statistical analyses were performed using Rstudio software (Version 1.4.1717).

3. Results

3.1. Selection and identification of studies

The step-by-step details of study identification and selection are illustrated in Figure 1. A total of 1162 potential citations were identified from the database search. After removing duplicated publications, 888 articles remained for title and abstracts screening. After 858 articles were excluded based on title and abstract screening, 30 articles were selected for the full-text evaluation. Twenty-four articles’ papers were excluded for the following reasons: non-RCT design, ineligible intervention, repeatedly publication, and no relevant outcomes. Ultimately, 6 articles were eligible for this systematic review, and 5 articles were used for meta-analysis.

Figure 1.

The PRISMA flow diagram of the study selection process.

3.2. Study characteristics

The detailed characteristics of the included RCTs were described in Table 1. The data were pooled from 6 RCTs with a sample size ranging from 40 to 67 subjects. Overall, 309 participants were included in these trials. The selected studies were published between 2009 and 2021. Four studies were conducted in China,^{[25,34,37,38]} and 2 studies were carried out in Brazil and Singapore.^[32,33] All trials had a parallel study design. The mean age of the participants ranges between 23 and 58 years old. Five studies were conducted in both sexes,^{[25,32,33,37,38]} and one study was conducted in healthy males.^[34] RCTs were performed in various populations, including type 2 diabetes mellitus (1 trial), metabolic syndrome (1 trial), and healthy people (4 trials). Different L. barbarum supplement forms were used for the intervention. Two studies used dried wolfberry with a healthy diet,^[32,33] 2 studies used LBP-standardized L. barbarum fruit juice,^[37,38] and the remaining 2 studies used LBP capsules.^[25,34] Overall, the intervention periods ranged from 4 to 16 weeks. The control groups also differed; placebo was used for the control group in 4 studies,^{[25,34,37,38]} whereas 2 studies used a healthy diet pattern.^[32,33] Four trials were conducted without requiring patients to change their previous eating habits or exercise patterns.^{[25,34,37,38]}

Table 1.

Characteristics of 6 randomized controlled trials selected for systematic review and meta-analysis.

Study	Country	Sample size (I/C)	Gender(M/F)	Mean Age	Health status	L. barbarum forms	Daily dosage	Control	Duration	Ref
Amagase, 2009a	USA	50 (25/25)	25/25	58.1	Healthy Chinese adults	LBP-standardized L. barbarum fruit juice	120 mL	Placebo	30 d	(Amagase, Sun, and Borek 2009)
Amagase, 2009b	USA	60 (30/30)	NA	58.9	Healthy Chinese adults	LBP-standardized L. barbarum fruit juice	120 mL	Placebo	30 d	(Amagase, Sun, and Nance 2009)
Cai, 2015	China	67 (37/30)	37/30	57	Type 2 diabetes	LBP capsule	300 mg	Placebo	3 mo	(Cai et al 2015)
Zanchet, 2017	Brazil	50 (25/25)	25/25	50.9	Metabolic syndrome	Dried Goji berry with a healthy dietary	14 g	Healthy diet pattern	45 d	(Zanchet et al 2017)
Xia, 2018	China	42 (21/21)	0/42	23.79	Healthy male	LBP capsule	300 mg	Placebo	4 wk	(Xia et al 2018)
Toh, 2021	Singapore	40 (22/18)	29/11	56	Healthy participants	Dried Goji berry with a healthy dietary	15 g	Healthy diet pattern	16 wk	(Toh et al 2021)

I/C = intervention/control, LBP = Lycium Barbarum Polysaccharides, M/F = male/female.

3.3. Risk of bias

The details of the risk of bias assessment are described in Table 2. We evaluated the risk of bias assessment for 6 RCTs according to the Cochrane Collaboration’s tool. In summary, the risk of performance bias, attrition bias, detection bias and reporting bias was low in most of the studies. Three studies had an unclear risk of bias for allocation concealment.^[25,33,34] Two of the 6 studies had a low risk of bias in the generation of random sequences,^[32,34] whereas the rest of the studies had an unclear risk of bias due to insufficient information.^{[25,33,37,38]} One study had an unclear risk of bias for participants and personnel blinding and outcome assessment blinding.^[33] All studies were had a low risk of other sources of bias.

Table 2.

The Cochrane collaboration’s tool for the risk of bias assessment of studies selected for systematic review and meta-analysis.

Study	Random sequence generation	Allocation concealment	Blinding of participants and personnel	Blinding of outcome assessment	Incomplete outcome data	Selective reporting	Other sources of bias
Amagase, 2009a	Unclear	Low	Low	Low	Low	Low	Low
Amagase, 2009b	Unclear	Low	Low	Low	Low	Low	Low
Cai, 2015	Unclear	Unclear	Low	Low	Low	Low	Low
Zanchet, 2017	Unclear	Unclear	Unclear	Unclear	Low	Low	Low
Xia, 2018	Low	Unclear	Low	Low	Low	Low	Low
Toh, 2021	Low	Low	Low	Low	Low	Low	Low

3.4. Effects of L. barbarum supplementation on the TC concentration

Four studies reported the TC concentration. The pooled effected size showed a non-significant reduction in the TC concentration after L. barbarum supplementation (Fig. 2A; MD: 0.11 mmol/L, 95% CI: −0.37 to 0.59) with significant between heterogeneity (I² = 70.2%, P = .12). Results of subgroup analyses showed that no significant reduction of TC levels was observed in both L. barbarum extracts (MD: −0.03 mmol/L, 95% CI: −0.48 to 0.42) and L. barbarum fruit. Heterogeneity (I² = 90%, P = .02) was significant in the subgroups with L. barbarum fruit. However, heterogeneity (I² = 0%, P = .97) was not observed in the subgroups with L. barbarum extract (Table 3).

Figure 2.

Forest plots for RCTs examining the effects of L. barbarum supplementation on lipid profiles in adults. HDL-C = high-density lipoprotein-cholesterol, LDL-C = low-density lipoprotein cholesterol, MD = mean difference, RCTs = randomized controlled trials, SD = standard deviation, TC = total cholesterol, TG = triglyceride.

Table 3.

Pooled effect of L. barbarum supplementation on lipid profiles in various subgroup.

Subgroup	No. of Trials	MD (95% CI)	I2 (%)	P for heterogeneity	P for between subgroup heterogeneity
TC
Health status
Healthy	3	−0.16 (−0.37; 0.05)	0	.81	<.01
Unhealthy	1		–	–
L. barbarum forms
L. barbarum extract	2	−0.03 (−0.48; 0.42)	0	.97	.61
L. barbarum fruit	2	0.22 (−0.66; 1.10)	90	.02
Duration
≤3 mo	3	0.29 (−0.32; 0.89)	56	.1	.14
>3 mo	1		–	–
TG
Health status
Healthy	3	0.07 (−0.12; 0.25)	0	.96	.41
Unhealthy	1		–	–
L. barbarum forms
L. barbarum extract	2	0.05 (−0.41; 0.51)	0	.77	.7
L. barbarum fruit	2	0.14 (0.08; 0.21)	0	.85
Duration
≤3 mo	3	0.15 (0.08; 0.22)	0	.88	.47
>3 mo	1		–	–
HDL-C
Health status
Healthy	2	−0.09 (−0.18; 0.01)	0	.93	.65
Unhealthy	2	−0.06 (−0.13; 0.01)	0	.92
L. barbarum forms
L. barbarum extract	2	−0.07 (−0.24; 0.10)	0	.98	.98
L. barbarum fruit	2	−0.07 (−0.13; −0.01)	0	.97
Duration
≤3 mo	2	−0.06 (−0.14; 0.02)	0	1	.66
>3 mo	2	−0.09 (−0.17; 0.00)	0	.84
LDL-C
Health status
Healthy	2	−0.15 (−0.37;0.08)	0	.88	<.01
Unhealthy	1
L. barbarum forms
L. barbarum extract	1				.75
L. barbarum fruit	2	0.26 (−0.06; 1.12)	90	<.01
Duration
≤3 mo	2	0.66 (0.20; 1.13)	0	.36	<.01
>3 mo	1

HDL-C = high-density lipoprotein cholesterol, LDL-C = low-density lipoprotein cholesterol, MD = mean difference, TC = total cholesterol, TG = triglyceride.

3.5. Effects of L. barbarum supplementation on TG concentration

A total of 4 studies reported the TG concentration. Based on the random-effect model, the combined results showed that L. barbarum supplementation significantly reduced the TG concentration (Fig. 2B; MD: 0.14 mmol/L, 95% CI: 0.08–0.20). We observed no heterogeneity among studies (I² = 0%, P = .85). Subgroup analyses (Table 3) showed that the TG concentration significantly decreased in the pooled data from individuals taking L. barbarum fruit with a healthy diet (MD: 0.14 mmol/L, 95% CI: 0.08–0.21). However, we observed no significant reduction in the TG concentration in the subgroups with L. barbarum extracts (MD: 0.05 mmol/L, 95% CI: −0.41 to 0.51).

3.6. Effects of L. barbarum supplementation on HDL-C concentration

The meta-analysis showed that the L. barbarum supplementation significantly increased the HDL-C concentrations (Fig. 2C; MD: −0.07 mmol/L, 95% CI: −0.13 to −0.01). Between-study heterogeneity was not observed (I² = 0%, P = .97). Subgroup analyses based on health status and intervention duration showed that L. barbarum supplementation did not affect the HDL-C concentration. However, L. barbarum fruit with a healthy diet significantly increased HDL-C concentration compared with LBP (Table 3).

3.7. Effects of L. barbarum supplementation on the LDL-C concentration

Four studies reported the LDL-C concentration, which included a total of 132 participants (68 interventions, and 64 control). The pooled effected size showed that L. barbarum supplementation did not reduce the LDL-C concentrations (Fig. 2D; MD: 0.21 mmol/L, 95% CI: −0.46 to 0.89). However, we found significant heterogeneity among the studies (I² = 80.5%, P = .0059). Because of the small body of literature, we did not have sufficient data to conduct a subgroup analysis to explore the source of this heterogeneity (Table 3).

3.8. Sensitivity analyses

To ensure that the meta-analysis was not sensitive to the selected correlation coefficient, all the analyses for each parameter were repeated using correlation coefficients of 0.25 and 0.75 (Table S1, Supplemental Digital Content, http://links.lww.com/MD/J660). To evaluate whether the results of the meta-analysis were driven by a single study, we performed sensitivity analyses by removing each study in turn. The sensitivity analyses showed that the pooled effects of L. barbarum supplementation on the TC concentration did not change by removing any of the studies. However, removing a single study^[33] changed the pooled effect of L. barbarum supplementation on the TG concentration (MD: 0.07 mmol/L, 95% CI: −0.12 to 0.25). Similarly, after removing the studies by Toh et al^[32] and Zanchet et al^[33] the effect of L. barbarum supplementation on the HDL-C concentration changed. The pooled effects after removing each of these studies were MD: −0.06 mmol/L (95% CI: −0.13 to 0.01) and MD: −0.07 mmol/L (95% CI: −0.17 to 0.00) for Toh et al^[32] and Zanchet et al,^[33] respectively. Furthermore, after removing the study by Toh et al the overall effect of L. barbarum supplementation on the LDL-C concentration (MD: 0.66 mmol/L, 95% CI: 0.20–1.13) became statistically significant.

4. Discussion

Lycium barbarum L. is one of the most widely used functional foods worldwide. It is available on the market in various forms of use, including L. barbarum juice and L. barbarum fruit, amongst others. The effects of L. barbarum on lipid profiles have remained controversial. We questioned whether the available RCTs support the lipid-lowering efficacy of L. barbarum. In addition, we questioned whether different forms of L. barbarum intake have similar effects on blood lipids. To answer this question and to gain a deeper understanding of the health benefits of L. barbarum in terms of blood lipid profiles, we conducted a meta-analysis of 5 RCTs. Overall, our findings demonstrated that L. barbarum supplements might have beneficial effects on lipid profiles, suggesting that L. barbarum supplements may reduce the cardiovascular risk of individuals.

The most important finding of this work is that L. barbarum supplementation had positive effects on TG and HDL-C concentrations. L. barbarum supplementation reduced the TC and LDL-C concentrations, but the results did not reach statistical significance. Elevated blood TG is associated with an increased risk of cardiovascular disease. Elevated HDL-C is associated with a lower risk of cardiovascular disease, but the mechanisms are unclear. However, our study did not find that L. barbarum supplementation reduced the LDL-C concentration, which is a major risk factor for atherosclerosis development. Therefore, we inferred that dried Goji berries themselves as part of a healthy diet have potential benefits in terms of lipid profile management, which are helpful for improving public cardiovascular health. It is worth mentioning that some study data require further review and investigation. The sensitivity analyses found altered effects of L. barbarum supplementation on TG, HDL-C and LDL-C after exclusion of the studies by Toh et al^[32] and Zanchet et al,^[33] indicating that the current findings are not statistically robust. Because of the limited number of studies involved, we were unable to confirm the exact effects of LBP supplementation on blood lipids.

Our study showed clear differences to previous studies. Only L. barbarum was included in our study, however, there are 2 similar species of Lycium barbarum L. This is a feature of our study. To measure the effect of the different forms of L. barbarum supplements on blood lipids, we included all currently retrievable studies related to L. barbarum, including studies on LBP extracted from L. barbarum fruit and wolfberry fruit. This study has some notable strengths. For example, we carefully conducted a comprehensive search of the literature in accordance with the PRISMA guidelines to minimize any bias in the review process. Moreover, we developed reasonable inclusion and exclusion criteria, which were strictly enforced. In addition, we excluded the non-RCTs,^[39] which were included in previous meta-analysis,^[23,30] improving the accuracy of the meta-analysis and the quality of the evidence.

The subgroup analyses in the present study showed that L. barbarum fruit with a healthy diet significantly reduced the TG and HDL-C concentrations, whereas the effect of L. barbarum extract on TG and HDL-C concentrations were nonsignificant. We speculated that 2 main reasons for the differences in the results of the subgroup analyses. Firstly, the fruit of L. barbarum contains several antioxidant components that seem to be associated with hypolipidemic effects. In addition to LBP, the fruit of L. barbarum contains other nutrients such as vitamins, zeaxanthin, carotenoids, and flavonoids.^[15] A previous experimental study showed that the aqueous decoction and crude LBP had better hypolipidemic effects than the purified polysaccharide fraction (LBP-X) in a rabbit model of hyperlipidemia.^[19] The antioxidant components in the aqueous extract of L. barbarum and LBP may play a synergistic role in lowering blood lipids. The bioavailability of L. barbarum fruit is higher than that of LBP. Secondly, the differences may have been related to the dietary background factors of the study populations. In the subgroup analyses, neither the health status of the subjects nor the duration of the intervention had a significant effect on the elevated HDL-C concentrations with L. barbarum supplementation. Dietary patterns may have influenced the results of these studies. The Chinese subjects consumed a typical Chinese diet during the studies conducted by Cai et.al and Xia et al,^[25,34] while the subjects from Singapore and Brazil had a healthy dietary pattern. The importance of maintaining a healthy lifestyle is emphasized in the current guidelines for the management of dyslipidemia. Maintaining a healthy lifestyle can significantly lower TG, HDL-C and LDL-C concentrations, in turn reducing cardiovascular risk. Therefore, the dietary context as a confounding factor is an issue that needs to be carefully considered in the RCT design. The results of the subgroup analysis in a previous meta-analysis^[23] showed that supplemental L. barbarum significantly reduced the TG and TC levels in subjects with a mean age of ≥ 60 years or an intervention period of ≥ 3 months. Our findings did not support this conclusion. Because of the different study types included, the same subgroup analysis could not be performed. The previous meta-analysis on LBP showed that LBP significantly reduced LDL-C concentrations.^[30] However, the results of our meta-analysis could not determine the effect of LBP supplementation on the LDL-C concentration. Three studies reported the results of LDL-C, and significant heterogeneity among these studies was observed. The source of this heterogeneity could not be identified in the present study.

Our study has several limitations. First, the sample size of available RCTs was small. The study population was mainly focused on healthy volunteers, and few clinical trials of L. barbarum supplementation have been conducted in subjects with diabetes, metabolic syndrome, or dyslipidemia. Second, the included study population did not involve an adequate sample of young adults and consisted mainly of middle-aged and older adults. Third, given the inconsistent dietary environment around the world and the fact that most of the studies were conducted in China, with a small number of trials conducted in studies in Singapore and Brazil, it is challenging to generalize the findings to other regions. Fourth, the randomized controlled trials of lipid indicators as primary outcome indicators were lacking. Finally, further studies are needed to elucidate the mechanistic pathways underlying the effects of L. barbarum supplementation on blood lipids concentrations in adults.

5. Conclusions

In conclusion, this meta-analysis provides potential evidence that L. barbarum supplementation can significantly reduce TG and increase HDL-C concentrations in adults and L. barbarum fruit has an even greater effect on TG and HDL-C concentrations. This meta-analysis provides basic knowledge for the management of dyslipidemia. Given the small number and low quality of studies examining the effects of LBP on lipid profiles in adults, further high-quality studies are still needed to confirm our findings.

Acknowledgments

We thank Emily Woodhouse, PhD, from Liwen Bianji (Edanz) (www.liwenbianji.cn) for editing the English text of a draft of this manuscript. We thank Yang Wei, PhD, Statisticians, from Graduate School of China Academy of Chinese Medical Science (http://www.yjstcm.ac.cn) for reviewing the statistical analysis of this manuscript.

Author contributions

Conceptualization: Xueyuan Zeng, Weimin Zhao, Jixiang Ren.

Data curation: Huazhong Xiong, Junliang Wu.

Formal analysis: Xueyuan Zeng.

Methodology: Xueyuan Zeng, Siming Wang.

Software: Xueyuan Zeng, Huazhong Xiong, Junliang Wu.

Supervision: Jixiang Ren.

Validation: Xueyuan Zeng, Huazhong Xiong.

Writing – original draft: Xueyuan Zeng.

Writing – review & editing: Weimin Zhao, Siming Wang, Jixiang Ren.