Comparisons of the effects of different flaxseed products consumption on lipid profiles, inflammatory cytokines and anthropometric indices in patients with dyslipidemia related diseases: systematic review and a dose–response meta-analysis of randomized controlled trials

Background Flaxseed is widely used as a functional food for its rich sources of linolenic acid, lignans and dietary fibers in the world. This systematic review and dose–response meta-analysis on randomized controlled trials (RCTs) is first to evaluate effects of different flaxseed products (whole flaxseed, oil and lignans) on lipid profiles, inflammatory and anthropometric parameters in patients with dyslipidemia related diseases. Methods Literature search was performed in PubMed, Embase, Cochrane Central, Scopus, and Web of Science from the inception dates to January, 2020. Weighted mean differences with the 95% confidence interval (CI) were pooled using fix or random-effects models. Results Thirty-one RCTs involving 1,698 participants were included. The present meta-analysis revealed that flaxseed consumption had an overall beneficial effect on serum TC, LDL-C, TG, apo B and IL-6 in patients with dyslipidemia related diseases, but not on apo A, HDL-C, hs-CRP, CRP and anthropometric indices. However, different flaxseed products showed obviously different effects. Whole flaxseed supplementation significantly reduced TC (− 11.85 mg/dl, 95% CI − 20.12 to − 3.57, P = 0.005), LDL-C (− 10.51 mg/dl, 95% CI − 14.96 to − 6.06, P < 0.001), TG (− 19.77 mg/dl, 95% CI − 33.61 to − 5.94, P = 0.005), apolipoprotein B (− 5.73 mg/dl, 95% CI − 7.53 to − 3.93, P < 0.001), TC/HDL-C (− 0.10, 95% CI − 0.19 to − 0.003, P = 0.044) and weight (− 0.40 kg, 95% CI − 0.76 to − 0.05, P = 0.027); Lignans supplementation significantly reduced TC (− 17.86 mg/dl, P = 0.004), LDL-C (− 15.47 mg/dl, P < 0.001) and TC/HDL-C (− 0.45, P = 0.04). Although flaxseed oil supplementation had no such lowering-effect on lipid, meta-analysis revealed its lowering-effect on IL-6 (− 0.35 pg/ml, P = 0.033) and hs-CRP (− 1.54 mg/l, P = 0.004). Subgroup analysis revealed that whole flaxseed decreased TC, LDL-C and TG levels irrespective of country and the intervention time prescribed, but was more pronounced when the dose of whole flaxseed was ≤ 30 g/day (TC: WMD − 13.61 mg/mL; LDL-C: WMD − 10.52 mg/mL; TG: WMD − 23.52 mg/mL), rather not a dose > 30 g/day. Moreover, a linear relationship between dose of whole flaxseed and absolute changes in C-reactive protein (P = 0.036) and a nonlinear relationship between with IL-6 (P < 0.001) were detected. Conclusions Flaxseed intervention suggested the positive effects on lipid profiles, inflammatory cytokines and anthropometric indices in patients with dyslipidemia related diseases. Of these, whole flaxseed and lignans play an important role in reducing blood lipid, while flaxseed oil mainly plays in anti-inflammatory. Lipid- and weight-lowering was significant when whole flaxseed was consumed at doses < 30 mg/d, for lipid status with mixed dyslipidemia and patients with BMI > 25. Graphic abstract Supplementary Information The online version contains supplementary material available at 10.1186/s12986-021-00619-3.


Introduction
Flaxseed (Linum usitatissimum L.) is widely used as a functional food in the world, because it is a rich source of linolenic acid, lignans (secoisolariciresinol diglucoside [SDG]), and soluble and both insoluble dietary fibers, which are related to cholesterol control [1,2]. Flaxseed lignans plays a role in antiatherosclerosis including antiinflammatory, antioxidant, potent angiogenic, and antiapoptotic properties [3,4]. Flaxseeds are also the best source of α-linolenic acid (ALA), which is associated with human health benefits [5]. The flaxseed fiber also plays roles on reducing the blood glucose and cholesterol levels by delaying and reducing their absorption from the intestine [6,7].
It has been observed that many lipid/lipoprotein abnormalities are prevalent in obesity and heart problems, such as atherosclerosis and coronary heart disease (CHD) [8], collectively termed as dyslipidemia; these dyslipidemias are often hyperlipidemia which is characterized by elevated serum total cholesterol (TC), triglyceride (TG), low-density lipoprotein cholesterol (LDL-C) and very LDL-C (VLDL-C), and decreased high-density lipoprotein cholesterol (HDL-C) levels. Hyperlipidemia has become one of the greatest risk factors contributing to CHD, and also is associated with lipid disorders, which are considered the cause of atherosclerotic cardiovascular disease [9]. Moreover, nonalcoholic fatty liver disease (NAFLD) also was regarded associated with dyslipidemia [10,11]. Abnormalities in the lipid profile, specifically hypertriglyceridemia and low levels of HDL-C have been shown to be a strong predisposing issue to dyslipidemia related diseases. Likely, chronic inflammation was represented anther important factor in the development of dyslipidemia related diseases, such as dyslipidemia, atherosclerosis, CVD, metabolic syndrome (MetSyn) and obesity [12].
In recent years, the health effect of flaxseed intervention has aroused great concern in China. There was also

Search strategy
A literature search was performed in PubMed, Embase, Cochrane Library, Scopus, and Web of Science (WOS) from the inception dates to January 2020. We systematically searched databases by three independent investigators (C.Y., H.X., and M.W.). The computer-based searches included the key words (flaxseed OR linseed OR "flaxseed oil" OR "flaxseed lignan") in combination with ("non-alcoholic fatty liver disease" OR obesity OR overweight OR dyslipidemia OR hyperlipidemia OR "coronary heart disease" OR atherosclerosis). Moreover, the reference or citation lists from the retrieved articles were checked to search for further relevant studies. The search was limited to studies in humans published in English.

Inclusion criteria and exclusion strategy
Original studies were included if they met the following criteria: (1) diseases reported were that dyslipidemia, overweight, obesity, MetSyn, NAFLD, and atherosclerosis, (2) randomized clinical trials with a parallel or cross-over design, (3) investigation to the effects of flaxseed products on lipid profiles (TC, LDL-C, HDL-C, TG, apo A and apo B), anthropometric indices (weight, BMI and waist circumference) and inflammatory cytokines (IL-6, hs-CRP, CRP and TNF-α), (4) above-mentioned indicators could be extracted from the report, or providing baseline and end-trial indicators concentrations in both flaxseed and control groups, and (5) having supplementation with flaxseed or its derivatives for at least 2 weeks.
Animal trials, cytological study and publications with insufficient data were excluded. Non-clinical studies, uncontrolled trials, and trials with insufficient data on lipid, anthropometric and inflammatory parameters in flaxseed and control groups were excluded from the meta-analysis.

Data extraction
Data extraction was conducted using a standardized data collection independently by two investigators (C.Y., X.Y.). Inconsistencies were resolved by discussion until a consensus was reached. Study characteristics [including authors, publication year, sample size, type of the intervention, control treatment, study duration, study design, and participant features (sex, age, BMI, blood lipid data and dyslipidemia state)] were extracted. If a study had some comparisons by different intervention time, different product-types or doses of intervention, we regarded these comparisons as multiple studies. There are three studies [19][20][21] with two dose comparisons, one study [22] with two intervention time comparisons, and one study [23] with two product types.

Quality assessment
Assessment of study quality was independently performed by two reviewers (Y.F.L. and D.F.X.) using the Cochrane Collaboration tool for assessing risk of bias in RCTs [24]. The Cochrane tool has seven domains: random sequence generation (selection bias), allocation sequence concealment (selection bias), blinding of participants and personnel (performance bias), blinding of outcome assessment (detection bias), incomplete outcome data (attrition bias), selective outcome reporting (reporting bias), and other potential sources of bias. Each item  [25].

Quantitative data synthesis
The estimate of principal effect was defined as the weighted mean difference (WMD) in lipid concentrations (net change in mg/dL), anthropometric indices (net change in kg, cm, or kg/cm 2 ) and inflammatory cytokines (net change in pg/mL or mg/L) between the subjects assigned to consume flaxseed or its derivatives and those assigned to the control regimens. Standard deviations (SDs) of the mean difference were calculated using correlation coefficient methods referenced in the Cochrane Handbook by the following formula: [24], assuming a correlation coefficient (R) equal to 0.5 [26]. If standard error of the mean (SE) was only reported in included articles, the following formula [SD = SEM × sqrt (n)] was used to estimate the SD, where n is the number of subjects. The net changes in lipid concentrations, anthropometric indices and inflammatory cytokines were calculated by subtracting the change value after control-intervention from that calculated after treatment-intervention.

Statistical analysis
The meta-analysis was performed by using STATA SE (StataCorp LP, College Station, TX, USA). Q test and I 2 index were used to assess the heterogeneity among the included studies. Substantial heterogeneity was indicated as P < 0.05 in the χ 2 test and an I 2 > 50%. The fixed-effect models and random-effects models were used to calculate the WMD and 95% confidence interval (CI) for each variable according to the level of heterogeneity. If significant heterogeneity was present, then a random-effect model was used. Meta-regression was implemented to examine characteristics of the studies that were hypothesized to influence the observed treatment effects. It was performed to assess the association between the overall estimate of effect sizes with potential moderator variables, including type of flaxseed, BMI categories, lipid status, study country, intervention time and study design (parallel or crossover).
Further subgroup analyses were performed to explore impacts of certain characteristics: type of flaxseed (whole flaxseed, flaxseed oil or flaxseed lignans), country (Asian or Westerner), interventional time (≤ 10 weeks or > 10 weeks), dose of flaxseed products (whole flaxseed: ≤ 30 g/d or > 30 g/d; flaxseed oil: ≤ 10 g/d or > 10 g/d), BMI categories (18.5-< 25 kg/m 2 , 25-< 30 kg/ m 2 , and ≥ 30 kg/m 2 ) and lipid status (hypercholesterolemia, hypertriglyceridemia, mixed dyslipidemia and non-dyslipidemia). Sensitivity analyses were completed to detect the robustness of the statistical results and analyze possible sources of heterogeneity by excluding studies on a one-by-one basis, followed by the exclusion of those with a high risk of bias. Egger's test was conducted to quantitatively explore the possible publication bias (significant level = 0.05). All tests were two-tailed and P < 0.05 indicated statistical significance.
There was high level of heterogeneity among the studies that explored effects of flaxseed products on serum TC (I 2 = 95.8%). Meta-regression was conducted to explore the heterogeneity, and found that the heterogeneity could be explained by type of flaxseed products (P = 0.01) (shown in Additional file 3: Table 1).

Flaxseed supplementation on LDL-C, HDL-C and TC/HDL-C
Concentrations of LDL-C were reported in 30 comparisons from 26 studies representing 1,559 participants. Meta-analysis suggested significant net change in LDL-C concentrations following supplementation with flaxseedcontaining products (WMD = − 6.92 mg/dL, 95% CI − 8.58, − 5.25; I 2 = 90.6%), in which whole flaxseed and lignan supplementation had significant lowering-effect on LDL-C concentrations, rather than flaxseed oil intervention (shown in Fig. 1b).
High level of heterogeneity was found among the studies that explored effects of flaxseed products on serum LDL-C (I 2 = 95.7%) and HDL-C (I 2 = 92.0%). Metaregression was conducted to explore the heterogeneity, and found that the heterogeneity of LDL-C related studies could be explained by factors that were lipid status (P = 0.013) and type of flaxseed products (P = 0.001), the heterogeneity of HDL-C related studies could be explained by intervention time (P = 0.036) (shown in Additional file 3: Table 1).

Flaxseed supplementation on TG
Twenty-nine comparisons from 24 studies reported the results for TG, which included 1,552 participants. The pooled result suggested that there was a significant net change in TG following supplementation with flaxseedcontaining products (WMD − 12.31 mg/dL, 95% CI − 23.04, − 1.58, P = 0.025; I 2 = 93.1%). Whole flaxseed supplementation had a significant effect on reducing TG concentration (WMD − 19.77 mg/dL, 95% CI − 33.61, − 5.94, P = 0.005; I 2 = 96.1%) compared to control group according to the meta-analysis for different type of flaxseed products (shown in Fig. 1d). There was high level of heterogeneity among the studies that explored effects of flaxseed products on serum TG (I 2 = 93.1%). Meta-regression was conducted to explore the heterogeneity, and found that the heterogeneity of TG related studies could be explained by lipid status (P = 0.044) and country factor (P = 0.041) (shown in Additional file 3: Table 1).

Flaxseed supplementation on Apo A, Apo B and Apo B/Apo A
Six comparisons from 5 studies reported the results for Apo A and Apo B, which referring to 440 participants. Results suggested that there was a significant reduction in Apo A and Apo B following supplementation with flaxseed-containing products (Apo A: − 4.14 mg/dL, 95% CI

Flaxseed supplementation on inflammatory cytokines in patients with dyslipidemia-related diseases Flaxseed supplementation on IL-6
The results for IL-6 were reported in 10 comparisons including 426 patients with dyslipidemia-related diseases. The pooled result showed a significant net change for IL-6 (WMD − 0.23 pg/mL, 95% CI − 0.45, − 0.01;   Fig. 3a).
There is moderate level of heterogeneity among the studies. The heterogeneity among the studies could be explained by study design (P = 0.006) in the meta-regression (shown in Additional file 3: Table 1).

Flaxseed supplementation on CRP and hs-CRP
The source of high level of heterogeneity was explained by country factor (P = 0.009) in the meta-regression of flaxseed intervention on CRP, but no factor could explain the heterogeneity in the meta-regression of flaxseed intervention on hs-CRP (shown in Additional file 3: Table 1).
There is moderate level of heterogeneity among the studies. The heterogeneity among the studies could not be explained by presupposed factors in the meta-regression (shown in Additional file 3: Table 1).

Flaxseed supplementation on anthropometric indices in patients with dyslipidemia-related diseases Flaxseed supplementation on weight and BMI
The results for weight were reported in 12 comparisons from 11 studies including 706 participants. The pooled result from these studies showed no significant effect of supplementation with flaxseed-containing products on weight change (WMD − 0.29 kg, 95% CI − 0.59 0.01; I 2 = 0.0%) (shown in Fig. 4a). But subgroup analysis revealed whole flaxseed supplementation significantly reduced the net weight change (WMD − 0.40 kg, 95% CI − 0.75, − 0.05; I 2 = 0.0%) compared with the control group, while flaxseed oil supplementation did not (P = 0.991).
Seventeen comparisons included 949 participants from 16 studies reported the results for net change on BMI value (shown in Fig. 4b). Meta-analysis revealed no significant effect of supplementation with flaxseed-containing products on net change of BMI (WMD − 0.15 kg/m 2 , 95% CI − 0.53, 0.24; I 2 = 67.4%), no matter whole flaxseed or flaxseed oil.
High level of heterogeneity was found among the studies that explored effects of flaxseed products on BMI (I 2 = 67.4%). The high level of heterogeneity among the studies was explained by lipid status (P = 0.001) and study country (P = 0.047) in the meta-regression (shown in Additional file 3: Table 1).

Flaxseed supplementation on waist circumference (WC) and waist-height ration (WHR)
Eleven comparisons from 10 studies reported the results for WC, which involved 621 participants. The pooled analysis suggested that there was no significant net change in WC following supplementation with flaxseedcontaining products (WMD − 0.60 cm, 95% CI − 1.21, 0.02; I 2 = 44.5%) (shown in Fig. 4c). However, subgroup analysis revealed consumption with flaxseed oil had a significant reduction in net change for WC (WMD − 1.61 cm, 95% CI − 2.69, − 0.53; I 2 = 50.3%) compared to control group.
Four comparisons included 200 participants from 3 studies reported the results for net change for WHR. Meta-analysis revealed no net change on WHR following supplementation with flaxseed-containing products (WMD 0.01, 95% CI − 0.01, 0.02; I 2 = 29.3%) (shown in Fig. 4d); But subgroup results revealed separate consumption with whole flaxseed had an increase on net change for WHR (WMD 0.02, 95% CI 0.00, 0.03; There is moderate level of heterogeneity among the studies that explored effects of flaxseed products on WC (I 2 = 44.5%). The moderate level of heterogeneity among the studies was explained by study country (P = 0.047) in the meta-regression (shown in Additional file 3: Table 1).

Subgroup analysis
Further subgroup analyses were performed to explore impacts of certain characteristics. As the effect of whole flaxseed, flaxseed oil and lignans on the patients with lipid metabolism disorder is different, authors made subgroup analysis after stratifying pooled results according different type of flaxseed intervention. As the subgroups with less than two studies are not considered to be comparable, we did not compare differences within a subgroup with less than 2 studies of net changes. Detailed results of subgroup analysis are described in the Additional file 1 and Additional file 4: Table 2.

Meta-regression analysis for whole flaxseed supplementation on lipid profiles, inflammatory cytokines and anthropometric indices
Meta-regression using the random-effects model was undertaken to investigate the potential association between a decrease in lipid profile, anthropometric indices, and inflammatory cytokines and dose of whole flaxseed powder (g/day). It indicated a linear relationship between dose and absolute changes in CRP (P = 0.036) (shown in Fig. 5 Table 3).

Curvilinear regression for whole flaxseed supplementation and lipid profile, inflammatory cytokines and anthropometric indices
Curvilinear regression was undertaken to investigate the potential association between a decrease in lipid profile, anthropometric indices, and inflammatory cytokines and dose of whole flaxseed powder (g/day). It indicated a nonlinear relationship between dose and absolute changes in IL-6 (P < 0.01; approximate Cubic formula: Y = − 0.897 + 0.109χ − 0.003χ 2 − 3.169χ 3 ) (shown in Fig. 6), but not in lipid profile, anthropometric indices, and other inflammatory cytokines (P > 0.05) (shown in Additional file 6: Table 4).

Sensitivity analyses
Findings regarding serum lipid, inflammatory and anthropometric parameters basically remained robust in the sensitivity analysis when excluding a trial carrying a high risk of bias [53].

Discussion
The present systematic review and meta-analysis revealed that flaxseed consumption had an overall beneficial effect on serum TC, LDL-C, TG, apo B and IL-6 in patients with dyslipidemia related diseases, but not on apo A, HDL-C, hs-CRP, CRP and anthropometric indices. Moreover, different flaxseed products (whole flaxseed, flaxseed oil and flaxseed lignan) had obviously different effects. Meta-analysis revealed a significant reduction on TC, LDL-C, TG, apo B, TC/HDL-C and weight loss in patients allocated to whole flaxseed; flaxseed lignan had a significant lowering effect on TC, LDL-C and TC/HDL-C; however, flaxseed oil showed a significant reduction on IL-6 and hs-CRP.
Our present results on TC, LDL-C and TG in patients with dyslipidemia related diseases were consistent with a meta-analysis [13] that studied on the whole population. Whole flaxseed was an essential source of high-quality protein and soluble fiber [54], which contained 41% fat, 20% protein, and 28% fiber [55]. Especially, dietary soluble of flaxseed has been reported repeatedly on its function of reducing blood lipid [56][57][58]. However, different bioactive components are present in different flaxseedderived products [59], our meta-analysis revealed flaxseed oil had no such protective effect compared with whole flaxseed or lignans. It could be an explanation for more significant effects of whole flaxseed than form of flaxseed oil, that whole flaxseed is a combination of multiple nutrients including not only α-linolenic acid (ALA), but also phytoestrogen, and lignans, together with highquality protein and soluble fiber which could result in synergistic interactions [60]. Another reason was that the cholesterol-lowering effects of flaxseed may be enhanced by baking, perhaps because of the effect of high temperature on the bioavailability of some of the flaxseed phytochemicals [30]. Additionally, flaxseeds are the best source of lignans (0.7-1.5%, secoisolariciresinol diglucoside [SDG]) [2,61]. Flaxseed SDG has antioxidant, antiinflammatory, and potent angiogenic and antiapoptotic properties, which plays a role in antiatherosclerosis [62]. Another possible explanation was that the effects of flaxseed oil may have been masked by the use of MUFA-or n-6 PUFA-enriched oils as the control regimen in these included studies [63]. Subgroups analysis after stratifying pooled results according different type of flaxseed intervention (as shown in Table 3). Subgroup analysis revealed that whole flaxseed had significant effects on the net reductions in TC, LDL-C, TG and apo B in patients with dyslipidemia and with abnormal weight. One of the plausible explanations for the findings is that patients with dyslipidemia were more sensitive to lipid-lowering products than healthy adults [64]. Moreover, apo (B)-lowering effects of whole flaxseed intervention were shown in present meta-analysis. Apolipoprotein B is an essential part of the very-low-density lipoprotein (VLDL) and low-density lipoprotein (LDL), which facilitates their binding to receptors. As previous studies observed, the changes in apo B are consistent with the decrease in LDL cholesterol [65,66]. Flaxseed is the richest plant source of SDG, which is one of the three major groups of phytoestrogens. The effects of estrogen and selective estrogen receptor modulators such as Tamoxifen on reduction in lipoprotein (a) have been explored in many experimental and human studies [67,68]. In this regard, phytoestrogens could increase bile acid secretion, enhance thyroid function, and modify hepatic metabolism to play a role on the cholesterol-lowering effect [69]. Interestingly, whole flaxseed decreased TC, LDL-C and TG levels irrespective of country and the intervention time prescribed, but was more pronounced when the dose of whole flaxseed was ≤ 30 g/day (TC: WMD − 13.61 mg/mL; LDL-C: WMD − 10.52 mg/mL; TG: WMD − 23.52 mg/mL), rather not a dose > 30 g/day. Our findings confirmed excessive dietary intake may not bring more health benefits. Consistently, one ounce (28 g) of flaxseed exceeds the Adequate Intake (AIs) for ALA, which is 1.1 g/day for women and 1.6 g/day for men, and 1.4 g/day and 1.3 g/ day during pregnancy and lactation, respectively [70][71][72].
We did not observe any significant effects of flaxseed or its derivatives on HDL cholesterol. These findings were completely consistent with that meta-analysis [13,63]. The results of most current, current studies still support this view [65,66,73,74]. Apo A1 is the primary protein component of HDL that plays a role in HDL metabolism [75]. Actually, the apo (A)-lowering effects of whole flaxseed intervention were revealed in our meta-analysis, and this result well explained that flaxseed intervention had no significant effect on HDL-C. However, HDL-C without increase seems be not a negative effect on patients with lipid metabolism. A recent register study of more than 1 million US veterans found a U-shaped relationship between HDL-C and total mortality with 50 mg/dL (1.25 mmol/L) as the nadir associated with the lowest mortality [76]. We carefully reviewed that the baseline HDL-C of the included population was basically maintained at about 1.25 mmol/L. Similar to us, a retrospective showed that very high concentrations of HDL-C might be not associated with decreased risk [77]. Secondly, several studies also have provided evidence that the vascular effects of HDL are variable and hardly correlate with HDL-C concentrations in the plasma. In patients with diabetes mellitus, coronary disease, chronic renal insufficiency, NAFLD, cardiovascular risk factors and disorders, the function of HDL is impaired [78,79].
Previous meta-analysis assessed the effects of flaxseed and flaxseed-derived products (flaxseed oil or lignans) on inflammatory cytokines, did not reach consistent conclusions [15,80,81]. The present meta-analysis indicated that flaxseed supplementation significantly decreased IL-6, rather not CRP, hs-CRP and TNF-α. Our meta-analysis was conducted in a specific population with dyslipidemia related diseases. As some studies suggested these patients with metabolic disorders may get more benefits from the intervention, for obese and unhealthy subjects tending to accompany with higher levels of inflammatory factors [82,83]; for instance, CRP is an indicator of general low-grade inflammation [84,85]. Flaxseed related active compounds, especially SDG and its metabolites, have shown anti-inflammatory and antioxidant activity, mainly through inhibition of lipid peroxidation. A study [52] revealed that flaxseed oil mainly plays roles in antioxidant activity to reduce oxidation of low-density lipoprotein cholesterol (LDL-C), rather than its concentrations. In accordance with this, the present study revealed flaxseed oil had the lowering-effect on IL-6 and hs-CRP, rather not lipid profiles. Flaxseed oil mainly contains α-linolenic acid (ALA), ALA competes with linoleic acid (LA) in their common metabolism pathways for producing their long-chain metabolites. It could decrease production of arachidonic acid (AA) and consequently decreased appearance of this fatty acid in tissues has been suggested to decrease proinflammatory eicosanoid, which may decrease inflammation [81]. Meta-regression revealed a linear relationship between dose of whole flaxseed consumption and absolute changes in CRP and a nonlinear relationship between with IL-6. In fact, interactions between diet, inflammation, and the microbiota should be considered [86]. Flaxseed is one of the richest sources of soluble fiber which has been shown to be fermented to short-chain fatty acid (SCFA) (acetate, propionate, and butyrate) by intestinal bacteria. SCFA, especially propionate, can plays as anti-inflammatory factors through interference in various inflammatory pathways [87].
Our meta-analysis found whole flaxseed consumption had a significant reduction on net change for weight (WMD = − 0.40 kg; P = 0.027; I 2 = 0.0%) compared to control group. Although we did not find its significant reduction on BMI value (P = 0.719), however, subgroup result suggested its significant net change in participants with dyslipidemia (n = 6, WMD = − 0.31, P = 0.007). These results suggest that flaxseed has a beneficial effect on anthropometric indices. Whole flaxseed rather than flaxseed-oil seems effective in weight and BMI reduction which is attributed to the fact that the flaxseed can control the energy intake and increase satiety by containing 28% fibers [88]. These findings also could explain by the subsequent increase in circulating ALA after flaxseed supplementation. The anti-obesity effects of eicosapentaenoic acid and docosahexaenoic acid have been shown in previous studies [89,90].

Strengths and limitations
The strengths of the current study were that it was the first to explore the effect of flaxseed on lipid profiles, inflammatory cytokines and anthropometric indices in patients with dyslipidemia related diseases, and to distinguish the different effects of different types of flaxseed products. In addition, authors made subgroups analysis after stratifying pooled results according different type of flaxseed intervention to find the differences between subgroups more persuasively. It is more noteworthy that this study found revealed a linear or nonlinear relationship between dose of whole flaxseed consumption and absolute changes in inflammatory factors. Although our meta-analysis provides stronger evidence on the function of flaxseed interventions, these findings must be cautioned to applicate for the reasons as following: (1) the most of SDs of the net change were estimated using correlation coefficient methods referenced in the Cochrane Handbook, rather than were exact value; (2) many characteristics that vary within studies could not remove to be factors of between-study heterogeneity, such as difference of the different background of the patients included, the varying control groups, or the uneven quality of the studies; (3) some relevant studies might be missing from our pooling because neither grey literature databases were used for identifying studies, nor non-English studies were considered; (4) influences of the other covariates could not be fully determined because of the lack of information of the quality of products and the amount of specific bioactive components in flaxseed and their bioavailability. Further studies are needed with large sample sizes, adequate durations, and solid study designs to investigate the effectiveness of flaxseed supplementation.

Conclusions
All in all, flaxseed intervention suggested the positive effects on lipid profiles, inflammatory cytokines and anthropometric indices in patients with dyslipidemia related diseases. Of these, whole flaxseed and lignans play an important role in reducing blood lipid, while flaxseed oil mainly plays in anti-inflammatory. Lipid-and weight-lowering was significant when whole flaxseed was consumed at doses < 30 mg/d, for lipid status with mixed dyslipidemia and patients with BMI > 25. Additionally, a linear association between whole flaxseed consumption and absolute changes in C-reactive protein and a nonlinear relationship between with IL-6 was observed.