Effect of probiotic Lactobacillus on lipid profile: A systematic review and meta-analysis of randomized, controlled trials

Objective To assess the efficacy of probiotic Lactobacillus on serum lipids using a meta-analysis of randomized, controlled trials. Methods Fifteen studies containing 15 trials, with 976 subjects were included. The pooled WMD was calculated by random effects model. Results Probiotic Lactobacillus consumption significantly reduced TC by 0.26mmol/l (95% CI, -0.40 to -0.12) and LDL-C by 0.23mmol/l (95% CI, -0.36 to -0.10). Subgroup analysis of trials found significantly reduction of TC using L. plantarum and reduction of LDL-C using L. plantarum or L. reuteri. No significant effects were found on TG and HDL-C levels after supplementation with probiotic Lactobacillus. While, subgroup analysis found significantly beneficial effects on TG and HDL-C by consuming synbiotic food, containing L. sporogenes and inulin. Conclusion Consuming probiotic Lactobacillus, especially L. reuteri and L. plantarm, could reduce TC and LDL-C significantly. The study also suggested significantly beneficial effects on TG and HDL-C by consuming synbiotic food, containing L. sporogenes and inulin.


Introduction
As we know, coronary artery disease (CAD) is one of the most important causes of deaths all over the world. Epidemiological reports have demonstrated that the high risk of CAD is associated with dyslipidemia, especially the high level of low density lipoprotein-cholesterol (LDL-C) PLOS ONE | https://doi.org/10.1371/journal.pone.0178868 June 8, 2017 1 / 15 a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 [1]. One of the considerable first line for treating dyslipidemia is dietary regulations and recommendations; however, using these methods, only a modest amelioration can be achieved [2]. Probiotics are called as live microorganisms which are beneficial to human health when consumed in adequate quantities [3]. Specific strains of probiotic bacteria, such as Lactobacillus, have been demonstrated to improve anti-diarrheal and anti-inflammatory, potentiate immune, and delay diabetes [4]. Several Lactobacillus strains have been found to reduce total cholesterol (TC) and triglyceride (TG) concentrations in rats [5,6], while in human clinical studies there is no consensus on the effects of consumption of Lactobacillus strains on lipid profile. One previous study has shown that three weeks daily administration of 200ml milk containing Lactobacillus acidophilus L1 was associated with the reduction of TC level in hypercholesterolemic individuals [7]. Further, Fuentes et al. reported that daily intake of L. plantarumin a capsule containing 1.2×10 9 CFU lowered TC and LDL-C concentrations in participants with hypercholesterolemia after 12 weeks[8]. However, Hove et al. reported that 12 weeks intake of milk fermented with L. helveticus had no effect on serum lipids in type 2 diabetes patients [9]. As we know, there has been no meta-analysis on the efficacy of probiotic Lactobacillus for lipid control. In this paper, a meta-analysis was carried out to assess the functional effects of supplementation of probiotic Lactobacillus on lipid profile, which may provide further information on the efficacy of specific strains of Lactobacillus required to lipid control.

Literature search
Studies exploring the effects of probiotic Lactobacillus consumption on serum lipid were searched in PubMed, Web of Knowledge, Cochrane Library, and Embase databases. The search was last updated in July 2016 and involved only full-text articles published in English. the one whose follow-up period is the longest. There is no restriction on minimum number of interventions or controls in our meta-analysis. The study flow is showed as Fig 1. Wu YC and Zhang QQ conducted the search, data extraction, and analysis of study quality.
Risk of bias S1 Table shows the risk of bias of included articles, which was assessed by the Cochrane Collaboration's tool. Two studies clearly describe the method of randomization (13.3%, n = 2). The risk of bias regarding concealed allocation and selective reporting was unclear (100%, n = 15). Low risk of bias was revealed on incomplete outcome data (93.3%, n = 14), blinding of participants (100%, n = 15), and other sources of bias (93.3%, n = 14).

Statistical analysis
SD of the changes of serum lipid from baseline, if not reported, was calculated using the formula below (1). Statistical analysis was conducted using STATA 11.0 (Stata, College Station, TX, USA). The effect of Lactobacillus strains on serum lipid was showed as the weight mean difference (WMD) of serum lipid changes between the treatment groups and control groups, which was calculated as the difference in the mean outcome between the groups divided by SD of outcome among participants and analyzed by the random effects model regardless of heterogeneity. Heterogeneity across the included studies was assessed using the I 2 statistics, representing the percentage of actual variation in relation to total variation. Subgroup analysis was also performed. P < 0.05 was considered statistically significant. Additionally we assessed the probability of publication bias with Begg's funnel plots and the Egger's test, with p value < 0.10 considered representative of statistically significant publication bias. Fifteen studies, with 976 subjects in all, were included in this meta-analysis. All the included studies were randomized, controlled trials with double-blind design. Of the 15 studies, 13 were parallel and the other 2 were cross-over. Table 1 shows the basic characteristics of included trials. Of the 15 studies, 5 used subjects with hypercholesterolemia, 3 were healthy participants, 2 were diabetes mellitus patients, 1 included smokers, 2 included obese subjects, 1 included pregnant women and 1 included patients with gestational diabetes mellitus. Of the 15 studies, 13 studies used probiotics, including L. plantarum, L. acidophilus, L. fermentum, L. helveticus, L. salivarius, L. reuteri, and L. rhamnosus, and the other 2 studies used synbiotics containing L. sporogenes and inulin. The duration and dose of probiotic Lactobacillus consumption varied between trials. The total daily dose changed from 10 7 to 10 11 colony-forming units. The duration changed from 3 to 24 weeks. To be mentioned, different trials give the different definitions about hypercholesterolemia that we didn't make a subgroup analysis, so it is not quite necessary to give a new definition.

Main outcomes
Of 13 RCTs, a random-effects meta-analysis did not show the effect of supplementation with probiotic Lactobacillus on TG levels (WMD = -0.04; CI,-0.16 to 0.07; p = 0.481). The forest plot of the effect was presented in Fig 2. The included studies showed highly significant homogeneity (I 2 = 42.0%; p = 0.055).
A pooled analysis of 2 RCTs demonstrated a significant decrease in TG levels and increase in HDL-C levels after the use of the synbiotic food containing L. sporogenes and inulin   Table 2). The included studies showed highly significant homogeneity (I 2 = 22.0%, p = 0.258; I 2 = 0.0%, p = 0.460, respectively). Besides, subgroup analysis with the givensubjects of non-pregnancy, non-diabetes/obesityordiabetes/obesitywas performed in supplemental material (S2 Table). Most of the results remained similar to the main outcomes, except LDL-C in the diabetes/obesity subjects became nonsignificant.

Publication bias diagnostics
We further identify the potential publication biases of literatures by Egger's test and Begg's funnel plot. In all trials, the shapes of funnel plot indicated no obvious asymmetry (Figs 6-9). And Egger's test provided statistical evidence for the funnel plot symmetry. No significant publication bias was found in the trials (p = 0.218 for TG; p = 0.599 for TC; p = 0.141 for LDL-C; p = 0.785 for HDL-C).

Discussion
The effects of probiotics consumption on lipid control have attracted increasing interest recently [22,23]. A meta-analysis of 13 randomized controlled trials reported that probiotics consumption was able to decrease total-cholesterol and LDL-cholesterol levels effectively [24]. Previous studies show that the gut microbiome may play an important role in the variation in serum lipids, supporting the potential of therapies altering the gut microbiome to control triglycerides and high-density lipoproteins [25,26]. While, there are many different probiotic strains on the market. It is of practical interest to know the effect of specific strains rather than probiotics as 'a whole group'. So, we conducted this meta-analysis to explore the effects of probiotic Lactobacillus consumption on lipid profile.
This study demonstrated that consumption of Lactobacillus has beneficial effects on serum TC and LDL-C levels, while no obvious changes in serum HDL-C and TG levels. As we know, this is the first meta-analysis assessing the effect of probiotics Lactobacillus on lipid profile. The mechanisms of reducing cholesterol levels may be as following: reduction of the enterohepatic circulation of bile salts; assimilation of cholesterol in the gastrointestinal tract; production of propionic acid which could decrease blood lipids; conversion of cholesterol into coprostanol in the gut [27][28][29]. Furthermore, it appears that activation of FXRα, nuclear receptor of bile acids, can also improve triglyceride control in animal models [30]. Besides, activation of the bile acid cell membrane receptor TGR5 may also activate thyroid hormone in brown adipose tissue and muscle, which increases energy expenditure and therefore improves serum lipids [31,32].
The present meta-analysis found that consumption of L. reuteri and L. plantarum could decrease the TC and LDL-C levels effectively. This is in accordance with the findings by Singh TP et al. [33]who reported that the values for TC and LDL-C of pigs were reduced significantly in group fed with L. reuteri LR6, and also in line with findings by Salaj R et al. [34]who showed that Lactobacillus plantarum LS/07 reduced TC and LDL in rats fed with a high fat diet. Few studies have explored the mechanisms of decreasing TC and LDL-C levels by L. reuteri and L. plantarum. Yamaoka-Tojo M et al. [35] provided possible mechanism that bile acid-binding resins decrease the concentration of bile acids returned to the liver via enterohepatic cycling and therefore stimulate the conversion of cholesterol to bile acids.   Moreover, in the present study, consumption of the synbiotic food, containing L. sporogenes and inulin, significantly decreased serum TG and increased serum HDL-C levels, while had no impact on serum TC and LDL-C compared with control. Similar findings have also been shown in hypercholesterolemic pigs fed with a synbiotic food containing L. acidophilus, fructooligosaccharide, inulin and mannitol for 8 weeks [36]. The mechanisms of decreasing TG and increasing HDL-C levels by synbiotics remained unclear. One of the possible mechanisms  Probiotic Lactobacillus on lipid profile was inhibition of synthesis of fatty acids in the liver by producing short-chain fatty acid (SCFA) [37]. Next, inulin played a determinant role in decreasing expression of lipogenic enzymes [38]. Moreover, inulin and probiotics may have a synergistic effect when administered as a synbiotic [39]. Besides, we also found that consumption of L. helveticus significantly decreased serum HDL-C levels, reminding us that not all Lactobacillus are beneficial to health. There are some limitations to this meta-analysis. Primarily, the bias of the included studies exists in some aspects such as the dietary restrictions and exercise. And crossover studies may bring additional biases. Besides, some studies had a small number of participants and a quite short duration of Lactobacillus consumption. Last, the insufficient databases were searched, and it was absence of papers in other languages and pre-published protocol. Further researches with larger sample and longer duration are required to verify the effect of probiotic Lactobacillus on lipid profile.

Conclusion
This meta-analysis showed that consumption of probiotic Lactobacillus, especially L. reuteri and L. plantarm, could reduce TC and LDL-C significantly. The study also suggested significantly beneficial effects on TG and HDL-C by consuming synbiotic food, containing L. sporogenes and inulin. The effect of probiotic Lactobacillus on serum lipids, as well as their mechanisms involved, required further investigation. This study provides useful information on the effects of probiotic Lactobacillus on serum lipids, which could make a contribution to the application of probiotic Lactobacillus, especially L. reuteri, L. plantarm and synbiotics, as novel therapies in lipids control.