The Effect of Probiotic Supplementation on Glucolipid Metabolism in Patients with Type 2 Diabetes: A Systematic Review and Meta-Analysis

Purpose: Type 2 diabetes mellitus (T2DM) is a persistent metabolic condition with an unknown pathophysiology. Moreover, T2DM remains a serious health risk despite advances in medication and preventive care. Randomised controlled trials (RCTs) have provided evidence that probiotics may have positive effects on glucolipid metabolism. Therefore, we performed a meta-analysis of RCTs to measure the effect of probiotic therapy on glucolipid metabolism in patients with T2DM. Methods: With no constraints on the language used in the literature, Excerpta Medica Database, PubMed, the Cochrane Library, and the Web of Science were searched for pertinent RCTs published between the date of creation and 18 August 2022. Stringent inclusion and exclusion criteria were applied by two reviewers to independently examine the literature. The risk of bias associated with the inclusion of the original studies was assessed using the Cochrane risk-of-bias tool, and Stata 15.0 was used to perform the meta-analysis. Results: Thirty-seven publications containing a total of 2502 research participants were included in the meta-analysis. The results showed that after a probiotic intervention, the experimental group showed a significant decrease in body mass index (standardised mean difference (SMD) = −0.42, 95% confidence interval (CI) [−0.76, −0.08]), fasting glucose concentration (SMD = −0.73, 95% CI [−0.97, −0.48]), fasting insulin concentration (SMD = −0.67, 95% CI [−0.99, −0.36]), glycated haemoglobin concentration (SMD = −0.55, 95% CI [−0.75, −0.35]), Homeostatic Model Assessment for Insulin Resistance score (SMD = −0.88, 95% CI [−1.17, −0.59]), triglyceride concentration (SMD = −0.30, 95% CI [−0.43, −0.17]), total cholesterol concentration (SMD = −0.27, 95% CI [−0.43, −0.11]), and low-density lipoprotein concentration (SMD = −0.20, 95% CI [−0.37, −0.04]), and an increase in high-density lipoprotein concentration (SMD = 0.31, 95% CI [0.08, 0.54]). Moreover, subgroup analyses showed that patients with a longer intervention time, or those who were treated with multiple strains of probiotics, may benefit more than those with a shorter intervention time or those who were treated with a single probiotic strain, respectively. Conclusion: Probiotic supplementation improves glucolipid metabolism in patients with T2DM, offering an alternative approach for the treatment of these patients.


Introduction
It is estimated that more than 700 million people aged from 20 to 79 will have diabetes by 2045, with 90% of cases being type 2 diabetes mellitus (T2DM) [1]. Low-grade

Intervention Measures
Interventions: the test group received probiotics and the control group received a placebo.

Retrieval Strategies
We incorporated mesh phrases and free keywords into our search strategy, in accordance with PICOS philosophy. For RCTs, regarding the impact of probiotics on T2DM, a thorough search of Web of Science, PubMed, Cochrane Library, and Embase was conducted. The data were retrieved, and they were published between the databases' creation and 18 August 2022. The search strategy used a combination of subject terms and free words, with search terms including ("probiotic" OR "probiotics" OR "lactobacillus " OR " bifidobacterial ") AND ("diabetes" OR "glycemia" OR "glucose" OR "Fasting blood sugar" OR "FBS" OR " Glycosylated haemoglobin A1c" OR "HbA1c" OR "insulin" OR insulin resistance " OR "HOMA-IR" OR "cholesterol" OR "lipids" OR "Total cholesterol" OR "TC" OR "Triglyceride" OR "TG " OR "High density lipoprotein cholesterol" OR "HDL-C" OR "Low density lipoprotein cholesterol" OR "LDL-C", etc. There were no restrictions on language or region. The search strategy is detailed in Supplementary S1.
The flowchart in Figure 1 shows the outcomes of the study selection procedure. In the original and additional searches, the aforementioned databases yielded 11,747 studies. The retrieved literature was imported into EndNote X9 for management, after which 3334 duplicate publications were eliminated, 8085 irrelevant publications were eliminated after reading the titles and abstracts, and 291 publications that did not meet the inclusion criteria were eliminated after reading the full text. This process yielded 37 studies for inclusion in this meta-analysis.

Basic Characteristics of the Included Studies
The 37 studies had English as their primary language and included 2502 patients (1262 in experimental groups and 1240 in control groups). Table 1 displays the key characteristics of these studies.

Basic Characteristics of the Included Studies
The 37 studies had English as their primary language and included 2502 patients (1262 in experimental groups and 1240 in control groups). Table 1 displays the key characteristics of these studies.

Quality Assessment of the Included Studies
The included studies were all randomised, double-blind studies. Three of the included studies [34,35,40] did not mention whether random numbers were used for group allocation; two [34,35] did not mention whether allocation concealment was used; two [20,22] did not clarify whether blinding was used; eight [11,20,23,24,28,29,49,51] did not clarify whether there was measurement bias; and two [11,20] did not provide trial pre-registration or publication, therefore, it was not clear which outcome indicators should be reported. Moreover, five [20,26,29,34,35] studies had missing visit bias, and therefore, high reporting bias, but none of the other studies had reporting bias or any other biases. The results of the evaluation concerning the risk of bias in the included studies are shown in Figure 2.

Quality Assessment of the Included Studies
The included studies were all randomised, double-blind studies. Three of the included studies [34,35,40] did not mention whether random numbers were used for group allocation; two [34,35] did not mention whether allocation concealment was used; two [20,22] did not clarify whether blinding was used; eight [11,20,23,24,28,29,49,51] did not clarify whether there was measurement bias; and two [11,20] did not provide trial preregistration or publication, therefore, it was not clear which outcome indicators should be reported. Moreover, five [20,26,29,34,35] studies had missing visit bias, and therefore, high reporting bias, but none of the other studies had reporting bias or any other biases. The results of the evaluation concerning the risk of bias in the included studies are shown in Figure 2. The risk of bias graph shows the percentages assigned to each risk of bias item across all included studies. Each point is assigned one of three bias assessment criteria, denoted by the colours green, red, or yellow, respectively: "Low", "High", and "Unclear" [11,[17][18][19][20][21][22][23][24].
We also conducted subgroup analyses on two intervention parameters, the duration of probiotic treatment ( Figure 3a The risk of bias graph shows the percentages assigned to each risk of bias item across all included studies. Each point is assigned one of three bias assessment criteria, denoted by the colours green, red, or yellow, respectively: "Low", "High", and "Unclear" [11,[17][18][19][20][21][22][23][24].
We also conducted subgroup analyses on two intervention parameters, the duration of probiotic treatment ( Figure 3a In the subgroup analysis based on different strains, the experimental interventions were more effective than the control interventions in reducing BMI when the former comprised multi-strain probiotics (SMD = −0.62, 95% CI (−1.13, −0.11]), but not when they comprised single-strain probiotics (SMD = −0.21, 95% CI [−0.70, 0.27]). Therefore, it can be inferred that treatment with multi-strain probiotics is superior to treatment with single-strain probiotics in terms of improving BMI in patients with T2DM. . Therefore, it can be inferred that treatment with multi-strain probiotics is superior to treatment with single-strain probiotics in terms of improving BMI in patients with T2DM.

Effects of Probiotic Therapy on Fasting Blood Glucose Concentration
Thirty-seven studies [11,[17][18][19][20][21][22][23][24] reported the effect of a probiotic intervention on fasting glucose concentration in patients with T2DM, and they were included in this analysis. After summarization, 1262 cases were included in the experimental group and 1240 cases were included in the control group. After combining effect sizes with a randomeffects model (I 2 = 89.5%, p = 0.000), it was revealed that compared with the control interventions, the experimental interventions significantly lowered fasting blood glucose concentrations (SMD = −0.73, 95% CI [−0.97, −0.48]; Figure 4).

Effects of Probiotic Therapy on Fasting Blood Glucose Concentration
Thirty-seven studies [11,[17][18][19][20][21][22][23][24] reported the effect of a probiotic intervention on fasting glucose concentration in patients with T2DM, and they were included in this analysis. After summarization, 1262 cases were included in the experimental group and 1240 cases were included in the control group. After combining effect sizes with a randomeffects model (I 2 = 89.5%, p = 0.000), it was revealed that compared with the control interventions, the experimental interventions significantly lowered fasting blood glucose concentrations (SMD = −0.73, 95% CI [−0.97, −0.48]; Figure 4).
Next, we performed subgroup analyses based on the duration of probiotic treatment ( Figure 4a) and the strains used (Figure 4b). The subgroup analysis, based on the duration of probiotic treatment, showed that there were significant differences between probiotic treatment durations of less than or equal to 2 months (SMD = −0.52, 95% CI [−0.83, −0.21]),
We also conducted subgroup analyses based on treatment duration ( Figure 5a) and the probiotic strains used (Figure 5b). This revealed that intervention durations of less than or
We also conducted subgroup analyses based on the duration of probiotic treatment (Figure 7a) and the probiotic strains used (Figure 7b
We also conducted subgroup analyses based on the duration of probiotic treatment (Figure 7a) and the probiotic strains used (Figure 7b
conclusions that can be drawn. Third item.
The text continues here.
We also performed subgroup analyses based on the duration of probiotic treatment (Figure 9a)  All figures and tables should be cited in the main text as Figure 1, Table 1, etc.
We also performed subgroup analyses based on the duration of probiotic treatment (Figure 10a) and the probiotic strains used (Figure 10b
We also performed subgroup analyses based on the duration of probiotic treatment (Figure 11a) and the probiotic strains used (Figure 11b). An intervention duration of 2 to 3 months (SMD = 0.34, 95% CI [0.14, 0.54]) was effective in increasing HDL concentration, but those of less than or equal to 2 months (SMD = 0. 38

Sensitivity Analysis and Publication Bias Test
The included studies exhibited no sensitivity problems. However, Egger's test was used to examine the funnel plots, and it demonstrated that there was publication bias for the HOMA-IR, TC, and TG data (p = 0.006, p = 0.003, and p = 0.02, respectively; Table 2). We employed funnel plots to depict publication bias, and we used the trim-and-fill method to evaluate these in light of the publication bias that existed in the included studies. Including eight studies in the TC model resulted in a symmetrical funnel plot, and the total effect size was −0.430 (−0.583, −0.277). No additional studies were required to make the funnel plots symmetrical for the IR and TG models, and the overall effect sizes were consistent (Table 3). Funnel plots and sensitivity analyses are provided in Supplementary S2.

Discussion
As the gut microbiota is crucial for preserving the body's normal metabolism, it is now recognised as a significant 'hidden' organ in the human body [53]. The makeup of the gut microbiota varies significantly between individuals, and it is influenced by a several variables, including genetics, diet, lifestyle, and health status [54]. The composition of the gut microbiota has been linked to the occurrence and development of T2DM, and an imbalanced gut microbiota stimulates the body to produce cytokines that reduce insulin sensitivity and accelerate the onset of diabetes [55]. To control the balance of the gut microbiota and reduce insulin resistance in patients with T2DM, Zhai LX et al. have proposed probiotic administration [56]. Probiotics are microorganisms that change the gut microbiota of their host, and therefore, they have a variety of effects; for example, they can preserve the gut microbiota's structural balance, increase the body's antioxidant concentrations, and reduce intestinal inflammation [57]. Bianchi et al. found that probiotics can safely and efficiently change a patient's gut microbiota, and they advised patients with metabolic illnesses to consume them frequently [58]. Lipopolysaccharides, which enter the blood to cause inflammation, disrupt intestinal integrity and influence the body's glucose metabolism, primarily by increasing concentrations of glycated haemoglobin; they are much more prevalent in patients with T2DM than healthy individuals [59]. However, Kim YA et al. found that patients with T2DM who consume probiotics have significantly lower lipopolysaccharide concentrations, less endoplasmic reticulum stress, and improved insulin sensitivity [60].
In this meta-analysis, the effects of probiotics on glucolipid metabolism were assessed in 37 RCTs involving a total of 2502 participants with T2DM. Analysis of the combined data showed that after treatment, patients in the probiotic group had significantly lower fasting glucose and fasting insulin concentrations than those in the control group. This suggests that probiotics can assist patients with T2DM to control their blood sugar concentrations. Problems with gut microbiota cause patients with T2DM to have significantly lower glucagon-like peptide-1 concentrations than people without T2DM, which promotes gastric emptying in these patients and makes them feel hungry [61]. A hypoglycaemic effect can be produced via probiotic administration, as this promotes the production of glucagon-like peptide-1, thus inhibiting glucagon secretion and delaying gastric emptying [62]. This suggests that oral probiotics help to control glycated haemoglobin concentrations by promoting a glucagon-like peptide-1 release. More specifically, after treatment, glycated haemoglobin concentrations decreased to a significantly greater extent in the experimental group than in the control group. Additionally, the findings of this meta-analysis demonstrated that the change in HOMA-IR score after probiotic treatment was significantly greater in the experimental group than the control group. The HOMA-IR score reflects the equilibrium between hepatic glucose output and insulin secretion, allowing physicians to gauge the severity of a patient's insulin resistance. The abovementioned finding implies that in patients with T2DM, oral probiotics may reduce insulin resistance and improve insulin sensitivity. However, publication bias was present in the meta-analysis of the HOMA-IR score (p = 0.006), which may be due to the calibre of the studies included.
Obesity is an independent risk factor for T2DM. Moreover, there are differences in the composition of the gut microbiota between obese and healthy patients, and disorders of the gut microbiota aggravate obesity [63,64]. Obese patients have elevated concentrations of cytokines, interleukins, the tumour necrosis factor, and lipopolysaccharides, resulting in a chronic inflammatory state, which leads to metabolic disorders and obesity-related diseases [65]. Oral probiotics can increase the abundance of certain intestinal microbiota to therefore alleviate metabolic syndrome problems and improve immune function in obese adults [66]. The results of this meta-analysis showed that compared with the control group, the experimental group exhibited significant improvements in BMI and in TG, TC, HDL, and LDL concentrations. However, there was publication bias in the meta-analysis of the TG and TC concentrations (p = 0.006) which may be related to the quality of the studies included. TG and HDL cholesterol concentrations can be controlled by altering the composition of the gut microbiota [67]. This could occur via several possible mechanisms, such as a reduction in the enterohepatic circulation of bile salts, the assimilation of cholesterol in the gastrointestinal tract, and the conversion of cholesterol to faecal sterols in the intestine [68]. In addition, the activation of farnesol receptors may also modulate TG concentrations [69], thereby improving lipid concentrations [70]. Mechanisms of cholesterol removal via lactic acid bacteria have been proposed in previous studies. For example, probiotics were found to purify bile salts by lowering cholesterol concentrations, due to the absorption of cholesterol by bacterial cell membranes via bile salt hydrolases [71]. Another study found that a probiotic mixture could modulate apolipoprotein synthesis via a mechanism that may be mediated by peroxisome proliferator-activated receptor/farnesoid X receptor upregulation during enterohepatic circulation [72]. Lactobacillus rhamnosus GG and the Bifidobacterium subfamily may partially prevent hepatic steatosis and damage by regulating the activation of hepatic adenylate-activated protein kinase [73,74]. Supplementation with Bifidobacterium shortum B-3 inhibits the accumulation of epithelial fat and upregulates the expression of genes related to lipid metabolism and insulin sensitivity [6]. The mechanisms of these effects may be diverse, and they need to be investigated further in vivo.
In accordance with our subgroup analysis, patients treated with multi-strain probiotics experienced greater reductions in fasting blood glucose concentrations than those treated with a single strain of probiotics. Chapman et al. observed that multi-strain probiotics combinations offer more health benefits than any constituent strain alone [75], which is consistent with our findings. Synergistic interactions between probiotic strains may account for this effect, and they may be the result of functional groups interacting with one another to enhance the host's glycaemic parameters [6].
The following limitations apply to this meta-analysis. The probiotic doses varied significantly between the included studies, and some studies did not specify the doses used. Therefore, the effect of probiotic dose levels on glycaemic control in patients with T2DM requires confirmation in other studies. Guerrero-Bonmatty et al. showed that the combination of monacolin K (a statin) and L. plantarum strains was more effective in reducing LDL-C and TC levels in the treatment of hypercholesterolemia [76]. Therefore, in the future, some studies concerning probiotics in combination with statins, versus probiotics alone, could be considered. Moreover, general information about patients' ages and BMIs varied between the included studies, which may have increased the clinical heterogeneity of the sample used for this meta-analysis.
In future research, there are many new probiotics, in addition to the traditional ones, worth considering for use at this stage. Since probiotics are excellent carriers or delivery devices, they can be recombined to orally deliver to antidiabetic targets [77]. Although probiotics are usually defined as living microorganisms, it might be worth considering whether dead strains are as potent, or even more potent, than live strains [78]. Although most of the probiotics that are widely used at this stage are lactic acid bacteria, there are also some potentially novel probiotics such as Akkermansia muciniphila, which also have antidiabetic properties [79]. In the future, further experiments and clinical trials will inevitably be needed to identify and compare the effects of different probiotic strains and dosages. Furthermore, the preparation technique and viability of probiotics, as well as the duration and regimen of good treatments, are also important factors that influence their antidiabetic activity.

Conclusions
In conclusion, this meta-analysis shows that oral probiotics aid in the regulation of glucolipid metabolism in patients with T2DM, primarily indicated by a marked decrease in glucose metabolism and lipid metabolism following treatment. These findings suggest that probiotic supplementation can be utilised as a complementary therapy to help prevent T2DM. To confirm the ability of probiotics to support glycaemic, lipid, and blood pressure regulation, additional clinical investigations with various patient profiles, probiotic dosages, and intervention durations are required.