The effect of probiotics supplementation on blood pressure: a systemic review and meta-analysis.

BACKGROUND: Fermented milk has over the last decade been intensively studied because of the putative antihypertensive effect. The aim of our study is to investigate the role of probiotics support therapy in blood pressure and, as a kind of convenient and economic drugs for prevention and auxiliary treatment of hypertension.
MATERIALS AND METHODS: We performed a systemic review and meta-analysis to examine the effect of probiotics consumption on blood pressure. Databases including MEDLINE, EMBASE, Clinical trials, CNKI and the Cochrane library were searched. Also, the grey literature and references were searched.
RESULTS: Twenty-three randomized controlled trials (RCTs) involving 2037 participants met the inclusion criteria and were included. Probiotic consumption significantly changed systolic blood pressure (SBP) by - 3.05 mmHg (95%CI: - 4.67, - 1.44; P < 0.001) and diastolic blood pressure (DBP) by - 1.51 mmHg (95%CI: - 2.38, - 0.65; P = 0.001). Subgroup analysis indicated that the benefit effect of probiotics supplementation in SBP was only observed in hypertension [weight mean difference (WMD) = - 3.31 mmHg, 95%CI: - 5.71, - 0.92; P = 0.007] or type 2 diabetes (WMD = -4.85 mmHg, 95%CI: - 9.28, - 0.42; P = 0.032) patients, and the decreased DBP level by probiotics supplementation was only observed in hypertension patients (WMD = -2.02 mmHg, 95%CI: - 3.68, - 0.36; P = 0.017).This effect could only last for a short-term time of 8 or 10 weeks, but not for a long-term time.
CONCLUSION: This meta-analysis found a moderate and statistically significant reduction for either SBP or DBP with probiotics supplement compared with controls. Thus, probiotics is a potential for the dietary treatment of hypertension.

Introduction

Hypertension is a risk factor for cardiovascular disease, which is becoming a worldwide health problem for human being [1, 2]. Recent studies have suggested the involvement of gut microbiot, rich in probiotics, has potential influence in the development of chronic diseases, such as inflammatory bowel disease, liver cirrhosis, arthritis and type 2 diabetes [3-6]. Moreover, previous studies found that dietary constituents and supplements, such as fermented milk, can improve blood pressure (BP) control. Probiotics has been intensively studied because of the putative antihypertensive effect. Clinical and experimental studies which were carried out in spontaneously hypertensive rats reported that, biologically active peptides which were derived from fermented milk, had a positive effect on lowering the blood pressure in hypertensive subjects [7-9]. Most of the studies examined the tripeptides isoleucine–proline–proline (IPP) and valine–proline–proline (VPP), which werereleased by fermentation of milk by lactic acid bacteria, showed angiotensin-converting enzyme (ACE)-inhibitor effect in vitro. ACE plays an important role in the regulation of blood pressure by converting angiotensin I into the vasoconstrictor angiotens in II, and inactivating the vasodilator bradykinin, thereby increasing the blood pressure. It has also been demonstrated that probiotics and their products can improve BP by improving endothelial dysfunction [10, 11], and reducing blood glucose level and insulin resistance [12, 13]. At present, whether probiotics supplementation can improve the blood pressure has increasingly attracted people’s attention. However, the exactly effect is still unclear. Prospective studies have reported conflicting results regarding the effect of probiotics on hypertension. One large randomized clinical trial (RCT), enrolled subjects with obesity and/or high blood pressure, do not support a causal role for probiotics in blood pressure regulation [14]. Whereas another study suggested that probiotic soymilk supplementation significantly, yet modestly, lowered blood pressure [15].In order to address this issue, we performed this meta-analysis of RCTs to explore the potential relationship between probiotics supplement and hypertension.

Method

Literature search

We conducted this meta-analysis of the current literature according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines [16].Two authors (Dan Qi and Xiaolu Nie) independently performed a comprehensive literature search with PubMed/EMBASE/MEDLINE/Clinical trials/CNKI and the Cochrane Database of Systematic Reviews from 1951 to July 2019. The following keywords were used in all fields as search strategy: (1) probiotics or lactobacill or bifidobacter or saccharomyces or enterococcus or streptococcus or yogurt or yoghourt or yoghurt or milk or yeast, (2) blood pressure or hypertension or hypertensive or blood pressure, (3) 1 and 2. The titles and abstracts of the selected articles were examined. Full-text articles were retrieved. We also searched the reference lists of included articles for additional studies. We also searched for gray literature using Google.

Inclusion and exclusion criteria

Studies were included if they met the following inclusion criteria: (1) study design: RCT; (2) study subject: adult patients who was aged over 18 years; (3) study intervention: probiotic products with live bacteria and described the type of probiotics which defined as live microorganisms that may have health benefits for the host if consumed in adequate amounts; (4) outcomes: provided the data regarding the relationship between probiotics and blood pressure. Animal test and review were not included. Studies that assessed the relationship between probiotics and blood pressure only in protocol or abstract form were not included. When the same study published in multiple publications, only the one with the most recent data was included.

Data extraction

The data extraction was independently performed by two reviewers (Dan Qi and Xiao-Lu Nie). The following information from each study was included: author, year of publication, original country, race, sample size, numbers of case and control, mean age, study population, study design, outcome, method of intervention, duration of follow-up, inclusion and exclusion criteria. The outcome measured with mean ± SD or MD with 95% confidence intervals (CIs). If there are discrepancies between reviewers, a joint reevaluation of the original article will be addressed.

Risk for Bias assessment

Two reviewers (Dan Qi and Jian-Jun Zhang) independently assessed the quality of each study according to the Cochrane risk of bias [17]: Random sequence generation, allocation concealment, blinding of participants and personnel, blinding of outcome assessment, incomplete outcome data, selective reporting and other bias. Each study was regarded as being low, unclear, or high risk of bias.

Data synthesis and analysis

The continuous variables were expressed as weight mean difference (WMD) and 95% CIs. Before the data were summarized, we first test the heterogeneity among the included studies using Q chi-square test [18], in which a P value < 0.10 or I2 > 50% was considered as significant heterogeneity [18].2 statistic was used to describe the percentage of the variability that attributed to heterogeneity across the studies rather than the chance. Studies with an 2 statistic of < 25%, ~ 50%, ~ 75%, ~ 100% are considered to have no, low, moderate, and high degree of heterogeneity, respectively [19]. When significant heterogeneity was identified, we used a random-effects model [20] to pool the data; otherwise, a fixed-effects model [21] was used. Moreover, we also performed sensitivity analysis, subgroup analysis, andmeta-regression to investigate the potential sources of heterogeneity. The assessment of publication bias was evaluated by using Egger [22] and Begger test [23]. A P value less than 0.05 was judged as statistically significant, except where specified. Data were analyzed using Stata version 12.0 (Stata Corporation, College Station, TX, USA).

Results

Search results

Our initial search yielded 2571 relevant publications, of which 1052 were excluded because of duplicate records, leaving 1519 publications for further review. Among these records, 1479 were deleted based on the title/abstract review. Then 40 publications were screened for full-text information, however, 17 of them excluded because of the following reasons: 15 articles were not RCTs, and 2 studies did not use probiotics as exposure. Finally, 23 articles [14, 15, 24–44] met the inclusion criteria and were included in the meta-analysis (Fig. 1).

Fig. 1

Eligibility of studies for inclusion in meta-analysis

Study characteristics

The main characteristics of the ten included RCTs are presented in Table 1. These articles were published between 1996 and 2018,with a total sample size of2037.Among these studies, four were conducted in Finland, four in Iran, three in the Japan, three in Denmark, two in Poland, two in Canada, and each one in Australia, America, Brazil, Korea, or Russia. The age of participants ranged from 18 to 86. All the included studies enrolled both male and female patients except one [44], which enrolled only female patients. Nine [24–26, 30–32, 36, 41, 43] of the included articles involved patients with high blood pressure, four [15, 28, 37, 39] recruited patients with type 2 diabetes, five [14, 27, 38, 42, 44] enrolled obese patients, four [29, 33, 34, 40] enrolled healthy patients, and the remaining one [35] enrolled hypercholesterolemic patients. Among the nine studies that involved hypertensive participants, patients in two of the studies received antihypertensive medicines, whereas in the remaining seven studies, they received only probiotic without antihypertensive drug therapy. The duration of intervention ranged from 3 weeks to 24 weeks. Lactobacillus was used as an intervention in most of the included studies, and results from most of the studied showed that probiotics did reduce 24-h ambulatory blood pressure or blood pressure. All studies provided estimates that were adjusted baseline systolic pressure and diastolic pressure. The normal BP was defined as SBP less than 140 mmHg and DBP less than 90 mmHg according to the definition of the world health Organization /International Society of Hypertension (WHO/ISH) Hypertension Guidelines from 1999.

Table 1

Baseline characteristics of patients in the trials included in the meta-analysis

Study	Country	No. of patients	Patients’ status	Intervention	Control	Age (range, y)	Duration
Ivey, KL [14]	Australia	156	Obese patients	L.acidophius La5	Placebo	> 55	6 weeks
Hariri, M [15]	Iran	40	Type 2 diabetes	L.plantarum	Placebo	25–65	8 weeks
Hata, Y [24]	Japan	30	Hypertension	L.helveticus	Placebo	40–86	8 weeks
Mizushima, S [25]	Japan	46	Hypertension	L.helveticus	Placebo	23–59	4 weeks
Aihara, K [26]	Japan	40	High-normal BP	L.helveticus	Placebo	52.8	4 weeks
	Japan	40	Mild hypertension	L.helveticus	Placebo	51.5	4 weeks
Agerholm-Larsen, L [27]	Denmark	70	Obesity	Enterococcus	Placebo	18–55	8 weeks
Hove, KD [28]	Denmark	41	Type 2 diabetes	L.helveticus card04	Placebo	40–70	12 weeks
Naruszewicz, M [29]	Poland	36	Healthy	L.plantrarum	Placebo	35–45	6 weeks
Seppo, L [30]	Finland	39	Hypertension	L.helveticus	Placebo	30.2–61.7	21 weeks
Tuomilehto, J [31]	Finland	40	Hypertension	L.helveticus	Placebo	51.3	10 weeks
Jauhiainen, T [32]	Finland	88	Hypertension	L.helveticus	Placebo	51	10 weeks
Chang, BJ [33]	Korea	101	Healthy	Streptococcus	Placebo	20–65	8 weeks
Savard, P [34]	Canada	58	Healthy	Yoptimal	Placebo	18–55	4 weeks
Jones, ML [35]	Canada	120	Hyperchalesterolemic	L.reuteri NCIMB	Placebo	18–74	6 weeks
Sharafedtinov, KK [36]	Russia	40	Hypertension	L.plantarum TENSIA	Placebo	30–69	3 weeks
Mahboobi, S [37]	Iran	55	Prediabetic	Lactobacillus	Placebo	25–66	8 weeks
Rabiei, S [38]	Iran	40	Obese patients	L.acidophius	Placebo	25–70	12 weeks
Bahmani, F [39]	Iran	81	Type 2 diabetes	L.acidophius	Placebo	–	8 weeks
Moller, CM [40]	America	105	Healthy	Bifidobacterium breve	Placebo	18–23	28 days
Usinger, L [41]	Denmark	59	Hypertension	L.helveticus card04	Placebo	54	8 weeks
Szulinska, M [42]	Poland	71	Obese postmenopausal women	lyophilisate powder	Placebo	56.38	12 weeks
Jauhiainen, T [43]	Finland	89	Hypertension	L.helveticus	Placebo	25–55	24 weeks
Barreto, FM [44]	Brazil	24	Obese postmenopausal women	L.plantarum	Placebo	63	90 days

Risk of bias assessment

The details of risk bias are summarized in Fig. 2. Overall, 8 of the included studies were regarded as being at low risk of bias [14, 27, 29, 33–35, 38, 40], 13 at unclear risk of bias [15, 24–26, 28, 30–32, 37, 39, 41–43], and 2 at high risk of bias [36, 44]. The reason for the studies being at high risk of bias was that, they did not perform the blind to outcome assessors, or other bias. The most common reason for studies being at unclear risk of bias was that they did not adequately describe the methods for random sequence generation, or allocation concealment, or blinding to participants.

Fig. 2

Risk of bias

SBP

All the included studie s[14, 15, 24–44] reported the data of SBP. Pooled estimate showed that probiotics supplementation significantly reduced SBP level as compared to controls (WMD = -3.05 mmHg, 95%CI: − 4.67, − 1.44; P < 0.001) (Fig. 3). The test for heterogeneity was significant (I2 = 91.1%, P < 0.001). Thus, we performed sensitivity analysis. When the trial with outlier was removed [38], the overall estimate did not change substantially (WMD = -2.81 mmHg, 95%CI: − 4.43, − 1.19; P = 0.001), but the heterogeneity was still present (I2 = 91.2%, P < 0.001). When we excluded the trial with small sample size [24], the pooled result changed a little (WMD = -2.50 mmHg, 95%CI: − 4.11, − 0.89; P < 0.001), but the heterogeneity did not disappear (I2 = 90.5%, P < 0.001). We further excluded a single study once at a time, but the overall estimate and heterogeneity did not alter substantially.

Fig. 3

Forest plot showing the effect of probiotics supplementation on systolic blood pressure

Subgroup analysis was conducted based on the patients’ disease. Results showed that probiotics supplementation significantly reduced the SBP level in patients with hypertension (WMD = -3.31 mmHg, 95%CI: − 5.71, − 0.92; P = 0.007) or type 2 diabetes (WMD = -4.85 mmHg, 95%CI: − 9.28, − 0.42; P = 0.032), but not in obese (WMD = -2.91 mmHg, 95%CI: − 6.74, 0.92; P = 0.14) or healthy (WMD = -0.74 mmHg, 95%CI: − 3.35, 1.87; P = 0.58) patients (Figs. 4, 5).

Fig. 4

Forest plot showing the subgroup analysis of probiotics supplementation on systolic blood pressure in hypertension patients

Fig. 5

Forest plot showing the subgroup analysis of probiotics supplementation on systolic blood pressure in obese, healthy, or type 2 diabetes patients

Subgroup analysis based on the treatment duration suggested that, probiotics supplementation was associated with significantly decreased SBP level at the 8 weeks (WMD = -5.00 mmHg, 95%CI: − 7.42, − 2.59; P < 0.001) and 10 weeks (WMD = -3.48 mmHg, 95%CI: − 5.43, − 1.52; P = 0.001), but not at 12 weeks (WMD = -3.93 mmHg, 95%CI: − 8.80, 0.93; P = 0.113) and 24 weeks (WMD = -2.85 mmHg, 95%CI: − 5.76, 0.06; P = 0.06).

DBP

All the included studies [14, 15, 24–44] reported the data of DBP. Pooled result suggested that, probiotics supplementation significantly reduced DBP level as compared to controls (WMD = -1.51 mmHg, 95%CI: − 2.38, − 0.65; P = 0.001) (Fig. 6). The test for heterogeneity was significant (I2 = 81.6%, P < 0.001). Sensitivity analysis was performed to explore the potential sources of heterogeneity. When we excluded the trial with outlier or small sample size, the overall estimate did not alter substantially, but the heterogeneity was still present (data not shown).

Fig. 6

Forest plot showing the effect of probiotics supplementation on diastolic blood pressure

Subgroup analysis based on patients’ disease showed that, the reduced DBP level by probiotics supplementation was observed in hypertension patients (WMD = -2.02 mmHg, 95%CI: − 3.68, − 0.36; P = 0.017), but not in healthy patients (WMD = -0.71 mmHg, 95%CI: − 2.18, 0.76; P = 0.342), or those with obese (WMD = -1.22 mmHg, 95%CI: − 3.24, 0.81; P = 0.238), type 2 diabetes (WMD = -1.71 mmHg, 95%CI: − 3.78, 0.36; P = 0.105).

Subgroup analysis based on treatment duration suggested that, the better effect of probiotics supplementation over controls in DBP was only observed at 8 weeks (WMD = -2.22 mmHg, 95%CI: − 4.01, − 0.43; P = 0.015), but not at 10 weeks (WMD = -0.51 mmHg, 95%CI: − 2.49, 1.46; P = 0.611), 12 weeks (WMD = -1.71 mmHg, 95%CI: − 4.51, 1.10; P = 0.233) or 24 weeks (WMD = -1.50 mmHg, 95%CI: − 3.30, 0.30; P = 0.103).

Meta-regression

We first conducted univariate meta-regression analyses for each of the following variables: duration of intervention, sample size, baseline disease status, study location, age, gender, use of antihypertensive drugs, obesity, drinking and use of antibiotics, and smoking. The results demonstrated that, there was no significant association of effect size with these variables for the SBP level (duration of intervention: t = − 0.34, 95%CI: − 2.95, 2.10; P = 0.735; baseline disease status: t = − 0.80, 95%CI: − 2.14, 0.94; P = 0.431; study location: t = − 0.60, 95%CI: − 0.97, 0.53; P = 0.552; age: t = − 1.72, P = 0.096; gender: t = − 0.86, P = 0.397;antihypertensive drugs: t = 0.22, 95%CI: − 3.66, 4.54; P = 0.827; obesity: t = 0.11, 95%CI: − 3.31, 3.67; P = 0.92; drinking: t = − 0.27, 95%CI: − 4.76, 1.34; P = 0.65; use of antibiotics: t = − 3.64, 95%CI: − 4.76, 1.34; P = 0.84; smoking: t = 1.41, 95%CI: − 1.27, 6.91; P = 0.48), but sample size was associated with the treatment effect size of SBP level (t = 2.95, 95%CI: 1.02, 5.65; P = 0.006). This indicated that sample size was significant and independent predictor for heterogeneity.

With regard to the effect size of DBP level, meta-regression revealed a not significant association with these variables (duration of intervention: t = − 0.05, 95%CI: − 1.53, 1.46; P = 0.960; sample size: t = 1.33, 95%CI: − 0.52, 2.46; P = 0.195; baseline disease status: t = − 0.85, 95%CI: − 1.37, 0.56; P = 0.400; study location: t = − 0.07, 95%CI: − 0.48, 0.44; P = 0.946; age: t = 0.09, 95%CI: − 2.64, 0.37, P = 0.931; gender: t = − 0.48, 95%CI: − 1.53, 0.69, P = 0.633; antihypertensive drugs: t = − 0.11, 95%CI: − 2.56, 2.30; P = 0.913; obesity: t = − 0.17, 95%CI: − 2.29, 1.94; P = 0.87; drinking: t = 0.97, 95%CI: − 1.81, 5.08; P = 0.340; use of antibiotics: t = − 3.00, 95%CI: − 2.55, 0.48; P = 0.85; smoking: t = 0.69, 95%CI: − 1.71, 3.45; P = 0.50). This demonstrated that none of these variables was independent predictor for heterogeneity.

Publication bias

We assessed the publication bias by using Egger’s and Begg test, and results showed that no publication bias existed among the included studies (Egger’s test: t = − 1.86, P = 0.375; Begg test: Z = 0.82, P = 0.393).

Discussion

The present meta-analysis with 23 RCTs assessed the effects of probiotics supplementation on the blood pressure. Pooled results from these trials showed that, probiotics supplementation significantly reduced the SBP and DBP levels, as compared with controls. Moreover, the benefit effect of probiotics supplementation in SBP was only observed in hypertension or type 2 diabetic patients, and the decreased DBP level by probiotics supplementation was only observed in hypertension patients. This effect could only last for a short-term time of 8 or 10 weeks, but not for a long-term time. Our results indicated the effects of probiotics supplementation in controlling the blood pressure, especially in hypertension patients. The reduction reported by the current meta-analysis is modest. However, even a decrease in systolic blood pressure by 2 mmHg would reduce the risk for stroke and myocardial infarction by 4% [45].

Nutraceutical is a product isolated or purified from foods that is generally sold in medicinal forms not usually associated with food. It is demonstrated to have a physiological benefit or provide protection against chronic disease [46]. Several studies have reported the effects of nutraceutical in the clinical practice. These effects include the decreased cardiovascular disease morbidity and mortality, and ameliorating dyslipidaemia. The main mechanisms that explain the positive effect on the cardiovascular system are not well understood. However, researchers have found that carotenoids decreased the incidence and prevalence of cardiovascular events through their antioxidant action on free radicals or by acting as anti-inflammatory molecules [47, 48]. Physicians have attempted to identify the mechanisms responsible for nutraceuticals actions, the exact pathophysiological mechanism still remains uncertain. Chen et al. [49] hypothesized the role of resveratrol in counteracting hypercholesterolaemia. They fed mice a hypercholesterolaemic diet and resveratrol (200 mg/kg/day) for 8 weeks, and found a reduction in main serum lipid parameters [49].

Hypertension is associated with a myriad of major cardiovascular disease as well as mortality, and is becoming a worldwide health problem. A recent meta-analysis suggested that blood pressure lowering in early ischemic stroke had a neutral effect on the prevention of death or dependency [50]. Evidence for Cardiovascular Prevention from Observational Cohorts in Japan showed SBP was positively associated with ischemic stroke and intraparenchymal hemorrhage death [51].

Many experimental and clinical observations have involved that the effect of intestinal macrobiotics on cardiovascular disease including ST-elevation myocardial infarction [52-54]. A recent meta-analysis suggested probiotics could significantly reduce the value of SBP (WMD = − 5.04 mmHg) and DBP (Standard mean difference = − 0.39 mmHg) [55]. Another systemic review and meta-analysis suggested that consuming probiotics may improve BP by a modest degree, with a potentially greater effect when baseline BP is elevated, multiple species of probiotics are consumed, the duration of intervention is ≥8 weeks, or daily consumption dose is ≥10 colony-forming units [56]. In our meta-analysis, compared with control, probiotics resulted in reduction on SBP (− 3.05 mmHg, 95% CI, − 4.67, − 1.44; P < 0.001) and DBP (− 1.51, 95%CI: − 2.38, − 0.65; P = 0.001). Another important finding of this meta-analysis was the differentiation in the effect of probiotics on BP based on baseline BP level. Subgroup analysis of those studies which enrolled hypertension patients showed a meaningful reduction on either SBP or DBP, but no significant reduction on non-hypertensive population.

Fermented milk has over the last decade been intensively studied because of the putative antihypertensive effect. Miguel et al. found the antihypertensive effect of peptides from Enterococcus faecalis-fermented milk in rats [8]. The milk-derived IPP and VPP lowered blood pressure and increased plasma renin activity (PRA) in spontaneously hypertensive rats after long-term oral intake [57, 58]. It has been suggested that the mechanism of the antihypertensive effect of probiotics may be the inhibition of the ACE by IPP and VPP [30, 59]. Jauhiainen et al. suggested that Lactobacillus helveticus LBK-16H fermented milk, in daily use, does have a BP-lowering effect in hypertensive subjects and is a potential for the dietary treatment of hypertension. The elevator of C-reactive protein levels in L. helveticus group slightly indicated that the important mechanism or parallel phenomenon for the development of hypertension might at least, in part, be the systemic inflammation [32]. In addition to these suggested mechanisms, the fermented milk could also influence the positive effect on arterial stiffness and arterial stiffness is an independent predictor of cardiovascular morbidity and mortality and has been associated with hypertension [43].

Another contributing factor of the antihypertensive effect of probiotics supplementation might be the minerals; however, it does not explain the whole difference between probiotics supplementation and controls. In a recent meta-analysis of clinical trials, results showed that calcium supplementation (1000 to 2000 mg/d) significantly reduced the SBP by 1.44 mmHg and DBP by 0.84 mmHg [60]. Another meta-analysis of 33 controlled clinical studies suggested that, potassium supplementation (about 2.9 g/d) significantly decreased SBP by 3.11 mmHg and DBP by 1.97 mmHg [61]. The probiotics supplementation contained somewhat more calcium, potassium, and magnesium than the control product. Therefore, probiotics supplementation showed effect in decreasing the blood pressure, and had greater effect in SBP than in DBP.

In the present study, we found that the SBP reduction was greater than DBP, and this was in consistent with the findings of previously published studies [24, 62, 63]. Aihara K, et al. [26] reported that the magnitude of the treatment reduction in SBP and DBP was very similar for both the treatment group (5 mmHg from baseline), and the authors contributed this to the relatively short treatment period. Similarly, in the DASH trial, the net reduction of SBP was smaller than DBP at 1 week of treatment [64]. If the treatment period in these two trials had been longer, the reduction in SBP may have been greater than that in DBP. It is worth noting that, some trials focusing on the effect of calcium supplementation on blood pressure found greater DBP reduction than SBP [65-67]. However, some authors thought the effect of calcium supplementation was effective in those with low serum calcium and high parathyroid hormone levels, caused by high sodium intake and subsequent volume expansion in sodium, sensitive, low rennin hypertensive [68]. Therefore, the difference between SBP and DBP might be attributed to the calcium level in the probiotics supplementation.

There were several potential limitations in the present study. First, significant heterogeneity was identified among the included trials. However, we should be surprising given the various differences in the study design, sample size, treatment duration, patients’ baseline characteristics. These factors might have an impact on the treatment effect, and account for the heterogeneity. Second, some of the included studies have relatively small sample size, which had lower statistical power to test the effect differences. Compared with larger trials, studies with small sample size were more likely to overestimate the treatment effect. Third, in this meta-analysis, we only included studies published with English or Chinese language, which might result in language bias.

Conclusion

This meta-analysis found a moderate and statistically significant reduction for either SBP or DBP with probiotics supplement as compared with controls. Thus, probiotics supplement should be used as an antihypertensive agent. Considering the potential limitations in this study, more larger-scale, long-time RCT are needed to confirm the accurate effect of probiotics on blood pressure.

Additional file 1:.