Probiotic and synbiotic supplements for osteoporosis: a meta-analysis of randomized trials

Abstract

Background

Probiotic/synbiotic, have been proposed as a supplement to improve bone health markers. However, findings evidence remains inconsistent. This meta-analysis aimed to assess the effects of Probiotic/synbiotic supplementation on osteoporosis.

Methods

A systematic search was carried out in PubMed, Embase, Web of Science, and the Cochrane Library up to June 2025 to evaluate the efficacy of probiotic or synbiotic supplementation for osteoporosis. Data were pooled using a random-effects model to calculate standardized mean differences (SMD) and 95% Confidence intervals (CIs).

Results

Ten studies were included in this meta-analysis. Probiotic/synbiotic supplementation significantly decreased PTH levels (SMD = -1.19; 95% CI: -2.13, -0.26; P = 0.012), and increased lumbar spine BMD (SMD = 0.86; 95% CI: 0.01, 1.71; P = 0.047). However, probiotic/synbiotic supplementation did not significantly affect BMD of the total hip (SMD = 0.64; 95% CI: -0.56, 1.83; P = 0.296), C-terminal telopeptide (CTX) (SMD = -0.37; 95% CI: -0.75, 0.02; P = 0.063), osteocalcin (SMD = 0.24; 95% CI: -0.34, 0.82; P = 0.415), alkaline phosphatase (ALP) (SMD = 0.60; 95% CI: -0.61, 1.81; P = 0.334), and osteoprotegerin (SMD = -1.04; 95% CI: -3.19, 1.10; P = 0.340). Begg’s test revealed no significant publication bias.

Conclusions

The current meta-analysis revealed that probiotic/synbiotic supplementation beneficial effect on PTH, and lumbar spine BMD. No favorable effect of probiotic/synbiotic supplementation on BMD of the total hip, CTX, osteocalcin, ALP, and osteoprotegerin was observed.

Supplementary Information

The online version contains supplementary material available at 10.1186/s41043-025-01136-2.

Introduction

Osteoporosis is a common bone disorder that leads to reduced bone mass, lower density, and damage to bone structure, leaving bones fragile and more prone to fractures [1]. The disease is primarily driven by increased osteoclast activity, which accelerates bone loss. In China, more and more people are affected by osteoporosis, which can lead to serious fractures. Catching and treating this early can make a big difference in preventing injuries [2]. There are several treatment options available, including calcium and vitamin D supplements, medications that help build bone, and others like hormone-based therapies, calcitonin, RANKL inhibitors, bisphosphonates, estrogen, and selective estrogen receptor modulators [3]. These treatments aim to support healthy bones. For older adults, managing osteoporosis can be more challenging because their bodies process these medications differently and age-related changes can affect treatment outcomes [4]. Serious side effects, such as jaw necrosis from bisphosphonates or issues with the breast and uterus from estrogen therapies, can be concerning enough to make patients consider stopping their treatment [5]. To help prevent and manage these problems, healthcare providers often recommend non-drug options like exercise, calcium, and vitamin D [6]. Despite these efforts, fractures remain quite common, and each patient’s response to treatment can be different. These ongoing challenges have motivated researchers to search for safer and more effective solutions.

Growing evidence from animal studies suggests that probiotics and synbiotics may help protect bone, though their value in clinical practice is still under debate [7, 8]. Probiotics affect gut microbiota in complex ways, and many strains do not persist long-term [9, 10]. Clinical results have been inconsistent, which makes some physicians hesitant to use them for osteoporosis [11]. This uncertainty highlights the need for systematic evaluation of clinical trials. Since then, several well-designed RCTs have reported new findings on probiotics and synbiotics, suggesting their role may be greater than previously thought. The present study updates the evidence through a new meta-analysis of RCTs to assess the efficacy and safety of probiotic and synbiotic supplementation in osteoporosis, providing an updated basis for clinical decision-making.

Methods

Literature search

This study followed the PRISMA 2020 guidelines [12]. A systematic search was carried out in PubMed, Embase, Web of Science, and the Cochrane Library up to June 2025 to evaluate the efficacy of probiotic or synbiotic supplementation for osteoporosis. Reference lists of included RCTs were also screened. Two investigators independently reviewed titles and abstracts to identify eligible studies, resolving disagreements through discussion. The whole search strategy is provided in Supplementary Table 1.

Inclusion and exclusion criteria

Studies were included if they involved patients with osteoporosis who were given probiotics or synbiotics. The studies compared these treatments to a placebo or standard care. The researchers looked at outcomes such as lumbar spine and total hip bone mineral density, as well as markers like alkaline phosphatase, osteocalcin, CTX, osteoprotegerin, parathyroid hormone, and any adverse events. Only RCTs were considered. We excluded protocols, unpublished work, non-original articles (abstracts, corrections, or correspondence), non-RCT designs, studies with insufficient data, and review papers.

Data abstraction

Two authors extracted data independently, with disagreements resolved by a third reviewer. Information collected included first author, publication year, study location, trial design, registration number, intervention details, control treatment, sample size, participant age and BMI, duration of intervention, lumbar spine and total hip BMD, biochemical markers (alkaline phosphatase, osteocalcin, CTX, osteoprotegerin, parathyroid hormone), and adverse events. When necessary, corresponding authors were contacted to obtain complete datasets.

Quality evaluation

The methodological quality of each RCT was evaluated using the Cochrane risk-of-bias tool (RoB2) [13]. Each was graded as low, high, or unclear risk. Trials with more low-risk ratings were considered higher quality. Assessments were performed independently by two reviewers, with disagreements resolved through discussion.

Statistical analysis

Continuous outcomes were reported using the standardized mean differences (SMD) and 95% confidence intervals (CI) [14]. We checked for heterogeneity using the chi-squared test (Cochran’s Q) and the I² statistic to evaluate the consistency across studies [15]. We considered the heterogeneity substantial if the P value was below 0.1 or the I² was greater than 50%. Pooled estimates (SMD) were calculated using a random-effects model. For outcomes reported in more than two studies, sensitivity analyses were performed [16]. If fewer than ten studies reported a given result, Begg’s test assessed publication bias [17] in Stata 15.1 (Stata Corp, College Station, Texas, USA), with P < 0.05 considered significant.

Results

Characteristics of included studies

A total of 625 citations were identified and reviewed. After reviewing titles and abstracts, 144 citations were identified as duplicates and, therefore, ineligible. Of the remaining 17 full-text records, seven articles were subsequently excluded. The final analysis included ten studies (Fig. 1) [18–27]. Participants aged 50 to 76 had body mass index values between 24 and 27 kg/m2. The most common probiotic species were Lactobacillus and Bifidobacterium. Four included studies were conducted in Europe, and six were conducted in Asia. Their publication years ranged from 2013 to 2023. Background information for each assessed study is comprehensively presented in Table 1.

Fig. 1.

PRISMA flow diagram

Table 1.

Characteristics of included studies

Study	Duration	Number of participants (Int/Con)	Mean of age (Int/Con)	BMI (Int/Con)	Intervention	Control
Lambert et al. 2013, Denmark [22]	12 Months	38/40	61/63	25/26	Isoflavones combined with probiotics	Water and brown food coloring
Nilsson et al. 2015, Sweden [19]	12 Months	34/36	76	25	L. reuteri	Placebo
Takimoto et al. 2015, Japan [18]	2015.12–2016.4.12.4	34/35	57/58	22	Bacillus subtilis	Placebo
Jafarnejad et al. 2017, Iran [23]	6 Months	20/21	58/57	25/23	L. casei, B longum, L. acidophilus, L. rhamnosus, L. bulgaricus, B. brevis, S.thermophilus	Corn starch
Morato-Martínez et al. 2015, Spain [21]	2015.1-7.1	33/32	50–60	NR	Dairy products rich in bioactive nutrients (calcium, vitamin D, vitamin K, vitamin C, zinc, magnesium, L-leucine and the probiotic Lactobacillus plantarum)	Eat one serving of the same product, but not fortified
Han et al. 2019, Korea [26]	6 Months	27/26	58/59	24/23	Lactobacillus	Placebo
Jansson et al. 2019, Sweden [20]	2016.4-8.4-11.11— 2017.3-10.19.3.19	116/118	59/58	24	L. paracasei, L. plantarum	Placebo
Desfita et al. 2021, Indonesia [27]	90 Days	17/18/20	56/65	27/25	Fermented soy milk honey from L. casei, fermented soy milk honey from L. plantarum	Soybean milk
Zhao et al. 2022, China [24]	3 Months	15/12	62/61	23	L. B animalis, calcium, calcitriol	Placebo material, calcium, calcitriol
Vanichanont et al. 2023, Thailand [25]	2023.3.1–9.30	20/20	62/64	23/24	L. reuteri, L.paracasei, L. rhamnosus, L. rhamnosus, L. animalis, L.longum lactobacillus. longum, Bacillus coagulans and inulin	Inulin

All the studies included were RCTs

Risk of bias assessment

Seven of the ten included RCTs were of high quality. Table 2 presents the risk of biased information.

Table 2.

Quality assessment

Studies	Randomization Process	Deviation from Intended Interventions	Selection of the Reported Result	Measurement of the Outcome	Missing Outcome Data	General risk of bias
Lambert et al. 2013, Denmark [22]	Low	Low	Low	Low	Low	Low
Nilsson et al. 2015, Sweden [19]	Low	Low	Low	Low	Low	Low
Takimoto et al. 2015, Japan [18]	Low	High	Low	Low	Low	High
Jafarnejad et al. 2017, Iran [23]	Low	Low	Low	Low	Low	Low
Morato-Martínez et al. 2015, Spain [21]	Low	High	Low	Low	Low	High
Han et al. 2019, Korea [26]	Low	Low	Low	Low	Low	Low
Jansson et al. [20], Sweden	Low	Low	Low	Low	Low	Low
Desfita et al. [27], Indonesia	Low	Some concern	Some concern	Some concern	Low	Low
Zhao et al. 2022, China [24]	Some concern	High	Some concern	Low	Low	High
Vanichanont et al. 2023, Thailand [25]	Low	Low	Low	Low	Low	Low

Effect of probiotic/synbiotic on BMD

The meta-analysis revealed that probiotic/synbiotic supplementation significantly increased lumbar spine BMD (SMD = 0.86; 95% CI: 0.01, 1.71; P = 0.047; I² = 92.5%, P-heterogeneity < 0.001) (Fig. 2), but had no significant effect on the BMD of the total hip (SMD = 0.64; 95% CI: −0.56, 1.83; P = 0.296; I² = 96.4%, P-heterogeneity < 0.001) (Fig. 2). Sensitivity analysis confirmed the robustness of our findings on glycemic indices. Publication bias was not demonstrated by the Begg’s test (P > 0.05).

Fig. 2.

Mean difference and 95% CIs presented in forest plot of the studies on the effects of probiotic/synbiotic on BMD

Effect of probiotic/synbiotic on CTX

Probiotic/synbiotic supplementation did not significantly decrease CTX, with a pooled SMD of −0.37 (95% CI: −0.75, 0.02; P = 0.063; Five RCTs) and with moderate heterogeneity (I² = 60.7%, P-heterogeneity = 0.038) (Fig. 3). Sensitivity analysis confirmed the overall outcomes. Begg’s test revealed no significant publication bias (P > 0.05).

Fig. 3.

Mean difference and 95% CIs presented in forest plot of the studies on the effects of probiotic/synbiotic on CTX

Effect of probiotic/synbiotic on osteocalcin

Results from six RCTs involving 287 patients showed that probiotics/synbiotics had no significant effect on changes in osteocalcin (SMD = 0.24; 95% CI: −0.34, 0.82; P = 0.415; I² = 82.5%, P-heterogeneity < 0.001) (Fig. 4). The overall results on osteocalcin were confirmed by sensitivity analysis. Begg’s test did not show Publication bias (P > 0.05).

Fig. 4.

Mean difference and 95% CIs presented in forest plot of the studies on the effects of probiotic/synbiotic on osteocalcin

Effect of probiotic/synbiotic on PTH, ALP, and osteoprotegerin

Probiotic/synbiotic supplementation significantly decreased PTH levels (SMD = −1.19; 95% CI: −2.13, −0.26; P = 0.012; I² = 84.6%, P-heterogeneity = 0.002) (Fig. 5). Overall, probiotic/synbiotic supplementation did not significantly improve ALP (SMD = 0.60; 95% CI: −0.61, 1.81; P = 0.334; I² = 88.1%, P-heterogeneity = 0.004), and osteoprotegerin (SMD = −1.04; 95% CI: −3.19, 1.10; P = 0.340; I² = 95.6%, P-heterogeneity < 0.001) (Fig. 5). The overall results of the lipid profile were confirmed by sensitivity analysis. Begg’s test did not show Publication bias (P > 0.05).

Fig. 5.

Mean difference and 95% CIs presented in forest plot of the studies on the effects of probiotic/synbiotic on PTH, ALP, and osteoprotegerin

Discussion

This meta-analysis examined the effects of probiotic and synbiotic supplementation on bone mineral density (BMD) and bone turnover markers. The combined data show a significant rise in lumbar spine BMD, giving us strong confidence in this finding. Interestingly, no similar benefit was seen in the total hip BMD. This site-specific difference is essential, it makes sense because the lumbar spine has more trabecular bone, which is more metabolically active and likely to respond quickly to changes that affect calcium absorption and bone turnover. The cortical bone at the hip seems less responsive to short-term metabolic changes.

The improvement seen in lumbar spine bone density seems to match with research suggesting that substances produced by gut bacteria, mainly short-chain fatty acids, help the body absorb calcium better and reduce overall systemic issues [28, 29]. These processes are thought to help create a more supportive environment within our bones, promoting better bone health. While research shows that gut microbiota can positively influence bone health, the effects on hip BMD can be inconsistent [30]. This variability is similar to what we sometimes see with nutritional supplements like vitamin D or calcium, which often work better in areas with more trabecular bone. Overall, these findings make us more confident that our gut bacteria play a meaningful role in maintaining healthy bones, suggesting it could be a promising focus for future treatments [29, 31].

Biochemical outcomes give us important insights, making it easier to understand the effects we’re seeing. Even though taking probiotics or synbiotics didn’t result in noticeable changes in markers like osteocalcin or CTXwhich reflect bone remodelingthe wide range of these markers and the relatively short duration of the study suggest that the lack of significant findings might just be due to not enough time passing, rather than an actual absence of effect [32]. On a positive note, we saw a consistent and encouraging decrease in parathyroid hormone (PTH) levels. This reduction could help explain the improvements we observed in spine bone mineral density [33].

Lower PTH levels might mean better calcium absorption or other positive changes in how the body maintains bone health [34]. This highlights the potential benefits of these supplements. However, markers like alkaline phosphatase (ALP) and osteoprotegerin didn’t show noticeable changes, which underscores just how complex the relationship between microbes and skeletal health really is [35]. These findings reinforce the importance of continued research to better understand these mechanisms and their implications.

Although half of the included RCTs were of high quality, there was considerable variability among the studies. This diversity probably stems from differences in probiotic strains, synbiotic formulations, participant characteristics, intervention durations, and baseline nutritional status. Despite these variations, sensitivity analyses confirmed that the main findings remained consistent, and there was no evidence of publication bias, reinforcing the reliability of these results.

The findings suggest that taking probiotics and synbiotics might help keep our bones healthy. Not only can these beneficial microbes improve the body’s ability to absorb minerals, but they may also reduce inflammation and support our immune system, both important for overall bone health [36]. Synbiotics might enhance these benefits by encouraging the growth of beneficial microbes through prebiotic substrates [37, 38]. However, it’s important to note that not all probiotic strains or combinations work equally well, which could explain the mixed results in different studies.

This study is quite reliable because it focused solely on high-quality RCTs. It took the time to carefully look at both skeletal and biochemical results, giving us a clear and thorough understanding. The researchers also carefully checked for any potential biases, which makes us feel more confident in the results. However, there are some limitations to consider: most of the trials included had small numbers of participants, the intervention periods were short, and there was a lot of variability between studies. Additionally, many studies used surrogate markers of bone metabolism, which makes it harder to draw conclusions about long-term outcomes like fractures.

Conclusion

Our findings suggest that probiotic and synbiotic supplementation may positively influence lumbar spine bone mineral density and help lower circulating PTH levels. While these results are promising, they do not yet show consistent effects on hip BMD or other markers of bone turnover. This early evidence offers some hope for microbiota-targeted strategies to support bone health, but it’s important to interpret these findings cautiously due to current limitations in our research. Larger and longer-term studies with standardized approaches are needed to confirm these benefits and better understand the mechanisms at play.

Author contributions

Designing this study: Zhiyuan Gong, Tianyu LiangPerformed this study: Cheng HeDrafted the article, Revised the article critically for important intellectual content, and Approved the version to be published: Zhiyuan Gong, Qi Hu, Zahra Sabagh, Tianyu Liang, Cheng He.

Funding

None.

Data availability

The original data used during the current study can be obtained by contacting the corresponding author.

Declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.