Parkinson’s disease (PD), a neurodegenerative illness, is increasingly prevalent across the globe, severely affecting patients’ lives and causing huge economic strains on society. The primary hallmark of PD is the gradual loss of dopaminergic neurons (DNs) in the substantia nigra (SN). This neuronal death leads to classical motor symptoms, including resting tremors, bradykinesia, and muscle rigidity. Various non-motor symptoms affect PD patients, including anxiety, depression, constipation, sleep disturbances, and hyposmia. These non-motor symptoms often occur years before motor symptoms and severely impact PD patients’ daily functioning and quality of life (Lang, 2011). As the global population continues to age, the number of PD patients is projected to exceed 12 million by 2050 (Rocca, 2018). Although the exact causes of PD remain unclear, genetic susceptibility, environmental exposure, oxidative stress (OS), neuroinflammation, and gut microbiota dysbiosis (GMD) are believed to be strongly linked to the pathogenesis of PD (Simon et al., 2020).

In recent years, there has been growing attention on the influence of gut microbiota (GM) on the development and progression of PD. According to the gut-brain axis theory, GM communicates bidirectionally with the central nervous system through neural, endocrine, immune, and metabolic pathways (Cryan et al., 2019).Alterations in the abundance of certain bacterial taxa, including a common observation of reduced levels of bacteria like lactobacilli and bifidobacteria (often associated with health benefits in some contexts) and an increase in others such as Enterobacteriaceae and Prevotella species (often linked to pro-inflammatory conditions). However, it is crucial to note that these classifications are not absolute in the context of PD (Hill-Burns et al., 2017). This dysbiosis can lead to PD progression via gut barrier disruption, chronic inflammation, and the disruption of the central nervous system resulting from the imbalanced metabolism of short-chain fatty acids (SCFAs) (Poewe et al., 2017). Holmqvist et al. provided direct evidence in rats that α-synuclein, a key pathological protein in PD, can be transmitted from the gut to the brain via the vagus nerve, suggesting a potential role of the gut in the initiation and progression of PD pathology (Holmqvist et al., 2014). Increasing evidence indicates that neurochemicals and metabolites, which may directly or indirectly impact the body’s physiological processes and the onset and progression of PD, can be synthesized and regulated by GM (Jameson et al., 2020). GM and metabolites in PD patients differ from those of healthy individuals, with more pro-inflammatory cytokines and fewer anti-inflammatory cytokines (Nuzum et al., 2020). A study suggested that GM may correlate with the severity of PD symptoms (Scheperjans et al., 2015).

GM-targeted therapies have become a research focus in recent years. Probiotics, prebiotics, and fecal microbiota transplantation (FMT) are considered to be able to improve PD symptoms. Animal studies showed that FMT can alleviate motor symptoms and reduce neuroinflammation and neuronal loss in PD mouse models (Sun et al., 2018). In PD mice, FMT can mitigate GM alterations and reduce inflammation by activating microglial and astrocytic cells in the SN. FMT may work by modulating GM balance, lowering OS, enhancing anti-inflammatory responses, and improving gut-brain axis function (Tamtaji et al., 2019). Nurrahma et al. (2021) demonstrated that probiotic supplementation in PD mice significantly improved their motor function and increased antioxidant enzyme activities (Tsao et al., 2021). Tan et al. found that in PD patients, probiotics can effectively alleviate such non-motor symptoms as constipation and anxiety (Tan et al., 2021). However, whether GM-targeted treatments can effectively improve PD symptoms and their safety remains unestablished. This meta-analysis seeks to appraise the efficacy and safety of GM-targeted therapies in PD, providing references for clinical trials and theoretical support for future applications of these therapies in treating PD.

Clinical trials revealed that probiotics can alleviate gastrointestinal issues in PD patients (Tan et al., 2021). A study by Cryan et al. suggested that probiotics may reduce neuroinflammation induced by GM and promote the survival of DNs (Cryan et al., 2019). Additionally, an RCT by Tamtaji et al. (2019) demonstrated that probiotics containing bifidobacteria and lactobacilli significantly reduced MDS-UPDRS scores in PD patients, indicating their potential to improve motor symptoms. Probiotics can alleviate both motor and non-motor symptoms by regulating the vagus nerve and gut hormone secretion, promoting the synthesis of neurotransmitters such as serotonin and dopamine (Zhou et al., 2019).

Furthermore, probiotics effectively alleviate such non-motor symptoms as constipation and anxiety (Tamtaji et al., 2019) by balancing GM, reducing intestinal inflammation, and improving intestinal barrier function (Mertsalmi et al., 2017). A study suggested that probiotics can improve sleep quality in PD patients. Sun et al. (2018) found in an animal test that probiotics can modulate GM metabolism by increasing the production of SCFAs, thereby raising levels of sleep-related neurotransmitters, such as serotonin (Sun et al., 2018). Moreover, the immunoregulatory function of probiotics in multiple sclerosis (MS) has been confirmed. A study found that probiotics can improve MS pathology by decreasing pro-inflammatory cytokines, which include TNF-α and IL-1β (Tan et al., 2021).

This study appraised the efficacy and safety of GM-targeted therapies in PD, aiming to provide insights for clinical treatment. GM-targeted therapies can significantly improve MDS-UPDRS I, MDS-UPDRS III, NMSS, PDQ-39, BMs, use of laxatives, and MDA, but had no clear effect on MDS-UPDRS II, MDS-UPDRS IV, MMSE, MoCA, TAC, GSH, or GDS-15. Considering this factor, we degraded the corresponding evidence in the GRADE analysis to provide more objective findings. Wu et al. conducted a meta-analysis on the efficacy and safety of probiotics in treating mild cognitive impairment and Alzheimer’s disease. It revealed that probiotics outperformed the control group in improving the MMSE scores, with the difference between the two groups being statistically significant. The difference in the incidence of adverse events between the two groups was not statistically significant. Similarly, the subgroup analysis revealed no statistically significant difference in MMSE scores between the single-strain probiotics group and the control group, while the composite probiotics outperformed the control group. In this study, we not only analyzed MMSE and MoCA but also examined PD-related motor and non-motor symptoms, filling data gaps from previous studies. However, our study found no significant difference in the effects of probiotics-targeted therapies on MMSE, possibly due to variations in the number of literature. Numerous studies demonstrated no significant differences in the incidence and type of adverse events between the FMT group and the control group. FMT treatment typically lasts 24 to 48 hours. Common gastrointestinal symptoms include discomfort, bloating, diarrhea, constipation, and fever. These symptoms were mild and self-limiting, usually resolving spontaneously within two days (Paramsothy et al., 2017). Despite probiotics being generally considered safe, their long-term risks remain unclear and require validation in large-scale, long-term follow-up studies.

A study by Hill-Burns et al. (2017) demonstrated significant changes in GM composition in PD patients, with beneficial bacteria (such as lactobacilli and bifidobacteria) reduced and potentially pathogenic bacteria (such as Enterobacteriaceae and Prevotella) greatly increased (Hill-Burns et al., 2017). This dysbiosis may impact PD pathogenesis via two mechanisms. Firstly, the initiation of neuroinflammation contributes to PD pathogenesis. The key pathological hallmarks of PD are neuroinflammation and the loss of midbrain dopamine neurons in the nigrostriatal pathway. Lipopolysaccharide (LPS) is considered a key trigger for neuroinflammation (Keshavarzian et al., 2015). Studies indicated that LPS exacerbated neurodegeneration by activating microglia and releasing pro-inflammatory cytokines, which included TNF-α and IL-6. Sun et al. (2018) found substantially reduced levels of SCFAs in PD patients. This reduction may compromise gut barrier integrity, allowing toxic metabolites to pass through the blood-brain barrier and aggravate brain inflammation. An animal test conducted by Sun et al. (2018) showed that FMT could restore SCFAs levels and reduce inflammation, thereby greatly improving motor function (Sun et al., 2018). Secondly, gut barrier dysfunction is considered a potential pathogenic mechanism of PD. In PD patients, compromised gut barrier integrity may lead to the misfolding of the α-synuclein protein. The increased intestinal permeability is mainly caused by α-syn expression, the presence of E. coli, and elevated levels of lipopolysaccharide-binding protein (Forsyth et al., 2011). The supplementation of probiotics and prebiotics can help restore gut barrier function.

This study has several limitations. A total of 11 RCTs were included, with sample sizes ranging from 11 to 128 participants, which may cause small sample size effects, thereby limiting the generalizability of the findings. Future studies can increase sample sizes or conduct multi-center studies to improve quality. Significant heterogeneity was observed in this study for the following outcomes: change in MDS-UPDRS I, change in MDS-UPDRS II, change in MDS-UPDRS IV, change in NMSS, change in MoCA, change in PDQ-39, change in BMs, change in GSH, and change in MDA. Although substantial heterogeneity was observed in several analyses, the limited number of studies included for each outcome precluded reliable subgroup analyses (such as by intervention type, treatment duration, or disease severity) to explore sources of heterogeneity. With the publication of additional studies in the future, subgroup analyses or meta-regression may allow a more in-depth investigation of these sources. However, the findings were objectively interpreted by considering heterogeneity in the GRADE analysis. Publication bias was observed in changes in BMs, potentially due to the small sample size and regional bias. Most studies were conducted in Europe and America, with fewer from Asia and none from other regions such as North America, Africa, or South America. This may introduce regional bias and limit the applicability of the findings to populations with different diets or genetic backgrounds. Future research should encourage more multi-center collaborations on a global scale.

Since regional bias may contribute to publication bias, more multicenter and international studies involving greater sample sizes are required to reduce publication bias. Although most adverse events reported in this meta-analysis were mild (mainly gastrointestinal symptoms such as bloating and diarrhea) and were generally self-limiting, it should be noted that the majority of included studies had short follow-up periods and lacked long-term safety data. Caution is still warranted regarding prolonged use of probiotics, the long-term risks of FMT (such as potential infection transmission or immune effects), and complex metabolic impacts from certain strains. Future large-scale, long-term RCTs should prioritize the assessment of the long-term safety and tolerability of these interventions.

Furthermore, the functional consequences of these microbial shifts are complex and not fully understood. For instance, while some species of Lactobacillus are commonly used as probiotics and considered beneficial, certain strains have been shown to possess decarboxylase activity capable of metabolizing exogenous levodopa (Maini Rekdal et al., 2019). This may potentially reduce the bioavailability of the primary medication for PD, highlighting that the effects of gut microbiota on PD are not straightforward and may be strain-specific and context-dependent. Therefore, simplistic categorization of bacteria as “beneficial” or “pathogenic” in PD is misleading. Future research and therapies should move beyond taxonomic associations to focus on specific bacterial functions, host-microbe interactions, and their net effect on drug metabolism and disease progression.

In this study, GM-targeted therapies were found to significantly improve MDS-UPDRS III, BMs, use of laxatives, and MDA in PD patients. According to the GRADE assessment, change in MDS-UPDRS I, change in MDS-UPDRS III, change in NMSS, change in MMSE, change in BMs, change in use of laxatives, change in TAC, and change in GDS-15 (PSA response rate) were rated as moderate evidence. Given the limitations of small sample sizes, potential heterogeneity, and publication bias, future studies should involve larger, multi-center RCTs to further evaluate the long-term efficacy and safety of GM-targeted therapies in patients with PD.