The Potential Benefits and Controversies of Probiotics Use in Patients at Different Stages of Chronic Kidney Disease

The therapeutic modulation of the gut microbiome has been suggested to be one of the tools in the integrated management of chronic kidney disease (CKD) in recent years. Lactobacillus and Bifidobacterium genera are the two most commonly used probiotics strains. Most of the probiotics used in studies are mixed formulation. There is no consensus on the dose and duration of the probiotic administration for CKD patients Increasing evidence indicates that patients with early stage (1–2) CKD have an altered quantitative and qualitative microbiota profile. However, there was a dearth of prospective controlled studies on the use of probiotics in the early stage of the CKD population. The association between gut microbiota disturbance and advanced CKD was reported. Most randomized controlled trials on probiotic treatment used in CKD stage 3–5ND patients reported positive results. The metabolites of abnormal gut microbiota are directly involved in the pathogenetic mechanisms of cardiovascular disease and inflammation. We summarized 13 studies performed in the dialysis population, including 10 in hemodialysis (HD) patients and 3 in peritoneal dialysis (PD). Some controversial results were concluded on the decreasing plasma concentration of uremic toxin, symptoms, inflammation, and cardiovascular risk. Only three randomized controlled trials on PD were reported to show the potential beneficial effects of probiotics on inflammation, uremic toxins and gastrointestinal symptoms. There is still no standard in the dosage and duration of the use of probiotics in CKD patients. Overall, the probiotic administration may have potential benefit in improving symptoms and quality of life, reducing inflammation, and delaying the progression of kidney failure. Further research studies using a larger sample size with longer follow-up durations and a greater focus on clinical outcomes—including survival—are warranted to elucidate the significant clinical impact of the use of probiotics in CKD patients.


Introduction
Gut microbiota is involved in metabolic homeostasis, as shown in studies with humans. In recent decades, there is increasing evidence of the dysbiotic microbiota in the chronic kidney disease (CKD) population. Crosstalk does exist between the intestine and the kidney, which is the so-called kidney-gut axis. The gut microbiota interacts with the kidneys by very complicated mechanisms, including diet, microbiota-derived uremic toxins, immunemediated factors and metabolites, such as short-chain fatty acids (SCFAs). The overgrowth of the proteolytic bacteria (actinobacteria, proteobacteria, and firmicutes) was promoted by urea [1], by way of inducing the translocation of bacteria or their fractions into the bloodstream, and increasing the permeability of the intestinal wall, which in turn may enhance accelerated atherosclerosis and systemic inflammation [2].
With the progression of renal dysfunction, the accumulation of uremic toxins may further amplify their deleterious effects. The diversity and quantity of bacteria and the

Type, Dose, and Intervention Duration of Probiotics
The International Scientific Association for Probiotics and Prebiotics (ISAPP) consensus defines probiotics as "live microorganisms which when administered in adequate amounts confer a health benefit on the host" [5]. Probiotics are live and vital microorganisms which may be beneficial as part of a food or a supplement. The most frequently administered probiotics are Bifidobacteria longum, B. bifidum, Lactobacillus acidophilus, L. casei, L. sakei, L. reuteri, and Streptococcus thermophilus. Bifidobacteria are capable of producing vitamins, such as vitamin B1, B4, B6, B12, folate, and nicotinic acid [6]. Evidence from both in vivo and in vitro models has shown that the Bifidobacteria could effectively ameliorate epithelial damage, restore epithelial function [7,8], and produce short-chain fatty acids, particularly butyrate, through a cross-feeding mechanism by stimulating the growth of other bacterial species such as Lactobacillales [7]. L. acidophilus and B. bifidum have immune enhancement and stabilization effects on the gut mucosal barrier [9]. Enterococcus faecalis can generate bacteriocins (antimicrobial peptides) [10], which can inhibit the overgrowth of pathogens in the colon of CKD patients.
We summarized 25 original research studies which examined the benefit of probiotics in CKD patients. We included 24 randomized controlled studies and one cross-sectional study, which were conducted in CKD 3-5 non-dialysis stage as well as in dialysis patients. Lactobacillus and Bifidobacterium genera were the two most commonly used probiotics strains. In these studies, most of the microorganisms belonged to the Lactobacillus (19 studies, 73%) and Bifidobacterium genera (17 studies, 65%). Most of the probiotics used in the studies were mixed formulation, accounting for 80% of the total. Multiple probiotic products are now marketed and the efficacy, safety, and tolerance were shown in some trials [11][12][13].
The exact dose of probiotic that should be administered to CKD patients and for how long remain unclear. Insufficient doses of probiotics can be a potential factor leading to insignificant effects. The choice of the strain of probiotic used has until now been mainly empirical, contributing to the discrepancies between studies. Interventions usually last several weeks, but could last between 4 weeks and 5 years. Similarly, the probiotic dosages ranged from 16 × 10 9 CFU to 2.0 × 10 12 CFU; the dosages strategy had no specific standard. The formulation of the probiotic agents varied; some were administered in capsules, some were given in bags/envelopes to be dissolved in water or milk, while others were added to yogurt. The administration times of the probiotics reported in the studies included with meals or right after meals.
Another interesting issue is whether the living microorganisms could "safely" arrive the intestines and become colonized. Taki et al. found that the Bifidobacteria in most medicinal products and health foods cannot usually survive at pH 1.2 when it was exposed to gastric juices through oral administration. A gastro-resistant seamless capsule protects acid-fragile Bifidobacterium longum from the acidic gastric juice until they reach the intestines.
The capsule protects Bifidobacteria from inactivation by acidic gastric juice, preserving its activity in the intestines [14].

Dietary Interventions other Than Probiotics
The colon-derived solutes included PCS and IS, which are produced by the bacterial metabolism of the amino acids tyrosine and tryptophan, respectively. The alteration of diet can influence the production of colon-derived solutes. In addition to probiotics, prebiotics, and synbiotics, the therapeutic interventions for gut flora regulation include the dietary use of fruits, vegetables, and high-fiber products. The Atherosclerosis Risk in Communities (ARIC) study [15] included 12,000 adults with normal kidney function. The use of a healthy diet including many fruits and vegetables, fish, legumes, whole grains, and fibers with decreased red meat, sodium, and refined sugar intake were associated with a reduced risk of incident CKD for those subjects consuming more vegetable proteins and conferred a lower mortality in CKD patients [15,16]. Another study showed that the red meat intake was associated with a higher risk of developing end-stage kidney disease [17]. In the general population with normal kidney function, a vegetarian diet reduces the urinary excretion of PCS and IS by approximately 60% [18]. However, a diet with a higher content of nondigestible fibers, as consumed by vegetarians, may explain the prebiotic effect modulating uremic solutes production rather than the reduction in animal proteins [19]. In other countries, several examples of traditional food made by fermentation process contained large quantity of probiotics were taken as a daily food by healthy people. Wagner et al. [20] studied 888 CKD stage 3-5 patients and noted that yoghurts and probiotics (irrespective of the frequency of intake) are associated with reduced inflammation. Compared to subjects not consuming yoghurt, the ORs [95% CI] for CRP > 6 or >7 mg/L were significantly lower for those consuming ordinary yoghurt (0. 58 [20].

Effect of Probiotics in Early Stage of CKD (Stage 1-2)
In patients in early stages of CKD, the quantitative and qualitative profile of microbiota might be changed [1,21]. There was marked differentiation in the levels of metabolites (free amino acids and organic volatile compounds) from fecal and urinary samples between the progressor and non-progressor of patients with IgA nephropathy [22]. In CKD, the levels of both urea and ammonium increase in the gut, raising the pH level and promoting aerobic bacteria growth. In turn, these aerobic bacteria produce uremic toxins such as PCS, IS, and trimethylamine N-oxide which decrease the number of healthy anaerobic bacteria in the gut [23]. There was a dearth of prospective controlled studies on the use of probiotics in the early stage (1-2) of the CKD population. A case-control study revealed that the bacteria were not related to triglyceride, cholesterol, BUN, and creatinine levels; however, a negative correlation between Roseburia spp., Faecalibacterium prausnitzii, and CRP and renal function suggested that the depletion of butyrate-producing bacteria may contribute to CKD-associated inflammation and CKD progression [24]. In a randomized control trial, soy milk with Lactobacillus plantarum A7 was found to significantly reduce the oxidized glutathione concentration when it had been administrated to diabetes patients with proteinuria > 300 mg/day and glomerular filtration rate > 90 mL/minute for 8 weeks [25].

Effect of Probiotics in Advanced Stage of CKD (Stage 3-5ND)
Patients with advanced CKD are commonly recommended that to restrict their vegetable and fruit intake to reduce the danger of hyperkalemia and fluid overload. The shortage of fiber predisposes to dysbiosis with intestinal transit slowing, intestinal wall edema, and metabolic acidosis. Furthermore, polypharmacy, which is common in patients with end-stage kidney disease (ESKD) (including the use of vitamin D analogs, potassium, iron, phosphate-lowering agents, and diuretics), induced pro-inflammatory gastrointesti-nal overload [21,26]. Fermentative dysbiosis may be attributed to non-absorbed sugar hydrolyzation by several bacteria strains in the ascending colon and in the caecum.
To date, most randomized controlled trials on probiotic treatment used in the CKD stage 3-5ND patients have reported positive results ( Table 1). The colon-derived uremic toxin levels, such as PCS, IS, phenylacetylglutamine, and serum trimethylamine N-oxide, were commonly studied and a significant reduction in these uremic toxins by probiotics were shown in patients entering an advanced stage of CKD [23,[27][28][29][30]. Guida et al. reported the effect of Probinul neutro ® on 30 patients in stage 3-4 CKD in a randomized control trial for 4 weeks. They observed a significant decrease in the total plasma p-cresol levels however, without any improvement in GI symptoms [31]. The SYNERGY trial, with 37 stage 4-5ND CKD patients, demonstrated a decrease in serum p-cresyl sulfate, however, not in indoxyl-sulfate and a favorable change in the stool microbiome [29]. A significant change in both gut microbiota composition and intestinal bacterial metabolism was found in most of patients after taking Lactobacillales and Bifidobacteria [32,33]. The fecal pH measured in a multi-centered pilot study showed that the probiotic bacteria cohort (pH 6.94) was much lower than the placebo cohort (pH 7.29), which could be partially explained by the production or generation of lactic acid Lactobacillus in the mixture administered to the probiotic cohort [30].
Given the close relationship between gut microbiota disturbance and the progression of renal dysfunction, it is hypothesized that administering probiotic bacteria to advanced CKD patients may have the benefit of delaying the deterioration of kidney function. Two animal studies provided evidence of probiotics attenuating renal fibrosis and improving renal function in mice with CKD. Zhu et al. [11] administered C57BL/6 mice with L. casei Zhang (a probiotic-producing bacterium that was isolated from Chinese fermented sour milk samples) or L. acidophilus for 4 weeks. Probiotic pretreatment resulted in lower serum BUN and creatinine with less pathological damage, such as necrosis, tubular dilatation, casts formation, and brush border loss [11]. Wang [34] et al. supplemented C57BL/6 mice with a high dose or low dose of probiotics containing Lactobacillus acidophilus (TYCA06), Bifidobacterium longum subspecies infantis (BLI-02), and B. bifidum (VDD088). Both low and high doses of probiotics significantly reduced the serum levels of BUN and creatinine. The inflammation in the renal cortex and glomerular corpuscles and normal compact renal tubules in the renal pelvis were all improved after treatment with probiotics, particularly in the high-dose group. In human studies, early research shows the effect of probiotic supplementation on CKD progression. The glomerular filtration rate was improved significantly in patients who were on the low-protein diet combined with prebiotics and probiotics [35]. Similar results were drawn from the other two clinical controlled trials on the probiotic intakes that were associated with a decline in the progression of CKD [13,36].
In addition, the metabolites of altered gut microbiota have been reported to be directly involved in the pathogenesis of cardiovascular disease and inflammation. A meta-analysis was performed on the effects of probiotics in CKD patients that supports the potential effect of probiotics supplementation in reducing levels of PCS [37], inflammation markers, and oxidative stress [38]. However, the meta-analysis in question reflected that the methodological quality varied across studies [38]. Several clinical trials have focused on the effects of probiotics, including dietary supplements, on inflammation in CKD, and yielded inconsistent results [8,20,23,29,32,[39][40][41][42][43]. One trial from Mexico studied the different dose of lactobacillus casei Shirota (LcS) in CKD stages 3 and 4 and observed the higher dosages of LcS yielded better outcomes on decreasing the levels of inflammatory markers [23]. There was no evidence of a dose-effect relationship. Chen et al. studied the effects of the combinations of probiotic (Bifico) on interleukin 10 gene deficient (IL-10 KO) mice and Caco-2 cell monolayers [8]. IL-10 KO mice receiving the Bifico treatment had reduced the mucosal secretion of tumor necrosis factor-α and interferon-γ. The treatment of Caco-2 monolayers with Bifico or single-strain probiotics in vitro reduced the secretion of pro-inflammatory cytokines.

Effect of Probiotics in Dialysis Patients (Stage 5D)
With the deterioration of the residual kidney function, complications of ESKD and the dialysis procedure per se could lead to the dysbiosis of gut microbiota. Therefore, some studies assessed the probiotic treatment in either PD or HD patients. The constipation is reported to be common in patients on peritoneal dialysis (29%) and on hemodialysis (63%). The slowed transit time through the GI tract associated with constipation would lead to bacterial overgrowth in the stool that may contribute to the dysbiosis. Luo et al. [4] and Hu et al. [44] both compared the intestinal flora genome 16S rDNA sequencing in healthy people, CKD non-dialysis, HD patients, and PD patients in Chinese population. They concluded a remarkable difference in gut microbiota diversity before and after dialysis and inferred that PD and HD altered signal transduction and metabolic pathways. Table 2 depicted the randomized controlled trials of probiotic treatment in both HD and PD patients. Twelve studies were listed, with nine of them on HD and three on PD. The sample size ranged from 18 to 116 cases. Thirteen studies used probiotics alone and one study combined and probiotics. Most interventions lasted several weeks, and a few lasted as long as half a year [12,39].  Probiotics could significantly reduce the serum levels of endotoxin, pro-inflammatory cytokines (TNF-α and IL-6), IL-5, increase the serum levels of anti-inflammatory cytokine (IL-10), and preserve residual renal function in PD patients.

HD
Many studies focused on the effect of probiotics and prebiotics in reducing uremic toxin with inconsistent results. Early studies have shown a distinct reduction in fecal p-cresol while plasma p-cresol was only slightly decreased in the hemodialysis population [53]. Another study found that probiotic consumption decreased the plasma concentration of IS (26) but only insignificantly decreased indoxyl glucuronide [12]. In a pilot study (n = 22 HD, uncontrolled), the investigators demonstrated a significant reduction in serum PCS, but not IS, when the prebiotic oligofructose-enriched inulin was taken [54]. In another study (n = 56 HD, placebo-controlled), Sirich et al. noted that resistant starch, a form of fiber that is resistant to digestion, significantly decreased serum IS, and possibly, also PCS [55]. Posen et al. found no effect of prebiotic arabinoxylan oligosaccharides (AXOS) on the serum levels of PCS, p-cresyl glucuronide (PCG), IS, and phenylacetylglutamine (PAG), and only a borderline significant effect (not adjusted for multi-comparison) on the serum levels of trimethylamine N-oxide (TMAO) [27]. Borges et al. drew the conclusion that, in the CKD population, probiotics failed to reduce uremic toxins and inflammatory markers [55]. Based on the above evidence, probiotic therapy should be chosen with caution in HD patients. Further studies should be performed on the use of probiotic therapy in patients on dialysis.
Cardiovascular disease is the most common cause for mortality in dialysis patients. During dialyses, only the free fraction of protein-bound solutes can be removed. Apart from traditional CV risk factors that are present in the vast majority of CKD patients, emerging evidence has suggested that non-traditional risk factors, such as oxidative stress and inflammation, may play a role in the pathogeneses of cardiovascular diseases [3]. Whether probiotics could reduce CVD risk remains elusive. The imbalance of intestinal microbiota and deleterious colonic microbial metabolism are closely associated with the production of microbiota-derived metabolites [56]. The majority of gut-derived uremic toxins have shown a direct role in mortality and complications under CKD. For example, IS and PCS are positively correlated with an increased CVD mortality in CKD patients [57]. Trimethylamine N-oxide (TMAO) was recognized as a pro-atherogenic metabolite that involved in developing CV disease in CKD patients [58]. Kaminski et al. observed that IS was independently associated with the markers of impaired endothelial function, oxidative stress, and monocyte activation determinants [59]. Taki et al. reported the oral administration of the Bifidobacterium longum in a gastroresistant seamless capsule significantly decreased the serum levels of homocysteine in HD patients [14]. There have been studies showing the relationship between gut-derived uremic toxins with inflammation as an important CVD risk factor. However, the causal effect of probiotics on reducing CVD mortality has not been confirmed by prospective randomized controlled trials.

PD
An alteration in the composition and function of the intestinal microbiome in PD patients has been reported [4,32]. Compared to non-dialysis patients or HD patients, PD patients have the continuous peritoneal dialysis fluid in the abdominal cavity affecting the physical and chemical environment of gastrointestinal tract, which may have an influence on intestinal flora and the effects of probiotics treatment. There have been few reports on the effects of prebiotic, probiotic, and/or synbiotic supplementation in PD patients [37,51,52]. It has been shown that oral probiotics could lower the serum levels of uric acid [52], endotoxin, and inflammatory cytokines [37], and improve the nutrition and quality of life [51]. All three of these studies suggested that probiotic supplementation in patients undergoing PD is safe and well tolerated. Some researchers administrated p-inulin (prebiotics) to PD patients [60,61] and revealed a preliminary association between the p-inulin treatment and the changes in microbiome, metabolic pathways, as well as the plasma metabolome in PD patients. No studies reported CVD events and infections as outcome measures. Overall, there is a substantial lack of evidence for the effects of probiotics on clinical outcomes including patient mortality in PD patients.

Limitations of the Studies Reviewed
There are limitations to the reviewed studies. Firstly, the sample size of the studies were usually not large and the follow-up times were not long enough. Secondly, the diversity of bacterial strains, dose, intervention period, and combined medications made it difficult to compare the studies. Thirdly, long-term prognosis and outcome, such as mortality, were not sufficiently focused upon in these studies. Additionally, the uremic milieu in the gut may not be favorable for the survival of probiotics limiting their potential health benefits.

Conclusions
To date, studies on the potential benefit of probiotics in CKD were generally performed among non-dialysis CKD patients. Most studies were performed with CKD patients and but less so in dialysis patients-especially peritoneal dialysis patients. Generally, the probiotic administration may have a potential benefit in improving symptoms and quality of life, reducing inflammation, and delaying the progression of kidney failure. Further research studies on using a larger sample size with longer follow-up duration and focus on clinical outcomes-including survival-are warranted to elucidate the significant clinical impact on the use of probiotics in CKD patients.