Effects of resistant starch supplementation on renal function and inflammatory markers in patients with chronic kidney disease: a meta-analysis of randomized controlled trials

Abstract

Background

Recent studies have shown that consumption of resistant starch (RS) has beneficial effects on the gut microbiota and immune function in patients with chronic kidney disease (CKD). The objective of this study was to evaluate the effects of RS on inflammation, uremic toxins, and renal function in patients with CKD through a systematic review and meta-analysis.

Methods

This systematic review was conducted according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA)-2020. We included randomized controlled trials comparing RS supplementation to placebo. The National Library of Medicine (PubMed), Excerpta Medica Database (Embase), Cochrane Library, Web of Science, China National Knowledge Internet (CNKI) databases, and two gray literature sources – Baidu and Research Gate, were used for search, up to 28 August 2024. There was no limitation on publication date, but only manuscripts published in English and Chinese were included.

Results

A total of 645 articles were retrieved. Ten articles met the inclusion criteria, and a total of 355 subjects were included. The analysis revealed that RS dietary intervention can significantly reduce indoxyl sulfate (IS) levels (SMD: −0.37, 95% confidence interval (CI): −0.70 to −0.04, p = .03) and blood urea nitrogen (BUN) levels (SMD: −0.30, 95% CI: −0.57 to −0.02, p = .03). There were no significant differences in the levels of interleukin-6 (IL-6), p-cresyl sulfate (p-CS), albumin, phosphorus, or tumor necrosis factor-α.

Conclusions

The RS diet has potential beneficial effects on uremic toxin levels and renal function indices in patients with CKD. RS supplementation can reduce uremic toxin levels and improve renal function but does not reduce the inflammatory response in patients with CKD. Nevertheless, results should be cautiously interpreted, because of the limited sample size and different treatment dosages. Further research is necessary to corroborate the beneficial effects of RS2 supplementation in this population.

Introduction

Chronic kidney disease (CKD) is a widespread condition that affects approximately 13% of the general population and approximately 36% of high-risk individuals. As kidney function declines, electrolytes, excess fluids, and nitrogenous wastes accumulate in the body, leading to progression of the disease and related complications [1–3]. Along with the decline in renal function, the accumulation of nitrogenous uremic toxins leads to the worsening of symptoms and to complications. Many uremic toxins, such as indoxyl and p-cresol sulfate, are produced entirely by the gut microbiota [4]. There is increasing evidence that therapeutic interventions aimed at strengthening the gut microbiome may slow the progression of CKD [5].

The human gastrointestinal microbiota play a critical role in the overall health of the host. Considered symbiotic ‘complementary organs’, the composition and metabolic activity of these microbial communities depend on factors such as the host’s genome, diet, health, and lifestyle [6]. While healthy individuals benefit from the protective and nutritional functions associated with the presence of a balanced microbial ecosystem, the dysbiome (altered microbiome) has been shown to contribute to a variety of pathophysiologies. Under normal circumstances, homeostasis is maintained by the interaction between the host and intestinal flora [6]. However, patients with CKD display quantitative and qualitative changes in the intestinal flora, and this imbalance in the intestinal flora increases the abundance of pathogenic flora relative to that of symbiotic flora [7]. In patients with CKD, dysbiosis is associated with inflammation, oxidative stress, and the production of reactive oxygen species (ROS). Together, these factors lead to further renal injury by affecting microcirculation and blood perfusion [8].

Resistant starch (RS), which is found in green bananas, raw potatoes, high-amylose corn starch, and other foods, is composed of glucose monomers bound by α-glycosidic bonds in the form of amylose and amylopectin polymers [8]. However, unlike the bonds in digestible starches, the α-glycosidic bonds in RS are inaccessible to intestinal digestive enzymes due to their physical and/or chemical properties [9]. RS reaches the colon through the upper digestive tract, where it serves as a digestible substrate for beneficial colon bacteria [10]. Dietary supplementation with RS has shown promising results in altering the microbiota in animal models of CKD. Human studies involving RS supplementation have shown that it can alter the human gut microbiota [11]. The human gastrointestinal microbiota plays a fundamental role in the overall health status of the host. Increasing evidence suggests that therapeutic interventions aimed at fortifying the intestinal microbiota may slow the progression of CKD.

RS is a prebiotic that promotes the proliferation of gut bacteria such as Bifidobacterium and Lactobacillus and can increase the production of metabolites (including short-chain fatty acids), thus providing health benefits [12]. Previous studies have shown that intestinal disorders, including systemic inflammation, accumulation of uremic toxins, and infection, have significant effects on renal physiology and pathophysiology and may lead to increased morbidity and mortality in ESRD patients [10]. RS is resistant to the activity of pancreatic alpha-amylase, is not absorbed by the small intestine, and is fermented by the gut microbiome after reaching the large intestine, minimizing the negative effects of its imbalance by regulating the gut microbiome. RS is generally low in cost and easy to obtain and may become an important part of the treatment regimen for patients with CKD [13]. Collectively, RS appears as a promising adjuvant therapy in patients with CKD. However, conclusions are inconsistent. Wu et al.’s meta-analysis showed that dietary fiber supplementation could significantly reduce plasma p-CS levels in CKD patients [14], and Khosroshahi et al. also reported that p-CS levels were significantly reduced in end-stage renal patients after RS treatment [15]. However, Esgalhado et al. reported opposite results [16]. Since no consensus for the clinical use of RS has been reached thus far, therefore, this study evaluated the therapeutic effect of RS in patients with CKD using an evidence-based approach [8].

Methods

Study protocol

This study was conducted in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines [17], and it was registered in INPLASY (DOI: 10.37766/inplasy2024.2.0018).

PICO question

We considered information regarding the effect of RS supplementation (I) on CKD patients (P) with or without a comparator (C). Our goal was to assess the effect of RS on inflammatory and renal function indicators in patients with CKD (O).

Search strategy

Two researchers (Zhang Y and Hu XY) independently searched the PubMed, Embase, Cochrane Library, Web of Science, and China National Knowledge Internet (CNKI) databases from their inception to 28 August 2024 using medical subject headings (MeSH), Emtree and text, with no language limitations.

The following keywords were used in the search strategy: ‘renal insufficiency’, ‘chronic’, ‘renal dialysis’, ‘hemodialysis’, ‘end-stage renal disease’, ‘CKD’, ‘resistant starch’, ‘RS’, and ‘randomized controlled trial’ (Supplementary Table S1). Reference lists from the identified studies were also searched for potentially eligible articles. The identified publications were imported into EndNote X9.1 (Clarivate Analytics, Philadelphia, PA). After removal of duplicate records and irrelevant literature, appropriate studies with detailed classifications were identified.

Eligibility criteria

Two authors (Duan K and Hu YC) independently performed the primary review to identify trials that met the inclusion criteria [18] (Supplementary Table S3). Any discrepancies were resolved by discussion and consensus (Figure 1). The following inclusion criteria were used: (1) the study type was a randomized controlled trial (RCT); (2) the study included patients with CKD; and (3) the intervention measure was RS supplementation. The exclusion criteria were as follows: (1) for repeated publications, the longest or most recent publication was selected; (2) literature with incomplete or unavailable research data, as well as abstracts, reviews, systematic reviews, experience summaries, theoretical discussions, case reports, and qualitative studies were excluded; (3) the study involved animal experiments or in vitro experiments; and (4) the study was a nonrandomized controlled clinical study.

Figure 1.

PRISMA 2009 flow diagram.

Data extraction

Two reviewers (Zhang Y and Hu YC) independently extracted the data from the same set of publications. The following information was extracted: author, year of publication, sample size, study design, and proportion of males. The main results included the following: (1) reported levels of the inflammatory indicators interleukin-6 (IL-6) and tumor necrosis factor-α (TNF-α) and (2) reported values of the renal function indicators indoxyl sulfate (IS), p-cresyl sulfate (p-CS), blood urea nitrogen (BUN), phosphorus, and serum albumin (Table 1).

Table 1.

Main characteristics of the included studies.

Study	Nation	Sample		Average age (years)		Intervention		BMI (kg/m−2)		Male ratio (%)		Outcome index	Follow-up
		T/C		T/C		T/C				T/C
de Andrade et al. [19]	Brazil	15	11	55 ± 12		UBF-48% RS	Waxy corn starch	26 ± 4	25 ± 4	Unclear		IL-6, TNF-α, PCS, IS, Alb, P	4 weeks
Esgalhado et al. [20]	Brazil	15	16	56 ± 7	53 ± 11	HAM-RS2	Manioc flour	26 ± 5	27 ± 5	46.7	68.8	IL-6, PCS, IS, Alb, P	4 weeks
Esgalhado et al. [16]	Brazil	15	16	56 ± 7	53 ± 11	HAM-RS2	Waxy corn starch	56 ± 7	53 ± 11	46.7	68.8	IL-6, PCS, IS	4 weeks
Kemp et al. [21]	Brazil	10	10	53 ± 12	55 ± 11	HAM-RS2	Manioc flour	53 ± 12	55 ± 11	30.0	80.0	BUN, Alb	4 weeks
Khosroshahi et al. [22]	USA	22	22	52 ± 11	60 ± 14	HAM-RS2	Regular wheat flour	52 ± 11	60 ± 14	54.5	72.7	IL-6, BUN, Alb, P	8 weeks
Khosroshahi et al. [15]	USA	23	21	53 ± 10	57 ± 13	HAM-RS2	Waxy corn starch	53 ± 10	57 ± 13	56.0	60	PCS, IS, BUN, Alb, P	8 weeks
Laffin et al. [23]	USA	9	11	53 ± 11	57 ± 9	HAM-RS2	Regular wheat flour	53 ± 11	57 ± 9	66.7	63.3	IL-6, TNF-α, BUN	16 weeks
Meng et al. [24]	China	34	36	62 ± 9	61 ± 9	HAM-RS2	Low-protein flour	62 ± 9	61 ± 9	52.9	58.3	IL-6, TNF-α, BUN, Alb, P	12 weeks
Meng et al. [25]	China	16	13	63 ± 7	64 ± 8	HAM-RS2	PS cookies	63 ± 7	64 ± 8	34.5	37.9	IL-6, TNF-α, BUN, Alb, P	12 weeks
Sirich et al. [26]	USA	20	20	54 ± 14	58 ± 13	HAM-RS2	Waxy corn starch	54 ± 14	58 ± 13	55.0	65.0	PCS, IS, BUN, Alb, P	6 weeks

BUN: blood urea nitrogen; BMI: body mass index; HAM-RS2: high amylose resistant starch; IL-6: interleukin-6; IS: indoxyl sulfate; P: phosphorus; PCS: p-cresyl sulfate; PS: Polygonatum sibiricum; T/C: treatment group/control group; UBF: unripe banana flour.

Summary of effect sizes

The standardized mean difference (SMD) and the 95% confidence interval (CI) were used as the effect size measures for dichotomous data. The weights of the enrolled studies were accounted for by taking into account the sizes of the treatment and control groups and the total sample size. The Q test and I² were used to evaluate the heterogeneity between the effect results, and the effect size of RS supplementation in patients with CKD was calculated using a fixed effects model or a random effects model.

Heterogeneity was determined as follows: an I² statistic of 0–25% was considered low heterogeneity; I² of 25–50% was considered medium heterogeneity; I² of 50–75% was considered high heterogeneity; and I² of 75–100% was considered powerful heterogeneity. The p value was determined using the χ² test; p < .05 was considered to indicate statistical significance [27].

Risk of bias

The quality of all trials were independently evaluated by two authors (Hu XY and Hu YC) according to the Cochrane Risk of Bias 2.0 (RoB 2) tool [28] (Supplementary Figure S1), which assessed five domains as follows: randomization process, deviations from intended interventions, missing outcome data, outcome measurements, and selection of the reported results. Each study was evaluated within these domains, and an overall assessment of the risk of bias was categorized into three classifications: ‘low’ for studies with minimal risk of bias across all domains, ‘some concerns’ for studies displaying substantial risk of bias across multiple domains, and ‘high’ for studies exhibiting high risk of bias in at least one domain. Any disagreements between the authors were settled by discussion with a third author (Yang SY). A weighted kappa value was calculated and used to assess the agreement between the reviewers for the overall risk of bias assessment [18] (Supplementary Table S4). Publication bias was evaluated using Begg’s plots, Egger’s tests, and funnel plots (if more than 10 studies were included).

GRADE quality assessment

The overall quality of evidence was evaluated by two authors (Hu YC and Hu XY) according to the Grading of Recommendations Assessment, Development and Evaluation (GRADE) criteria by evaluating the following evidence [29, 30]: (1) study limitations; (2) inconsistency; (3) imprecision; (4) indirectness; and (5) publication bias. Any disagreements between the two authors were first resolved by discussion and then by consulting with a third author (Yang SY) or by the senior author (Zhang Y). The results and the overall quality of evidence are presented in Supplementary Table S2.

Statistical analysis

STATA 16.0 (Stata Corp LP, College Station, TX) was used to perform the statistical analyses. L’Abbe plots and meta-regression were used for intuitive assessment of heterogeneity. For the remaining studies, a random effect model was used to pool the effect sizes to calculate the statistical heterogeneity. Heterogeneity was analyzed using I² and χ² statistics. If there was significant heterogeneity, a Galbraith plot was generated to evaluate the consistency and quality of the results. Sensitivity analysis, subgroup analysis, and meta-regression were performed to determine sources of heterogeneity.

Results

Study selection

After exclusion of duplicate records, a total of 645 studies were identified during the initial search. Of those, 160 articles were excluded following screening of the titles/abstracts to ascertain the presence or absence of the inclusion and exclusion criteria. We read the full texts of the remaining articles; 449 records were excluded, and 10 RCTs [15, 16, 19–26] involving a total of 355 patients were included in the quantitative synthesis (Figure 1). The main characteristics of the included RCTs (country, sample size, average age, intervention measures, male ratio, follow-up time, and main results) are presented in Table 1.

Effect on levels of inflammatory markers

Interleukin-6

Six RCTs (n = 213) [16, 19, 20, 23–25] were included in the evaluation of the levels of inflammatory markers in CKD patients in the RS and placebo groups. Statistical analysis revealed no significant difference between the RS group and the placebo group (SMD: −0.10, 95% CI: −0.43 to 0.24, p = .56, I² = 33.68%, Figure 2).

Figure 2.

Forest plot of interleukin-6.

Sensitivity analysis (Supplementary Figure S2) and Galbraith’s plots (Supplementary Figure S3) were generated to evaluate the stability of the results. The results of the sensitivity analysis suggested that none of the individual studies significantly affected the pooled odds ratio (OR), indicating that the results were statistically robust. No significant publication bias was found according to Begg’s plots (p = .108, Supplementary Figure S4) and Egger’s test (p = .406, Supplementary Figure S5). Subgroup analysis showed that there was no significant difference in crossover or parallel trial group (Supplementary Figure S6).

TNF-α

Three RCTs (n = 116) [19, 23, 25] were included to evaluate the associations between the use of RS or a placebo and the levels of inflammatory markers in CKD patients. Statistical analysis revealed no significant difference between the RS group and the placebo group (SMD: −0.08, 95% CI: −0.44 to 0.28, p = .67, I² = 0%, Figure 3).

Figure 3.

Forest plot of TNF-α.

A sensitivity analysis (Supplementary Figure S7) was performed, and Galbraith’s plots (Supplementary Figure S8) were generated to evaluate the stability of the results. The results of the analysis suggested that none of the individual studies significantly affected the pooled OR, indicating that the results were statistically robust. No significant publication bias was found according to Begg’s plots (p = .117, Supplementary Figure S9) and Egger’s test (p = .323, Supplementary Figure S10).

Effects on renal function

Blood urea nitrogen

A pooled analysis of five studies (n = 203) [15, 23–26] was used in the present review to evaluate the association between BUN levels in the RS group and those in the placebo group. The overall effect size indicated that RS supplementation was associated with a significant decrease in BUN (standard mean difference (SMD): −0.30, 95% CI: −0.57 to −0.02, p = .03, I² = 0%, Figure 4).

Figure 4.

Forest plot of blood urea nitrogen.

A sensitivity analysis (Supplementary Figure S11) was performed, and Galbraith’s plots (Supplementary Figure S12) were generated and used to evaluate the stability of the results. The results of this analysis suggested that none of the individual studies significantly affected the pooled OR, indicating that the results were statistically robust. No significant publication bias was found according to Begg’s plots (p = .142, Supplementary Figure S13) and Egger’s test (p = .142, Supplementary Figure S14).

Indoxyl sulfate

Five RCTs (n = 172) [15, 16, 19, 20, 26] were included in the analysis of the association between IS levels in the RS group and those in the placebo group. Our results showed that RS supplementation significantly reduces IS levels in patients with CKD (SMD: −0.37, 95% CI: −0.70 to −0.04, p = .03, I² = 0%, Figure 5).

Figure 5.

Forest plot of indoxyl sulfate.

To evaluate the stability of the results, a sensitivity analysis (Supplementary Figure S15) was performed, and Galbraith’s plots (Supplementary Figure S16) were generated. The results of this analysis suggest that none of the individual studies significantly affected the pooled OR, indicating that the results were statistically robust. No significant publication bias was found according to Begg’s plots (p = .642, Supplementary Figure S17) and Egger’s test (p = .331, Supplementary Figure S18). Subgroup analysis showed that there was no significant difference in crossover or parallel trial group (Supplementary Figure S19).

Phosphorus

A pooled analysis of six studies (n = 205) [15, 19, 20, 22, 23, 26] was performed to estimate the association between phosphorus levels in the RS group and those in the placebo group. The overall effect size indicated that the phosphorus levels in the two groups did not differ significantly (SMD: −0.25, 95% CI: −0.70 to 0.20, p = .27, I² = 61.86%, Figure 6). Meta-regression by bubble plot revealed no significant heterogeneity in sample size per study (p = .985, Supplementary Figure S20), publication year (p = .296, Supplementary Figure S21), or country of publication (p = .531, Supplementary Figure S22).

Figure 6.

Forest plot of phosphorus.

A sensitivity analysis was performed (Supplementary Figure S23), and Galbraith’s plots (Supplementary Figure S24) were generated to evaluate the stability of the results. This analysis suggested that none of the individual studies significantly affected the pooled OR, indicating that the results were statistically robust. No significant publication bias was found according to Begg’s plots (p = .851, Supplementary Figure S25) and Egger’s test (p = .946, Supplementary Figure S26).

Serum albumin

Seven RCTs (n = 284) [15, 19, 20, 22, 24–26] were analyzed to estimate the association between serum albumin levels in the RS group and those in the placebo group. Statistical analysis revealed no significant difference between the RS group and the placebo group (SMD: 0.11, 95% CI: −0.12 to 0.34, p = .35, I² = 0%, Figure 7).

Figure 7.

Forest plot of serum albumin.

A sensitivity analysis was performed (Supplementary Figure S27), and Galbraith’s plots (Supplementary Figure S28) were generated to evaluate the stability of the results. This analysis suggested that none of the individual studies significantly affected the pooled OR, indicating that the results were statistically robust. No significant publication bias was found according to Begg’s plots (p = .881, Supplementary Figure S29) and Egger’s test (p = .992, Supplementary Figure S30).

Effects on uremic toxin levels

p-Cresyl sulfate

Five RCTs (n = 172) [15, 16, 19, 20, 26] were included to estimate the association between p-CS levels in the RS group and those in the placebo group. Statistical analysis revealed no significant difference in p-CS levels between the RS group and the placebo group (SMD: −0.09, 95% CI: −0.38 to 0.21, p = .56, I² = 0%, Figure 8).

Figure 8.

Forest plot of p-cresyl sulfate.

A sensitivity analysis (Supplementary Figure S31) and Galbraith’s plots (Supplementary Figure S32) were used to evaluate the stability of the results. The results of this evaluation suggested that none of the individual studies significantly affected the pooled OR, indicating that the results were statistically robust. No significant publication bias was found according to Begg’s plots (p = .624, Supplementary Figure S33) and Egger’s test (p = .969, Supplementary Figure S34). Subgroup analysis showed that there was no significant difference in crossover or parallel trial group (Supplementary Figure S35).

Discussion

An imbalance in the intestinal microbiome is one of the risk factors for the progression of advanced CKD to uremia [31]. According to the core points of the theory of the ‘entero-renal axis’ [32] and the concept of the ‘chronic nephropathy-colon axis’ [33], an imbalance in intestinal microbes induces systemic microinflammation through the accumulation of uremic toxins, and this exacerbates kidney damage [34]. In recent years, an increasing number of studies have shown that alteration and destruction of the intestinal flora may lead to the production of uremic toxins and affect the progression of CKD [35]. p-CS and IS, as the two main enterogenous uremic toxins, are mainly produced in the intestine [36]. Due to their high protein-binding ability, they are difficult to completely clear by peritoneal dialysis or hemodialysis, and the continuous accumulation of these toxins in the body will further aggravate renal microinflammation [37]. p-CS, one of the main enteric-borne uremic toxins, has been found to be closely related to the progression of CKD [14, 38, 39].

The level of IS in the blood is associated with coronary atherosclerosis, coronary artery disease, and heart failure [40]. In vitro experiments have shown that IS can promote the production of TNF-α and IL-1β by THP-1-derived macrophages [41] and that it is a predictor of cardiovascular disease [42] and mortality from cardiovascular disease [26] in patients with CKD. The Sirich study showed that free IS levels in plasma decreased significantly 6 weeks after the RS content of the diet was increased [26]. de Andrade et al. [19] did not observe an effect of adding immature banana powder (48% RS) to the diet on uremic toxin markers and speculated that a higher dose of RS is required to promote a reduction in the levels of uremic toxins. The results of the current meta-analysis show that IS levels decreased significantly after RS supplementation of the diet but that p-CS levels did not change significantly.

Supplementation with RS is a nutritional strategy that can decrease inflammation and oxidative stress levels in patients with CKD. An experiment in rats with CKD showed that intake of dietary fiber significantly delayed the progression of CKD and reduced oxidative stress and inflammation [43]. RS is considered a type of dietary fiber, and it may be involved in the anti-inflammatory process in individuals with various diseases, but the mechanism underlying this effect is not clear [44]. Wen et al. proposed that different types of RSs differentially regulate the intestinal flora. In the future, it will be necessary to establish structural characteristics and quantitative methods for identifying RSs both in vitro and in vivo to provide a strong theoretical basis for the application of RSs, the development of functional foods and the treatment of various chronic diseases [45].

The results of this study show that RS supplementation can improve symptoms associated with the presence of uremic toxins and renal function in patients with CKD; in these patients, IS and BUN levels are decreased. IS, IL-6, TNF-α, albumin, and phosphorus levels are not significantly altered by RS supplementation. The reason for our failure to observe changes in IS, IL-6, TNF-α, albumin, and phosphorus levels might be that the sample sizes of the included studies were too small, increasing reporting bias and decreasing the level of significance of the evidence. Second, due to insufficient data, differences in the duration and doses of RS supplementation given to the patients in the meta-analysis may have affected the results. RS supplementation has a positive effect on patients with CKD by reducing the inflammatory response and improving urinary toxin symptoms and renal function. However, because this systematic review included several studies in which small samples were used, caution should be exercised when interpreting the results. It will be necessary to verify the results of this study using larger samples and high-quality RCTs. Third, when developing strategies that target the gut microbiota of patients with CKD, the impact of different doses of RS, especially combinations of different types of prebiotics, needs to be further evaluated to allow us to better understand their impact on gut-derived metabolites, thereby benefiting patients with CKD.

This systematic review has several limitations. (1) The number of included studies was small. Currently, there are few clinical studies on the use of RS in individuals with CKD. (2) The sample sizes of some of the studies in the included literature were small; thus, there may be random errors, and the data obtained by some studies may not be sufficiently stable without use of the original data. (3) Only published Chinese and English literature was searched; thus, there may be publication bias caused by incomplete literature inclusion, and this might affect the results of the meta-analysis. (4) The randomized cross-controlled studies included in this systematic review used a short washout period, possibly leading to residual effects. (5) Among the included studies, Andrade’s study may have had an impact on the final results due to the lack of description of the characteristics of the participants. (6) Since the vast majority of studies originated from Western countries, thus, extrapolation of these results to Eastern populations needs to be further verified. (7) Various dialysis regimens, doses, durations, center settings, populations enrolled, etc. contributed to significant heterogeneity in some of the results, and many of the studies suffer from significant sources of bias, which may influence our conclusions.

Conclusions

Our review highlights the effects of RS supplementation among patients with CKD, especially those not yet on dialysis. The results of this study show that RS supplementation can improve symptoms associated with the presence of uremic toxins and renal function in patients with CKD. Nevertheless, results should be cautiously interpreted; large and pragmatic multicenter trials are thus necessary to corroborate before RS consumption could be considered a legitimate therapy in CKD.

Funding Statement

None.

Author contributions

Conceptualization: Kai Duan, Yong Zhang, and Ying-Chun Hu. Data curation: Shi-Yun Yang, Kai Duan, and Ying-Chun Hu. Formal analysis: Shi-Yun Yang, Yong Zhang, and Ying-Chun Hu. Investigation: Xiang-Yang Hu, Yong Zhang, and Ying-Chun Hu. Methodology: Kai Duan and Ying-Chun Hu. Project administration: Kai Duan, Yong Zhang, and Ying-Chun Hu. Resources: Xiang-Yang Hu and Ying-Chun Hu. Software: Xiang-Yang Hu and Ying-Chun Hu. Supervision: Xiang-Yang Hu, Yong Zhang, and Ying-Chun Hu. Validation: Xiang-Yang Hu and Ying-Chun Hu. Visualization: Xiang-Yang Hu and Ying-Chun Hu. Roles/writing – original draft: Xiang-Yang Hu and Ying-Chun Hu. Writing – review and editing: Shi-Yun Yang, Yong Zhang, and Ying-Chun Hu.

Ethical approval

All the data used in this meta-analysis have been published; ethical approval is not available.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability statement

Data are available from the authors upon reasonable request. The original contributions presented in the study are included in the article/Supplementary Material; further inquiries can be directed to the corresponding authors.