We systematically searched, extracted, and analyzed the data from reported systematic reviews and meta-analyses that focused on the effects of dietary intervention on DN according to the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) guidelines (40). The present umbrella review adhered to the methodological guidance outlined in the Joanna Briggs Institute Manual for Evidence Synthesis of Umbrella Reviews (41) and followed the procedures delineated in the Cochrane Handbook for Conducting Systematic Reviews (42). Furthermore, we proactively enrolled our umbrella review in the International Prospective Register of Systematic Reviews (PROSPERO), with the registration number CRD42024512670. (https://www.crd.york.ac.uk/PROSPERO/).

Systematic reviews and meta-analyses of RCTs evaluating the effects of dietary intervention on DN incidence in individuals of any ethnicity or sex in all countries and settings were eligible for inclusion. Data on individual dietary interventions were extracted separately if two or more dietary interventions were reported in a single meta-analysis. If two or more meta-analyses (those published more than 24 months apart) were performed on the same dietary intervention and clinical outcome of DN, we included the latest meta-analysis for data analysis. In the event that multiple meta-analyses were conducted within a 24-month timeframe, preference was given to the meta-analysis encompassing the highest number of RCTs. If an equal number of RCTs existed, priority was assigned to the meta-analysis with a superior AMSTAR score. In addition, if the latest meta-analysis did not perform dose-response analysis, while another meta-analysis did, both studies were included for data extraction. Non-English studies and animal and cell culture studies were also excluded.

This umbrella review is centered on systematically reviewing meta-analyses that assess the effects of dietary intervention on DN. The primary focus of the original articles incorporated within these systematic reviews and meta-analyses should be directed toward identifying dietary interventions that have the potential to either improve or exacerbate the clinical outcomes of DN. Studies evaluating the efficacy of a certain dietary intervention for the risk of DN were excluded.

We included a meta-analysis that reported at least 1 type of dietary intervention for DN, including probiotics, a salt restriction diet, vitamin D, soy isoflavone, and low-protein diets. The efficacy of dietary intervention on the clinical outcomes of DN was evaluated by the risk ratio (RR), mean difference (MD) or standard mean difference (SMD) with 95% confidence intervals (CIs).

The outcomes of this umbrella review included endocrine metabolic outcomes, including the urinary albumin excretion rate (UAER), serum creatinine (Scr), blood urea nitrogen (BUN), the urinary albumin–creatinine ratio (UACR), fasting blood glucose (FBG), total cholesterol (TC), triglycerides (TG), low-density lipoprotein cholesterol (LDL-C), high-density lipoprotein cholesterol (HDL-C), systolic blood pressure (SBP), diastolic blood pressure (DBP), coenzyme Q10 (CoQ10), the glomerular filtration rate (GFR), and the creatinine clearance rate (CrCl).

Only systematic reviews and meta-analyses of RCTs evaluating the effects of dietary intervention on DN incidence in individuals of any ethnicity or sex in all countries and settings were eligible for inclusion. All the included systematic reviews and meta-analyses needed to focus on dietary intervention in DN patients and describe the meta-analysis methods in detail, including the complete search strategy, inclusion and exclusion criteria, literature quality evaluation criteria, result evaluation methods, analysis methods and procedures, and interpretation criteria.

In our study, we systematically searched PubMed, Embase, the Web of Science, and the Cochrane Database of Systematic Reviews until July 2023 for relevant systematic reviews and meta-analyses of RCTs. We also reviewed the reference lists of the included meta-analyses to find additional relevant articles.

The databases were accessed using Medical Subject Headings (MeSH), keywords, and text terms related to dietary intervention and DN, following the Scottish Intercollegiate Guidelines Network (SIGN) recommendations for literature search methodology (43). The detailed search strategy for PubMed was as follows: (((“Diabetic Nephropathies”[Mesh]) OR (((((((((((((((((Nephropathies, Diabetic) OR (Nephropathy, Diabetic)) OR (Diabetic nephropathy)) OR (Diabetic Kidney Disease)) OR (Diabetic Kidney Diseases)) OR (Kidney Disease, Diabetic)) OR (Kidney Diseases, Diabetic)) OR (Diabetic Glomerulosclerosis)) OR (Glomerulosclerosis, Diabetic)) OR (Intracapillary Glomerulosclerosis)) OR (Nodular Glomerulosclerosis)) OR (Glomerulosclerosis, Nodular)) OR (Kimmelstiel-Wilson Syndrome)) OR (Kimmelstiel Wilson Syndrome)) OR (Syndrome, Kimmelstiel-Wilson)) OR (Kimmelstiel-Wilson Disease)) OR (Kimmelstiel Wilson Disease))) AND (((“Diet”[Mesh]) OR (diets)) OR ((“Food”[Mesh]) OR (foods)))) AND (meta-analysis OR systematic review).

All the retrieved literature was screened using Endnote X9. After excluding duplicates, two authors screened the titles and abstracts and identified meta-analyses that met the inclusion criteria through full-text reading independently. All disagreements between the two authors were resolved by a third author. In addition, we hand-searched studies from the reference lists to identify meta-analyses that might have been excluded ( Figure 1 ).

The methodological quality of each meta-analysis was assessed by two authors using AMSTAR, a validated, stringent, and reliable tool for evaluating systematic reviews and meta-analyses (44, 45). In addition, according to the Grading of Recommendations, Assessment, Development and Evaluation (GRADE), evidence of each health outcome was evaluated and graded as “high”, “moderate”, “low” or “very low” quality to draw conclusions (46). Additionally, we classified the evidence of outcomes into 4 categories following the evidence classification criteria: class I (convincing evidence), class II (highly suggestive evidence), class III (suggestive evidence), class IV (weak evidence) and NS (nonsignificant) (47–50). The detailed criteria for evidence classification are shown in Table 1 .

Two investigators autonomously retrieved the pertinent data from each qualifying study: 1) name of the author, 2) publication date, 3) dietary intervention, 4) control, 5) outcomes, 6) number of included studies, 7) sample size, 8) length of follow-up, and 9) MD or SMD estimates with 95% CIs. In addition, we extracted the meta-analytic model used (random or fixed), estimate of heterogeneity (I ² and Cochran’s Q test) and small-study assessment (Egger’s test, Begg’s test and funnel plot). When dose response analysis and subgroup analysis were performed, we extracted the P value for nonlinearity and any reported estimate for subgroup analysis. Any disagreements were resolved by a third author.

We recalculated the RR, MD or SMD with 95% CIs through random or fixed effects models and evaluated the heterogeneity (I ² and Cochran’s Q test) and small-study effects (Egger or Begg test for each systematic review and meta-analysis with more than 10 studies) in each meta-analysis when sufficient data were provided (51–53). For dietary interventions identified as class I-II evidence, high-quality evidence or moderate-quality evidence, we conducted sensitivity analysis when sufficient data were available to determine the effect of some individual studies on the total significance of the evidence. Dose-response analysis of DN incidence associated with any dietary intervention was also performed. Furthermore, if the most recent meta-analysis did not involve clinical studies that involved other meta-analyses, we combined the data of these studies and performed a reanalysis. A P < 0.10 was considered to indicate heterogeneity, and for other tests, P < 0.05 was considered to indicate statistical significance. Review Manager v5.4.1 (Cochrane Collaboration, Oxford, UK) was used for evidence synthesis. Egger and Begg tests, along with sensitivity analysis, were performed using Stata v15.1.

A flowchart of the literature search and selection process is presented in Figure 1 . After a systematic literature search, 501 unique articles were identified. A total of 36 meta-analyses were yielded based on our inclusion criteria. We extracted 9 unique dietary interventions (including probiotics, a salt restriction diet, vitamin D, soy isoflavone, CoQ10, ketoanalog, dietary polyphenols, antioxidant vitamins, and low-protein diets) and 55 corresponding outcomes in meta-analyses, including 34 significantly associated outcomes and 21 nonsignificantly associated outcomes ( Table 2 ). After a careful evaluation of evidence quality using established criteria, all outcomes were classified as IV or NS (nonsignificant) evidence. In addition, according to the GRADE rating criteria, only five dietary interventions were rated as moderate-quality evidence, 33 were rated as low-quality evidence, and 17 were rated as very low-quality evidence ( Table 2 ). Moderate-quality evidence and low-quality evidence for dietary interventions that could significantly improve clinical outcomes in patients with DN are presented in Figure 2 .

A total of 5 meta-analyses (11, 16, 17, 19, 21) studied the efficacy of probiotic intervention for DN. The meta-analysis published by Dai et al. in 2022 (16) included 6 RCTs describing 446 patients with DN, in which Lactobacillus acidophilus, Lactobacillus casei, Lactobacillus lactis, Bifidobacterium bifidum, Bifidobacterium longum, Bifidobacterium infants, Lactobacillus plantarum A7, Lactobacillus fermentum strain ZT-L3, Bacillus coagulans T11, and Streptococcus thermophilus were included for pooled analysis. An umbrella review found that probiotic intervention could significantly improve LDL-C (MD -7.14, 95% CI -11.03 to -3.24) (moderate-quality evidence), TC (MD -6.93, 95% CI -11.67 to -2.19) (moderate-quality evidence), BUN (MD -1.36, 95% CI -2.20 to -0.52) (moderate-quality evidence), Scr (MD -0.17, 95% CI -0.29 to -0.05) (low-quality evidence), UACR (MD -16.05, 95% CI -27.12 to -4.99) (low-quality evidence), FBG (MD -13.53, 95% CI -19.85 to -7.21) (low-quality evidence), HbA1c (MD -0.12, 95% CI -0.20 to -0.04) (low-quality evidence), and HDL-C (MD 2.72, 95% CI 0.47 to 4.97) (low-quality evidence) in DN patients compared with conventional care without probiotics. However, an umbrella review revealed that probiotic intervention had no significant improvement on the glomerular filtration rate (GFR) (MD 4.51, 95% CI -0.03 to 9.06) (low-quality evidence) in DN patients ( Figure 2 ) ( Table 2 ).

A total of 3 meta-analyses (10, 13, 30) studied the effect of a salt restriction diet on DN incidence. The meta-analysis of Hodson et al. published in 2023 (13) included 12 RCTs describing 400 patients with DN. An umbrella review revealed that, compared with a usual or high-salt diet, a salt restriction diet significantly improved SBP (MD -7.36, 95% CI -10.75 to -3.98) (very low-quality evidence), DBP (MD -3.17, 95% CI -4.58 to -1.76) (very low-quality evidence), CrCl (MD -6.05, 95% CI -10.00 to -2.10) (low-quality evidence), and body weight (MD -1.21, 95% CI -1.73 to -0.68) (very low-quality evidence) in DN patients. However, an umbrella review revealed that a salt restriction diet had no significant improvement on the glomerular filtration rate (GFR) (MD -1.87, 95% CI -5.05 to 1.31) (low-quality evidence) or HbA1c (SMD -0.62, 95% CI -1.49 to 0.26) (very low-quality evidence) in DN patients ( Figure 2 ) ( Table 2 ).

A total of 1 meta-analysis studied the effect of soy isoflavone supplementation on DN incidence. The meta-analysis of Wang et al. published in 2021 (5) included eight RCTs involving 261 patients with DN. An umbrella review revealed that, compared with no supplementation with soy isoflavones, supplementation with soy isoflavones significantly improved BUN (SMD -0.67, 95% CI -0.94 to -0.41) (low-quality evidence), FBG (SMD -0.39, 95% CI -0.68 to -0.10) (low-quality evidence), total cholesterol (TC) (SMD -0.58, 95% CI -0.83 to -0.33) (low-quality evidence), total glucose (TG) (SMD -0.41, 95% CI -0.66 to -0.16) (low-quality evidence), LDL-C (SMD -0.68, 95% CI -0.94 to -0.42) (low-quality evidence) and 24-hour urine protein (SMD -2.58, 95% CI -3.94 to -1.22) (very low-quality evidence) in DN patients. However, an umbrella review revealed that supplementation with soy isoflavones had no significant improvement on body weight (SMD -0.05, 95% CI -0.32 to 0.21; low-quality evidence), Scr (SMD -0.24, 95% CI -0.49 to 0.01; low-quality evidence), CrCl (SMD -0.36, 95% CI -0.83 to 0.10; low-quality evidence), GFR (SMD -0.07, 95% CI -0.35 to 0.20; low-quality evidence) or HDL-C (SMD 0.16, 95% CI -0.09 to 0.41; low-quality evidence) in DN patients ( Figure 2 ) ( Table 2 ).

A total of 4 meta-analyses (4, 22, 24, 28) studied the effect of vitamin D supplementation on DN incidence. The meta-analysis of He et al. published in 2022 (4) included 6 RCTs involving 874 patients with DN. A review of Umbrella medicine showed that, compared with placebo, vitamin D supplementation significantly improved the UACR (SMD -0.24, 95% CI -0.39 to -0.09) (low-quality evidence), UAER (SMD -0.42, 95% CI -0.53 to -0.32) (very low-quality evidence), and 24-hour urine protein (MD -0.26, 95% CI -0.34 to -0.17) (very low-quality evidence) in DN patients. However, an umbrella review of the meta-analysis of Wang et al. published in 2019 (22) revealed that vitamin D supplementation had no significant improvement on Scr (MD -0.83, 95% CI -3.67 to 2.02) (low-quality evidence), GFR (MD 2.13, 95% CI -2.06 to 6.32) (low-quality evidence), HbA1c (MD 0.01, 95% CI -0.09 to 0.11) (moderate-quality evidence), or FBG (MD -0.05, 95% CI -0.29 to 0.20) (low-quality evidence) in DN patients ( Figure 2 ) ( Table 2 ).

In addition, a total of 1 meta-analysis studied the effect of antioxidant vitamin supplementation on DN. The meta-analysis of Chen et al. published in 2020 (9) included 6 RCTs involving 315 patients with DN. An umbrella review revealed that, compared with placebo, supplementation with antioxidant vitamins significantly improved Scr (MD -0.11, 95% CI -0.19 to -0.03) (low-quality evidence), SBP (MD -6.02, 95% CI -9.65 to -2.40) (low-quality evidence), and HbA1c (MD -0.22, 95% CI -0.43 to -0.001) (low-quality evidence) in DN patients. However, an umbrella review revealed that supplementation with antioxidant vitamins did not significantly improve DBP (MD -1.19, 95% CI -3.91 to 1.52) (low-quality evidence) or FBG (MD -1.12, 95% CI -13.24 to 10.99) (low-quality evidence) in DN patients ( Figure 2 ) ( Table 2 ).

A total of 1 meta-analysis studied the effect of CoQ10 supplementation on DN incidence. The meta-analysis of Zhang et al. published in 2019 (7) included 3 RCTs involving 135 patients with DN. An umbrella review revealed that, compared with placebo, supplementation with CoQ10 significantly improved FBG (SMD -2.04, 95% CI -3.90 to -0.18) (very low-quality evidence), HbA1c (MD -1.83, 95% CI -3.39 to -0.27) (low-quality evidence), total cholesterol (TC) (SMD -1.73, 95% CI -3.41 to -0.05) (low-quality evidence), and high-density lipoprotein cholesterol (HDL-C) (MD 0.09, 95% CI 0.01 to 0.18) (low-quality evidence) in DN patients. However, an umbrella review revealed that supplementation with CoQ10 did not significantly improve LDL-C levels (SMD -0.27, 95% CI -0.62 to 0.07) (moderate-quality evidence) in DN patients ( Figure 2 ) ( Table 2 ).

A total of 19 meta-analyses (6, 12, 15, 18, 20, 23, 25–27, 29, 31–39) studied the effect of low-protein diets on DN incidence. The meta-analysis of Jiang et al. published in 2023 (12) included 7 RCTs and included 367 patients with DN. An umbrella review revealed that low-protein diets significantly improved the UAER (standardized mean difference (SMD) of 0.68, 95% CI of 0.08 to 1.29) (very low-quality evidence) in DN patients compared with the usual diet. However, the umbrella review revealed that low-protein diets had no significant improvement on all-cause mortality (RR 0.38, 95% CI 0.10 to 1.44) (low-quality evidence), renal failure (RR 1.16, 95% CI 0.38 to 3.59) (low-quality evidence), GFR (MD -0.73, 95% CI -2.30 to 0.83) (very low-quality evidence), CrCl (MD -2.39, 95% CI -5.87 to 1.08) (very low-quality evidence), or 24-hour urinary albumin excretion (MD 0.00, 95% CI -0.07 to 0.07) (very low-quality evidence) in DN patients ( Figure 2 ) ( Table 2 ).

A total of 1 meta-analysis studied the effect of dietary polyphenol supplementation on DN incidence. The meta-analysis of Macena et al. published in 2022 (14) included 7 RCTs describing 333 patients with DN. An umbrella review revealed that dietary polyphenol supplementation significantly improved HbA1c levels (MD -0.28, 95% CI -0.51 to -0.04) (low-quality evidence), glomerular filtration rate (GFR) (MD 3.66, 95% CI 0.16 to 7.15) (very low-quality evidence) and 24-hour urine protein levels (MD -109.10, 95% CI -216.57 to -1.63) (very low-quality evidence) in DN patients compared with those in patients receiving no polyphenols or placebo ( Figure 2 ) ( Table 2 ).

A total of 1 meta-analysis studied the effect of ketoanalogue supplementation on DN incidence. The meta-analysis of Bellizzi et al. published in 2022 (8) included 7 RCTs of 310 patients with DN. An umbrella review revealed that, compared with the usual diet, ketoanalog supplementation significantly improved 24-hour urine protein (MD -1.41, 95% CI -2.74 to -0.08) (very low-quality evidence) and fasting blood glucose (FBG) (MD -27.57, 95% CI -39.20 to -15.94) (very low-quality evidence) in DN patients. However, an umbrella review revealed that ketoanalogue supplementation had no significant improvement on the glomerular filtration rate (GFR) (MD 4.06, 95% CI -1.84 to 9.97) (very low-quality evidence) in DN patients ( Figure 2 ) ( Table 2 ).

In our study, 74.5% of the outcomes were reanalyzed using a random or fixed effects model. The reanalysis revealed that approximately 36.6% of the examined outcomes exhibited significant heterogeneity (I ² > 50% or Cochran’s Q test P < 0.1). The heterogeneity of most of the outcomes could be attributed to various potential factors, such as study setting, geographical region, ethnicity, sex, age, study quality, sample size, follow-up duration, and adjustment for confounding variables. For the remaining 25.5% of the unanalyzed outcomes, approximately 50% exhibited significant heterogeneity.

In our reanalysis, Egger’s test assessed publication bias for 19.5% of the total outcomes, revealing publication bias in 1 of them. For nonreanalyzed outcomes, publication bias was detected in 35.7% of the outcomes via statistical tests or funnel plots. Importantly, other outcomes either showed no significant publication bias or lacked reported bias assessments.

The median AMSTAR score for all outcomes was 8 (8-11), and further detailed AMSTAR scores specific to each outcome can be found in Supplementary Table S1 . For the GRADE, five outcomes (change in BUN (probiotics), change in TC (probiotics), change in LDL-C (probiotics), change in HbA1c (vitamin D), change in LDL-C (CoQ10)) were downgraded to “moderate” quality given the imprecision, and the remaining outcomes were downgraded to “low” or “very low” due to the risk of bias, inconsistency, indirectness, or imprecision. Supplementary Table S2 shows the detailed GRADE classification for each outcome. In terms of evidence, all outcomes were classified as IV or NS (nonsignificant) because of the small sample size.

Relevant studies have shown that the incidence of DN is increasing rapidly, and patients with DN accounted for 20% to 40% of type 2 diabetes patients in the community from 2009 to 2012 (54). It is not only the main cause of death in type 1 diabetes patients but also an important factor threatening the health of type 2 diabetes patients (55). At present, it is believed that the disease progression of DN is difficult to reverse, the risk factors involved in the progression of DN cannot be identified, and effective measures cannot be taken to delay the progression of disease to end-stage nephropathy (56). With the increase in the number of DN patients, the disease burden on society and families will also increase (57). In recent years, due to the deepening of basic research, the treatment of DN has taken a new direction. Several scholars have proposed that probiotics may improve and prevent metabolic diseases such as DN through changes in the human intestinal flora (58). In addition, some animal model studies have shown that soy foods can prevent kidney disease and delay the deterioration of kidney function (59, 60). Giving soy foods instead of meat to DN patients can improve kidney function (61, 62). Furthermore, a large number of animal and cellular experiments and clinical studies have shown that active vitamin D has a renoprotective effect and may play a role in inhibiting the inflammatory response, antioxidative stress, and renal fibrosis; inhibiting the renin-angiotensin system; and improving insulin resistance (4, 22, 24, 28).

To date, a large number of researchers worldwide have carried out clinical research and evidence-based medical research on the effects of dietary intervention on DN. This umbrella evaluation evaluated the advantages and disadvantages of existing evidence-based medical methods from systematic reviews and meta-analyses on the effects of dietary intervention on DN, helped us to understand the potential effective dietary management strategies for the prevention and treatment of DN in a more comprehensive way from multiple dimensions, provided a theoretical basis for the development of more clinically effective prevention and control measures for DN, and provided directions for further clinical research.

The present umbrella review extracted 9 unique dietary interventions (including probiotics, a salt restriction diet, vitamin D, soy isoflavone, CoQ10, ketoanalog, dietary polyphenols, antioxidant vitamins, and low-protein diets) and 55 corresponding outcomes in meta-analyses, including 34 significantly associated outcomes and 21 nonsignificantly associated outcomes. All outcomes were classified as IV or NS (nonsignificant), and only five dietary interventions were rated as moderate-quality evidence.

First, compared with conventional care without probiotics, probiotic intervention significantly improved LDL-C (moderate-quality evidence), TC (moderate-quality evidence), BUN (moderate-quality evidence), Scr (low-quality evidence), UACR (low-quality evidence), FBG (low-quality evidence), HbA1c (low-quality evidence), and HDL-C (low-quality evidence) in DN patients. He et al. (63) reported that probiotic supplementation can reduce the abundance of conditioned pathogenic bacteria, increase the abundance of beneficial intestinal bacteria, and reduce the release of enterogenic endotoxin, thus effectively improving blood sugar and blood lipid levels and kidney function. In recent years, an increasing number of studies have shown that inflammatory factors play a certain role in the pathogenesis of DN. Inflammation in DN patients is characterized by increased expression of inflammatory factors, inflammatory chemokines and adhesion factors; inflammatory cell infiltration; and increased CRP levels. Compared with that of classical inflammation, the severity of DN is mild, and DN is associated with a state of microinflammation (64). Firouzi et al. (65) showed that probiotic supplementation could reduce the content of enteric-borne urotoxins (such as para-cresol and indoxyl sulfate) in the blood of DN patients, inhibit the microinflammatory state of the whole body, and delay the deterioration of renal function. Proteinuria and changes in glomerular filtration membrane permeability in DN patients are closely related to vascular endothelial injury caused by oxidative stress, and DN patients often exhibit damage to the antioxidant defense system and an increase in free radical products. Probiotics can exert antioxidant effects through their own antioxidant system, such as regulating signaling pathways to produce various metabolites with antioxidant activity, such as glutathione (66).

Second, we found that compared with the usual or high-salt diet, the salt restriction diet significantly improved SBP (very low-quality evidence), DBP (very low-quality evidence), CrCl (low-quality evidence), and body weight (very low-quality evidence) in DN patients. High salt intake leads to elevated blood pressure caused by high sodium intake, which increases the risk of cardiovascular events in patients with DN. People with DN can lower their blood pressure by restricting salt, and in both type 1 and type 2 diabetes, salt restriction lasting 1 week leads to lower blood pressure (7.11/3.13 mmHg in type 1 diabetes patients and 6.90/2.87 mmHg in type 2 diabetes patients) (67). Current nutritional guidelines for patients with DN consistently recommend limiting dietary sodium intake to < 1.5 to 2.3 g/d (5 g NaCl). However, too low of a sodium intake may reduce insulin sensitivity and is not conducive to glucose homeostasis (68).

Third, the present umbrella review showed that supplementation with soy isoflavones significantly improved BUN (low-quality evidence), FBG (low-quality evidence), total cholesterol (TC) (low-quality evidence), LDL-C (low-quality evidence) and 24-hour urine protein (very low-quality evidence) in DN patients compared with no supplementation with soy isoflavones. Studies have shown that soy foods can regulate blood lipid metabolism in the body to reduce low-density lipoprotein levels and increase high-density lipoprotein levels. Moreover, plant sterols contained in soybeans can competitively inhibit the body’s cholesterol synthesis and reduce serum cholesterol levels (69). To improve kidney function, soy foods can reduce 24-h urinary protein levels. Replacing animal protein with a portion of soy protein in the diet does not adversely affect kidney function but also improves kidney hemodynamic function and reduces the elimination of urinary protein (5). Soybean protein itself is a high-quality protein and has a relatively high raw price. After the digestibility of soybean food is significantly improved, soybean protein and animal protein play the same nutritional role. Moreover, soy protein is lower in fat than animal protein is, which helps people with diabetes control the total calories in their diet and reduce the intake of too much fat, especially saturated fat, due to the consumption of human animal protein (69). More importantly, the unique nutrients of soy protein contribute to the stability of blood sugar and blood lipids in diabetic patients and can also remove excess free radicals in diabetic patients, reduce oxidative stress in the body, reduce the attack of glycoylation end products on the body’s target organs, and prevent complications (70).

However, the effect of a low-protein diet on DN has been controversial. The basic principle of low-protein diet therapy is to reverse glomerular filtration and reduce uremic symptoms. Studies on patients with chronic kidney disease and advanced DN have shown that a low-protein diet can lead to malnutrition, which is a risk factor for mortality from this disease (71). Therefore, the beneficial effect of a low-protein diet on renal prognosis may be offset by the malnutrition of the treatment itself, and more importantly, a low-protein diet may increase the mortality of DN patients (72). The results of this study showed that a low-protein diet was not significantly associated with improved kidney function in patients with DN. Although these results do not completely negate other potential benefits of a low-protein diet for DN patients, the benefits of a low-protein diet on renal function are not significant (71). Urinary tract infection is also one of the common complications in patients with DN (73, 74). However, the existing studies have not reported a significantly effective dietary intervention that can reduce the risk of urinary tract infection in patients with DN. The study by Chen et al. (73) found that vegetarianism was a protective factor for urinary tract infections, but the protective effect was not significant in the subgroup of patients with diabetes. In addition, Zaragoza-Marti et al. (75) believe that the Mediterranean diet can significantly reduce the risk of gestational diabetes and urinary tract infections, but there are no data on the effect of the Mediterranean diet on the development of urinary tract infections in diabetic patients. In addition, this study revealed that nutritional supplements such as CoQ10, dietary polyphenols and ketoanalog can effectively improve the clinical outcomes of DN patients, but the quality of evidence is low.