Unraveling the Role of Tumor Necrosis Factor-Alpha in Diabetic Peripheral Neuropathy: A Systematic Review and Meta-Analysis

Diabetic peripheral neuropathy (DPN) is a prevalent and debilitating complication of diabetes mellitus, leading to sensory abnormalities, decreased balance, and increased risk of foot problems. Although tumor necrosis factor-alpha (TNF-α) has emerged as a potential factor in the pathogenesis of DPN, its role remains contested. This study intends to thoroughly analyze the association between TNF-α and DPN by combining data from various global studies. This systematic review and meta-analysis adhered to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines and included 23 articles investigating TNF-α levels in DPN patients for systematic review and 11 articles for meta-analysis. Data were extracted, and heterogeneity was examined. A random-effect model was chosen due to high heterogeneity. The major outcome measure across studies was serum TNF-α levels. The meta-analysis found a significant mean difference of 15.2464 (95% confidence interval = 4.4963; 25.9965) under the random-effect model due to the substantial heterogeneity (I2 = 98.1%) among included studies. The meta-analysis indicates a consistent elevation in TNF-α levels in individuals with DPN compared to those without neuropathy. This underlines the potential of TNF-α as a biomarker and contributor to diabetic neuropathy. Despite heterogeneity, the study’s extensive scope and systematic approach enhance the trustworthiness and generalizability of the findings.


Introduction And Background
Diabetic peripheral neuropathy (DPN) is a prevalent and often under-recognized consequence of diabetes mellitus.It often presents as sensory abnormalities, including numbness, lack of sensation, and decreased balance, which can greatly reduce a patient's quality of life.Moreover, DPN raises the risk of foot ulcers and, in extreme situations, may lead to amputation [1].In India alone, a survey conducted in 2020 indicated that 26-31% of diabetes patients get DPN, with a rising frequency of amputations linked to the illness [2].The American Diabetes Association has released recommendations for diabetic foot care to address these concerns [3].
In the convoluted pathophysiology of DPN, there is a rising interest in the role of tumor necrosis factoralpha (TNF-α), a versatile cytokine with significance in inflammatory responses.While factors such as glucotoxicity and the production of advanced glycated end products have been implicated in the neurotoxic pathogenesis of diabetes, the significance of TNF-α in neurotoxic consequences cannot be overlooked [4].TNF-α, previously identified for its immunological and inflammatory effects, is now being recognized to play a major role in the complicated pathways contributing to neuropathic problems in diabetes [5].Several inflammatory, free radicals, antioxidant, coagulation, and lipid peroxidation markers have been studied concerning DPN but have not proven helpful as standalone biomarkers.Numerous studies have investigated the potential link between higher TNF-α levels and the development and progression of DPN [6][7][8][9].Inflammation, a hallmark of diabetes-induced problems, is strongly linked with the neurodegenerative processes identified in DPN.TNF-α, as a proinflammatory cytokine, is implicated in the beginning and persistence of inflammatory cascades that lead to nerve injury, demyelination, and decreased nerve conduction [5].
The significance of TNF-α in DPN is a subject of substantial study and controversy, with some studies proposing it as a potential marker for early diagnosis and others exploring its therapeutic potential as a target for intervention.However, the existing literature on TNF-α and DPN shows various findings and interpretations, leading to inconsistent conclusions.To address this diversity and provide a thorough perspective, this systematic review and meta-analysis aims to synthesize and critically examine studies of various study designs to present a comprehensive overview of the connection between TNF-α and DPN.

Review Methodology
The systematic review followed the guidelines outlined by Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA), a framework designed to facilitate the reporting of systematic reviews and meta-analyses [10].The protocol for the review was registered on PROSPERO (registration ID: CRD42023441858) before the commencement of the study.

Selection Procedure
The systematic review aimed to investigate the role of TNF-α in DPN.Eligible study designs included studies exploring the role of TNF-α in the development, progression, or management of DPN, with no specific restrictions on study types.The participants of the included study were individuals with DPN.The exposure under scrutiny was TNF-α levels, with a comparator group consisting of individuals with diabetes mellitus without neuropathy or healthy volunteers.The study context encompassed settings relevant to understanding the implications of TNF-α in DPN.The primary outcomes of the included study involved assessing neuropathy severity through nerve conduction studies or neurological examination scores.

Search Strategy
We conducted a systematic search of the literature on PubMed, Scopus, Google Scholar, and CINAHL from their inception until July 2023.Furthermore, we delved into the bibliographies of relevant articles for additional insights.Our inclusion criteria focused on studies published exclusively in the English language, employing the subsequent search strategy depicted in Table 1.

Screening and Data Analysis
Studies identified underwent initial screening based on their titles and abstracts by two independent authors.Those deemed potentially relevant underwent further scrutiny, with full texts reviewed by two independent authors using the open-access online tool CADIMA version 2.2.3 from the Julius Kühn Institute in Quedlinburg, Germany.Any disagreements were resolved through discussion or consultation with a third author.Data extraction from selected studies was performed independently by two authors using a standardized form, with discrepancies resolved through consensus or consultation with another author.The extracted data encompassed participant demographics, sample characteristics, exposure details, and study specifics, all recorded in an Excel spreadsheet.R statistical software was employed for data analysis, with mean difference (MD) and 95% confidence intervals (CIs) expressed for continuous variables using inverse variance methods.Heterogeneity among studies was assessed through Cochran's Q and I² statistics, guiding the choice between fixed-effect or random-effect models based on heterogeneity.

Quality Assessment
Quality assessment and identification of biases in the reviewed studies were performed using the critical appraisal checklist for cross-sectional studies by the Joanna Briggs Institute [11], and the quality assessment graph (Figure 1) and summary (Figure 2) was generated using RevMan 5.4 from the Cochrane Collaboration.

Results
The search strategy initially identified 1,066 articles, distributed across PubMed (284), Scopus (694), Google Scholar (38), and CINAHL (50).After eliminating duplicates and consolidating results, a total of 818 articles were identified.The PRISMA flowchart in Figure 3 outlines the search process.Subsequent screening of titles and abstracts narrowed down the selection to 63 articles for full-text screening.Nine reports were unretrievable, and 31 were excluded for reasons detailed in Figure 3. Ultimately, 23 articles were included in the systematic review, with 11 featuring TNF-α expressed as MD selected for meta-analysis.The studies, conducted in 13 countries globally, demonstrated true diversity, primarily focusing on DPN patients, with sample sizes ranging from 50 to 483 patients.The main outcome measure across studies was serum TNF-α levels.The results from the meta-analysis of 11 included studies revealed varying MDs and corresponding 95% CIs across different studies (Table 5).Notably, the forest plot (Figure 4) illustrated substantial heterogeneity among the studies, as indicated by a high I 2 value of 98.1%.The heterogeneity was assessed by tau2 and tau, which were 324.5915 and 18.0164, respectively.These results, coupled with the Q-profile method, revealed the necessity for a more conservative strategy to accommodate the heterogeneity among the included studies.inverse variance method, restricted maximum-likelihood estimator for tau2, Q-profile method for confidence interval of tau2 and tau.

MD
Under the common-effect model, the pooled MD was 0.8210 (95% CI = 0.6843; 0.9576), with a z-value of 11.77 and a p-value less than 0.0001.The random-effect model showed a higher mean difference of 15.2464 (95% CI =4.4963; 25.9965), with a z-value of 2.78 and a p-value of 0.0054.The use of the random-effect model is justified due to the significant heterogeneity observed among the studies.
The forest plot provides a visual depiction of the effect sizes of individual studies, coupled with the combined effect size from the meta-analysis (Figure 4).It is obvious from the plot that the overall effect is positive, demonstrating increased TNF-α levels in DPN patients compared to diabetic individuals without neuropathy.
The sensitivity analysis was performed by eliminating one study at a time and estimating the overall MD for both fixed-effect and random-effect models.The overall effect sizes were consistent across varied exclusions, showing the robustness of the meta-analysis results.No single study appeared to have a disproportionate influence on the total effect size.

Discussion
We performed this systematic review and meta-analysis to assess the association between TNF-α and DPN risk in diabetic patients.We found 23 studies where serum TNP-α levels were studied in diabetic patients with peripheral neuropathy compared to patients without neuropathy.Among them, 19 studies showed an increase in TNF-α levels, and four studies showed no significant change in serum TNP-α levels in patients with DPN.
DPN is the most predominant neuropathic syndrome in diabetic mellitus patients.It can lead to severe neuropathic pain, ulceration, and lower limb amputation.The exact mechanism for the development of DPN is not fully understood but may include derangement in pathways related to hyperglycemia, dyslipidemia, microvascular complications caused by oxidative stress, neuronal inflammation, mitochondrial damage, and cell death [4].Chronic low-grade inflammation is the major determinant for the development of DPN.There occurs activation of innate immune response, alteration in the insulin signaling pathway, insulin resistance, and mitochondrial and endoplasmic reticulum stress due to chronic low-grade inflammation.All these lead to the production of various kinases such as protein kinase C, jun-N terminal kinase, mitogen-activated protein kinase, and the release of various proinflammatory cytokines and interleukins such as 1b, 2, 6, 8, and TNP-α which play a vital role in the pathophysiology of DPN [32].
Numerous markers are used to predict the risk of developing DPN in patients with diabetes.These include macrophage-colony stimulating factor (CSF-1), monocyte chemoattractant protein-1 (MCP-1), vascular cell adhesion molecule-1, intercellular adhesion molecule-1 (ICAM-1), TNP-α, C-reactive protein (CRP), and Eselectin [33].There is a significant alteration in innate immunity in diabetes patients.Toll-like receptor 4 (TLR4) receptors play an important role in recognizing the pathogen and initiating the immune response.Two significant pathways, namely, mitogen-activated protein kinase (MAPK) and nuclear factor kappa B (NF-κB), are activated by TLR4.They lead to the release of TNF-α and IL-6 and play a vital role in the regulation of neuronal plasticity [30].
A proinflammatory cytokine, TNP-α produced by monocytes and activated macrophages contributes significantly to the DPN pathophysiology.Patients with prolonged hyperglycemia have dysregulated TNF-α levels.In DPN patients, myelinated nerves show demyelination while the unmyelinated ones suffer axonal loss.T cells are activated due to immune response which produces TNP-α, mediating the inflammatory reactions.It causes oligodendrocyte toxicity leading to demyelination.The secretions of other cytokines such as IL-1β and IL-6 are increased from endothelial cells and monocytes that adds to damage in neurons.
It also leads to hemodynamic abnormalities, endothelial dysfunctions and increased expression of various cell adhesion molecules in blood vessels, and decreased nitric oxide synthase activity in endothelial cells [9].
Anti-TNF-α medications may be evaluated as a viable method for controlling DPN.In DPN rats, the suppression of the TNF-α pathway utilizing a recombinant human TNF-α receptor-antibody fusion protein (rhTNFR:Fc) demonstrated recovery from nerve injury [34].Other drugs such as infliximab, etanercept, sesamol, and rimonabant, which can either decrease or modify TNF-α levels, have shown promising benefits in DPN rats [35].However, further study is necessary to explore the impact of anti-TNF-α therapy in DPN patients.
While this meta-analysis provides valuable insights, it is important to note some limitations, with the primary one being the heterogeneity among the studies.Differences in population demographics, study design, and diagnostic methods may impact the generalizability of our findings.Additional constraints include the fact that there are merely three prospective cohort studies examining the correlation between serum TNF-α levels and DPN in patients with diabetes mellitus.The majority of studies included in this meta-analysis were either case-control or cross-sectional studies which might lead to bias to a certain extent as all the confounding factors were not taken into consideration.Thus, there is a need for more prospective cohort studies to get a more accurate relation between serum TNF-α levels and the risk of DPN in diabetes patients.Another major limitation of this meta-analysis is that there is only one study that evaluated the association between serum TNF-α levels and the risk of DPN in type 1 diabetes mellitus.

Conclusions
This systematic review and meta-analysis of diverse studies suggests a substantial relationship between higher TNF-α levels and DPN.The findings underline the potential function of TNF-α as a biomarker and contributor to diabetic neuropathy.Despite study variability, the systematic review provides a complete overview, emphasizing the need for more research to explain the processes of TNF-α in DPN and its therapeutic implications.

FIGURE 1 :
FIGURE 1: Quality assessment graph of the included studies using the critical appraisal checklist for cross-sectional studies by the Joanna Briggs Institute.

FIGURE 2 :
FIGURE 2: Quality assessment summary of the included studies using the critical appraisal checklist for cross-sectional studies by the Joanna Briggs Institute.

FIGURE 3 :
FIGURE 3: Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) flow diagram.

TABLE 1 : Search strategy.
CINAHL: Cumulative Index to Nursing and Allied Health Literature

Table 2
presents a complete overview of studies examining DPN in varied groups.Studies ranged throughout the United States, Turkey, Egypt, Greece, China, Germany, India, Japan, Hungary, Denmark, Mexico, and Indonesia, and involved both type 2 and type 1 diabetes patients, prediabetic individuals, and healthy controls.Duration of diabetes varied, with major variances in the measures adopted for assessment.Common measures included the Neuropathy Symptom Score (NSS), Neuropathy Disability Score (NDS), nerve conduction investigations, clinical examinations, and specialized scoring systems such as the Toronto Clinical Scoring System (TCSS) and Michigan Neuropathy Screening Instrument (MNSI).

Country Population Duration of diabetes Measures of DPN
China T2DM patients with and without DPN 5.33 ± 1.79 years in the non-DPN group, 7.34 ± 1.99 years in the DPN group Symptoms, physical examination, electrophysiology 2023 Sen et al.Cureus 15(12): e49926.DOI 10.7759/cureus.49926Zeng et al., 2018 [27] China Patients with prediabetes, diabetes and healthy controls Mean 1.24 years in diabetes group NSS, NDS, and NCS Not reported for the whole cohort, 15.04 ± 10.26 years in the DPN group vs. 12.35 ± 8.97 years in the non-DPN group

TABLE 2 : Basic characteristics of the included studies.
*: Data are presented as mean ± SD or median (range).threshold; DN4: Douleur Neuropathique 4 questionnaire; KORA: Cooperative Health Research in the Region of Augsburg (German study); PAD: peripheral artery disease

Table 3
provides a complete summary of sample sizes, mean ages, gender distributions, and study designs.The study designs ranged from cross-sectional, case-control, to longitudinal, demonstrating the multidimensional approach to understanding DPN across different groups and circumstances.years in the diabetic group, 54.5 years in the alcoholic group, and 49.1 years in controls 21 M/8 F in the diabetic group, and 23 M/8 F in the alcoholic group A case-control study Nádró et al., 2021 [21] 54 DPN, 24 Non-DPN 64.1 years in the DPN group, and 63.58 years in the non-DPN group 22 M, 32 F in the neuropathy group; 11 M, and 13 F in the control group 2023 Sen et al.Cureus 15(12): e49926.DOI 10.7759/cureus.49926