Neuropathies encompass a variety of conditions characterized by nerve damage that arises from multiple etiologies. These conditions and the extent to which they impact different neural pathways result in a range of dysfunctions in nerve transmission, leading to diverse health consequences for patients [1].

Among the spectrum of neuropathies, peripheral neuropathies are the most prevalent, affecting approximately 2.4% of the global population. This prevalence increases to up to 8% among the elderly [2]. Peripheral neuropathies comprise disorders of the peripheral neurons and fibers, manifesting from a broad spectrum of underlying causes [3]. These conditions can be categorized based on the extent of injury (mononeuropathies, multifocal neuropathies, polyneuropathies), the nature of the neuropathy (compressive or non-compressive, axonal, or demyelinating), and the disease progression (chronic or acute) [4]. Various factors, including genetic predispositions, nerve compression or injury (as in carpal tunnel syndrome), exposure to toxins (e.g., methanol, alcohol), nutritional deficiencies, and diabetes mellitus, can trigger these neuropathies [5]. Notably, peripheral neuropathy remains idiopathic in 25% to 46% of cases [6].

Translating pathology into symptomatology, many of the signs depend on the extent and type of neuropathy, ranging from motor and cognitive alterations to sensory dysfunction and painful neuropathies.

Neuropathic pain is defined by the International Association for the Study of Pain (IASP) [7] as pain caused by a lesion or disease of the somatosensory nervous system. Most of the time these lesions are chronic and peripheral sensitization magnifies the condition. Peripheral sensitization refers to an increase in the magnitude of responsiveness at the peripheral ends of sensory nerve fibers to any stimulus. This results from the release of a chemical mediator by nociceptors and other non-neuronal cells such as immune cells, endothelial cells, keratinocytes, and fibroblasts at the site of tissue injury or inflammation [8], thereby resulting in increased nociceptor excitability through changes in protein phosphorylation and activity, contributing to both acute and long-term pain sensitivity [9, 10, 11, 12].

Neuropathic pain significantly diminishes patients’ quality of life and is characterized by persistent or intermittent spontaneous pain, development of allodynia, and hyperalgesia [1]. Despite extensive characterization, many aspects of the underlying mechanisms of neuropathic pain remain elusive, underscoring the need for further research. Moreover, the process of peripheral sensitization, crucial for the diagnosis, treatment, and progression of neuropathy beyond clinical signs, highlights the importance of identifying biomarkers. Biomarkers, including proteins, metabolites, and genetic polymorphisms, offer pathophysiological insights into the condition, exhibit minimal variability, and change in expression in response to disease, treatment, or pathological states. Once validated, biomarkers enable the prediction of disease outcomes or responses to interventions, facilitating disorder screening, diagnosis, and personalized therapeutic strategies [13].

This review therefore investigated inflammatory mediators in blood and/or cerebrospinal fluid to identify potential differential and prognostic biomarkers of neuropathic pain among patients with varying presentations of neuropathy.

There is compelling evidence linking glial activation, immune response, and the development of chronic pain in animal models, yet translating these findings to human conditions remains an intricate task [41]. The search for a convincing pain biomarker continues to be a formidable challenge for the scientific community, yet the need to offer potential breakthroughs in understanding and treating patient conditions continues.

The immune system’s role in neuropathies, particularly within inflammatory and autoimmune contexts, is significant [42]. Hence, improper or excessive activation can lead to the infiltration of immune cells, such as lymphocytes and macrophages, into peripheral nerves, causing damage and dysfunction. Microglial shift from a surveillance state to a reactive state [43] leads to a secretion of inflammatory molecules, such as cytokines and chemokines, that trigger an inflammatory response and also contribute to nerve damage. Moreover, these molecules contribute to the recruitment and activation of more immune cells at the site of injury, amplifying the inflammatory response and perpetuating the damage [44]. Additionally, another key factor in the pathogenesis and progression of these conditions may be the peripheral leukocyte infiltration and cytotoxic enzyme secretion that also take place.

Although neuropathic pain is a complex and multifactorial condition, several immune mechanisms, some of which have already been mentioned, may contribute to the generation and maintenance of pain. One of them is the sensitization of the lesion site causing pain thresholds to decline significantly. Conversely, chronification of an injury may lead to a reactive state of microglia responding against nerve endings, leading to neuropathic pain [42, 43, 44, 45] and increased permeability of inflammatory molecules at the blood-brain barrier [46]. In the case of neuropathies that can be painful or non-painful, the pathogenesis is multifactorial and complex, hence the discovery of accurate biomarkers is required to achieve individualized treatments for each condition and each patient.

Of the 68 biomarkers identified in our systematic review, 12 could be used for comparisons as they were the only ones common across several articles, and only three of these were related to inflammation. This outcome revealed an absence of significant results in the comparison of painful and painless neuropathies, which could stem from the challenges of standardizing analysis conditions across studies, even when measuring the same protein. This highlights one of the significant limitations of our work, further compounded by our selection of studies reporting data solely as mean $\pm$ SD or mean $\pm$ SEM, excluding those utilizing proximity extension analysis for proteomics. Such studies report proteins in terms of normalized expression and employ multivariate analysis techniques such as principal component analysis and hierarchical clustering, providing an alternative insight into the proteomic landscape of neuropathic pain. By not incorporating these studies, our analysis may have overlooked crucial biomarkers that are implicated in low-grade inflammation associated with neuropathic pain. This oversight underscores the need for a more inclusive methodological approach in future research, aiming to bridge the gap between different proteomic data types. Nevertheless, the results we have presented for cytokines remain relevant for understanding painful neuropathies, despite the methodological constraints.

Among the 12 studies reviewed, none included data regarding cerebrospinal fluid samples. This scarcity of results does not reflect a lack of importance of biomarkers in cerebrospinal fluid (CSF), as they may be of great interest amid the invasiveness of sample collection, but is probably a result of the rigid inclusion criteria and the huge heterogeneity of the studies. Significantly, studies measuring proteins in blood and those measuring CSF biomarkers were not mixed and analyzed in the same comparison and, considering the limited number of the latter, only a standardized pool from blood was available for analyses. This highlights a critical limitation of our meta-analysis, which stems from the inherent heterogeneity among the included studies and underscores the potential benefits of well-conducted individual studies that might provide more definitive guidance. Despite this, meta-analysis remains a valuable tool in our field, offering a comprehensive overview that individual studies cannot provide alone. Among the studies determining biomarkers in blood, most of the articles highlighted the role of TNF-alpha, IL-6, C-reactive protein, and IL-1beta in inflammatory diseases, provoking our further interest in them. Nevertheless, the sample size was small throughout the whole study, and probably contributed to the absence of significant results. Nevertheless, the role of these molecules in the aforementioned pathologies must be taken into consideration, opening the way for future research. An increase in IL-6 and TNF-alpha levels has been shown to occur in inflammatory-like diseases [41, 45, 46, 47]. Briefly, both mediators have been found to be up-regulated after nerve injury, inflammatory processes, chronic exposure to various drugs, and other conditions leading to inflammatory reactions; hence, this increased secretome possibly plays an important role in the development and maintenance of pain.

IL-6 is a multifunctional cytokine that is a key player in inflammatory regulation. After a traumatic event in the nervous system, the levels of this interleukin rise, leading to an upregulation of astrocytic activity as a part of the activation process, shifting to a reactive state. In addition, it has been proposed that IL-6 may have a modulatory effect on neuronal excitability in the central and peripheral nervous systems. This could influence the perception of symptoms in neuropathies. Besides, IL-6 can increase the excitability of sensory neurons and modulate the transmission of nerve signals [45], which can translate into an increase in pain severity, as suggested by a large number of studies [26, 29, 36, 37, 48].

Moreover, IL-6 and TNF-alpha level increases were correlated with the intensity of allodynia in a study comparing painful and painless polyneuropathy [49]. Given the difficulty of finding studies comparing these two sample groups, samples of patients with chronic pain vs healthy controls have also been examined, showing increases in these cytokines in subjects suffering painful conditions [48].

TNF-alpha is an inflammatory cytokine that plays an important role in various functions of the immune system and is involved in several physiological and pathological processes. In the context of neuropathies, TNF-alpha has been the subject of research due to its possible contribution to both painful and non-painful neuropathies. It has been shown to stimulate astrocytes in the release of chemoattractant factors leading to a chronification of the inflammatory process [50]. In animal models, supplementation with this mediator has been proven to be correlated with the development of neuropathies and subsequent nociceptive processes [51]. Further, several authors have described a significant increase in both TNF-alpha and its microRNA in rodent neuropathic pain models [27, 36, 37, 40, 49]. As a result of this, the search for microRNA-type biomarkers is becoming increasingly widespread. An article comparing serum levels of markers such as TNF-alpha and other interleukins with their respective microRNA levels was included in the present metanalysis, although no significant differences were found [40].

MicroRNAs are short RNA sequences that do not code for proteins but are instead responsible for regulating gene expression following gene transcription. They often modulate the expression or repression of messenger RNA (mRNA), thus raising the possibility that they are biomarkers of evolution or the pathogenesis of a disease [52]. As such, several studies suggest a positive up-regulation of specific microRNAs in patients with painful neuropathies compared with non-painful diabetic conditions [53, 54]. Specifically, in the field of neuropathic pain, Tramullas et al. [55] detected a microRNA in an animal model that could help predict the development of neuropathy in patients by its determination in plasma and cerebrospinal fluid samples.

Finally, the scarcity of data to be analyzed led us to consider not including neuropathies that were not subtyped by specific pathology. If that were the case, the present work would have been missing a second analysis by subgroup, which would have shed light on whether each pathology should be examined with respect to a specific and different pain biomarker. This premise is held by some authors regarding CRP [27, 28]. This protein is produced by the liver in response to inflammation and activation of the immune system. In the context of both painful and painless neuropathies, the role of CRP has been investigated as a marker of inflammation and as a possible indicator of disease severity or prognosis [56, 57, 58]. However, the direct involvement of CRP in neuropathies and its relationship with neuropathic pain is not yet fully understood and remains under investigation.

Furthermore, in some painful neuropathies such as diabetic neuropathy, several authors have described an elevation of CRP levels, suggesting the presence of an inflammatory component. Inflammation may play a role in the generation and maintenance of neuropathic pain in these conditions. Despite our results not confirming any correlations with CRP, some authors found elevated CRP in type 1 [59] and type 2 [60] diabetes mellitus. It is important to note that the pathogenesis of type 2 diabetes mellitus is considerably heterogeneous and both glycation stress and inflammation-related processes could be cooperatively driving the development of late complications [28].

Finally, the information yielded by the funnel plots is essential to fully understand publication bias. Funnel plots are graphs used in meta-analyses and systematic reviews to visualize publication bias or any other type of bias. These graphs allow the assessment of potential asymmetry in the data, which would indicate the presence of such bias. In the context of a systematic review of pain markers in patients with painful neuropathies, a symmetric funnel plot indicates that the studies included in the review are not affected by publication bias or other types of bias.

Our comprehensive review highlights the ongoing gaps in the understanding of neuropathic pain and the need for identifying effective biomarkers for diagnosis and treatment. Despite thorough searches across six databases and analysis of 1403 subjects, our findings show no significant differences in the levels of CRP, IL-6, and TNF-alpha between patients with and without neuropathic pain. This result underscores the complexity of neuropathic pain and the challenges in using these inflammatory biomarkers for diagnosis or treatment. The high inter-rater reliability and adequate bias assessment strengthen the validity of our study, but the observed publication bias and study heterogeneity necessitate caution in interpreting our findings. Consequently, our review emphasizes the urgent need for more targeted research to unravel the underlying mechanisms of neuropathic pain and to identify more definitive biomarkers in blood and cerebrospinal fluid. This could ultimately lead to improved patient outcomes through better diagnostic tools and personalized treatment approaches.

Conceptualization, MM-A, MS-G and CG; methodology, MM-A, MS-G, DP, and CR-R; validation, MM-A, MS-G and CR-R; investigation, MM-A and MS-G; resources, CG, MM-A and MS-G; data curation, MM-A, MS-G and CR-R; writing—original draft preparation, MS-G; writing—review and editing, MM-A, MS-G, DP and CR-R; supervision, MM-A, CG, DP and CR-R; project administration, MS-G. All authors contributed to editorial changes in the manuscript. All authors read and approved the final manuscript. All authors have participated sufficiently in the work and agreed to be accountable for all aspects of the work.

Not applicable.

Not applicable.

This research received no external funding.

The authors declare no conflict of interest.