Investigating the neurological function and anti-inflammatory effects of resveratrol in small mammal spinal cord injury models: A systematic review and meta-analysis

Abstract

Background:

Spinal cord injury (SCI) produces serious neurological and medical problems. Resveratrol (RES), a natural polyphenol, possesses anti-inflammatory and antioxidant characteristics and has revealed neuroprotective ability in central nervous system diseases. This study investigated the therapeutic potential of RES in rat, mouse, and rabbit SCI models.

Methods:

Articles were identified from Medline, Embase, Scopus, and Web of Science until April 25, 2025. Studies evaluating RES with SCI models were included. Main results were locomotor function, inflammation, oxidative stress, apoptosis, and neuronal survival. Random-effects meta-analysis showed standardized mean differences (SMDs) and their p-values. SYRCLE’s risk of bias tool and I² statistic were used for study quality and heterogeneity, respectively.

Results:

Twenty-six studies were included. RES improved locomotor function (SMD = 3.41, p < 0.0001). RES was associated with anti-inflammatory signaling, including decreased tumor necrosis factor-alpha (SMD = −4.64, p < 0.0001), interleukin (IL)-6 (SMD = −3.19, p < 0.0001), and IL-1β (SMD = −4.10, p < 0.0001). RES was associated with antioxidant signaling, including increased superoxide dismutase (SMD = 4.74, p < 0.0001) and glutathione (SMD = 3.21, p < 0.0001) and decreased malondialdehyde (SMD = −2.41, p < 0.0001). RES had significant effects on apoptotic signaling, including increased B-cell Lymphoma 2 (Bcl-2) (SMD = 3.55, p < 0.0001) and decreased Bcl-2-associated X protein (Bax) (SMD = −3.71, p = 0.010), Bax/Bcl-2 ratio (SMD = −6.83, p < 0.0001), and cleaved caspase-3 (SMD = −4.39, p < 0.0001), and terminal deoxynucleotidyl transferase dUTP nick end labeling (SMD = −3.81, p < 0.0001). RES treatment was associated with improved neuronal survival markers, including Nissl (SMD = 3.39, p < 0.0001), neuronal nuclei (SMD = 2.00, p < 0.0001), and growth-associated protein 43 (SMD = 3.13, p < 0.0001).

Conclusion:

In preclinical SCI models, RES is a promising neuroprotective agent. This suggests that further research is warranted regarding the potential role of RES in SCI treatment.

INTRODUCTION

Spinal cord injury (SCI) is a major cause of disability worldwide, resulting from primary mechanical trauma followed by secondary processes such as inflammation, oxidative stress, and apoptosis, yet no definitive therapy currently exists to promote nerve regeneration or locomotor recovery.[6,12,16] Resveratrol (RES), a natural polyphenolic compound with anti-inflammatory, antioxidant, and anti-apoptotic properties, has shown promise in preclinical SCI models for improving locomotor outcomes, although its precise mechanisms remain unclear.[8,28] This systematic review and meta-analysis aims to evaluate the effects of RES on inflammation, oxidative stress, apoptosis, autophagy, and locomotor function following SCI.

MATERIALS AND METHODS

Study design

This systematic review and meta-analysis followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines and was registered in the International Prospective Registry of Systematic Reviews (CRD42024569681).[22] The population, intervention, comparison, outcome framework was: Population as rat, mouse, or rabbit models of SCI (compression, transection, hemi-section, or contusion); Intervention as RES; Comparison as untreated or standard-treatment controls; Outcome as locomotor recovery scales and biomarkers of inflammation, apoptosis, and oxidative stress.

Search strategy

Authors searched PubMed, Embase, Scopus, and Web of Science until April 25, 2025, using MeSH/Emtree terms for “resveratrol” and “spinal cord injury” and their synonyms [full strategy in Supplementary Material 1]. Google Scholar, gray literature, and reference lists of relevant articles were also screened to identify additional studies.

Eligibility criteria

Included studies used RES as the intervention in rat, mouse, or rabbit SCI models with a control group. Excluded were human studies, non-SCI or ischemia-reperfusion models, transgenic-animal studies, protocols without original data, studies with insufficient outcome data, duplicates, and reviews/meta-analyses.

Screening and data collection

Records were managed in EndNote X9. Two independent, blinded reviewers screened titles/abstracts and then full texts against the eligibility criteria; disagreements were resolved by a third reviewer. When multiple reports used overlapping data, the study with the largest sample or longest follow-up was retained. Corresponding authors were contacted for missing data, and non-English articles were translated as needed. Extracted data included study design, animal characteristics, SCI model/severity, timing and route of RES administration, follow-up duration, and outcome measures.

Risk of bias and quality of evidence

Two reviewers independently assessed risk of bias using SYRCLE’s tool, resolving disagreements by discussion or third-reviewer arbitration. Evidence quality was graded using the Grading of Recommendations Assessment, Development, and Evaluation (GRADE).[7,23,29]

Statistical analysis

Analyses were performed in STATA 18.0. Standardized mean differences (SMD) with a 95% confidence interval (95% CI) were calculated per outcome and pooled using a random-effects model, given anticipated heterogeneity. Meta-analysis required ≥3 studies per outcome; for scales where lower scores indicate better outcomes, absolute SMD values were used. Heterogeneity was assessed through I2 and Chi-squared tests (I² > 50% or p < 0.1 indicating significant heterogeneity). Subgroup analyses and meta-regression explored heterogeneity by species, SCI level/model/severity, and intervention timing/duration. Publication bias was evaluated using funnel plots and Egger’s test, per Doleman et al.[2]

RESULTS

Study selection

Of 665 records identified, 193 duplicates were removed, and 405 were excluded at title/abstract screening. Of the remaining 67 full-text articles, 26 met all inclusion criteria [Figure 1].[1,4-6,8,10-18,20,24,26-28,30-36]

Figure 1:

Preferred reporting items for systematic reviews and meta-analyses 2 flow diagram of the screening process.

Study characteristics

The 26 included studies (23 from China, 2 from Türkiye, 1 from Argentina) used rats (n = 20), mice (n = 5), and rabbits (n = 1), predominantly male, with thoracic SCI models (below T6) of moderate (n = 22) or severe (n = 4) severity. RES was most commonly administered intraperitoneally (50–400 mg/kg, 21 studies), with oral (10–400 mg/kg, 2 studies) and intravenous (5–100 mg/kg, 3 studies) routes used less frequently. The treatment began within 0–2 h post-injury in most studies (21/26), with follow-up ranging from 1 to 56 days (median 21 days). Full details are provided in Table 1.

Table 1:

Summary characteristics of the included studies.

Locomotor function

Pooled analysis of 17 studies showed that RES significantly improved locomotor recovery after SCI (SMD = 3.41, 95% CI: 2.084.74, p < 0.001; I² = 92.55%) [Figure 2]. Given high heterogeneity, subgroup and meta-regression analyses identified dosing frequency, treatment duration, and daily dose as significant moderators: three-times-daily dosing outperformed once-daily dosing (meta-regression p = 0.012), and both shorter (<1 week) and longer (>1 week) treatment courses showed larger effects than 1-week regimens (p = 0.006–0.029). Higher daily dose was independently associated with greater effect size (coefficient = 0.0065, 95% CI: 0.0007–0.0123, p = 0.029), whereas the treatment duration in days was not (p = 0.387) [Supplementary Table 1 and Supplementary Figure 1].

Figure 2:

The forest plot illustrate the efficacy of resveratrol on locomotor function. CI: Confidence interval, SD: Standard deviation

Inflammation

Across 16 studies, RES significantly reduced tumor necrosis factor-alpha (TNF-α) (SMD = −4.64, p < 0.001), interleukin (IL)-6 (SMD = −3.19, p = 0.03), and IL-1β (SMD = −4.10, p < 0.001), though heterogeneity was substantial for IL-1β and IL-10 [Figure 3]. Subgroup analysis of IL-1β (12 studies) showed no difference between intraperitoneal and intravenous administration (p = 0.775), and no significant relationship with treatment duration or follow-up time [Supplementary Table 2, Supplementary Figure 1].

Figure 3:

The forest plots illustrate the efficacy of resveratrol on inflammatory, , CI: Confidence interval, SD: Standard deviation, IL-1β: Interleukin-1 beta, IL-6: Interleukin-6, TNF-α: Tumor necrosis factor-alpha.

Antioxidative effects

RES significantly increased superoxide dismutase (SOD) (SMD = 4.74, p < 0.001; 8 studies), glutathione (GSH) (SMD = 3.21, p = 0.01; 5 studies), and decreased malondialdehyde (MDA) (SMD = −2.41, p < 0.001; 11 studies) [Figure 4]. No significant moderating effects of administration route or species were identified for SOD or MDA [Supplementary Tables 3 and 4].

Figure 4:

The forest plot illustrate the efficacy of resveratrol on oxidative stress , CI: Confidence interval, GSH: Glutathione, MDA: Malondialdehyde, SD: Standard deviation, SOD: Superoxide dismutase.

Apoptosis biomarkers

RES exerted a consistent anti-apoptotic effect: B-cell lymphoma 2 (Bcl-2) increased (SMD = 3.55, p < 0.001) while Bcl-2-associated X protein (Bax) (SMD = −3.71), Bax/Bcl-2 ratio (SMD = −6.83), and cleaved caspase-3 (SMD = −4.39) decreased (all p < 0.001 except Bax/Bcl-2, p = 0.23). Terminal deoxynucleotidyl transferase dUTP nick end labeling (TUNEL)-positive cells (SMD = −3.81, p < 0.001) and apoptotic index (SMD = −7.16, p < 0.001) were also significantly reduced [Figure 5 and 6a]. Heterogeneity was high for most markers; insufficient study numbers precluded further subgroup analysis.

Figure 5:

The forest plot illustrate the efficacy of resveratrol on apoptosis biomarkers Bax: Bcl-2-associated X protein, Bcl-2: B-cell lymphoma 2, CI: Confidence interval, SD: Standard deviation

Figure 6:

The forest plot illustrate the efficacy of resveratrol on (a) apoptosis assays, and (b) neuronal biomarkers following spinal cord injury. CI: Confidence interval, Nissl: Nissl substance (rough endoplasmic reticulum), SD: Standard deviation, TUNEL: Terminal deoxynucleotidyl transferase dUTP nick end labeling. NeuN: Neuronal nuclei.

Neuronal markers

RES significantly increased Nissl-positive neurons (SMD = 3.39, p = 0.02), neuronal nuclei (NeuN)-positive neurons (SMD = 2.00, p = 0.14), and growth-associated protein 43 (GAP-43) expression (SMD = 3.13, p = 0.04) compared with controls [Figure 6b].

Sensitivity analysis

Excluding the single rabbit study did not materially change any pooled estimates or heterogeneity levels, confirming the robustness of the primary findings [Supplementary Table 5].

Risk of bias in the studies

Using the SYRCLE tool, most domains were rated as unclear risk across all 26 studies, with low risk for incomplete outcome data; overall risk of bias was judged as “some concerns” [Supplementary Table 6].

Publication bias

Funnel plots and Egger’s test showed no evidence of small-study effects for most outcomes. Significant asymmetry was detected for cleaved caspase-3 (p = 0.038), consistent with visual funnel plot asymmetry, while significant Egger’s slopes without corresponding intercept effects were noted for IL-6, IL-1β, and GSH, suggesting heterogeneity rather than true publication bias [Figure 7].

Figure 7:

The funnel plot illustrates the distribution of outcomes in studies for assessment of publication bias.

Certainty of evidence

Using GRADE, certainty of evidence was low for IL-10 and cleaved caspase-3, moderate for TNF-α, SOD, MDA, Bcl-2, Bax, and apoptotic index, and high for all remaining outcomes [Supplementary Table 7].

DISCUSSION

RES is a natural phytoalexin polyphenol with anti-oxidative and anti-inflammatory properties.[1] In this systematic review and meta-analysis, RES administration was associated with improved locomotor recovery and favorable effects on apoptosis, inflammation, oxidative stress, and neuronal survival in rodent models of SCI.

Apoptosis is a major contributor to neuronal damage after SCI.[14] RES treatment significantly reduced Bax, cleaved Caspase-3, the Bax/Bcl-2 ratio, apoptotic index, and TUNEL-positive cells, while increasing Bcl-2. Given that apoptosis regulation is a key benchmark for evaluating SCI therapeutics before clinical translation, these findings support RES’s neuroprotective potential.[21]

RES also reduced pro-inflammatory markers (IL-1β, TNF-α, IL-6), consistent with its known suppression of cytokines that drive secondary injury in SCI. However, IL-10 was unexpectedly lower in the RES group, and IL-1β levels rose with longer treatment duration and follow-up. Combined with high heterogeneity in these outcomes, this raises uncertainty about the durability of RES’s anti-inflammatory effect over extended periods.[3,19]

On oxidative stress, RES lowered MDA, while increasing SOD and GSH.[25] These effects paralleled improved neuronal survival, reflected in higher NeuN, Nissl-positive cell counts, and GAP-43 expression, markers associated with axonal regrowth and therapeutic efficacy in SCI.[9] Functionally, RES improved locomotor recovery (Basso, Beattie, Bresnahan/Basso mouse scale scores), with higher doses associated with greater improvement, consistent with prior reports of reduced lesion size, demyelination, and gliosis.[12]

Several limitations should be noted. Substantial heterogeneity across studies, stemming from differences in experimental protocols and RES dosing, limits the strength of these conclusions and emphasizes the need to define an optimal therapeutic window. Further studies are needed to clarify RES’s long-term anti-inflammatory effects, both alone and in combination with other neuroprotective agents.

CONCLUSION

RES demonstrates promising neuroprotective effects in preclinical small mammal SCI models, reducing oxidative stress, inflammation, and apoptosis while enhancing neuronal survival and locomotor recovery, though further standardized preclinical and clinical studies are needed to confirm its efficacy and optimize dosing, delivery, and timing.

Ethical approval:

Institutional Review Board approval is not required.

Declaration of patient consent:

Patient’s consent is not required as there are no patients in this study.

Financial support and sponsorship:

Publication of this article was made possible by the James I. and Carolyn R. Ausman Educational Foundation.

Conflicts of interest:

There are no conflicts of interest.

Use of artificial intelligence (AI)-assisted technology for manuscript preparation:

The authors confirm that there was use of artificial intelligence (AI)-assisted technology. The authors employed Claude Sonnet 5 for final language editing in this paper. After utilizing these tools, they carefully reviewed and refined the content to ensure it met all necessary standards and took full responsibility for the published work.

Supplementary Data available on:

https://dx.doi.org/10.25259/SNI_815_2026

Disclaimer

The views and opinions expressed in this article are those of the authors and do not necessarily reflect the official policy or position of the Journal or its management. The information contained in this article should not be considered to be medical advice; patients should consult their own physicians for advice as to their specific medical needs.