Registration has been completed in the PROSPERO database before data extraction (CRD420251035172). The protocol is reported in compliance with PRISMA-P (Preferred Reporting Items for Systematic Reviews and Meta-Analyses Protocols) [29], and the completed PRISMA-P checklist is provided in Checklist 1. The study will be conducted strictly in accordance with the Cochrane Handbook for Systematic Reviews of Interventions [30] and will adhere to the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) 2020 guidelines [31] to ensure transparent and complete reporting.

We intend to incorporate RCTs that meet predefined eligibility criteria to evaluate the effectiveness and safety of electroacupuncture in preventing PND among older patients undergoing general anesthesia.

Studies will be selected according to the criteria outlined in the following sections.

A comprehensive search will be conducted in 8 electronic databases, including PubMed, Embase, Cochrane Central Register of Controlled Trials, Web of Science, China National Knowledge Infrastructure, Chongqing VIP Chinese Science and Technology Periodical Database, Wan Fang, and China Biomedical Literature Database, from inception to March 16, 2025. The literature search will be updated to include all studies available up to just before the manuscript’s final submission to capture the most recent evidence. To maximize the scope of review, the ClinicalTrials.gov, Chinese Clinical Trial Registry, International Traditional Medicine Clinical Trial Registry, and gray literature (eg, dissertations) will be additionally searched to identify eligible trials. Furthermore, the reference lists of included studies will be manually screened to supplement relevant literature.

The search strategy will integrate both Medical Subject Headings (MeSH) and free-text terms across four critical domains: (1) population—“aged,” “elderly,” “geriatric,” “perioperative neurocognitive disorder,” “postoperative cognitive dysfunction,” “delirium,” and “cognitive decline”; (2) intervention—“electroacupuncture,” “acupuncture,” and “EA”; (3) context—“general anesthesia,” “surgery,” and “postoperative period”; and (4) study design—“randomized controlled trial” and “RCT.” Boolean operators (AND, OR, NOT) and truncation symbols (*) will be used to ensure optimal balance between search sensitivity and specificity. The detailed search strategy for PubMed is provided in Table 1. The complete, reproducible search strategies for all other databases and clinical trial registries are provided in Multimedia Appendix 1, ensuring full reproducibility as per PRISMA-P guidelines.

All identified records will be imported into EndNote (version X20; Clarivate Analytics) for reference management and deduplication. Two independent reviewers (CW and XW) will perform screening based on title, abstract, and full-text assessment. Discrepancies will be discussed and adjudicated by a third reviewer (KW).

Two independent researchers (CW and XW) will perform data extraction using a standardized template developed in accordance with the Cochrane Handbook [33], with disagreements resolved by KW. Researchers will extract (1) study characteristics (first author, publication year, nationality, etc), (2) participant demographics (age, sex ratio, sample size, surgical type, etc), (3) intervention details (acupoints, stimulation parameters [waveform, frequency, and intensity], electroacupuncture intervention timing [preoperative, intraoperative, and postoperative], sessions, duration, etc), and (4) outcomes (evaluation indicators [POCD and POD], assessment time points, measurement tools, diagnostic criteria for outcome events, etc). For all biomarker outcomes from each included study, we will systematically extract detailed methodology including the specific assay technique (eg, enzyme-linked immunosorbent assay or chemiluminescence immunoassay), sampling time points, biomarker reference ranges, and units of measurement. Missing data will be retrieved from corresponding authors via email. Studies with irretrievable critical data will be excluded from quantitative synthesis. For duplicate publication of outcome data, we will retain only the most comprehensive or recent publication to avoid data overlap.

Two investigators (CW and XW) will independently assess the methodological quality of included studies using the revised Cochrane risk of bias tool for randomized trials (RoB 2) [34]. The evaluation will cover 7 domains: random sequence generation, allocation concealment, blinding of participants and personnel, blinding of outcome assessment, incomplete outcome data, selective reporting, and other biases. Searching for and comparing preregistered trial protocols will be an integral part of identifying selective reporting. Each domain will be judged as “low risk,” “high risk,” or “unclear risk,” with discrepancies arbitrated by KW.

All meta-analyses will be performed using RevMan (version 5.4; Cochrane Collaboration). Mean differences and 95% CI will be used to calculate the effect size for continuous variables, and relative risk estimates with 95% CI will be used for dichotomous variables. When the units of continuous variables are not standardized, the standardized mean differences with 95% CI will be used. Heterogeneity will be assessed via I² statistic and Q test. If I²≤50% and P≥.1, indicating acceptable heterogeneity, a fixed-effect model will be used; if I²>50% or P<.1, suggesting substantial heterogeneity, a random-effects model will be selected [35]. Forest plots will be generated to visually present the meta-analysis results, with P<.05 considered statistically significant (2-tailed). Where feasible, independent analyses for POD and POCD will be performed to investigate the potential differential effects of electroacupuncture. For biomarker outcomes, quantitative synthesis will be attempted only for studies with comparable methodologies and sampling time points (eg, enzyme-linked immunosorbent assay results from samples taken 24 hours postoperatively). Where quantitative synthesis is precluded by clinical or methodological heterogeneity (eg, vast differences in timing and methods or too few studies per group), we will perform a systematic narrative synthesis and summarize key findings in structured evidence tables.

Given the anticipated clinical heterogeneity in electroacupuncture interventions, we will use the following strategies: (1) detailed description of electroacupuncture parameters (eg, acupoints, stimulation parameters, and timing) for all included studies; (2) prespecified subgroup analyses to explore the impact of key sources of heterogeneity, including acupoint selection (eg, scalp vs body acupoints and standardized vs individualized prescriptions), stimulation frequency (eg, 2 Hz vs >20 Hz), the number of electroacupuncture sessions (multiple sessions [≥2] vs single session), and electroacupuncture intervention timing (preoperative, intraoperative, postoperative, or multiphase perioperative intervention); and (3) if a sufficient number of studies are available (eg, ≥10 studies), meta-regression will be conducted to assess the contribution of continuous variables (eg, the total number of electroacupuncture sessions) to the observed heterogeneity. The results of these analyses will be integral to the interpretation of the pooled findings. Additionally, procedure-stratified and evaluation time–stratified subgroup analyses will also be conducted to reveal differential effect sizes contributed by surgical trauma intensity and outcome assessment time points. Sensitivity analyses will be performed by sequentially excluding each low-quality study with high overall risk of bias, aiming to evaluate the robustness of the findings and determine whether potential methodological bias exists in these studies.

When more than 10 studies are included for a specific outcome, we will assess publication bias through funnel plot asymmetry test and Egger linear regression test, with a statistical significance threshold of P<.05 indicating potential publication bias [36].

Two reviewers (CW and XW) will independently grade the evidence certainty for each critical outcome using the Grading of Recommendations Assessments, Developments and Evaluations (GRADE) approach [37], yielding classifications of “high,” “moderate,” “low,” or “very low” confidence. Disagreements will be resolved through discussion or third-party (KW) adjudication. The GRADE assessment will evaluate 5 domains: risk of bias, inconsistency, indirectness, imprecision, and other considerations. The assessment of “risk of bias” will be informed by the overall RoB 2 judgment. The certainty of evidence will be downgraded for risk of bias if a substantial proportion of the contributing studies are judged to be at high risk in key domains. The certainty of evidence will be downgraded for imprecision if the included studies have limited sample sizes that fail to meet the optimal information size criterion. If the included studies are found to have potential publication bias, the certainty of evidence will be downgraded within the “other considerations” domain.

Patients and the public will not be involved in the design, conduct, data analysis, interpretation of results, reporting, and dissemination plans of this study.

This study does not involve the direct participation of patients and the public, so ethics approval and informed consent are not applicable and required. Because the systematic review relies on previously published data, no original data will be collected.

The study will be rigorously conducted following the Cochrane Handbook for Systematic Reviews of Interventions [30] and will comply with the PRISMA 2020 guidelines [31] to guarantee transparency and comprehensive reporting.

The protocol for this systematic review will be amended when necessary, and the revisions will be documented in PROSPERO.

Advanced age (≥60 years, particularly ≥70 years) and general anesthesia are established independent risk factors for PND [1,3]. Epidemiological data demonstrate a marked age-dependent increase in PND incidence, with patients aged 70 years and older experiencing twice the risk compared with those aged 60 to 69 years [38]. The confluence of population aging and rising surgical volumes has rendered PND a critical public health challenge, adversely affecting both short-term recovery trajectories and long-term quality of life, while compounding societal health care burdens [1,39]. To date, no effective treatment exists for PND, underscoring the critical importance of exploring a novel preventive strategy.

Accumulating evidence supports acupuncture’s neuroprotective potential, with studies reporting significant cognitive improvements across multiple disorders including Alzheimer disease [18], mild cognitive impairment [40], vascular dementia [20], vascular cognitive impairment [41], poststroke cognitive impairment [42], schizophrenia [43], and cognitive dysfunction following acute myocardial ischemia-reperfusion injury [17]. Clinical studies [12,13,16] have confirmed electroacupuncture’s potential to reduce PND incidence and improve cognitive recovery in older patients receiving general anesthesia. Electroacupuncture may exert neuroprotective effects through multifaceted mechanisms such as (1) attenuation of neuroinflammation and systemic inflammatory responses [18,44], potentially mediated via the vagal-adrenal axis activation [44]; (2) remodeling of synaptic structural and functional plasticity [18]; (3) upregulation of brain-derived neurotrophic factor related to neuronal survival and synaptic plasticity [45,46]; (4) modulation of gut microbiota homeostasis [19]; (5) inhibition of microglial activation [17,19]; (6) prevention of neuronal apoptosis [17]; (7) mitigation of oxidative stress [21]; and (8) enhancement of cerebral perfusion [20]. Neuroimaging evidence indicates that acupuncture can activate and enhance neural activity in cognition-related brain regions (eg, prefrontal cortex and hypothalamus) [18,40], while strengthening functional connectivity to modulate brain networks [47]. This suggests that the molecular effects of electroacupuncture ultimately translate into enhanced neural efficiency and network integration, providing a plausible substrate for observed cognitive benefits. Furthermore, acupuncture provides complementary perioperative benefits [48,49] such as preoperative anxiety relief, intraoperative anesthetic requirement reduction, attenuation of surgical stress responses, and postoperative complication decrease, collectively contributing to PND prevention [1]. In addition, electroacupuncture demonstrates unique integrative advantages by combining acupoint specificity, electrical stimulation, and needling effects. Its favorable safety profile, minimal invasiveness, and lower cost-effectiveness render it particularly suitable for perioperative application.

Current evidence remains fragmented and insufficient due to clinical heterogeneity, methodological heterogeneity, and limited sample sizes in existing studies [12-16,22], which consequently restricts the widespread perioperative application of electroacupuncture. While previous systematic reviews and meta-analyses [23-28] have primarily focused on transcutaneous electrical acupoint stimulation for PND, there remains an evidence gap regarding electroacupuncture’s potential for PND prevention in older patients (≥60 years) undergoing general anesthesia. Therefore, we propose this review to objectively evaluate the effectiveness and safety of perioperative electroacupuncture for PND prevention in this high-risk population, with the ultimate goal of providing updated evidence-based recommendations for clinical practice.

The incidence of PND will serve as the primary outcome, providing direct evidence on whether electroacupuncture reduces the risk of PND. The MMSE scores are selected as a secondary outcome due to its well-established sensitivity and specificity in assessing global cognitive function [32], thereby comprehensively capturing electroacupuncture’s potential benefits on postoperative cognitive recovery. Neuroinflammation represents one of the most critical pathological mechanisms underlying PND [50]. Specifically, the systemic inflammatory response syndrome triggered by surgical trauma and stress exacerbates central nervous system inflammation through multiple pathways: (1) increased blood-brain barrier permeability, (2) activation of microglia and astrocytes, and (3) subsequent release of proinflammatory cytokines (eg, IL-1β, IL-6, and TNF-α) into the central nervous system. These processes collectively contribute to synaptic dysfunction, neuronal injury or death, impaired hippocampal neurogenesis, and loss of synaptic plasticity, ultimately leading to PND. Neuronal damage is further reflected by elevated levels of brain injury biomarkers (NSE and S100β). These proteins enter systemic circulation via a compromised blood-brain barrier. Clinical studies have consistently demonstrated significantly increased serum concentrations of NSE and S100β in patients with PND following coronary artery bypass grafting [51], abdominal surgery [52], and orthopedic procedures [53]. Therefore, this study selects serum levels of IL-1β, IL-6, TNF-α, NSE, and S100β as secondary outcome measures to evaluate whether electroacupuncture exerts neuroprotective effects by mitigating peripheral inflammatory responses and reducing brain injury severity. These outcomes will help establish an “intervention-effectiveness-mechanism” framework, providing critical insights into electroacupuncture’s perioperative neuroprotection.

This study will address a critical evidence gap by specifically investigating electroacupuncture’s effectiveness for PND prevention in older patients (≥60 years) undergoing elective general anesthesia—a previously unexamined yet clinically significant population. Through prespecified subgroup and sensitivity analyses, we will determine how key covariates influence intervention effects and verify the robustness of results. Furthermore, the evaluation of methodological rigor in existing RCTs will identify crucial knowledge gaps to inform the design of future high-quality clinical trials.

Should meta-analysis confirm electroacupuncture’s effectiveness and safety for PND prevention among older patients undergoing surgery, the GRADE-assessed evidence from this review will (1) qualify electroacupuncture as a candidate intervention for inclusion in clinical guidelines, (2) offer a novel nonpharmacological strategy to optimize perioperative care protocols for older adults, and (3) potentially mitigate the cascade of adverse outcomes associated with PND through enhanced cognitive recovery in high-risk populations. Such confirmation would represent a significant advancement in clinical practice, particularly for anesthesia and surgical teams managing older patients.

This study has several inherent limitations: First, clinical heterogeneity is anticipated due to variations across studies in both demographic factors (anesthesia regimens and surgical types) and electroacupuncture intervention protocols (eg, acupuncture prescription, intervention timing, sessions, and stimulation parameters), which may preclude pooled quantitative analysis for certain outcomes. Second, methodological limitations in the included studies may compromise the strength of the evidence. Third, the generalizability of findings may be restricted to older patients undergoing elective surgery under general anesthesia, excluding those requiring emergency procedures. Future research should (1) conduct multicenter RCTs with standardized electroacupuncture protocols (including optimal acupoint selection, waveform, frequency, and intervention timing) and improved methodology (eg, standardized outcome assessment protocols), (2) elucidate electroacupuncture’s mechanisms of action by correlating biomarkers with multimodal neuroimaging data, and (3) develop risk stratification tools to identify patients most likely to benefit from electroacupuncture intervention.

This systematic review will provide a comprehensive, GRADE-assessed evidence synthesis specifically focused on the efficacy and safety of electroacupuncture for PND prevention in older patients undergoing general anesthesia. The findings are expected to either support electroacupuncture’s integration into ERAS protocols if effectiveness is demonstrated or identify critical knowledge gaps requiring further RCTs if evidence remains inconclusive.