Previous Article in Journal
Prevalence and Associated Factors of Depression Among Elderly Hypertensive Patients in Vietnam
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Geriatrics
Volume 10
Issue 5
10.3390/geriatrics10050130
geriatrics-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Systematic Review

Fall-Related Adverse Events of Anti-Epileptic Drugs Used for Neuropathic Pain in Older Adults: A Systematic Review and Meta-Analysis

by
Arun Vamadevan
1,2,*,†,
Vijesh Vijayan
3,*,†,
Fellisha Marwein
1 and
Nishad Yoosuf
4
1
NIHR Clinical Research Facility, Liverpool University Hospitals NHS Foundation Trust, Liverpool L7 8XP, UK
2
School of Health and Society, University of Salford, Salford M5 4WT, UK
3
Stockport NHS Foundation Trust, Stockport SK2 7JE, UK
4
School of Medical Sciences, University of Hyderabad, Hyderabad 500046, India
*
Authors to whom correspondence should be addressed.
†
These authors contributed equally to this work.
Geriatrics 2025, 10(5), 130; https://doi.org/10.3390/geriatrics10050130 (registering DOI)
Submission received: 12 July 2025 / Revised: 27 August 2025 / Accepted: 22 September 2025 / Published: 11 October 2025
Browse Figures
Versions Notes

Abstract

Background: Older adults are at elevated risk of falls, especially when prescribed AEDs (AEDs) for neuropathic pain. The sedative and neuropsychiatric effects of these agents contribute significantly to fall-related morbidity. However, existing studies often lack stratification by age and dose. Objective: To systematically evaluate the incidence and drug-specific risk of falls and fall-related adverse events (AEs) in older adults prescribed AEDs for neuropathic pain. Methods: A systematic search was performed across PubMed, Scopus, CINAHL, ScienceDirect, and Cochrane Library databases up to May 2025. Studies were selected using PICOS criteria and included RCTs and controlled cohort studies reporting on AED-related AEs among participants aged ≥60 years. The methodological quality was assessed using RoB 2, ROBINS-I, and GRADE frameworks. Meta-analyses were performed using logit event rates and fixed-effects modeling via Comprehensive Meta-Analysis v3.7. Publication bias was evaluated using Begg’s and Egger’s tests. Results: Twenty-three studies met the inclusion criteria. The pooled logit event rate for falls was −1.693 (95% CI: −1.993 to −1.393), corresponding to a 15.5% incidence. Gabapentin showed the lowest fall risk (~10%), while pregabalin and carbamazepine were associated with higher rates of dizziness (up to 21.6%), sedation (~15.5%), and ataxia (~17.8%). Heterogeneity was low (I2 = 0–22.3%) across outcomes. Conclusions: AEDs carry a clinically significant fall risk in older adults, with dose-dependent patterns. Gabapentin may present a safer profile, while pregabalin and carbamazepine warrant cautious use and monitoring. These findings inform individualized prescribing and fall prevention strategies in geriatric neuropathic pain management.

1. Introduction

Neuropathic pain is a common and disabling condition among older adults, with prevalence rates ranging from 7% to 10% in community-dwelling populations and even higher among institutionalized older adult individuals [1]. Initial treatments for neuropathic pain frequently include antiepileptic drugs (AEDs) such as gabapentin, pregabalin, and carbamazepine, which act by modulating calcium channels and sodium conductance in nociceptive pathways [2]. However, these medications also exert significant central nervous system (CNS) effects, resulting in common adverse events such as dizziness, sedation, ataxia, and somnolence—factors strongly associated with increased fall risk in older adults [3].
Falls are the leading cause of injury, hospitalization, and loss of independence in people aged ≥65, with roughly 30% of older adults experiencing at least one fall annually [4]. The World Health Organization’s report identifies falls as the second leading cause of unintentional injury deaths globally, accounting for over 684,000 fatalities and 37.3 million medically treated cases annually, underscoring the timeliness and urgency of this public health issue [5]. Importantly, medications are among the most modifiable risk factors for falls, and polypharmacy involving CNS-active drugs such as AEDs and antidepressants is particularly hazardous [6]. Despite this, older adults are frequently underrepresented or excluded from randomized controlled trials (RCTs), especially in dose-finding studies, resulting in an evidence gap regarding drug safety in real-world geriatric populations [7].
Although observational studies have linked AEDs to increased fall risk, with odds ratios as high as 2.55 [8], these studies often lack dosage stratification and are susceptible to confounding. A recent review by Zia et al. (2017) [9] emphasized the urgent need for high-quality studies that delineate drug-specific and dose-dependent risks in older populations. In the absence of such data, clinicians face significant uncertainty in balancing analgesic efficacy with fall-related harms [10]. Most existing literature either aggregates data across diverse CNS-active drug classes or fails to stratify by age and dose, limiting clinical applicability [11]. This lack of granular evidence leaves prescribers without robust guidance on which AEDs may pose higher risks at certain doses in geriatric populations [12].
This study aims to address critical knowledge gaps by systematically evaluating and quantifying the incidence of fall-related adverse events in older adults receiving AEDs for neuropathic pain. Given the heightened vulnerability of older populations to falls—often triggered by CNS-active medications—this analysis focuses specifically on identifying drug- and dose-dependent risk patterns. Using data from randomized controlled trials and controlled observational studies, the review quantifies the incidence of key fall-associated adverse events, including dizziness, somnolence, sedation, vertigo, and ataxia. Importantly, it not only examines aggregate risk but also stratifies by individual AEDs and their corresponding dosages to determine differential safety profiles.

2. Methods

This study was conducted in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines (22) for conducting systematic reviews and meta-analyses. Compliance with the principles of the Declaration of Helsinki further reinforced the ethical foundation of this review. This review exclusively analyzed secondary data from existing studies and, therefore, qualifies for exemption from informed consent or institutional review board approval. The study protocol was registered on the PROSPERO online database with the registration number CRD420251048827.

2.1. Search Strategy

A comprehensive and systematic literature search was conducted by two authors (AV and VV) across five electronic databases, PubMed, Cochrane Library, Scopus, ScienceDirect and CINAHL, covering studies published up to May 2025 as described in Table 1. The search strategy was developed using a combination of MeSH terms and free-text keywords related to AED (e.g., “gabapentin,” “pregabalin,” “carbamazepine,” “lamotrigine”), adverse events (e.g., “falls,” “dizziness,” “ataxia,” “somnolence,” “sedation,” “vertigo”), and older populations (e.g., “elderly,” “aged,” “older adults”). Boolean operators (AND/OR) were used to structure search strings, with database-specific syntax adjustments. Filters were applied to restrict results to human studies involving participants aged ≥60 years. Only English-language publications were included. Duplicates were removed using automated reference management tools, followed by manual screening. Records lacking free full-text availability were excluded. Titles and abstracts were screened for relevance, and full texts were assessed against predefined inclusion criteria. Reference lists of included studies were also manually searched to identify additional eligible studies. The complete electronic search strategy for all databases is available in Supplementary Table S1.

2.2. Study Selection

Study selection was performed using the PICOS (Participants, Intervention, Comparisons, Outcomes, and Study Design) framework to define inclusion and exclusion criteria. The inclusion and exclusion criteria for this systematic review were defined using the PICOS framework to ensure methodological rigor and relevance to the research objectives. Population criteria included studies involving older adults aged >50 years, reflecting the population most vulnerable to antiepileptic drug (AED)-associated adverse events (AEs) such as falls. Interventions were restricted to AEDs prescribed for neuropathic pain, including gabapentin, pregabalin, carbamazepine, oxcarbazepine, lamotrigine, lacosamide, levetiracetam, and related agents. Comparators included placebo, usual care, or other AEDs, allowing for assessment of relative safety profiles. Outcomes required clear reporting of fall-related AEs including falls, dizziness, ataxia, vertigo, somnolence, or sedation. Study design inclusion was limited to randomized controlled trials (RCTs), cohort studies, and controlled observational studies that provided extractable quantitative data. Studies were excluded if they did not report on older populations or failed to stratify data by age, focused on other indications such as epilepsy without mention of the adverse events relating to fall, assessed non-AED interventions, did not report on falls or neuropsychiatric AEs relevant to fall risk, or were reviews, editorials, letters, case reports, or non-English publications.

2.3. Data Collection and Quality Evaluation

Data collection and quality evaluation were carried out independently by two reviewers, who screened the titles and abstracts of studies in Mendeley Cite based on the inclusion criteria. Full texts of potentially eligible studies were reviewed, with disagreements resolved through decision-making by a blinded third author (NY). Data were extracted into a pre-tested Microsoft Excel sheet for consistency, documenting key details such as author, year, study design, country, diagnosis, sample size, mean age, interventions and measured outcomes. Methodological rigor and potential biases were also documented to ensure systematic data collection and analysis. The methodological quality of the included studies was rigorously evaluated using established tools tailored to study design. For randomized controlled trials (RCTs), the Revised Cochrane Risk of Bias tool (RoB 2) [13] was applied to assess bias across domains such as randomization, deviations from intended interventions, missing outcome data, measurement of outcomes, and selection of reported results. Non-randomized studies were assessed using the ROBINS-I (Risk Of Bias In Non-randomized Studies of Interventions) [14] tool, which evaluates confounding, selection bias, classification of interventions, deviations, missing data, measurement of outcomes, and reporting bias. Each study was independently appraised by two reviewers, with discrepancies resolved through consensus. Additionally, the overall certainty of the evidence for each outcome was graded using the GRADE (Grading of Recommendations, Assessment, Development, and Evaluations) [15] framework, which considers study limitations, inconsistency, indirectness, imprecision, and publication bias. This comprehensive evaluation ensured the reliability and interpretability of the synthesized evidence.

2.4. Statistical Analysis

The statistical analysis for this meta-analysis was conducted using Comprehensive Meta-Analysis (CMA) software version 3.7. Pooled estimates of logit event rates and corresponding standard errors were calculated using a fixed-effects model, which is appropriate due to the generally low heterogeneity observed across studies. The primary measure of effect was the logit event rate, which allowed for consistent transformation of event proportions across studies, particularly when event rates were near 0 or 1. Standard errors and 95% confidence intervals were calculated for each logit event rate. Subgroup analyses were conducted based on drug type and dose to evaluate variation in adverse events such as falls, dizziness, somnolence, sedation, vertigo, and ataxia among older patients treated with AEDs. A Z-test determined the statistical significance of pooled outcomes, with p < 0.05 considered significant. Heterogeneity among studies was quantified using the Q statistic, p-value, and I2 index, with I2 values below 30% considered low. Publication bias was assessed through both Begg’s rank correlation test and Egger’s regression intercept test. In cases where asymmetry was detected in funnel plots, potential bias was further explored through sensitivity analyses.

3. Results

3.1. Study Selection

The PRISMA flowchart illustrates the systematic process of literature identification, screening, and selection applied in this study as shown in Figure 1. A total of 4447 records were initially retrieved from five electronic databases: PubMed (n = 2326), Cochrane Library (n = 137), Scopus (n = 1018), ScienceDirect (n = 898), and CINAHL (n = 68). Following the removal of duplicates (n = 2407) and records lacking free full-text access (n = 253), 1787 records were screened. An automated exclusion tool removed 873 records based on title and abstract relevance. The remaining 914 records were sought for full-text retrieval, but 542 could not be accessed or retrieved due to lack of open-access or institutional availability and were excluded, leaving 372 reports for full eligibility assessment. Among these, 286 were excluded for not specifically addressing older adult populations, 33 due to unrelated interventions, and 30 for lacking data on falls or related adverse events. Parallelly, 77 additional records were identified through other methods, including websites (n = 24) and citation searching (n = 53). Of these, 52 were not retrieved, leaving 25 for eligibility assessment. Thirteen were excluded for irrelevant outcomes and 12 for inappropriate study design. Ultimately, 23 studies met the inclusion criteria and were incorporated into the final review.

3.2. Study Characteristics

The included studies encompass a diverse range of randomized controlled trials and observational designs evaluating the effects of antiepileptics and CNS-active medications on outcomes such as neuropathic pain, psychiatric symptoms, and fall risk in older adults as depicted in Table 1. Sample sizes varied widely—from as small as 8 participants [16] to over 41,000 subjects [17], with most studies focusing on older adults aged 60 years and above. Diagnoses addressed included chronic sciatica, painful diabetic neuropathy, epilepsy, bipolar depression, and dementia-related agitation. Interventions spanned a broad pharmacologic spectrum, including gabapentin, pregabalin, lamotrigine, carbamazepine, lacosamide, mirtazapine, and combinations with B vitamins or antidepressants. Several studies employed head-to-head comparisons or combination therapy strategies.

3.3. Risk of Bias Assessment

The risk of bias for the 15 randomized controlled trials (RCTs) included in this analysis was evaluated using the RoB 2 tool in Figure 2. Most RCTs demonstrated a low overall risk of bias, particularly in domains related to randomization, adherence to intended interventions, and outcome measurement. Studies by Robertson et al. (2018) [18], Saetre et al. (2007, 2009) [20,25], and Tesfaye et al. (2022) [31] were rigorously designed with appropriate allocation concealment, blinding, and complete outcome reporting. A few studies, such as Sajatovic et al. (2011) [22] and Wymer et al. (2009) [30], showed “some concerns” due to moderate dropout rates or adverse event-related attrition, potentially influencing outcome reliability. Holbech et al. (2015) also raised concerns due to higher dropout in the combination arm [28].
For the eight observational studies and cohort designs, the ROBINS-I tool was applied to assess bias in Figure 3. Most prospective cohort studies, including those by Luukinen et al. (1995) [32], Tromp et al. (1998) [33], and Ensrud et al. (2002) [36], were deemed low risk across domains due to well-defined participant selection, standardized outcome assessment, and control of key confounders. Conversely, retrospective studies such as Dustin et al. (2006) [17] and Mayo et al. (1989) [35] were rated as having moderate bias due to limitations in confounder adjustment, potential selection bias, and missing data handling. Cross-sectional data from Masud et al. (2013) [37] and retrospective registry-based data in Titler et al. (2011) [38] also indicated moderate bias due to measurement and reporting variability. These findings underscore the need for cautious interpretation of results from non-randomized studies, although they provide valuable real-world insights into fall risk associations with CNS medication use in the older adults.

3.4. GRADE Assessment

The GRADE assessment of included studies revealed a moderate to high certainty of evidence for most randomized controlled trials (RCTs), particularly those evaluating antiepileptic and neuropathic pain medications in older populations. Few Studies [25,26,31] demonstrated high methodological rigor, consistent results, and direct clinical applicability, thereby achieving a high certainty rating. Conversely, trials with small sample sizes, high dropout rates, or imprecision in outcome estimates—such as those by Jensen-Dahm et al. (2011) [16] and Roose et al. (2003) [23]—were downgraded to low or moderate certainty. Among observational studies, the certainty of evidence was generally moderate, as seen in large-scale cohort studies like Ensrud et al. (2002) [36] and Masud et al. (2013) [37], which provided robust associations between CNS-active drugs and fall risk. However, studies such as Mayo et al. (1989) [35] and Titler et al. (2011) [38] were rated low due to potential confounding and imprecision. A detailed summary of the GRADE ratings for each outcome is presented in Supplementary Table S2.

3.5. Incidence on Falls

The Forest Plot in Figure 4 evaluated the incidence of falls among older patients receiving AEDs for neuropathic pain, synthesizing data from eight studies and subgrouping by drug type. The overall pooled logit event rate was −1.663 (SE = 0.153; 95% CI: −1.993 to −1.393), which corresponds to an estimated fall incidence of approximately 15.5%. The heterogeneity across studies was low (Q = 6.83; df = 7; p = 0.45; I2 = 0%), suggesting consistency in effect estimates. Among subgroups, Gabapentin was represented by two studies [19,27], yielding a pooled logit of −2.20 (SE = 0.32), equating to a lower fall incidence (~10%) and moderate heterogeneity (I2 = 32.8%). Pregabalin, analyzed in two studies [16,31], showed a higher logit rate of −1.53 (SE = 0.68), suggesting ~18% fall risk, with no significant heterogeneity (I2 = 0%). Lamotrigine [22], Mirtazapine [23], and Oxcarbazepine [29] were each evaluated in single studies, with logit event rates of −1.85, −1.52, and −1.44, respectively, translating to a fall incidence ranging between 14% and 18%. The consistency in directionality across AEDs highlights a clinically meaningful risk of falls, particularly relevant in geriatric populations with neuropathic pain. Notably, gabapentin was associated with a comparatively lower risk, suggesting it may be a safer option in fall-prone older adult individuals. These findings emphasize the importance of fall-risk assessment and individualized drug selection in older adults, especially when initiating or titrating AED therapy for neuropathic pain. Further studies should explore dose–response relationships and mechanisms underlying differential fall risks across AEDs.
To strengthen the findings, we performed sensitivity analyses using both random-effects modeling and leave-one-out procedures, as shown in Supplementary Figure S1. The overall pooled logit event rate remained highly consistent across models, indicating that the choice of statistical approach did not materially influence results. In the leave-one-s heterogeneity, the effect stays negligible (I2 = 0%). These findings suggest that no single trial disproportionately affected the meta-analysis and confirm that the observed association between AED use and falls is stable, reliable, and not driven by outlier studies.

3.6. AE Relating to Falls

3.6.1. Incidence of Dizziness

The Forest plot in Figure 5 revealed that dizziness is a prevalent adverse event (AE) among older patients prescribed AEDs for neuropathic pain, with an overall fixed-effect logit event rate of −1.438 (SE = 0.066), translating to an incidence of approximately 18.4%. The analysis demonstrated low heterogeneity (I2 = 22.3%), indicating consistency across studies. Drug-specific analysis showed that Lacosamide had the highest dizziness incidence (~21.6%) with significant heterogeneity (I2 = 90.2%), reflecting dose-related variability between 200 mg and 600 mg regimens. Gabapentin (300–3600 mg/day) also demonstrated a high incidence (~19.7%) with moderate heterogeneity (I2 = 60.6%), likely due to wide dosing variations. In contrast, drugs like Lamotrigine (25–500 mg/day) and Carbamazepine (300–2000 mg/day) showed lower dizziness rates (~16–17%) and minimal heterogeneity (I2 = 0%), suggesting more stable safety profiles. Pregabalin (150–600 mg/day) had a pooled incidence of ~20.6% with consistent results across studies (I2 = 0%). Single-study estimates for Levetiracetam, Mirtazapine, and Oxcarbazepine showed dizziness incidences ranging from 13% to 18%. The findings indicate a dose-dependent pattern, where higher or more rapidly titrated doses (notably for Gabapentin and Lacosamide) are associated with increased dizziness. Clinically, this underscores the importance of individualized dosing and cautious titration, especially in frail older adults at risk of falls. Agents like Lamotrigine and Carbamazepine may be preferable when aiming to minimize vestibular side effects. These results highlight the need for fall-prevention strategies and close monitoring of patients, particularly when initiating or up-titrating AED therapy in geriatric neuropathic pain management.
Sensitivity analysis was performed to evaluate the robustness of the dizziness findings by excluding three influential trials, as shown in Supplementary Figure S2 (Dworkin et al., 2009 [27], and Wymer et al., 2009 [30]). After their removal, the pooled logit event rate for dizziness remained consistent with the primary analysis, indicating minimal change in the overall incidence estimate. Importantly, heterogeneity was further reduced, with the I2 value declining from 22.3% in the initial model to 0% in the sensitivity analysis. This substantial reduction confirms that variability across studies was largely attributable to these outlying trials, which involved higher dosing regimens or distinct trial populations. The persistence of similar effect sizes alongside the disappearance of heterogeneity underscores the stability and reliability of the association between AED use and dizziness. These results suggest that the observed prevalence of dizziness is not dependent on any single study and reflects a reproducible and robust safety signal.

3.6.2. Incidence of Somnolence

This meta-analysis in Figure 6 examined the incidence of somnolence—a critical adverse event (AE) linked to fall risk—in older patients receiving AED for neuropathic pain. Thirteen studies were included, stratified by AED class and dosage. The overall pooled logit event rate was −1.633 (SE = 0.079), which corresponds to an approximate somnolence event rate of 15.9%. This indicates that somnolence is a clinically significant and relatively common AE in this population. The statistical association was robust (Z = −20.99, p < 0.001), and overall heterogeneity was low (I2 = 16.7%, Q = 14.41, df = 12, p = 0.275), suggesting consistent results across studies. Subgroup analysis by drug and dose showed notable trends. Carbamazepine (300–2000 mg) had a pooled logit of −1.66 with mild heterogeneity (I2 = 41.5%). Gabapentin (600–3600 mg) and lamotrigine (75–500 mg) had similar effects (logit ~−1.55 to −1.60) and negligible heterogeneity. Pregabalin (300–600 mg) showed more variability (I2 = 85.9%), possibly due to differing study designs and dose responses. Single-study data for levetiracetam (3000 mg), oxcarbazepine (2400 mg), and mirtazapine (15–45 mg) also demonstrated strong associations with somnolence. These findings underscore the importance of careful AED selection and dose titration in the older adults, where even modest somnolence increases the risk of falls and associated morbidity. Clinicians should consider fall-prevention strategies and prefer agents with lower sedative burden. These results highlight the need for individualized risk–benefit assessments, especially in frail older Patients with polypharmacy.
The sensitivity analysis using a random-effects model demonstrated stable estimates of somnolence incidence across all AED classes, with heterogeneity markedly reduced as shown in Supplementary Figure S3. For carbamazepine, the pooled logit was −1.44 (SE = 0.22) with I2 = 0%, confirming consistency despite dose variation. Gabapentin (2 studies) also showed a logit of −1.60 (SE = 0.18) with I2 = 0%, indicating robust agreement. Similarly, lamotrigine yielded a logit of −1.51 (SE = 0.30) and I2 = 0%, with no residual heterogeneity. For single-study outcomes—levetiracetam, pregabalin, oxcarbazepine, and mirtazapine—I2 was inherently 0% due to the absence of within-group variability. Overall, the pooled incidence remained consistent with the fixed-effect model (logit −1.59; incidence ~16%), while I2 fell from 16.7% to 0%. This confirms the somnolence findings are robust and not sensitive to model choice or study-level variability.

3.6.3. Incidence of Sedation

The meta-analysis in Figure 7 examined the incidence of sedation as an adverse event (AE)—a key contributor to fall risk—in older patients using AED (s) for neuropathic pain. The overall pooled logit event rate was −1.700 (SE = 0.158, p < 0.0001), translating to an estimated sedation rate of approximately 15.5%. The low heterogeneity (Q = 7.60, df = 7, p = 0.37; I2 = 0%) suggests consistency across the eight included studies. Subgroup analysis by drug type and dosage revealed clinically relevant differences in sedation risk. For Gabapentin, sedation risk was observed across doses from 600 mg to 2400 mg, with the highest incidence noted at 1800 mg (logit event rate −1.96, SE = 0.57, p = 0.0005, Dworkin et al [27]). At 600 mg [19], although the logit rate was −1.64 (SE = 0.32), statistical significance was higher (p = 3.3 × 10−7), indicating even lower doses warrant caution. For Pregabalin, the highest sedation incidence was seen at 600 mg [31] (logit −2.31, SE = 0.34, p = 7.4 × 10−12). Notably, 150 mg [16] showed a high logit rate (−2.59) but with wide confidence intervals, likely due to small sample size. Oxcarbazepine, evaluated at 900 mg, also showed significant sedation (logit −1.10, SE = 0.32, p = 0.0006). These findings underscore the need for careful dose selection and monitoring, especially at or above 600–1800 mg for Gabapentin and 600 mg for Pregabalin. Clinicians must weigh analgesic benefit against sedation-induced fall risk in older patients, ensuring individualized therapy and fall-prevention strategies.
The sensitivity analysis using a random-effects model demonstrated that sedation incidence remained stable and robust across AED subgroups after exclusion of Tesfaye et al., 2022 [31] as shown in Supplementary Figure S4. The overall pooled logit event rate was −1.53 (SE = 0.18), corresponding to an estimated sedation rate of approximately 15.5%. Importantly, heterogeneity was minimal, with I2 = 0% (Q = 3.38, df = 6, p = 0.76), indicating consistent results across the seven remaining studies. Within subgroups, gabapentin (four studies, doses 600–2400 mg) showed a pooled logit of −1.73 (SE = 0.24) with I2 = 0%, while pregabalin (two studies, 150–600 mg) yielded a logit of −1.69 (SE = 0.45) with I2 = 0%. Oxcarbazepine (one study, 900 mg) remained unchanged, also with I2 = 0%. The close alignment of fixed- and random-effects estimates and the complete absence of residual heterogeneity underscore the robustness of the sedation findings, supporting their reliability across drug classes and doses in older adults.

3.6.4. Vertigo Incidence

The Forest Plot in Figure 8 assessed the incidence of vertigo—a key contributor to fall risk—as an adverse event in older patients receiving AEDs for neuropathic pain. The overall pooled logit event rate was −2.513 (SE = 0.134, p < 0.0001), corresponding to an estimated vertigo incidence of approximately 8%. The analysis demonstrated no significant heterogeneity across the nine included studies (Q = 4.62, df = 8, p = 0.80; I2 = 0%), suggesting consistent results across AED types and dosages. Carbamazepine at 2000 mg (Brodie et al., 1999 [21]) showed a logit event rate of −2.11, translating to the highest vertigo incidence (≈11.1%), with a strong statistical signal (p = 3.6 × 10−11). For Gabapentin, the 3600 mg dose (Alvarado et al., 2016 [24]) had a notably high event rate (logit −2.42; ≈8.2%), whereas 2400 mg (Robertson et al., 2018 [18]) showed a lower and non-significant effect (p = 0.08), suggesting a possible threshold for increased risk beyond 3000 mg. Lacosamide at both 600 mg and 200 mg (Wymer et al., 2009 [30]) had a combined logit of −2.61 (≈7.2%), showing a dose-independent consistent risk. Lamotrigine at 500 mg (Brodie et al., 1999 [21]) showed an event rate of approximately 10.5%. Among all, Pregabalin 600 mg (Alvarado et al., 2016 [24]) showed the lowest logit of −3.18 (≈4%), while the combined doses from three studies yielded a pooled rate of −2.61 (≈6.7%). Clinically, these findings suggest that higher doses of Carbamazepine (2000 mg), Gabapentin (≥3600 mg), and Lamotrigine (500 mg) are associated with higher vertigo incidence in older patients. Careful dose titration and vertigo monitoring are essential to minimize fall-related morbidity. Since the I2 statistic was 0% and the test of heterogeneity was non-significant, there was no evidence of between-study variability for vertigo incidence. Accordingly, a sensitivity analysis was not required, as the findings are stable and robust across the included studies.

3.6.5. Incidence of Ataxia

Figure 9 shows the results of the meta-analysis, which evaluated the incidence of ataxia—a critical adverse event (AE) contributing to fall risk—in older patients treated with AEDs for neuropathic pain. The overall pooled logit event rate was −1.875 (SE = 0.170, p < 0.0001), which corresponds to an ataxia incidence of approximately 13.3% (using inverse logit transformation). This indicates a clinically meaningful rate of motor coordination impairment attributable to AED use in this population. Notably, heterogeneity was low and non-significant across the five included studies (Q = 5.22, df = 4, p = 0.27; I2 = 23.4%), suggesting robust and consistent findings. Carbamazepine (2000 mg) showed a significant association with ataxia (logit −1.61, p = 1.38 × 10−9), translating to an incidence of ~17.8%. Similarly, Lamotrigine (500 mg) reported a logit of −1.77 (p = 1.54 × 10−5; ≈14.5% incidence), and Oxcarbazepine (900 mg) yielded a logit of −1.71 (p = 9.17 × 10−6; ≈15.3% incidence). Among Pregabalin doses, although both 300 mg and 600 mg were evaluated, the pooled logit rate was −2.59 (p = 4.83 × 10−13), corresponding to a lower incidence (~7%), suggesting that Pregabalin may be better tolerated with respect to ataxia. Clinically, these findings underscore the need for careful dose selection and monitoring when prescribing AEDs to older adults. Higher doses of Carbamazepine, Lamotrigine, and Oxcarbazepine were associated with greater ataxia risk, which can significantly increase fall-related morbidity. Therefore, Pregabalin at moderate doses may be a safer alternative, although individualized risk–benefit assessments remain crucial.
The sensitivity analysis, conducted after exclusion of Tesfaye et al. (2022) [31], confirmed the robustness of findings on ataxia risk associated with antiepileptic drugs (AEDs) in older adults with neuropathic pain, as shown in Supplementary Figure S5. The random-effects model demonstrated a pooled logit event rate of −1.88 (SE = 0.17, p < 0.0001), corresponding to an estimated ataxia incidence of 13.3%. Heterogeneity was low and non-significant (Q = 5.22, df = 4, p = 0.27; I2 = 23.4%), reinforcing the consistency of outcomes across studies. Individual drug estimates revealed that higher doses of Carbamazepine (2000 mg; ~17.8% incidence), Lamotrigine (500 mg; ~14.5%), and Oxcarbazepine (900 mg; ~15.3%) were significantly associated with ataxia, whereas Pregabalin showed a comparatively lower incidence (~7%). The random-effects model appropriately accounted for between-study variation and confirmed that removal of a single study did not alter the overall direction or significance of results, thereby strengthening confidence in the observed associations.

3.7. Publication Bias

Publication bias across the outcomes, falls, dizziness, somnolence, sedation, and vertigo, was assessed using Begg and Mazumdar’s rank correlation test and Egger’s regression intercept and visually represented as Funnel plots in Supplementary Figures S5–S10. For fall outcome, Kendall’s tau-b was −0.321 (p = 0.265), and Egger’s intercept was −0.889 (95% CI: −2.79 to 1.01; p = 0.296), suggesting no significant publication bias. Similarly, dizziness outcome assessment showed Kendall’s tau-b of −0.208 (p = 0.176) and Egger’s intercept of −0.874 (95% CI: −2.14 to 0.39; p = 0.164), again indicating no evidence of bias. For somnolence outcome, Begg’s tau-b was −0.244 (p = 0.246) and Egger’s intercept was −1.474 (95% CI: −4.31 to 1.36; p = 0.277), also non-significant. Sedation outcome demonstrated Kendall’s tau-b of −0.393 (p = 0.174) and Egger’s intercept of −0.591 (95% CI: −3.02 to 1.83; p = 0.573), with no bias indicated. Finally, vertigo outcome showed the strongest evidence against bias, with Kendall’s tau-b of −0.028 (p = 0.917) and Egger’s intercept of −0.196 (95% CI: −2.26 to 1.87; p = 0.829). Notably, only vertigo’s Trim and Fill suggested missing studies (n = 2), prompting a re-estimation of the pooled effect, which remained robust (from −2.442 to −2.552). For all other outcomes, the Trim and Fill analysis showed no missing studies. Overall, none of the Egger or Begg tests reached statistical significance, and no corrections were needed except for vertigo, where the adjusted estimate was still within the confidence bounds, supporting the reliability of the meta-analytic findings. These results suggest minimal to no publication bias across the outcomes, strengthening the validity of the observed adverse event rates.

4. Discussion

This comprehensive meta-analysis evaluated the incidence of key AEs related to falls in older patients prescribed AED for neuropathic pain. With fall incidence estimated at approximately 15.5% across RCTs, the results align with observational evidence highlighting AED-related fall risk in older adults. For example, Ensrud et al. (2002) [8,36] identified a 2.56-fold increased risk of frequent falls among women using anticonvulsants in a large U.S. cohort, while Dustin et al. (2006) [17] similarly reported significantly higher CNS drug use—including AEDs—among fall-related outpatient visits in veterans. Masud et al. (2013) [37] extended these findings to men, revealing 2–3-fold increased fall risks with anticonvulsants, SSRIs, and TCAs. Our RCT-based analysis echoes these associations, particularly noting elevated dizziness and sedation rates with gabapentin and pregabalin, and higher vertigo and ataxia rates at upper therapeutic doses of carbamazepine and lamotrigine. Notably, the pooled RCT data suggest a somewhat lower incidence of vertigo (~8%) and ataxia (~13%) than observational estimates, potentially due to stricter trial inclusion criteria, more controlled dosing, and shorter follow-up periods. Nonetheless, the convergence of evidence reinforces AEDs as key pharmacologic contributors to fall risk, especially when combined with polypharmacy and comorbid frailty. Furthermore, the more favorable AE profiles observed with pregabalin in terms of ataxia, and with lamotrigine in terms of dizziness, corroborate prior reports suggesting improved tolerability with newer AEDs.
The findings from this meta-analysis are strongly corroborated by extensive observational evidence demonstrating consistent associations between antiepileptic drug use and fall risk in older populations. Multiple systematic reviews have confirmed that AED use in ambulatory older adults leads to significantly increased fall risk, with relative risks ranging from 1.29 to 1.62 and odds ratios from 1.75 to 6.2 [8,39]. The meta-analysis findings align particularly well with large-scale epidemiological studies, including Ensrud et al.’s [36] prospective cohort of over 8000 women, which documented a 2.56-fold increased risk of frequent falls among anticonvulsant users. Similarly, Masud et al.’s [37] cross-sectional study of Danish men aged 60–75 years reported odds ratios of 2.8 for falls and 2.6 for recurrent falls with AED use. The observed drug-specific patterns are also supported by real-world evidence, with gabapentin and pregabalin showing dose-dependent associations with altered mental status and falls in hemodialysis patients, where gabapentin demonstrated 50–55% higher hazards of falls in the highest dose categories [40]. Furthermore, observational studies in diabetic peripheral neuropathy patients have confirmed that anticonvulsant use significantly increases fall risk (HR 1.34, 95% CI 1.18–1.51) [41], while polypharmacy studies demonstrate that fall risk is particularly elevated when AEDs are combined with other CNS-active medications in frail older adult populations.
A prior systematic review reported an odds ratio of 1.88 (95% CI: 1.02–3.49) for falls among community-dwelling older adults using AEDs, reinforcing the robustness of our pooled estimate of fall incidence at 15.5% across AED types [42]. Similarly, de Vries et al. confirmed that both opioids and AEDs are significantly associated with increased fall risk, validating our subgroup findings of drug-specific and dose-dependent fall-related adverse event (AE) profiles [43]. Notably, our analysis identified dizziness, somnolence, and sedation as frequent AEs—mirroring clinical reports where such symptoms are most commonly linked to gabapentin and pregabalin [44]. The heterogeneity observed in lacosamide, and gabapentin subgroups may be attributable to such dose variability. For example, pregabalin at 300 mg/day or higher has been shown to result in withdrawal due to AEs in 32% of patients, compared to only 5% in placebo arms [26]—a pattern closely aligned with our observed dropout-linked AEs. Collectively, these converging lines of evidence underscore the need for individualized prescribing, incorporating careful dose titration, fall-risk screening, and shared decision-making in older patients receiving AEDs for neuropathic pain.
In addition to pharmacologic considerations, these results underscore the need for routine fall-risk assessment when prescribing CNS-active medications to older patients [8,39]. Integrating evidence-based fall prevention strategies—such as balance training, home hazard assessment, and regular medication review—may mitigate risk. Emerging evidence also suggests that non-pharmacological pain interventions and cautious deprescribing of CNS-active drugs could further reduce harm, an area that warrants pragmatic trials with falls as primary endpoints. Our dose-stratified analyses further expand on these findings, identifying elevated risks of dizziness, sedation, and ataxia—particularly at higher AED doses such as gabapentin (≥2400 mg) and pregabalin (600 mg). While Seppälä et al. [8] identified substantial heterogeneity across studies, our low heterogeneity supports consistency in specific AED-related adverse event patterns.
In addition, potential drug–drug interactions warrant consideration. AEDs are frequently prescribed alongside other CNS-active agents such as benzodiazepines, hypnotics, and sleep medications in older adults [45,46]. The combined sedative and vestibular effects of these agents may synergistically increase fall risk, compounding the dose-dependent risks we observed with AEDs alone [47]. This highlights the need for careful medication reconciliation, deprescribing strategies where appropriate, and close monitoring of patients receiving multiple CNS-active drugs.
This analysis has several limitations. Heterogeneity in study designs and variable definitions of adverse events may have introduced residual confounding. Many studies reported falls as secondary outcomes, possibly underestimating true incidence, and a substantial number of full texts (n = 542) could not be accessed due to subscription or institutional restrictions, and only English-language publications were included. While comprehensive database searches, manual reference screening, and cross-study comparisons were undertaken to minimize selection bias, it remains possible that relevant evidence was excluded. Nonetheless, the overall low heterogeneity and consistency across drug-specific estimates strengthen confidence in these findings.
In summary, this review reinforces the significant and dose-dependent fall risk posed by AEDs in older adults. Clinicians should prioritize shared decision-making, conservative titration, and individualized prescribing, balancing analgesic benefit against fall-related harms. Future research should further delineate dose–response relationships, identify subgroups at highest risk, and evaluate integrated deprescribing strategies to optimize safe neuropathic pain management in geriatric populations.

5. Conclusions

This comprehensive meta-analysis confirms that antiepileptic drugs prescribed for neuropathic pain pose a significant and clinically meaningful fall risk for older adults, with an overall incidence of approximately 15% and clear dose-dependent patterns. While gabapentin appears to have a comparatively lower risk, higher doses of carbamazepine and lamotrigine are associated with greater neuropsychiatric adverse events such as dizziness, sedation, and ataxia. These findings highlight the importance of routine fall-risk screening, conservative dose titration, and preference for safer agents when appropriate. Clinicians should carefully balance analgesic benefit against fall-related harm, incorporating non-pharmacological interventions and deprescribing strategies where feasible. Future research should prioritize pragmatic trials that directly evaluate fall outcomes and inform evidence-based guidelines for optimizing neuropathic pain management in frail older populations.

Supplementary Materials

The following supporting information can be downloaded at https://www.mdpi.com/article/10.3390/geriatrics10050130/s1, Figures S1–S5 (sensitivity analyses for falls, dizziness, somnolence, sedation, and ataxia outcomes); Figures S6–S11 (funnel plots assessing publication bias); Table S1 (detailed electronic search strategies); and Table S2 (GRADE evidence summary of included studies).

Author Contributions

A.V.: Conceptualization, Methodology, Data curation, Formal analysis, Investigation, Writing—Original Draft, Project administration, Supervision, Validation. V.V.: Conceptualization, Methodology, Data curation, Formal analysis, Investigation, Writing—Original Draft, Project administration, Supervision, Validation. F.M.: Literature search, screening, and manuscript review. N.Y.: Data extraction, quality appraisal, and manuscript review. All authors have read and agreed to the published version of the manuscript.

Funding

No specific funding was received for this work. The review was conducted as part of the authors’ academic and clinical research development.

Institutional Review Board Statement

Not applicable. This systematic review and meta-analysis did not involve the use of human participants, human data, or human tissue.

Informed Consent Statement

Not applicable. The manuscript does not contain data from any individual person.

Data Availability Statement

All data generated or analyzed during this study are included in this published article and its Supplementary Files. Forest plots, risk of bias assessments, and GRADE evidence tables are available within manuscript.

Acknowledgments

We would like to thank the NIHR Clinical Research Facility at Liverpool University Hospitals NHS Foundation Trust for supporting the lead author’s academic development. We also acknowledge institutional training support from the University of Salford.

Conflicts of Interest

The authors declare that they have no competing interests.

References

  1. Van Hecke, O.; Austin, S.K.; Khan, R.A.; Smith, B.H.; Torrance, N. Neuropathic pain in the general population: A systematic review of epidemiological studies. Pain 2014, 155, 654–662. [Google Scholar] [CrossRef] [PubMed]
  2. Schug, S.A.; Stannard, K.J. Treatment of Neuropathic Pain. J. Korean Med. Assoc. 2011, 64, 484–490. [Google Scholar] [CrossRef]
  3. Haasum, Y.; Johnell, K. Use of antiepileptic drugs and risk of falls in old age: A systematic review. Epilepsy Res. 2017, 138, 98–104. [Google Scholar] [CrossRef]
  4. Ambrose, A.F.; Paul, G.; Hausdorff, J.M. Risk factors for falls among older adults: A review of the literature. Maturitas 2013, 75, 51–61. [Google Scholar] [CrossRef]
  5. Falls [Online]. Available online: https://www.who.int/news-room/fact-sheets/detail/falls (accessed on 2 October 2025).
  6. Hartikainen, S.; Lönnroos, E.; Louhivuori, K. Medication as a risk factor for falls: Critical systematic review. J. Gerontol. Ser. A Biol. Sci. Med. Sci. 2007, 62, 1172–1181. [Google Scholar] [CrossRef]
  7. Florisson, S.; Aagesen, E.K.; Bertelsen, A.S.; Nielsen, L.P.; Rosholm, J.U. Are older adults insufficiently included in clinical trials?—An umbrella review. Basic. Clin. Pharmacol. Toxicol. 2021, 128, 213–223. [Google Scholar] [CrossRef]
  8. Seppala, L.J.; van de Glind, E.M.; Daams, J.G.; Ploegmakers, K.J.; de Vries, M.; Wermelink, A.M.; van der Velde, N.; Blain, H.; Bousquet, J.; Bucht, G.; et al. Fall-Risk-Increasing Drugs: A Systematic Review and Meta-analysis: III. Others. J. Am. Med. Dir. Assoc. 2018, 19, 372.e1–372.e8. [Google Scholar] [CrossRef]
  9. Zia, A.; Kamaruzzaman, S.B.; Tan, M.P. The consumption of two or more fall risk-increasing drugs rather than polypharmacy is associated with falls. Geriatr. Gerontol. Int. 2017, 17, 463–470. [Google Scholar] [CrossRef]
  10. Dassieu, L.; Paul-Savoie, E.; Develay, É.; Guilhon, A.C.V.; Lacasse, A.; Guénette, L.; Perreault, K.; Beaudry, H.; Dupuis, L.; The Quebec Consortium on Adverse Effects of Pain Medications. Swallowing the pill of adverse effects: A qualitative study of patients’ and pharmacists’ experiences and decision-making regarding the adverse effects of chronic pain medications. Health Expect. 2021, 25, 394. [Google Scholar] [CrossRef] [PubMed]
  11. Hilmer, S.N.; Gnjidic, D. The effects of polypharmacy in older adults. Clin. Pharmacol. Ther. 2009, 85, 86–88. [Google Scholar] [CrossRef] [PubMed]
  12. Zeber, J.E.; Copeland, L.A.; Pugh, M.J.V. Variation in antiepileptic drug adherence among older patients with new-onset epilepsy. Ann. Pharmacother. 2010, 44, 1896–1904. [Google Scholar] [CrossRef] [PubMed]
  13. Sterne, J.A.C.; Savović, J.; Page, M.J.; Elbers, R.G.; Blencowe, N.S.; Boutron, I.; Cates, C.J.; Cheng, H.Y.; Corbett, M.S.; Eldridge, S.M.; et al. RoB 2: A revised tool for assessing risk of bias in randomised trials. BMJ 2019, 366, l4898. [Google Scholar] [CrossRef]
  14. Sterne, J.A.C.; Hernán, M.A.; Reeves, B.C.; Savović, J.; Berkman, N.D.; Viswanathan, M.; Henry, D.; Altman, D.G.; Ansari, M.T.; Boutron, I.; et al. ROBINS-I: A tool for assessing risk of bias in non-randomised studies of interventions. BMJ 2016, 355, I4919. [Google Scholar] [CrossRef]
  15. Guyatt, G.H.; Oxman, A.D.; Vist, G.E.; Kunz, R.; Falck-Ytter, Y.; Alonso-Coello, P.; Schünemann, H.J. GRADE: An emerging consensus on rating quality of evidence and strength of recommendations. BMJ Br. Med. J. 2008, 336, 924. [Google Scholar] [CrossRef]
  16. Jensen-Dahm, C.; Rowbotham, M.C.; Reda, H.; Petersen, K.L. Effect of a single dose of pregabalin on herpes zoster pain. Trials 2011, 12, 55. [Google Scholar] [CrossRef]
  17. French, D.D.; Campbell, R.; Spehar, A.; Cunningham, F.; Bulat, T.; Luther, S.L. Drugs and falls in community-dwelling older people: A national veterans study. Clin. Ther. 2006, 28, 619–630. [Google Scholar] [CrossRef]
  18. Robertson, K.; Marshman, L.A.G.; Plummer, D.; Downs, E. Effect of Gabapentin vs Pregabalin on Pain Intensity in Adults With Chronic Sciatica: A Randomized Clinical Trial. JAMA Neurol. 2018, 76, 28. [Google Scholar] [CrossRef]
  19. Richards, K.C.; Fry, L.M.; Lozano, A.J.; Ji, W.; Morrison, J.D.; Britt, K.C.; Bliwise, D.L.; Gooneratne, N.S.; Hanlon, A.L. Treatment of Restless Legs Syndrome Improves Agitation and Sleep in Persons with Dementia: A Randomized Trial. J. Am. Med. Dir. Assoc. 2025, 26, 105485. [Google Scholar] [CrossRef]
  20. Saetre, E.; Abdelnoor, M.; Amlie, J.P.; Tossebro, M.; Perucca, E.; Taubøll, E.; Anfinsen, O.G.; Isojärvi, J.; Gjerstad, L. Cardiac function and antiepileptic drug treatment in the elderly: A comparison between lamotrigine and sustained-release carbamazepine. Epilepsia 2009, 50, 1841–1849. [Google Scholar] [CrossRef] [PubMed]
  21. Brodie, M.J.; Overstall, P.W.; Giorgi, L. Multicentre, double-blind, randomised comparison between lamotrigine and carbamazepine in elderly patients with newly diagnosed epilepsy. Epilepsy Res. 1999, 37, 81–87. [Google Scholar] [CrossRef] [PubMed]
  22. Sajatovic, M.; Gildengers, A.; Al Jurdi, R.K.; Gyulai, L.; Cassidy, K.A.; Greenberg, R.L.; Bruce, M.L.; Mulsant, B.H.; Have, T.T.; Young, R.C. Multisite, open-label, prospective trial of lamotrigine for geriatric bipolar depression: A preliminary report. Bipolar Disord. 2011, 13, 294–302. [Google Scholar] [CrossRef]
  23. Roose, S.P.; Nelson, J.C.; Salzman, C.; Hollander, S.B.; Rodrigues, H. Open-label study of mirtazapine orally disintegrating tablets in depressed patients in the nursing home. Curr. Med. Res. Opin. 2003, 19, 737–746. [Google Scholar] [CrossRef]
  24. Alvarado, A.M.; Navarro, S.A. Clinical trial assessing the efficacy of gabapentin plus B complex (B1/B12) versus pregabalin for treating painful diabetic neuropathy. J. Diabetes Res. 2016, 2016, 1–8. [Google Scholar] [CrossRef]
  25. Saetre, E.; Perucca, E.; Isojärvi, J.; Gjerstad, L. An international multicenter randomized double-blind controlled trial of lamotrigine and sustained-release carbamazepine in the treatment of newly diagnosed epilepsy in the elderly. Epilepsia 2007, 48, 1292–1302. [Google Scholar] [CrossRef] [PubMed]
  26. Freynhagen, R.; Strojek, K.; Griesing, T.; Whalen, E.; Balkenohl, M. Efficacy of pregabalin in neuropathic pain evaluated in a 12-week, randomised, double-blind, multicentre, placebo-controlled trial of flexible- and fixed-dose regimens. Pain 2005, 115, 254–263. [Google Scholar] [CrossRef]
  27. Dworkin, R.H.; Barbano, R.L.; Tyring, S.K.; Betts, R.F.; McDermott, M.P.; Pennella-Vaughan, J.; Bennett, G.J.; Berber, E.; Gnann, J.W.; Irvine, C.; et al. A randomized, placebo-controlled trial of oxycodone and of gabapentin for acute pain in herpes zoster. Pain 2009, 142, 209–217. [Google Scholar] [CrossRef] [PubMed]
  28. Holbech, J.V.; Bach, F.W.; Finnerup, N.B.; Brøsen, K.; Jensen, T.S.; Sindrup, S.H. Imipramine and pregabalin combination for painful polyneuropathy: A randomized controlled trial. Pain 2015, 156, 958–966. [Google Scholar] [CrossRef]
  29. Sommer, O.H.; Aga, O.; Cvancarova, M.; Olsen, I.C.; Selbaek, G.; Engedal, K. Effect of oxcarbazepine in the treatment of agitation and aggression in severe dementia. Dement. Geriatr. Cogn. Disord. 2009, 27, 155–163. [Google Scholar] [CrossRef]
  30. Wymer, J.P.; Simpson, J.; Sen, D.; Bongardt, S. Efficacy and safety of lacosamide in diabetic neuropathic pain: An 18-week double-blind placebo-controlled trial of fixed-dose regimens. Clin. J. Pain 2009, 25, 376–385. [Google Scholar] [CrossRef] [PubMed]
  31. Tesfaye, S.; Sloan, G.; Petrie, J.; White, D.; Bradburn, M.; Julious, S.; Rajbhandari, S.; Sharma, S.; Rayman, G.; Gouni, R.; et al. Comparison of amitriptyline supplemented with pregabalin, pregabalin supplemented with amitriptyline, and duloxetine supplemented with pregabalin for the treatment of diabetic peripheral neuropathic pain (OPTION-DM): A multicentre, double-blind, randomised crossover trial. Lancet 2022, 400, 680–690. [Google Scholar] [CrossRef]
  32. Luukinen, H.; Koski, K.; Laippala, P.; Kivelä, S.L. Predictors for recurrent falls among the home-dwelling elderly. Scand. J. Prim. Health Care 1995, 13, 294–299. [Google Scholar] [CrossRef] [PubMed]
  33. Tromp, A.M.; Smit, J.H.; Deeg, D.J.H.; Bouter, L.M.; Lips, P. Predictors for falls and fractures in the longitudinal aging study Amsterdam. J. Bone Miner. Res. 1998, 13, 1932–1939. [Google Scholar] [CrossRef] [PubMed]
  34. Kelly, K.D.; Pickett, W.; Yiannakoulias, N.; Rowe, B.H.; Schopflocher, D.P.; Svenson, L.; Voaklander, D.C. Medication use and falls in community-dwelling older persons. Age Ageing 2003, 32, 503–509. [Google Scholar] [CrossRef]
  35. Mayo, N.E.; Korner-Bitensky, N.; Becker, R.; Georges, P. Predicting falls among patients in a rehabilitation hospital. Am. J. Phys. Med. Rehabil. 1989, 68, 139–146. [Google Scholar] [CrossRef]
  36. Ensrud, K.E.; Blackwell, T.L.; Mangione, C.M.; Bowman, P.J.; Whooley, M.A.; Bauer, D.C.; Schwartz, A.V.; Hanlon, J.T.; Nevitt, M.C. Central nervous system-active medications and risk for falls in older women. J. Am. Geriatr. Soc. 2002, 50, 1629–1637. [Google Scholar] [CrossRef]
  37. Masud, T.; Frost, M.; Ryg, J.; Matzen, L.; Ibsen, M.; Abrahamsen, B.; Brixen, K. Central nervous system medications and falls risk in men aged 60-75 years: The study on male osteoporosis and aging (SOMA). Age Ageing 2013, 42, 121–124. [Google Scholar] [CrossRef]
  38. Titler, M.G.; Shever, L.L.; Kanak, M.F.; Picone, D.M.; Qin, R. Factors associated with falls during hospitalization in an older adult population. Res. Theory Nurs. Pract. 2011, 25, 127–148. [Google Scholar] [CrossRef]
  39. Maximos, M.; Chang, F.; Patel, T. Risk of falls associated with antiepileptic drug use in ambulatory elderly populations: A systematic review. Can. Pharm. J. CPJ 2017, 150, 101. [Google Scholar] [CrossRef]
  40. Ishida, J.H.; McCulloch, C.E.; Steinman, M.A.; Grimes, B.A.; Johansen, K.L. Gabapentin and Pregabalin Use and Association with Adverse Outcomes among Hemodialysis Patients. J. Am. Soc. Nephrol. 2018, 29, 1970–1978. [Google Scholar] [CrossRef]
  41. Randolph, A.; Lin, Y.; Volpi, E.; Kuo, Y. Impact of Anticonvulsant Use on Incidence of Falls in Older Patients with Diabetic Peripheral Neuropathy. Innov. Aging 2018, 2, 151. [Google Scholar] [CrossRef]
  42. Woolcott, J.C.; Richardson, K.J.; Wiens, M.O.; Patel, B.; Marin, J.; Khan, K.M.; Marra, C.A. Meta-analysis of the impact of 9 medication classes on falls in elderly persons. Arch. Intern. Med. 2009, 169, 1952–1960. [Google Scholar] [CrossRef]
  43. de Vries, M.; Seppala, L.J.; Daams, J.G.; van de Glind, E.M.; Masud, T.; van der Velde, N.; Blain, H.; Bousquet, J.; Bucht, G.; Caballero-Mora, M.A.; et al. Fall-Risk-Increasing Drugs: A Systematic Review and Meta-Analysis: I. Cardiovascular Drugs. J. Am. Med. Dir. Assoc. 2018, 19, 371.e1–371.e9. [Google Scholar] [CrossRef] [PubMed]
  44. Zaccara, G.; Gangemi, P.; Perucca, P.; Specchio, L. The adverse event profile of pregabalin: A systematic review and meta-analysis of randomized controlled trials. Epilepsia 2011, 52, 826–836. [Google Scholar] [CrossRef] [PubMed]
  45. Bauquier, S.H. Benzodiazepines. In Pharmacology in Veterinary Anesthesia and Analgesia; Wiley: Hoboken, NJ, USA, 2024; pp. 60–74. [Google Scholar] [CrossRef]
  46. Rozing, M.P.; Wium-Andersen, M.K.; Wium-Andersen, I.K.; Jørgensen, T.S.H.; Jørgensen, M.B.; Osler, M. Use of hypnotic-sedative medication and risk of falls and fractures in adults: A self-controlled case series study. Acta Psychiatr. Scand. 2023, 148, 394–404. [Google Scholar] [CrossRef] [PubMed]
  47. Masse, M.; Henry, H.; Cuvelier, E.; Pinçon, C.; Pavy, M.; Beeuwsaert, A.; Barthélémy, C.; Cuny, D.; Gautier, S.; Kambia, N.; et al. Sleep Medication in Older Adults: Identifying the Need for Support by a Community Pharmacist. Healthcare 2022, 10, 147. [Google Scholar] [CrossRef]
Geriatrics 10 00130 g001
Figure 1. PRISMA flow chart.
Figure 1. PRISMA flow chart.
Geriatrics 10 00130 g001
Geriatrics 10 00130 g002
Figure 2. Traffic Light Plots on RoB of included RCTs [16,18,19,21,22,23,24,25,26,27,28,29,30].
Figure 2. Traffic Light Plots on RoB of included RCTs [16,18,19,21,22,23,24,25,26,27,28,29,30].
Geriatrics 10 00130 g002
Geriatrics 10 00130 g003
Figure 3. Traffic Light plots on the RoB assessment on included non-RCT studies [17,32,33,34,35,36,37,38].
Figure 3. Traffic Light plots on the RoB assessment on included non-RCT studies [17,32,33,34,35,36,37,38].
Geriatrics 10 00130 g003
Geriatrics 10 00130 g004
Figure 4. Forest plot showing pooled incidence of falls in older adults treated with AEDs from the included studies [16,19,22,23,27,29,31].
Figure 4. Forest plot showing pooled incidence of falls in older adults treated with AEDs from the included studies [16,19,22,23,27,29,31].
Geriatrics 10 00130 g004
Geriatrics 10 00130 g005
Figure 5. Forest plot showing incidence of dizziness associated with AED use in older adults from included studies [16,18,20,21,23,24,26,27,28,29,30,31].
Figure 5. Forest plot showing incidence of dizziness associated with AED use in older adults from included studies [16,18,20,21,23,24,26,27,28,29,30,31].
Geriatrics 10 00130 g005
Geriatrics 10 00130 g006
Figure 6. Forest plot showing incidence of somnolence (drowsiness) with AED use in older adults included in the studies [19,20,21,23,24,25,26].
Figure 6. Forest plot showing incidence of somnolence (drowsiness) with AED use in older adults included in the studies [19,20,21,23,24,25,26].
Geriatrics 10 00130 g006
Geriatrics 10 00130 g007
Figure 7. Forest plot of sedation incidence linked to AEDs in older adults involved in the included studies [16,18,19,27,29,31].
Figure 7. Forest plot of sedation incidence linked to AEDs in older adults involved in the included studies [16,18,19,27,29,31].
Geriatrics 10 00130 g007
Geriatrics 10 00130 g008
Figure 8. Forest plot showing incidence of vertigo following AED treatment in older populations from various studies [18,21,24,26,30].
Figure 8. Forest plot showing incidence of vertigo following AED treatment in older populations from various studies [18,21,24,26,30].
Geriatrics 10 00130 g008
Geriatrics 10 00130 g009
Figure 9. Forest plot summarizing ataxia incidence across included studies of AEDs in older adults [21,28,29,31].
Figure 9. Forest plot summarizing ataxia incidence across included studies of AEDs in older adults [21,28,29,31].
Geriatrics 10 00130 g009
Table 1. Characteristics of included studies evaluating AED-related fall risk in older adults.
Table 1. Characteristics of included studies evaluating AED-related fall risk in older adults.
StudyCountrySample SizeMean AgeDiagnosisInterventionsKey Findings
Robertson et al., 2018 [18]Australia1857 ± 16.5Chronic SciaticaGabapentin vs. PregabalinGabapentin superior in pain reduction and fewer adverse events compared to pregabalin.
Richards et al., 2025 [19]USA14783.4 ± 9.1RLS with dementiaGabapentin enacarbil vs. PlaceboReduced nighttime agitation; trend toward more falls in the gabapentin group (p = 0.066).
Saetre et al., 2009 [20]Norway108≥65Newly diagnosed epilepsyCarbamazepine vs. LamotrigineNo significant ECG changes; both drugs tolerated well in older Adults.
Brodie et al., 1999 [21]UK/Europe15077Newly diagnosed epilepsyLamotrigine vs. CarbamazepineLamotrigine had fewer adverse events and better continuation than carbamazepine.
Sajatovic et al., 2011 [22]USA5766.5 ± 6.7Bipolar depressionLamotrigine (augmentation)Improvement in depression and function; some adverse events including unsteady gait.
Roose et al., 2003 [23]USA119≥70DepressionMirtazapine (15–45 mg/day)Effective in depression; common AEs were falls (18%) and somnolence (12%).
Alvarado et al., 2016 [24]Mexico270>50Painful diabetic neuropathyGabapentin + B1/B12 vs. PregabalinComparable efficacy; less vertigo with gabapentin/B1/B12 (p = 0.014); lower gabapentin doses needed for pain relief.
Saetre et al., 2007 [25]Norway/Multicenter185≥65Newly diagnosed epilepsyLamotrigine (25–500 mg/day), Carbamazepine SR (100–2000 mg/day)LTG and CBZ had similar effectiveness; CBZ had higher seizure freedom, LTG better tolerability.
Freynhagen et al., 2005 [26]Multicenter33862.7 ± 10.6DPN or PHN (neuropathic pain)Pregabalin (flexible: 150–600 mg/day or fixed: 300–600 mg/day)Both regimens reduced pain and sleep interference; dizziness and somnolence common AEs.
Dworkin et al., 2009 [27]USA87≥50Acute pain in herpes zosterGabapentin (1200–1800 mg/day), CR-oxycodone, placeboOxycodone reduced pain more than placebo; gabapentin had modest effect; common AEs included constipation.
Jensen-Dahm et al., 2011 [16]USA865Acute herpes zoster painPregabalin (150 mg single dose)33% pain reduction vs. 14% placebo; well tolerated; effects on allodynia not significant.
Holbech et al. (2015) [28]Multicenter (EU)7320–85 yearsPainful polyneuropathyImipramine 75 mg/day vs. Pregabalin 300 mg/day vs. Combination vs. PlaceboCombination significantly more effective than either monotherapy but had more adverse events.
Sommer et al. (2009) [29]Germany10383Agitation/aggression in dementiaOxcarbazepine vs. PlaceboNo significant difference in agitation/aggression; slight trend favoring oxcarbazepine.
Wymer et al. (2009) [30]InternationalNR58Painful diabetic neuropathyLacosamide 200 mg/d, 400 mg/d, 600 mg/d vs. PlaceboLacosamide 400 mg/d showed optimal efficacy/tolerability; 600 mg/d had high AE dropout rate.
Tesfaye et al. (2022) [31]UK (13 sites)130Mean: 60 sDiabetic peripheral neuropathic pain (DPNP)A-P, P-A, D-P pathways combining amitriptyline, duloxetine, pregabalinAll had similar efficacy; combination better than monotherapy if pain relief suboptimal.
Dustin et al., 2006 [17]USA41,102 (20,551 cases, 20,551 controls)>65Fall-related outpatient visits vs. nonspecific chest painAnalysis of CNS, CVS, and MSS drugs prescribedCNS drugs (e.g., antidepressants, anticonvulsants, antipsychotics) more common in fall patients. CVS drugs more common in controls. Highlighted medication-related fall risk.
Luukinen et al., 1995 [32]Finland1016≥70Recurrent falls in home-dwelling older AdultsCommunity-based observation and fall historyPrior falls, peripheral neuropathy, psychotropics, and slow gait identified as predictors of recurrent falls.
Tromp et al., 1998 [33]Netherlands1469>60 (Born before 1931)Falls, recurrent falls, fracturesPopulation-based cohort analysisImpaired mobility, analgesics, and antiepileptics predicted recurrent falls. Fractures predicted by inactivity, female gender, and prior fractures.
Kelly et al., 2003 [34]Netherlands1469>60Recurrent falls and fracturesBaseline interview with follow-up over 38 monthsSimilarly to Tromp et al.: identified analgesics and inactivity as strong predictors. Recurrent falls occurred in 15% of the cohort.
Mayo et al., 1989 [35]Canada402 (201 cases, 201 controls)>60Falls in rehabilitation hospitalRetrospective case–control from admission recordsStroke, incontinence, anticonvulsants, and topical eye meds predicted falls. Risk model validated in second cohort year.
Ensrud et al., 2002 [36]USA8127 women≥65 yearsCommunity-dwelling older womenCNS-active drugs: benzodiazepines, antidepressants, anticonvulsants, narcoticsUse of benzodiazepines (OR: 1.51), antidepressants (OR: 1.54), and anticonvulsants (OR: 2.56) significantly increased risk of frequent falls. Narcotics not associated.
Masud et al., 2013 [37]Denmark4696 menMedian: 66.3 yearsMen aged 60–75 years, general populationCNS drugs: opiates, antidepressants, anxiolytics, SSRIs, TCAsOpiates (OR: 2.4), antidepressants (OR: 2.8), SSRIs (OR: 3.1), TCAs (OR: 2.2), and antiepileptics (OR: 2.8) significantly associated with falls.
Titler et al., 2011 [38]USA10,187 hospitalizations≥60 yearsHospitalized older adultsMedical/pharmacy/nursing treatments; CNS meds; fall prevention interventionsAntidepressants, antipsychotics, benzodiazepines, restraints, and neurologic monitoring linked to falls. RN skill mix, ulcer care, and pain management inversely related.
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Vamadevan, A.; Vijayan, V.; Marwein, F.; Yoosuf, N. Fall-Related Adverse Events of Anti-Epileptic Drugs Used for Neuropathic Pain in Older Adults: A Systematic Review and Meta-Analysis. Geriatrics 2025, 10, 130. https://doi.org/10.3390/geriatrics10050130

AMA Style

Vamadevan A, Vijayan V, Marwein F, Yoosuf N. Fall-Related Adverse Events of Anti-Epileptic Drugs Used for Neuropathic Pain in Older Adults: A Systematic Review and Meta-Analysis. Geriatrics. 2025; 10(5):130. https://doi.org/10.3390/geriatrics10050130

Chicago/Turabian Style

Vamadevan, Arun, Vijesh Vijayan, Fellisha Marwein, and Nishad Yoosuf. 2025. "Fall-Related Adverse Events of Anti-Epileptic Drugs Used for Neuropathic Pain in Older Adults: A Systematic Review and Meta-Analysis" Geriatrics 10, no. 5: 130. https://doi.org/10.3390/geriatrics10050130

APA Style

Vamadevan, A., Vijayan, V., Marwein, F., & Yoosuf, N. (2025). Fall-Related Adverse Events of Anti-Epileptic Drugs Used for Neuropathic Pain in Older Adults: A Systematic Review and Meta-Analysis. Geriatrics, 10(5), 130. https://doi.org/10.3390/geriatrics10050130

Article Metrics

Article metric data becomes available approximately 24 hours after publication online.
Back to TopTop