Background:Heterogeneity in schizophrenia (SCZ) remains a major obstacle to the development of effective treatment. Aims:This study aimed to identify reproducible neurobiological subtypes of SCZ using source-localised, high-density resting-state electroencephalography (EEG) functional connectivity patterns. Methods:We analysed high-density resting-state EEG data from a discovery dataset of 86 patients with SCZ and an independent replication dataset of 89 patients. We applied a sparse k-means clustering algorithm to power envelope correlations, computed across five frequency bands and two resting conditions in the source space. Subtype stability was rigorously assessed using split-half validation and cross-cohort replication. Results:The analysis consistently identified two reproducible SCZ subtypes. One subtype was characterised by significant temporoparietal disconnection compared with healthy controls, while the second subtype exhibited a relatively preserved functional connectivity pattern. These subtypes were stable within the cohort and successfully replicated in the independent dataset. Crucially, connectivity features significantly predicted symptom severity, but this association was specific to the more disconnected subtype. Conclusions:Source-space EEG connectivity enables the identification of robust, reproducible, and mechanistically distinct subtypes of SCZ. These findings highlight underlying neurobiological heterogeneity that is not captured by standard clinical assessments and suggest that these subtypes may provide a basis for future biomarker development and patient stratification studies.
Background:Alzheimer's disease (AD), a neurodegenerative disorder, is pathologically defined by the accumulation of amyloid-β (Aβ) plaques, hyperphosphorylated tau tangles and sustained neuroinflammation. Glucagon-like peptide-1 (GLP-1), originally characterised as an incretin hormone, is also expressed endogenously in the brain. Although GLP-1 receptor agonists (GLP-1RAs) have demonstrated compelling neuroprotective effects in preclinical AD models, their efficacy in human clinical trials has, thus far, been inconsistent and limited. Moreover, the physiological role and dynamics of endogenous GLP-1 during AD progression remain poorly understood, with studies reporting conflicting findings across different patient cohorts. Aims:This study aimed to characterise serum GLP-1 levels across the Alzheimer's disease cognitive continuum and to examine their associations with neuroinflammation, Aβ pathology, tau pathology, brain atrophy and cognitive decline. Methods:We analysed two independent cohorts (Renji: n = 123; SheMountain: n = 127), including participants with cognitive unimpairment, mild cognitive impairment and AD dementia. Serum biomarkers (GLP-1, glial fibrillary acidic protein [GFAP], Aβ40, Aβ42 and tau phosphorylated at threonine 217 [pTau217]) were quantified, along with APOE genotyping, neuropsychological assessments, brain magnetic resonance imaging and Aβ-PET imaging. Postmortem brain tissue GLP-1 expression was analysed in the prefrontal cortex (PFC) and hippocampus (HP). Statistical analyses included partial correlations, mediation analysis and machine learning models. Results:GLP-1 levels were significantly elevated in both brain tissue and serum of patients with AD compared to controls, with concentrations increasing progressively from cognitively normal individuals through mild cognitive impairment (MCI) to AD dementia. Serum GLP-1 negatively correlated with Montreal Cognitive Assessment (MoCA) cognitive performance and positively correlated with GFAP, pTau217 and brain amyloid burden. Serial mediation analysis suggested that GLP-1 and GFAP were statistically positioned along the association between amyloid burden and tau pathology. Machine learning models combining GLP-1 with other biomarkers achieved diagnostic performance (area under the curve = 0.896). Conclusions:The elevation of endogenous GLP-1 in Alzheimer's disease likely represents a compensatory yet ultimately insufficient response to the underlying pathology. It functions as a multidimensional biomarker, reflecting the integrated neuroinflammatory burden, amyloid deposition and clinical disease severity, and may have potential utility in diagnostic applications.
Background:The comparative liability for abuse and dependence of drugs used for attention deficit hyperactivity disorder (ADHD) has been poorly explored so far. Aims:To evaluate the relative potential of medications used for ADHD for generating spontaneous reports of abuse and dependence in international pharmacovigilance data. Methods:A disproportionality analysis was conducted using data recorded in VigiBase®, the World Health Organisation global database of individual case safety reports (ICSRs). ICSRs recorded between 1 January 2002 and 30 June 2023, referring to paediatric and adult subjects with known sex and aged ≥ 15 years and involving suspect drugs used for ADHD, were included. Reporting odds ratios (RORs) for 'Drug abuse and dependence', along with their 95% confidence intervals (CIs), were calculated using methylphenidate as the reference drug and with adjustment for age, sex, seriousness, reporter qualification and continent of reporting. Several sensitivity analyses were performed. Results:Among the 55 219 ICSRs included, 2633 were reports of drug abuse and dependence. The proportion of drug abuse and dependence reports was significantly lower for all studied drugs than for methylphenidate, except for lisdexamfetamine (ROR: 1.06, 95% CI 0.91-1.24). Sensitivity analyses yielded results comparable to the primary analysis, but with a slightly higher ROR for lisdexamfetamine when excluding US data (ROR: 1.26, 95% CI 1.03-1.54). Conclusions:Methylphenidate and lisdexamfetamine emerged as the two drugs with the highest likelihood of safety reports related to abuse and dependence. These findings challenge previous statements suggesting that the abuse potential of lisdexamfetamine was lower than that of other psychostimulant drugs used for ADHD and advocate for further pharmacoepidemiological studies in this field.
Objective:Estimating clinically important differences (CIDs) for patient-reported outcome measures (PROMs) is essential for translating statistical findings into clinically meaningful treatment effects. We developed estimating clinically important differences (ESTICID), a cross-platform toolkit implemented in both R Shiny and statistical analysis system, to estimate and visualise CIDs in longitudinal studies. Methods:ESTICID applies an anchor-based approach using linear mixed-effects models with random intercepts. The framework accommodates repeated measures, supports both continuous and categorical anchors and handles incomplete follow-up data under a missing-at-random assumption. A multicenter randomised controlled trial dataset was used for illustration. Examples:Using the illustrative dataset, we show that ESTICID generates clinically interpretable CID estimates across different levels of improvement and provides graphical outputs that facilitate interpretation. In simulation analyses, the estimates remained stable under increasing levels of missing follow-up data. The R Shiny implementation further enables interactive, browser-based use without programming. Conclusions:ESTICID provides a practical and accessible framework for anchor-based CID estimation in longitudinal PROM research. By integrating methodological rigour with cross-platform implementation, it may improve the interpretation of treatment effects, responder definitions, sample size estimation and evidence-to-decision processes in psychiatry and related clinical fields.
Background:Response inhibition is a fundamental component of executive control and a transdiagnostic mechanism underpinning numerous mental disorders. Transcranial direct current stimulation (tDCS) has been investigated as a potential neuromodulatory intervention to enhance response inhibition; however, findings remain inconsistent due to methodological heterogeneity in stimulation parameters and outcome measures. Aims:To quantitatively evaluate the effects of tDCS on response inhibition as measured by the stop-signal task (SST) and to explore potential moderators influencing tDCS efficacy. Methods:This meta-analysis was conducted in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses. We systematically searched PubMed/MEDLINE, Web of Science, PsycINFO, Embase and Scopus up to 1 April 2026. Quality assessment was conducted utilising the Cochrane risk of bias 2 tool, with random-effects meta-analysis, moderator and subgroup analyses and sensitivity analysis for statistical evaluation. Results:The search yielded 2982 articles, of which 35 were considered eligible for inclusion, including 1768 participants. The risk of bias 2 assessment indicated acceptable methodological quality, with the majority of studies rated as low risk of bias or as having some concerns. The overall effect of tDCS on response inhibition was modest yet statistically significant (Hedges' g = -0.22, 95% confidence interval (CI) -0.34 to -0.11, p < 0.001). Subgroup analysis indicated a small-to-moderate effect of anodal tDCS (Hedges' g = -0.38, 95% CI -0.51 to -0.26, p < 0.001), especially when targeting the right inferior frontal gyrus, right dorsolateral prefrontal cortex or motor-related cortex; additionally, online tDCS demonstrated greater efficacy than offline tDCS. Cathodal tDCS demonstrated a small but statistically significant detrimental effect on response inhibition (Hedges' g = 0.19, 95% CI 0.03-0.35, p = 0.023). Conclusions:Our findings support the potential of anodal tDCS as a neuromodulatory approach for enhancing response inhibition. The polarity-, target- and timing-specific effects highlight key methodological considerations for optimising tDCS protocols in research on response inhibition impairments. PROSPERO Registration Number:CRD42024565038.
Background:Identifying reliable predictors for dementia remains a critical unmet need. Light exposure plays a crucial role in regulating circadian rhythms, which influence cognitive function. However, the association between light exposure and dementia risk remains unclear. Aims:This study examined the associations of daytime and nighttime light exposure with dementia risk. Methods:A total of 87 577 dementia-free participants (mean age: 62.36 years; 56.98% female) were included. Daytime and nighttime light exposures were measured using 7-day free-living wrist-worn accelerometry. Incident dementia was identified from primary care, hospital inpatient admissions and death registry data. Cox proportional hazards models assessed associations, and mediation analyses evaluated circadian rest-activity rhythms (CRARs), brain structures and vitamin D as potential mediators. Results:Over a median follow-up of 8.1 years, 741 participants developed dementia. Daytime light exposure above 1000 lux was associated with reduced dementia risk (hazard ratio [HR] 0.84, 95% confidence interval [CI] 0.71-0.99, p = 0.039). Longer exposure to brighter light (e.g., ≥ 0.70 h at ≥ 5000 lux; HR 0.83, p = 0.036) was associated with a further reduction in risk. In exploratory analyses, CRARs and brain structures mediated up to 33% of the association. Protective associations were stronger in those with high levels of nighttime light exposure, an evening chronotype or apolipoprotein E (APOE) ε4 carrier status, with a risk reduction of up to 41%. Furthermore, < 0.70 h per day of bright daytime light (≥ 5000 lux) outperformed six established dementia predictors (e.g., obesity, alcohol consumption, traumatic brain injury and so on). Nighttime light showed no significant association with dementia risk. Conclusions:High levels of daytime light exposure were significantly associated with lower dementia risk. Further research should explore its role in dementia screening and inform the development of light-based interventions.
Background:Although experimental evidence supports links between exercise and certain mental health indicators, epidemiological research has focused primarily on the dose-response effect of a single type of exercise, overlooking potential differences across exercise types in their associations with overall and specific mental health outcomes. Aims:To examine the associations between different types of exercise and multiple mental health outcomes, including depression, anxiety, post-traumatic stress disorder, suicidal ideation, suicide attempts and nonsuicidal self-injury, and to explore their dose-response relationships in terms of exercise frequency and duration. Methods:Data were drawn from a large-scale, cross-sectional study of 79 011 university students in Jilin Province, China, in 2021 via an online survey. Mental health outcomes were evaluated using validated scales and diagnoses. Propensity score matching was first applied to balance covariates between exercise types. Logistic regression analysis was then used to examine the associations between exercise types and mental health outcomes. Additionally, generalised additive models were employed to explore the dose-response relationships between exercise frequency/duration and mental health outcomes, adjusting for sociodemographic factors. Results:Among the 79 011 students reporting participation in physical exercise, team ball sports showed the strongest protective associations, followed by single anaerobic activities, two-player racket sports, single moderate-intensity aerobic sports and single low-intensity aerobic sports. U-shaped associations were observed between exercise frequency, duration and mental health outcomes. Exercising three to four times per week for 90-120 min was associated with a lower probability of reporting mental health problems, whereas excessive exercise was associated with poorer outcomes. Conclusion:Participation in different types of exercise is differentially associated with mental health outcomes, with team ball sports showing the most favourable associations. Optimal levels of exercise dosage vary between exercise types, suggesting that individuals may benefit from selecting exercise patterns that best support their mental health.
Background:Depression and obesity are major global health burdens, yet the co-occurrence patterns and temporal trends of their comorbidity remain poorly characterised at the global level. Aims:To map the global patterns of co-occurring depression and obesity, quantify temporal trends of depression, obesity and comorbidity, forecast the prevalence of comorbidity up to 2035 and compare social indicators across groups stratified by their combined temporal trends of depression and obesity. Methods:Data were sourced from the Global Burden of Disease Study 2023, the NCD Risk Factor Collaboration and the World Bank Open Data repository. Countries and territories (n = 199) were classified into four co-occurrence patterns based on the current prevalence of depression and obesity. Average annual percentage changes (AAPCs) were calculated to quantify temporal trends and define four trend groups. We forecast the prevalence of depression-obesity comorbidity across 199 countries and territories up to 2035 using an autoregressive integrated moving average model. The Kruskal-Wallis H test and Mann-Whitney U test were used to compare seven social indicators relevant to socioeconomic status and healthcare resources across these trend groups. Results:For the four co-occurrence patterns, the consistently-low-prevalence group was primarily concentrated in Asia, the consistently-high-prevalence group in the Americas, the depression-dominant group in Africa and the obesity-dominant group in the Americas and Oceania. Concerning temporal trends, the prevalence of depression-obesity comorbidity increased globally (median AAPC: 4.66%; interquartile range [IQR]: 3.49%-6.30%). Notably, countries and territories in East Asia, Southeast Asia and sub-Saharan Africa had higher comorbidity AAPCs despite their low current prevalence. Temporal trends were categorised into four patterns: 44 (22.11%) countries and territories were classified into the double-fast-growing group (Southeast Asia, sub-Saharan Africa), 56 (28.14%) into the depression-driven-growing group (West Asia, North Africa, North America, Oceania), 56 (28.14%) into the obesity-driven-growing group (East Asia, Latin America and the Caribbean) and the remaining 43 (21.61%) into the double-slow-growing group (Europe, Australia and New Zealand). Predictive analysis indicated the median prevalence of depression-obesity comorbidity may increase to 833.32 (IQR: 532.07-1367.42) per 100 000 in 2030 and 869.35 (IQR: 544.11-1403.71) per 100 000 in 2035. Compared with the double-slow-growing group, the obesity-driven and double-fast-growing groups showed poorer socioeconomic status and healthcare resources; the depression-driven-growing group showed poorer healthcare resources. Conclusions:The prevalence of depression-obesity comorbidity increased globally. Particular attention is warranted for countries and territories with currently low comorbidity prevalence but rapidly increasing trends. Our findings also highlight the need for accessible interventions for countries and territories with poorer socioeconomic status and healthcare resources.
Background:Major depressive disorder (MDD) and chronic pain (CP) demonstrate marked sex disparities, with higher prevalence and severity in females. However, the underlying mechanisms remain poorly understood. Aims:We aimed to examine the intricate interplay among anxiety, physical pain and sleep disruptions in inpatients with depression and chronic pain, specifically addressing potential gender differences. Methods:This cross-sectional study enrolled 122 participants and split them into three groups: the comorbidity group (inpatients with MDD and CP, n = 37), the MDD group (inpatients with MDD without CP, n = 37) and the control group (healthy controls, n = 48). Sex-stratified piecewise regression tested pain/latent variable threshold effects in depression-comorbidity subgroups. Generalised additive models (GAMs) with tensor-product smoothing and S-splines were used to model associations between a standardised somatic anxiety subscore and pain intensity, capturing nonlinear patterns and subgroup-specific interactions. Results:The comorbidity group exhibited greater impairments in pain intensity, anxiety and depressive symptoms, and quality of life compared to other groups. Piecewise regression analysis revealed a nonlinear relationship between the somatic anxiety subscore and pain intensity, with female inpatients in the comorbidity group experiencing higher pain intensity and anxiety severity compared to those in the MDD group. The GAM results showed a peak in pain intensity at a moderate-low threshold in the MDD and comorbidity groups, suggesting an optimal clinical breakpoint (8 points on the somatic anxiety subscore of Hamilton Anxiety Rating Scale-14) for targeted interventions in females. Conclusions:Our findings offer novel insights into the sex-specific dynamics linking anxiety, somatic pain and sleep disturbances in MDD-CP comorbidity, providing an empirical foundation for personalised therapy.
Background Current research suggests that genetic risk for psychiatric disorders is largely due to distinct combinations of many common variants shared by different disorders. This points to the existence of latent components affecting different dimensions of psychopathology.Aims The aim of this study is to identify and characterise latent genetic components involved in psychopathology using a data-driven approach and evaluate their potential as principal component-based polygenic scores (PC-PGSs).Methods Singular value decomposition was applied to a matrix of summary statistics from the largest available genome-wide association studies (GWASs) for eight psychiatric disorders to identify latent components. The components were characterised by gene mapping of the top contributing variants, enrichment analysis and genetic correlation with external traits. PC-PGSs were evaluated in the FinnGen dataset by computing group-wise PGSs from summary statistics using Reconstructing Allelic Count.Results The different components were mainly involved in synapse organisation and neurodevelopment. The first latent component (PC1) explained 30.5% of the total variance and represented a broad transdiagnostic dimension. Neuroticism was the most strongly correlated external trait. Substance use traits and other psychiatric disorders were positively correlated, whereas cognitive traits were negatively correlated. The second latent component (14.7% of the variance) contrasted thought disorders with childhood-onset neurodevelopmental disorders. Educational attainment and creativity were the most correlated external traits. Other components were more related to a single disorder or differentiated between two related disorders. PC-PGSs in FinnGen largely showed associations in the expected direction, indicating a consistent overall trend for the different PC-PGSs. PC1-PGS was associated with all disorders. Other PC-PGSs showed the expected association in case-case comparisons.Conclusions Decomposing GWAS summary statistic matrices can reveal functionally coherent dimensions of psychiatric genetic risk that could be clinically relevant, offering a potential framework to refine diagnosis, improve prediction and inform personalised treatment.
Background The adult attention deficit hyperactivity disorder (ADHD) Self-Report Scale (ASRS) is one of the most widely used patient-reported outcome measures (PROMs) for evaluating ADHD symptom severity and treatment-related changes in clinical trials of adults with ADHD; however, clinically important differences (CIDs) for the ASRS remain poorly defined. Aims This study aimed to review how CIDs for the ASRS have been reported in randomised controlled trials (RCTs) of adult ADHD and to establish ASRS-specific CID values for this population. Methods We first conducted a cross-sectional literature search to examine the reporting of CID values in RCTs of adult ADHD interventions. We then conducted a secondary analysis of an RCT involving 56 adults with ADHD who received either active or sham transcranial alternating current stimulation. The ASRS total score served as the PROM. An anchor-based approach using the Clinical Global Impression-Improvement scale was applied. General linear mixed-effects modelling estimated mean differences across patient-defined improvement levels to identify categorical CID thresholds.Results None of the 11 identified studies reported CID values for the ASRS as a PROM in adults with ADHD. The mixed-effects model indicated that a reduction of approximately 13.39 points in ASRS total score represents a clinically meaningful improvement (CIDwithin-group), and a reduction of 19.51 points represents substantial improvement, corresponding to reductions of 29.2% and 42.6% from the baseline, respectively. Additionally, a between-group difference of 6.13 points in ASRS total score represents a minimal clinically meaningful difference (CIDbetween-group), and a 12.25-point difference represents a moderately meaningful between-group difference. Conclusions This study provides adult-ADHD-specific CID estimates that offer practical benchmarks for interpreting treatment response and informing power calculations in future clinical trials.
Background:Dementia poses a growing global public health burden, particularly in low- and middle-income countries where cognitive screening coverage remains limited. Current risk estimation tools often depend on cognitive testing or biomarkers, restricting their applicability in community and primary care settings. Aims:To establish and validate a data-driven analytical framework for developing a cognitive-testing-free dementia risk estimation tool (Cog-Free) using routinely collected health examination data. Methods:For this prospective cohort study, we developed Cog-Free, an internet-based dementia risk estimation tool, using 38 Least Absolute Shrinkage and Selection Operator-selected risk-associated variables and the optimal machine-learning algorithm (logistic regression). The optimal algorithm was internally validated with bootstrap resampling and externally tested in the Chinese Longitudinal Healthy Longevity Survey cohort. The tool was trained and internally validated in 2962 dementia-free adults aged ≥ 65 years (2018-2024), and its performance was compared with three established cognitive-testing-free tools. Results:Cog-Free achieved the highest area under the receiver operating characteristics curve in the internal validation set (0.86 [95% confidence interval (CI) 0.82-0.89]) with an accuracy of 0.81 (95% CI 0.78-0.83), sensitivity 0.78 (95% CI 0.68-0.85) and specificity 0.81 (95% CI 0.78-0.84), significantly outperforming three existing tools (DeLong's test, p < 0.001). Several previously under-recognized risk-associated variables for dementia were identified, such as right-hand grip strength, cognitive activity, having worse memory than peers, nighttime awakenings and income satisfaction. Conclusions:Cog-Free provides a data-driven, cognitive-testing-free and easily accessible approach for early dementia risk screening using routine health data. Its performance and web-based design suggest potential utility as a pre-screening and risk stratification tool within community health systems, including settings with limited access to cognitive testing.
Background:Major depressive disorder (MDD) is characterised by interhemispheric functional imbalance, yet the neuromodulatory mechanisms underlying its correction remain poorly understood. Repetitive transcranial magnetic stimulation (rTMS) has shown promise in restoring hemispheric balance, but evidence regarding its effects on alpha-band functional connectivity (FC) is limited. Aims:This randomised, two-arm controlled trial aims to elucidate the neural mechanisms by which bilateral rTMS ameliorates interhemispheric functional imbalance in patients with MDD by reducing alpha-band functional hyperconnectivity in the left hemisphere. Methods:Fifty-two patients with MDD were randomised to active (n = 31) or sham (n = 21) bilateral rTMS targeting the dorsolateral prefrontal cortex (dlPFC). Active stimulation involved 20 Hz over the left dlPFC and 1 Hz over the right dlPFC. Participants underwent emotional attention network task testing with simultaneous electroencephalography recording before and after 10 sessions of rTMS. Phase slope index (PSI) was used to quantify directional FC in the alpha band (8-13 Hz). Clinical symptoms and behavioural performance were also assessed. Results:Active rTMS significantly improved sleep quality (PSQI, F = 5.52, p = 0.023) and reduced reaction times under both congruent and incongruent emotional conditions (-in-: t = -2.23, p = 0.031, +in-: t = -2.01, p = 0.05). Alpha-band FC analysis revealed decreased left hemispheric hyperconnectivity in the active group, particularly in the occipital, left temporoparietal and left frontocentral regions (t ranging from 2.42 to 3.65, all p = 0.038). The sham group showed placebo-driven improvements in subjective mood but no significant FC changes. Conclusions:Bilateral rTMS modulates interhemispheric imbalance in patients with MDD by reducing pathological alpha-band FC in left hemispheric regions, which is correlated with improved emotional processing and attentional performance. These findings support the use of alpha-band FC as a biomarker for rTMS efficacy and highlight the potential of network-targeted stimulation protocols for the individualised treatment of MDD.
Background Nutraceutical supplementation targeting mitochondrial function has been proposed as a beneficial therapeutic strategy to improve physical and mental health in psychiatric patients.Aims To summarise the results of studies evaluating nutraceutical supplementation targeting mitochondrial function in patients with psychiatric disorders.Methods Following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines, we searched PubMed, Embase and Scopus databases from 1 January 2007 to 30 April 2024. Reports were included if they evaluated outcomes of nutraceutical supplementation in patients with psychiatric disorders or related conditions. Additionally, we performed a risk-of-bias analysis of the studies compatible with the RoB2 tool.Results Of the 2061 records identified, 122 studies met the inclusion criteria, evaluating vitamin D3, N-acetylcysteine, acetyl-L-carnitine, coenzyme Q10, alpha-lipoic acid, magnesium, vitamin B6, vitamin B7, folic acid, vitamin B12, vitamin E, vitamin A, vitamin C and vitamin B3. The most studied nutraceuticals were vitamin D3 (27.05%) and N-acetylcysteine (15.6%). Among randomised controlled clinical trials (RCTs), vitamin D3 was the most extensively investigated and accounted for the highest number of trials reporting improvements in clinical outcomes, although findings were heterogeneous. Notably, 14.8% of the studies evaluated combinations of three or more nutraceuticals. Dietary supplements were extensively evaluated for autism spectrum disorder (28 studies), schizophrenia spectrum disorder (27 studies), major depressive disorder or related depressive symptoms (22 studies), attention-deficit hyperactivity disorder (9 studies) and bipolar spectrum disorder (6 studies). A substantial proportion of studies were not RCTs but open-label single-arm trials or case reports. Significant heterogeneity was observed in the nutraceutical components used, treatment duration and the outcomes assessed. Overall, the risk of bias was high, and the methodological quality was generally low.Conclusions Promising findings in nutraceutical studies for psychiatric disorders face challenges, including small sample sizes, short follow-up periods and a lack of treatment standardisation. Future research requires robust RCTs with standardised protocols and validated biomarkers of efficacy.
Background:Delayed neurocognitive recovery (dNCR) is a prevalent complication in older patients undergoing surgery. It may progress to long-term cognitive impairment and increase the risk of Alzheimer's disease. Aims:This study aimed to evaluate the effects of olfactory enrichment on dNCR and to examine the association between olfactory function and dNCR. Methods:This sham-controlled, assessor-blind, parallel-group randomised trial enrolled 149 participants aged 65 or older undergoing elective total knee or hip replacement under general anaesthesia. Participants were assigned to either the olfactory enrichment group or the sham group. The intervention group received daily olfactory enrichment from 3 days preoperatively to 7 days postoperatively. Cognitive function was evaluated using a neuropsychological test battery 3 days before and 7 days after surgery. Olfactory identification ability was assessed by five-odour olfactory detection arrays. Propensity score matching analysis was employed to mitigate potential confounding and selection bias. Results:A total of 131 patients completed the study (62 in the olfactory enrichment group and 69 in the sham group). The overall incidence of dNCR was 26.7% (35 out of 131). In the intention-to-treat analysis, the difference between groups was not statistically significant (19.4% vs. 33.3%; χ 2 = 3.259; p = 0.071). However, in the 1:1 propensity score-matched cohort (n = 82), the incidence of dNCR was significantly lower in the olfactory enrichment group than in the sham group (12.2% vs. 39.0%; χ 2 = 7.476; p = 0.005). Raw postoperative cognitive scores and individual change scores did not differ between the groups. Participants with decreased olfactory identification scores (n = 32) had a significantly higher incidence of dNCR than those with stable or improved scores (40.6% vs. 22.2%; χ 2 = 4.183; p = 0.041). Conclusions:In older patients undergoing major orthopaedic surgery, perioperative olfactory dysfunction is associated with an increased risk of dNCR. Olfactory enrichment may represent a potential nonpharmacological strategy for reducing postoperative cognitive decline in this population.
