Depression, Not Anxiety, Tied to Poor Cognition Post-Stroke

TOPLINEDepression, but not anxiety, was independently linked to poorer executive function over 5 years after ischaemic stroke. The association between depression and visuospatial function was less robust. Anxiety was present in 27%-36% of patients at each assessment. METHODOLOGYResearchers analysed data from the single-centre, observational STROKDEM cohort, which included 191 patients with ischaemic stroke (119 men and 72 women) assessed at 6, 12, 36, and 60 months after stroke.Data on anxiety were available for 157 patients at 6 months and 85 patients at 60 months; patients with poorer baseline cognition were more likely to be lost to follow-up.Patients completed neuropsychological testing across five domains (memory, executive function, attention, visuospatial abilities, and language), with results expressed as domain-specific z-scores.Anxiety was measured using the Hamilton Anxiety Rating Scale (scores ≥ 8 indicated clinically relevant anxiety), depression was assessed using the Center for Epidemiologic Studies Depression Scale (scores ≥ 16 indicated clinically relevant depression), and stroke severity was evaluated using the National Institutes of Health Stroke Scale.Researchers fitted two models, each using one affective score as the outcome and the five cognitive domains as predictors, with adjustment for the other affective score, time, sex, age, education, and stroke severity. Sensitivity analyses used multiple imputation and pattern-mixture models to address attrition.TAKEAWAYAfter mutual adjustment and false discovery rate correction, depression was independently associated with executive function (standardised regression coefficient [b], -0.167; P = .021) and visuospatial abilities (b, -0.208; P = .043). No cognitive domain was independently associated with anxiety after adjustment for depression. The association between depression and executive function held across all sensitivity analyses, including multiple imputation (b = -0.178; P = .015) and pattern-mixture analysis (P = .008 to .025). The association between depression and visuospatial abilities was not significant after multiple imputation (P = .195). Anxiety prevalence ranged from 26.6% to 36.1% across assessments (28.7% at 6 months and 31.8% at 60 months), whereas depression prevalence declined from 27.1% to 17.8%. IN PRACTICE"Distinguishing between [depression and anxiety] requires joint modelling; failing to do so risks misattributing depression-related cognitive impairment to anxiety. For clinical practice, screening for depression is the more informative route for identifying stroke survivors at cognitive risk," the authors wrote. "This does not diminish the importance of anxiety, which is prevalent, persistent, and detrimental to quality of life and rehabilitation participation in its own right," they added. SOURCEThe study was led by Florine Ruthmann, Université de Lille, Inserm, Lille, France. It was published online on October 12 in the Journal of Affective Disorders.LIMITATIONSAttrition was substantial, and patients with poorer baseline cognition were more likely to be lost to follow-up; the COVID pandemic further contributed to the loss to follow-up at 60 months. Few patients had persistent anxiety. Psychiatric history was not available, psychotropic medication use was not adjusted for, and education was the only measure of cognitive reserve. The study was observational and conducted at a single centre, and the cohort included mostly men; effect sizes were modest.DISCLOSURESHauts-de-France Regional Council funded the study. The authors reported having no conflicts of interest.This article was created using several editorial tools, including AI, as part of the process. Human editors reviewed this content before publication.