Cognitive function in late-life depression after modified electroconvulsive therapy: A stratified analysis

BACKGROUND: Modified electroconvulsive therapy (MECT) yields heterogeneous cognitive outcomes in late-life depression (LLD). Recent advances highlight that MECT-induced cognitive outcomes are linked to metabolic and mitochondrial health, suggesting baseline cognitive impairment may reflect reversible metabolic suppression rather than structural damage. Our study aims to find baseline clinical and cognitive features that predict cognitive change after MECT, with three novel contributions: (1) first demonstration that specific baseline deficits predict improvement rather than decline; (2) identification of early post-treatment markers; and (3) integration with modern neuroplasticity theory, providing evidence for tailored LLD monitoring and intervention. METHODS: In a 65-patient LLD cohort, cognitive status one week after acute MECT was assessed using the Montreal Cognitive Assessment (MoCA). A ≥2-point increase defined improvement; <2 points indicated stability/decline. Baseline data included demographics, HAMD-17, HAMA, MMSE, and MoCA scores. Binomial logistic regression analyzed predictors of improvement. RESULTS: Both groups showed similar antidepressant efficacy. Baseline, the improved group had higher cognitive disturbance scores, lower MMSE, and MoCA scores in several domains. Post-treatment, they scored higher on total MoCA and orientation. Logistic regression identified naming (OR 0.08) and delayed recall (OR 0.22) as negative predictors, and post-treatment orientation (OR 7.70) as a positive predictor. LIMITATIONS: Limited sample size. CONCLUSION: MECT effectively relieves depressive and anxiety symptoms in LLD. Patients with greater baseline cognitive deficits-especially impaired naming and delayed recall-are most likely to gain cognitively, whereas those with preserved cognition require closer surveillance. Serial MoCA tracking, focusing on orientation, facilitates individualized risk-benefit discussions and treatment planning for MECT-treated LLD. TRIAL REGISTRATION: ClinicalTrials.gov identifier: ChiCTR2300078063.