Network meta-analysis of the efficacy of pharmacological treatments for post-stroke cognitive impairment and vascular cognitive impairment

Background Based on recent reviews, vascular cognitive impairment (VCI) encompasses a spectrum of cognitive deficits caused by cerebrovascular disease and its risk factors, ranging from mild cognitive impairment to dementia, and often coexists with neurodegenerative conditions like Alzheimer’s disease. VCI is categorized into four clinical-imaging subtypes, including post-stroke cognitive impairment (PSCI)—a common stroke complication and major VCI subtype. Current guidelines recommend cholinesterase inhibitors and NMDA receptor antagonists as first-line treatments for VCI, with expert consensus supporting donepezil and rivastigmine for PSCI. However, existing evidence primarily derives from placebo-controlled or head-to-head drug comparisons, lacking comprehensive evaluations of multiple cognitive enhancers. This study aims to systematically assess the efficacy and safety of cognitive-enhancing drugs in VCI, with a focused analysis on PSCI, to better inform clinical decision-making and improve patient outcomes. Methods We systematically searched four databases using predefined search strategies. Eligible studies were selected based on predetermined criteria. The included studies were analyzed with StataSE 16.0, RevMan 5.3, and Grade software to compare the efficacy and safety of cognitive-enhancing drugs to identify the optimal treatment for VCI and PSCI. Results Sixteen studies (5,599 participants) were included. In terms of cognitive outcomes, sailuotong was superior to placebo on the Alzheimer’s Disease Assessment Scale-Cognitive Subscale (ADAS-cog) (MD = −3.00, 95% CI: −4.50, −1.50) and ranked best (SUCRA 88.5%). Memantine was most effective on the Mini-Mental State Examination (MMSE) (MD = 1.23, 95% CI: 0.23–2.23; SUCRA 80.8%). For the secondary outcome, the MoCA assessment showed that Ginkgo biloba extract significantly improved Montreal Cognitive Assessment (MoCA) scores compared to placebo (MD = 1.29, 95% CI: 1.24, 1.35). Regarding safety, donepezil significantly increased the risk of overall adverse events compared to placebo (OR: 1.57; 95% CI: 1.19–2.06). Conclusion Our network meta-analysis suggests that memantine might have the best effect for PSCI, with sailuotong potentially serving as a secondary option. However, these estimates are based on a small randomized controlled trial and a sparse network. Therefore, the current evidence is limited, highlighting the need for more high-quality studies to robustly validate the therapeutic potential of these interventions for VCI and PSCI. Systematic review registration https://www.crd.york.ac.uk/PROSPERO/ , identifier in PROSPERO (CRD420250627957).