ObjectiveClinical trials investigating the association between acylcarnitines and cognitive function have yielded conflicting results. We conducted a Mendelian randomization (MR) study to examine the causal associations between 20 acylcarnitines and neurocognitive outcomes.MethodsSingle-nucleotide polymorphisms of 20 acylcarnitines were extracted from primary European ancestry-based genome-wide association studies. Two-sample MR (TSMR) was performed to initially screen for acylcarnitines potentially associated with cognitive performance. Further TSMR analysis was conducted to investigate the associations between specific acylcarnitines and other cognitive outcomes, dementia, and brain structure.ResultsAmong the 20 acylcarnitines, we observed that lower genetically predicted levels of butyrylcarnitine (beta = -0.06, 95% CI: [-0.11 to -0.02], p = 0.003) and acetyl-L-carnitine (beta = -0.02, 95% CI: [-0.04 to 0], p = 0.04) were associated with adverse neurocognitive effects. Furthermore, we found a negative correlation between low genetically predicted levels of butyrylcarnitine and cognitive function (beta = -0.17, 95% CI: [-0.3 to -0.05], p = 0.01) as well as intelligence (beta = -0.05, 95% CI: [-0.09 to -0.02], p = 0.003). However, there was no evidence supporting any association between these two acylcarnitines and dementia or brain structure.ConclusionsOur results suggest a causal association between acetyl-L-carnitine and butyrylcarnitine levels and adverse neurocognitive effects. Multicenter, multiregional, and large sample studies are needed to further validate these findings.