BackgroundAge-related macular degeneration (AMD) is the leading cause of visual impairment in the elderly population. Accumulating evidence has revealed the possible association between metabolites and AMD. This study aimed to assess the effect of plasma metabolites on AMD and its two subtypes using a bidirectional two-sample Mendelian randomization approach.MethodsThe causality between plasma metabolites and AMD was assessed by a bidirectional two-sample Mendelian randomization (MR) analysis using the genome-wide association studies (GWAS) summary statistics of 1400 genetically determined metabolites (GDMs) and AMD. For this MR analysis, inverse variance weighted (IVW) was used as the primary method, with weighted median, MR-Egger, weighted mode, and simple mode as supplementary methods to examine the causality. MR-Egger intercept, Cochran's Q, and MR-PRESSO test were employed to evaluate possible pleiotropy and heterogeneity.ResultsThe results of IVW showed significant causal associations between 13 GDMs and AMD. 1-stearoyl-GPE (18:0), androstenediol (3 beta,17 beta) monosulfate, stearoyl sphingomyelin (d18:1/18:0), xylose, and X-11,850 exhibited a protective effect on AMD, while gulonate and mannonate increased the risk of AMD. 1-stearoyl-GPE (18:0) and X-11,850 exhibited protective effects on dry AMD. DHEAS, 1-stearoyl-GPE (18:0), 5 alpha-androstan-3 beta,17 beta-diol disulfate, xylose, androstenediol (3 beta,17 beta) monosulfate, and N2-acetyl, N6, N6-dimethyllysine exhibited a protective effect on wet AMD, while succinimide, 16a-hydroxy DHEA 3-sulfate, and X-13,553 increased the risk of wet AMD. Horizontal pleiotropy and heterogeneity did not distort the causal estimates. In the reverse MR analysis, AMD reduced the androstenediol (3 beta,17 beta) monosulfate level, and increased the stearoyl sphingomyelin(d18:1/18:0) level.ConclusionThis study supported the effect of plasma metabolites on AMD, providing novel insights for clinical diagnosis and prevention strategy.