Many observational studies imply elevated blood pressure (BP) as a leading risk factor for incident myocardial infarction (MI), but whether this relationship is causal remains unknown. In this study, we used bidirectional Mendelian randomization (MR) to investigate the potential causal association of BP levels with the risk of MI. Genetic variants associated with BP and MI traits were retrieved from the International Consortium of Blood Pressure (N = 7,57,601) and UKB (N = 3,61,194), obtaining 1,26,40,541 variants. We used two-sample MR (TSMR) analyses to examine the potential bidirectional causal association of systolic BP (SBP), diastolic BP (DBP) and pulse pressure (PP) with MI. The forward MR analysis identified a potentially causal association between MI and BP except PP[odds ratio (OR) SBP: 1.0008, P = 1.911 × 10-22; ORDBP: 1.0014, P = 1.788 × 10-28;odds ratio (OR)pp: 1.0092, P = 0.179]. However, the reverse analysis suggested no causal relation (betaSBP: 5.469, P = 0.763; betaDBP: 3.624, P = 0.588; betaPP: -0.074, P = 0.912). These findings were robust in sensitivity analyses such as the MR-Egger method, the maximum likelihood method and the MR pleiotropy residual sum and outlier test (MR-PRESSO). No horizontal pleiotropy (p = 0.869 for SBP, p = 0.109 for DBP and p = 0.978 for PP in the forward results and p = 0.168 for SBP, P = 0.892 for DBP and p = 0.989 for PP in the reverse results) was observed. Elevated SBP or DBP levels increase the risk of MI, but there is no causal relationship between MI and changes in BP including PP. Independent of other risk factors, optimal BP control might represent an important therapeutic target for MI prevention in the general population.