Background: Sarcopenia (SP) is an aging-related loss of muscle mass and function, affecting the respiratory system. However, the causality of the association between sarcopenia on lung diseases remains elusive. Methods: The bidirectional univariate Mendelian randomization (UVMR), multivariate MR (MVMR) analysis, and mediation MR were utilized to systematically investigate the genetic causal relationship of SP and 11 respiratory diseases. Independent genomic variants related to sarcopenia or respiratory diseases were identified as instrumental variables (IVs), and the summary level data of genome-wide associated studies (GWAS) were obtained from the UK biobank and FinnGen. MVMR analysis was conducted to explore the mediation effects of body mass index (BMI), Alcohol Use Disorders Identification Test (AUDIT), smoking, education attainment (EA), physical activity, and Type 2 Diabetes Mellitus (T2DM). Results: Forward UVMR analysis based on the primary method revealed that pneumoconiosis was associated with a higher risk of appendicular lean mass (ALM) (OR = 1.01, p = 0.03), and BMI (10.65%), smoking (10.65%), and physical activity (17.70%) had a mediating role in the effect of pneumoconiosis on ALM. In reverse MR analysis, we found that genetically predicted ALM was significantly associated with an increased risk of pulmonary embolism (PE) (OR = 1.24, p = 7.21E-05). Chronic obstructive pulmonary disease (COPD) (OR = 0.98, p = 0.002) and sarcoidosis (OR = 1.01, p = 0.004) were identified to increase the loss of left-hand grip strength (HGS). Conversely, the increase in left- HGS presented a protective effect on chronic bronchitis (CB) (OR = 0.35, p = 0.03), (OR = 0.80, p = 0.02), and asthma (OR = 0.78, p = 0.04). Similarly, the loss of the right-HGS elevated the risk of low respiratory tract infection (LRTI) (OR = 0.97, p = 0.02) and bronchiectasis (OR = 1.01, p = 0.03), which is also an independent protective factor for LRTI and asthma. In the aspects of low HGS, the risk of LRTI was increased after MVMR analysis, and the risk of sarcoidosis and pneumoconiosis was elevated in the reverse analysis. Lastly, asthma was found to be related to the loss of the usual walking pace, and the reverse MR analysis suggested a causal relationship between the usual walking pace and LRTI (OR = 0.32, p = 2.79 × 10−5), asthma (OR = 0.24, p = 2.09 × 10−6), COPD (OR = 0.22, p = 6.64 × 10−4), and PE(OR = 0.35, p = 0.03). Conclusions: This data-driven MR analysis revealed SP was bidirectional causally associated with lung diseases, providing genetic evidence for further mechanistic and clinical studies to understand the crosstalk between SP and lung diseases. [ABSTRACT FROM AUTHOR]