Objective To evaluate the clinical diagnostic value of knee MRI at 90° flexed position for Ramp lesions of medial meniscus. Methods A total of 228 patients with knee pain as the main complaint who were admitted between September 2021 and September 2023 was selected as the research subjects, of which 51 patients met the selection criteria and were enrolled in the study. There were 31 males and 20 females with an average age of 38.6 years (range, 15-67 years). Body mass index was 17.2-28.7 kg/m2 (mean, 23.9 kg/m2). There were 25 cases of left knee and 36 cases of right knee. The time from injury to admission was 0.1-14.3 weeks (mean, 2.1 weeks). Preoperative knee MRI at fully extended position (knee extension position) and 90° flexed position (knee flexion position) were performed to determine the presence of irregular signs at the posterior edge of the medial meniscus, and PHMM fluid high signal [i.e. complete fluid filling between the posterior horn of the medial meniscus (PHMM) and the capsule margin]. Findings obtained under arthroscopy served as the “gold standard” to analyze the sensitivity, specificity, accuracy, positive predictive value (PPV), and negative predictive value (NPV) of MRI at knee extension and flexion positions for the two specific signs of Ramp lesion.Results Twenty-one patients (41.2%) were diagnosed with Ramp lesions by using arthroscopy, including 1 case of Thaunat type Ⅰ, 2 cases of type Ⅱ, 6 cases of type Ⅲ, 7 cases of type Ⅳ, and 5 cases of type Ⅴ. The positive rates of irregular signs at the posterior edge of the medial meniscus on MRI at knee extension and flexion positions were significantly different from the diagnosis of Ramp injury under arthroscopy (P<0.05). The sensitivity, specificity, accuracy, PPV, and NPV of MRI in the diagnosis of irregular signs were 76.1%, 60.0%, 66.7%, 57.1%, and 78.3% respectively at knee extension position, and 85.7%, 73.3%, 78.4%, 69.2%, and 88.0% respectively at knee flexion position. The positive rates of PHMM fluid high signal on MRI at knee extension and flexion positions were significantly different from the diagnosis of Ramp injury under arthroscopy (P<0.05). The sensitivity, specificity, accuracy, PPV, and NPV of MRI in diagnosing PHMM fluid high signal were 38.1%, 100%, 74.5%, 100%, and 69.8% respectively at knee extension position, and 85.7%, 100%, 94.1%, 100%, and 90.9% respectively at knee flexion position. ConclusionKnee MRI at 90° flexed position improves the diagnostic performance of the detection of medial meniscal Ramp lesions compared with MRI at fully extended position.