Objective To evaluate the feasibility of assessing osteoarthritis (OA) in hip dysplasia using 3D delayed gadolinium-enhanced MRI of cartilage (dGEMRIC).Methods Thirty-five hips in 20 patients with radiographic evidence of hip dysplasia underwent 3D-dGEMRIC scanning.Clinical symptoms were assessed with the Western Ontario and McMaster Universities Osteoarthritis ( WOMAC ) questionnaire.Radiographic measurement of lateral center-edge angle and T(o)nnis grading were performed on the X-rays.Hips of T(o)nnis grade 1were included in the group of hips with early OA,while the hips with no evidence of OA and without pain symptom were included in the group of hips with normal morphology.The 3D-dGEMRC scans were completed on a 1.5 T MR scanner.The data of 3D-dGEMRIC was reconstructed radically.The dGEMRIC indices were measured on six sites of periphery zones of hip cartilage on reconstructed images.The dGEMRIC indices among different groups were analyzed by non-parametric tests.The differences of dGEMRIC indices among six sites in the group of early OA or the group of normal morphology were analyzed by Wilcoxon test.Results The mean dGEMRIC indices of six sites were lower in group of T(o)nnis grade 1than in group of T(o)nnis grade 0 ( Z =- 2.149,P =0.032 ),and lower in group of T(o)nnis grade 2 than in group of T(o)nnis grade 1( Z =- 1.990,P =0.047 ).The dGEMRIC indices of the anterior site,anterosuperior site,superior-anterior site,and superior site were significantly different between the group of hips with early OA and the group of hips with normal morphology (Z =-2.333--2.041,all of the P values were lower than 0.05).In the group of hips with normal morphology,the dGEMRIC indices of superior-anterior site of hip were lower than superior site(P =0.028).In the group of hips with early OA,the dGEMRIC indices of superior-anterior site were lower than the other sites except for anterior-superior site ( Z =- 3.041- - 2.277,all of the P values were lower than 0.05 ).Conclusions 3 D-dGEMRIC might be a sensitive technique for detection of glycosaminoglycans alteration in early OA and staging of OA in hip dysplasia.Radial reconstruction could provide an accurate assessment of OA,and the results demonstrated that early cartilage alteration could be detected in the anterior to superior sites of hips,and the earliest cartilage alteration may occur in the superior-anterior site of hips.

Objective: To evaluate the prognostic impact of tumor deposits on the overall survival (OS) of gastric cancer (GC) patients. Methods: Between January 2007 and December 2012, 312 GC patients undergoing curative resection in The First Affiliated Hospital of Hainan Medical University were enrolled. Patients were categorized into two groups based on the tumor deposit status of postopera-tive pathology: positive group, presence of tumor deposits and negative group, absence of tumor deposits. The correlations of tumor deposit status with clinicopathological and potential prognostic factors were analyzed. Results: Eighty-four (26.9%) patients had tumor deposits. There were significant differences in Borrmann type, tumor size, depth of invasion, N stage, tumor-node-metastasis (TNM) stage, and lymphovascular invasion between the two groups on univariate analysis. Multivariate analysis revealed that Borrmann type, N stage, and lymphovascular invasion were independently associated with the presence of tumor deposits. In univariate survival analy-sis, age, tumor location, Borrmann type, tumor size, TNM stage, type of gastrectomy, lymphovascular invasion, and presence of tumor deposits were found to be significant prognostic factors. GC patients with tumor deposits had a significantly lower 5-year OS rate than those without tumor deposits (5-year OS: 34.5% vs . 67.5% , P<0.001). Multivariate analysis revealed that age, Borrmann type III/IV, TNM stage, lymphovascular invasion, and presence of tumor deposits were independent prognostic factors for this cohort. Further stratified analysis demonstrated that the significant prognostic differences between the two groups were only observed in patients with stage N0-3a disease. There were no significant differences in survival between patients with and without tumor deposits at the N3b stage. The prognosis of GC patients with tumor deposits was independently correlated with N stage, lymphovascular invasion, and postoperative chemotherapy. Conclusions: The presence of tumor deposits was an independent prognostic factor in GC patients and can be used as a prognostic indicator for GC patients with stage N0-3a disease. GC patients with tumor deposits should receive postop-erative chemotherapy regardless of TNM stage.