The Preoperative EUS Staging And Clinical Pathology In Patients With Gastric Cancer

Gastric cancer is one of the most common carcinoma in the world. As its untypical symptoms in early time, many patients were at the advanced stage when gastric cancer is finally diagnosed, thus resulted in poor prognosis. Surgery is the main therapeutic method for the patients with gastric cancer, and the surgery method is different in terms of different stage of carcinoma. Therefore, accurate preoperative staging is important for the determination of appropriate treatment. It is more important when some new technologies are introduced such as endoscopy, laparotomy and adjunctive treatment.The traditional diagnostic and staging methods include barium meal, endoscpy, computed tomography(CT),ultrasonography, and biopsy. Endoscopic ultrasonography(EUS) is a new technology which combines the strongpoint of both endoscopy and ultrasonography, it can observe the morphological structures of lesions closely through endoscope, as well as the layer structures of gastric wall and invasion of perigastric organs through ultrasound, thus achieved rapid development among this years, and has been confirmed as routine method of diagnosis and preoperative staging of gastric cancer.EUS has been widely used in the preoperative staging of gastric cancer. However, the reported accuracy of EUS for staging varies greatly among different studie, and these studie are mainly retrospective diagnostic experiments.This article aims to compare patients’ clinical data, laboratory and imaging examination (including EUS and CT) with postoperative pathologic result, hoping to evaluate the value of EUS in preoperative staging, and analyse the factors which influence the accuracy for EUS staging and gastric cancer pathology.This study includes two parts, which will be discussed in detail as below:Part I The accuracy of EUS for preoperative TN staging of gastric cancer and possible influencing factorsObjective To evaluate the accuracy of EUS for preoperative staging of gastric cancer through comparison of EUS results and pathologic results, and to find factors influencing EUS accuracy by analyzing patients’ common data, endoscopy and EUS results, as well as pathologic findings.Methods Prospectively analyze patients who were diagnosed as gastric tumor by EUS and confirmed as gastric cancer by operation between January2006and July2012. Postoperative TNM staging was made according to the sixth edition established by American Joint Committee on Cancer (AJCC). This is used as golden standard to evaluate the accuracy of EUS for preoperative staging. And use case-control study to find some meaningful factors influencing this accuracy.Results1. A total of303cases were followed, and144cases were suitable for this study, which includes99male and45female, aging from19to84, with an average age of54.3.2. The overall accuracy of EUS for preoperative T staging is57.6%,61cases (42.4%)do not stage correctly, which includes16cases(11.1%) understaged and45cases overstaged(31.3%), mainly T2overstaged as T3by EUS.3. The overall accuracy of EUS for preoperative N staging is53.5%, EUS can determine N-and N+with accuracy of41.6%and81.4%. The overall accuracy of CT for preoperative N staging is51.4%, the combination of EUS and CT can increase the accuracy to59.7%, but there is no statistic difference compared with EUS alone(p>0.05).4. The accuracy for T and N staging was50%and44.1%from January2006to December2009, which increased significantly to64.5%and61.8%between January2010and July2012(for T staging:p<0.1; for N staging:p<0.05).6. The accuracy of EUS for T staging between early gastric cancer(EGC) and advanced gastric cancer(AGC) is of no difference, while N staging is significantly lower in AGC(46.9%vs77.4%, p=0.003), as well as the presence of local lymph node metastasis(84.0%vs37.2%, p<0.001).7. The univariate and multivariate analyses reveal that different location of tumor affects the accuracy of EUS for T staging. There is a decreasing tendency of accuracy from cardia to pylorus(cardia:77.8%, pylorus:47.4%, p=0.032). The accuracy is of no significant difference in different gender, age, lesion type, gastric wall thickness, tumor size, and pathologic type.8. The univariate analysis reveals that different gende affects the accuracy of EUS for N staging, the accuracy is low in female patients(40.0%vs60.0%, p=0.029). Multivariate analysis do not find meaningful factors affecting N staging accuracy.Conclusion1. The accuracy of EUS for T and N staging is lower than expect, there is an increasing tendency along time.2. The accuracy of CT for N staging is a bit less than EUS, the combination of CT and EUS does not increasing the accuracy significantly.3. Tumor location affects T staging accuracy, which is lowest at pylorus.4. Tumor invasion depth and local lymph node metastasis affect N staging accuracy, deep tumor invasion and present of lymph node metastasis result in lower accuracy for N staging. Part II Influencing factors of different gastric cancer pathologic stagingObjective Find out factors influencing different pathologic staging of gastric cancer by analyzing patient’s clinical data, laboratory and imaging examination, and postoperative pathologic result, analyse relationship between this factors as well.Methods Using the data of144cases in part one, cases can be divided to EGC and AGC according to T staging, and to positive/negative lymph node metastasis according to N staging. Univariate and multivariate analyses are used to find factors influencing different pathologic staging of gastric cancer, spearman correlation analysis is used to assess the relationship between factors.Results1.144cases of gastric cancer include31cases of EGC and113cases of AGC,50cases have none lymph node metastasis and94cases have lymph node metastasis.2. There is a marked correlation between T and N staging, higher T staging may lead to high staging of corresponding N staging.3. Univariate analysis shows that gender, tumor lacation, gastric wall thickness, tumor size, and pathologic type are related with pathologic T staging. In female patients, tumor at cardia and stoma, wall thickness greater than20mm, tumor size greater than3cm, mediate and low differentiate carcinoma, AGC are more common.4. Multivariate analysis shows that tumor size and pathologic type are two independent factors affecting T staging. Patients with tumor size larger than3cm(3-4cm: OR12.884;>4cm:OR49.571), mediate and low differentiate carcinoma (mediate-differentiate carcinoma:OR11.289; low-differentiated carcinoma:OR22.655)are more likely to be diagnosed as AGC.5. Univariate analysis shows that gender, tumor size, and pathologic type are related with pathologic N staging. In female patients, tumor size greater than4cm, mediate and low differentiate carcinoma, lymph node metastasis are more common.6. Multivariate analysis shows that tumor size and pathologic type are two independent factors affecting N staging. Patients with tumor size larger than4cm(OR5.063), mediate and low differentiate carcinoma (mediate-differentiate carcinoma:OR5.096; low-differentiated carcinoma:OR6.179)are more likely to occur local lymph node metastasis.7. Correlation analysis reveals that relationship exists between gender and lesion type, age and pathologic type, lesion type and pathologic type, as well as gastric wall thickness and tumor size.Conclusion1. There is a correlation between pathologic T and N staging of gastric cancer, patients with deeper tumor invasion may have more amounts of lymph node metastasis.2. Patients with large tumor size, mediate and low differentiated carcinoma are more likely to be diagnosed as AGC, and easier to occur local lymph node metastasis.3. Relationship exists between factors influencing pathologic T and N staging of gastric cancer, which should be taken into consideration when explaining the results.