Font Size: a A A

Discussion Of Application And Influential Factors Of Risk Of Malignancy Index-4 In Ovarian Epithelial Tumor

Posted on:2016-12-15Degree:MasterType:Thesis
Country:ChinaCandidate:P HuFull Text:PDF
GTID:2284330461970649Subject:Gynecologic oncology
Abstract/Summary:PDF Full Text Request
Objective:To evaluate the diagnostic value of risk of malignancy index-4 (RMI-4), clinical symptom, pelvic examination, ultrasound examination, serum CA125 in ovarian epithelial tumor and influential factors of RMI-4’s diagnostic performance.Methods:This is a retrospective study of 150 women with ovarian epithelial tumor were admitted between January 2004 and January 2014 for surgical treatment in tumor hospital affiliated to Guangxi medical university.Among them,50 cases were benign tumor,50 cases were borderline tumor and 50 cases were malignant tumor.Collecting information about patients including clinical symptom, symptom index, pelvic examination, ultrasound examination, ultrasound findings, tumor size, menopausal status, serum CA125 value.According to RMI-4’s formula,we can know RMI-4’s value of all patients. After we bring borderline group into malignant group,by calculating the area under the Receiver Operating Characteristic curves (ROC curves), we can know RMI-4, clinical symptom, pelvic examination, ultrasound examination and serum CA125, which is more effective in judging ovarian epithelial tumor between benign and malignant.After RMI-4 separately distinguish benign group from borderline group (group A), benign group from malignant group (group B), we can assess the differences of sensitivity, specificity, positive predict value, negative predict value and area under ROC curves.Statistical analysis is performed by software SPSS 20.0 and MedCalc 13.0. Counting material is analyzed by t test and difference in masculine rate is analyzed by x2 test.ROC curve is used to evaluate the diagnostic accuracy of all criteria, the difference is analyzed by Z test.With P= 0.05 for inspection standards, there are significant statistical differences (P<0.05).Results:l.Age and menopausal status:The number and percentage of postmenopausal patients of benign group, borderline group, malignant group was separately 10(20.0%),14(28.0%),28(56.0%).The statistical difference between borderline group and malignant group about the number and percentage of postmenopausal patients is significant (P<0.05).2.Histopathologic classification and clinical stage:Serous and mucinous tumor in benign and borderline group were in the majority, unclassified adenocarcinoma and serous carcinoma in malignant group were in the majority.The statistical difference between borderline group and malignant group about the number and percentage of FIGO stage III-IV is significant(P<0.05).3.Ultrasound performance:Unilocular or multilocular cystic lesions were more common in benign group,borderline group show multilocular cystic lesions or cystic solid echo,malignant group were more presented with cystic solid echo or solid echo, and were often accompanied with ascites and intra-abdominal metastasis.92% patients of malignant group had at least 2 points about ultrasound findings,they were evident higher than borderline group(54.0%),thus the statistical difference is significant(P<0.05).But the statistical difference between borderline group and malignant group about the number and percentage of tumor size≥7cm is not significant (P>0.05).4.Serum CA125:The statistical differences among benign group, borderline group and malignant group about serum CA125 value and positive rate of serum CA125 are all significant(P<0.05).The statistical differences between borderline group and malignant group about the number and percentage of advanced stage,ascites,lymph node positive are all significant(P<0.05).5.RMI-4’s sensitivity is evident higher than clinical symptom and pelvic examination, the statistical difference is significant (P<0.05).RMI-4’s specificity is evident higher than clinical symptom, pelvic examination, ultrasound examination and serum CA125, the statistical difference is significant (P<0.05).RMI-4’s positive predict value is evident higher than clinical symptom and pelvic examination, the statistical difference is significant.We can see that area under the ROC curve of RMI-4, clinical symptom, pelvic examination, ultrasound examination and serum CA125 is 0.915,0.670,0.740,0.840,0.839.By using Z test,we can find that RMI-4 has the largest area under ROC curve,is evident larger than clinical symptom (Z= 4.884, P< 0.001),pelvic examination (Z= 3.957, P< 0.001),ultrasound examination(Z=2.239, P=0.025)and serum CA125(Z=3.166, P=0.001),the statistical differences are all significant(P<0.05).6.After RMI-4 separately distinguishes benign group from borderline group (group A), benign group from malignant group (group B), we compare differences about diagnostic performances. The sensitivity, negative predict value and area under ROC curves of group B is evident higher than group A,the statistical difference is significant(P<0.001).Conclusion:1.Malignant group is evident higher than borderline group about the number and percentage of postmenopausal patients,FIGO stage III-IV,ultrasound findings≥2 points, advanced stage,ascites, lymph node positive,serum CA125 value and positive rate of serum CA125,these statistical differences are all significant(P<0.05).2.RMI-4 places high value on judging ovarian epithelial tumor between benign and malignant preoperatively, which is superior to ovarian cancer screening parameters in diagnosing performance.3.Borderline ovarian tumor is an independent factor on affecting RMI-4’s diagnostic performance, which affects serum CA125,ultrasound findings,menopausal status.Tumor size is not a significant factor on affecting RMI-4’s diagnostic performance.
Keywords/Search Tags:epithelial ovarian tumor, borderline ovarian tumor, the risk of malignancy index, diagnostic performance
PDF Full Text Request
Related items