| Background:With the development of the endoscopic diagnosis and treatment technology, the detection rate of colorectal intraepithelial neoplasia(IEN) lesions increased significantly. Timely treatment can effectively prevent disease progressing to colorectal cancer. Currently colonoscopy and endoscopic forceps biopsy(EFB) are considered as gold standard for diagnosis of IEN of large intestine. However, in recent years, some researchers showed discrepancies between EFB and ER specimens in IEN. Our country still lacked research about discrepancy in colorectal IEN and there is no uniform conclusion about it. Object:To investigate consistency rate and risk factors of histologic diagnoses between EFB and ER specimens by comparing pathology histologic diagnoses and provide a theoretical basis for improving the preoperative diagnostic accuracy of colorectal IEN lesions. Methods:Clinicopathological data of patients who had a treatment of colorectal IEN lesions by ER from January 1, 2013 to December 31, 2015 in the endoscopy center of the Bethune First Hospital of Jilin University were retrospective analyzed. A total of 336 patients and 350 lesions were in accordance with the inclusive criteria. To investigate consistency rate of histologic diagnoses between EFB and ER. The subjects were divided into two groups according to the pathological diagnosis and to investigate risk factors of the discrepancy. Results:(1)A total of 336 patients with colorectal IEN lesions were in accordance with the inclusive criteria, 227 were male, 109 were female. The average age was 58.3±10.5. A total of 350 lesions and 91 cases were diagnosed in other hospitals, 259 cases were diagnosed in our hospital.(2)The concordance rate of pathological diagnoses between EFB and ER specimens was 60.8%(213/350). The concordance rate of Low-grade intraepithelial neoplasia(LGIN) and High-grade intraepithelial neoplasia(HGIN) were 62.5%(168/269) and 55.6%(45/81) respectively and there was no significant difference among two groups(P=0.299). There were 32.9%(115/350) of cases whose pathology diagnoses after ER were more severe than biopsy, including 76 cases(28.3%) of HGIN and 17 cases(6.3%) of early cancer from LGIN and 22 cases(27.2%) of early cancer from HGIN. There were 6.3%(22/350) of patients whose final pathology diagnoses were downgraded than EFB, including 8 cases(3.0%) of non-neoplasms from LGIN and 14 cases(17.3%) of LGIN from HGIN.(3)The analysis of univariate and logistic regression revealed that tumor size≥30 mm(P=0.013, OR=2.171), depressed(P=0.030, OR=4.183) and mixed type of surface appearance(P=0.029, OR=2.709), surface nodularity(P=0.002, OR=2.185) and biopey in other hospital(P<0.001, OR=2.834) were independent risk factors which were significantly associated with the discrepancy between EFB and ER.(4)The analysis of univariate and logistic regression revealed that tumor size≥30 mm(P=0.012, OR=2.688), depressed(P=0.020, OR=8.647) and mixed type of surface appearance(P=0.008,OR=4.132), nodularity surface(P=0.005,OR=2.382) were independent risk factors which were significantly associated with the discrepancy in our hospital. Conclusion:(1)The concordance rate of pathological diagnoses between EFB and ER specimens was 60.8%, so we could not depend entirely on EFB pathology to select clinical treatment.(2)The tumor size≥30 mm, depressed or mixed type of surface appearance and surface nodularity were independent risk factors which influence the accuracy of the endoscopic biopsy. If we find such lesions, we should observe carefully and choose the best way and position of endoscopic biopsy in order to improve the diagnostic rate of preoperative endoscopic biopsy. |