Abstract: Objective： To analyze the efficiency of ureteroscope and biopsy in the diagnosis of tumor grade, muscle-invasiveness and multifocality in suspected upper tract urinary carcinoma (UTUC) patients in order to find out whether it can be used in the risk stratification of UTUC patients. Methods: A retrospective study of 76 UTUC patients who underwent preoperative ureteroscope and/or biopsy and received radical nephroureterectomy in Peking University Third Hospital during January 2014 to December 2016 was undertaken. Results: In this study, 76 patients were included. There were 31 males (40.8%), and 45 females (59.2%). The median age was 64.5 years (31-88), and 51 patients had the symptom of hematuresis. The tumor was located in renal pelvis in 39 patients, and in ureter in 37 patients. Post-operative pathology confirmed that all the 76 patients included in this study suffered from UTUC, of whom 21 (21.6%) were of low-grade, 51 (67.1%) were of high-grade, 4 (5.3%) were undetermined, and 47 (61.9%) patients were muscle-invasive, and 27 (35.5%) were not, and 2 (2.6%) were undetermined. Among the 50 patients, in whom the grade of the tumor could be diagnosed by biopsy, the sensitivity, specificity, accuracy, positive predictive value and negative predictive value for low-grade tumor was 88.2%, 69.7%, 76.0%, 60.0% and 92.0%, respectively. Among the 27 patients, in whom the muscle-invasiveness could be diagnosed by biopsy, 5 patients were diagnosed with muscle-invasiveness, all confirmed by pathology after surgery and 22 patients were diagnosed with none muscle-invasiveness, turned out to be 50% muscle-invasive and 50% none-muscle invasive after surgery. The accuracy was 59.3%. The accuracy of ureteroscopic biopsy to diagnosis multifocality was 61.0%. On univariate ana-lysis, biopsy grade was associated with postoperative pathology (P=0.001), while gender, age, side, body mass index (BMI), hematuresis, preoperative estimated glomerular filtration rate (eGFR), hydronephrosis, tumor size, location, multifocality and sessile were not associated with postoperative pathology grade. Biopsy grade (P=0.02), preoperative eGFR＜90 mL/(min·1.73 m2)(P=0.025) and tumor located in pelvis (P=0.049) were associated with muscle invasiveness. Gender, age, side, BMI, hematuresis, hydronephrosis, tumor size, multifocality and sessile were not significantly associated with muscle invasiveness. Conclusion: Ureteroscope and biopsy can assist risk stratification in upper tract urothelial carcinoma patients.

Key words: Urothelial carcinoma, Ureteroscopy, Biopsy, Neoplasm grading