The preliminary study about feasibility and accuracy of the frozen section during targeted prostate biopsy / 中华泌尿外科杂志

ABSTRACT

Objective:To investigate the feasibility and reliability of the frozen section during targeted prostate biopsy.Methods:The clinical and pathological information of patients who received cognitive fusion transperineal targeted plus systematic biopsy and frozen section of 1-2 core targeted biopsy were consecutively collected and retrospectively studied. The median age was 70 (ranging 64-78) years, with the median prostate-specific antigen (PSA) level of 11.00 (ranging 6.63-16.52) ng/ml and the median prostate volume of 35.72 (ranging 22.59-47.71) ml. All patients received bi-parametric magnetic resonance imaging (bp-MRI) and have Prostate Imaging Reporting and Data System (PI-RADS) 3 or higher lesions diagnosed on bp-MRI. The suspected lesions would be taken by targeted biopsy of which one or two cores would be sent to prepare for the frozen sections. Then a cognitive fusion targeted and systematic biopsy covering the above targeted zones would be routinely administered under a transperineal approach as a standard protocol. The total time used for diagnosis of the frozen sections, the pathological diagnosis and the International Society of Urological Pathology (ISUP) grade groups (GG) would be recorded. The sensitivity, the positive predictive value, and the accuracy on grade groups would be analyzed, using the pathological diagnosis based on standard sections from the same targeted lesion.Results:A total of 29 patients were included in this study. Accordingly, 29 suspected lesions were identified on bp-MRI. A total of 20 lesions were finally diagnosed of PCa on frozen section, with the detection rate of 69.0%. Of those, 9(45.0%) cases were ISUP GG 1 diseases, 5(25.0%) cases were GG 2 diseases, 1(5.0%) case was GG 3 disease, and 5(25.0%) cases were GG 4-5 diseases. A total of 22 lesions were diagnosed with PCa on standard sections of cores from the same targeted lesions, with the detection rate of 75.9%. Of those, 6(27.3%) cases were GG 1 disease, 11(50.0%) cases were GG 2 diseases, 1(4.5) case was GG 3 disease, and 4(18.2%) cases were GG 4-5 diseases. The sensitivity and the positive predictive value of frozen section were 90.9% and 100%, respectively. No false positive diagnosis was made by frozen section. Compared to diagnosis from frozen sections, the GG diagnosed from final standard sections were found to upgrade and downgrade in 2 and 2 cases, respectively. The accuracy rate on GG of frozen sections was 80%. The time used for the diagnosis of frozen sections was (11±2) minutes. The histology quality control of four specimens was dissatisfactory. Two were due to tissue loss and deformation during sampling, and the other two were due to cytoclasis during low-temperature transferring.Conclusion:It is feasible and reliable to make a pathological diagnosis from frozen section of prostate targeted biopsy.