Premium
Feasibility and accuracy of pathological diagnosis in en‐bloc transurethral resection specimens versus conventional transurethral resection specimens of bladder tumour: evaluation with pT1 substaging by 10 pathologists
Author(s) -
Yanagisawa Takafumi,
Yorozu Takashi,
Miki Jun,
Iwatani Kosuke,
Obayashi Koki,
Sato Shun,
Kimura Takahiro,
Takahashi Hiroyuki,
Egawa Shin
Publication year - 2021
Publication title -
histopathology
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.626
H-Index - 124
eISSN - 1365-2559
pISSN - 0309-0167
DOI - 10.1111/his.14307
Subject(s) - concordance , medicine , interquartile range , bladder cancer , pathological , cancer , urology , bladder neoplasm , carcinoma in situ , pathology , surgery
Aims En‐bloc transurethral resection (TUR) of bladder tumour (ERBT) is designed to provide more accurate pathological diagnosis of specimens than conventional TUR of bladder tumour (cTURBT). Some studies have reported that T1 bladder cancer substage could be a prognostic factor in assessing tumour progression, but such substaging has not been widely adopted because of problems with pathological diagnosis using cTURBT specimens. The aim of this study was to evaluate the possible advantages of en‐bloc TUR specimens in T1 substaging following assessment by a panel of 10 pathologists. Methods and results We assessed the substages in 123 patients (cTURBT, n = 91; ERBT, n = 32) who were diagnosed with pT1 bladder cancer. We randomly selected 10 ERBT specimens and 10 cTURBT specimens with cancer invasion areas equivalent to those of their corresponding ERBT specimens. Ten pathologists performed pT1 substaging for pT1a/b/c and pT1m/e in 20 patients (cTURBT, n = 10; ERBT, n = 10). We evaluated diagnostic times and rates of diagnostic concordance among these pathologists, comparing cTURBT and ERBT. The median diagnostic times per slide were 87.7 s [interquartile range (IQR) 71.9–109.2 s) for cTURBT and 54.7 s (IQR 46.0–59.6 s) for ERBT ( P = 0.009). The rate of diagnostic concordance was significantly better for ERBT specimens. For pT1a/b/c, the median concordance rates were 50% for cTURBT and 80% for ERBT ( P = 0.02); for pT1m/e, the median concordance rates were 70% for cTURBT and 90% for ERBT ( P = 0.05). For pT1a/b/c, the average κ ‐values between the pathologist and the standard diagnosis were 0.04 for cTURBT and 0.47 for ERBT. Conclusions The use of ERBT specimens shortened the diagnostic time and minimised interobserver variability for T1 substaging compared with the use of cTURBT specimens.