Prospective Evaluation of the Accuracy of A Training Program in Image Recognition by Narrow-Band Imaging Guided Hysteroscopy of Endometrial Neoplasms

Abstract

Introduction: Narrow-band imaging (NBI) hysteroscopy by experienced hysteroscopists (EH) is useful for diagnosing endometrial neoplasms. Objective: We investigated whether the diagnostic reliability of NBI could be improved by specific professional training. Methods: Three levels of trainees who were Surgeons at our hospital were selected. Level I: 6 trainees had no prior hysteroscopic experience; level II: 6 trainees had experience with <100 cases; and level III: 6 trainees had <500 cases. The two-hour training program for white light hysteroscopy (WLH) and NBI included information on the classifications of diseases of the uterine cavity and on the features of diagnostic images. Images from 529 patients were evaluated independently by trainees with 3 levels of before and after training, and by EH. Trainees and EHs had to analyze and arrive at a hysteroscopic diagnosis for each image that was compared to the pathological diagnosis for diagnostic accuracy. Results: After training, all levels achieved higher diagnostic accuracy with NBI than was seen with WLH. Level III trainees achieved diagnostic accuracy and kappa values for NBI that were equivalent to those of EH. Conclusions: Training can increase the diagnostic skill of all trainees using NBI, especially for trainees with prior hysteroscopic experience.

Improving tissue characterization, differentiation and diagnosis in gynecology with the narrow‑band imaging technique: A systematic review

“…NBI has proven to be an efficacious approach for the diagnosis of endometrial cancer and hyperplasia, while thorough training improves the trainee's diagnostic skills to an extent depending on their previous hysteroscopic experience ( 37 ). However, it must be stressed that NBI laparoscopy was not superior in the detection of peritoneal metastases in comparison with standard WL laparoscopy alone, which was reported by Aloisi et al ( 39 ) and Schnelldorfer et al ( 45 ).…”

“…Wang et al ( 37 ) reported that the physician who performs hysteroscopy must be familiar with endometrial lesions, which are influenced by estrogen and progesterone secretion. Endometrial necrosis may not only be observed in malignant lesions, but also frequently appears in benign hyperplastic lesions associated with abnormal uterine bleeding.…”

“…A total of 11 studies described the use of NBI in hysteroscopy for the detection of endometrial pathologies (9,10,(19)(20)(21)(22)(23)(24)(25)36,37) and another 11 studies reported the use of NBI in laparoscopy for the detection of peritoneal endometriosis (11)(12)(13)(26)(27)(28)(29)(30)33,34,37,38). Furthermore, three articles dealt with the use of NBI in digital colposcopy for the detection of lower genital tract pathologies (14,31,32) and three articles reported on the use of NBI in laparoscopy for the detection of peritoneal metastases associated with ovarian cancer and other gynecological malignancies (15)(16)(17).…”

“…Details on the selected studies reporting the use of NBI in hysteroscopy are presented in Table I. A total of 11 studies were included in the review, dating from 2009-2020 with a combined cohort of 2,424 female patients (9,10,(19)(20)(21)(22)(23)(24)(25)36,37). Of these, eight studies were prospective studies (Canadian Task Force classification, II-2) (19,23,36) and two studies were case reports (9,10).…”

“…4 these, 17 had a prospective design lacking randomization and two were randomized controlled trials with control groups. A total of 9 studies were indicated to have a high risk of bias (24,25,27,29,(30)(31)(32)34,36) and 10 studies were determined to have a low risk of bias (19)(20)(21)23,28,30,32,(37)(38)(39).…”