Learning Curve for the Detection of Deep Infiltrating Endometriosis and Adenomyosis with 3-D Transvaginal Water Contrast Sonography

Learning Curve for Ultrasound Assessment of Myometrial Infiltration in Endometrial Cancer Visualizing Videoclips: Potential Implications for Training

BACKGROUND

Diagnostic accuracy for estimating myometrial infiltration by ultrasound in endometrial cancer requires experience. The objective of this study is to determine the learning curve (LC) for assessing myometrial infiltration in cases of endometrial cancer using transvaginal ultrasound (TVS).

METHODS

Five trainees (one staff radiologist and four fourth-year OB/GYN residents) participated in this study. All trainees had experience in performing TVS, but none of them had specific training on the assessment of myometrial infiltration. Trainees were given one specific lecture about the topic, and then they observed videoclips from 10 cases explained by the trainer. After this, all trainees visualized 45 videoclips of uterine ultrasound scans of endometrial cancer cases. The assessment of myometrial infiltration was based on the subjective impression. Definitive histology was used as a reference standard. Trainees stated whether myometrial infiltration was ≥50% or <50%. LC-CUSUM and standard CUSUM graphics were plotted to determine how many cases were needed to reach competence, allowing a mistake rate of 15%.

RESULTS

All trainees completed the study. LC-CUSUM graphics showed that three trainees reached competence at the 33rd, 35th and 36th case, respectively. All three of them kept the process under control after reaching competence. One trainee reached competence but did not maintain it in the cumulative analysis. One trainee did not reach competence.

CONCLUSION

Our study suggests that 30-40 cases would be needed to be trained for assessing myometrial infiltration by TVS by visual interpretation of videoclips by most trainees.

Evaluation of an e-learning program for the diagnosis of rectosigmoid endometriosis with rectal water contrast transvaginal ultrasonography (rectosonography)

This study aimed to evaluate the value of an e-learning program for the diagnosis of rectosigmoid endometriosis lesions using rectal water contrast transvaginal ultrasonography (rectosonography/RSG). Theoretical RSG training using videos with a commentary was offered online to healthcare professionals involved in ultrasound screening for endometriosis. A test (without correction) with 24 RSG video loops was used to assess the participants' baseline level before the training and their improvement afterwards. If the success rate post-training was below 80 %, the participant could start over with another series of 24 videos. Between February and June 2020, thirty participants took the training course (of which 80 % were obstetrics-gynaecology residents). The e-learning program resulted in a significant performance increase in the diagnosis of rectosigmoid endometriosis lesions, with a higher test success rate after the training compared to before (74.4 % and 63.6 % respectively; +10.8 %; 95 % CI [6,6; 15]; p < 0.001). Significant improvement was also observed regarding the overall skills involved in the ultrasound diagnosis of deep infiltrating endometriosis (+9.2 %; p < 0.001), the accurate diagnosis of the height of bowel lesions (+14.7 %; p < 0.001) and uterosacral ligament lesions (+8%; p < 0.005). In conclusion, our e-learning program led to a significant improvement of the diagnostic performance of digestive endometriosis using transvaginal ultrasound with intrarectal water contrast (rectosonography). Adding feedback to the post-test video loops could further increase the efficacy of the e-learning training.