The aim of this study was to perform an external validation of GI-RADS classification system to differentiate between benign and malignant adnexal masses. This is a retrospective diagnostic accuracy study of data collected prospectively from patients with adnexal pathology who underwent transvaginal ultrasound, performed by gynecologists and radiologists, experts in ultrasound, prior to surgery. This study was conducted in a tertiary university oncology hospital. Data were collected between January of 2012 and December of 2018. We used the GI-RADS classification system based on the subjective impression of the examiner. Definitive pathology after tumour surgical removal was the reference standard used in this study. Sensitivity, specificity, positive (PPV) and negative predictive values (NPV) were calculated. For this calculation, GI-RADS 1-3 were considered as benign and GI-RADS 4-5 were considered as malignant. A total of 587 patients underwent an ultrasound examination in our institution. 24 women were excluded for they were not operated. Thus, a total of 563 women were ultimately included (415 premenopausal and 148 postmenopausal). 35 women had bilateral masses, so 608 masses were evaluated. Final pathology revealed 65 malignant and 543 benign tumors. GI-RADS classification system had a sensitivity, specificity, PPV and values NPV of 91%, 85%, 44% and 99%, respectively. In this external retrospective validation study GI-RADS classification system showed good diagnostic performance. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.