Updated O-RADS US 2022: Better Specificity, Accuracy
Study confirms update improves risk scores, reduces false positives
A study from the University of Chicago (UChicago) confirms that the updated 2022 version of the Ovarian-Adnexal Reporting and Data System for Ultrasound (O-RADS US) is performing as intended, providing more accurate patient risk scores with fewer false-positive results.
In the study, 5.3% of patients were reclassified at a lower cancer risk for ovarian and adnexal lesions, based on imaging features when the 2022 version of O-RADS US was used, compared to the 2019 guidelines.
“I was very excited by the results of this study,” said Lori M. Strachowski, MD, co-chair of the O-RADS initiative governed by the American College of Radiology. “It’s exactly what we hoped would happen.”
Introduced in 2019, O-RADS US is based on sonographic morphologic features and an evidence-based lexicon. However, concerns about an observed overestimation of malignancy led an O-RADS US subcommittee to issue a 2022 update.
“Validation studies on the 2019 version of O-RADS US showed that lesions in the O-RADS 3 and 4 categories were turning out to be at the very low end of the risk malignancy range,” said Dr. Strachowski, professor emeritus, radiology and biomedical imaging, and obstetrics, gynecology and reproductive sciences at the University of California, San Francisco. “We realized that we needed to find some other descriptors that would suggest benignity and move these lesions into a lower category.”
Key updates to the 2022 O-RADS version include:
- New descriptors: bilocular cysts and acoustic shadowing to improve diagnostic specificity of lower-risk lesions.
- Expanded lexicon for classic benign lesions.
- Updated management guidelines, allowing the option for short-term US follow-up for O-RADS 3 category cystic lesions.
- Better alignment with existing consensus statements.
In the UChicago retrospective, single-center study, radiologists stratified consecutive US studies of 547 patients performed between 2017 and 2022 using both O-RADS US 2019 and 2022.
Of the patients within the cohort, 43% were postmenopausal and 17% were conservatively managed. The prevalence of malignancy was 19%.
According to lead researcher Ryan E. Longman, MD, section chief of OB/GYN, US, Genetics and Fetal Neonatal Care Center at UChicago, both O-RADS US versions were effective for stratifying patients into malignancy risk scores with high sensitivity and high negative predictive value at the 10% risk threshold.
However, the specificity and accuracy significantly increased with the updated version, demonstrating an AUC for the 2019 and 2022 versions at 0.901 and 0.905, respectively.
Among patients reclassified as lower-risk using O-RADS US 2022, the new classification was accurate in 28 of 29 cases. Sixteen patients were downgraded from a risk score of 3 to 2, and 13 patients moved from 4 to 3.
“The O-RADS US version 2022 is superior to version 2019,” Dr. Longman said. “It maximizes sensitivity while significantly improving the specificity of lower-risk lesions. The updated lexicon for describing cysts provides more criteria, allowing us to be more specific and accurate in our assessments.”
He noted that lesions downgraded under O-RADS US 2022 will spare some patients from surgery and reduce the need for follow-up US in others.
Research related to the optimal approach for confirming the stability and true benign nature of lesions managed conservatively remains ongoing.
“The O-RADS US version 2022 is superior to version 2019. It maximizes sensitivity while significantly improving the specificity of lower-risk lesions. The updated lexicon for describing cysts provides more criteria, allowing us to be more specific and accurate in our assessments.”
— RYAN E. LONGMAN, MD
Aligning the Major Guidelines
For O-RADS US 2022, the committee also considered user feedback and opportunities to align with other major guidelines.
In the original O-RADS US system, a lesion with one or more complete septations was considered multilocular and automatically assigned to a higher risk of malignancy range, particularly if it lacked solid tissue.
“We realized that it can be very challenging for the interpreter to distinguish between two cysts or two follicles adjacent to each other, and that normal intervening tissue might appear like a septation,” Dr. Strachowski said. “We were probably inadvertently mischaracterizing this as multilocularity.”
Following the success of the Society of Radiologists in Ultrasound (SRU) guidelines for simple adnexal cysts, O-RADS US 2022 now recommends the same management for a cyst with an apparent single septation and smooth inner walls (bilocular cyst) and a simple cyst. Recent large studies have demonstrated an exceptionally low risk of cancer associated with simple adnexal cysts.
“Aligning the major guidelines is a goal of the committee and will help eliminate confusion,” Dr. Strachowski said.
The updated guidelines also offer another management option for O-RADS category 3, which in the original version recommended interpretation by an “US specialist” or by MRI.
“We found that radiologists were recommending MRIs for all lesions in this category,” Dr. Strachowski said. “For those lesions that undergo active surveillance, there is now an option for short-term follow-up with US. The exception is the solid lesion in the O-RADS 3 category, that warrants an MRI.”
Dr. Strachowski said the committee is currently working on additional updates to O-RADS, including management recommendations following MRI as well as better aligning the lexicon terminology between the US and MRI arms of the system. The new version is expected in the spring of 2026.
“We hope this University of Chicago study is the first of many more validation trials of O-RADS US 2022 from other institutions,” she added. “We also encourage researchers to consider a cost-benefit analysis from the reduction of MRI for low-risk lesions.”
For More Information
Access the Radiology study, “O-RADS US Version 2022 Improves Patient Risk Stratification When Compared with O-RADS US Version 2019.”
Read previous RSNA News stories on women’s imaging: