The Development of a Reporting and Data System Using Ultrasound: My ACR O-RADS Journey

Supervising the development of the American College of Radiology Ovarian-Adnexal Imaging-Reporting and Data System for Ultrasound (ACR O-RADS US)1 has been a journey that has challenged and substantially improved my leadership and management skills.Rochelle F. Andreotti, MD

O-RADS is a quality assurance tool and clinical decision support system for the standardized description of ovarian/adnexal pathology and its management consisting of a lexicon and risk stratification system. It is 1 of 10 Reporting and Data Systems (RADS) sponsored by the American College of Radiology (ACR). The committee was formed in 2015 under the direction of the ACR Ultrasound Commission and Commissioner, Beverly Coleman. I was asked to Chair the committee with Dr. Phyllis Glanc from Toronto, Canada, as Vice-chair.

“The best and the brightest”

O-RADS is an international initiative that has involved extensive collaboration with competing national and international societies. We began in the summer of 2015 developing our mission and membership. Our membership was primarily derived from several major initiatives that prompted our formation. These included the SRU Consensus Statement, a North American initiative helpful in determining management of cystic lesions, the International Consensus, the first collaboration of European and North American management approaches promoting a more conservative, standardized approach while optimizing the referral pattern to a GYN-oncologist when malignancy is suspected and terms and risk stratification models developed by the International Ovarian Tumor Analysis Group (IOTA). It was also highly recommended that the committee consist of members representing national and international related societies who could contribute to and eventually help promote our system. As a result, from the beginning, I was facing highly opinionated, accomplished colleagues so that there would need to be lots of creative thinking to navigate the pathway going forward.

Lumper, not a splitter

I can see the overall picture and am an accomplished problem solver but concentrating on the smaller details is not my forte and I often find them cumbersome. In order to achieve group consensus, the next 2 years that we spent establishing the lexicon was a thought-provoking and prolonged experience in which both of these qualities were essential.

Ergo, I needed to step up my game.

Evaluating quality of evidence using a comprehensive scoring system was an early point of contention, but fairly quickly we were able to come to agreement that scoring articles for quality would not be of much concern in the lexicon phase, although evaluating the quality of the study would be useful if the article added support to the risk management phase.  The method chosen to develop the lexicon became a tedious process of culling evidence-based and frequently used terms from the literature using a survey, then through a consensus process, narrowing down the list to a workable group. Inevitably, since the IOTA terms were the most evidence-based, this became the foundation of the lexicon.

Looking back at other approaches, perhaps there may have been an easier, less time-intensive pathway that would also have led to the same results. Nevertheless, the process taught me that no matter how well thought out a strategy, always be prepared for others who, out of their own desire to work toward the greater good, will complicate the plan.

Let’s keep this as simple as possible

On a similar note to the “lumper” versus “splitter” mindset, we vigorously debated the specific modalities to be included in this system. There was no question that ultrasound (US) as the primary modality and magnetic resonance imaging (MRI) as a problem-solving tool were key. However, would it be prudent to add CT/PET, tools not recommended for these adnexal mass diagnoses, although occasionally demonstrating incidental findings?

Limiting our bandwidth to the two tracks was my recommendation. However, this high-spirited deliberation came close to splintering our fledgling committee, be it not for the ACR staff’s suggestion of a vote that finally put to bed the possibility of a third O-RADS track. The vote left us with the two original parallel US and MRI working groups, preventing much added unnecessary work and anxiety. From this encounter, I learned the value of highly polished social skills.

The European mathematical model and the North American pattern approach- the challenge of working internationally

The relationship of the Ultrasound Working Group of the ACR O-RADS Committee with the IOTA Group has been collaborative but, at times, complicated and contentious. The reasons for this were two-fold. Foremost, the IOTA Group had already developed a set of applicable terms that were evidence-based as well as validated mathematical models to risk stratify lesions and were most interested in expanding their influence. However, these European models, while highly accurate, were less accepted in North America where a pattern-recognition approach is generally more desirable. Since IOTA provided their cohort of over 5900 surgically proven lesions, to support our pattern approach, compromise needed to be reached regarding further incorporation into the O-RADS Ultrasound System.

In the early development of the risk stratification system at our 2017 meeting at ACR headquarters in Reston Virginia, Dr. Dirk Timmerman from Leuven, Belgium, our IOTA representative, first presented to the group a proposal of a dual approach with addition of the IOTA Simple Rules2. After further work using a more generalized pattern approach based upon IOTA data, this was not pursued.

However, later in 2019, we were confronted with the need to incorporate the more accurate, well-validated IOTA ADNEX mathematical model3 into the O-RADS system as an alternate approach. In this way, we were able to obtain acknowledgment from key players representing IOTA with the hope of allowing O-RADS US to be launched internationally in addition to North American acceptance.

With continued use of the system, I have found an extra advantage of incorporating the ADNEX model when evaluating higher risk lesions in that it adds additional specificity to the diagnosis, information greatly appreciated by the gynecologic oncologists.

Impact factor

Any success that I have had in the field of medicine can be attributed to a desire to influence and leave this world, in some way, a little better for it. My hope is that this data system will prove to be something that will make a meaningful contribution and be my legacy to women’s healthcare.

 

References:

  1. Andreotti RF, Timmerman D, Strachowski LM, et al. O-RADS US risk stratification and management system: A consensus guideline from the ACR Ovarian-Adnexal reporting and data system committee. Radiology 2020;294:168–185.
  1. Timmerman D, Van Calster B, Testa A, et al. Predicting the risk of malignancy in adnexal masses based on the Simple Rules from the International Ovarian Tumor Analysis group. Am J Obstet Gynecol 2016;214(4):424–437.
  1. Van Calster B, Van Hoorde K, Valentin L, et al. Evaluating the risk of ovarian cancer before surgery using the ADNEX model to differentiate between benign, borderline, early and advanced stage invasive, and secondary metastatic tumours: prospective multicentre diagnostic study. BMJ 2014;349:g5920.

 

Rochelle F. Andreotti, MD, is a Professor of Clinical Radiology and Obstetrics and Gynecology at Vanderbilt University College of Medicine in Nashville, Tennessee.

 

Interested in learning more about using O-RADS? Be on the lookout for the virtual course being held on September 26, 2020, New Approaches to Adnexal Mass Evaluation in North America: The Use of IOTA and O-RADS Systems; registration opens soon. Contact learn@aium.org for more information.

 

https://connect.aium.org/home

Leave a Reply

Please log in using one of these methods to post your comment:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.