Ovarian cysts are incredibly common. It is important for imaging professionals to issue accurate reports so that benign lesions are not worked up unnecessarily, whilst not dismissing potentially significant ones.
Many ovarian lesions are physiological (especially in pre-menopausal women) and will resolve on their own. Some are benign but may need follow-up due to potential for complications such as dermoid cysts (teratomas) and endometriomas. Some lesions are worrying enough to be referred straight to a gynaecological oncologist.
Ovarian malignancies are uncommon but are the leading cause of death amongst gynaecological malignancies. There is a much better survival rate if operated on by sub-specialist gynaecological oncologists. However, many lesions can be managed either by a general gynaecologist, or by repeat imaging alone.
Radiologists rely on pattern recognition to decide if a lesion is worrying or not. Until recently, much of the data to determine accuracy has been largely based on expert opinion.
The International Tumour Analysis group (IOTA) has led the way over the past 20 years in trying to standardise nomenclature for adnexal lesions. Other groups, such as the Society of Radiologists in Ultrasound (SRU) have attempted to produce guidelines to help radiologists in their recommendations for management of adnexal lesions.
The IOTA group have produced a risk stratification tool called the Assessment of Different NEoplasias in the AdNEXa (ADNEX) model using clinical and ultrasound features to determine a lesion’s risk of malignancy. Its accuracy has been validated in multiple trials and has won the hearts of many users (including mine!) for its ease of use and practicality.
It has also been popular with clinicians as it helps guide whether the lesion requires referral to a specialist and, if so, to whom (i.e. a general gynaecologist or gynaecological oncologist).
Multiple national societies have recommended the use of the IOTA ADNEX model, including the American College of Obstetricians and Gynaecologists, and the Royal College of Obstetricians and Gynaecologists. The IOTA ADNEX model has been shown prospectively to have an accuracy rate of 92-95% with a false positive rate of 10%. It is much more accurate that the Risk of Malignancy Index (RMI). CA-125 can be incorporated into the calculation if available, but this will only differentiate between Stage 1 Ovarian cancer from Stage 2-4 Ovarian cancer.
In 2018, the American College of Radiology published a risk stratification classification system called O-RADS, which also has guidelines for management.
O-RADS can be used for ultrasound and MRI, but is most useful for ultrasound, as the usual first line modality for investigating pelvic lesions. MRI is usually used for problem solving. O-RADS can be used for all adnexal lesions and has been vigorously studied in low- and high-risk populations.
There are 5 categories of risk in the O-RADS classification:
- O-RADS 1 – Normal ovary
- O-RADS 2 – Almost certainly benign (<1% risk of malignancy
- O-RADS 3 – Low risk of malignancy (1 to <10% risk of malignancy)
- O-RADS 4 – Intermediate risk of malignancy (10 to <50% risk of malignancy)
- O-RADS 5 – High risk of malignancy (>50% risk of malignancy).
Faced with an adnexal lesion, a radiologist will use pattern recognition to classify a lesion. Most lesions will be easily categorised as normal (i.e. simple follicle or corpus luteum – O-RADS 1 lesions), or simple cysts/ haemorrhagic cysts/ endometriomas (O-RADS 2 lesions) or dermoids (O-RADS 3 lesions).
Most lesions seen on imaging are benign, and are in the O-RADS 1, 2 or 3 categories. Some lesions are harder to categorise. If unsure what the lesion is, the IOTA-ADNEX can help identify the risk category. This then gives a lesion a numerical risk of malignancy (e.g. 25%), and then it can be classified into one of the O-RADS categories of risk (e.g. O-RADS 3).
Management will then be dictated on this basis, or by using the O-RADS Chart which is available on the ACR website as a free resource (https://www.acr.org/Clinical-Resources/Reporting-and-Data-Systems/O-Rads).
Radiologists are increasingly using the IOTA ADNEX model for risk calculation of adnexal lesions and the O-RADS reporting system as a simple, user-friendly and practical tool to ensure accurate reporting and appropriate follow-up for women with adnexal lesions.
Key messages
- Ovarian lesions are common and assessment challenging
- The O-RADS system has advanced risk stratification
- Better assessment leads to better outcomes.
– References available on request
Author competing interests – nil