During routine antenatal ultrasound examinations, an ovarian mass can be found incidentally. In clinical practice, the differential diagnosis between benign and malignant ovarian masses is essential for planning further management. Ultrasound imaging has become the most popular diagnostic tool during pregnancy, with the recent development of ultrasonography. In non-pregnant women, several methods have been used to predict malignant ovarian masses before surgery. The International Ovarian Tumor Analysis (IOTA) group reported several scoring systems, such as the IOTA simple rules, IOTA logistic regression models, and IOTA assessment of different NEoplasias in the adneXa. Other researchers have also evaluated the malignancy of ovarian masses before surgery using scoring systems such as the Sassone score, pelvic mass score, DePriest score, Lerner score, and Ovarian-Adnexal Reporting and Data System. These researchers suggested specific features of ovarian masses that can be used for differential diagnosis, including size, proportion of solid tissue, papillary projections, inner wall structure, locules, wall thickness, septa, echogenicity, acoustic shadows, and presence of ascites. Although these factors can also be measured in pregnant women using ultrasound, only a few studies have applied ovarian scoring systems in pregnant women. In this article, we reviewed various scoring systems for predicting malignant tumors of the ovary and determined whether they can be applied to pregnant women.
Malignant ovarian tumors are the seventh most common malignancy in women worldwide, accounting for 3.7% of all female cancer cases. These tumors have a high mortality rate owing to late detection, with 67% of the patients already having progressive disease at the time of diagnosis [
Although there have been no definite ultrasound-based diagnostic criteria for ovarian cancer, many efforts have been made to distinguish benign from malignant tumors using ultrasonographic findings. Various ovarian scoring systems have been proposed for non-pregnant gynecological patients. With recent advances in ultrasonography, some studies have reported the detailed features of tissue composition using shades of colors and attempted to diagnose the histological and clinical stages of ovarian cancer before surgery [
Ultrasonography is the most commonly used tool for evaluating ovarian tumors during pregnancy because of its relative safety. As other diagnostic modalities, such as computed tomography (CT) or magnetic resonance angiography with contrast, cannot be used during pregnancy, the application of an ovarian scoring system based on the findings of ultrasonography, which can be safely performed during pregnancy, may help determine the timing of surgery during or after pregnancy.
In this article, we reviewed several scoring systems for predicting malignant ovarian tumors and discussed the applicability of the scoring systems during pregnancy.
The IOTA study is the largest study on the accuracy of ultrasound for the diagnosis of ovarian tumors, which has brought many benefits to the field of transvaginal ultrasonography (TVS). Studies have been conducted to explain the morphological and Doppler ultrasound characteristics of these tumors [
The IOTA group was started 15 years ago to create an appropriate “evidence-based” algorithm for all types of ovarian tumors. IOTA includes several scoring systems, including the IOTA simple rules, IOTA logistic regression (LR) models, and IOTA assessment of different NEoplasias in the adneXa (ADNEX) model.
The first scoring system is the IOTA simple rules. This is based on a set of five ultrasound features indicative of a benign tumor (B-features) and five ultrasound features indicative of a malignant tumor (M-features). If only B-features are observed, the ovarian tumor is classified as benign. If the ovarian tumor shows only M-features, it is classified as malignant. However, this scoring system has a disadvantage when both B- and M-features or features that do not correspond to these criteria are observed [
The simple rules system was reported to have a sensitivity of 93% and a specificity of 90% [
In 2005, more universally useful LR models, IOTA LR1 and LR2, were developed to distinguish between malignant and benign adnexal tumors before surgery [
The IOTA LR2 is calculated based on six of the above criteria: (3), (6), (7), (9), (10), and (11). The formula used to determine the probability of malignancy is as follows: y=1/(1+exp−z), where z=−5.3718+0.0354 (3) +1.6159 (6) +1.1768 (7) +0.0697 (9) +0.9586 (10) −2.9486 (11). As with LR1, the probability y is dichotomized at a score of 0.1 to make a predictive diagnosis of cancer [
The IOTA ADNEX is the first risk model to differentiate between benign and four types of malignant ovarian tumors (borderline, stage I cancer, stage II–IV cancer, and secondary metastatic cancers). The ADNEX model includes three clinical factors and six ultrasound factors. The clinical factors include age (years), serum CA125 level (U/mL), and type of the center where the patient underwent ultrasound. The six predictors in the ADNEX model are ultrasound variables, as follows: maximal diameter of the lesion (mm), percentage of solid tissue (%), number of papillary projections (0, 1, 2, 3, and >3), presence of more than 10 cyst locules (yes/no), acoustic shadows (yes/no), and presence of ascites (yes/no) [
The Sassone score was developed by Sassone et al. [
PMS is a scoring system proposed by Rossi et al. [
In this formula, SASS is the numeric value of the Sassone score, log(CA125) is the base 10 logarithm of the CA125 level, VAS is the type of vascularization (peripheral=1, central/septal=2), MS is the menopausal state (pre-menopausal=1, post-menopausal=2), and RI is the numeric value of the resistance index of the pelvic mass (
DePriest et al. [
In 1993, the Lerner score was developed. It included four parameters: wall structure, shadowing, septa, and echogenicity. The modified scoring system proposed by Lerner et al. [
Scores of 0 to 3 were developed for each category according to specific criteria. First, the wall structure was divided into three categories: smooth or small irregularities <3 mm (0 points), solid or non-applicable (2 points), and papillarities ≥3 mm (3 points). Second, the shadowing score consisted of “yes” (0 points) and “no” (1 point) categories. Third, the septa were divided into two categories: none or thin (0 points) and thick (1 point). Fourth, echogenicity was also divided into two categories: sonolucent or low-level echo or echogenic core (0 points) and mixed or high (3 points) (
In 2018, the American College of Radiology divided ovarian and adnexal tumors into six categories to construct a unified imaging report for communication and quality improvement. The six categories were as follows: 1) description of major categories, physiological category or not; 2) description of size; 3) description of solid lesions; 4) description of cystic lesions; 5) description of vascularity; and 6) general and extraovarian findings. First, the major categories were divided into two categories: physiological category (follicle and corpus luteum) and lesion category (unilocular, unilocular cyst with solid component, multilocular cyst without solid elements, multilocular cyst with solid component, and solid or solid appearance). Second, size was defined as the maximum diameter. Third, solid lesions were divided into two categories: external contour (smooth and irregular) and internal content (acoustic shadowing). Fourth, cystic lesions included the inner margin (papillary projection or nodule, and smooth/irregular) and internal content (anechoic fluid and hyperechoic components). Fifth, vascularity was divided into four color scores (CS): 1 (no flow), 2 (minimal flow), 3 (moderate flow), and 4 (very strong flow). Sixth, the general and extra-ovarian findings were as follows: cul-de-sec fluid, ascites, and peritoneal thickening or nodules.
With reference to these descriptions, risk classification was performed for tumors and six O-RADS scores were created. The meaning of each score was as follows: O-RADS 0, incomplete evaluation; O-RADS 1, physiological category (normal premenopausal ovary); O-RADS 2, almost certainly benign category (1% risk of malignancy) [
In the validation study of O-RADS, the optimal cutoff value for predicting malignancy was O-RADS >3. The sensitivity of O-RADS was 98.7%, and the specificity of this system was 83.2% [
Several authors have reported that 51% to 92% of adnexal masses resolve during pregnancy [
Previous studies evaluated only a few ultrasound functions during pregnancy, and most studies focused on outcomes in the presence of an ovarian mass (
Bernhard et al. [
The most recent study (published in 2021) evaluated the accuracy of malignancy prediction between ovarian scoring systems and subjective assessment [
Recently, we reported a multicenter retrospective study on ovarian mass scoring systems during pregnancy [
The factors of representative ovarian scoring systems studied thus far (IOTA, Sassone score, PMS, DePriest score, Lerner score, and O-RADS) are compared in
In the Sassone scoring system, high echogenicity had the highest score, and the larger the solid portion, the higher the probability of malignancy. The DePriest and Lerner scores also increased as the size of the solid portion increased. In the Lerner score, high scores were assigned to mixed or high echogenicity and papillarity in the wall structure. In addition, only the IOTA ADNEX and PMS included CA125 level.
In the analysis using the Lerner scoring system, it was also found that papillary masses within the inner wall structure category had a greater percentage of malignancy than masses that were mostly solid. For this reason, papillary masses were assigned a higher weight. Significant values were obtained in the regression analysis using these four variables, and regression coefficients were used to calculate the final weight of each variable [
In pregnant women, the incidence of adnexal masses is approximately 0.05% to 3.2% [
Most ovarian masses are discovered incidentally during routine ultrasound examinations in pregnant women [
With the developments in ultrasound technology, many studies have been conducted on scoring systems for malignancy evaluation using ultrasound characteristics in non-pregnant women; however, no scoring system has been applied to pregnant women.
When making clinical decisions in pregnant women with ovarian masses, both the mother and the fetus should be considered, which makes the clinical decision more complex. Pregnancy complications and malignancy are the most important factors in the decision making process.
If the ovarian scoring systems mentioned above are applied to pregnant women, the following should be noted: first, as the gestational weeks increase, ascites become difficult to detect with ultrasound in pregnant women because they are hidden in the uterus. Second, the diagnostic utility of serum tumor markers in women with ovarian masses found during pregnancy remains controversial. The CA125 level is elevated during pregnancy. It peaks in the first trimester (range, 7.251 units/mL) and steadily decreases thereafter [
Ultrasound is helpful in evaluating pregnant women with ovarian masses. Various ultrasound features of ovarian masses of various etiologies can be evaluated in pregnant women in actual clinical practice. Some ultrasound features may raise the suspicion of malignancy, including but not limited to the presence of solid components, increased wall thickness, multi-location large tumors with a maximum diameter of >6 cm, and total internal septum >2–3 mm. The presence of papillary protrusions, increased vascularity during Doppler examination, and ascites can also be clues to suspect malignancy [
Other imaging tests, such as CT, are inappropriate for assessing ovarian masses during pregnancy because of fetal radiation effects. Pelvic ultrasonography is considered the modality of choice for evaluating ovarian masses found during pregnancy and is suitable for guiding surgical intervention, if necessary [
In this review, we investigated the usefulness of existing ovarian scoring systems in pregnant women with ovarian masses. Unlike for general gynecological patients, the available imaging modalities for pregnant women are limited. Ultrasonography can be the optimal tool considering that pregnant women undergo ultrasound examinations several times during prenatal care. However, few research results have been published on whether the prediction of malignancy using ovarian scoring systems in pregnant women is accurate. Hence, more studies are needed on the application of scoring systems for the evaluation of ovarian mass malignancy in pregnant women.
We are grateful to the patients included in this study. We thank the medical staff for their assistance.
Han Sung Hwang has been an Editorial Board of Obstetrics & Gynecology Science; however, he was not involved in the peer reviewer selection, evaluation, or decision process of this article. Otherwise, no other potential conflicts of interest relevant to this article were reported.
No potential conflict of interest relevant to this article was reported.
This study does not require approval of the Institutional Review Board because no patient data is contained in this article.
Not applicable.
None.
Ovarian mass scoring system: IOTA
Scoring system | Score | |
---|---|---|
IOTA simple rules | Benign tumor (B-features) | Malignant tumor (M-features) |
B1: Unilocular | M1: Irregular solid tumor | |
B2: Presence of solid components where the largest solid component has a largest diameter <7 mm | M2: Presence of ascites | |
B3: Presence of acoustic shadows | M3: At least four papillary structures | |
B4: Smooth multilocular tumor with largest diameter <100 mm | M4: Irregular multilocular solid tumor with largest diameter ≥100 mm | |
B5: No blood flow (color score 1) | M5: Very strong blood flow (color score 4) | |
|
||
IOTA LR |
y=1/(1+e−z), where z=−6.7468+1.5985 (1) −0.9983 (2) +0.0326 (3) +0.00841 (4) −0.8577 (5) +1.5513 (6) +1.1737 (7) +0.9281 (8) +0.0496 (9) +1.1421 (10) −2.3550 (11) +0.4916 (12) | |
|
||
IOTA LR |
y=1/(1+exp–z), where z=−5.3718 +0.0354 (3) +1.6159 (6) +1.1768 (7) +0.0697 (9) +0.9586 (10) −2.9486 (11) | |
|
||
IOTA ADNEX | They used IOTA ADNEXA calculator. The factors are follows; age of the patient at examination, oncology center, maximal diameter of the lesion, maximal diameter of the largest solid part, more than 10 locules, number of papillations acoustic shadows present, ascites present, and serum CA125 or without serum CA125. |
IOTA, International Ovarian Tumor Analysis; LR, logistic regression; ADNEX, assessment of different NEoplasias in the adneXa.
LR factor: (1) personal history of ovarian cancer (yes=1, no=0), (2) current hormonal therapy (yes=1, no=0), (3) age of the patient (in years), (4) maximum diameter of the lesion (in millimeters), (5) the presence of pain during the examination (yes=1, no=0), (6) the presence of ascites (yes=1, no=0), (7) the presence of blood flow within a solid papillary projection (yes=1, no=0), (8) the presence of a purely solid tumor (yes=1, no=0), (9) maximal diameter of the solid component (expressed in millimeters, but with no increase >50 mm), (10) irregular internal cyst walls (yes=1, no=0), (11) the presence of acoustic shadows (yes=1, no=0), and (12) the color score (1, 2, 3, or 4).
Ovarian mass scoring system: SASS and PMS
Scoring system | Score | ||||
---|---|---|---|---|---|
SASS | |||||
Inner wall structure (mm) | Score 1: smooth | Score 2: irregularities ≤3 | Score 3: papillarities >3 | Score 4: not applicable, mostly solid | - |
Wall thickness (mm) | Score 1: thin (≤3) | Score 2: thick (>3) | Score 3: not applicable, mostly solid | - | - |
Septa (mm) | Score 1: none | Score 2: thin (≤3) | Score 3: thick (>3) | - | - |
Echogenicity | Score 1: sonolucent | Score 2: low echogenicity | Score 3: low echogenicity with echogenic core | Score 4: mixed echogenicity | Score 5: high echogenicity |
PMS |
|
SASS, Sassone score; PMS, pelvic mass score; log(CA125), 10 logarithm of the CA125 levels; VAS, type of vascularization (peripheral=1; central/septal=2); MS, menopausal state (pre-menopausal=1; post-menopausal=2); RI, numeric value of the resistance index of the pelvic mass.
Ovarian mass scoring system: DePriest score and Lerner score
Scoring system | Score | ||||
---|---|---|---|---|---|
DePriest score | 0 | 1 | 2 | 3 | 4 |
Volume | <10 cm3 | 10–50 cm3 | >50–200 cm3 | >200–500 cm3 | >500 cm3 |
Cyst wall structure | Smooth (<3 mm thickness) | Smooth (≥3 mm thickness) | Papillary projection (<3 mm) | Papillary projection (≥3 mm) | Predominantly solid |
Septa structure | No septa | Thin septa (<3 mm) | Thick septa (3 mm to 1 cm) | Solid area (≥1 cm) | Predominantly solid |
Lerner score | |||||
Wall structure (mm) | Smooth or small irregularities <3 | - | Solid or not applicable | Papillarities ≥3 | |
Shadowing | Yes | No | - | - | |
Septa (mm) | None or thin (<3) | Thick (≥3) | - | - | |
Echogenicity | Sonolucent or low-level echo or echogenic core | - | - | Mixed or high |
Ovarian mass scoring system: ORADS
O-RADS 0 | O-RADS 1 | O-RADS 2 | O-RADS 3 | O-RADS 4 | O-RADS 5 | |
---|---|---|---|---|---|---|
Definition | Incomplete evaluation | The physiologic category (normal premenopausal ovary) | The almost certainly benign category (1% risk of malignancy) | Lesions with low risk of malignancy (1% to 10%) | Lesions with intermediate risk of malignancy (10% to 50%) | Lesions with high risk of malignancy (>50%) |
Characteristic | - | - | Simple cysts, Unilocular cysts with smooth walls, and maximal size of ovarian mass is less than 10 cm | Unilocular cyst (≥10 cm), typical dermoid cysts, endometriomas, and hemorrhagic cysts (≥10 cm); unilocular cyst, any size with irregular inner wall (<3 mm height); multilocular cyst (<10 cm); Smooth inner wall, CS=1–3; any size solid smooth mass with CS=1 | Multilocular cyst without solid component, Unilocular cyst with solid component, multilocular cyst with solid component, and solid | Unilocular cyst, any size, ≥4 papillary projection, CS=any, any size multilocular cyst with solid component, CS=3–4, any size solid smooth with CS=4, any size solid irregular mass with CS=any; ascites and/or peritoneal nodules |
O-RADS, Ovarian-Adnexal Reporting and Data System.
Previous studies on ultrasound imaging of ovarian tumors during pregnancy
Study | Study population | Type of studies | Objective | Conclusion |
---|---|---|---|---|
Platek et al. [ |
31 | Retrospective study (in USA, Bronx Municipal Hospital Center and the Weiler Hospital of the Albert Einstein College of Medicine) | To evaluate the pathological features and outcomes of pregnancies complicated by persistent adnexal masses managed conservatively or surgically |
- Ovarian cancer in pregnancy is rare, but the frequency of ovarian cysts diagnosed in pregnancy is much higher - Complications of abdominal surgery may be increased in pregnancy |
Bromley and Benatcerraf [ |
125 | Retrospective study (single hospital in USA, Massachusetts General Hospital) | To determine the accuracy of ultrasound diagnosis and perinatal results in pregnant women with ovarian mass |
- Prenatal sonography can accurately characterize maternal adnexal lesions - Sonographic appearance; simple cyst, dermoid appearance, complex cyst, septate cyst etc. |
Bernhard et al. [ |
422 | Retrospective study (single hospital in USA, Washington University Medical Center) | To determine factors predicting the persistence of sonographically identified adnexal masses in pregnancy |
- Most adnexal masses were small, simple cysts that did not pose a risk to the pregnancy - Even most large-scale or ultrasonically complex ovarian masses have spontaneously disappeared - The best predictors of persistent ovarian mass were ultrasound shape and size |
Schmeler et al. [ |
59 | Retrospective study (single hospital in USA, Women & Infants’ Hospital of Rhode Island, Brown University Medical School) | To determine whether there is a risk of adverse maternal and fetal outcomes when surgery is delayed in pregnant women with ovarian mass |
- Close observation of antenatal surgery in pregnant women with ovarian mass is a reasonable alternative - Evaluated for mass size, presence/absence of septae, echogenicity, wall thickness, papillary projections, border characteristics, and vascularity |
Czekierdowski et al. [ |
36 | Retrospective multicenter study (two departments of obstetrics and gynecology, one department of obstetrics and pathological pregnancy, and one gynecological oncology center in Poland) | To evaluate the accuracy of preoperative diagnosis of ovarian mass in pregnant women in SA, the IOTA group SRR and ADNEX models | Subjective assessment is the best predictive method for complex adnexal masses found on prenatal ultrasonography in pregnant women. For inexperienced sonographers, the SRR and ADNEX scoring systems can also be used for the characterization of these tumors, but the serum tumor markers CA125 and HE4 and the ROMA algorithms appear less accurate |
Lee et al. [ |
236 | Retrospective multicenter study (eleven referral hospitals in South Korea) | To compare ultrasonographic ovarian mass scoring systems (IOTA, Sassone, and Lerner) and evaluate which factors can help predict the malignancy risk in pregnant women |
- Among the ultrasound characteristics, six factors showed statistically significant differences (maximal diameter of ovarian mass, maximal diameter of ovarian solid mass, inner wall structure, wall thickness, thickness of septation, and papillartity) - The combined model was developed with these six components - Malignant ovarian tumors can be predicted with high accuracy using either the Sassone scoring system or a combined model in pregnant |
Usui et al. [ |
69 | Retrospective study (single hospital in Japan) | To assess the impact on patients who underwent prenatal surgery and fetal outcomes |
- Ovarian mass may be associated with adverse fetal outcomes - Surgical interventions less than 24 weeks of gestation may not themselves have been associated with adverse outcomes |
Mascilini et al. [ |
34 | Retrospective observational study (four ultrasound centers in Italy) | To elucidate the ultrasound features that can discriminate between benign and malignant ovarian cysts with papillary projections but no other solid component in pregnant women |
- Ground-glass echogenicity and papillations with a smooth contour on ultrasound are most likely to be decidualized endometriomas - Cysts with anechoic or low-level echogenicity and papillations with an irregular contour suggest borderline malignancy |
SA, subjective assessment; IOTA, International Ovarian Tumor Analysis; SRR, simple rules risk; ADNEX, assessment of different NEoplasias in the adneXa; HE4, human epididymis protein 4; ROMA, Risk of Ovarian Malignancy Algorithm.
Factors included in each scoring system
Scoring system | Factor | Tumor marker |
---|---|---|
IOTA simple rules |
- Locule - Locule - Proportion of solid tissue (%) - Acoustic shadows (yes/no) - Irregularity - Number of papillary projections - Blood flow (color score) - Presence of ascites (yes/no) |
- |
IOTA LR |
- Personal history of ovarian cancer (yes=1, no=0) - Current hormonal therapy (yes =1, no=0) - Age of the patient (in years) - Maximum diameter of the lesion (in mm) - Presence of pain during the examination (yes=1, no=0) - Presence of ascites (yes=1, no=0) - Presence of blood flow within a solid papillary projection (yes=1, no=0) - Presence of a purely solid tumor (yes=1, no=0) - Maximal diameter of the solid component (expressed in millimeters, but with no increase >50 mm) - Irregular internal cyst walls (yes=1, no=0) - Presence of acoustic shadows (yes=1, no=0) - Color score (1, 2, 3, or 4) |
- |
IOTA ADNEX |
- Age of the patient at examination - In oncology center - Maximal diameter of the lesion (mm) - Proportion of solid tissue (%) - Number of papillary projections (0, 1, 2, 3, >3) - Presence of more than 10 cyst locules (yes/no) - Acoustic shadows (yes/no) - Presence of ascites (yes/no) |
CA125 |
Sassone |
- Inner wall structure - Wall thickness - Septa - Echogenicity |
- |
PMS |
- Sassone score - Type of vascularization (peripheral=1; central/septal=2) - Numeric value of the resistance index of the pelvic mass |
CA125 |
DePriest |
- Volume - Cystic wall structure - Septa structure |
- |
Lerner |
- Wall structure - Shadowing - Septa - Echogenicity |
- |
O-RADS |
- Major category (locule, solid elements) - Size - Solid or solid-appearing lesions (irregularity, acoustic shadowing) - Cystic lesions (papillarity, irregularity) |
- |
IOTA, International Ovarian Tumor Analysis; LR, logistic regression; ADNEX, assessment of different NEoplasias in the adneXa; PMS, pelvic mass score; O-RADS, Ovarian-Adnexal Reporting and Data System.