Adnexal masses are a common problem facing obstetricians/gynecologists in practice and although most of these masses will prove to be benign, it is important to identify the ones more likely to be malignant while avoiding over-treatment of those likely to be benign. Masses with a high likelihood of being malignant require rapid further evaluation and proper surgical planning, either primary referral to a gynecologic oncologist, intra-operative surgical consultation as deemed necessary, or a discussion with the patient that a second surgery may be needed if the mass is found to be malignant [1].

The discrimination between benign and malignant adnexal masses has long been challenging and findings such as pelvic exam, tumor marker, and diagnostic imaging (i.e., computer tomography (CT) scan, ultrasound, MRI) taken alone have inadequate sensitivity and specificity [2, 3].

Efforts to improve on discrimination of individual tests led to the development of algorithms combining tumor markers, clinical factors, and ultrasound findings. One such algorithm is the Risk of Malignancy Index (RMI), initially described by Jacobs et al. [4] in 1990. This index uses the CA-125 level, menopausal status and ultrasound features to calculate a score above which the likelihood of malignancy is higher. A higher score is associated with menopausal status, higher CA-125 levels and ultrasound findings such as bilaterality, solid areas, multilocularity, and ascites. The RMI 1 of Jacobs et al. [4] with a cut-off of 200 showed a sensitivity of 85% and specificity of 97% in discriminating between malignant and benign lesions. The Risk of Malignancy Index 1 and subsequent modifications are simple, practical, and sensitive tools which have entered clinical practice after being studied and validated in larger patient cohorts [5, 6, 7, 8, 9].

Among imaging modalities, transvaginal ultrasound remains the method of choice for evaluating adnexal masses, based on its ease, availability, and patient tolerance. Other methods, such as CT scan and MRI, are not superior for the initial evaluation and are reserved for further evaluation in patients when the suspicion for malignancy is high to assess for lymph node enlargement, peritoneal disease, or possible distant metastases [3, 10, 11, 12].

Many research efforts have been made into using ultrasound morphologic criteria and scoring algorithms to differentiate between benign and malignant adnexal masses. In 2005, the International Ovarian Tumor Analysis (IOTA) group described two models (M1 and M2) based on a logistic regression analysis of clinical and ultrasound parameters collected in a standard fashion in a large patient cohort by experienced sonographers in a small number of centers using a common protocol. The larger model M1, using 12 parameters, achieved a sensitivity of 93% and a specificity of 76% in distinguishing between benign and malignant masses [13]. To obtain a simpler and clinically useful model, the IOTA group found in a later analysis that 10 of the ultrasound parameters had the highest predictive value and were easily and widely applicable. This set of 10 ultrasound findings—the Simple Rules—was published in 2008 [14]. Five were indicative of benign masses: unilocular; solid components with the largest being 7 mm; smooth; multilocular with largest diameter 10 cm; presence of acoustic shadows; no blood flow. The other five were indicative of malignancy: irregular solid tumor; presence of ascites; $>$4 papillary structures; irregular multilocular solid tumor with largest diameter $>$10 cm; strong blood flow signal.

Several studies have validated the Simple Rules prospectively [15, 16] and a 2014 systematic review and meta-analysis found the IOTA Simple Rules algorithm to be the best risk assessment algorithm, with a sensitivity of 0.93 [95% confidence interval (CI), 0.89–0.95] and a specificity of 0.81 [95% CI, 0.76–0.85] based on pooled data [17]. One drawback of the Simple Rules is that approximately 25% of adnexal masses cannot be classified, either because none of the rules apply or because rules from both benign and malignant categories apply. Despite this drawback, the Simple Rules remain one of best simple tools to assess 75% of adnexal masses for the likelihood of being benign or malignant [18].

Among adnexal masses judged to be benign according to ultrasound appearance, using different models including the IOTA Simple Rules, the risk of malignancy is very low (less than 1–8%) [19, 20, 21]. The rate at which a benign-appearing adnexal mass grows has not been described as a risk factor for malignancy and this observation is the basis of this report.

All patients in our obstetrics and gynecology group practice found to have a presumably benign adnexal mass requiring surgery between 2015 and 2020 were reviewed for the evaluation. Patients with presumably malignant adnexal masses were referred directly for surgery in gynecologic oncology centers and were not considered for the analysis. Forty-eight masses were classified as benign according to ultrasound findings using IOTA Simple Rules and formed the study group. Benign masses could have only benign criteria and none of the malignant. Benign criteria are: unilocular masses; masses with solid components the largest being 7 mm; smooth, multilocular masses with a maximum diameter 10 cm; presence of acoustic shadows and no blood flow. Approximately 10 masses, presumably benign, could not be classified according to the Simple Rules and were not included in the analysis.

Benign-appearing masses were initially followed by ultrasound, if asymptomatic and small (3 cm). The indications for surgery included patient symptoms (pain, pressure, abnormal bleeding), persistence or changes on ultrasound (i.e., growth or morphologic features).

Ultrasound exams were performed by the three practice partners (gynecologists each having over 15 years of sonography training and experience) following a generally standard procedure. The exam was begun transvaginally with a General Electric (GE) Voluson, 730 Pro or Voluson E6 6.5 MHz transvaginal probe and if the mass was too large to be assessed transvaginally, a transabdominal approach was performed. Color Doppler was performed at low velocity (5–10 cm/s), low pulse repetition frequency and high sensitivity settings.

A total of forty-eight consecutive cases meeting the criteria were found and included in the evaluation. Because of its retrospective chart review nature of data which is anonymous and personally unidentifiable, IRB and ethics approval was not required.

Ultrasound findings of the adnexa up to 3 years prior to surgery and CA-125 levels when available were reviewed. The ultrasound findings (adnexal mass morphology) were assessed by the authors and described according to standard terminology and definitions as described by Timmerman et al. [22] in 2000.

A Risk of Malignancy Index was calculated using the following formula: RMI 2 (Risk of Malignancy Index) = M $\times$ U $\times$ CA-125 according to Tingulstad et al. [5], where M (menopausal score) is 1 point for premenopausal patients or 4 points for postmenopausal patients. Menopause is defined as $>$1 yr amenorrhea or age $>$50 if status post hysterectomy. The ultrasound score (U) is based on the ultrasound morphologic criteria: bilaterality, solid areas, multilocularity, ascites and metastases. For 0–1 criteria, U = 1, for $>$2 criteria, U = 4. CA-125 levels are measured in U/mL. CA-125 levels were not performed in every patient, but rather at the discretion of the treating physicians.

A rapid rate of growth was defined when a mass increased in size by more than 50% in 6–12 months or when a mass felt to indicate surgery first appeared 6–12 months prior to surgery.

Between 2015 and 2020, forty-eight patients in our obstetrics and gynecology group practice (consecutive cases) were operated for adnexal masses benign in appearance on ultrasound according to IOTA Simple Rules. Patient characteristics, types of surgery performed, and histology results are shown in Table 1 (Ref. [5, 14]). CA-125 levels, when performed, and the Risk of Malignancy Index 2 [5] are also reported in Table 1. Small, asymptomatic, benign-appearing masses were followed initially with serial ultrasound. Once patients became symptomatic or masses grew larger than 3 to 4 cm, the indication for surgery was made. The histology confirmed benign pathology such as endometriosis, benign teratoma, cystadenoma etc. in most cases.

Patients ranged in age from 22 to 84; 27 were pre-menopausal and 21 were postmenopausal. All adnexal masses appeared benign on ultrasound using IOTA Simple Rules[14]. Masses were generally measured in three dimensions. The largest diameter in centimeters is reported and used in the analysis. CA-125 levels ranged from 4 to 127 U/mL. The Risk of Malignancy Index 2 ranged from 7 to 127 and was never above 200 (indicative of malignancy).

Ultrasound findings up to three years prior to surgery were available for 43 patients. These findings (at 0–6 months, 6–12 months, 12–18months, 18–24 months and 24–36 months prior to surgery) are also reported in Table 1. In 39 of these patients, the adnexal mass was present and growing slowly (increased in size 50%) over several months prior to surgery. The surgical histology in these patients was benign.

In the other 4 patients, the adnexal mass grew rapidly—either increasing in size $>$50% over 6–12 months prior to surgery or first appearing 6–12 months prior to surgery. Two of these patients were premenopausal (aged 40–50), the other two were postmenopausal (aged 65–75). In three of the four patients, the benign-appearing mass measured 6 to 9 cm in largest diameter at presentation and had not been present on an ultrasound 6 to 12 months prior. In the other patient, the mass measured 3cm in largest diameter at initial presentation and had grown to 4.5 cm 3–6 months later. Surgery was performed either due to pain or size greater than 3–4 cm. In these four patients, all were found to have borderline or early stage ovarian or tubal carcinomas. The rate of malignancy was 9.3%.

The management of adnexal masses is dependent on the symptoms they produce as well as on their likely nature (benign versus suspicious for malignancy). Much research has been dedicated to determining which clinical factors, imaging modalities and laboratory markers can best distinguish the nature of adnexal masses. Among imaging modalities, transvaginal ultrasound remains the best first-line method due to its convenience, ease of use and ability to distinguish likelihood of malignancy based on certain morphologic criteria. CT scan and MRI have been shown to have higher specificity but lower sensitivity than ultrasound at predicting malignancy [3, 11] and are better suited for evaluating extent of disease if suspicion is high for malignancy.

Diagnostic algorithms which have been found to have a high sensitivity and specificity for differentiating benign from malignant adnexal masses include the Risk of Malignancy Index (RMI). First described in 1990 by Jacobs et al. [4], the RMI 1 is calculated using the product of the serum CA-125 level (U/mL), the ultrasound scan result (expressed as a score of 0, 1 or 3 depending on ultrasound findings) and the menopausal status score (1 if premenopausal and 3 if postmenopausal). Ultrasound findings included bilaterality, solid areas, multilocularity, ascites and metastases. If no abnormality was found, the score was 0. For one abnormality, the score was 1 and for 2 or more abnormalities, the score was 3. Menopause was defined as one year of amenorrhea or the age of 50 for women who had had a hysterectomy. The RMI 1 with a cut-off of 200 showed a sensitivity of 85% and specificity of 97% in discriminating between malignant and benign lesions.

Several modifications of the Risk of Malignancy Index have been developed to improve on sensitivity and specificity. Tingulstad et al. [5] described the RMI 2 in 1996, which is calculated similarly to the RMI 1, but differing in weight of the ultrasound score (1 if $\le$1 ultrasound feature and 4 if $\ge$2 ultrasound features) and the menopausal status (1 if premenopausal and 4 if postmenopausal). The RMI 3 was another modification by Tingulstad described in 1999 [6] and the RMI 4 was developed by Yamamoto in 2009 [7]. Several prospective studies and systematic reviews comparing the four indices have since been published. In 2009, Geomini et al. [8] found the RMI 1 and 2, both with a cut-off of 200, to be the best predictors with an estimated sensitivity of 78% [95% confidence interval (CI), 0.71–0.85] and a specificity of 87% [95% CI, 0.83–0.91] for malignancy. In 2014, Yamamoto et al. [9] found RMI 2 to have the highest sensitivity for malignancy, and RMI 2 and 4 to have similar ability to distinguish between benign and malignant adnexal masses. Because of these considerations, we chose to use the RMI 2 in our analysis. None of the patients in our analysis had a RMI 2 above 200.

Another algorithm based on ultrasound criteria alone is called the IOTA Simple Rules and was developed by the IOTA group in 2008. Five were indicative of benign masses: unilocular; solid components less than 7 millimeters; smooth, multilocular less than 10 centimeters; presence of acoustic shadows; no blood flow. The other five were indicative of malignancy: irregular solid tumor; presence of ascites; more than 4 papillary structures; irregular, solid tumor greater than 10 centimeters; strong blood flow signal. The IOTA Simple Rules have high sensitivity (0.93) and specificity (0.81) as found by a systematic review and meta-analysis in 2014 [8]. Furthermore, they are easy to apply and approximately 75% of adnexal masses can be classified with a high likelihood either as benign or malignant. For a quantification of risk of malignancy, the IOTA group published new data in 2016 showing that the Simple Rules, if applicable, individually or in certain combinations, can accurately estimate an individual patient’s risk of malignancy [19]. We chose to use the IOTA Simple Rules to assess the ultrasound morphology of the adnexal masses and only those masses which could be classified by the Simple Rules and which were judged to be benign were included.

Adnexal masses predicted to be benign according to these diagnostic algorithms have an extremely low likelihood of being malignant and can in general be followed by serial ultrasound if patients are asymptomatic. The unexpected finding of malignancy after surgery for an anticipated benign adnexal mass led to this evaluation and review of the literature. In two large series of unilocular adnexal masses followed with ultrasound, the rate of malignancy was less than 1% and factors associated with malignancy were menopausal status, history of breast and ovarian cancer, as well as the development of abnormal morphologic features on ultrasound [19, 20]. Furthermore, the authors noted that intra-cystic solid structures were indeed found by pathology after removal and may have been missed on ultrasound, making a point for careful ultrasound evaluation prior to surgery looking specifically for intra-cystic solid/papillary structures.

A factor which we noted to be associated with malignancy in our consecutive series of patients operated for benign appearing adnexal masses was the rate of growth. Patients in whom the adnexal mass had grown by over 50% in 6–12 months prior to surgery or in whom the adnexal mass first appeared 6–12 months prior to surgery had a higher likelihood of malignant histology as compared to those in whom the adnexal mass had been growing at a slower pace. The rate of malignancy was 9.3%. This observation is biologically plausible since most ovarian cancers grow rapidly and present in advanced stages. Despite lack of elevated CA-125 levels or Risk of Malignancy Index and despite benign ultrasound features, rapid growth was associated with a high likelihood of malignancy.

Our study has several limitations: its small size, it’s retrospective nature, the lack of CA-125 values and ultrasound findings in several patients, the lack of a formal statistical analysis. It is not meant to change current practice, but rather to report an interesting and clinically relevant finding. These findings need to be validated in a larger group of patients evaluated and treated in a standard fashion before a practice-changing conclusion can be drawn.

In our series of forty-eight consecutive patients taken to surgery for adnexal masses judged to be benign according to ultrasound criteria (IOTA Simple Rules) and low Risk of Malignancy Index, those adnexal masses with a rapid rate of growth (increases in size greater than 50% in 6–12 months prior to surgery) were associated with a high risk of malignancy (9.3%). This finding warrants further study and may be useful in the planning of surgery for adnexal masses.

CM-T contributed to patient management, to data collection and analysis, and to writing of the manuscript. NV contributed to patient management. NG provided conceptual guidance and review of the manuscript. All authors read and approved the final manuscript.

Because this is a retrospective observational analysis of anonymous, personally unidentifiable clinical data, IRB and ethics approval was not sought at the outset of performing this study.

We would like to extend a special thank you to NG for the steadfast collaboration and friendship over so many years, and to our patients for entrusting us with their care.

This research received no external funding.

The authors declare no conflict of interest. NG is the Editor of this journal, given his role as Editor, had no involvement in the peer-review of this article and has no access to information regarding its peer-review.