A retrospective case series at a tertiary hospital in china: ultrasonographic features of Struma ovarii.

Table  1 delineates the clinical characteristics of 103 patients. Among the 103 patients diagnosed with SO, 95 were benign SO (median age: 44.9 years; age range,15–83 years) and 8 were malignant struma ovarii (MSO) (median age: 49.6 years; age range: 42–58 years). MSO cases accounted for 7.8% of the cohort. No statistically significant difference was observed in the median age between patients with benign SO and those with MSO ( p  = 0.346). Among the SO patients, 53 cases (51.5%) had a tumor in the right ovary, and 47 cases (45.6%) had a tumor in the left ovary, 2 cases (1.9%) had bilateral SO, and one patient was uncertain which side of the source was derived. Patients with SO often have no specific clinical manifestations and often have a pelvic mass found on gynecologic ultrasonography. Among patients with SO, 41 cases were found on physical examination, 10 cases were found to have abdominal pain or bloating, 7 cases were due to self-palpation of the mass, 6 cases were found to have other gynecological diseases (2 cases of uterine fibroids, 4 cases of cesarean section), and 1 case was found to have vaginal bleeding. Table 1 Clinical manifestations of 103 patients with SO Variable Benign ( n  = 95) Malignant ( n  = 8) P value Age at diagnosis, years 44.9(15–83) 49.6 (42–58) 0.346 Location Left 43(45.7%) 4(50.0%) >0.05 Right 49(52.1%) 4(50.0%) Bilateral 2(2.1%) 0(0%) Presenting symptom No symptom/incidental finding 45 3 < 0.001 Abdominal/pelvic pain 10 0 Abnormal vaginal bleeding 1 0 Others/unclear 33 5 Discovered due to other medical conditions 6 0 Data are expressed as median (range) or numbers (%) Clinical manifestations of 103 patients with SO Data are expressed as median (range) or numbers (%) Most of the patients routinely undergo the following serum tumor marker test before surgery: CEA, AFP, CA125, CA153, CA199, HE4, and SCC. Table  2 presents the tumor marker outcomes for the SO patients. Among them, 43 patients (52.4%) had elevated CA125 levels, while 5/79 (6.3%) had elevated CEA, 3/75 (4%) had elevated AFP, and HE4 and CA199 levels were within the normal range. Table 2 Tumor marker outcomes of the SO patients Variable Benign ( n  = 95) Malignant ( n  = 8) CEA elevation (4/76, 5.3%) (1/3, 33.3%) AFP elevation (3/72, 4.2%) (0/3, 0%) CA125 elevation (32/78, 41.0%) (3/4, 75.0%) CA153 elevation (3/29, 10.3%) (0/1, 0%) CA199 elevation (2/75, 2.7%) (0/4, 0%) HE4 elevation (0/69, 0%) (0/4, 0%) SCC elevation (2/31, 6.4%) (0/1, 0%) Values are expressed as numbers (%) Tumor marker outcomes of the SO patients Values are expressed as numbers (%) All 98 patients underwent ultrasonography before surgery. The possibility of SO was suggested for only 2 patients via ultrasonography, and the diagnostic results of preoperative ultrasonography are detailed in Table  3 . The ultrasound manifestations were mostly cystic masses or cystic predominantly cystic-solid masses (78/98), and 18 cases were solid masses or solid predominantly cystic-solid masses. Most of the masses are completely separated and multilocular, with smooth outer margins. Nearly one-half of the masses have punctate or short-line echoes with comet tails, papillary protrusions can be seen on the inner wall of some lesions, and there is usually no blood flow inside the tumor or a few dotted blood flow signals (79/96). Eighteen cases were recorded by pulsed spectral Doppler imaging, including 7 cases with a resistance index (RI) of < 0.5, 9 cases with 0.5 < RI  0.7. The margins of the benign SO were clear, while the interior of the multilocular-solid mass was more irregular, with multiple short, grainy strong echoes visible on the wall (shown in Fig.  1 A). The solid component was usually slightly hyperechoic, resembling the echo of thyroid tissue (shown in Fig.  1 B). Solid components are often located on the side of the tumor. When accompanied by colloidal components, it can present as a strong spot-like echo with a “comet tail” sign behind (shown in Fig.  1 C). Among the 6 patients with MSO which have ultrasound imaging, 4 patients showed a multilocular cystic mass on ultrasound, with poor sound echo in the cystic part, and 2 patients showed a solid mass on ultrasound with abundant blood flow signals (shown in Fig.  1 D-F). This series of cases is classified into O-RADS categories ranging from 2 to 5, comprising 10 cases in O-RADS 2, 43 cases in O-RADS 3, 30 cases in O-RADS 4, and 14 cases in O-RADS 5. Table  4 summarized the distribution of O-RADS categories and their corresponding ultrasound features in our cohort. While benign SO predominantly fell into O-RADS 2–3 categories (51/92, 55.4%), malignant cases were more frequently classified as O-RADS 4–5 (3/6, 50%), highlighting the challenges in distinguishing benign from malignant lesions based solely on imaging. Table 3 Preoperative ultrasound diagnostic findings Ultrasound diagnostic findings Benign ( n  = 92) Malignant ( n  = 6) SO 2 (2.2%) 0 (0%) Mature cystic teratoma 14 (15.2%) 1 (16.7%) Endometriosis cyst 2 (2.2%) 0 (0%) Ovarian cysts 3 (3.3%) 0 (0%) Cystadenoma 24 (26.1%) 2 (33.3%) Hydrosalpinx 2 (2.2%) 0 (0%) Borderline or malignant tumor 17 (18.5%) 2 (33.3%) No obvious diagnosis 28 (30.4%) 1 (16.7%) Values are expressed as numbers (%).Values are expressed as numbers Preoperative ultrasound diagnostic findings Values are expressed as numbers (%).Values are expressed as numbers Fig. 1 A-C: Benign struma ovarii. A: Yellow arrow indicates short, grainy hyperechoic foci. B: Yellow arrow denotes solid component with thyroid tissue-like echogenicity. C: Yellow arrow marks colloid components showing hyperechoic spots with posterior “comet-tail” artifact. D-F: Malignant struma ovarii. D: Predominantly cystic mass with heterogeneous echogenicity, showing poor acoustic transmission in most cystic areas (“snowstorm” appearance). E and F: Solid-predominant cystic-solid mass A-C: Benign struma ovarii. A: Yellow arrow indicates short, grainy hyperechoic foci. B: Yellow arrow denotes solid component with thyroid tissue-like echogenicity. C: Yellow arrow marks colloid components showing hyperechoic spots with posterior “comet-tail” artifact. D-F: Malignant struma ovarii. D: Predominantly cystic mass with heterogeneous echogenicity, showing poor acoustic transmission in most cystic areas (“snowstorm” appearance). E and F: Solid-predominant cystic-solid mass Table 4 O-RADS classification and ultrasound features of SO O-RADS Malignancy risk Ultrasound features Benign SO ( n  = 92) MSO ( n  = 6) 2 < 1% Unilocular cyst, smooth walls, no solid, compomemts or vascularity 10 (10.9%) 0 (0%) 3 1–10% Multilocular cystic mass with thin septations, hyperechoic foci (“pearl signs”), or comet-tail artifacts 41 (44.6%) 2 (33.3%) 4 10–50% Solid-cystic mass with irregular walls, papillary projections (≤ 3), moderate vascularity 28 (30.4%) 2 (33.3%) 5 ≥ 50% Predominantly solid mass with irregular margins, rich vascularity, or necrotic areas 13 (14.1%) 1 (16.7% In the MSO group, one patient showed no detectable lesion on ultrasound examination but was incidentally discovered during unrelated gynecological surgery O-RADS classification and ultrasound features of SO In the MSO group, one patient showed no detectable lesion on ultrasound examination but was incidentally discovered during unrelated gynecological surgery A total of 103 patients underwent surgery and were diagnosed via routine pathology after surgery. The average diameter of the tumor during surgery was approximately 9.45 cm (1.5–40 cm). Ninety-seven patients underwent laparoscopic surgery, including 36 cases of ovarian lesion resection, 21 cases of unilateral salpingectomy, 15 cases of bilateral salpingectomy, and 25 cases of total hysterectomy + unilateral or bilateral salpingectomy. Three patients underwent laparotomy, and three patients were identified via cesarean section. The pathological results of 103 SO patients revealed that 63 patients had SO, 6 had cystic changes, 5 had ipsilateral/isolateral cysts, 6 had adenomatous hyperplasia, 1 had calcification, 9 had ipsilateral/isolateral mature teratomas, 2 had ipsilateral/isolateral adenofibromas, 1 had thecoma of the ovary, 4 had ipsilateral/isolateral serous/mucinous cystadenomas, 3 had ipsilateral/isolateral borderline tumors, 1 had bilateral low-grade serous adenocarcinoma, 1 had SO papillary carcinoma, and 3 had SO carcinoma. Immunohistochemistry results showed positive expression rates for thyroid transcription factor-1 (TTF-1) (96%, 24/25), thyroglobulin (TG) (90.5%, 19/21), PAX8 (94.7%, 18/19), and Ki-67 (100% 16/16), though Ki-67 expression levels were low, with 15 cases < 5% and 1 case at 50%.

This study was conducted in accordance with the Declaration of Helsinki and approved by the Ethics Committee of the Affiliated Hospital of Qingdao University (QYFYWZLL29486). Due to the nature of the study, the need for informed consent was waived. We retrospectively reviewed the sonographic and histopathological features of 103 patients with histopathologically confirmed SO treated at the Affiliated Hospital of Qingdao University from May 2013 to November 2023. The exclusion criteria for patients were as follows: had undergone an adnexal mass biopsy with an incomplete mass or who had an unclear sonogram that interfered with the diagnosis. Transvaginal ultrasound was performed using a GE Voluson E8/E10 or Logiq E9 color Doppler ultrasound scanner with an intraluminal probe at a frequency of 5–9 MHz. The transvaginal modality is preferred, and abdominal ultrasonography is performed if the patient is not sexually active. Conventional ultrasonography divides lesions into cystic, cystic-predominant, cystic-predominant, and solid lesions according to their internal echogenicity. In addition to recording the location of lesions, the O-RADS standardized dictionary focused on observing and recording the type of lesions, maximum diameters, internal margins or walls, external contours, separations, solid components, number of locules, acoustic shadowing, number of papillary projections, and blood flow scores. According to the O-RADS US Risk Stratification and Management System, adnexal masses were divided into categories 0–5, among which O-RADS category 2 represented the identified benign lesions (malignant risk < 1%); and O-RADS classes 3–5 represented low-grade, moderate, and high-grade malignancies, respectively (malignant risk 1–10%, 10–50%, and ≥ 50%, respectively). The blood flow score follows the criteria proposed by the International Ovarian Tumor Analysis (IOTA) team and is divided into 1 to 4 points: no blood flow, micro blood flow, moderate blood flow, and rich blood flow. The 103 SO cases were classified according to the O-RADS risk stratification by 2 attending physicians with more than 5 years of gynecological ultrasound experience without knowledge of the pathology. If there was a disagreement between the two sonographers, all the details were discussed with the help of a senior sonographer until a consensus was reached.

In summary, the diagnosis of SO presents significant challenges due to its rarity occurrence and diverse ultrasound manifestations. SO may be suspected when ultrasonography reveals multilocular cystic or cystic predominantly cystic-solid masses, septations of varying thickness, nodular solid protrusions, hyperechoic foci resembling thyroid tissue, “pearl signs” and punctate hyperechoic areas with distinctive “comet tail” artifacts. Notably, our findings demonstrate considerable overlap in imaging characteristics between benign SO and MSO, making preoperative differentiation particularly difficult. This diagnostic ambiguity underscores the need for comprehensive evaluation incorporating clinical, imaging, and histopathological assessments to ensure accurate diagnosis and appropriate management of this rare ovarian tumor. The current study highlights both the diagnostic value and limitations of ultrasound in SO evaluation. It emphasizes the importance of maintaining a high index of suspicion when encountering these characteristic yet non-pathognomonic sonographic features. Future research should focus on establishing more definitive imaging criteria to improve preoperative diagnostic accuracy and guide clinical decision-making.

SO, a specialized monodermal teratoma, exhibits distinct pathological features that differentiate it from conventional mature teratomas. Unlike typical teratomas comprising tissues from all three germ layers, SO is predominantly composed of thyroid tissue derived from the endoderm (> 50%) [ 4 ]. This study systematically analyzed 103 pathologically confirmed SO cases, integrating the latest evidence-based findings to comprehensively explore the clinical-pathological characteristics, diagnostic challenges, and therapeutic strategies, thereby providing critical insights for the standardized management of this rare disease. The median age of the cohort was 45.3 years (range 15 to 83), with a malignant transformation rate of 7.8%, consistent with the previously reported range of 0.3–10% [ 2 , 3 , 10 ]. Notably, while the mean age of the malignant group (49.6 years) was higher than that of the benign group (44.9 years), the difference was not statistically significant ( p  = 0.346). This finding aligns with the large-scale study by Rybak et al. [ 11 ], suggesting that age may not be an independent predictor of malignancy. Clinical manifestations of SO exhibit marked heterogeneity. While most cases are incidentally detected on imaging, patients may also present with symptoms related to pelvic masses, including lower abdominal pain, palpable masses, or abnormal vaginal bleeding [ 1 , 12 – 15 ]. In our study cohort, 11 patients (10.7%) manifested such pelvic mass-related symptoms. Notably, approximately 5% of cases demonstrate hormonal activity (e.g., hyperthyroidism) [ 13 , 16 ], underscoring the importance of thyroid function evaluation. A tumor size exceeding 12.5 cm with predominant thyroid tissue composition has been associated with malignant potential [ 13 , 16 ]-a finding corroborated by our data showing larger mean diameters in MSO cases (14.2 cm) than in benign ones (8.9 cm). Recent work by Sengul and Sengul in the field of thyroidology has highlighted both the rarity and diagnostic challenges of SO, emphasizing the need for standardized diagnostic and management protocols given its uncommon yet clinically significant nature [ 17 ]. Unlike polycystic ovary syndrome (PCOS), which manifests as systemic metabolic dysfunction, the hormonal activity of SO is localized (such as hyperthyroidism) and mostly subclinical, thus requiring targeted thyroid function tests [ 18 ]. In this study, elevated CA125 levels were observed in 43 patients (52.44%), likely reflecting mechanical irritation or inflammatory responses rather than malignant transformation [ 19 ]. This nonspecific finding parallels the overlapping ultrasonographic features between benign SO and MSO: while benign cases predominantly presented as multilocular cystic masses with “comet tail” signs, MSO could also manifest as either cystic or hypervascular solid masses (Fig.  1 D-F). This diagnostic ambiguity underscores the need for pathological confirmation, particularly given the high positive rates of thyroid-specific markers (TTF-1: 96%; TG: 90.5%) [ 20 , 21 ]. Our findings corroborate the report by Christiana Oikonomou et al. that SO represents the sole ovarian tumor capable of producing TG [ 21 ], further validating its diagnostic significance. Despite these diagnostic challenges, certain characteristic sonographic features may suggest SO. The most common benign presentation manifests as a multilocular cystic mass with the following features: lobulated margins, septations of variable thickness, hyperechoic solid components resembling thyroid tissue, “struma pearls,” and punctate hyperechoic foci with “comet tail” artifacts (Fig.  1 A-C). However, the following features should raise suspicion for malignancy: infiltrative borders of solid components, markedly increased vascularity (Fig.  1 D-F), thick (> 3 mm) and irregular septations with nodularity, and rapid interval growth on follow-up imaging. Notably, neither “struma pearls” nor “comet tail” artifacts can reliably exclude malignancy [ 4 , 19 ]. The diagnostic difficulty stems from the pathological spectrum: benign SO demonstrates thyroid tissue hyperplasia, while MSO exhibit architectural and nuclear atypia resembling thyroid carcinoma [ 4 ]. These diagnostic limitations highlight the critical importance of comprehensive evaluation. As demonstrated in Table  4 , the overlapping imaging features in O-RADS classifications between benign SO and MSO not only highlight the necessity of incorporating pathological and immunohistochemical findings for definitive diagnosis, but also emphasize the critical importance of comprehensive evaluation integrating multiple factors including tumor markers (albeit with limited value), clinical manifestations, and ultimately pathological assessment. Recent studies from the Serbian Oncology Center (including preliminary survival outcomes) recommend individualized surgical approaches based on patient age, fertility requirements, and malignancy risk [ 22 ]. For benign SO, fertility-sparing surgery is generally sufficient, while MSO may require more aggressive interventions (e.g., total hysterectomy with bilateral salpingo-oophorectomy), particularly in postmenopausal women [ 21 , 23 ]. It is particularly noteworthy that the “3R” theory (Rare, Risky, Redebated) recently proposed by Sengul et al. [ 17 ] aptly encapsulates the clinical characteristics of SO. Our study findings provide empirical support for this theoretical framework while simultaneously highlighting several critical unresolved issues: (1) the specificity of sonographic features requires validation in larger cohorts, (2) reliable predictors of malignant transformation remain undefined, and (3) standardized long-term follow-up protocols are currently lacking. Through this large-scale analysis, our study systematically elucidates the clinicopathological characteristics of struma ovarii (SO), with particular emphasis on both the diagnostic value and limitations of ultrasonography. The findings not only validate the “3R” theory (Rare, Risky, Redebated) proposed by Sengul et al. [ 17 ], but also provide substantial evidence to inform clinical practice. Future multicenter collaborative studies are warranted to establish more precise diagnostic criteria and standardized treatment protocols, thereby improving the overall management of this rare disease entity.

Struma ovarii (SO) is a rare, highly specific monodermal teratoma that accounts for 1% of all ovarian tumors [ 1 ]. Despite their commonly benign behavior, these lesions also have malignant potential similar to other teratomas, accounting for approximately 5–10% of all cases [ 2 , 3 ]. Similar to that in the thyroid gland, the most common malignancy associated with SO is papillary thyroid carcinoma (PTC) [ 4 ]. SO usually occurs in women of childbearing age and is uncommon in children and adolescents. The clinical manifestations of SO are mostly nonspecific and are often found by physical examination, similar to those of other types of ovarian tumors, and are mostly nonspecific pelvic mass symptoms. No specific tumor markers associated with SO have been reported. Previously reported magnetic resonance imaging (MRI) results revealed lobulated multicystic lesions, a glass-stained appearance, and an enhanced solid composition [ 5 , 6 ]. The main images on CT scans include solid, cystic, or cystic-solid images, which are common unilaterally, with hyperdensity and calcification, and are markedly enhanced on contrast-enhanced scans [ 7 , 8 ]. Preoperative diagnosis remains challenging due to heterogeneous imaging findings and often requires postoperative pathological confirmation. It is often overtreated because it is difficult to distinguish it from malignancy. Owing to the relative rarity of SO and the fact that most previous studies have been limited to case reports, the accuracy of preoperative diagnosis and surgical decision - making is not well understood. In addition, accurate preoperative diagnosis is essential for selecting the best surgical decision. Ultrasound is the first choice for evaluating ovarian masses by imaging tests [ 9 ]. The Ovarian-Adnexal Reporting and Data System (O-RADS), proposed by the United States College of Radiology, involves the risk stratification and management of ovarian and adnexal masses on the basis of ultrasound characteristics to more accurately determine the malignant risk of ovarian and adnexal masses. Therefore, we retrospectively analyzed the ultrasound manifestations, and clinical and pathological characteristics of 103 SO patients at the Affiliated Hospital of Qingdao University over 10 years, and studied the O-RADS classification to better understand of the disease and improve the accuracy of the preoperative diagnosis.