Abstract
Background & Objectives:
The ultrasound features of ovarian mature teratomas are well defined. However, literature regarding the ultrasound features of immature teratomas is lacking and features can often be confused with those of mature teratomas. The objective of this study was to evaluate the subjective assessment of ultrasound examiners of anonymised static B mode ultrasound images of immature and mature teratomas.
Method:
An electronic questionnaire was distributed to levels 1 to 3 ultrasound examiners containing 23 anonymised ultrasound images including nine different immature teratomas and nine different mature teratomas. The images were collected retrospectively, and the reference test was the final histology of the surgically removed masses. The sensitivity and specificity test of the subjective assessment for each histological group was calculated. The live (at the time the scan was performed) subjective impression of each tumour was also recorded and the level of the examiner that performed the scan was noted.
Results:
There were a total of 43 responses (30 from level 1, 9 from level 2 and 4 from level 3 examiners. The sensitivity and specificity of subjective assessment ranged from 16.3% to 97.7% and 81.6% to 96.1%, respectively, for all histological groups. The sensitivity and specificity to correctly diagnose mature teratomas and immature teratomas were 41.3% (95% CI: 36.3%–46.4%), 82.2% (95% CI: 78.9%–85.1%) and 41.2 % (95% CI: 36.2%–46.3%), 81.6 % (95% CI: 78.3%–84.7%), respectively. There was no statistical difference in the diagnostic performance of the subjective assessment of examiners between mature and immature teratomas (p = 0.975).
Conclusions:
Diagnostic performance of static 2D ultrasound images for mature and immature teratomas is low. Immature teratoma still pose a diagnostic challenge.
Introduction
Ovarian teratomas are the most common type of germ cell tumours and account for approximately 20% of adult ovarian tumours and 50% of ovarian masses in children.1–3
Teratomas comprise of several different types of tumours which are composed of mature or immature tissues and are derived from the three pluripotent tissue layers (mesoderm, ectoderm and endoderm). 1
Mature ovarian teratomas (dermoid cysts), account for more than 95% of all ovarian teratomas and approximately 70% of all germ cell tumours. 2 They are the most common benign ovarian mass in women less than 45 years old and generally grow slowly, at approximately 1.8 mm per year.1,4 Most mature ovarian teratomas can be diagnosed at ultrasound and have common appearances such as dots and lines referred to as dermoid mesh, fat fluid levels and a focal hypertuberance from the cyst wall (Rokitansky nodule).5,6 However, the ultrasound appearance of mature teratomas can vary. 6
Immature teratomas are malignant and account for less than 1% of ovarian teratomas. 6 They occur most commonly in the first two decades of life and are usually unilateral tumours.1,7 Pathologically, immature teratomas are usually larger (140–250 mm) than mature teratomas (mean diameter of 70 mm).8,9 However, the ultrasound characteristics of immature teratomas are not as clearly defined as mature teratomas in the literature. 6 Immature teratomas are more easily defined on CT and MRI. 6 The last type of teratoma are monodermal teratomas which are derived from a single embryonic layer. There are three main types of monodermal teratomas: neural tumours, struma ovarii and carcinoid tumours. 2
There is a lack of evidence defining the ultrasound features of immature teratomas. The differentiation of mature teratomas from immature teratomas on ultrasound can influence patient management and outcomes significantly.
The aim of this study was to assess the diagnostic performance of subjective assessment of levels 1–3 ultrasound examiners in interpreting anonymised static B mode ultrasound images of immature and mature teratomas. The US characteristics of the immature and mature teratomas diagnosed during this study period were also analysed.
The subjective assessment of levels 1 and 2 ultrasound examiners was also compared to the subjective assessment of level 3 ultrasound examiners and the diagnostic performance of correctly identifying the tumours as malignant and benign was also analysed.
Methods
The ultrasound images for this analysis were collected from a tertiary referral centre for gynae-oncology in London, United Kingdom.
Anonymised ultrasound images of nine immature teratomas, nine mature teratomas, one endometrioma, one endometrioid ovarian carcinoma, one serous borderline tumour, one granulosa cell tumour and one simple ovarian cyst were retrospectively identified from surgical histology from 2011 to 2019 and were included in an online electronic questionnaire. 10 Each individual tumour had one image included. This questionnaire was distributed to members of the British Medical Ultrasound Society (BMUS) and the London School of Obstetrics and Gynaecology South London Deanery. Examiners were asked to give their subjective assessment for each ultrasound image from the seven possible options. These options were immature teratoma, mature teratoma, endometrioid ovarian tumour, serous borderline tumour, granulosa cell tumour, simple ovarian cyst and endometrioma.
Levels 1 to 3 ultrasound examiners were included in the study and these levels were defined as per the European societies for ultrasound in medicine and biology (EFSUMB). 11
In short, level 1 examiners are those who recognise common anatomy and common abnormalities and are aware when a referral onto a level 2 or 3 examiner is required. 11 Level 2 examiners are those that manage referrals from level 1 examiners, teach and conduct some research. Level 3 examiners are those who perform specialist scans, teach and conduct significant research. 11
A retrospective review was conducted to identify all patients who had a histological diagnosis of the subtypes of interest from 2011 to 2019. Once identified, a retrospective review of the hospitals electronic ultrasound system was performed to identify 2D static images of the mass pre-operatively. The best and most appropriate images of each tumour were chosen. These images were assessed by AS, a level 3 examiner, for their suitability.
All the tumours were classified according to the WHO (World Health Organisation) classification of tumours for histological subtypes and the malignant tumours were classified according to the FIGO (The International Federation of Gynaecology and Obstetrics) criteria.12,13
Outcomes of the subjective assessment from the questionnaire were compared to the corresponding histological subtypes. A copy of the questionnaire with the images used in the study is shown in the appendix.
Diagnostic performances for identification of the images as malignant and benign were also calculated and the subjective assessment of levels 1 and 2 examiners were also compared to the subjective assessment of level 3 examiners.
The level of the examiner who performed the real time scan was documented. The subjective assessment of the examiner who performed the real time scan (live scan subjective assessment of the examiner scanning/live interpretation) with the ability to utilise Doppler was documented. This was compared to the interpretations from the questionnaire. The most common descriptions of the mature and immature teratomas from the live interpretation were also extracted.
Approval of the study as a ‘service improvement’ was granted by the audit department at Guy’s and St Thomas’ hospital and therefore formal ethics approval was not required.
Statistical analysis
Statistical analysis was performed using Stata MP v17.0 software (USA, 2023). 14 To calculate the diagnostic performances of the subjective assessment of the examiners, the performance measures sensitivity, specificity, positive and negative likelihood ratios were calculated.
The 95% confidence intervals were calculated for the sensitivities, specificities, and positive and negative predictive values.
Results
A total of 43 responses were received (30 of the 43 responses from level 1 examiners, 9 of the 43 responses from level 2 and 4 of the 43 responses from level 3 examiners).
Between 2011 and 2019, nine ultrasound examinations of immature teratomas were identified.
The mean age of the patients with immature teratomas and mature teratomas were 25 (range 14–53) and 33 (range 23–49), respectively, and eight of the nine cases were premenopausal in both the immature and mature teratoma groups.
The median largest diameter of the immature teratomas was 140 mm and the median largest diameter of the mature teratomas was 69 mm.
All the immature teratomas were confined to the ovaries and diagnosed at FIGO stage 1(A-C) and histopathological grading of the immature teratomas were from grades 1 to 3. Just over 50% (5/9) were grade 1, 33% (3/9) were grade 2 and 11% (1/9) was grade 3.
From the live interpretation, the mature teratomas were mainly described as thin-walled avascular cyst containing a solid hyperechoic component/hyperechoic mixed contents with no acoustic shadowing. The immature teratomas were described as large complex cystic masses with solid and cystic components with noted vascularity. In some cases, the cysts were reported to have nodular small solid/hyperechoic components with irregular septations.
Examiners were able to correctly identify immature teratomas using subjective assessment with a sensitivity of 41.2 % (95% CI: 36.2%–46.3%) and a specificity of 81.6 % (95% CI: 78.3%–84.7%). Mature teratomas were correctly identified with a sensitivity of 41.3% (95% CI: 36.3%–46.4%) and a specificity of 82.2% (95% CI: 78.9%–85.1%). There was no statistical difference in the diagnostic performance of the subjective impression of examiners between mature and immature teratomas (p = 0.975, χ2).
The sensitivities and specificities for the remaining histological types of tumours are shown in Table 1.
Diagnostic performance – Sensitivity, Specificity, Positive predictive value, Negative predictive value of all histological subtypes.
The simple cyst was most correctly identified with a sensitivity of 97.7% (95% CI: 87.7%–99.9%) and a specificity of 96.1% (95% CI: 94.6%–97.2%).
The endometrioid ovarian carcinoma was most misdiagnosed by subjective impression with a sensitivity of 16.3% (95% CI: 6.81%–30.7%) and a specificity of 93.6% (95 CI: 91.9%–95.1%).
The accuracy of subjective assessment for each individual image for immature and mature teratoma from the questionnaire are defined in Tables 2 and 3. The live subjective assessment of the immature and mature ovarian tumours and level of clinician performing the scan was recorded in Tables 2 and 3.
Images Immature teratomas.
Accuracy of individual images(n) as specific histologies and benign and malignant with the interpretation and level of clinician at the time of the live scan.
Images Mature teratomas.
Accuracy of individual images(n) as specific histologies and benign and malignant with the interpretation and level of clinician at the time of the live scan.
When analysing the live interpretations of the teratoma group alone into benign and malignant groups and assuming ‘unclassified’ as malignant, the sensitivity and specificity were 75% (95% CI: 19.4%–99.4%) and 100% (95% CI: 54.1%–100%), respectively, compared to 66.2% (95% CI: 61.3%–70.9%) and 63.4% (95% CI: 58.3%–68.2%) for the static interpretation of these masses (p > 0.1).
However, at live interpretation, the correct subjective impression was recorded in one of the nine immature teratomas and six of the nine mature teratomas (p = 0.02).
The live interpretation and level of clinician performing the scan for the remaining tumours is shown in Table 4.
Live interpretation and level of clinician performing the live scan.
There were specific ultrasound features which were easier to diagnose than others. In over 50% of the cases, the ultrasound examiners were able to correctly identify immature teratomas as such in questions 7, 15 and 16. In all these images, there were large irregular solid components, some with coarse calcifications all of which is usually associated with an immature teratoma. 15 The live interpretation of questions 7, 15 and 16 was malignant (unclassified), abnormal unclassified and immature teratoma, respectively. Questions 7 and 16 were scanned by level 3 clinicians and question 15 by a level 1 clinician. The images which appeared more challenging to diagnose in this study were those with a large anechoic cystic components (questions 6 and 19) and solid like projections arising from the cyst wall (question 10). These were also difficult to diagnose on live interpretation with all images unclassified.
The subjective impression of level 3 examiners for immature teratomas was compared to levels 1 and 2 examiners. Level 3 examiners correctly identified immature teratomas with a sensitivity 34.2% (95% CI: 19.6%–51.4%) compared to 42% (95% CI: 36.7%–47.3%) for levels 1 and 2 examiners. The remaining diagnostic accuracies are documented in Table 5.
Diagnostic performance of the subjective impression of level 3 examiners compared to level 1 and 2 examiners combined for immature teratomas – Sensitivity, Specificity, Positive predictive value and Negative predictive value.
Examiners were able to correctly identify immature teratomas as any malignant tumour using subjective assessment with a sensitivity of 62.1% (56%–67.9%) and specificity of 66.7% (62.2%–71%). The sensitivities and specificities for identifying the remaining malignant histological groups as malignant are shown in Table 6.
Sensitivity, Specificity, Positive predictive value, Negative predictive value for the identification of malignant histologies as malignant.
Mature teratomas were correctly identified as benign using subjective assessment with a sensitivity of 61.3% (55.1–67.2) and specificity of 68.6% (64.6%–72.7%).
The sensitivities and specificities for identifying the remaining benign histological groups as benign are outlined in Table 7.
Sensitivity, Specificity, Positive predictive value, Negative predictive value for the identification of benign histologies as benign.
Discussion
This study demonstrates the diagnostic challenge faced by all levels of ultrasound examiners when assessing mature and immature teratomas.
To our knowledge, this is the first cross-sectional study evaluating the subjective assessment of immature and mature teratomas by ultrasound examiners of varied experience.
Ascertaining the ultrasound features diagnostic of immature teratomas is difficult in part, due to their rarity, however ultrasound is often the first imaging modality used as it is inexpensive and easily accessible. 16
They are predominantly unilocular, large tumours, 6 which are fast growing.2,17 They are complex, heterogenous masses that contain fluid with low-level echogenicity 6 and solid components with multiple small cystic spaces.2,15,18 The mean age at diagnosis, median diameter of tumour and ultrasound features in this study were all comparable to other documented studies.6,15,18
The difference between the accuracy of the subjective assessment of ultrasound examiners between immature teratomas and mature teratomas was not statistically significant. This may suggest that these malignant tumours are not appropriately triaged and referred to gynae-oncology centres and this could influence the patient’s survival rates and morbidity. In our data set, immature teratomas were correctly identified as a malignant tumour with a sensitivity of 62.1% and specificity of 66.7%.
In this study, the only options on the questionnaire were a specific histological diagnosis and therefore a diagnosis had to be chosen. In practice, the tumours may be classified as uncertain prompting further investigation and imaging.
On live interpretation, some of the ultrasound images of immature teratomas appeared easier to diagnose than others. The variation in ultrasound appearances of immature teratomas within this study reflects the variation of immature teratomas in clinical practice.
With regards to the level of the examiner, the sensitivity for the subjective impression of level 3 examiners for immature teratomas was 34.2% compared to 42% for levels 1 and 2 examiners combined. Although small numbers (n = 4), this further highlights the difficulty of US interpretation of these tumours, even for the most experienced of examiners. It is also important to acknowledge the difficulty in assessing the US experience of examiners as there is overlap between the levels. 11
The detection of mature teratomas is reported to have high sensitivity (77%–86%) and specificity (94%–100%). 19 In another paper, the sensitivity and specificity of ultrasound was between 58% and 92.7% and 87.5% and 99%, respectively. 20 The results in this study are lower than that reported in the literature. This may be because some of the mature teratomas included in our study were atypical and did not contain the dot-dash sign, which has a positive predictive value (PPV) of 98% for mature teratomas or typical Rokitansky nodules. 21 This included images in questions 4, 11, 18, 22 and 24. For these images, fewer than half of the examiners were able to identify the mature teratomas as such. A large cystic component and an area of hyperechoic mixed content, not typical of a usual Rokitansky nodule were mainly seen in these images. This finding is consistent with the live interpretation of the cyst where two of the five cysts were classified as mature teratomas. The others were unclassified.
As a tertiary referral centre, there is a higher referral rate of patients with atypical mature teratomas. In this paper, a higher incidence of atypical mature teratomas may be included when compared to previous studies.
Diagnostic sensitivities on subjective assessment for specific adnexal masses has been published in previous IOTA studies (39.5% vs 77% for endometriomas, 61.9% vs 29% for borderline tumours, simple ovarian cyst 97.7% vs 100%).19,22
Endometriomas have varying ultrasound appearances which is clearly evident between pre- and post-menopausal endometriomas.16,23–25 Typically, they are described as unilocular cyst with ground glass content, however they can be multilocular solid tumours and contain cyst fluid with mixed echogenicity. 25 The cyst in this study was not typical and this could account for the differing sensitivities between the two studies. An ‘atypical’ endometrioma was chosen to reflect the complexity of endometriosis patients treated at our hospital.
That said, the endometrioma was correctly identified as benign with a sensitivity of 65.2% in this study, which closer resembles data from previous studies. Even with the presence of atypical features, subjective assessment would in most cases initiate the correct follow up plan and prevent over investigation.
Borderline tumours have two main histological subtypes, serous and mucinous. Serous borderline tumours are often small and anechoic with a significant number containing a solid part or a papillary projection. 26
In comparison, mucinous borderline tumours are often multiloculated and contain a solid part and papillary projection in 40%. Endocervical and intestinal mucinous borderline tumours also have differing ultrasound appearances. 22
Here, the borderline tumour included was serous which differs from previous studies where both serous and mucinous tumours were included. Given the more defining characteristic of serous borderline tumours, the difference between the studies could be attributed to differences in tumour population.
Endometrioid ovarian carcinoma is a rare diagnosis and sensitivity for this tumour was low (16.3%). Malignant endometrioid cyst have a higher prevalence of solid tissue with positive Doppler signals and a high number (>3) papillary projections within the cysts. 27 Although the image of the endometrioid carcinoma in this study was not a Doppler image, there were several papillary projections which are typical of these tumours. As they are rare, many level 1 examiners may not have seen such tumour. However, it was correctly identified as a malignant tumour with a sensitivity of 61.2%, thus demonstrating the presence of multiple papillary projections among other features would correctly ensure appropriate triage.
Granulosa are the most common sex cord stromal tumour. 28 The most common ultrasound appearance is a multilocular solid cyst that contains many small locules (>10) with no papillary projections.28–30 The tumour included in this study had characteristic ultrasound appearances.
There are a few limitations of this study to note. A weakness of this study is the lack of colour Doppler images used in this study. A characteristic of the solid components in mature teratomas compared to malignant is reduced vascularity. 30 Conversely, increased colour content on Doppler examination during ultrasound is associated with malignancy. 31 It is possible that the presence of colour Doppler may have increased the examiners sensitivity to the malignant tumours in this series and improved the subjective assessment. However, the study has enabled a discussion of the other salient features that differ between these tumours.
Ultrasound is a dynamic speciality and in live interpretation, is often used in conjunction with a patient’s clinical history and if available, biochemical results to help provide a diagnosis. We excluded these in this study as this was solely an ultrasound based study, however the use of static B mode images for this study has removed this dynamic assessment of ultrasound and thus may have influenced the examiners interpretations. In the future, the consideration of live ultrasound video clips could help with the examiners subjective assessment.
Given immature teratomas are rare tumours and the availability of ultrasound imaging of these tumours is not common, inferential statistics were conducted and included in the study. As this is a small sample size, this could be considered a limitation.
The small number of examiners, in particular level 3 examiners, is also important to note, although this could map many gynaecology units as patients with immature teratomas are likely to have the first imaging with levels 1 and 2 examiners. To further improve the generalisability of these results a larger, multicentre study would build on this current data.
Contribution
What are the novel findings of this work?
This work demonstrates that immature teratomas remain difficult to diagnose on ultrasound among all level of ultrasound examiners (levels 1–3). It has further confirmed that the ultrasound features of teratomas are varied.
What are the clinical implications of this work?
This work suggests that further education and training is required to improve the diagnostic accuracy of immature teratomas to reduce patient anxiety and to improve patients’ outcomes.
Conclusion
In conclusion, immature teratomas have varying ultrasound appearances, adding to this diagnostic challenge. Although a diagnosis of malignancy is often multifaceted, caution must be taken if there is uncertainty about the ultrasound diagnosis of immature teratomas or atypical mature teratomas on an individual scan and where possible, second opinions or further imaging must be sought.
To improve the diagnostic performance and morbidity for these patients, further awareness for all sonographers and sonologists on the specific ultrasound features of these tumours are required.
Supplemental Material
sj-pdf-1-ult-10.1177_1742271X251353712 – Supplemental material for Immature teratomas remain difficult to diagnose: A cross-sectional study
Supplemental material, sj-pdf-1-ult-10.1177_1742271X251353712 for Immature teratomas remain difficult to diagnose: A cross-sectional study by Sian Mitchell, Frances Bailey, Alison Smith, Jonathan Gaughran, Tom Holland, Mustafa Zelal Muallem and Ahmad Sayasneh in Ultrasound
Footnotes
Acknowledgements
Nil.
Contributors
Sian Mitchell and Ahmad Sayasneh contributed to the conception, design, acquisition of data, analysis, interpretation, drafting the article, revisions and final approval of the manuscript. Frances Bailey, Alison Smith, Jonathan Gaughran, Tom Holland and Mustafa Zelal Muallem contributed to drafting the article, revisions and final approval of the manuscript
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship and/or publication of this article.
Ethics Approval
Approval of the study as a service improvement was granted by the audit department at Guy’s and St Thomas’ NHS trust. Formal ethical approval was not required.
Permission from patient(s) or subject(s) obtained in writing for publishing their case report
NO, anonymised 2D US images used with no patient identifiable information or history.
Permission obtained in writing from patient or any person whose photo is included for publishing their photographs and images
NO – anonymised 2D US images with no patient identifiable information.
Confirm that you are aware that permission from a previous publisher for reproducing any previously published material will be required should your article be accepted for publication and that you will be responsible for obtaining that permission
NO.
Guarantor
None.
Supplemental Material
Supplemental material for this article is available online.
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
