Ovarian teratomas constitute the most frequent nonepithelial benign ovarian tumor, accounting for 95% of all germ-cell ovarian tumors.1 Prevalence reported in large series of surgically removed benign ovarian masses ranged from 14.3% to 28.6%.2–4 Most benign teratomas occur in premenopausal women (mean age 32 years old) but they can be found in premenarcheal girls and elderly postmenopausal women.1
Approximately 30% of women diagnosed with a benign ovarian teratoma present with symptoms with abdominal or pelvic pain being the most frequent.5 In the majority of women, these are incidentally discovered by physical examination, imaging, or surgical evaluations done for unrelated indications. Transvaginal ultrasonography is an excellent tool for characterizing ovarian tumors6: for benign teratoma, the reported sensitivity and specificity are as high as 86% and 99%, respectively.3,4
Traditionally, the management of diagnosed teratomas has been surgical removal.7 Expectant management of these tumors in asymptomatic women is gaining acceptance8 because the risk of complications such as torsion or cyst rupture has been shown to be low.9,10 We performed a systematic search of papers published in English language using the terms “Ovary,” “Dermoid,” “Teratoma,” and “management” in several databases (PubMed, Web of Science, and Google Scholar) from 1985 to 2016 and we observed that data regarding expectant management for ovarian teratomas were scarce.11,12
The aim of the present study is to analyze long-term results for expectant management of ultrasonographically diagnosed benign ovarian teratomas.
MATERIALS AND METHODS
This is a retrospective observational study. The study comprises a cohort of consecutive asymptomatic nonpregnant women diagnosed as having a benign ovarian teratoma by ultrasound examination at a single tertiary care university hospital (Hospital Universitari Dexeus, Barcelona, Spain) from January 2003 to December 2013.
Women were recruited at the outpatient gynecologic unit of this institution. Ultrasonography was performed as part of a routine gynecologic examination (a routine practice at this institution), to evaluate a pelvic mass found on pelvic examination, or as a second opinion after ultrasound examination performed outside the institution.
Most women underwent transvaginal ultrasound evaluations supplemented by transabdominal ultrasonography in large lesions. Eight women with intact hymen underwent only transabdominal ultrasonography. All ultrasound examinations were performed by expert sonologists with more than 15 years' experience and exclusively dedicated to gynecologic ultrasonography (M.A.P., C.P., B.G.) using a Voluson 730 Expert or E8 ultrasound machines and according to a predefined scanning protocol.13
The diagnosis of a benign ovarian teratoma required the finding of at least one of the following features: a cystic mass with mixed echogenicity, thick band-like internal echoes, a fat–fluid level, or an echogenic tubercle with posterior shadowing (the so-called Rokitansky protuberance).14,15Figures 1–3 show some examples of these ultrasonographic features.
Tumor size was measured in three orthogonal planes with calipers placed on the inner surface of the tumor. The largest diameter, reported in millimeters, was used for analysis. When normal ovarian tissue on the affected side was noted (the so-called crescent sign), it was not included in tumor size measurement.
The protocol used for treatment and follow-up recommendations is outlined in Box 1. Expectant management consisted of a follow-up scan at 3 months, 6 months, and yearly thereafter. Only women with more than 6 months’ follow-up were included.
Protocol for Management for Asymptomatic Women Diagnosed Ultrasonographically With Benign Ovarian Teratomas
- A. Cysts 3 cm or greater in maximal diameter: consider laparoscopic removal if the cyst causes anxiety to the patient.
- a. If the patient declines surgical treatment: repeat ultrasonography in 3 and 6 mo to confirm stability.
- B. Cysts greater than 6 cm: case reviewed by the surgical committee.*
- C. Cysts less than 3 cm in maximal diameter: follow-up scan at 3 mo, 6 mo, and yearly thereafter.
- D. Bilateral cysts: provide information on fertility preservation.†
- a. Cysts greater than 6 cm: case reviewed by the surgical committee.*
- b. Cysts less than 3 cm in maximal diameter: follow-up scan at 3 mo, 6 mo, and yearly thereafter.
- A. Ipsilateral or contralateral new lesion: ultrasound follow-up scan at 3 mo, 6 mo, and yearly thereafter.
- For new lesions, tumor markers (CA 125, CA 19.9, HE4) followed by laparoscopic salpingo-oophorectomy of the affected side.
- For chronic lesions
- a. Yearly follow-up, and if changes occur, surgery.
- b. Lesion removal if the patient requests it.
*Our surgery committee is held weekly and ponders pros and cons of follow-up compared with surgery. If the committee requests surgery and the patient has no surgical risk as a result of other medical or personal reasons, the procedure will take place. If the committee requests follow-up, this could be reevaluated if important changes in the scenario occur.
†Information provided regarding preservation of fertility includes cryopreservation of oocytes in case of damage to ovaries as a result of surgery or the teratoma itself.
Surgery was recommended if the patient developed symptoms that could be related to the teratoma such as pelvic or abdominal pain (either acute or chronic), if tumor size increased 10 mm or greater during the first two follow-ups scans (at 3 and 6 months), or if ultrasound characteristics changed suggesting malignancy such as the appearance of vascularized solid areas or ascites. Surgery was also performed in case of the patient's decision, referring physician's recommendation according to our protocol, or concomitant to another surgery (for example, hysterectomy for leiomyomas). In case of surgery, a definitive histologic diagnosis was obtained and it was considered the gold standard.
The possible benefits and risks of expectant management were discussed with the patients. All women included gave verbal informed consent for conservative management. The study was approved by the institutional review board (Càtedra d'Investigació en Obstetricia I Ginecologia of Universitat Autònoma de Barcelona).
To identify the patients diagnosed with teratoma, an SQL query was performed on the database of the Electronic Clinical Record of the Department of Obstetrics, Gynecology and Reproduction of our institution.
Categorical variables are expressed as absolute numbers and percentage; continuous variables are expressed as mean with SD or median with interquartile range, depending on normal distribution or not. Categorical variables were compared using χ2 with Yates correction. Continuous variables were compared using two-tailed Student t test or Mann-Whitney U tests, where appropriate.
A Cox regression model was used to assess the probability of surgery (hazard ratio) as a function of time using the patient's age and tumor size as covariates. Model was adjusted to menopausal status. A P value <.05 was considered statistically significant for all comparisons. Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines were used for this study.16
During the study period, 613 asymptomatic women were diagnosed as having a benign ovarian teratoma by ultrasonography. Patients' mean age was 36.6 years old (SD 10.6), ranging from 14 to 81 years old. Forty-six (7.5%) women had bilateral ovarian teratomas on ultrasound scan.
Two hundred five (33.4%) women underwent surgery shortly (within 4 weeks) after diagnosis. Reasons for surgery were as follows: referring physician's recommendation according to our protocol (n=189), patient's decision (n=11), and concomitant to another surgery (n=5). Histologic diagnoses of those lesions are shown in Table 1: 183 of 205 (89.2%) were confirmed as benign ovarian teratomas. These 205 cases are excluded from follow-up analysis; results of surgery are presented as additional information in Table 2.
Therefore, 408 women were ultimately included and constitute the basis of the present report (Fig. 4). Three hundred sixty-four women (89.2%) were premenopausal and 44 (10.8%) were postmenopausal. When comparing women who underwent surgery shortly after diagnosis and those women managed expectantly, we found that women managed expectantly were significantly older, had smaller lesions, and bilaterality was less frequent (Table 2).
During follow-up, 130 (31.8%) women underwent surgery. Histologic diagnoses are shown in Table 1. Reasons for surgery were as follows: physician's recommendation according to our protocol (n=115), pelvic pain (n=4), patient's decision (n=6), and concomitant to another surgery (n=5). One of 408 (0.2%) patients had a confirmed ovarian torsion (1/1,541: 6.5 cases/10,000 woman-years). Histologically 103 of 130 (79.2%) were confirmed teratomas; of the remaining 20%, only two serous borderline tumors and no malignant tumor were found. The total number of histologically confirmed teratomas was 286 of 335 (85.4%).
Mean tumor size at diagnosis was significantly larger (43.5 mm, SD 22.5) in women who had surgery during follow-up compared with those women who did not have surgery (27.7 mm, SD 20.2) (P<.001). In women who underwent surgery, mean tumor size before surgery (51.4 mm, SD 23.7) was significantly larger that mean tumor size at diagnosis scan (43.5 mm, SD 22.5) (P<.01). In these cases, we observed a growth rate of 4.8 mm/y.
Among those 278 women who did not undergo surgery, median follow-up was 45.6 months, ranging from 6 to 147 months. In this group, there was no significant change in mean tumor size between the scan at diagnosis (20.5 mm, SD 10.8) and last scan during follow-up (22.8 mm, SD 4.7) (P>.05), a growth rate of 0.6 mm/y.
In the expectant group, most surgeries were performed within the first 5 years after diagnosis (112/130 [86.2%]) (Fig. 5). Main factors associated with surgery were patient's age at diagnosis (hazard ratio 0.97, 95% CI 0.95–0.98) and tumor size at diagnosis (hazard ratio 1.05, 95% CI 1.04–1.06) adjusted by menopausal status as a covariate in the final model. Measured at diagnosis, risk of surgery augmented 5% for every millimeter of the lesion's size and dropped 3% relative to every year of age. Bilaterality was not related to the risk for undergoing surgery in the expectant cohort.
In the present study, we assessed the results of expectant management of ultrasonographically diagnosed benign ovarian teratomas in asymptomatic nonpregnant women.
Among those lesions that were removed surgically, we observed that sensitivity of ultrasonography for diagnosing ovarian teratoma was 286 of 335 (85.4%), similar to others reported in the literature.2,3 We could not calculate specificity because our series analysis is based only on ovarian lesions ultrasonographically diagnosed as benign ovarian teratomas. Of relevance, we had no invasive malignant tumor, but three borderline tumors (0.9% of those lesions removed surgically). In our experience, the risk of missing an ovarian cancer is very low when a diagnosis of benign ovarian teratoma is achieved at ultrasound examination. This observation confirms data from previous studies.11,12 According to our data, endometriomas and serous cysts were the lesions more frequently confused for teratomas on ultrasound examination. There were two cases of struma ovarii, a type of monodermal teratoma.
The vast majority of women remained asymptomatic during follow-up. We observed a low growth rate (0.6 mm/y) in women who did not undergo surgery, but a significant growth rate (4.8 mm/y) in women who underwent surgery during follow-up. This finding is not unexpected, because the faster growing teratomas had a higher chance of being removed. We found no specific factor that could predict this evolution other than age and tumor size at diagnosis. Caspi et al11 reported on a series of 86 women with a ultrasonographic diagnosis of benign ovarian teratomas. Mean follow-up was 34.8 (SD 21.6 months). They observed a mean cyst growth rate of 1.8 mm/y in premenopausal women (n=72) but no size increase in postmenopausal women (n=14). Hoo et al12 reported on 93 women who underwent expectant management for a benign ovarian teratoma. Median follow-up was 12.6 months (range 3–73 months). These authors found a cyst growth rate of 1.7 mm/y. They made no distinction between premenopausal and postmenopausal women.
Torsion has been reported as high as 3.5% for ovarian benign teratomas.5 However, we observed a much lower rate of torsion (0.2%). We do not have a clear explanation for this finding. However, it should be noted that there were no reports of ovarian torsion in Caspi's11 and Hoo's12 studies. Thus, we can conclude that the actual risk for torsion of ovarian teratoma is lower than that reported in early studies.
We observed that younger women and women with larger cysts at initial diagnosis had higher probabilities of undergoing surgery during follow-up. We detected no significant relationship between bilaterality and the risk of surgery. We did not assess parity or medical history. In the study from Caspi et al,11 21 women (24.4%) underwent surgery, most of them because of the cyst's increase in size. In Hoo's study, the authors reported that older and multiparous women, women with a history of an ovarian mass, and women with bilateral or large cysts had higher odds of having surgery. Therefore, increase in size seems to be a critical factor for surgical treatment, although most women will remain asymptomatic during follow-up. However, there is not a clear lesions size cutoff or growing rate that could be used in clinical practice for advising surgery. At our center, with a very low rate of complications (ovarian torsion for instance), we arbitrarily advise surgery for lesions larger than 6 cm. Our results show that, in women with more than 5 years of follow-up, the risk of surgery decreases significantly.
The main strengths of our study were the sample size and median follow-up. After performing a systematic search of papers published in English language using the terms “Ovary,” “Dermoid,” “Teratoma,” and “management” in several databases (PubMed, Web of Science, and Google Scholar), we observed that our series is the largest one (n=278) with the longest follow-up (median 45.6 months) reported to date analyzing long-term results of expectant management for benign ovarian teratomas.
The main weakness of our study is its retrospective design, which precludes assessing some potential factors that could be related to the natural history of this kind of lesion such as hormonal treatments, medical history, or contraceptive use. Despite these weaknesses, we do think that the results of our study might be relevant for clinical practice because we observed that most benign teratomas do not grow and the risk of surgery becomes low after 5 years from diagnosis.
In conclusion, our results demonstrate that expectant management might be a reasonable option for managing asymptomatic women who have an ultrasonographic diagnosis of a benign ovarian teratoma.
1. World Health Organization. Tumors of the ovary and peritoneum. In: Tavassoli FA, Devilee P, editors. Pathology and genetics. Tumors of the breast and female genital organs. Lyon (France): IARC Press; 2003.
2. Sokalska A, Timmerman D, Testa AC, Van Holsbeke C, Lissoni AA, Leone FP, et al. Diagnostic accuracy of transvaginal ultrasound examination for assigning a specific diagnosis to adnexal masses. Ultrasound Obstet Gynecol 2009;34:462–70.
3. Alcázar JL, Guerriero S, Laparte C, Ajossa S, Ruiz-Zambrana A, Melis GB. Diagnostic performance of transvaginal gray-scale ultrasound for specific diagnosis of benign ovarian cysts in relation to menopausal status. Maturitas 2011;68:182–8.
4. Timmerman D, Ameye L, Fischerova D, Epstein E, Melis GB, Guerriero S, et al. Simple ultrasound rules to distinguish between benign and malignant adnexal masses before surgery: prospective validation by IOTA group. BMJ 2010;341:c6839.
5. Comerci JT Jr, Licciardi F, Bergh PA, Gregori C, Breen JL. Mature cystic teratoma: a clinicopathologic evaluation of 517 cases and review of the literature. Obstet Gynecol 1994;84:22–8.
6. Alcázar JL, Aubá M, Ruiz-Zambrana A, Olartecoechea B, Diaz D, Hidalgo JJ, et al. Ultrasound assessment in adnexal masses: an update. Expert Rev Obstet Gynecol 2012;7:441–9.
7. Chapron C, Dubuisson JB, Samouh N, Foulot H, Aubriot FX, Amsquer Y, et al. Treatment of ovarian dermoid cysts. Place and modalities of operative laparoscopy. Surg Endosc 1994;8:1092–5.
8. Alcázar JL, Olartecoechea B, Guerriero S, Jurado M. Expectant management of adnexal masses in selected premenopausal women: a prospective observational study. Ultrasound Obstet Gynecol 2013;41:582–8.
9. Ayhan A, Aksu T, Develioglu O, Tuncer ZS, Ayhan A. Complications and bilaterality of mature ovarian teratomas (clinicopathological evaluation of 286 cases). Aust N Z J Obstet Gynaecol 1991;31:83–5.
10. Stern JL, Buscema J, Rosenshein NB, Woodruff JD. Spontaneous rupture of benign cystic teratomas. Obstet Gynecol 1981;57:363–6.
11. Caspi B, Appelman Z, Rabinerson D, Zalel Y, Tulandi T, Shoham Z. The growth pattern of ovarian dermoid cysts: a prospective study in premenopausal and postmenopausal women. Fertil Steril 1997;68:501–5.
12. Hoo WL, Yazbek J, Holland T, Mavrelos D, Tong EN, Jurkovic D. Expectant management of ultrasonically diagnosed ovarian dermoid cysts: is it possible to predict outcome? Ultrasound Obstet Gynecol 2010;36:235–40.
13. Pascual MA, Hereter L, Tresserra F, Carreras O, Ubeda A, Dexeus S. Transvaginal sonographic appearance of functional ovarian cysts. Hum Reprod 1997;12:1246–9.
14. Mais V, Guerriero S, Ajossa S, Angiolucci M, Paoletti AM, Melis GB. Transvaginal ultrasonography in the diagnosis of cystic teratoma. Obstet Gynecol 1995;85:48–52.
15. Guerriero S, Ajossa S, Gerada M, Virgilio B, Pilloni M, Galvan R, et al. Transvaginal ultrasound in the diagnosis of extrauterine pelvic disease. Expert Rev Obstet Gynecol 2008;3:731–52.
16. von Elm E, Altman DG, Egger M, Pocock SJ, Gøtzsche PC, Vandenbroucke JP, et al. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies. Lancet 2007;370:1453–7.