Simple ovarian cysts in premenopausal patients

International Journal of Gynecology & Obstetrics 57 (1997) 49-55

Article

Simple ovarian cysts in premenopausal B. Gerber*,

H. Miiller,

patients

T. Kiilz, A. Krause, T. Reimer

Depamnent of Obstem’csand Gynecology Universi~ of Restock, Restock, kennatty

Received3 June1996;revised20November1996;accepted 25November1996

Abstract Objective: To compareclinical, ultrasonographical,and cytological findingswith the histopathologicaldiagnosisof unilocular, anechoic smooth-walledcystic ovarian tumors (‘simple ovarian cysts’). Method: In 140 premenopausal womensimpleovarian cystswere removedby laparoscopyfollowing ultrasoundevaluation. In this retrospectivestudy the histopathologicaldiagnosiswascorrelated with clinical data, sonographiccharacteristics,macroscopicimpression and with cytological findings.Results: Histopathologyrevealed21 (15.0%)functional cysts,31 (22.1%) retention cysts, 9 (6.4%) endometriomas,3 (2.1%) cystic teratomas, 12 (8.6%) undifferentiated cystsand 64 (45.7%) cystadenomas. No mentionable differences were correlated with the patient’s age or the size of the cyst as determined by ultrasound. Classically,‘chocolate-like’ cystic fluid characterizesendometriomas.However, in the presentstudy cysts with different histopathologicalclassificationsexhibited similar fluid characteristics.The cytological diagnosiswas correct in only 53 (37.9%) of all 140 cases.Conchion: In premenopausalwomen differential diagnosisof ovarian cystsis not possibleby clinical characterization, either by ultrasound or cytological evaluation. Simpleovarian cysts shouldbe observedfor at least 8 weeksor 2 menstrualcycles,respectively.If persistingover that period, the ovarian cyst shouldbe removedby laparoscopy,but not by cyst aspiration. 0 1997International Federation of Gynecology and Obstetrics

Keywords: Ovarian cyst; Ultrasound; Observation time; Laparoscopy;Cytology; Histology

1. Introduction A ‘simple ovarian cyst’ is defined by ultrasound as a unilocular anechoic smooth-walled mass in the adnex region. However, the histological status of the cyst cannot be determined by this diagno-

*Corresponding author.Tel.: +49 3814948188; fax: +49 3814948102.

sis. The clinical management offers considerable differences and ranges from observation and follow-up to surgical removal. In recent years some authors recommend the aspiration of ovarian cysts and cytological examination of cyst content as a valid alternative to surgical removal. The accuracy of cytology from ovarian cyst aspirates to rule out malignancy is still arguable. The aim of this study was to assessthe diagnostic value of cytology in connection with clinical features in

0020-7292/97/$17.00Q 1997InternationalFederationof Gynecology andObstetrics PII SOO20-7292(97)

02832-9

50

B. Gerber

et al. /International

Journal

the management of simple ovarian cysts in premenopausal women. 2. Materials and methods

Between March 1990 and March 1996 we have treated 154 patients with a single and ‘simple cyst’ in the ovarian region. The total numbers of premenopausal patients subjected to operation due to ovarian pathology and ultrasound findings during the same time are shown in Table 1. Only 10.5% (142/1358) of ovarian tumors were evaluated as ‘simple ovarian cysts’, whereas the other ones offered septations or internal echogenicity. All ‘cysts’ fulfilled specific sonographical criteria: they were unilateral, unilocular, anechoic, smooth-walled and contained no septa or solid areas in transvaginal and abdominal ultrasound. All 154 patients underwent laparoscopical removal of simple ovarian cyst. In 14 (9.1%) of the 154 patients the sonographically suspected ovarian cyst was not confirmed intraoperatively. Therefore paraovarian cysts (n = 5), paratubal cysts (n = 41, hydrosactosalpinx (n = 21, peritoneal cysts (n = 2) and retroperitoneal cyst (n = 1) were found. Surgical removal was performed under general anaesthesia by video-laparoscopy (Karl Storz GmbH and Co., Tuttlingen, Germany). After surgical removal, the cyst was macroscopitally evaluated and 15-20 ml fluid were aspirated by a 20-gauge needle. The smears were then stained according to the method of Papanicolaou.

of Gynecology

& Obstetrics

57 (1997)

49-55

Cytological findings were divided into two groups, suspicious or not suspicious, according to the criteria of Selvaggi [l] and Davila [2] (Table 2). Cyst walls were completely removed either by conserving the ovary or by unilateral adnexectomy (n = 39). For laparoscopical removal of the specimens an Endo-CatchTM (Autosuture, Tonisvorst, Germany) was used. The age of patients ranged from 13 to 56 years (jz + S.D.: 33.8 + 8.7). The histological findings were correlated with age, recorded time of persistence, measures by ultrasound, color and consistency (clear/yellow, mutinous, sanguineous/chocolatelike, sebaceous) of fluid content, use of oral contraceptive, and hormonal treatment. Follicular cysts, luteinized follicular cysts, hemorrhagic corpus luteum cysts as well as luteal cyst were summarized as functional cysts. Elderly corpus luteum cysts, also called retention cysts, were analyzed separately. Their histological characteristics are defined as a small border of yet existing luteinized granulosa cells and low lipid content. Differences were evaluated for significance by chi-square test and t-test. 3. Results

In 140 simple ovarian cysts histopathological examination revealed 21 (15.0%) functional cysts, 31 (22.1%) retention cysts, 9 (6.4%) endometric

Table 1 Total number of different ovary pathology and preoperative ultrasound findings in premenopausal patients subjected to operation due to the suspected pathology at our department between March 1990 and March 1996 Histology

Functional cysts Retention cysts Endometriomas Cystadenomas Cystic teratomas Not differentiated benign cysts Ovarian cancer Total

Patients

Ultrasound findings Not done Ovarian ‘cysts’ with internal echogenicity, septations etc.

‘Simple ovarian cyst’

62 167 91 618 95 67

2 6 5 32 3 8

39 (62.9%) 130 (77.8%) 77 (84.6%) 522 (84.5%) 89 (93.7%) 47 (70.1%)

21(33.9%) 31 (18.6%) 9 (9.9%) 64 (10.4%) 3 (3.2%) 12 (17.9%)

258 1358

3 59

253 (98.1%) 1157 (85.2%)

2 (0.8%) 142 (10.5%)

B. Gerber et al. /International Journal of Gynecology & Obstetrics 57 (1997) 49-55 Table 2 cytological

findings

which

are indicative

of histology

Histology

WohY

Follicular cysts Luteinized follicular cysts Hemorrhagic corpus luteum Corpus luteum cysts Retention cysts Endometriomas Cystadenomas Not differentiated benign cysts

Granulosa Granulosa Luteinized Luteinized Luteinized Erythrocytes, Epithelial No specific

[1,21

cells cells, luteinized granulosa cells granulosa cells, erythrocytes granulosa cells, fibrin, hemosiderin-pigmented macrophages granulosa cells, histiocytes, fibroblasts, fibrin, macrophages with hematoidin hemosiderin-pigmented macrophages, endometrial cells cells or clusters with small oval nuclei, frequently nude nuclei, macrophages signs

cysts, 64 (45.7%) cystadenomas, 3 (2.1%) cystic teratomas and 12 (8.6%) not differentiated benign cysts. The differentiation of cystadenomas offered 48 serous, 9 mutinous and 7 seropapillary cystadenomas. The mean age extended from 31.5 _+ 8.8 years in women with endometriotic cysts to 35.3 f 9.5 years in patients with cystic teratomas and was not statistically different between the groups (P > 0.05). Nearly two-thirds (64.3%) of all ‘cysts’ had been observed more than 8 weeks. Only among patients with functional cysts, 8 (38.1%) of 21 patients,was the observation time 8 weeks and more. These differences did not prove to be statistically significant (P > 0.05) in comparison to the retention cysts group (64.5%), but was different with P < 0.001 from the cystadenomas group (70.3%). Fifty-seven percent of the cysts removed were larger than 5 cm in diameter. All three cystic teratomas and 64.1% of cystadenomas were larger than 5 cm, while only 38.7% of retention cysts were larger than 5 cm (Table 3). Table 3 Histological

findings,

age, time of observation

Histology

Functional cysts Retention cysts Endometriomas Cystadenomas Cystic teratomas Not differentiated Total

n

benign

cysts

51

21(15.0%) 31(22.1%) 9 (6.4%) 64 (45.7%) 3 (2.1%) 12 (8.6%) 140 (100.0%)

and sonometry

Lower abdominal pain at time of referral was reported by 13 (9.3%) of 140 patients. The histological diagnosis for these patients revealed functional cysts (n = 51, cystadenomas (n = 3), retention cysts (n = 31, and endometriomas (n = 2). The fact, that 5 (23.8%) of 21 patients with functional cysts had complaints, may contribute to the relatively short observation time in this group. A contorted cyst was found in one case per group for functional cysts, retention cysts and cystadenomas. Hormonal treatment after sonographical diagnosis of a simple ovarian cyst was performed in 29 (20.7%) of 140 patients. Among these women, only one patient had a functional cyst, but 18 patients had cystadenomas and 7 patients had retention cysts. The intraoperative evaluation of the color and consistency of the cyst’s content is shown in Table 4. In 55.7% of all cases the cyst content was described as clear or yellow. Among the functional cysts, 9 corpus rubrum cysts were found, con-

of ovarian

cysts

Age of patients

Observation

(years

I 8 weeks

32.6 34.4 31.5 34.1 35.3 33.7

+ S.D.) f & + + f +

13.1 10.6 8.8 8.2 9.5 8.3

33.8 f 8.7

pigment

time

Sonometry > 8 weeks

15cm

>5CRl

13 11 2 19 1 4

8 20 7 45 2 8

(38.1%) (64.5%) (77.8%) (70.3%) (66.7%) (66.7%)

8 19 4 23 0 5

13 (61.9%) 12 (38.7%) 5 (55.6%) 41(64.1%) 3 WO%o) 7 (58.3%)

50

90 (64.3%)

59

81(57.9%)

52 Table 4 Macroscopical

B. Gerber

evaluation

et al. /International

of cyst content

Histology

n

and color

Journal

by the operating

benign

cysts

Total

21 31 9 64 3 12

11 (52.4%) 12 45 (70.3%)

140

78 (55.7%)

10 (83.3%)

taining bloody fluid. ‘Chocolate-like’ content was observed in 9 (100%) of the endometriomas, but was found in other diagnosis groups as well. Numerous retention cysts and cystadenomas were misjudged as endometriomas according to macroscopical evaluation of the cyst content. The mutinous consistency of cyst content led to the correct tentative diagnosis in only 7 (77.8%) of 9 mutinous cystadenomas. Typical epithelial structures and sebaceous cyst content were seen only in 2 of 3 cystic teratomas. Cytology by itself pointed to the correct diagnosis in only 53 (37.9%) of all 140 cases. Based on cytological findings all three cystic teratomas, 66.7% of the endometrioma (hemosiderin-pigmented macrophages), 45.2% of retention cysts and 37.5% of the cystadenomas (epithelial cells) would have been diagnosed correctly. In the functional cysts group, cytology would have predicted the histological findings in less than one third (Table 5). 4. Discussion

4.1. Probability

& Obstem’cs 57 (1997) 49-55

gynecologist

Color/consistency Clear/yellow

cysts Functional Retention cysts Endometriomas Cystadenomas Cystic teratomas Not differentiated

of Gynecology

of malignancy

Unilocular, anechoic, smooth-walled ovarian cysts have a very low probability of being malignant. This risk has been reported to be less than 2% in premenopausal women [3,4], and extend to 6% [5] in postmenopausal women. Morphological characteristics, determined by ultrasound, can further discriminate the likelihood of malignancy. Among 1317 patients, Schillinger [4] found malignant ovarian tumors in 0.9% of the clearly out-

Mutinous 1 8 1 1 11 (7.9%)

Sanguineous/chocolate-like

Sebaceous

9 19 (61.3%) 9 (lOO%o) 10 1

1 2 (66.7%) -

48 (34.3%)

3 (2.1%)

Table 5 Histology and cytological ing cytology was present, present Histology

tindings in cyst fluid. A demonstratif cytology findings of Table 1 were

n

Cytology indicative of histological diagnosis Yes

Functional cysts Retention cysts Endometriomas Qstadenomas Cystic teratomas Not differentiated benign cyst Total

21 31 9 64 3 12 140

6 14 6 24 3 -

No (28.6%) (45.2%) (66.7%) (37.5%) (100%)

15 17 3 40 0 12 (100%)

53 (37.9%)

87 (62.1%)

lined solitary cysts, in 1.9% of the clearly outlined homogeneous tumors, in 17% of the poorly defined or slightly heterogeneous tumors, in 58% of the marked heterogeneous tumors, and in 75% of the completely heterogeneous tumors. Nevertheless, the risk of ovarian cancer, appearing as a simple ovarian cyst, remains. Puls et al. [6] examined 96 ovarian cystadenocarcinomas, observing benign epithelium adjacent to an area of borderline or malignant epithelium in 74 tumors (79%) and a site of epithelial transition in 38 cases (40%). This demonstrates the possibility of carcinogenesis within benign cystic ovarian tumors. In the presented study of 140 simple ovarian cysts treated by laparoscopy there was no case of malignancy, although - during the time of this study - a patient was referred to our hospital, in whom ovarian cancer was misjudged as a

B. Gerber

et al. /International

Journal

benign ovarian cyst by ultrasound as well as by laparoscopy. In two of our own cases of ‘simple ovarian cysts’ the malignancy was suspected due to minimal papillary structures at the ovarian surface. The puncture of a benign appearing cyst can always lead to tumor cell spill, thus resulting in extensively disseminated ovarian carcinoma [7]. 4.2. Diagnostic value of cytology The diagnostic value of cytological examination of fluids from ovarian cysts is still arguable. Inadequate material (acellular fluid, blood cells, debris, degenerated cells, or scanty mucus> impeding the correct cytological diagnosis has been reported in 12% [2], 14% 181, 18% [91, 19.5% [lo] or even 72% [ 111 of cases. Hasson [12] analyzed the literature and concluded that a correct cytological interpretation of the aspirates is difficult: firstly because of the great diversity of normal and pathologic cells, secondly because of the inability to differentiate between benign and borderline tumors or between various types of benign cysts lined with a single layer of cuboidal cells, and thirdly because a localized ovarian malignancy may be missed. According to Greenebaum et al. [8] and Davila [2], highly cellular follicular cysts may resemble granulosa cell tumors, luteal cysts may mimic adenocarcinoma, and epithelial atypia in endometriosis or serous cystadenoma may also lead to a false positive cytological diagnosis of malignancy. These potential pitfalls limit the diagnostic value of cytology. Dordoni et al. [13] missed 3 of 5 malignancies in a series of 204 cases. The negative predictive value with respect to malignancy was found by Granberg et al. [14] to be 77%. Our results show a low accuracy of cytological examination of ovarian cyst aspirates. In only 53 (37.9%) of 140 cases cytological findings correlated with the histological type of ovarian cyst. Macroscopical evaluation of the cyst content and color could not contribute to the improvement of diagnostic accuracy. Several attempts have been made to improve cytological methods. Biochemical characteristics of the aspirate and tumor markers in peripheral blood were combined with cytology [15-171. Andolf et al. [15] found correlation of fibrinolytic

of Gynecologv

& Obstetrics

57 (1997)

49-55

53

activity and estradiol concentration in cyst fluid with histopathology and recurrence rate after puncture. Kreuzer et al. [161 compared cytology, ultrasound, and the estradiol content of the cyst’s fluid with the histological diagnosis in 203 cases. In hormonal analysis of the cysts content, estradiol levels less than 800 pg/ml suggested cystic neoplasia. Although in all cystic carcinomas (15/15) and in 96.7% of the mutinous cystadenomas (29/30) cytology yielded the correct diagnosis; only 52.5% of the serous cystadenomas (21/40) and 68.7% of the dermoid cysts were diagnosed correctly with cytology. Overall, in 101 cystic neoplasms, all three methods indicated cystic neoplasia in 42.6%, were contradictory in 56.4% and failed completely in 1.0%. Greenebaum et al. [8] found analysis of DNA ploidy of ovarian cyst fluid a useful adjunct to cytology, showing that 90% (47/52) of the benign cysts were diploid, all 6 fully malignant cysts were non-diploid, and 1 of 3 borderline cysts were polyploid. 4.3. Diagnostic value of ultrasound Ultrasound contributes substantially to the preoperative diagnosis. According to Herrmarm et al. [3,18], ultrasound predicted 82.6% (38/46) of the ovarian carcinomas and 95.6% (177/185) of the benign tumors. The sonographically determined size of cyst correlates poorly with histology. Luxman et al. [5] reported, in a series of 102 patients, malignant tumors in 2 simple ovarian cysts smaller than 5 cm. Thus, sonographical scores have been established to improve precision and reliability of the ultrasonographical diagnosis [19], all relating to wall structure and thickness, septations, and echogenicity. For optimal results, transvaginal and transabdominal ultrasound are performed, thus combining the superior resolution of the first with the larger scanning depth and overview of the latter method. Color Doppler ultrasound can add information about the localization of blood vessels within the tumor and about blood flow patterns, such as resistance index, pulsatility index, and diastolic notch [20]. Three-dimensional (3D) ultrasound can further improve the reliability of

54

B. Gerber

et al. /International

Journal

sonography, as reported by Bonilla-Musoles et al. [21]. All 5 ovarian malignancies in a series of 76 ovarian masses were correctly predicted by 3-Dultrasound. Magnetic resonance (MR) imaging represents yet another technique for preoperative evaluation of ovarian masses. Yamashita et al. [22] reported that 95% of malignancies were correctly identified by contrast-enhanced MR imaging, as compared to 88% detected by transvaginal ultrasound. For all other ovarian tumors, this rather invasive and expensive method was less than 80% accurate. It seems that all imaging techniques have similar drawbacks by reflecting either morphology or vascularization of ovarian tumors. These structural or functional features of malignant and benign tumors overlap substantially, thus obstructing the precise prediction of histology [23]. 4.4. Management of functional

cysts

The rate of extirpated functional cysts reflects the rate of false-positive preoperative diagnosis. Our rate of 15.0% is similar to the percentage reported in the literature: Canis et al. [241 found 18%, Kreuzer et al. [16] 15%, and Seltzer [251 18.7%. Expectative management of simple ovarian cysts can keep the number of unnecessarily removed functional cysts down, but requires close follow-up [ 161. Asymptomatic ovarian cysts in premenopausal patients should be observed at least for 8 weeks or 2 hormonal cycles, respectively. Conservative management of ovarian cysts in premenopausal women can be supported by the administration of high-dose monophasic contraceptive pills (> 35 pg estradiol). The success of such a treatment depends on the age of the functional cyst, because elderly cysts offered a loss of steroid hormone receptors. 4.5. Recurrence rate afrer ultrasound-guidedpuncture

Ultrasound-guided puncture and aspiration of ovarian cysts is associated with high recurrence rates. Incomplete cyst emptying or recurrences after ultrasound guided aspiration have been reported in 25% [lo], 54% [26], 57% [15], or even 65% [13]. Andolf et al. [15] found that high fibrinolytic activity and low estrogen levels in the cyst

of Gynecology

& Obstetrics

57 (1997) 49-55

content as well as a large cyst volume can predict recurrence. 5. Conclusions

Neither ultrasound nor cytology can reliably predict histological diagnosis of ovarian cysts. In addition, the presented data demonstrate that the patient’s age as well the cyst size, color or consistency of cyst content or the cytological evaluation are not reliable indicators for the histology of a cyst. Although the probability of ovarian cancer in premenopausal simple ovarian cysts is extremely low, its occurrence must be considered. To avoid an excessive rate of unnecessarily removed functional cysts, asymptomatic unilocular anechoic smooth-walled masses in the adnex region should be observed for at least 2-3 menstrual cycles. A hormonal treatment, preferably by high-dose monophasic pills, may also decrease the number of functional cysts. If the cyst persists, complete laparoscopical removal is indicated. References

111 Selvaggi SM. Cytology of nonneoplastic cysts of the ovary. Diagn Cytopathol 1990; 6: 77-85.

121 Davila RM. Cytology of benign cystic uterine adnexal

masses. Acta Cytol 1993; 37: 385-390. Herrman UJ Jr, Lecher GW, Goldhirsch A. Sonographic patterns of ovarian tumors: prediction of malignancy. Obstet Gynecol 1987; 69: 777-781. [41 Schillinger H, Kliem M, Klosa W, Pohl J, Wilhelm C, Madjar H, Zalasa J. Assessing the extent of gynecologic tumors by a sonographic tumor score with special reference to ovarian cancer. (German) Arch Gynecol Obstet 1989; 245: 596-600. 151 Luxman D, Bergmann A, Sagi J, David M. The postmenopausal adnexal mass: correlation between ultrasonic and pathologic findings. Obstet Gynecol 1991; 75:

t31

726-728.

El Puls LE, Powell DE, DePriest PD, Gallion HH, Hunter

JE, van Nagell JR. Transition from benign to malignant epithelium in mutinous and serous ovarian cystadenocarcinoma. Gynecol Oncol 1992; 47: 53-57. [71 Trimbos JB, Hacker NF. The case against aspirating ovarian cysts. Cancer 1993; 72: 828-831. Bl Greenebaum E, Yee HT, Liu J. DNA ploidy of ovarian and adnexal cyst fluid. A useful adjunct to cytology. Acta Cytol 1994; 38: 201-208. [91 Wojcik EM, Selvaggi SM. Fine-needle aspiration cy-

B. Gerber

[lOI

[111 WI 1131

[141

1151

M [171

et al. /International

Journal

tology of cystic ovarian lesions. Diagn Qtopathol 1994; 11: 9-14. Bonilla Musoles F, Ballester MJ, Simon C, Serra V, Raga F. Is avoidance of surgery possible in patients with perimenopausal ovarian tumors using transvaginal ultrasound and duplex color doppler sonography? J Ultrasound Med 1993; 12: 33-39. De Crespigny L, Robinson HP, Davoren RAM, Fortune D. The ‘simple’ ovarian cyst: aspirate or operate? Br J Obstet Gynaecol 1989; 96: 1035-1039. Hasson HM. Ovarian surgery. In: Sanfilippo JS, Levine RL, editors, Operative Gynecological Endoscopy. New York: Springer, 1989: 86-106. Dordoni D, Zaglio S, Zucca S, Favalli G. The role of sonographically guided aspiration in the clinical management of ovarial cysts. J Ultrasound Med 1993; 12: 27-31. Granberg S, Norstrom A, Wikland M. Comparison of endovaginal ultrasound and cytological evaluation of cystic ovarian tumors. J Ultrasound Med 1991; 10: 9-14. Andolf E, Casslen B, Jorgensen C, Buchhave P, Lecander I. Fluid characteristics of benign ovarian cysts: correlation with recurrence after puncture. Obstet Gynecol 1995; 86: 529-535. Kreuzer GF, Paradowski T, Wurche KD, Flerker H. Neoplastic or nonneoplastic ovarian cyst? The role of cytology. Acta Cytol 1995; 39: 882-886. Schwartz PE. The role of tumor markers in the preoperative diagnosis of ovarian cysts. Clin Obstet Gynecol 1993; 36: 384-394.

of Gynecology

& Obstetrics

57 (1997)

49-55

55

W31 Herrmann UJ. Sonographic patterns of ovarian tumors.

Clin Obstet Gynecol 1993; 36: 375-383. Lamer JP, Timor-Tritsch IE, Federman A, Abramovich G. Transvaginal ultrasonographic characterization of ovarian masses with an improved, weighted scoring system. Am J Obstet Gynecol 1994; 170: 81-85. DO1 Maly Z, Riss P, Deutinger J. Localization of blood vessels and qualitative assessment of blood flow in ovarian tumors. Obstet Gynecol 1995; 85: 33-36. Pll Bonilla-Musoles F, Raga F, Osborne NG. Three-dimensional ultrasound evaluation of ovarian masses. Gynecol Oncol 1995; 59: 129-135. La Yamashita Y, Torashima M, Hatanaka Y, Harada M, Higashida Y, Takahashi M et al. Adnexal masses: accuracy of characterization with transvaginal US and precontrast and postcontrast MR imaging. Radiology 1995; 194: 557-565. t231 Granberg S. Relationship of macroscopic appearance to the histologic diagnosis of ovarian tumors. Clin Obstet Gynecol 1993; 36: 363-374. 1241 Canis M, Mage G, Pouly JL, Wattiez A, Manhes H, Bruhat MA. Laparoscopic diagnosis of adnexal cystic masses: a 12-year experience with long-term follow-up. Obstet Gynecol 1994: 83: 707-712. [=I Seltzer V. Laparoscopic surgery for ovarian lesions: potential pitfalls. Clin Obstet Gynecol 1993; 36: 402-412. Dl Lipitz S, Seidmann DS, Menczer J, Bider D, Oelsner G, Moran 0, Shalev J. Recurrence rate after fluid aspiration from sonographically benign-appearing cysts. J Reprod Med 1992; 37: 845-848. [191