Laparoendoscopic Ovarian-Sparing Surgery of Adnexal Tumors in Children and Adolescents by General Gynecologists: A 10-Year,Retrospective Cohort Study

Introduction

Pediatric and adolescent adnexal tumors are uncommon, but important, and the most frequent tumors of the female genital tract. ¹ The majority of these cases are functional cysts and hemorrhagic corpus luteal cysts; however, ovarian cystectomy is still only performed in 60% of cases.^2–4 Gynecological tumors during the first two decades of life are not common and they constitute 5%–10% of all tumors and, of those, malignant ovarian cysts occur in only 4.5%–16%. ⁴

In pediatric and adolescent gynecology (PAG), ovarian-sparing surgery (OSS) is an approach for preserving the ovaries affected by tumors and torsion during surgical treatment, such as ovarian cystectomy, to maximize future fertility, quality of life, and future sexual function.^3,5 The ovary has a great ability to regenerate, so although most ovaries appear to be removed during cystectomy, any remaining part is still sufficient to maintain the ovarian function. Therefore, identification of the normal-appearing ovarian cortex versus the cyst/tumor wall is one of the major issues in the surgery. ⁶ Several studies on diagnostic protocols for the evaluation of adnexal tumors in young patients reported different preservation rates, with a significant number of uncomplicated functional cysts that were managed surgically.^2,3,6 Management of adnexal tumors in children and adolescents is currently a controversial topic in the PAG literature. ³ The literature lacks or rarely includes research that explored the management of adnexal tumors by a general gynecologist, so our research will contribute to filling that gap.

The aim of this study is to investigate the laparoendoscopic management of adnexal tumors in children and adolescents and significance of general/adult gynecologists in OSS of adnexal tumors in children and adolescents.

Materials and Methods

A retrospective cohort study was performed at the University Clinical Center, Clinic for Gynecology and Obstetrics, during the 10-year period between January 2008 and December 2017. Ethical approval for this study was obtained from the institutional review board. Using medical records, we identified 40 (aged < or = 19 years, from 12 to 19 years) adolescents with adnexal tumors. Inclusion criteria were adolescents with an operative diagnosis of adnexal tumors, confirmed by ultrasound, and who had undergone a scheduled surgical procedure or emergency surgery, after which the diagnosis was pathologically confirmed. The exclusion criterion was patients who had undergone diagnostic laparoscopic surgery, but without evident adnexal tumors (2 patients).

The patients were stratified based on the presentation and diagnoses and outcomes based on the rate of OSS.

Data on the course of the disease and treatment were collected on the basis of available medical documentation. The analysis included age, age at menarche, occupation, marital status, family history of gynecological tumors (adnexal tumors), virginity, symptoms of the disease, and ultrasound findings (mass size), and elevated values of the tumor marker (CA-125 > 35 U/mL) were observed. All patients underwent a pelvic ultrasound examination—transabdominal, transrectal, or less frequently transvaginal. The adnexal tumor volume was calculated according to the following formula: (length × height × width × 0.523). Adnexal tumor volumes >10 cm³ for premenarchal and >20 cm³ for postmenarchal patients were considered to be abnormal. ⁷

In adolescents with adnexal tumors who underwent scheduled surgery after a decision was made at the general gynecological advisory board (due to lower abdominal pain, menstrual irregularities, urinary disturbances, and ultrasound-verified adnexal masses) or emergency surgery because of an acute abdomen (abdominal pain, nausea, and vomiting), the method of access (laparotomy/laparoscopy) as well as the type of surgical procedure (cystectomy, oophorectomy, or salpingo-oophorectomy), pathological diagnosis, and type of treatment after the presentation of gynecological oncological advisory board for malignant tumors were analyzed. The general gynecology advisory board consists of 3 general gynecologists with special interest in PAG. Decisions about OSS were made based on clinical, ultrasound, tumor marker, and intraoperative findings and sometimes on frozen section (extempore) pathological diagnosis. ⁷

Adnexal tumors include non-neoplastic cysts (corpus luteal cysts, paraovarian cysts, and endometriomas); neoplastic benign cysts (teratomas, cystadenomas, and cystadenofibromas); and borderline and malignant (epithelial, germ cell, and sex cord) tumors.^1,2

Descriptive statistics, mean value, standard deviation, and percentage were used in statistical data processing. Data were statistically assessed with the Student t-test, chi-square analysis, or Fisher exact test to compare a set of two data cohorts where appropriate. Statistical significance was set at P < .05.

Results

A total of 40 patients were identified as potentially eligible for inclusion in the study. During a review of medical records, 2 patients were excluded because of diagnostic laparoscopic surgery. Consequently, the study population comprised 38 consecutive female patients, aged < or = 19 years, who underwent surgery for a preoperative diagnosis of an adnexal tumor during the study interval.

The average age of patients included in the study was 16.7 ± 2.1 years, from 12 to 19 years.

The average age at menarche was 13.05 ± 0.92 years, from 12 to 16 years (Table 1). When we analyzed the occupation and education level, 27 (71%) were elementary and high school students, 6 (15.7%) were university students, and 5 (13.1%) were housewives (P < .000092). Thirty (78.9%) adolescents were virgins (P < .000092). Three patients were married (7.8%) and 35 were unmarried (92.1%) (P < .000092). Family history of gynecological tumors was found in 2 (5.2%) cases. The average adnexal tumor size was 8.7 ± 3.1 cm, from 5 to 19 cm. Elevated CA-125 tumor marker values were observed in 3 (7.8%) cases.

Scheduled surgery was performed in 22 (57.8%) patients, while 16 (42.1%) patients underwent emergency surgery. The laparoscopic approach was performed in 28 (73.6%) patients and laparotomic approach in 10 (26.3%) patients (P < .0001) (Table 1). Of the 10 laparotomies, 4 were for malignant disease, scheduled laparotomies, and six were for the large size of adnexal tumors and dominant solid component, and all six were emergency laparotomies.

OSS was performed in 29 (76.3%), oophorectomy in 1 (2.6%), and salpingo-oophorectomy in 8 (21.05%) patients (Table 1). Reasons for performing oophorectomy/salpingo-oophorectomy were as follows: one case because of a borderline tumor; one case because of a huge, benign, cystic ovarian teratoma (19 × 16 × 11 cm) without signs of preserved ovarian tissue; and three cases because of irreversible necrotic changes caused by torsion. Four salpingo-oophorectomy procedures were performed because of malignant tumors; two adenocarcinomas, one granulosa cell ovarian tumor, and one ovarian angiosarcoma.

A significantly higher number of patients underwent OSS with laparoscopy in the elective and emergency surgery groups (P = .021 and P = .028) (Table 1). OSS was performed more frequently when the surgical approach was laparoscopy (25, 86.2%) compared with laparotomy (4, 13.7%) (Table 1). Significantly younger patients were in the salpingo-oophorectomy/oophorectomy versus OSS (15.6 ± 3.04 years, P = .036) and laparotomy versus laparoscopy (15.1 ± 2.6 years, P = .001) groups. Older patients were more likely to undergo OSS with laparoscopy in the scheduled and emergency surgery groups. A significantly earlier age at menarche (12.8 ± 0.85) was found in OSS (P = .03). The categorization of OSS according to clinical variables is shown in Table 2.

Torsion was seen in 8 (21.05%) patients (Table 1). Overall, 34 (89.4%) tumors were unilateral: 18 (47.3%) were left-sided tumors and 16 (42.1%) were right-sided tumors. Of the bilateral tumors, 4 (10.5%) were mature teratoma, and luteal cysts and paraovarian cysts were found in 2 patients.

There were 31 (81.5%) benign ovarian tumors: 22 (57.8) non-neoplastic cysts and 9 (23.6%) neoplastic cysts—borderline in 3 (7.8%) and malignant in 4 (10.5%) (Table 3). The most frequent benign ovarian tumors were paraovarian cysts in 9 (29.03%) and hemorrhagic luteal cysts in 6 (19.3%) adolescents. Three patients (7.8%) underwent chemotherapy after surgery. Two patients (5.2%) died during a 5-year follow-up period, 1 with ovarian angiosarcoma and 1 with granulosa cell tumor FIGO IIIc stage.

Discussion

Our study demonstrated a high rate of OSS using the laparoscopic approach compared with many studies,^5,7–9 most probably because in our institution, we have a team consisting of 2 general gynecologists who deal mostly with PAG, in addition to other areas of gynecology and obstetrics. In our study, a patient was more likely to have her ovary conserved if she underwent a scheduled surgery compared with an emergency surgery, and a scheduled surgery was more likely to be performed laparoscopically, but without statistical significance.

The least wanted surgical outcome is unnecessary oophorectomy due to ovarian torsion, ovarian cysts, and benign ovarian tumors that are not indicative for ovariectomy as a result of lack of understanding or awareness.^10,11 Most pediatric surgeons tend to perform oophorectomy in girls presenting with ovarian torsion, in which the ovary appears necrotic where the gynecologist tended to perform detorsion and preserve the ovary.^6,12 Detorsion is the procedure of choice in the treatment of ovarian torsion in children and adolescents. ¹² Detorsion was not accompanied by an increase in morbidity, and most patients had functioning ovarian tissue on follow-up despite the surgeon's finding and belief that a grossly blue–black hemorrhagic adnexa was irreversibly damaged.^3,5,12 However, PAG literature suggests that many ovaries can be treated by detorsion alone because torsion occurs with normal ovaries (42%–50%), benign tumors, functional cysts, and malignant tumors (0.5% −1.8%).^11,12

In our study of 8 patients, 7 (87.5%) had functional cysts and 1 patient was with benign tumors (adult teratoma). No malignant tumors were identified in any of our patients with torsion, similar to the study by Aziz et al. ¹² In Aziz et al.'s study, researchers found 45% of detorsion cases with or without cystectomy and 55% of oophorectomies in case of ovarian torsion, indicating a lower rate of detorsion and higher rate of oophorectomies than in our study (75% and 25%, respectively), probably because in Aziz et al.'s study, surgery was performed by a general or pediatric surgeon or gynecologist. ¹²

Current literature indicates a significant trend over the recent decades toward laparoscopy in OSS.^2,3 Laparoscopy and OSS have become the gold standard of care for benign adnexal tumors in PAG with excellent results in terms of ovarian function and few local relapses.^4,6,13–15 The use of laparoscopy has increased from 32.1% to 57.9%, and patients undergoing laparoscopy are two times more likely to undergo cystectomy.

The presence of a gynecologist as a primary surgeon increases the use of laparoscopy and OSS compared with pediatric surgeons.^2,8,9 These views were also confirmed in a study by Peeraully et al. with procedures performed laparoscopically in 50% patients and 65% cases underwent cystectomy, which are lower than our findings (73.6% and 76.3%, respectively). ⁵

Nowadays, more and more pediatric literature and pediatric surgeons advocate a laparoscopic and OSS approach.^14,15 In our study, we avoid spillage during laparoscopy by using endobags. Tricks and techniques to preserve ovaries include the incision of the ovarian capsule, where the dissection plane between the tumor wall and healthy ovarian parenchyma and cortex must be verified, followed by tumor enucleation. This step was key in OSS, which correlates with literature recommendations.^14,15

In our study, we have no conversion from laparoscopy to laparotomy, most probably because our selection criteria for laparoscopy proved to be adequate.^14,15 As reported by Guillén et al., one of the selection criteria for laparoscopy was a solid tumor size of around 5 cm; however, bigger solid lesions require a laparotomy. ¹⁵

The reported rate of oophorectomy and/or salpingo-oophorectomy was from 50.5% to 58%, which is higher compared with 1 oophorectomy and 8 salpingo-oophorectomies (23.6%) in our study.^8–10 We succeeded in preserving the ovaries in 27 benign tumors (87%) compared with 93% reported in a study by Stanković et al. ⁷ In the study by Hernon et al., oophorectomies were performed for all patients with malignancy, similar to our study. ¹⁰ Salpingo-oophorectomy was performed for one borderline ovarian tumor and two borderline ovarian tumors were treated with cystectomy in our study. Oophorectomy was also performed in 75 (48%) cases for benign tumors or functional cysts, which is higher compared with 1 oophorectomy (for adult teratoma) and 3 salpingo-oophorectomies (10.5%) (for luteal hemorrhagic cysts) in our study. ¹⁰

It is important to differentiate between malignant and benign adnexal tumors to ensure optimal surgical management. Solid/complex tumors in the younger population are more likely to be malignant, so oophorectomy is appropriate. OSS/cystectomy is a preferred approach when the mass is likely to represent a benign entity.^14–17

Median age at diagnosis in our study was 16.78 years, which correlates with other studies.^5,6,8,9,12

The most common presenting symptom of adnexal tumors in most studies was acute abdominal pain and, consequently, emergency surgery.^5,8–10 However, incidental findings of adnexal tumors on ultrasound are also frequently encountered, with scheduled surgery management.^5,8–10

The average ultrasound size of adnexal tumors was very similar to Bergeron et al.'s study (8.7 versus 8.8, respectively). ⁶ In our study, a larger adnexal tumor size was found in the salpingo-oophorectomy/oophorectomy group than in the OSS group, but without statistical significance.

In Hernon et al.'s study, the tumor marker, CA-125, was evaluated in only 16 (10%) patients, which is lower compared with 12 (31.5%) patients in our study. ¹⁰ A low rate of abnormality in tumor markers in adnexal tumors was confirmed in an Indian study (12.5%), similar to our study (7.8%). ¹³ In our study, we did not include all tumor markers (alpha-fetoprotein, beta-human chorionic gonadotropin, carcinoembryonic antigen, inhibin B, and lactate dehydrogenase) during preoperative studies because we are a developing and poor country. We evaluated only those tumor markers that our institution always has at its disposal because (occasionally) we do not have reagents for all the necessary tumor markers. Another reason is that it is a large number of emergency surgical procedures, and it was not possible to wait for the results of tumor markers.

Studies further confirmed that the use of CA-125 in addition to ultrasound findings does not improve the diagnostic accuracy in young patients, which justifies such a low usage of tumor markers in our study. CA-125 is a marker for epithelial neoplasms; however, in younger populations, neoplasms tend not to be epithelial.^2,7 However, nowadays, preoperative assessment of tumor markers should be mandatory in scheduled and emergency surgeries for diagnostics and later for follow-up and academic purposes, as reported by Guillén et al. ¹⁵ It is of academic importance to have all markers tested before scheduled and emergency OSS for later follow-up levels of markers and success of OSS in different adnexal masses.

In a review of 67 cases, Schultz et al. found 93% of unilateral tumors, which correlates with our study. They reported dominant right-sided tumors, in contrast to our left-sided tumors. ¹ This is an important finding in differential diagnosis between acute appendicitis and torsed adnexa, which is frequent in PAG.

Hassan et al. reported 77.2% benign; 15.8% malignant, and 7% borderline tumors, which is a distribution similar to our data. ¹⁸ Hernon et al. reported 9 paraovarian (5.8%) and 4 hemorrhagic (2.5%) cysts, which are lower than in our study with 8 (21.05%) and 6 (15.7%), respectively. ¹⁰ The study by Hermans et al. reported 40.5% of functional cysts, 34.2% of benign tumors, and 25.2% of malignant tumors, similar to our data. ⁹

A retrospective study by Major et al. reported similar proportions of malignant epithelial and granulosa cell tumors compared with our study. ¹⁹ They performed one total hysterectomy and bilateral salpingo-oophorectomy for an epithelial tumor FIGO III and three unilateral salpingo-oophorectomies, similar to all three unilateral salpingo-oophorectomies in our study. ¹⁹ All cases of malignancy in the analyzed studies underwent laparotomy, and this surgical approach is recommended when preoperative findings suggest a high risk of malignancy.^5,16,17,19

Adjuvant chemotherapy was administered to all patients in both studies. ¹⁹ Hernon et al. and Bristow et al. reported 10 (6.4%) and 12 (14.6%) malignant tumors, respectively, which is similar to our study, 4 (10.5%).^8,10 Hermans et al. reported a higher incidence of granulosa cell tumors, which was 21% of all malignancies, similar to our study (25%), whereas in previous studies, sex cord stromal tumors represented 5%–8% of all malignancies. ⁹

A surgical dilemma for adnexal tumors of unknown pathology (solid/complex) is the risk of insufficient or aggressive access. ¹⁰ Possible surgical errors include the preservation of ovarian tissue in malignant tumors or oophorectomy in benign ovarian tumors. ¹⁰

Schultz et al. noted 2 deaths (2.9%) in patients in their study that included 37 malignancies, which is lower when compared with 2 deaths (5.2%) of 38 malignancies in our study. ¹

When reading the recent literature, our clinic seems to struggle with the same difficulties experienced by other medical centers treating adnexal tumors in the PAG group because adnexal tumors in the PAG group are operated upon in pediatric clinics, sometimes without the presence of a gynecologist.^9,10,16,20 A key factor in improving the treatment of adolescents with adnexal tumors and eliminating unnecessary oophorectomies beyond adequate preoperative diagnostics is to increase the number of educated and experienced pediatric gynecologists in hospitals.^10,16 In our developing country, there is no specialization or subspecialization in pediatric and adolescent gynecology.

There are several limitations that should be addressed in association with this study: its retrospective design, a small number of patients, and single-center design. However, the low incidence of adnexal tumors in children and adolescents is somewhat prohibitive for this study design. A minority of patients underwent preoperative magnetic resonance imaging or evaluation of tumor markers before surgery.

Despite limitations associated with this study, we have reported several interesting findings. The results of our study should serve as a basis for defining appropriate surgical treatment of adnexal tumors in PAG. Our study confirms the findings of previous researchers that the majority of adnexal tumors were of benign etiology.^4,16,20 Strengths of this study include its long period of recruitment and the surgeries were performed by 2 general gynecologists with PAG as a special field of interest.

Conclusion

We expect this study to be helpful in raising awareness of the importance of distinguishing between benign and malignant adnexal tumors, assisting gynecologists in preserving more ovarian tissue during surgery of adnexal tumors and torsion, and in reducing the rate of surgical management of uncomplicated ovarian cysts.

Our study shows a high trend of OSS using the endoscopic approach in management of adnexal tumors, despite the fact that management was done by general gynecologists with PAG as the special field of interest. However, the practice of removing ovaries for benign adnexal tumors may be overcome by establishing a specialization or subspecialization, especially for young gynecologists, or appointing more gynecologists with advanced training and skills in PAG.

IRB Approval

Approval for this study was obtained from the institutional review board. All procedures performed on human participants were in accordance with the ethical standards of the institutional and national research committees and with the 1964 Declaration of Helsinki and its later amendments or comparable ethical standards.