- Email: [email protected]
Ovarian-Sparing Surgery in Pediatric Benign Ovarian Tumors
Original Study Ovarian-Sparing Surgery in Pediatric Benign Ovarian Tumors Paulette I. Abbas MD 1, Jennifer E. Dietrich MD 2, Jessica A. Francis MD 2, Mary L. Brandt MD 1, Darrell L. Cass MD 1, Monica E. Lopez MD 1,* 1 2
Division of Pediatric Surgery, Michael E. DeBakey Department of Surgery, Baylor College of Medicine, Houston, Texas Division of Pediatric and Adolescent Gynecology, Department of Obstetrics and Gynecology, Baylor College of Medicine, Houston, Texas
a b s t r a c t Study Objective: To evaluate outcomes of children after ovarian-sparing surgery (OSS) for non-neoplastic and benign neoplastic ovarian lesions. Design: Retrospective cohort study from January 2003 to January 2012. Setting: Single, high-volume, tertiary care hospital. Participants: Children 18 years of age and younger. Interventions: None. Main Outcome Measures: Postoperative complications and tumor recurrence after OSS. Results: One hundred nine patients underwent OSS with a median age of 13.3 years (interquartile range [IQR], 11.4-15.1 years). Eighty-two patients were treated laparoscopically with 4 conversions to an open procedure. Postoperative complications included surgical site infections in 7 patients (6%). Pathology most commonly revealed functional ovarian cysts (n 5 57) and mature teratomas (n 5 37). Ninetyfour patients (86%) were followed for a median of 10.4 months (IQR, 0.72-30.8 months). Fifty-five patients (60%) had subsequent imaging surveillance a median of 7.6 months postoperatively (IQR, 3.9-13 months). Ten patients (10%) developed a second ipsilateral lesion within a median time of 11 months (IQR, 7.7-24 months), of whom 5 girls had repeated surgery for mass enlargement or persistent abdominal pain at a median time of 10.5 months (IQR, 8.0-12.65 months). Fifty-eight patients (63%) began or resumed menses at their most recent followup. Three girls became pregnant after OSS at a median follow-up of 5 years (range, 2.4-6.7 years). Conclusion: Benign ovarian lesions in children can be treated successfully with OSS with low recurrence and repeat surgery rates. Key Words: Pediatric ovarian tumors, Ovarian-sparing surgery, Fertility, Adolescent gynecology, Pediatric surgery
Introduction
Ovarian lesions are frequently incidentally identified in pediatric patients during the imaging evaluation for nonovarian etiologies of abdominal pain.1 Commonly, ultrasonographic characteristics can differentiate ovarian neoplasms from cystic lesions. Although ovarian tumors only comprise 1% of childhood cancers, neoplastic lesions carry a 10% risk of malignancy.2e7 Non-neoplastic cystic lesions are benign and generally asymptomatic, thus they are monitored for spontaneous resolution; however, surgery might be warranted for cystic lesions that persist or grow to become symptomatic. Common complications of cystic lesions include rupture, hemorrhage, or torsion with large cysts.8e10 Alternately, benign neoplastic lesions, such as mature teratomas, require surgical intervention as the primary method of therapy. The approach for the resection of these benign lesions was the focus of this study. The authors indicate no conflicts of interest. Dr Paulette I. Abbas and Dr Monica E. Lopez conceptualized and designed the study, drafted the initial manuscript, reviewed and revised the manuscript and approved the final manuscript as submitted. Dr Jennifer E. Dietrich, Dr Jessica A. Francis, Dr Mary L. Brandt, and Dr Darrell L. Cass reviewed and revised the manuscript for critical intellectual content and approved the final manuscript as submitted. * Address correspondence to: Monica E. Lopez, MD, Texas Children's Hospital, 6701 Fannin St, Suite 1210, Houston, TX 77030; Phone: (832) 822-3135 E-mail address: [email protected] (M.E. Lopez).
Currently, the conservative and minimally invasive approach of ovarian-sparing surgery (OSS) has gained popularity over oophorectomy for benign ovarian disease, mainly for its purported fertility-sparing benefits.5,11,12 Among women who seek treatment at infertility clinics, the proportion of women with a single ovary is higher than women with both ovaries, drawing attention to the importance of preservation of the ovary.13 However, despite growing concerns about future fertility, the adoption of OSS for benign disease in the pediatric patient still lags behind. According to a retrospective review of the Perspective nationwide database in 2012 by Berger-Chen et al, 40% of adolescent patients continue to undergo oophorectomy for benign ovarian lesions.6 This delay in the incorporation of the ovarian-sparing approach might be partly because of concerns of incomplete resection and potential recurrence of the lesion.14 In our study we aimed to analyze the clinical outcomes of pediatric patients treated for benign ovarian disease with OSS. Materials and Methods Patient Population
After institutional review board approval (H-31281), pediatric patients (18 years of age or younger) with benign ovarian tumors treated with OSS from January 2003 to
1083-3188/$ - see front matter Ó 2016 North American Society for Pediatric and Adolescent Gynecology. Published by Elsevier Inc. http://dx.doi.org/10.1016/j.jpag.2016.03.009
P.I. Abbas et al. / J Pediatr Adolesc Gynecol 29 (2016) 506e510
January 2012 were identified using Current Procedure Terminology codes for laparoscopic and open ovarian cystectomy. All surgeries were performed at a single tertiary care children's hospital. Benign etiology was confirmed through pathology reports. Patients treated by pediatric gynecologists or pediatric surgeons were included. Patients with malignant pathology or those who underwent oophorectomy were excluded. Study Design and Clinical Variables
The study was a retrospective chart review. We collected demographic data such as patient age at the time of surgery and menarchal status. Collection of operative details included surgical technique (open vs laparoscopic), intraoperative spillage of fluid from the mass, and documentation of incomplete resection. Incomplete resection was defined as a remnant of the cyst wall left in situ secondary to difficulty removing the wall without damaging remaining ovarian tissue. Maximum cyst size was collected from standard preoperative sonographic measurements and final pathology was also evaluated. Data on tumor recurrence were also collected. Recurrence was defined as an ipsilateral development of a second lesion with the same pathology. Patients were stratified according to age categories: infants (younger than 1 year), children (1-9 years), and adolescents (9.1-18 years). This differentiation is important because various age groups are susceptible to different ovarian tumor etiologies on the basis of menarche status and hormonal levels.9 Additionally, patients were stratified according to pathology because underlying etiology for recurrence might differ between functional cyst and benign neoplastic lesions. Lastly, resumption of menses after OSS was evaluated as a marker of ovarian function. Statistical Analyses
Statistical analysis was performed using SPSS (version 22.0; SPSS Inc, Armonk, NY). Frequency distributions between categorical values were compared using c2. Summary statistics were presented as frequency and proportions for categorical variables. Continuous data are presented as median with interquartile range (IQR). Univariate analyses were performed on the overall cohort to evaluate individual characteristics for an association with lesion recurrence. Factors associated with recurrence (age, cyst size, menarchal status, laparoscopic approach, intraoperative spillage, incomplete resection)5,12,15e18 were included as covariates in a multivariate regression model to evaluate predictive factors of recurrence. Additionally, because recent literature has pointed to increased recurrence with a cyst diameter greater than 8 cm,19 this threshold was analyzed in our cohort. Patients were then stratified according to age and pathology and further analyzed with similar univariate and multivariate regressions. A P value of ! .05 was determined to be statistically significant.
507
Results Patient Population
During the study period, 143 patients underwent surgery for benign ovarian lesions; 109 patients (76%) underwent OSS and were ultimately included for analysis. Of note, 3 patients who underwent primary OSS for presumed benign disease were ultimately diagnosed with malignant pathology and excluded from analysis. The median age at time of surgery for included patients was 13.3 years (IQR, 11.4-15.1 years). When stratified according to age, there were 5 infants, 8 children, and 96 adolescents. Eighty-two patients were treated laparoscopically with 4 conversions to open technique because of anatomic aberrations or difficulty with safely completing the dissection laparoscopically. Twenty-seven patients underwent an open procedure as their primary procedure. Twelve patients (11%) had documented intraoperative spillage of cyst contents and 10 patients (9%) had incomplete resection. Postoperative complication of superficial and deep/organ surgical site infections, as defined by the American College of Surgeons National Surgical Quality Improvement Program (ACS NSQIP), occurred in 7 patients (6%). Pathology most commonly revealed functional ovarian cysts (n 5 57 patients) and mature teratomas (n 5 37 patients). The remaining 15 benign neoplastic lesions included: 8 serous cystadenomas, 4 mucinous cystadenomas, 2 cystadenofibromas, and 1 papillary cystadenoma. Pathology stratified according to age is shown in Figure 1. Ninety-four patients (86%) were followed postoperatively for a median of 10.4 months (IQR, 0.72-30.8 months). Fifty-five patients (60%) had subsequent imaging surveillance at a median time of 7.6 months postoperatively (IQR, 3.9-13 months; Fig. 2). Of the 55 patients with repeat imaging, 25 patients had an initial diagnosis of benign neoplastic lesion and the remaining 30 patients had functional cysts. Ten patients (18%) developed a second ipsilateral lesion detected on follow-up imaging. The median time to presence of a
Fig. 1. Distribution of ovarian lesion pathology stratified by age.Ă
508
P.I. Abbas et al. / J Pediatr Adolesc Gynecol 29 (2016) 506e510
Fig. 2. Flow diagram detailing study participant follow-up.Ă
second lesion was 11 months (IQR, 7.7-24 months). This recurrence rate was significantly different between benign neoplastic lesions (n 5 1) compared with non-neoplastic lesions (n 5 9; 4% vs 30%; P 5 .013). Five of the patients (50%) who developed an ipsilateral second lesion underwent repeated surgery for mass enlargement or persistent abdominal pain with a median time of 10.5 months (IQR, 8.0-12.65 months).
independent predictor of recurrence among all-comers (Table 1). When assessing the lesion size threshold of 8 cm, there was no difference in recurrence rates. Patients with masses less than 8 cm (n 5 29) had a 17% recurrence rate compared with the 19% recurrence rate in the 26 patients with lesions greater than 8 cm (P 5 .849).
Overall Ovarian Lesion Recurrence
Although the 9 adolescent patients with recurrence were more likely to have incomplete resection, other factors such as age, menarchal status, surgical approach, and tumor spillage did not significantly affect recurrence (Table 2). After adjusting for other variables, incomplete resection remained an independent predictor of developing a second lesion (Table 2). Laparoscopic approach was not included in the regression for adolescents because there were not enough data on open resection for comparison. When further stratifying the recurrence rates in adolescent
In univariate analysis, the only feature found to be associated with overall development of a second ovarian lesion was documentation of incomplete lesion removal (Table 1). There was no difference in ability to perform a complete resection regardless of surgical approach (laparoscopy, 90.2% vs open, 91.3%; P 5 .878). Nevertheless, in multivariate regression analysis adjusting for all clinically relevant features, incomplete resection was not an
Overall Ovarian Lesion Recurrence in Adolescents
Table 1 Univariate and Multivariate Analyses of Predictors for Development of a Second Ipsilateral Ovarian Lesion in All Patients Variable
Univariate Analyses Second Lesion (n 5 10)
Median age (IQR), years Median largest cyst diameter (IQR), cm Laparoscopic approach, n (%) Postmenarchal, n (%) Intraoperative spillage, n (%) Incomplete resection, n (%) Bold indicates statistical significance. IQR, interquartile range.
12.1 7.5 9 8 2 3
(10.9-14.0) (5.9-10.3) (90) (80) (20) (30)
No second Lesion (n 5 45) 13.2 7.3 34 364 7 3
(11.2-15.0) (5.4-12.0) (76) (80) (16) (6.8)
Multivariate Analyses P
Adjusted Hazard Ratio
95% Confidence Interval
.348 .974 .317 1.0 .754 .035
0.89 0.978 1.6 2.5 1.9 5.6
0.67-1.2 0.77-1.2 0.15-16.3 0.2-32.1 0.28-12.9 0.77-40
P.I. Abbas et al. / J Pediatr Adolesc Gynecol 29 (2016) 506e510
509
Table 2 Univariate and Multivariate Analysis of Predictors for Developing an Ipsilateral Second Ovarian Lesion in Adolescent (8-18 years) Patients Variable
Univariate Analyses Second Lesion (n 5 9)
Median age (IQR), years Median largest cyst diameter (IQR), cm Laparoscopic approach, n (%) Postmenarchal, n (%) Intraoperative spillage, n (%) Incomplete resection, n (%)
12.3 8.1 9 8 2 3
(10.7-13.6) (5.4-10.3) (100) (89) (22) (33)
No Second Lesion (n 5 43) 13.2 7.3 32 36 7 2
(11.7-15.2) (5.4-12.6) (74) (84) (16) (5)
Multivariate Analyses P
Adjusted Hazard Ratio
95% Confidence Interval
.475 .775 .087 .696 .692 .009
0.77 0.97 * 4.6 2.8 14.6
0.48-1.2 0.79-1.2 * 0.29-73.6 0.36-21.2 1.6-134.4
Bold indicates statistical significance. IQR, interquartile range. * Unable to calculate secondary to insufficient data on open approach in this cohort.
patients according to pathology of the lesion, age and incomplete resection remained significant for adolescent patients with functional cysts. Incomplete cyst resection remained an independent predictor of functional cyst recurrence in adolescent patients when adjusted for other variables (adjusted hazard ratio: 27.5; 95% confidence interval, 1.09-694). No features were predictive of recurrent benign neoplastic lesions in adolescents. Recurrence of Overall Ovarian Lesions Stratified According to Pathology
Of the 57 patients with functional cysts, 30 received a follow-up ultrasound examination. Patients with functional cysts who developed a second lesion (n 5 9) were younger than their counterparts who did not (11.9 years vs 13.8 years; P 5 .045). However, after adjustment for other clinical features, age was no longer an independent predictor (adjusted hazard ratio, 0.86; 95% confidence interval, 0.586-1.3). When the 26 patients with benign neoplastic lesions who received imaging follow-up were analyzed independently, only 1 patient developed a recurrent ipsilateral teratoma. No significant factors were identified in univariate analyses in determining the development of a second teratoma. Multivariate regression was not performed for benign neoplastic lesions because only 1 patient developed a recurrent lesion in our cohort. Resumption of Menses
Of the 86 adolescent patients followed clinically, 58 (67 %) of the patients had documented initiation or resumption of their menses at the time of most recent follow-up. Three patients became pregnant after undergoing OSS with a median follow-up of 5 years (range, 2.4-6.7 years) from surgery to conception. Discussion
Among a typical population of pediatric patients with benign ovarian lesions treated at a single center, the ovarian-sparing surgical approach conferred successful clinical outcomes of low recurrence and repeated surgery rates. Results from this study adds to the body of evidence showing that OSS is a viable approach in children and adolescents with benign ovarian lesions in an attempt to
preserve future fertility.3,5e7,11,12 In this large cohort of 109 patients with ovarian lesions ranging from 3 to 27 centimeters, OSS was performed successfully with few operative complications. We showed that the laparoscopic approach is feasible for most lesions, because 70% of the patients were able to undergo laparoscopic excision of the lesion without the need to convert to an open procedure. Although OSS has been advocated for benefits of fertility preservation, studies have shown that surgeons continue to perform oophorectomies for benign lesions. Large mass size5 and inability to separate the tumor from normal ovarian tissue12 precluding complete resection have been cited as potential factors in the decision to perform an oophorectomy. To investigate the concerns for recurrence with the use of OSS, we examined rates of developing a second ipsilateral lesion in this patient population. We found an overall ipsilateral recurrence rate of 10% (10 patients) with 9 recurrences occurring in functional cysts and only 1 recurrent benign teratoma; only 5% (5 patients) required repeated surgery. This appears to be lower than the 30.3% recurrence rate reported in the adult literature after OSS.14 Furthermore, the observed maximal cyst size was not significantly different between the cohort of patients who developed a second lesion and those who did not, irrespective of underlying pathology. Interestingly, the lack of complete resection was not associated with laparoscopic approach, as was previously described in literature.16 Additionally, when applying the threshold of 8 cm for recurrence determined by Harada et al19 to our cohort, we found no difference in recurrence rate. Both of these findings suggest that a large lesion size should not preclude the attempt of an OSS or a laparoscopic approach. Although the recurrence rate might be underestimated, because longterm follow-up was not achieved in all patients, we had a large percentage of our patients (86%) return to our institution for follow-up. In further identifying potential predictive factors of recurrence, patients were stratified according to age categories and pathology. Adolescents were found to have a higher rate of recurrence with incomplete resection and this remained an independent predictor of recurrence among adolescents. No predictors for recurrence were seen in our cohort of children or neonates because of the small sample size within these age groups. When the adolescent cohort was further stratified according to pathology, incomplete resection was again found to be associated with
510
P.I. Abbas et al. / J Pediatr Adolesc Gynecol 29 (2016) 506e510
recurrence in patients with functional cysts but not for those with benign neoplastic lesions. Furthermore, younger age at the time of surgery was associated with the development of recurrent functional cysts. This increased recurrence of functional cysts in younger and potentially early menarchal girls might reflect hormonal fluctuations and dysfunctional ovulation from puberty that increases the likelihood of developing functional cysts.9,20 Conversely, in the benign neoplastic cohort, no patient characteristics or operative detail, including incomplete resection or intraoperative spillage, was associated with the 1 recurrent teratoma. Although resumption of menses is often reported as a clinical outcome because it might act as a surrogate for ovarian function, it has an unknown implication in future fertility. In adults, a study after menstrual cycles after conservative ovarian surgery confirmed no significant effects of the surgery on ovulatory function, menstrual function, or conception rates.21 We examined these factors in our patient cohort and found similar results. Most patients of pubertal age resumed or began menstruation postoperatively; 3 patients became pregnant. Longer follow-up is required to determine the utility of menstruation as a predictor of fertility. Limitations of our study include its retrospective nature and lack of uniform follow-up on all patients, which could bias the interpretation of the findings related to recurrence and repeat surgery rates. Because of the inherent variation in practice within our institution with regard to surveillance imaging in patients with a history of ovarian lesions, detection bias could also influence some of our findings. To ensure accurate follow-up information, we carefully screened medical record notes pertaining to all clinical encounters for the ovarian lesion or for abdominal complaints. Nevertheless, this highlights an opportunity to standardize our follow-up protocol and to study it prospectively. Furthermore, our use of resumption of menses as a metric to delineate the restoration or preservation of ovarian function might be of unknown clinical significance because laboratory values confirming ovarian function were not explicitly measured in our study. Additionally, we had a small number of patients with benign neoplastic lesions who developed a second ipsilateral lesion, thus limiting our ability to determine predictors of recurrence in this group. Last, we were unable to differentiate between a true recurrent lesion or the presence of a metachronous lesion on the ipsilateral ovary. Despite these limitations, this study adds to the pediatric surgical literature by providing data
from a large cohort of patients who successfully underwent OSS for various types of benign lesions. Future studies are needed to validate the finding of higher recurrence of functional cysts in adolescent patients. Conclusions
OSS is a feasible surgical approach that should be considered for all pediatric patients with benign ovarian tumors, because it can attain low recurrence and repeat surgery rates for benign neoplastic lesions. References 1. Helmrath M, Shin C, Warner B: Ovarian cysts in the pediatric population. Semin Pediatr Surg 1998; 7:19 2. Zolton J, Maseelall P: Evaluation of ovarian cysts in adolescents. Open J Obstet Gynecol 2013; 3:12 3. Cass D, Hawkins E, Brandt M, et al: Surgery for ovarian masses in infants, children, and adolescents: 102 consecutive patients treated in a 15-year period. J Pediatr Surg 2001; 36:693 4. Oltmann S, Garcia N, Barber R, et al: Can we preoperatively risk stratify ovarian masses for malignancy? J Pediatr Surg 2010; 45:130 5. Ozcan R, Kuruoglu S, Dervisoglu S, et al: Ovary-sparing surgery for teratomas in children. Pediatr Surg Int 2013; 29:233 6. Berger-Chen S, Herzog T, Lewin S, et al: Access to conservative surgical therapy for adolescents with benign ovarian masses. Obstet Gynecol 2012; 119:270 7. Brookfield K, Cheung M, Koniaris L, et al: A population-based analysis of 1037 malignant ovarian tumors in the pediatric population. J Surg Res 2009; 156:45 8. Akkoyun I, Gulen S: Laparoscopic cystectomy for the treatment of benign ovarian cysts in children: an analysis of 21 cases. J Pediatr Adolesc Gynecol 2012; 25:364 9. Strickland J: Ovarian cysts in neonates, children and adolescents. Curr Opin Obstet Gynecol 2002; 14:459 10. Cass D: Ovarian torsion. Semin Pediatr Surg 2005; 14:86 11. Abdel-Hady el-S, Abdel-Hady R, Gamal A, et al: Fertility sparing surgery for ovarian tumors in children and young adults. Arch Gynecol Obstet 2012; 285: 469 12. Chabaud-Williamson M, Netchine I, Fasola S, et al: Ovarian-sparing surgery for ovarian teratoma in children. Pediatr Blood Cancer 2011; 57:429 13. Lass A: The fertility potential of women with a single ovary. Hum Reprod Update 1999; 5:546 14. Poncelet C, Fauvet R, Boccara J, et al: Recurrence after cystectomy for borderline ovarian tumors: results of a French multicenter study. Ann Surg Oncol 2006; 13:565 15. Laberge P, Levesque S: Short-term morbidity and long-term recurrence rate of ovarian dermoid cysts treated by laparoscopy versus laparotomy. J Obstet Gynaecol Can 2006; 28:789 16. Lin P, Falcone T, Tulandi T: Excision of ovarian dermoid cyst by laparoscopy and by laparotomy. Am J Obstet Gynecol 1995; 173:769 17. Rogers E, Allen L, Kives S: The recurrence rate of ovarian dermoid cysts in pediatric and adolescent girls. J Pediatr Adolesc Gynecol 2014; 27:222 18. Savasi I, Lacy J, Gerstle J, et al: Management of ovarian dermoid cysts in the pediatric and adolescent population. J Pediatr Adolesc Gynecol 2009; 22:360 19. Harada M, Osuga Y, Fujimoto A, et al: Predictive factors for recurrence of ovarian mature cystic teratomas after surgical excision. Eur J Obstet Gynecol Reprod Biol 2013; 171:325 20. Kanizsai B, Orley J, Szigetvari I, et al: Ovarian cysts in children and adolescents: their occurrence, behavior, and management. J Pediatr Adolesc Gynecol 1998; 11:85 21. Sayegh R, Garcia C: Ovarian function after conservational ovarian surgery: a long-term follow-up study. Int J Gynaecol Obstet 1992; 39:303
Division of Pediatric Surgery, Michael E. DeBakey Department of Surgery, Baylor College of Medicine, Houston, Texas Division of Pediatric and Adolescent Gynecology, Department of Obstetrics and Gynecology, Baylor College of Medicine, Houston, Texas
a b s t r a c t Study Objective: To evaluate outcomes of children after ovarian-sparing surgery (OSS) for non-neoplastic and benign neoplastic ovarian lesions. Design: Retrospective cohort study from January 2003 to January 2012. Setting: Single, high-volume, tertiary care hospital. Participants: Children 18 years of age and younger. Interventions: None. Main Outcome Measures: Postoperative complications and tumor recurrence after OSS. Results: One hundred nine patients underwent OSS with a median age of 13.3 years (interquartile range [IQR], 11.4-15.1 years). Eighty-two patients were treated laparoscopically with 4 conversions to an open procedure. Postoperative complications included surgical site infections in 7 patients (6%). Pathology most commonly revealed functional ovarian cysts (n 5 57) and mature teratomas (n 5 37). Ninetyfour patients (86%) were followed for a median of 10.4 months (IQR, 0.72-30.8 months). Fifty-five patients (60%) had subsequent imaging surveillance a median of 7.6 months postoperatively (IQR, 3.9-13 months). Ten patients (10%) developed a second ipsilateral lesion within a median time of 11 months (IQR, 7.7-24 months), of whom 5 girls had repeated surgery for mass enlargement or persistent abdominal pain at a median time of 10.5 months (IQR, 8.0-12.65 months). Fifty-eight patients (63%) began or resumed menses at their most recent followup. Three girls became pregnant after OSS at a median follow-up of 5 years (range, 2.4-6.7 years). Conclusion: Benign ovarian lesions in children can be treated successfully with OSS with low recurrence and repeat surgery rates. Key Words: Pediatric ovarian tumors, Ovarian-sparing surgery, Fertility, Adolescent gynecology, Pediatric surgery
Introduction
Ovarian lesions are frequently incidentally identified in pediatric patients during the imaging evaluation for nonovarian etiologies of abdominal pain.1 Commonly, ultrasonographic characteristics can differentiate ovarian neoplasms from cystic lesions. Although ovarian tumors only comprise 1% of childhood cancers, neoplastic lesions carry a 10% risk of malignancy.2e7 Non-neoplastic cystic lesions are benign and generally asymptomatic, thus they are monitored for spontaneous resolution; however, surgery might be warranted for cystic lesions that persist or grow to become symptomatic. Common complications of cystic lesions include rupture, hemorrhage, or torsion with large cysts.8e10 Alternately, benign neoplastic lesions, such as mature teratomas, require surgical intervention as the primary method of therapy. The approach for the resection of these benign lesions was the focus of this study. The authors indicate no conflicts of interest. Dr Paulette I. Abbas and Dr Monica E. Lopez conceptualized and designed the study, drafted the initial manuscript, reviewed and revised the manuscript and approved the final manuscript as submitted. Dr Jennifer E. Dietrich, Dr Jessica A. Francis, Dr Mary L. Brandt, and Dr Darrell L. Cass reviewed and revised the manuscript for critical intellectual content and approved the final manuscript as submitted. * Address correspondence to: Monica E. Lopez, MD, Texas Children's Hospital, 6701 Fannin St, Suite 1210, Houston, TX 77030; Phone: (832) 822-3135 E-mail address: [email protected] (M.E. Lopez).
Currently, the conservative and minimally invasive approach of ovarian-sparing surgery (OSS) has gained popularity over oophorectomy for benign ovarian disease, mainly for its purported fertility-sparing benefits.5,11,12 Among women who seek treatment at infertility clinics, the proportion of women with a single ovary is higher than women with both ovaries, drawing attention to the importance of preservation of the ovary.13 However, despite growing concerns about future fertility, the adoption of OSS for benign disease in the pediatric patient still lags behind. According to a retrospective review of the Perspective nationwide database in 2012 by Berger-Chen et al, 40% of adolescent patients continue to undergo oophorectomy for benign ovarian lesions.6 This delay in the incorporation of the ovarian-sparing approach might be partly because of concerns of incomplete resection and potential recurrence of the lesion.14 In our study we aimed to analyze the clinical outcomes of pediatric patients treated for benign ovarian disease with OSS. Materials and Methods Patient Population
After institutional review board approval (H-31281), pediatric patients (18 years of age or younger) with benign ovarian tumors treated with OSS from January 2003 to
1083-3188/$ - see front matter Ó 2016 North American Society for Pediatric and Adolescent Gynecology. Published by Elsevier Inc. http://dx.doi.org/10.1016/j.jpag.2016.03.009
P.I. Abbas et al. / J Pediatr Adolesc Gynecol 29 (2016) 506e510
January 2012 were identified using Current Procedure Terminology codes for laparoscopic and open ovarian cystectomy. All surgeries were performed at a single tertiary care children's hospital. Benign etiology was confirmed through pathology reports. Patients treated by pediatric gynecologists or pediatric surgeons were included. Patients with malignant pathology or those who underwent oophorectomy were excluded. Study Design and Clinical Variables
The study was a retrospective chart review. We collected demographic data such as patient age at the time of surgery and menarchal status. Collection of operative details included surgical technique (open vs laparoscopic), intraoperative spillage of fluid from the mass, and documentation of incomplete resection. Incomplete resection was defined as a remnant of the cyst wall left in situ secondary to difficulty removing the wall without damaging remaining ovarian tissue. Maximum cyst size was collected from standard preoperative sonographic measurements and final pathology was also evaluated. Data on tumor recurrence were also collected. Recurrence was defined as an ipsilateral development of a second lesion with the same pathology. Patients were stratified according to age categories: infants (younger than 1 year), children (1-9 years), and adolescents (9.1-18 years). This differentiation is important because various age groups are susceptible to different ovarian tumor etiologies on the basis of menarche status and hormonal levels.9 Additionally, patients were stratified according to pathology because underlying etiology for recurrence might differ between functional cyst and benign neoplastic lesions. Lastly, resumption of menses after OSS was evaluated as a marker of ovarian function. Statistical Analyses
Statistical analysis was performed using SPSS (version 22.0; SPSS Inc, Armonk, NY). Frequency distributions between categorical values were compared using c2. Summary statistics were presented as frequency and proportions for categorical variables. Continuous data are presented as median with interquartile range (IQR). Univariate analyses were performed on the overall cohort to evaluate individual characteristics for an association with lesion recurrence. Factors associated with recurrence (age, cyst size, menarchal status, laparoscopic approach, intraoperative spillage, incomplete resection)5,12,15e18 were included as covariates in a multivariate regression model to evaluate predictive factors of recurrence. Additionally, because recent literature has pointed to increased recurrence with a cyst diameter greater than 8 cm,19 this threshold was analyzed in our cohort. Patients were then stratified according to age and pathology and further analyzed with similar univariate and multivariate regressions. A P value of ! .05 was determined to be statistically significant.
507
Results Patient Population
During the study period, 143 patients underwent surgery for benign ovarian lesions; 109 patients (76%) underwent OSS and were ultimately included for analysis. Of note, 3 patients who underwent primary OSS for presumed benign disease were ultimately diagnosed with malignant pathology and excluded from analysis. The median age at time of surgery for included patients was 13.3 years (IQR, 11.4-15.1 years). When stratified according to age, there were 5 infants, 8 children, and 96 adolescents. Eighty-two patients were treated laparoscopically with 4 conversions to open technique because of anatomic aberrations or difficulty with safely completing the dissection laparoscopically. Twenty-seven patients underwent an open procedure as their primary procedure. Twelve patients (11%) had documented intraoperative spillage of cyst contents and 10 patients (9%) had incomplete resection. Postoperative complication of superficial and deep/organ surgical site infections, as defined by the American College of Surgeons National Surgical Quality Improvement Program (ACS NSQIP), occurred in 7 patients (6%). Pathology most commonly revealed functional ovarian cysts (n 5 57 patients) and mature teratomas (n 5 37 patients). The remaining 15 benign neoplastic lesions included: 8 serous cystadenomas, 4 mucinous cystadenomas, 2 cystadenofibromas, and 1 papillary cystadenoma. Pathology stratified according to age is shown in Figure 1. Ninety-four patients (86%) were followed postoperatively for a median of 10.4 months (IQR, 0.72-30.8 months). Fifty-five patients (60%) had subsequent imaging surveillance at a median time of 7.6 months postoperatively (IQR, 3.9-13 months; Fig. 2). Of the 55 patients with repeat imaging, 25 patients had an initial diagnosis of benign neoplastic lesion and the remaining 30 patients had functional cysts. Ten patients (18%) developed a second ipsilateral lesion detected on follow-up imaging. The median time to presence of a
Fig. 1. Distribution of ovarian lesion pathology stratified by age.Ă
508
P.I. Abbas et al. / J Pediatr Adolesc Gynecol 29 (2016) 506e510
Fig. 2. Flow diagram detailing study participant follow-up.Ă
second lesion was 11 months (IQR, 7.7-24 months). This recurrence rate was significantly different between benign neoplastic lesions (n 5 1) compared with non-neoplastic lesions (n 5 9; 4% vs 30%; P 5 .013). Five of the patients (50%) who developed an ipsilateral second lesion underwent repeated surgery for mass enlargement or persistent abdominal pain with a median time of 10.5 months (IQR, 8.0-12.65 months).
independent predictor of recurrence among all-comers (Table 1). When assessing the lesion size threshold of 8 cm, there was no difference in recurrence rates. Patients with masses less than 8 cm (n 5 29) had a 17% recurrence rate compared with the 19% recurrence rate in the 26 patients with lesions greater than 8 cm (P 5 .849).
Overall Ovarian Lesion Recurrence
Although the 9 adolescent patients with recurrence were more likely to have incomplete resection, other factors such as age, menarchal status, surgical approach, and tumor spillage did not significantly affect recurrence (Table 2). After adjusting for other variables, incomplete resection remained an independent predictor of developing a second lesion (Table 2). Laparoscopic approach was not included in the regression for adolescents because there were not enough data on open resection for comparison. When further stratifying the recurrence rates in adolescent
In univariate analysis, the only feature found to be associated with overall development of a second ovarian lesion was documentation of incomplete lesion removal (Table 1). There was no difference in ability to perform a complete resection regardless of surgical approach (laparoscopy, 90.2% vs open, 91.3%; P 5 .878). Nevertheless, in multivariate regression analysis adjusting for all clinically relevant features, incomplete resection was not an
Overall Ovarian Lesion Recurrence in Adolescents
Table 1 Univariate and Multivariate Analyses of Predictors for Development of a Second Ipsilateral Ovarian Lesion in All Patients Variable
Univariate Analyses Second Lesion (n 5 10)
Median age (IQR), years Median largest cyst diameter (IQR), cm Laparoscopic approach, n (%) Postmenarchal, n (%) Intraoperative spillage, n (%) Incomplete resection, n (%) Bold indicates statistical significance. IQR, interquartile range.
12.1 7.5 9 8 2 3
(10.9-14.0) (5.9-10.3) (90) (80) (20) (30)
No second Lesion (n 5 45) 13.2 7.3 34 364 7 3
(11.2-15.0) (5.4-12.0) (76) (80) (16) (6.8)
Multivariate Analyses P
Adjusted Hazard Ratio
95% Confidence Interval
.348 .974 .317 1.0 .754 .035
0.89 0.978 1.6 2.5 1.9 5.6
0.67-1.2 0.77-1.2 0.15-16.3 0.2-32.1 0.28-12.9 0.77-40
P.I. Abbas et al. / J Pediatr Adolesc Gynecol 29 (2016) 506e510
509
Table 2 Univariate and Multivariate Analysis of Predictors for Developing an Ipsilateral Second Ovarian Lesion in Adolescent (8-18 years) Patients Variable
Univariate Analyses Second Lesion (n 5 9)
Median age (IQR), years Median largest cyst diameter (IQR), cm Laparoscopic approach, n (%) Postmenarchal, n (%) Intraoperative spillage, n (%) Incomplete resection, n (%)
12.3 8.1 9 8 2 3
(10.7-13.6) (5.4-10.3) (100) (89) (22) (33)
No Second Lesion (n 5 43) 13.2 7.3 32 36 7 2
(11.7-15.2) (5.4-12.6) (74) (84) (16) (5)
Multivariate Analyses P
Adjusted Hazard Ratio
95% Confidence Interval
.475 .775 .087 .696 .692 .009
0.77 0.97 * 4.6 2.8 14.6
0.48-1.2 0.79-1.2 * 0.29-73.6 0.36-21.2 1.6-134.4
Bold indicates statistical significance. IQR, interquartile range. * Unable to calculate secondary to insufficient data on open approach in this cohort.
patients according to pathology of the lesion, age and incomplete resection remained significant for adolescent patients with functional cysts. Incomplete cyst resection remained an independent predictor of functional cyst recurrence in adolescent patients when adjusted for other variables (adjusted hazard ratio: 27.5; 95% confidence interval, 1.09-694). No features were predictive of recurrent benign neoplastic lesions in adolescents. Recurrence of Overall Ovarian Lesions Stratified According to Pathology
Of the 57 patients with functional cysts, 30 received a follow-up ultrasound examination. Patients with functional cysts who developed a second lesion (n 5 9) were younger than their counterparts who did not (11.9 years vs 13.8 years; P 5 .045). However, after adjustment for other clinical features, age was no longer an independent predictor (adjusted hazard ratio, 0.86; 95% confidence interval, 0.586-1.3). When the 26 patients with benign neoplastic lesions who received imaging follow-up were analyzed independently, only 1 patient developed a recurrent ipsilateral teratoma. No significant factors were identified in univariate analyses in determining the development of a second teratoma. Multivariate regression was not performed for benign neoplastic lesions because only 1 patient developed a recurrent lesion in our cohort. Resumption of Menses
Of the 86 adolescent patients followed clinically, 58 (67 %) of the patients had documented initiation or resumption of their menses at the time of most recent follow-up. Three patients became pregnant after undergoing OSS with a median follow-up of 5 years (range, 2.4-6.7 years) from surgery to conception. Discussion
Among a typical population of pediatric patients with benign ovarian lesions treated at a single center, the ovarian-sparing surgical approach conferred successful clinical outcomes of low recurrence and repeated surgery rates. Results from this study adds to the body of evidence showing that OSS is a viable approach in children and adolescents with benign ovarian lesions in an attempt to
preserve future fertility.3,5e7,11,12 In this large cohort of 109 patients with ovarian lesions ranging from 3 to 27 centimeters, OSS was performed successfully with few operative complications. We showed that the laparoscopic approach is feasible for most lesions, because 70% of the patients were able to undergo laparoscopic excision of the lesion without the need to convert to an open procedure. Although OSS has been advocated for benefits of fertility preservation, studies have shown that surgeons continue to perform oophorectomies for benign lesions. Large mass size5 and inability to separate the tumor from normal ovarian tissue12 precluding complete resection have been cited as potential factors in the decision to perform an oophorectomy. To investigate the concerns for recurrence with the use of OSS, we examined rates of developing a second ipsilateral lesion in this patient population. We found an overall ipsilateral recurrence rate of 10% (10 patients) with 9 recurrences occurring in functional cysts and only 1 recurrent benign teratoma; only 5% (5 patients) required repeated surgery. This appears to be lower than the 30.3% recurrence rate reported in the adult literature after OSS.14 Furthermore, the observed maximal cyst size was not significantly different between the cohort of patients who developed a second lesion and those who did not, irrespective of underlying pathology. Interestingly, the lack of complete resection was not associated with laparoscopic approach, as was previously described in literature.16 Additionally, when applying the threshold of 8 cm for recurrence determined by Harada et al19 to our cohort, we found no difference in recurrence rate. Both of these findings suggest that a large lesion size should not preclude the attempt of an OSS or a laparoscopic approach. Although the recurrence rate might be underestimated, because longterm follow-up was not achieved in all patients, we had a large percentage of our patients (86%) return to our institution for follow-up. In further identifying potential predictive factors of recurrence, patients were stratified according to age categories and pathology. Adolescents were found to have a higher rate of recurrence with incomplete resection and this remained an independent predictor of recurrence among adolescents. No predictors for recurrence were seen in our cohort of children or neonates because of the small sample size within these age groups. When the adolescent cohort was further stratified according to pathology, incomplete resection was again found to be associated with
510
P.I. Abbas et al. / J Pediatr Adolesc Gynecol 29 (2016) 506e510
recurrence in patients with functional cysts but not for those with benign neoplastic lesions. Furthermore, younger age at the time of surgery was associated with the development of recurrent functional cysts. This increased recurrence of functional cysts in younger and potentially early menarchal girls might reflect hormonal fluctuations and dysfunctional ovulation from puberty that increases the likelihood of developing functional cysts.9,20 Conversely, in the benign neoplastic cohort, no patient characteristics or operative detail, including incomplete resection or intraoperative spillage, was associated with the 1 recurrent teratoma. Although resumption of menses is often reported as a clinical outcome because it might act as a surrogate for ovarian function, it has an unknown implication in future fertility. In adults, a study after menstrual cycles after conservative ovarian surgery confirmed no significant effects of the surgery on ovulatory function, menstrual function, or conception rates.21 We examined these factors in our patient cohort and found similar results. Most patients of pubertal age resumed or began menstruation postoperatively; 3 patients became pregnant. Longer follow-up is required to determine the utility of menstruation as a predictor of fertility. Limitations of our study include its retrospective nature and lack of uniform follow-up on all patients, which could bias the interpretation of the findings related to recurrence and repeat surgery rates. Because of the inherent variation in practice within our institution with regard to surveillance imaging in patients with a history of ovarian lesions, detection bias could also influence some of our findings. To ensure accurate follow-up information, we carefully screened medical record notes pertaining to all clinical encounters for the ovarian lesion or for abdominal complaints. Nevertheless, this highlights an opportunity to standardize our follow-up protocol and to study it prospectively. Furthermore, our use of resumption of menses as a metric to delineate the restoration or preservation of ovarian function might be of unknown clinical significance because laboratory values confirming ovarian function were not explicitly measured in our study. Additionally, we had a small number of patients with benign neoplastic lesions who developed a second ipsilateral lesion, thus limiting our ability to determine predictors of recurrence in this group. Last, we were unable to differentiate between a true recurrent lesion or the presence of a metachronous lesion on the ipsilateral ovary. Despite these limitations, this study adds to the pediatric surgical literature by providing data
from a large cohort of patients who successfully underwent OSS for various types of benign lesions. Future studies are needed to validate the finding of higher recurrence of functional cysts in adolescent patients. Conclusions
OSS is a feasible surgical approach that should be considered for all pediatric patients with benign ovarian tumors, because it can attain low recurrence and repeat surgery rates for benign neoplastic lesions. References 1. Helmrath M, Shin C, Warner B: Ovarian cysts in the pediatric population. Semin Pediatr Surg 1998; 7:19 2. Zolton J, Maseelall P: Evaluation of ovarian cysts in adolescents. Open J Obstet Gynecol 2013; 3:12 3. Cass D, Hawkins E, Brandt M, et al: Surgery for ovarian masses in infants, children, and adolescents: 102 consecutive patients treated in a 15-year period. J Pediatr Surg 2001; 36:693 4. Oltmann S, Garcia N, Barber R, et al: Can we preoperatively risk stratify ovarian masses for malignancy? J Pediatr Surg 2010; 45:130 5. Ozcan R, Kuruoglu S, Dervisoglu S, et al: Ovary-sparing surgery for teratomas in children. Pediatr Surg Int 2013; 29:233 6. Berger-Chen S, Herzog T, Lewin S, et al: Access to conservative surgical therapy for adolescents with benign ovarian masses. Obstet Gynecol 2012; 119:270 7. Brookfield K, Cheung M, Koniaris L, et al: A population-based analysis of 1037 malignant ovarian tumors in the pediatric population. J Surg Res 2009; 156:45 8. Akkoyun I, Gulen S: Laparoscopic cystectomy for the treatment of benign ovarian cysts in children: an analysis of 21 cases. J Pediatr Adolesc Gynecol 2012; 25:364 9. Strickland J: Ovarian cysts in neonates, children and adolescents. Curr Opin Obstet Gynecol 2002; 14:459 10. Cass D: Ovarian torsion. Semin Pediatr Surg 2005; 14:86 11. Abdel-Hady el-S, Abdel-Hady R, Gamal A, et al: Fertility sparing surgery for ovarian tumors in children and young adults. Arch Gynecol Obstet 2012; 285: 469 12. Chabaud-Williamson M, Netchine I, Fasola S, et al: Ovarian-sparing surgery for ovarian teratoma in children. Pediatr Blood Cancer 2011; 57:429 13. Lass A: The fertility potential of women with a single ovary. Hum Reprod Update 1999; 5:546 14. Poncelet C, Fauvet R, Boccara J, et al: Recurrence after cystectomy for borderline ovarian tumors: results of a French multicenter study. Ann Surg Oncol 2006; 13:565 15. Laberge P, Levesque S: Short-term morbidity and long-term recurrence rate of ovarian dermoid cysts treated by laparoscopy versus laparotomy. J Obstet Gynaecol Can 2006; 28:789 16. Lin P, Falcone T, Tulandi T: Excision of ovarian dermoid cyst by laparoscopy and by laparotomy. Am J Obstet Gynecol 1995; 173:769 17. Rogers E, Allen L, Kives S: The recurrence rate of ovarian dermoid cysts in pediatric and adolescent girls. J Pediatr Adolesc Gynecol 2014; 27:222 18. Savasi I, Lacy J, Gerstle J, et al: Management of ovarian dermoid cysts in the pediatric and adolescent population. J Pediatr Adolesc Gynecol 2009; 22:360 19. Harada M, Osuga Y, Fujimoto A, et al: Predictive factors for recurrence of ovarian mature cystic teratomas after surgical excision. Eur J Obstet Gynecol Reprod Biol 2013; 171:325 20. Kanizsai B, Orley J, Szigetvari I, et al: Ovarian cysts in children and adolescents: their occurrence, behavior, and management. J Pediatr Adolesc Gynecol 1998; 11:85 21. Sayegh R, Garcia C: Ovarian function after conservational ovarian surgery: a long-term follow-up study. Int J Gynaecol Obstet 1992; 39:303