Abstract
Study objective:
Limited data exists on the morphologic and physiologic effect on the remaining ovary after unilateral oophorectomy, especially in the pediatric population. Our aim is to evaluate ovarian volumes following unilateral oophorectomy to determine if compensatory ovarian hypertrophy occurs in the remaining contralateral ovary.
Design:
Retrospective chart review of ovarian volume measured on ultrasounds which were completed after unilateral oophorectomy. Postoperative ovarian volumes were compared to established radiologic standards.
Setting:
Large tertiary care academic children’s hospital in Atlanta, Georgia
Participants:
Female patients less than 21 years old who underwent unilateral oophorectomy.
Main Outcome Measures:
Ovarian volumes measured on postoperative ultrasounds.
Results:
93 patients met inclusion criteria for this study. Serial ultrasounds were performed on most patients (n=48, 51.6%) totaling 193 post-operative ovarian volumes. The average age of oophorectomy was 10.8 years. Prior to surgery, the majority of patients presented with abdominal pain (n=51, 54.8%) or pelvic mass (n=51, 54.8%) and most (n=77, 82.8%) had benign final pathology. Ovarian volumes were compared to 4 published radiologic ultrasound standards. Greater than 62.2% of ovarian volumes from girls who had previously had unilateral oophorectomy were larger than age-matched standard ovarian volumes.
Conclusion:
Ovarian enlargement occurs in the contralateral ovary following unilateral oophorectomy in the pediatric and adolescent population. This supports the concept of compensatory ovarian hypertrophy. This knowledge provides valuable information for interpretation of radiologic images in young females who have undergone oophorectomy and can assist with counseling on the risk of adnexal complications due to ovarian hypertrophy after unilateral oophorectomy.
Keywords: ovarian volume, oophorectomy, ovarian hypertrophy, compensatory hypertrophy
Introduction
Although the true prevalence of unilateral oophorectomy (UO) in the pediatric and adolescent gynecology (PAG) population is unknown, literature has shown that UO is too often performed for indications such as ovarian masses presumed to be malignant or adnexal torsion1–3. Long-term implications of UO in the pediatric population are not well known and limited data exists on the morphologic and physiologic effects on the remaining ovary after oophorectomy, particularly in the PAG population.
Several studies in animals, including those in pigs, rats, and opossums, have demonstrated compensatory ovarian hypertrophy following UO4–7. Similarly, ultrasound studies in human boys have demonstrated testicular hypertrophy following gonadectomy for unilateral cryptorchidism8. In adult females, small studies have indicated that a compensatory process may occur in the remaining ovary. For example, women undergoing invitro fertilization treatments with a history of UO were shown to have higher antral follicle counts and oocytes compared to ipsilateral ovaries in women who had not undergone oophorectomy9. However, limited definitive evidence for compensatory hypertrophy of the remaining ovary after UO has been documented, particularly in PAG. A recent study of ovarian volumes measured on ultrasounds of pediatric patients did demonstrate higher mean ovarian volumes on post-operative ultrasounds compared to age-matched controls10. This study was limited by sample size and only had comparisons of ovarian volume to decade of life instead of age by year, which does not account for dynamic changes in ovarian volumes which occur secondary to puberty.
Accepted radiologic standards for ovarian volumes have been published11–15. These standards have been criticized as they were developed using small sample sizes per age and less advanced ultrasound technology13. These standards were also developed under the assumption that mean ovarian volumes are symmetric within an individual and they do not account for the possibility of hypertrophy of the contralateral ovary in those with history of unilateral oophorectomy10,13.
To bridge the gap in knowledge on this topic, the aim of this study is to compare ovarian volumes to multiple established radiologic standards for ovarian volumes in girls who have not undergone UO. These comparisons will allow us to evaluate for changes in ovarian volumes specifically in this population following UO in order to determine if compensatory ovarian hypertrophy occurs in the contralateral ovary following UO. If ovarian hypertrophy after UO is found, this information provides valuable information for providers during interpretation of radiologic imaging, patient counseling and clinical management regarding risk of adnexal complications such as adnexal torsion in girls with ovarian enlargement after UO,
Materials and Methods
After institutional review board approval, a retrospective chart review was completed at a large tertiary academic children’s hospital. Female patients under the age of 21 years old were included. Medical records were queried using CPT, ICD-9 and ICD-10 medical codes for oophorectomy. Medical records were reviewed from January 2007, the start of electronic medical records at this institution, through January 2019. On review of medical records, patients were excluded if oophorectomy was not performed, if they subsequently underwent contralateral oophorectomy, if post-surgical ultrasounds were not performed, if post-surgical ultrasounds demonstrated sonographic abnormality such as a mass or cyst greater than 2 cm in diameter. Additionally, patients were excluded if their medical chart documented previous diagnosis which could impact pubertal development or timing such as precocious puberty.
The protocol for pelvic ultrasounds at this institution is a transabdominal approach with a full bladder performed by certified pediatric sonographers using the highest frequency transducer appropriate for patient size. In general, the ultrasounds performed at this institution would be performed with a curved transducer, but occasionally in small children a linear transducer could be used. Most radiologic standards for ovarian volume in the PAG population are established from transabdominal ultrasound, however, notably previous studies in adult and pediatric populations have supported that there is no significant difference in ovarian evaluation on transabdominal versus transvaginal ultrasound imaging13,16,17.
Post-operative ultrasound reports were reviewed by board certified pediatric radiologists. Ovarian volumes were recorded in three dimensions and calculated using standard formula (length in centimeters × width in centimeters × height in centimeters × 0.523). Secondary review of these ultrasound was completed by a board certified pediatric radiologist associated with our research team. In the event of discrepancy in ovarian volume between reviewers, review by a third board certified pediatric radiologist was planned, however, no discrepancies in ovarian volume were reported.
All post-operative ultrasounds were recorded for data collection. In those patients with serial post-operative ultrasounds, ovarian volumes for each postoperative ultrasound were recorded along with the patient age at the time that the ultrasound was performed. Ovarian volumes were pooled based on patient age and analyzed in two categories: first ultrasound and all ultrasounds. The first ultrasound category included only ovarian volumes recorded from the from one post-operative ultrasound, regardless of completion of serial ultrasounds. Ovarian volumes included in this category were recorded from only the first ultrasound that was performed post-operatively. The all ultrasound category included all ultrasounds that were completed in the post-operative period. In this category, patients who had serial ultrasounds completed could have multiple ovarian volumes recorded across age groups.
Post-operative ovarian volumes were compared to four established radiologic standards which were identified on literature review and are cited as the only available standards for mean ovarian volume in children and adolescents12–15. Radiologic standards for ovarian volume are based on female patients with no history of oophorectomy and evidence of bilateral ovaries visualized on ultrasound at the time of ovarian volume measurement11,14–16. Mean ovarian volumes and confidence intervals for mean ovarian volumes were calculated based on age group. Exact confidence intervals were calculated when there was one ultrasound reviewed per patient and bootstrapping confidence intervals were built when there was more than one ultrasound reviewed per patient. Mixed effects regression to account for multiple readings per person was performed to test for potential confounders of ovarian volume such as race, presence of comorbid conditions, laterality of surgery performed, route of surgery performed, presence of ovarian torsion at time of oophorectomy, and receipt of chemotherapy. Statistical significant was assessed at the alpha < 0.05 level and all analyses were done using STATA version 1518.
Results
Of the 328 patients who met the initial medical coding criteria, 93 patients met the inclusion criteria. Most patients were excluded due to lack of post-surgical ultrasounds or sonographic abnormalities. Three patients were excluded for medical comorbidities (two with precocious puberty and 1 with congenital adrenal hyperplasia). Forty-five percent (n=42) were White, 41.9% (n=39) were African American, 12.9% (n=12) were Hispanic, Asian, or unknown. Of the 93 patients included in the study, each had at least one post-operative ultrasound completed. This amounted to 93 ovaries reviewed from surgery to first post-operative ultrasound. Forty-eight (51.6%) patients had more than one post-operative ultrasound completed. When all post-operative ultrasounds were included, a total of 193 post-operative ovaries were available for review.
The average age of oophorectomy was 10.8 years (range 4 months to 18 years). The average time from surgery to first post-operative ultrasound was 12 months (range 0 to 129 months). Forty-eight percent (n=45) had only one post-operative ultrasound performed; in this group, the average time from surgery to post-operative ultrasound was 15 months (range 0 to 129 months). Of those patients with multiple (two or more) post-operative ultrasounds (n=48), most (n=19, 39.6%) received two serial post-operative ultrasounds. The highest number of postoperative ultrasounds performed was 7 (n=1, 1.1%). On average the interval between serial ultrasounds was 11 months (mode 6 months, range 0 to 54 months).
Descriptive statistics are documented in Table 1. Most (n=71, 76.3%) did not have other medical comorbidities. None had record of previous visits to pediatric gynecologist, which was likely a function of limited pediatric gynecology providers at the time of chart review. Few (n=1, 1.1%) had previously seen pediatric endocrinology. Thirty-four percent (n=32) were documented to be menarchal at time of oophorectomy, however, the majority of patients (n=39, 41.9%) did not have documentation of menarchal status at time of oophorectomy. Most underwent surgery for abdominal pain (n=51, 54.8%) or pelvic mass (n=51, 54.8%), and oophorectomy was more commonly performed via exploratory laparotomy (n=64, 68.8%) over laparoscopy (n=24, 25.8%). Laterality of oophorectomy was evenly distributed. Final pathology was most often benign (n=77, 82.8%). Notably, 17 patients (18.3%) underwent oophorectomy with final pathology demonstrating normal ovarian tissue only. Most of these cases (n=10, 58.8%) had ovarian torsion documented on operative findings, although 41.2% with normal ovarian tissue on pathology did not have documentation of presence of ovarian torsion on operative report or postoperative documentation. Thirteen patients (14.0%) received post-operative chemotherapy in the setting of malignancy on pathologic findings. A small number of patients (n=11, 11.5%) underwent serum testing of ovarian function following oophorectomy. Monitoring ovarian function following oophorectomy was more common in girls who subsequently underwent chemotherapy.
Table 1:
Demographic Data
| Characteristic | #(%) |
|---|---|
| Indication for Surgery | |
| Abdominal pain | 51 (54.8%) |
| Pelvic mass | 51 (54.8%) |
| Abdominal distention | 13 (14.0%) |
| Other | 12 (12.9%) |
| Surgery performed | |
| Exploratory laparotomy | 64 (68.8%) |
| Mini laparotomy | 4 (4.3%) |
| Laparoscopy | 24 (25.8%) |
| Other/unknown | 1 (1.1%) |
| Laterality of surgery performed | |
| Right | 45 (48.4%) |
| Left | 48 (51.6%) |
| Torsion on Surgical Findings | |
| Yes | 25 (26.9%) |
| No | 68 (73.1%) |
| Final Pathology | |
| Benign teratoma | 52 (55.9%) |
| Benign ovary | 17 (18.3%) |
| Malignant teratoma | 1 (1.1%) |
| Germ cell tumor | 3 (3.2%) |
| Dysgerminoma | 5 (5.4%) |
| Sertoli-Leydig tumor | 1 (1.1%) |
| Mucinous cystadenoma | 7 (7.5%) |
| Yolk sac tumor | 1 (1.1%) |
| Granulosa cell tumor | 1 (1.1%) |
| Serous cystadenoma | 1 (1.1%) |
| Other/unknown | 4 (4.3%) |
| Salpingectomy performed with oophorectomy | |
| Yes | 37 (39.8%) |
| No | 56 (60.2%) |
| Additional adnexal surgery following oophorectomy | |
| Yes | 2 (2.2%) |
| No | 91 (97.8%) |
Mean post-operative ovarian volumes by age are shown in Table 2. The mean ovarian volumes in our sample remained small until age 6. Mean ovarian volumes then increased sharply during expected ages of pubertal development, around ages 10–11, and then stabilized into early adulthood. In confounding analysis, mean ovarian volumes were not correlated with race, presence of comorbid conditions, laterality of surgery performed, route of surgery performed, presence of torsion at time of oophorectomy, or receipt of chemotherapy.
Table 2:
Mean ovarian volumes following unilateral oophorectomy
| Age (years) | 1st ultrasound | All ultrasounds | ||
|---|---|---|---|---|
| Number of ovaries | Mean ovarian volume in cm3 (SD) | Number of ovaries | Mean ovarian volume in cm3 (SD) | |
| < 1 | 7 | 1.13 (1.03) | 8 | 1.06 (0.97) |
| 1 | 1 | 0.63 | 1 | 0.63 |
| 2 | 1 | 0.39 | 1 | 0.39 |
| 3 | 0 | 1 | 0.30 | |
| 4 | 0 | 0 | ||
| 5 | 2 | 1.5 (1.70) | 2 | 1.5 (1.70) |
| 6 | 2 | 1.57 (1.03) | 4 | 1.31 (0.96) |
| 7 | 2 | 10.95 (12.79) | 3 | 7.84 (10.54) |
| 8 | 2 | 1.17 (0.25) | 5 | 1.85 (1.14) |
| 9 | 6 | 2.31 (2.05) | 10 | 2.76 (2.16) |
| 10 | 5 | 6.64 (4.12) | 9 | 7.76 (5.45) |
| 11 | 8 | 11.25 (4.83) | 10 | 10.43 (4.60) |
| 12 | 11 | 10.81 (7.56) | 20 | 12.84 (9.23) |
| 13 | 4 | 9.67 (9.17) | 17 | 9.70 (8.36) |
| 14 | 13 | 17.54 (8.42) | 24 | 15.55 (9.64) |
| 15 | 12 | 22.96 (37.51) | 22 | 25.47 (32.75) |
| 16 | 7 | 8.80 (5.77) | 21 | 10.74 (5.19) |
| 17 | 4 | 9.69 (3.89) | 11 | 7.13 (4.09) |
| 18 | 3 | 15.79 (16.81) | 9 | 13.06 (10.09) |
| 19 | 2 | 17.95 (17.04) | 6 | 20.24 (15.33) |
| 20 | 1 | 9.4 | 3 | 12.90 (6.77) |
Cohen et al published standards of ovarian volume in patients without history of UO for two age categories: those less than or equal to 10 or those greater than age 1012. In comparison to this standard, post-operative ovarian volumes in girls who had undergone oophorectomy were larger in both age categories (Table 3 and Graph 1A). Sixty-two percent (n=117) of postoperative ovarian volumes in our study were larger than this age-matched standard. Of enlarged ovaries, 81.20% (n=95) had more than 25% greater volume, 65.81% (n=77) had more than 50% greater volume, and 52.14% (n=61) had more than 75% greater volume than this standard.
Table 3:
Mean ovarian volumes following unilateral oophorectomy in comparison to Cohen et al standard
| Age (years) | Cohen et al | 1st ultrasound | All ultrasounds | |||
|---|---|---|---|---|---|---|
| Number of ovaries | Volume in cm3 (SD) | Number of ovaries | Mean ovarian volume in cm3 (SD) [95% CI] | Number of ovaries | Mean ovarian volume in cm3 and [95%CI]* | |
| ≤10 | 19 | 1.7 (1.4) | 28 | 3.09 (4.35) [1.40 to 4.77]) | 44 | 3.36 [2.09 to 4.65] |
| >10 | 83 | 7.8 (4.4) | 65 | 14.53 (17.72) [10.13 to 18.92] | 143 | 14.27 [11.81 to 16.73] |
CI is calculated using bootstrapping (1000 samples) to account for some repeated measures within subjects
Graph 1: Mean ovarian volumes following unilateral oophorectomy in comparison to published standards.
A: Compared to Cohen et al standard, B: Compared to Haber et al standard, C: Compared to Orsini et al standard, D: Compared to Gilligan et al, left ovaries only, E: Compared to Gilligan et al, right ovaries only
Haber et al published standards of ovarian volume for girls without history of UO aged 2 to 15 years14. In comparison to this standard, post-operative ovarian volumes in our study in girls who had undergone oophorectomy were larger in almost all age categories and variations in ovarian volume were more pronounced after age 6 (Table 4 and Graph 1B). The 95% confidence interval for mean ovarian volumes did not cross the standard ovarian volume after age 8, indicating a true difference in mean ovarian volumes between this sample and the established standard. Eighty-four percent (n=108) of post-operative ovarian volumes were larger than this age-matched standard. Of enlarged ovaries 94.44% (n=102) had more than 25% greater volume, 89.81% (n=97) had more than 50% greater volume, and 80.56% (n=87) had more than 75% greater volume.
Table 4:
Mean ovarian volumes following unilateral oophorectomy in comparison to Haber et al standard
| Age (years) | Haber et al | 1st ultrasound | All ultrasounds | |||
|---|---|---|---|---|---|---|
| Number of ovaries | Volume in cm3 (SD) | Number of ovaries | Mean ovarian volume in cm3 (SD) [95% CI] | Number of ovaries | Mean ovarian volume in cm3 (SD) [95% CI] | |
| 2+3 | 17 | 0.7 (0.4) | 1 | 0.39 | 2 | 0.35 [0.28 to 0.41] |
| 4+5 | 13 | 0.7 (0.5) | 2 | 1.5 (1.70) [0 to 16.75] | 2 | 1.50 [0 to 3.30] |
| 6+7 | 15 | 0.8 (0.6) | 4 | 6.26 (9.18) [0 to 20.87] | 7 | 4.11 [0 to 9.44] |
| 8+9 | 12 | 0.6 (0.4) | 8 | 2.03 (1.81) [0.51 to 3.54] | 15 | 2.45 [1.50 to 3.41] |
| 10+11 | 10 | 1.3 (1.0) | 13 | 9.48 (4.97) [6.47 to 12.48] | 19 | 9.16 [6.82 to 11.51] |
| 12+13 | 8 | 3.7 (2.1) | 15 | 10.51 (7.69) [6.25 to 14.77] | 37 | 11.39 [8.57 to 14.21] |
| 14+15 | 9 | 6.7 (4.8) | 25 | 20.14 (26.23) [9.31 to 30.97] | 46 | 20.29 [13.83 to 26.74] |
Orsini et al published standards of ovarian volume for girls without history of UO aged 2 to 13 years15. In comparison to this standard, post-operative ovarian volumes in girls who had undergone oophorectomy were larger particularly at age 7 and in all ages after age 10 (Table 5 and Graph 1C). The 95% confidence interval for mean ovarian volumes did not cross the standard ovarian volume after age 11, indicating a true difference in mean ovarian volumes between this sample and the established standard. 78.05% (n=64) of post-operative ovarian volumes were larger than this age-matched standard. Of enlarged ovaries 98.44% (n=63) had more than 25% greater volume, 85.94% (n=55) had more than 50% greater volume, and 75.00% (n=48) had more than 75% greater volume.
Table 5:
Mean ovarian volumes following unilateral oophorectomy in comparison to Orsini et al standard
| Age (years) | Orsini et al | 1st ultrasound | All ultrasounds | |||
|---|---|---|---|---|---|---|
| Number of ovaries | Mean volume in cm3 (SD) | Number of ovaries | Mean ovarian volume in cm3 (SD) [95% CI] | Number of ovaries | Mean ovarian volume in cm3 (SD) [95% CI] | |
| 2 | 5 | 0.75 (0.410 | 1 | 0.39 | 1 | 0.39 |
| 3 | 6 | 0.66 (0.17) | 1 | 0.30 | 1 | 0.30 |
| 4 | 14 | 0.82 (0.36) | 0 | 0 | ||
| 5 | 4 | 0.86 (0.02) | 2 | 1.5 (1.70) [0 to 16.75] | 2 | 1.5 (1.70) [0 to 3.29] |
| 6 | 9 | 1.19 (0.36) | 2 | 1.57 (1.03) [0 to 10.80] | 4 | 1.31 (0.96) [0.35 to 2.27] |
| 7 | 8 | 1.26 (0.59) | 2 | 10.95 (12.79) [0 to 125.89] | 3 | 7.84 (10.54) [0 to 18.86] |
| 8 | 10 | 1.05 (0.50) | 2 | 1.17 (0.25) [0 to 3.41] | 5 | 1.85 (1.14) [0.75 to 2.94] |
| 9 | 11 | 1.98 (0.76) | 6 | 2.31 (2.05) [0.16 to 4.46] | 10 | 2.76 (2.16) [1.43 to 4.08] |
| 10 | 12 | 2.22 (0.69) | 5 | 6.64 (4.12) [1.52 to 11.76] | 9 | 7.76 (5.45) [4.12 to 11.39] |
| 11 | 12 | 2.52 (1.30) | 8 | 11.25 (4.83) [7.22 to 15.28] | 10 | 10.43 (4.60) [7.06 to 13.80] |
| 12 | 6 | 3.80 (1.40) | 11 | 10.81 (7.56) [5.73 to 15.89] | 20 | 12.84 (9.23) [8.68 to 16.99] |
Gilligan et al published standards for ovarian volume for girls without history of UO aged 0 to 20 with ovarian volumes separated by laterality of ovary13. In comparison to this standard, post-operative ovarian volumes in girls who had undergone oophorectomy in our study were larger when matched for age and laterality of ovary. However, volumes were not uniformly larger across all age categories; enlargement in ovarian volume in the girls who had undergone oophorectomy was most pronounced in older age categories (Table 6, Graph 1D and 1E). Sixty-seven percent (n=127) of all ovarian volumes were greater than this age-matched and laterality-matched standard. Of enlarged ovaries 83.46% (n=106) had more than 25% greater volume, 71.65% (n=91) had more than 50% greater volume, and 55.91% (n=71) had more than 75% greater volume. For left ovaries only, 69.70% (n=69) of ovarian volumes were greater than this age- and laterality-matched standard. 84.06% (n=58) had more than 25% greater volume, 78.26% (54) had more than 50% greater volume, and 57.97% (n=40) had more than 75% greater volume. For right ovaries only, 65.17% (n=58) of ovarian volumes were greater than this ageand laterality-matched standard. 82.76% (n=48) had more than 25% greater volume, 63.79% (n=37) had more than 50% greater volume and 53.45% (n=31) had more than 75% greater volume.
Table 6:
Mean ovarian volumes following unilateral oophorectomy in comparison to Gilligan et al standard
| Age (years) | Gilligan et al | 1st Ultrasound | All Ultrasounds | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Left Ovary | Right Ovary | Left Ovary | Right Ovary | Left Ovary | Right Ovary | |||||||
| Number of ovaries | Mean ovarian volume in cm3 (SD) | Number of ovaries | Mean ovarian volume in cm3 (SD) | Number of ovaries | Mean ovarian volume in cm3 (SD) [95% CI] | Number of ovaries | Mean ovarian volume in cm3 (SD) [95% CI] | Number of ovaries | Mean ovarian volume in cm3 (SD) [95% CI] | Number of ovaries | Mean ovarian volume in cm3 (SD) [95% CI] | |
| 0 | 31 | 1.2 (2.1) | 31 | 1.4 (1.7) | 4 | 0.72 (1.06) [0 to 2.41] | 3 | 1.7 (0.79) [0 to 3.67] | 4 | 0.73 (1.06) [0 to 1.84] | 4 | 1.42 (0.85) [0.58 to 2.27] |
| 1 | 30 | 0.6 (0.4) | 30 | 0.6 (0.5) | 1 | 0.6 | 0 | - | 1 | 0.6 | 0 | - |
| 2 | 21 | 0.6 (0.5) | 21 | 0.8 (0.5) | 1 | 0.4 | 0 | - | 1 | 0.4 | 0 | - |
| 3 | 32 | 0.8 (0.5) | 32 | 0.9 (1.5) | 0 | - | 0 | - | 1 | 0.3 | 0 | - |
| 4 | 32 | 0.9 (1.2) | 32 | 0.9 (0.6) | 0 | - | 0 | - | 0 | - | 0 | - |
| 5 | 35 | 0.9 (0.8) | 35 | 0.8 (0.4) | 2 | 1.5 (1.70) [0 to 16.75] | 0 | - | 2 | 1.5 (1.7) [0 to 3.24] | 0 | - |
| 6 | 36 | 1.0 (0.7) | 36 | 1.1 (0.7) | 0 | - | 2 | 1.55 (1.06) [0 to 11.08] | 2 | 1.05 (1.20) [0 to 2.41] | 2 | 1.55 (1.06) [0.46 to 2.64] |
| 7 | 46 | 1.4 (1.3) | 46 | 1.7 (2.1) | 1 | 20 | 1 | 1.9 | 2 | 10.8 (13.01) [0 to 25.07] | 1 | 1.9 |
| 8 | 45 | 1.7 (1.4) | 45 | 1.7 (1.2) | 0 | - | 2 | 1.2 (0.28) [0 to 3.74] | 2 | 3.05 (0.49) [2.51 to 3.59] | 3 | 1.07 (0.31) [0.73 to 1.41] |
| 9 | 35 | 2.0(1.1) | 35 | 2.1 (1.0) | 4 | 1.1 (0.78) [0 to 2.34] | 2 | 4.7 (1.27) [0 to 16.14] | 4 | 1.1 (0.79) [0.38 to 1.82] | 6 | 3.85 (2.10) [2.13 to 5.57] |
| 10 | 35 | 2.6 (2.1) | 35 | 2.7 (2.1) | 3 | 4.1 (1.04) [1.51 to 6.69] | 2 | 10.45 (4.17) [0 to 47.93] | 5 | 5.4 (5.21) [0 to 11.17] | 4 | 10.7 (4.71) [5.96 to 15.44] |
| 11 | 37 | 4.7 (4.2) | 37 | 4.7 (3.3) | 4 | 12.75 (6.60) [2.26 to 23.24] | 4 | 9.75 (2.20) [6.25 to 13.25] | 6 | 10.88 (5.88) [5.61 to 16.15] | 4 | 9.75 (2.20) [7.56 to 11.94] |
| 12 | 36 | 4.9 (2.6) | 36 | 6.2 (3.5) | 7 | 12.17 (9.15) [3.71 to 20.64] | 4 | 8.45 (3.34) [3.13 to 13.77] | 12 | 13.53 (11.13) [7.34 to 19.72] | 8 | 11.79 (5.85) [7.70 to 15.87] |
| 13 | 37 | 5.6 (2.8) | 37 | 6.9 (3.6) | 3 | 8.73 (11.01) [0 to 36.08] | 1 | 12.5 | 12 | 10.50 (9.80) [4.68 to 16.33] | 5 | 7.76 (2.98) [5.04 to 10.48] |
| 14 | 45 | 7.1 (5.0) | 45 | 9.5 (7.3) | 5 | 20.88 (8.40) [10.45 to 31.31] | 8 | 15.46(8.27) [8.55 to 22.38] | 13 | 17.24 (11.02) [11.27 to 23.21] | 11 | 13.54 (7.74) [8.99 to 18.10] |
| 15 | 40 | 7.1 (4.4) | 40 | 7.4 (4.6) | 6 | 38.88 (49.69) [0 to 91.03] | 6 | 7.03 (4.33) [2.49 to 11.58] | 11 | 35.51 (42.87) [8.99 to 62.03] | 11 | 15.43 (13.87) [6.64 to 24.21] |
| 16 | 47 | 7.2 (4.0) | 47 | 8.2 (5.5) | 3 | 5.43 (2.05) [0.33 to 10.53] | 4 | 11.34 (6.59) [0.86 to 21.84] | 10 | 10.81 (6.09) [6.85 to 14.77] | 11 | 10.7 (4.51) [8.02 to 13.38] |
| 17 | 45 | 7.0 (4.2) | 45 | 8.0 (4.5) | 2 | 12.1 (4.38) [0 to 51.49] | 2 | 7.25 (1.48) [0 to 20.59] | 6 | 8.38 (4.74) [4.75 to 12.01] | 5 | 5.6 (2.85) [3.01 to 8.19] |
| 18 | 40 | 6.7 (4.3) | 40 | 7.5 (4.1) | 1 | 5.6 | 2 | 20.9 (20.22) [0 to 202.60] | 2 | 9.5 (5.15) [3.36 to 15.64] | 7 | 14.09 (11.19) [5.81 to 22.36] |
| 19 | 36 | 6.7 (5.4) | 36 | 7.5 (4.5) | 1 | 30 | 1 | 5.9 | 2 | 17.6 (17.53) [0 to 36.33] | 4 | 21.55 (16.80) [5.04 to 38.06] |
| 20 | 37 | 7.2 (3.8) | 37 | 7.9 (4.3) | 0 | - | 1 | 9.4 | 0 | - | 3 | 12.9 (6.77) [5.51 to 20.29] |
Conclusions
The mean ovarian volumes after UO demonstrated in our study were larger than those ovarian volumes published as accepted standards for girls without history of UO. Differences in ovarian volumes in girls following UO compared to accepted standards of ovarian volumes developed from ultrasounds in girls without history of oophorectomy were especially evident with increasing age, particularly after age 10. The larger ovarian volumes above standard norms following UO in our study of demonstrate the possibility of a compensatory process.
Compensatory unilateral ovarian hypertrophy has been demonstrated previously in non-human animal models4–7. Human studies have demonstrated the ability of the adult female ovary to physiologically compensate to meet fertility needs after UO9. Our data further supports these findings. Additionally, this data supports a previous small study of ultrasound volumes in girls following UO which indicated that enlargement of the remaining ovary can occur. However, in that study, comparison of ovarian volume was limited to age by decade or menarchal status, and thus, did not delineate specific differences in ovarian size by age throughout childhood and adolescence in girls who had undergone UO10.
It is important to note that several of these ovarian volume standards used for comparison of ovarian volume in our study have limitations, particularly the Cohen et al, Haber et al, Orsini et al standards. The most notable is that these standards were developed using small sample sizes, some with fewer than 10 ovaries used to determine mean ovarian volume per age category15. These standards have also been published for several decades with publication dates ranging from 1984 to 1994, and concern has been raised that alterations in pubertal timing and ultrasound technology in recent decades could alter these standard values13,19,20. Additionally, question has been raised regarding the appropriateness of combining ovarian volumes of the left and right ovary, as some evidence has demonstrated larger ovarian volumes on the right in comparison to the left13. These standards, however, are regularly used in the literature and in daily practice to outline expected ovarian volumes by age and are the only published standards available for comparison. Notably, while our study compares ovarian volumes following UO to these more controversial standards, this study also uses a recently proposed standard from 2019 demonstrated in Gilligan et al. When compared to this more recently published study which used larger samples sizes, more stringent inclusion criteria, and allowed for variations in laterality for impact on ovarian volumes, the mean ovarian volumes demonstrated in our study remained larger than those documented in this standard13.
Limitations of our study include the retrospective nature which is at risk of selection bias. In addition, the lack of standardization of post-oophorectomy ultrasound timing prevented comparison of pre-operative and post-operative ovarian volumes as many patients who received post-operative ultrasounds did not have pre-operative ultrasound imaging (and instead had imaging via alternative such as CT or MRI). Therefore, published standards for ovarian volume were the only available comparison group. In examining the mean ovarian volumes of girls who received serial ultrasounds, there was not sufficient power to compare changes in ovarian volume over time and postoperative ultrasounds could not be controlled for age or time between ultrasounds. A future prospective study could better address this.
Many patients did not have documentation of menarchal status or last menstrual period, inhibiting analysis of changes in ovarian volume in relation to phases of the menstrual cycle. While it is widely accepted that ovarian volumes fluctuate throughout the menstrual cycle, a previous small study has demonstrated persistent ovarian enlargement in patients with unilateral oophorectomy compared to controls in the same menstrual cycle phase10. The lack of documentation of menstrual cycle at the time of pelvic ultrasound in our study is likely due to the fact that PAG was a new specialty at this children’s hospital and the standard radiology template for ultrasound images did not include last menstrual cycle during the time periods this data was collected. Notably, this data was presented the radiology division at the institution where this data was collected and has subsequently led to a change in protocol and improved documentation of menarchal status with performance of pelvic ultrasounds within this age group.
Additionally, few patients underwent serum testing of ovarian function following oophorectomy, limiting the ability to draw conclusions about alterations in ovarian function when compensatory ovarian hypertrophy occurs. Future research would include evaluation of serum anti-mullerian hormone, follicular stimulating hormone, luteinizing hormone, and estradiol levels as well as scheduled interval ultrasounds following oophorectomy to trend changes in ovarian function along with ovarian size.
Another important observation from this study is that 18% of patients underwent oophorectomy with final pathology ultimately demonstrating normal ovarian tissue. Additionally, the majority (55.9%) of cases returned with final pathology of benign teratoma, a condition regularly treated with ovarian cystectomy instead oophorectomy given the benign nature of this condition, low recurrence risk and concerns regarding impact on future hormonal support and fertility with oophorectomy in young girls21. Although not an aim of our study, this is an important observation as it indicates that 69 (74.2%) patients potentially underwent oophorectomy unnecessarily. While specific conclusions cannot be drawn from this data regarding decision-making for oophorectomy as well as expertise or surgeon comfort with ovarian masses, it is important to note that previous research has demonstrated that surgeon specialty may play a role in decision-making for cystectomy versus oophorectomy for benign ovarian masses and may alter surgical decision for ovarian preservation in the setting of ovarian torsion1–3. Furthermore, the large number of laparotomies in comparison to laparoscopies performed for oophorectomy in this study may also indicate a difference in preference based on provider specialty. However, given that information on surgeon decision-making regarding route of surgery based on pre-operative ovarian size or concern for malignancy was not consistently available during data collection, confounding factors cannot be eliminated. Notably, during the time of data collection for this study, a pediatric gynecologist was not on staff at our institution and all ovarian masses were managed solely by pediatric surgery. Additionally, none of the patients in our study had records of previous visits to pediatric gynecologist prior to their surgery.
Despite limitations, this study contributes to sparse literature on the topic of oophorectomy in the pediatric and adolescent population and demonstrates the possibility of compensatory ovarian hypertrophy following UO in the PAG population. Understanding alterations in ovarian size following UO is important for radiologists and surgeons as it may impact radiologic imaging interpretation of the “normal” ovarian size and, therefore, change the threshold for volumes used to evaluate for potential pathology is girls who have undergo oophorectomy. Additionally, understanding variations in ovarian volume following oophorectomy may allow for more detailed counseling of patients, as some ovarian volumes in our study were larger than 20 cm3, which is the generally accepted volume over which an ovary may be at risk for adnexal torsion in adolescents22. This knowledge of ovarian hypertrophy may assist providers in their clinical management of patients with prior UO especially in regard to protection of the remaining ovary. Ovarian hypertrophy and increased risk of adnexal torsion based on ovarian volume may warrant consideration of oophoropexy to reduce risk of future adnexal torsion, though there is limited literature regarding guidance on when to perform oophoropexy in patients23–25. Some providers may even consider counseling regarding fertility preservation, particularly if ovarian torsion was noted intraoperatively during contralateral oophorectomy while others may consider use of post-operative oral contraceptives to reduce ovarian stimulation or risk of future physiologic cyst formation, theoretically decreasing risk of future ovarian torsion with enlarged ovary26.
Continued research is needed to better understand the pathophysiology of compensatory process within the ovary. Moreover, additional larger prospective studies are needed to evaluate ovarian volumes in the PAG population. Prospective evaluation of changes in ovarian volumes over time after UO could eliminate many of the limitations of this retrospective study and contribute to continued understanding of the compensatory process of contralateral ovarian hypertrophy following UO demonstrated in this study. Radiologists and surgeons should be aware of the potential for compensatory ovarian hypertrophy following UO to assist in counseling and clinical management in the PAG population.
Footnotes
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Previously presented: Oral presentation at ASRM Scientific Congress and Expo, October 2019, Philadelphia PA; poster accepted for presentation at NASPAG Annual Clinical and Scientific Meeting, April 2020, Grapevine TX (conference was subsequently cancelled)
Conflict of Interest Statement: our authors do not report any conflicts of interest
References
- 1.Berger-Chen S, Herzog TJ, Lewin SN, et al. Access to conservative surgical therapy for adolescents with benign ovarian masses. Obstet Gynecol 2012;119(2 Pt 1):270–275. [DOI] [PubMed] [Google Scholar]
- 2.Bristow RE, Nugent AC, Zahurak ML, Khouzhami V, Fox HE. Impact of surgeon specialty on ovarian-conserving surgery in young females with an adnexal mass. J Adolesc Health. 2006;39(3):411–416. [DOI] [PubMed] [Google Scholar]
- 3.Peeraully R, Henderson K, Fairbrother K, et al. Effect of Surgical Specialty on Management of Adnexal Masses in Children and Adolescents: An 8-Year Single-Center Review. J Pediatr Adolesc Gynecol. 2020;33(1):89–92. [DOI] [PubMed] [Google Scholar]
- 4.Dailey RA, Peters JB, First NL, Chapman AB, Casida LE. Effect of unilateral ovariectomy in the Yorkshire and Poland China prepuberal gilt. J Anim Sci. 1969;28(6):775–779. [DOI] [PubMed] [Google Scholar]
- 5.Staigmiller RB, First NL, Casida LE. Ovarian compensatory hypertrophy following unilateral ovariectomy in hysterectomized and early pregnant gilts. J Anim Sci. 1972;35(4):809–813. [DOI] [PubMed] [Google Scholar]
- 6.Mandl AM, Zuckerman S, Patterson HD. The number of oocytes in ovarian fragments after compensatory hypertrophy. J Endocrinol. 1952;8(4):347–356. [DOI] [PubMed] [Google Scholar]
- 7.Hartman CG. Observations on the Functional Compensatory Hypertrophy of the Opossum Ovary. American Journal of Anatomy. 1925(35):1–24. [Google Scholar]
- 8.Laron Z, Zilka E. Compensatory hypertrophy of testicle in unilateral cryptorchidism. J Clin Endocrinol Metab. 1969;29(11):1409–1413. [DOI] [PubMed] [Google Scholar]
- 9.Khan Z, Gada RP, Tabbaa ZM, et al. Unilateral oophorectomy results in compensatory follicular recruitment in the remaining ovary at time of ovarian stimulation for in vitro fertilization. Fertil Steril. 2014;101(3):722–727. [DOI] [PubMed] [Google Scholar]
- 10.Dao KA, Tadros S, Chan S, Poremski PM, Squires JH. Pediatric ovarian volumes measured at ultrasound after contralateral unilateral oophorectomy. Pediatr Radiol. 2019;49(5):632–637. [DOI] [PubMed] [Google Scholar]
- 11.Cohen HL, Shapiro MA, Mandel FS, Shapiro ML. Normal ovaries in neonates and infants: a sonographic study of 77 patients 1 day to 24 months old. AJR Am J Roentgenol. 1993;160(3):583–586. [DOI] [PubMed] [Google Scholar]
- 12.Cohen HL, Tice HM, Mandel FS. Ovarian volumes measured by US: bigger than we think. Radiology. 1990;177(1):189–192. [DOI] [PubMed] [Google Scholar]
- 13.Gilligan LA, Trout AT, Schuster JG, et al. Normative values for ultrasound measurements of the female pelvic organs throughout childhood and adolescence. Pediatr Radiol. 2019;49(8):1042–1050. [DOI] [PubMed] [Google Scholar]
- 14.Haber HP, Mayer EI. Ultrasound evaluation of uterine and ovarian size from birth to puberty. Pediatr Radiol. 1994;24(1):11–13. [DOI] [PubMed] [Google Scholar]
- 15.Orsini LF, Salardi S, Pilu G, Bovicelli L, Cacciari E. Pelvic organs in premenarcheal girls: real-time ultrasonography. Radiology. 1984;153(1):113–116. [DOI] [PubMed] [Google Scholar]
- 16.Andolf E, Jorgensen C. A prospective comparison of transabdominal and transvaginal ultrasound with surgical findings in gynecologic disease. J Ultrasound Med. 1990;9(2):71–75. [DOI] [PubMed] [Google Scholar]
- 17.Nazario AC, Nicolau SM, Nishimura CM. [Comparison between pelvic endovaginal and transabdominal sonography in the measurement of the uterus and ovaries]. Rev Paul Med. 1991;109(2):51–54. [PubMed] [Google Scholar]
- 18.Stata Statistical Software: Release 15 [computer program]. College Station, TX: StataCorp LLC; 2017. [Google Scholar]
- 19.Euling SY, Herman-Giddens ME, Lee PA, et al. Examination of US puberty-timing data from 1940 to 1994 for secular trends: panel findings. Pediatrics. 2008;121 Suppl 3:S172–191. [DOI] [PubMed] [Google Scholar]
- 20.McDowell MA, Brody DJ, Hughes JP. Has age at menarche changed? Results from the National Health and Nutrition Examination Survey (NHANES) 1999–2004. J Adolesc Health. 2007;40(3):227–231. [DOI] [PubMed] [Google Scholar]
- 21.Rogers EM, Allen L, Kives S. The recurrence rate of ovarian dermoid cysts in pediatric and adolescent girls. J Pediatr Adolesc Gynecol. 2014;27(4):222–226. [DOI] [PubMed] [Google Scholar]
- 22.Adeyemi-Fowode O, McCracken KA, Todd NJ. Adnexal Torsion. J Pediatr Adolesc Gynecol. 2018;31(4):333–338. [DOI] [PubMed] [Google Scholar]
- 23.Abes M, Sarihan H. Oophoropexy in children with ovarian torsion. Eur J Pediatr Surg. 2004;14(3):168–171. [DOI] [PubMed] [Google Scholar]
- 24.Comeau IM, Hubner N, Kives SL, Allen LM. Rates and Technique for Oophoropexy in Pediatric Ovarian Torsion: A Single-Institution Case Series. J Pediatr Adolesc Gynecol. 2017;30(3):418–421. [DOI] [PubMed] [Google Scholar]
- 25.Crouch NS, Gyampoh B, Cutner AS, Creighton SM. Ovarian torsion: to pex or not to pex? Case report and review of the literature. J Pediatr Adolesc Gynecol. 2003;16(6):381–384. [DOI] [PubMed] [Google Scholar]
- 26.Murray S, London S. Management of ovarian cysts in neonates, children, and adolescents. Adolescent and Pediatric Gynecology. 1995;8(2):64–70. [Google Scholar]