ORIGINAL RESEARCH • PEDIATRIC IMAGING

Physiologic Ovarian Cysts versus Other Ovarian

and Adnexal Pathologic Changes in the
Preadolescent and Adolescent Population:
US and Surgical Follow-up
Erica K. Schallert, MD  •  Paulette I. Abbas, MD  •  Amy R. Mehollin-Ray, MD  •  Martin C. Price, MD  • 
Jennifer E. Dietrich, MD  •  Robert C. Orth, MD, MPH, PhD
From the Department of Radiology, Texas Children’s Hospital, 6701 Fannin St, Suite 470, Houston, TX 77030 (E.K.S., A.R.M.R., R.C.O.); Department of Surgery,
Children’s Hospital of Michigan, Detroit, Mich (P.I.A.); Department of Radiology, Geisinger Wyoming Valley Medical Center, Wilkes Barre, Pa (M.C.P.); and Department
of Pediatric and Adolescent Gynecology, Texas Children’s Hospital, Houston, Tex (J.E.D.). Received November 30, 2018; revision requested January 23, 2019; revision
received March 31; accepted April 3. Address correspondence to E.K.S. (e-mail: ekschall@texaschildrens.org).

Conflicts of interest are listed at the end of this article.

Radiology 2019; 292:172–178 • https://doi.org/10.1148/radiol.2019182563 • Content code:

Background:  Ovarian and adnexal cysts are frequently encountered at US examinations performed in preadolescent and adolescent
patients, yet there are few published studies regarding the outcomes of cysts in this population.

Purpose:  To identify characteristics at US that help to distinguish physiologic ovarian cysts from nonphysiologic entities.

Materials and Methods:  Female patients who underwent pelvic US with or without Doppler from January 2009 through December
2013 were identified by using a centralized imaging database. Patients older than 7 years and younger than 18 years with ovarian or
adnexal cysts at least 2.5 cm were included. Demographic characteristics, date of surgery, surgical notes, and pathologic reports were
extracted from the electronic medical record. Initial and follow-up dates of US, cyst size and complexity, imaging diagnosis, and
change on subsequent US images were recorded. Statistical analysis was performed with the Wilcoxon rank sum and Kruskal-Wallis
tests for continuous variables and the Fisher exact test for categorical variables.

Results:  Of 754 patients who met inclusion criteria (age, 8–18 years; mean age, 14.6 years 6 1.9 [standard deviation]; mean cyst
size, 5 cm 6 3.3), 409 patients underwent complete follow-up that included resolution at imaging (n = 250) or surgery (n = 159).
In the patients with complete imaging follow-up, mean time to US documentation of resolution was 194 days 6 321; 59.6% (149
of 250) patients had nonsimple cyst characteristics. One-hundred fifty-nine patients underwent surgical intervention (mean cyst
size, 8.5 cm 6 5.3), and 69.8% (111 of 159) of the cysts had simple characteristics. Of the 159 cysts, 100 (62.8%) were defined in
the pathologic report as paratubal cysts. Of 409 patients, no malignancies were encountered in this study population with surgical
or imaging resolution.

Conclusion:  No malignancies were encountered in the study population and the majority of cysts resolved at follow-up imaging.
Large size, persistence, and separability from the ovary were most helpful for identification of nonphysiologic paratubal cysts.
© RSNA, 2019

Ring of ovarian cysts, adnexal cysts, cystic masses, and ad-

ecent efforts to standardize characterization and report-
nexal masses at US have focused on providing meaningful
risk stratification (1–3). The resulting published imaging
guidelines to direct the follow-up and treatment of ovar-
ian lesions are largely on the basis of adult data (2). Often,
these adult-based guidelines are extrapolated to children
without evidence to support this practice.
Ovarian cysts are frequently incidentally encountered
in the pediatric population (4). Millar et al (5) found small
ovarian cysts (defined as echo-free structures within the vi-
cinity of the ovary, .10 mm) in 2%–5% of their cohort of
prepubertal patients (age, 0–8 years) and most cysts larger
than 2 cm were in patients younger than 1 year. In an as-
ymptomatic patient population aged 10–19 years imaged
with US for a year, Porcu et al (6) reported a frequency
of 12% for simple ovarian cysts. The majority of ovarian
cysts resolve without treatment in 6 months inherent to
their physiologic nature (4,7–10). Nonphysiologic cystic
This copy is for personal use only. To order printed copies, contact reprints@rsna.org
entities in the pediatric population include benign and ma-
lignant ovarian neoplasms, infection, and paratubal cysts
(11–13). Kanizsai et al (8) found no malignancies in 144
ovarian cysts in an adolescent population, and Templeman
et al (9) found a low incidence of malignancy (4.4%) in
a study of ovarian surgical cases in an adolescent popula-
tion. Paratubal cysts are most often referred to in the radi-
ology literature as paraovarian cysts and can be identified
as separate from the ipsilateral ovary by transabdominal US
(14). Paratubal cysts are located in the mesosalpinx of the
broad ligament separate from the ovary (15) and are rem-
nants of paramesonephric or mesonephric ducts, Wolffian
structures intended to become male reproductive anatomy
when stimulated by androgen secretion in the embryo
(16,17). Because of the normal hormonally-driven stimu-
lation of the Wolffian structures in males, it is hypothe-
sized that paratubal cysts (as Wolffian remnants in females)
would also be hormonally sensitive and may increase in size
(18). Recent adolescent gynecology and pediatric surgery

Summary
Adnexal cysts in adolescents and preadolescents are usually benign
and the majority resolve on their own.
Key Points
nn Of 409 adolescent and preadolescent patients with cystic adnexal
lesions at least 2.5 cm in diameter, no malignancies were encoun-
tered.
nn In the 250 of 409 patients with imaging-only follow-up, the mean
time to documentation at US of cyst resolution was 194 days 6
321 (standard deviation).
nn The incidence of paratubal cysts was 24.4% (100 of 409 cystic
adnexal lesions . 2.5 cm).
studies support this theory and have found a correlation between
obesity, paratubal cyst size, and hyperandrogenism (13,19,20).
Because paratubal cysts are nonphysiologic cysts, they will not
regress spontaneously (13).
Clinical treatment of ovarian and adnexal cysts in the pediat-
ric population is on the basis of size, complexity, and symptoms.
Surgical intervention is indicated for acute symptoms, solid or
complex masses, and large (5 cm) masses that do not undergo
regression (4).
With this in mind, the purpose of our study was to identify char-
acteristics at US that would help differentiate between physiologic
ovarian cysts and nonphysiologic ovarian and adnexal cysts.
Materials and Methods
Institutional review board approval was obtained for this retro-
spective cohort study, and the requirement to obtain informed
consent was waived. This study was Health Insurance Portabil-
ity and Accountability Act compliant.
Patient Population
Patients in our tertiary care, stand-alone children’s hospital who
had been evaluated with transabdominal pelvic US (with or with-
out Doppler) and with positive key word search of cyst or cystic
lesion in the diagnostic report at our institution from January 9,
2009, through December 31, 2013, were identified by using our
centralized imaging database (M.C.P., pediatric radiology fel-
low, and P.I.A., pediatric surgery postdoctoral research fellow).
This yielded 1235 patients. Patients younger than 8 years were
excluded to concentrate on the peri- and postpubertal age group
because our focus was on distinguishing physiologic cysts from
nonphysiologic cysts. Adult patients, defined as 18 years or older,
were excluded. Both diagnostic reports and US images were re-
viewed (E.K.S., pediatric radiologist with 5 years of experience).
Patients who had ovarian or other adnexal cysts with maximum
axis measuring at least 2.5 cm in diameter were enrolled in the
study. If bilateral or multiple unilateral cysts were manifest, the
largest was selected as a single patient entry. The lower size limit
was set at 2.5 cm because normal follicles can enlarge up to 2–2.5
cm before ovulation (1). Cysts with solid components or imag-
ing features typical of a dermoid cyst and documented as such
in the radiology report were excluded. Descriptive features at US
that the primary interpreting radiologist used when prospectively
diagnosing a dermoid cyst (confirmed at histologic analysis) in-
Radiology: Volume 292: Number 1—July 2019  n  radiology.rsna.org 173
Schallert et al
cluded the following: round hyperechoic masses with attenuation
of the sound beam posteriorly, dot-dash echo pattern, and absent
internal color Doppler flow (2). Finally, patients were excluded if
imaging follow-up was incomplete, defined as last US with per-
sistent cyst and no surgery. Figure 1 highlights the inclusion and
exclusion criteria, and the path of patients through study.
Study Design and Clinical Variables
Retrospective review of the electronic medical record in the
study population was performed for demographics, dates of
pelvic US, dates of surgery, surgical notes, and pathologic re-
ports (M.C.P., P.I.A., and E.K.S.). Menarchal status was not
recorded because of the longitudinal nature of the study.
One radiologist (E.K.S.) reviewed US images in the picture
archiving and communication system to corroborate the diag-
nostic US report; collect terminology of the originally interpret-
ing radiologist’s impression; and verify maximal cyst size, com-
plexity, and change at follow-up US when applicable. Lesion
complexity was stratified into two categories, simple and non-
simple. Simple was defined as round or oval, anechoic, smooth,
thin wall, no solid component or septation, posterior acoustic
enhancement, and no internal flow (2). The nonsimple cyst cat-
egory included all other lesions in which no solid or internal vas-
cular components were detected. For example, nonsimple cysts
included findings such as thin (2–3 mm) avascular septations,
uniform internal echogenicity, heterogeneous internal echo-
genicity, reticular internal echoes, peripheral wall vascularity, and
homogeneous avascular echoes (2). If follow-up imaging was
performed, the date of the follow-up US and change in size (ie,
increased in size, decreased in size, unchanged, or resolved) were
documented. If multiple follow-up imaging examinations were
performed, the date of the examination that demonstrated cyst
resolution or the examination with the longest time to follow-up
was selected to provide maximal chance of resolution. Follow-up
examination dates were included until December 2017.
Patients were placed in two main categories of definitive fol-
low-up: complete resolution at US or surgical intervention. Pa-
tients with surgical intervention either underwent surgery after
the initial US or after one or more follow-up US examinations.
Statistical Analysis
Continuous variables were summarized as means and standard
deviations. Categorical variables were summarized as counts
and percentages. Wilcoxon rank-sum and Kruskal-Wallis tests
were applied to compare continuous variables, whereas Fisher
exact test was applied to compare categorical variables. P values
less than .05 were considered to indicate statistical significance.
We used statistical software for all statistical analyses (SAS 9.4;
SAS Institute for Statistical Computing, Cary, NC).
Results
A total of 754 patients met initial inclusion criteria (mean
age, 14.6 years 6 1.9 [standard deviation]). We subsequently
excluded 345 of 754 (45.8%) patients either because they
underwent a single US examination and did not undergo
surgery (n = 290) or they underwent more than one US
examination without resolution at imaging or surgery (n =
Physiologic Ovarian Cysts versus Other Ovarian and Adnexal Pathologic Changes
Figure 1:  Flowchart of the inclusion and exclusion criteria, and the path of patients
through the study.
55) because we could not be definitive about the etiologic
nature of their cyst. The final study population (409 of 754
patients; 54.2%) included patients who underwent com-
plete imaging or surgical follow-up (Table 1).
Follow-up with US Imaging Only
Of the 464 patients with follow-up, 305 (65.7%) underwent
imaging only. In 250 of those 305 patients (82.0%), ovarian
cysts resolved (mean time to follow-up imaging, 194 days 6
321). Twenty-six of the 305 adnexal cysts (8.5%) decreased
in size (mean time to follow-up imaging, 66 days 6 61), 20
(6.6%) stayed the same size (mean time to follow-up imaging,
256 days 6 406), and nine (3.0%) increased in size (mean time
to follow-up imaging, 470 days 6 351).
Surgical Outcomes and Histologic Analysis
The 159 patients who underwent surgical intervention either
underwent surgery after the initial US (100 patients; 62.8%)
or after at least one follow-up US examination (59 patients;
37.1%). The decision for surgical intervention is complex and
individualized by patient, but primary indications are pain or
persistence and in consideration of cysts 5 cm or larger, which
174 radiology.rsna.org  n  Radiology: Volume 292: Number 1—July 2019
increase the risk of torsion. Among the 100 pa-
tients who underwent surgery after the initial US
examination, findings at histologic examination
confirmed a benign paratubal cyst in 62 (62.0%)
and benign ovarian etiologic causes in the other
38.0% (mean time to surgery, 10 days 6 60).
Twenty-seven patients (27.0%) were noted to
have concurrent torsion involving the following
benign etiologic causes: paratubal cyst (n = 16),
hemorrhagic cyst (n = 1), corpus luteum (n = 1),
dermoid (n = 1), ovarian cyst (n = 2), paraovarian
cyst (n = 1), adnexal cyst (n = 1), mucinous cyst-
adenoma (n = 1), benign papillary cystadenoma
(n = 1), and ischemic ovarian parenchyma (n = 2).
In the 59 patients who underwent additional US
imaging before surgery, 96% (57 of 59) of cysts ei-
ther increased in size (22 of 59) or stayed the same
size (35 of 59). The cysts decreased in size in two pa-
tients: one from 4.8 cm to 3.5 cm during 8 months
and the second from 7.6 cm to 5.8 cm during 5 days.
Surgical intervention was on the basis of pain and
persistence in the first patient (paratubal cyst) and
pain with clinical concern for torsion in the second
(hemorrhagic corpus luteum). All of the 59 patients
with cysts and additional US imaging eventually un-
derwent surgical procedure, and results of histologic
examination confirmed paratubal cysts in 38 of the
59 patients (64%) and benign other in 21 patients
(35%). Four patients had concurrent torsion, all in-
volving paratubal cysts. Mean time from initial US
to surgical intervention in this group who underwent
additional US imaging was 258 days 6 297.
Histologic findings from both surgical groups
showed paratubal cyst in 100 of 159 patients
(62.9%) and other benign etiologic causes in 59
(37.1%) (Table 2). Examples of terminology pulled from the
impression section of the initial pelvic US radiology report
in cystic lesions that were diagnosed at histologic analysis as
paratubal cysts are in Table 3, with the typical appearance at
imaging in Figure 2. Nine patients were found to have cystad-
enomas (Table 4, Fig 3). The mean size of the torsed paratubal
cysts (n = 31) versus nontorsed cysts (n = 69) was 8 cm 6 4.1
versus 8.5 cm 6 5.3, respectively (P , .76).
Cyst Size and US Characteristics
The mean size of the cysts that underwent imaging only (the im-
aging-only group) was 4.2 cm 6 1.3, and the mean size of all cysts
treated with surgery was 8.5 cm 6 5.4 (P , .001). Of the cysts
that did not undergo surgical procedure, 46.2% (141 of 305) were
simple, whereas 69.8% (111 of 159) of the cysts that underwent
surgical procedure were simple (P , .001). Of the cysts that re-
solved on US images, 40.4% (101 of 250) were simple.
A summary of cyst size and characteristics at US across the
three patient groups is in Figures 4 and 5. Briefly, mean cyst size
was different (P , .001) between all groups: mean cyst sizes for
the imaging-only group, the group who underwent surgery after
initial US, and the group who underwent surgery after follow-up

Table 1: Study Population
Patient Group Mean Age (y)
Resolution at US (n = 250)
14.5 6 1.8
Surgery after initial US (n = 59) 14.6 6 1.9
Surgery after follow-up US (n = 100) 13.9 6 1.6
Note.—Data are 6 standard deviation. Population consists of
female patients (age 8–18 years) who underwent an initial pelvic
US examination that depicted ovarian or other adnexal cyst 
2.5 cm, with either resolution of cyst at follow-up US or who
underwent surgical follow-up.
Table 2: Summary of Benign Adnexal Histologic Types
Histologic Analysis
Paratubal cysts
Dysfunctional ovarian cysts
  Corpus luteum (with or without hemorrhage)
  Follicular cyst (with or without hemorrhage)
Nonspecific location and histologic type
  Adnexal cyst
  Paraovarian cyst
  Epithelium-lined inclusion cyst
  Hemorrhagic cyst
Mature cystic teratoma (dermoid)
Other: benign, origin masked by ischemic change
Serous cystadenoma
Mucinous cystadenoma
Ovarian simple cyst
Insufficient sample for definitive diagnosis
Fibroma versus fibrothecoma
Cystadenofibroma
Hematometrocolpos
Endometrioma
Mesonephric cyst (location not specified)
Mullerian cyst (paraovarian versus ovarian)
Papillary cystadenoma
Paramesonephric cyst
US were 4.2 cm 6 1.3, 9.9 cm 6 5.9, and 6.1 cm 6 3.0, respec-
tively. The percentage of cysts with simple imaging characteris-
tics was 46.2% (141 of 305) for the imaging-only group, 68.0%
(68 of 100) for those who underwent surgery after the first US,
and 71% (42 of 59) for cysts that underwent US follow-up be-
fore surgery (P , .001).
Discussion
Few studies exist regarding the outcomes of ovarian and ad-
nexal cysts in the pediatric population, and current recom-
mendations are on the basis of adult guidelines. The purpose
of our study was to identify characteristics at US that would
allow for differentiation of physiologic ovarian cysts, which
will resolve on their own, from nonphysiologic entities. In-
cluded were 754 of 1235 patients who met inclusion criteria
as preadolescent and adolescent girls (mean age, 14.6 years
Radiology: Volume 292: Number 1—July 2019  n  radiology.rsna.org 175
Schallert et al
Table 3: Terminology from Initial Pelvic US Radiology
Report
Simple ovarian cyst
Ovarian versus paraovarian
Ovarian versus mesenteric versus lymphatic cyst
Cystic structure located eccentrically within the ovary, likely of
follicular origin
Paraovarian/paratubal cyst
Exophytic cyst
Exophytic or paraovarian
Indeterminate cystic structure
Physiologic, paratubal, hydrosalpinx, cystic teratoma, serous
cystadenoma
Tubo-ovarian abscess versus complex hydrosalpinx versus
No. of paratubal cyst
Patients CSF pseudocyst versus paraovarian cyst
100 Sequela of PID, endometrioma or other complex adnexal mass
Adnexal cyst
10 Large cystic structure…which limits evaluation of origin. Right
4 ovary not seen
Exophytic ovarian cyst, dominant follicle, or simple cyst
1 Cystic structure within ovary likely representing physiologic cyst
9 Two physiologic adnexal cyst with torsed ovary versus large
1 hemorrhagic cyst
2 Cystadenoma versus paraovarian
6 Cyst intimately associated with torsed ovary
5 Note.—Reports were from cystic lesions that were diagnosed as
4 paratubal cysts at histologic analysis; the terminology is from the
4 Impression section of the radiology reports. CSF = cerebrospinal
3 fluid, PID = pelvic inflammatory disease.
2
1
1
1
1
1
1
1
1
Figure 2:  Transabdominal grayscale transverse US
image in the right lower quadrant in a 15-year-old
girl shows imaging features typical of a paratubal cyst
(between calipers) including simple unilocular charac-
teristics at US. It is located between the ovary (Ov) and
uterus (Ut), separate from the ovary.
6 1.9) with an adnexal cyst 2.5 cm or larger identified at
US from 2009–2013. These patients were followed through
2017, and 409 of 754 patients underwent definitive follow-
up that consisted of either complete resolution at US (250
patients) or surgery with a histologic diagnosis (159 pa-
Physiologic Ovarian Cysts versus Other Ovarian and Adnexal Pathologic Changes

Table 4: Details of Patients with Cystadenomas

Patient Mean Cystadenoma Characteristic

Pathologic Diagnosis Age (y)* Size (cm)* at US Change in Size† US Interval (mo) Torsed
Papillary cystadenoma 13 16.2 NS NA NA Yes
Serous cystadenoma 14.7 18 S NA NA No
Serous cystadenoma 13.1 14.5 S NA NA No
Serous cystadenoma 15.6 20 NS NA NA No
Serous cystadenoma 14.6 10.1 S No 1.5 No
Mucinous cystadenoma 16.7 11.4 S NA NA Yes
Mucinous cystadenoma 16.1 7.2 NS I 2.3 No
Mucinous cystadenoma 16.1 6.2 NS No 3 No
Mucinous cystadenoma 16.4 4.1 S I 10.2 No
Mean of all patients 15.2 12 NA NA NA NA
Note.—I = increased, NA = not applicable, NS = nonsimple, S = simple.
* At initial US.
†
At follow-up US.

Nonsimple characteristics at US did not help to predict sur-

Figure 3:  Transabdominal grayscale transverse
US image in the right lower quadrant in a 16-year-
old girl shows a 9.4-cm simple unilocular cyst (be-
tween calipers). This is a surgically proven mucinous
cystadenoma.
tients). Cyst resolution occurred on average within 6 months
of the initial US. This finding is similar to the existing litera-
ture (4) regarding physiologic follicular involution. Of those
250 patients in whom cysts resolved at follow-up US, 149
patients (59.6%) had cysts that showed nonsimple charac-
teristics at US. This can be explained by superimposed hem-
orrhage and the evolution of blood products in the expected
course of functional and dysfunctional follicular growth and
atresia (2,4).
Cyst size was a statistically significant variable (P , .001) in
patients who underwent a surgical procedure (mean cyst size, 8.5
cm) versus patients with cysts who did not undergo a surgical
procedure (mean cyst size, 4.2 cm). This supports the association
of larger cyst size with potential risk for torsion in the pediatric
population (21). This observation is consistent with the Soci-
ety of Radiologists in Ultrasound consensus guideline option of
surgical evaluation in reproductive-age adult women (age, 18
years) with simple cysts larger than 7 cm (2).
gical over nonsurgical treatment because they represented only
30.8% (49 of 159) of the patients who underwent surgical in-
tervention compared with 59.6% (149 of 250) of the patients
whose cyst demonstrated resolution at US. The odds ratio of a
simple cyst undergoing surgical procedure was 2.6. This observa-
tion was unexpected, but a potential explanation would be that
hemorrhage (and therefore complexity) is common in the cycle
of follicular development and atresia; however, the high percent-
age of paratubal cysts encountered in our study are a nonphysi-
ologic entity, not expected to regress (ultimately requiring surgi-
cal treatment) and typically unilocular, simple cyst morphologic
structure (4,13).
Our results are similar to existing studies (8,9) that show
that the majority of adnexal cysts in the pediatric population are
benign. No malignancies were encountered in our study popu-
lation of 409 patients with either surgical correlation or com-
plete imaging resolution, which is similar to previous studies,
although to our knowledge there are no studies with cohorts as
large as our cohort (8).
A unique finding in our study was the high percentage of
paratubal cysts encountered in the surgical group (62.8% [100
of 159]) and incidence in cysts with follow-up (24.4% [100 of
409]). Size was not helpful in predicting those who had torsed
cysts (8 cm) versus those who did not (8.5 cm) (P = .76). The
Society of Radiologists in Ultrasound consensus guidelines state
that paraovarian and paratubal cysts were considered together
with ovarian cysts; per the guidelines, “while they are not likely to
resolve, simple paraovarian cysts generally are inconsequential in
asymptomatic women” and “that using the same size thresholds
as for ovarian cysts was reasonable” (2). The American College
of Obstetricians and Gynecologists Practice Bulletin lists para-
tubal cysts as a separate entry in the benign category and does
not give a specific recommendation for this entity, but rather
a general guideline for benign disease (22). For the pediatric
population, this is an important distinction. Such cysts are fre-
quently a source of abdominal pain, and because paratubal cysts
are not physiologic, they are unlikely to resolve on their own,

176 radiology.rsna.org  n  Radiology: Volume 292: Number 1—July 2019


Figure 4:  Box-and-whisker plot shows cyst size across patient groups (P , .01).
Figure 5:  Bar graph compares US characteristics across patient
groups (P , .01 for comparison of groups).
which necessitates clinical correlation of size and symptoms to
determine timing of what will ultimately be operative interven-
tion (13). Because of recent studies that correlate paratubal cyst
size, obesity, and hyperandrogenism (19,20), the incidence of
this entity will likely parallel that of the growing pediatric obesity
epidemic (23).
Limitations of our study included its retrospective design,
which limited ability to accurately identify cyst location as
intra- or extraovarian especially because of the large size of many
of the cysts. Many patients were excluded from final analysis be-
cause of lack of surgical or imaging follow-up (total number of
patients excluded, 345 of 754 [45.8%]; patients with cysts, 290
of 754 [38.5%]; patients in whom cysts did not resolve at follow-
up US and who did not undergo surgery, 55 of 305 [18.0%]).
Additionally, single radiologist review of the US images was a
weakness, but we did not see it as a major limitation because
their role was to corroborate size and classify cysts as simple
Radiology: Volume 292: Number 1—July 2019  n  radiology.rsna.org 177
Schallert et al
versus nonsimple rather than provide a primary
interpretation of the imaging findings. In terms of
histologic diagnosis, limitations included two cysts
with insufficient sample for definitive diagnosis and
nonstandardized language in the pathologic reports,
which made classification of some cysts difficult.
Additionally, histologic analysis–confirmed entities
in this study were benign, but serous and mucinous
cystadenomas may have a percentage of atypical
cells that would classify them as borderline ovarian
tumors. One of our patients had atypia of less than
10% with a comment in the patient record of fo-
cal epithelial proliferation. Potentially mucinous or
serous cystadenoma (found in eight patients) could
recur as borderline ovarian tumor or malignant tu-
mor if there was an undersampling error. A study
from our institution found 14 borderline ovarian
tumors in a 15-year period, which overlapped with
our study period by 7 years, in which US charac-
teristics were not specified but generalized that
simple-to-complex features were noted on imaging
studies (24). Radiologist language for description of
paratubal cysts was nonuniform; paratubal, rather
than paraovarian or exophytic descriptions, would
facilitate accurate communication with surgical and pathology
department staff.
In conclusion, ovarian and other adnexal cysts were benign
in our adolescent and preadolescent population who underwent
surgical intervention or in whom cysts resolved (ie, no longer
visible) at imaging. Large size, persistence, and separability from
the ovary are the most helpful clues for identification of non-
physiologic paratubal cysts.
Acknowledgment: Thank you to Ninad Patel, MD, for his expertise in gyne-
cologic pathology.
Author contributions: Guarantors of integrity of entire study, E.K.S.,
A.R.M.R.; study concepts/study design or data acquisition or data analysis/inter-
pretation, all authors; manuscript drafting or manuscript revision for important
intellectual content, all authors; approval of final version of submitted manuscript,
all authors; agrees to ensure any questions related to the work are appropriately
resolved, all authors; literature research, E.K.S., J.E.D., R.C.O.; clinical studies,
E.K.S., A.R.M.R., M.C.P., R.C.O.; statistical analysis, R.C.O.; and manuscript ed-
iting, E.K.S., P.I.A., A.R.M.R., J.E.D., R.C.O.
Disclosures of Conflicts of Interest: E.K.S. disclosed no relevant relation-
ships. P.I.A disclosed no relevant relationships. A.R.M.R. disclosed no relevant re-
lationships. M.C.P. disclosed no relevant relationships. J.E.D. disclosed no relevant
relationships. R.C.O. disclosed no relevant relationships.
References
1. Brown DL, Dudiak KM, Laing FC. Adnexal masses: US characterization and report-
ing. Radiology 2010;254(2):342–354.
2. Levine D, Brown DL, Andreotti RF, et al. Management of asymptomatic ovarian
and other adnexal cysts imaged at US: Society of Radiologists in Ultrasound Con-
sensus Conference Statement. Radiology 2010;256(3):943–954.
3. Maturen KE, Blaty AD, Wasnik AP, et al. Risk stratification of adnexal cysts and cys-
tic masses: Clinical performance of Society of Radiologists in Ultrasound Guidelines.
Radiology 2017;285(2):650–659.
4. Strickland JL. Ovarian cysts in neonates, children and adolescents. Curr Opin Ob-
stet Gynecol 2002;14(5):459–465.
5. Millar DM, Blake JM, Stringer DA, Hara H, Babiak C. Prepubertal ovarian cyst
formation: 5 years’ experience. Obstet Gynecol 1993;81(3):434–438.
6. Porcu E, Venturoli S, Dal Prato L, Fabbri R, Paradisi R, Flamigni C. Frequency and
treatment of ovarian cysts in adolescence. Arch Gynecol Obstet 1994;255(2):69–72.
Physiologic Ovarian Cysts versus Other Ovarian and Adnexal Pathologic Changes
7. Qublan HST, Abdel-hadi J. Simple ovarian cysts: frequency and outcome in girls
aged 2-9 years. Clin Exp Obstet Gynecol 2000;27(1):51–53.
8. Kanizsai B, Orley J, Szigetvári I, Doszpod J. Ovarian cysts in children and adolescents:
their occurrence, behavior, and management. J Pediatr Adolesc Gynecol 1998;11(2):
85–88.
9. Templeman C, Fallat ME, Blinchevsky A, Hertweck SP. Noninflammatory ovarian
masses in girls and young women. Obstet Gynecol 2000;96(2):229–233.
10. Brandt ML, Luks FI, Filiatrault D, Garel L, Desjardins JG, Youssef S. Surgical indi-
cations in antenatally diagnosed ovarian cysts. J Pediatr Surg 1991;26(3):276–281;
discussion 281–282.
11. Epelman M, Chikwava KR, Chauvin N, Servaes S. Imaging of pediatric ovarian
neoplasms. Pediatr Radiol 2011;41(9):1085–1099.
12. Anthony EY, Caserta MP, Singh J, Chen MY. Adnexal masses in female pediatric
patients. AJR Am J Roentgenol 2012;198(5):W426–W431.
13. Muolokwu E, Sanchez J, Bercaw JL, et al. The incidence and surgical manage-
ment of paratubal cysts in a pediatric and adolescent population. J Pediatr Surg
2011;46(11):2161–2163.
14. Kim JS, Woo SK, Suh SJ, Morettin LB. Sonographic diagnosis of paraovarian cysts: value
of detecting a separate ipsilateral ovary. AJR Am J Roentgenol 1995;164(6):1441–1444.
15. Perlman S, Hertweck P, Fallat ME. Paratubal and tubal abnormalities. Semin Pediatr
Surg 2005;14(2):124–134.
16. Stenbäck F, Kauppila A. Development and classification of parovarian cysts. An ul-
trastructural study. Gynecol Obstet Invest 1981;12(1):1–10.
178 radiology.rsna.org  n  Radiology: Volume 292: Number 1—July 2019
17. Stock RJ. Large intraligamentous cysts. Technique of surgical removal and correla-
tion of surgical observations and histologic findings pertaining to origin. J Reprod
Med 1987;32(5):347–352.
18. Devouassoux-Shisheboran M, Silver SA, Tavassoli FA. Wolffian adnexal tumor, so-
called female adnexal tumor of probable Wolffian origin (FATWO): immunohisto-
chemical evidence in support of a Wolffian origin. Hum Pathol 1999;30(7):856–863.
19. Dietrich JE, Adeyemi O, Hakim J, et al. Paratubal cyst size correlates with obesity and
dysregulation of the Wnt signaling pathway. J Pediatr Adolesc Gynecol 2017;30(5):
571–577.
20. Muolokwu E, Sanchez J, Bercaw JL, et al. Paratubal cysts, obesity, and hyperandrogen-
ism. J Pediatr Surg 2011;46(11):2164–2167.
21. Oltmann SC, Fischer A, Barber R, Huang R, Hicks B, Garcia N. Cannot exclude
torsion--a 15-year review. J Pediatr Surg 2009;44(6):1212–1216; discussion 1217.
22. American College of Obstetricians and Gynecologists’ Committee on Practice Bulle-
tins—Gynecology. Practice bulletin no. 174: evaluation and management of adnexal
masses. Obstet Gynecol 2016;128(5):e210–e226.
23. Skinner AC, Ravanbakht SN, Skelton JA, Perrin EM, Armstrong SC. Prevalence of obe-
sity and severe obesity in US children, 1999-2016. Pediatrics 2018;141(3):e20173459.
24. Childress KJ, Patil NM, Muscal JA, Dietrich JE, Venkatramani R. Borderline ovar-
ian tumor in the pediatric and adolescent population: a case series and literature
review. J Pediatr Adolesc Gynecol 2018;31(1):48–54.