| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Clinical Observations |
1 Pediatric Radiology Department, Hospital Materno-infantil Vall d'Hebron, Ps.
Vall d'Hebron 119-129, Barcelona E-08035, Spain.
2 Department of Radiology, Consorci Hospitalari Parc Taulí, Sabadell,
Spain.
3 Department of Pathology, Hospital Vall d'Hebron, Barcelona, Spain.
4 Department of Obstetrics, Hospital Materno-infantil Vall d'Hebron, Barcelona,
Spain.
5 Department of Pediatric Surgery, Hospital Materno-infantil Vall d'Hebron,
Barcelona, Spain.
Received July 24, 2004; accepted after revision October 15, 2004.
Address correspondence to G. Enríquez
(genriquez{at}vhebron.net).
OBJECTIVE. Prenatally diagnosed complex ovarian cysts are most often managed surgically in an attempt to save the ovary. Nevertheless, published surgical results disclose that most patients undergo oophorectomy or salpingo-oophorectomy. We assessed whether a surgical or conservative approach was more appropriate by comparing the long-term outcome of infants treated by both methods. A hypothesis for the cause of complex cysts is presented.
CONCLUSION. Clinical evidence questions the use of surgery for asymptomatic complex ovarian cysts. Histologic analysis suggests gonad maldevelopment as the origin of complex neonatal ovarian cysts.
Before the routine use of sonography, only large ovarian cysts palpable on physical examination could be detected. Such cysts were originally thought to represent approximately 5% of all abdominal masses in infant girls in the first month of life [1]. Thanks to sonography, the detection rate of ovarian cysts has risen significantly in Western countries. The incidence is uncertain but has been estimated at 1 in 2,625 pregnancies [2]. Visualization of these cysts on prenatal sonography poses a dilemma regarding treatment because considerable debate persists as to their management. The source of this controversy lies in the following considerations: Most patients are asymptomatic at birth; the literature emphasizes that ovarian cysts carry a high potential risk of complications, the most severe being torsion of the pedicle, which theoretically could result in loss of the affected ovary [3–8]; and although maternal or placental hormone stimulation is thought to be the cause of fetal ovarian cysts, this theory has not been proven by scientific evidence, so the etiopathogenesis remains uncertain. As a result, the treatment of ovarian cysts is not standardized and is seldom based on the clinical status of the patient (usually asymptomatic) but rather on the size and appearance of the cyst on sonographic examination. On the basis of their sonographic features, the cysts are divided into two types: simple, which are completely anechoic and have imperceptible walls, and complex, also known as complicated, twisted, or hemorrhagic, which show specific characteristics such as a debris–fluid level, septa, a retracting clot or mural nodule, or echogenicity mimicking a solid appearance, as described by Nussbaum et al. [3].
The consensus on the management of simple cysts is to adopt a wait-and-see approach with sonographic monitoring because they tend to regress spontaneously, although some authors advocate obstetric procedures such as intrauterine puncture of the cyst [6, 9, 10] or induction of labor [11]. In the case of simple cysts more than 5 cm in diameter and complex cysts, surgical removal of the mass is most often recommended in the clinical [8, 10], surgical [4, 7, 12–15], and radiologic literature [16–18] in an attempt to preserve viable ovarian tissue. Nevertheless, published surgical results disclose that most patients undergo oophorectomy or salpingo-oophorectomy [11, 13, 14, 19–22]. Two groups of investigators, however [20, 23], have recommended conservative management for all types of cysts (simple and complex), regardless of their sonographic features.
The present study was designed to investigate these discrepancies in concept and management by comparing two groups of patients with complex ovarian cysts treated surgically or conservatively. Sonographic findings for all patients, the surgical and histologic data of the operated patients, and the long-term outcome of those managed conservatively are presented. In addition, we propose a hypothesis for the etiopathogenesis of neonatal ovarian cysts based on a review of the surgical, histologic, and follow-up data of our patients and an analysis of human ovary development.
Materials and Methods
The study comprised two groups of patients with complex ovarian cysts: the retrospective surgical group, which included nine patients, eight of whom underwent surgery according to the early protocol of our hospital and one of whom underwent surgery later because of clinical symptoms related to the size of the cyst; and the prospective conservative group, which included 11 consecutive newborns managed conservatively with sonographic monitoring according to our revised hospital protocol.
Surgical Group
Patients—The medical charts of nine infant girls who had
undergone surgery for nine complex ovarian cysts at the Pediatric Surgery Unit
of our center were retrospectively reviewed. Eight of the surgeries took place
between 1981 and 1991 and one in 1998. These nine infants were selected from
13 operated on for ovarian cysts during the study period (1981–1999).
Selection criteria included complete clinical records, a postnatal sonographic
study identifying the cyst as complex, surgical data on the location and
appearance of the cyst and on the presence or absence of the pedicle and
pedicle torsion, and histopathologic analysis of the surgical specimen. In
addition, maternal charts were reviewed to ascertain whether the mother had a
history of diabetes, toxemia, or Rh isoimmunization.
At our institution, neither the approval of the Ethics Committee nor the informed consent of parents is required for retrospective review of medical records.
In patients 1 to 7, the cysts were discovered on prenatal sonography at gestational ages ranging from 32 to 39 weeks. The infants were operated on shortly after birth, six between 1981 and 1991, conforming to the routine for management of this condition at our center during that period, and one (patient 7) in 1998 because of clinical symptoms related to the size of the mass. In the remaining two patients, operated on in 1987 and 1989, the cyst was discovered on abdominal (patient 8) or hip (patient 9) radiographs obtained for unrelated reasons at the ages of 3 and 4 months, respectively. These two patients were subsequently examined by sonography, and the masses were surgically removed. Laparotomy through a Pfannenstiel's incision was performed on all patients by a pediatric surgeon.
Radiography—Abdominal radiography was performed on two patients to investigate vomiting (patient 7) or diarrhea (patient 8). Radiographs of the hip were obtained for patient 9 to investigate hip luxation.
Sonography—Prenatal sonography was performed on seven of the nine patients by obstetric sonographers according to the protocol for pregnancy in our country, which includes first-, second-, and third-trimester routine studies. All postnatal sonographic studies were performed by two of the authors—between 1 and 10 days old for patients 1–7 and at 3 and 4 months old for patients 8 and 9, respectively—using a sonographic scanner (Aloka 650 or Acuson 128) equipped with high-frequency sector and linear transducers (7–5 MHz). Sonographic examination included a complete survey of the abdomen and longitudinal and transverse views of the cyst in all nine patients.
Sonographic images and the original sonographic reports were reviewed for the purposes of this study by three pediatric radiologists.
Histopathology—All the specimens were evaluated by pediatric pathologists. For the present study, one pediatric pathologist reassessed the six available preserved specimens with special focus on characterization of the cyst wall and pedicle. For this purpose, four to seven 5-µm sections of the tissue blocks were stained by H and E, the Masson stain, and periodic acid–Schiff and examined by light microscopy.
Conservative Group
Patients—Eleven consecutive newborns with 11 complex ovarian
cysts managed conservatively were prospectively studied between 1993 and 1999.
All patients underwent sonographic follow-up and clinical monitoring according
to our revised hospital protocol based on analysis of the surgical and
histologic data of the surgical group, presented herein. All the parents gave
written informed consent for conservative management with sonographic
monitoring of the cysts.
The cysts were detected by prenatal sonography in all patients and, in addition, by prenatal MRI in one. Intrauterine puncture of the cyst had been performed on patient 8, and delivery was induced at the gestational age of 37 weeks in patient 4. Postnatally, the characteristics of the masses were defined by sonography (11 patients) or MRI (two patients).
Sonography—Prenatal sonographic studies were performed by obstetricians according to the same protocol as described for the surgical group.
Postnatal and follow-up sonographic examinations were done by the same pediatric radiologists using an Aloka 650, Acuson 128, or Sequoia 512 (Acuson) sonographic scanner equipped with high-frequency sector and linear transducers.
The first study was conducted when the infants were between one and six days old and included a routine survey of the abdomen, longitudinal and transverse views of the cysts measured in two diameters, and pelvic scans to reveal the ovaries. Sector (5–8 MHz) and linear (8–13 MHz) probes were routinely used. In particular, linear probes were highly useful for defining the cyst wall. The field of view was extended in some patients to better show the position of the cyst with respect to other organs such as the liver and kidney. Bladder catheterization was not used to study the pelvis in any patient.
Follow-up sonographic studies (four to seven per patient) were performed monthly during the first 3 months and at 3-month intervals thereafter. The examinations focused on detecting changes in the size, position, or echotexture of the cysts. At the end of follow-up, the entire pelvis was reexamined.
Results
Surgical Group
Clinical and surgical findings—Of the nine patients in this
group, only patient 7 had symptoms (abdominal distention and vomiting) that
were due to the large size of the cyst (11.2 x 6.1 cm). The remaining
eight patients (89%) were asymptomatic (patients 1–6) or had
gastroenteritis (patient 8) or asymmetric gluteal folds (patient 9), neither
being related to the cyst. On physical examination, the cyst was palpable in
patients 5 and 7. No maternal history of diabetes, toxemia, or isoimmunization
had been recorded for any patient.
Surgical data are presented in Table 1. The cysts were found in the right upper quadrant (two patients), right lower quadrant (five patients), left lower quadrant (one patient), and occupying both sides of the abdomen above the bladder (one patient). All specimens were soft, rounded, and yellowish or brown. A hard nodule corresponding to a calcification was present in two specimens.
|
|
|
|
|
Radiography findings—Radiographs revealed a mass occupying the lower half of the abdomen and displacing the intestinal loops in patient 7, a rounded mass with an internal calcification in the right hemiabdomen in patient 8, and a calcification without mass effect, in left pelvis, in patient 9 (Fig. 1A).
|
|
|
Histopathology findings—No mature ovarian tissue was histologically documented in any of the nine surgical specimens. On focused reevaluation of the six available preserved specimens, the cysts were found to be covered with a cortex of celomic mesothelium with vestiges of germinal epithelium, consisting of a single layer of cuboidal and, on occasion, peglike epithelial cells. Within this covering were found scattered oogonia positive for glycogen on periodic acid–Schiff staining (Fig. 2A) and, in two specimens, a few primordial follicles and germ cells with flattened primitive granulosa. No hilus cells, ovarian medulla, or rete ovarii were observed in any specimen. The pedicle was composed of vascularized connective tissue with thrombosis of some vessels (Fig. 2B).
Conservative Group
Clinical findings—The 11 patients were clinically
asymptomatic. Ten were full-term neonates with normal vaginal delivery. Labor
was induced in one by the attending obstetrician at 37 weeks of gestation
because of discovery of the cyst. At birth, all 11 neonates were physically
examined by a neonatologist and masses were palpated in three. The infants
were closely followed up as outpatients by a pediatrician, and none presented
clinical symptoms related to the cyst.
Sonography findings—Nine cystic and two solid-appearing abdominal masses were detected in the 11 patients on prenatal examination performed between 33 and 37 weeks of gestation. Sonography-guided intrauterine puncture performed on one patient yielded 26 mL of serohematic fluid.
The size of the cysts varied from 2.4 x 2.6 cm (the cyst punctured in utero) to 5.3 x 6.1 cm. The cyst was located in the subhepatic region in two patients, anterior to the right kidney in two, anterior to the left kidney in one, in the right lower quadrant in four, and in the left lower quadrant in two. All cysts were complex according to the criteria of Nussbaum et al. [3]: An echogenic wall was found in nine patients (82%), a debris–fluid level in eight (73%), a mural nodule in two (18%), and septa in two (18%). In two patients (18%), the lesion had a solid appearance and showed a central hyperechogenic nodule (Fig. 3A). The hemorrhagic content of the cyst was established by MRI in these two patients (Fig. 3B).
Follow-up sonographic results are summarized in Table 2. A progressive decrease in cyst size became more evident on follow-up studies after the age of 3 months. Involution of the cyst occurred in most patients between 3 and 15 months old. The mural nodules and debris–fluid levels disappeared from all patients, new septa appeared in five, and a crescent-shaped calcification was visualized at the base of the mass in seven (Figs. 4A, 4B, 4C). Remarkably, the cyst was found to be mobile in five patients and even was seen on one side of the abdomen or the other on different examinations. A pelvic study conducted after involution of the cyst revealed a single normal ovary (left in seven patients and right in four patients), with a normal sonographic appearance (Fig. 3C). Only two residual calcifications, one in the cul-de-sac and the other adjacent to the contralateral ovary, were visible.
|
|
|
|
Discussion
This study showed that complex neonatal ovarian cysts have a natural tendency to involute spontaneously and that, therefore, conservative management with clinical and sonographic monitoring would be reasonable. Nevertheless, there is still a tendency to operate early to preserve the ovary. Surgery in neonates is not free of risk and should be performed only if salvage of the gonad reasonably can be expected. The ovary was not saved in any of our seven newborns undergoing immediate postnatal surgery, and this outcome coincides with extensive data retrieved from the literature.
In 1991, Brandt et al. [19] reviewed 257 reports of simple and complex ovarian cysts published between 1975 and 1990. One hundred seventy patients were treated surgically, and oophorectomy was performed on 145 (85.3%). In 1992, Müller-Leisse et al. [20] reported 49 ovarian cysts from a multicenter study at five German institutions; 15 (58%) of the 26 patients who underwent surgery required oophorectomy. A review of the 1992–2003 literature [5, 11, 14, 21–24] yielded reports of 49 complex ovarian cysts from 49 patients, 46 (93.9%) of whom underwent oophorectomy. Although cystectomy was the surgical procedure performed on some patients [3, 12, 15, 17], it is unknown whether ovarian function and morphology were preserved. In several articles, the type of surgery was not mentioned [5, 8]. All these data led us to question the need for surgery in these patients and to establish conservative management for asymptomatic newborns with complex ovarian cysts discovered in utero. In addition, the surgical and histologic findings for our operated patients and a review of human gonadal development prompted us to suggest another etiopathogenesis for this entity.
The human ovary originates high in the abdomen in a pararenal location. The
primitive gonad is embedded in mixoid vascularized connective tissue that runs
from the diaphragm to the cloaca, in close contact with the mesonephric
tubules that stem from the mesonephric duct (duct of Wolff). The
paramesonephric duct (Müller's canal) is parallel to this group of
structures. During differentiation of the primitive gonad, the upper portion
of the paramesonephric duct develops into the fallopian tube, and the
mesonephric duct atrophies. As it matures, the ovary (followed by the
fallopian tube) descends until it reaches its normal location in the neonatal
pelvis, where its position changes from vertical to transverse
(Fig. 5A). The position of the
cyst high in the abdomen and its attachment to the pelvis by a thin stalk in
our operated patients closely resembled the pattern of early ovarian
development (Fig. 5B),
suggesting that these gonads were still in their initial position and had not
descended to the pelvis. Moreover, no mature ovarian tissue was found in any
of our nine surgical specimens, a previously reported finding
[3,
4,
25]. Focused histologic study
of the six available specimens disclosed features corresponding to the 5- to
6
-week embryonic stage (celomic mesothelium and oogonia) in four
specimens and to the 20-week embryonic stage (primordial follicles and germ
cells) in two [26]. In
addition, histologic examination of the pedicles showed them to be composed of
vascularized connective tissue, probably corresponding to embryonic
mesenchyma.
|
|
In view of these observations, we speculate that complex ovarian cysts may not be caused by hormonal stimulation in a mature ovary, as is generally considered, but instead result from abnormal development of the primitive gonad, probably because of disruption of its vascularization. This possibility would also explain why the ipsilateral fallopian tube, which shares with the gonad the primitive capillary vascularization, is often atretic [3, 24, 25, 27]. According to this concept, formation of the cyst would be a secondary event caused by germinal epithelial secretion in the dysgenic gonad. Twisting of the pedicle, observed in around half the reported patients [4, 5, 17, 20], would also be a secondary event and not the primary cause of the hemorrhagic content of the cysts. Experimental studies would be required to prove this etiopathologic hypothesis and to clarify the sequence of events leading to hemorrhage.
Involution of the cysts was documented in all 11 conservatively managed patients monitored by sonography, in most by the age of 1 year. The cysts showed progressive shrinkage, and an echogenic, crescent-shaped calcification appeared at the base of the cyst in seven patients. At this point of the evolution, a radiograph of the abdomen could show features in common with those of omental cysts. The ovarian origin of the cyst, which can become amputated and adhere to the omentum, may not be recognized even at surgery. Some omental cysts previously reported in girls may actually have been partially involuted hemorrhagic ovarian cysts. We had one such experience: A girl in whom an omental cyst [28] had been diagnosed was found to have a single ovary on sonographic study at the age of 21 years. Study of the ovaries either by sonography or during surgery in cases of calcified cystic abdominal masses is essential for correct diagnosis and appropriate management.
Because malignancy is uncommon in newborns, the initial option of conservative treatment does not imply substantial risk. The differential diagnosis of ovarian masses in neonates is limited, with gastrointestinal tract anomalies being the main consideration [29]. Enteric duplication cysts, which can occur at any level of the gastrointestinal tract, present a characteristic "double wall" sonographic sign due to enteric mucosa and the hypoechoic muscularis layer in 50% of patients. The "daughter cyst" sign, defined as a small cyst in the wall of a primary cyst, favors the diagnosis of an ovarian cyst [30]. Omental cysts can bleed and thus be indistinguishable from complex ovarian cysts. An omental cyst should be considered in female neonates who have a complex cystic mass and in whom both ovaries are visualized. Ovarian tumors are so rare in infants that a hemorrhagic ovarian cyst should be considered even when the mass is echogenic. Ovarian teratoma is practically nonexistent in infants less than 1 year old [31]. Cystic adenoma is also rare, with only two cases described in the literature [17].
In conclusion, conservative management with clinical and sonographic monitoring should be considered strongly for complex neonatal ovarian cysts. Surgery would be reserved for very large lesions, those causing clinical symptoms attributable to the mass, and those not involuting spontaneously. Significant surgical and histologic evidence suggests that a complex ovarian cyst is actually a form of ovarian dysgenesis. In our opinion, the implication of pedicle torsion in neonatal ovarian cysts has been overstated, thereby contributing to the controversy surrounding their management.
Acknowledgments
We thank Montserrat Martí and Roser Camós for technical assistance and Celine Cavallo and Christine O'Hara for language editing.
References
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  B H. Guvenc, B. Azman, and B. Erkus Management of ovarian cysts during infancy: autoamputation presenting as a possible pitfall BMJ Case Reports, March 17, 2009; 2009(mar08_1): bcr0720080447 - bcr0720080447. [Abstract] [Full Text]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
