Pathophysiology and Clinical Implications of Ovarian Endometriomas

Ovarian endometriomas often require surgical management for diagnosis and treatment, with an emerging role for fertility optimization and cancer prevention.

Abstract

Ovarian endometriomas affect many patients with endometriosis and have significant effects on quality of life, fertility, and risk of malignancy. Endometriomas range from small (1–3 cm), densely fibrotic cysts to large (20 cm or greater) cysts with varying degrees of fibrosis. Endometriomas are hypothesized to form from endometriotic invasion or metaplasia of functional cysts or alternatively from ovarian surface endometriosis that bleeds into the ovarian cortex. Different mechanisms of endometrioma formation may help explain the phenotypic variability observed among endometriomas. Laparoscopic surgery is the preferred first-line modality of diagnosis and treatment of endometriomas. Ovarian cystectomy is preferred over cyst ablation or sclerotherapy for enabling pathologic diagnosis, improving symptoms, preventing recurrence, and optimizing fertility outcomes. Cystectomy for small, densely adherent endometriomas is made challenging by dense fibrosis of the cyst capsule obliterating the plane with normal ovarian cortex, whereas cystectomy for large endometriomas can carry unique challenges as a result of adhesions between the cyst and pelvic structures. Preoperative and postoperative hormonal suppression can improve operative outcomes and decrease the risk of endometrioma recurrence. Whether the optimal management, fertility consequences, and malignant potential of endometriomas vary on the basis of size and phenotype remains to be fully explored.

Endometriosis is a systemic, estrogen-dependent, inflammatory condition of endometrium-like glands and stroma found outside the uterus. Endometriotic cysts, or endometriomas, are found nearly exclusively in the ovaries.^1,2 This differs from deep infiltrating endometriosis, superficial implants, and solid nodules, which are observed elsewhere inside and outside the pelvis.³ Here we review the etiology, evaluation, and treatment of endometriomas (Figs. 1–3).

Fig. 1. Small endometrioma filled with dark brown material type I (solid arrow) superimposed on larger endometrioma type IIB (dashed arrow).

Nezhat. Ovarian Endometriomas. Obstet Gynecol 2024.

Fig. 3. Bilateral type IIC endometriomas with dense adhesions.

Nezhat. Ovarian Endometriomas. Obstet Gynecol 2024.

Fig. 2. Large type IIA endometrioma. Cortical ovarian endometriosis with corpus luteum.

Nezhat. Ovarian Endometriomas. Obstet Gynecol 2024.

ETIOLOGY OF ENDOMETRIOMAS

The 17th century anatomist Frederik Ruysch was among the first to suggest that refluxed blood from the uterus could be the source of hemorrhagic ovarian cysts. This theory was reintroduced in the early 20th century by John Sampson, whose work systematically studying the clinical and histopathologic features of “chocolate cysts” gave rise to a modern understanding of endometrioma formation.⁴ In 1921, Sampson⁵ proposed that endometriomas arise from functional cysts that rupture or perforate and are subsequently invaded by endometrial tissue through the perforation. As the invading endometrial epithelium advances, Sampson theorized, the luteal cyst lining regresses. Sampson described a series of cysts possessing, in varying proportions, both endometrial and luteal lining. He speculated that cysts containing relatively less endometrial tissue represented earlier stages of endometrioma development, whereas cysts lined completely by endometrial tissue represented a later stage of development.

Sampson⁵ conjectured that cyst rupture events were critical to both initiation of endometriomas and formation of surrounding adhesions. He observed a correlation between the location of the endometrial lining within a cyst and suspected sites of rupture. Luteal lining, if present, was generally found in the deeper portion of the cyst away from the suspected rupture. These findings supported his theory of cyst invasion by endometrial tissue through a prior site of rupture. In 1927 and 1940, Sampson published theories of endometriosis spread through the venous circulation and, more famously, through retrograde menstruation.^6–8 These theories on the genesis of endometriosis broadly suggested sources of the endometrial tissue that would come to perforate functional ovarian cysts and form endometriomas.

In contrast to Sampson's theory of cyst invasion by metastatic endometrial tissue, studies by Czernobilsky and Morris⁹ in 1979 and Nisolle-Pochet et al¹⁰ in 1988 led to an alternative, competing theory that metaplastic ovarian surface epithelium could be the source of the epithelial lining of chocolate cysts. In histologic series of endometriomas, both groups found, somewhat unexpectedly, fallopian tube–like epithelium in a subset of endometriomas. At the time, the ovarian surface epithelium was believed to possess high intrinsic potential for müllerian metaplasia, including to endosalpingeal, endometrial, and endocervical epithelium. This potential was thought to be particularly pronounced within the hormonal and chemical milieu of ovarian cysts.^11,12 More recent data have challenged the assumption that the ovarian surface epithelium possesses such significant müllerian metaplastic potential.^13,14 At the time, the finding of concurrent endosalpingeal and endometrial epithelium inside endometriomas led the authors to conclude that metaplastic ovarian surface was the common source of endometrioma lining.⁹

Histologic studies describing endometriomas that feature luteal lining and invasive endometriosis implants are consistent with a hypothesis of endometrioma origin as follicular or luteal ovarian cysts that become involved by endometriotic tissue, whether through invasion or metaplasia.^3,5,15–19 This origin theory is further supported by the findings that large endometrioma formation can be prevented with hormonal therapies that suppress ovulation, and inversely, large endometrioma formation is not uncommon among patients with endometriosis undergoing ovarian hyperstimulation for assisted reproduction.^20,21 An alternative mechanism of endometrioma formation, in contrast to that of invasion or metaplasia of preexisting cysts, is through ovarian surface endometriosis that bleeds or invaginates into the ovarian stroma, creating in particular small and densely fibrotic endometriomas.^3,19,22,23 Histologic findings among some small cysts showing a lining of exclusively endometrial-type tissue and no luteal tissue support this alternative origin hypothesis for a subset of cysts.³ Based on the findings of a systematic clinical–pathologic study of a large series of endometriomas by Nezhat et al, a classification schema has been proposed diving endometriomas into type I and type II cysts (A, B, and C). Type I cysts (primary) encompass small, densely fibrotic cysts proposed to arise from ovarian surface endometriosis that bleeds into the stroma, and type II cysts (secondary) are larger endometriomas proposed to originate as follicular or luteal cysts invaded by endometriosis (Figs. 1–3).³

A fibrotic reaction to bleeding endometrial tissue is believed to create characteristic adhesions between endometriomas and the surrounding ovary and other pelvic structures, which can range from small focal areas of adhesion to thick and densely adherent capsules enveloping the cyst. The cyst wall of endometriomas is rarely more than 2 mm thick.¹⁵ A study involving ovarioscopy and direct biopsy of a series of large endometriomas has demonstrated that fibrosis and adhesions colocalize with the location of endometriotic tissue and away from functional tissue inside these cysts.²⁴ Molecularly, the process of fibrosis has been found to implicate reactions to free iron and reactive oxygen species, as well as cellular factors, including transforming growth factor-β, Wnt1, activin A, connective tissue growth factor, sphingosine-1-phosphate, and plasminogen activator inhibitor-1.^25–27

With few exceptions, endometriomas are found exclusively in the ovaries and not in other anatomic locations. Fibrogenesis and scar formation around endometriotic implants in other areas of the pelvis may help explain why large cysts are not found elsewhere in the pelvis, whereas the relatively loose connective tissue of the ovarian medulla, dynamic expression of genes regulating extracellular matrix hardness, rapid vascular remodeling, and basement membrane mechanics may contribute to the biomechanic hospitality of the ovary to cyst formation.^28–30 The enriched concentration of hormones, growth factors, and cytokines in the ovary that is not present elsewhere has also been proposed to drive endometrioma growth in the ovaries specifically.²⁹

TREATMENT INDICATIONS

Goals of treatment for endometrioma include relieving symptoms, ruling out malignancy, treating or preventing ovarian torsion, improving fertility outcomes, preserving ovarian hormonal function, and minimizing recurrence. Because ovarian endometriomas are rarely isolated and more commonly occur concurrently with nonovarian endometriosis, surgical excision of both ovarian and nonovarian lesions is often indicated. In some circumstances, surveillance may be appropriate for asymptomatic patients who are evaluated to be at low risk for torsion or malignancy on the basis of mass size (generally less than 5 cm) and appearance and patient characteristics, including age and family history.³¹

MEDICAL MANAGEMENT

Hormonal treatment of endometriomas may be associated with cyst size reduction and symptom improvement but is unlikely to impart cyst resolution.^32–37 A recent meta-analysis has comprehensively reviewed the effect of medical management on ovarian endometrioma size and concluded that dienogest, oral contraceptive pills, gonadotropin-releasing hormone agonists, norethindrone acetate, and danazol have all been shown to reduce cyst diameter by a mean of 0.6–1.95 cm, with oral contraceptive pills having been studied most extensively and danazol showing the largest mean reduction in cyst size.³⁸ Medical management alone foregoes the opportunity for histologic diagnosis, and surgery is generally preferred for both diagnosis and treatment.^39,40 However, medical therapy before surgery can provide symptom management and improve operative outcomes. For most patients, we recommend hormonal suppressive therapy for at least 4–8 weeks before surgery to prevent ovulation and corpus luteum formation and to decrease risks of bleeding and ovarian damage during cystectomy.

SURGICAL MANAGEMENT

Definitive diagnosis of endometrioma is histologic, and surgery can be both diagnostic and therapeutic. Since the 1980s, video laparoscopy with and without robotic assistance has become the preferred approach over laparotomy on account of improved operative outcomes and visualization of anatomy.^41–44 In 1986, Nezhat et al⁴⁴ reported video laparoscopic management of all stages of endometriosis, including endometriomas. Risk of conversion to open surgery is low but relatively higher for patients with endometriomas than with other types of adnexal masses (5.1% vs 3.5%).⁴⁵

Surgical approaches include ovarian cystectomy with stripping of the cyst wall, cyst drainage and ablation, and sclerotherapy.^46,47 Cyst aspiration alone is associated with high recurrence rates (greater than 80%) and is not recommended as sole treatment.⁴⁸ Cystectomy is associated with reduced rates of recurrence, reoperation, dysmenorrhea, dyspareunia, and nonmenstrual pelvic pain, as well as increased spontaneous pregnancy rate, compared with ablation.^49–54 In the short term, postoperative anti-müllerian hormone decline may be the highest after cystectomy, but cystectomy is also associated with higher postoperative conception rates and lower rates of cyst recurrence compared with ablation and sclerotherapy.⁴⁹

Endometrioma surgery begins with inspection of the pelvis and adhesiolysis. Peritoneal washings should be collected before cyst manipulation.⁵⁵ Cystectomy typically proceeds with creation of an incision along the cyst away from the hilum.^22,55 Laparoscopic graspers are then used to create traction and countertraction and to strip the cyst wall from the ovary. Care should be taken to avoid excess trauma to the normal ovarian tissue and damage to the ovarian hilum. Injection of dilute vasopressin under the cyst capsule defines the plane between the cyst and the normal ovary and decreases bleeding. To achieve hemostasis of the ovarian bed, hemostatic sealants, suture, and sparing electrosurgery may be used. A number of small, randomized trials have demonstrated greater decline in ovarian reserve markers with the use of electrosurgery for hemostasis compared with hemostatic agents or suture.^56,57

Small, densely fibrotic endometriomas (type I) often must be removed piecemeal because of circumferential fibrosis. In contrast, large, less adherent endometriomas (type II) often can be stripped more easily from the surrounding ovarian stroma. The oldest endometriotic portion of the cyst, which may be very adherent to the ovary, is typically the most difficult portion to excise.²²

In clinical scenarios in which cystectomy would result in excessive loss of healthy ovarian tissue, cyst drainage and vaporization or ablation may be appropriate.²² When possible, as much of the cyst wall as possible is removed and sent for histologic evaluation, and vaporization or ablation is used to destroy any remaining foci of endometriotic tissue. The choice of vaporization or ablation (CO₂ laser, plasma jet, or radiofrequency) is guided by surgeon expertise, and all techniques may not be widely available.⁵⁸ In general, the depth of penetration and level of tissue trauma are least with CO₂ laser or plasma jet and highest with radiofrequency. Excess damage to adjacent ovarian tissue can theoretically be avoided by limiting exposure time and periodically cooling the ovary with irrigation fluid.^59,60

Sclerotherapy is a less common modality of endometrioma treatment in the United States but has been studied extensively in the global literature.⁶¹ Sclerosing agents function by direct damage to cyst wall epithelial tissue and to the endothelium of feeding blood vessels.⁶² Through either a laparoscopic or an ultrasound-guided approach, the endometrioma is punctured, aspirated, irrigated, and instilled with a sclerosing agent (ethanol 20–100%, tetracycline 1–5%, or methotrexate 1%).^61,63,64 Ethanol dosing should be limited to 1 mL/kg to avoid alcohol intoxication.⁶² The sclerosing agent is left in situ or reaspirated. Several studies have demonstrated decreased recurrence if an ethanol indwelling time of at least 10 minutes is used to maximize tissue necrosis.^62,65–67 In a cyst that is densely adherent, adhesiolysis resulting in multiple cyst perforations may complicate the ability to achieve adequate indwelling time. The most common complication of sclerotherapy is abdominal pain; rare complications include postoperative fever and abscess formation.⁶¹

Recurrence of endometrioma after surgical management is on the order of 30–50%.^68–70 Risk factors for cyst recurrence include younger age, prior treatment of endometriosis, larger cyst size, bilateral ovarian involvement, ablation or sclerotherapy rather than cystectomy, and overall burden of endometriosis at surgery.^68,71,72 Ovulation inhibition, through either postoperative hormonal treatment or pregnancy, has been shown to reduce the risk of recurrence, and medical ovulation inhibition should be considered in patients who do not desire immediate pregnancy.^{39,69,73–76}

FERTILITY CONSIDERATIONS

Endometriosis in general is associated with infertility, although a causal relationship has not been proven.⁷⁷ The specific contributions of endometriomas and other features of endometriosis to infertility are difficult to separate. Endometriomas contain increased concentrations of free iron, reactive oxygen species, and proteolytic enzymes that have been demonstrated to negatively affect surrounding ovarian tissue.²⁷ A number of studies have documented decreased anti-müllerian hormone levels in patients with unoperated endometriomas.⁷⁸ Among a series of patients with unilateral endometriomas undergoing oocyte retrieval, a decreased response to ovarian stimulation was observed on the endometrioma-containing ovary compared with the contralateral ovary.⁷⁹ However, patients with and without endometriomas undergoing assisted reproduction cycles appear to have similar live birth rates.⁸⁰ This may be caused in part by compensation by the contralateral ovary in patients with unilateral endometriomas, as well as a selection bias in which patients with endometriomas are more likely to cancel assisted reproduction cycles.

Cystectomy is associated with an increased spontaneous pregnancy rate for patients with a history of subfertility. For patients planning medically assisted reproduction for endometriosis-related infertility, no benefit of endometrioma surgery on pregnancy rate or live birth rates has been shown in two large meta-analyses.^80,81 In addition, the rate of cycle cancellation attributable to poor ovarian response or failed oocyte retrieval has been shown to be significantly higher among patients who underwent surgery for endometrioma compared with patients who did no undergo surgery.⁸² The European Society of Human Reproduction and Embryology guidelines (2022) recommend against routine surgery for endometrioma before assisted reproduction cycles for the sole purpose of improving reproductive outcomes, given the lack of benefit and potential for harm, although surgery may still be appropriate for treatment of symptoms or in cases in which the location of cysts prevents oocyte retrieval.^39,83 In contrast, the American Society for Reproductive Medicine 2012 guidelines recommend that patients with American Society for Reproductive Medicine stage III–IV endometriosis with endometrioma larger than 4 cm should consider cystectomy for diagnosis, oocyte access, and possibly ovarian response.^32,77 Of note, some studies comparing pregnancy rates after second endometriosis surgery with proceeding with in vitro fertilization have not shown a benefit of repeat surgery.^84–86

SURGICAL CONSIDERATIONS FOR PATIENTS NOT DESIRING FERTILITY

For patients not desiring fertility, oophorectomy is sometimes considered in place of cystectomy, particularly in the setting of large or recurrent cysts. Patients should be counseled on the risk of contralateral ovary recurrence after unilateral oophorectomy. In one study of 50 patients who underwent laparoscopic unilateral salpingo-oophorectomy for unilateral endometrioma, the risk of recurrence on the contralateral ovary during the first 5 years after unilateral salpingo-oophorectomy was 24.7%.⁸⁷

A distinct benefit of salpingo-oophorectomy is cancer risk reduction. A population-based study in Sweden found a significant negative association between history of oophorectomy or history of radical endometriosis surgery and ovarian cancer incidence after controlling for salpingectomy, hysterectomy, and use of medical therapies.⁸⁸ In addition, one retrospective case series of 33 patients in Japan with histologically confirmed ovarian cancer arising in the setting of an endometrioma found that 6 of the 33 patients with ovarian cancer had a history of cystectomy of the ipsilateral ovary, suggesting that cystectomy alone was not protective against ovarian cancer in those cases.⁸⁹ The majority of cancers were of clear cell or endometrioid histology, consistent with the well-established association between endometriosis and these ovarian cancer subtypes.^90–93

An additional opportunity for ovarian cancer risk reduction at the time of endometrioma surgery for patients who have completed childbearing is concurrent bilateral salpingectomy. Histologic and molecular studies over the past 20 years have provided compelling evidence that the high-grade serous subtype of ovarian cancer commonly originates in the fallopian tube epithelium.⁹⁴ Accordingly, retrospective studies show a risk reduction for ovarian cancer after bilateral salpingectomy on the order of 65%.⁹⁵ An increasing number of international societies recommend that patients who do not desire future fertility be counseled on the option to have concurrent bilateral salpingectomy at the time of other gynecologic surgery.^96–98

CONCLUSIONS

Endometriomas affect up to 6% of patients with ovaries and carry risks of chronic pain, infertility, pregnancy complications, and potential for malignant transformation. Two pathways of endometrioma formation have been proposed, endometriotic invasion or metaplasia of preexisting functional ovarian cysts and bleeding of ovarian endometriotic implants directly into the ovarian cortex, that may account for the phenotypic variability observed among endometriomas. Future research evaluating the recapitulation of these mechanisms at a molecular level is anticipated to shed light on the genesis of endometriomas, with likely implications for the prevention of endometrioma formation. Whether endometriomas of different phenotypes have identifiably disparate effects on fertility or risk of malignancy is an important clinical question that has not been fully explored.

Medical management may shrink endometrioma size but rarely leads to cyst resolution and forgoes the opportunity for pathologic diagnosis; therefore, surgical management is often indicated for purposes of diagnosis and symptom management. Benefits of surgery for fertility optimization and cancer risk reduction have been suggested by retrospective studies. The precise role of surgery for these goals is yet to be defined, particularly with regard to a stratification of benefit by cyst phenotype or size. Moving forward, continued attention to cyst phenotype or size in the design of studies on the management and implications of endometriomas will continue to better inform ability to make evidence-based clinical recommendations tailored to patients' unique clinical presentations.