Thoracic Endometriosis

Thoracic endometriosis is a clinical syndrome whereby ectopic endometrial tissue is deposited in thoracic structures. Growth of endometriotic implants is dependent on the presence of ovarian steroids. The four major presentations of thoracic endometriosis include catamenial pneumothorax, catamenial hemothorax, catamenial hemoptysis, and asymptomatic pulmonary nodules. Catamenial pneumothorax accounts for 73% of the cases of thoracic endometriosis described.

Symptoms occur within 24–48 h after the onset of menses. Current theories on the pathogenesis of thoracic endometriosis involve microembolization of endometrial tissue from pelvic veins and the movement of endometrial tissue into the pleural space via congenital diaphragmatic defects. The latter mechanism explains the right-side predilection of catamenial pneumothorax. The diagnosis is typically established clinically, as pathologic documentation is generally not required. The diagnosis rarely can be confirmed by pleural fluid cytology, cytology from bronchoscopic aspiration, and needle aspiration of lung nodules.

Elevated levels of serum CA-125 and measurements of endometrial antibody have been reported in patients with thoracic endometriosis. Treatment consists of suppression of the ectopic endometrium by interfering with ovarian estrogen secretion. Despite estrogen suppression, recurrence rates remain high, and pleurodesis should be considered in those who develop recurrent pneumothorax or hemothorax.


Endometriosis is defined as the presence of endometrial glands and stroma outside of the uterus. This clinical entity is most often diagnosed in women between the ages of 25 and 35 years. The prevalence of endometriosis in the general population is not known. The incidence of pelvic endometriosis is estimated at 1% for women undergoing surgery for all gynecologic indications but may be as high as 50% if the indication for surgery is to evaluate the cause for pelvic pain or infertility. Growth and maintenance of endometriotic implants are dependent upon the presence of ovarian steroids. Endometriosis most commonly involves structures within the pelvis resulting in dysmenorrhea, dyspareunia, and infertility. Endometrial tissue can occur outside the pelvis and has been described in the abdomen, central nervous system, skin, aorta, and thoracic cavity.

In 1958, Maurer and colleagues were the first to report thoracic endometriosis in a patient with recurrent spontaneous pneumothorax associated with menstruation. At thoracostomy, their patient was found to have a normal-appearing lung; however, the right hemidiaphragm had a notable defect, which was studded with endometrial tissue. They postulated that air entered the peritoneal cavity by way of the genital tract during menses. Air within the peritoneal cavity could transverse into the pleural space via the diaphragmatic defect to induce a pneumothorax. Since this initial description, thoracic endometriosis remains a rare disease with varying clinical presentations. To date, there are slightly more than 100 cases reported in the medical literature. The most common presentation of thoracic endometriosis is right-side pneumothorax. Hemothorax, hemoptysis, chest pain, and asymptomatic pulmonary nodule occur less frequently.

The symptoms of thoracic endometriosis are catamenial. ‘Catamenial’ is a Greek word meaning monthly or periodic. Symptoms occur with 24–48 h after the onset of menses. Chest pain is the most common symptom occurring in 90% of patients with thoracic endometriosis.

Successful treatment requires both eradication and suppression of existing thoracic endometrial tissue and prevention of reseeding from the pelvis. Suppressing ovulation using oral contraceptives, progestins, danazol, or gonadotropin-releasing hormone (GNRH) analogs are currently the most accepted medical strategies. Hysterectomy with bilateral salpingo-oophorectomy should be considered only in the patient who is also suffering from disabling endometriosis–pelvic pain. Despite estrogen suppression, pneumothorax recurrence approaches 50%. Pleurodesis should be considered in those who develop a recurrent pneumothorax or hemothorax.


The causative and genetic factors for endometriosis are not well understood. Growth and maintenance of endometrial tissue is dependent upon ovarian steroids. Previous epidemiologic studies have suggested that delayed pregnancy is a risk factor for the development of endometriosis. Therefore, Caucasian women are reported to have a higher occurrence of endometriosis compared to other races. This trend, however, was a result of this group having greater access to medical care. Most believe there is no difference between the incidence of endometriosis in different ethnic groups.

Genetic factors probably impact a woman’s susceptibility to endometriosis. A familial tendency for endometriosis has been recognized, and concordance in twins has been observed. Multiple genes interacting with environmental triggers are probably operative in conferring the disease. However, future studies are needed to identify the genes and environmental triggers involved in pathogenesis.


Thoracic endometriosis can be classified according to its clinical presentation. These clinical presentations include: (1) catamenial pneumothorax; (2) catamenial hemothorax; (3) catamenial hemoptysis; (4) asymptomatic pulmonary nodules; and (5) aortic dissection. Histopathological diagnosis of thoracic endometriosis involves identifying endometrial glands and stroma within the thoracic cavity. Usually, histopathologic documentation of thoracic endometrial tissue is not required and signs are catamenial.

Clinical Features

Of the 111 cases of thoracic endometriosis reported in the English language since 1958, catamenial pneumothorax is by far the most common presentation of thoracic endometriosis occurring in 80 (73%). Catamenial hemothorax occurred in 15 (14%), catamenial hemoptysis in 8 (7%), lung nodules in 7 (5%), and aortic dissection in 1 (<1%).

Thoracic endometriosis most commonly presents in premenopausal women with a mean age of 35 years and a reported age range of 19 to 54 years. Postmenopausal women on hormonal replacement therapy can also develop thoracic endometriosis. The vast majority of women with thoracic endometriosis will have evidence of pelvic endometriosis. In 15% of cases, pelvic endometriosis will be lacking in those with thoracic involvement. In catamenial pneumothorax, endometriotic implants are seen in less than 15% of patients examined by thoracoscopy or thoracostomy.

The symptoms of thoracic endometriosis are catamenial, presenting in 24–48 h after the onset of menses. Chest pain is the most common symptom, occurring in 90% of patients with catamenial pneumothorax. In more than 90% of cases of catamenial pneumothorax, the pneumothorax is seen on the right side. Two cases of bilateral pneumothorax and one of isolated left-side occurrence has been reported.

There should be a high index of suspicion for thoracic endometriosis in women of childbearing age who present with recurrent chest pain, pneumothorax, or hemoptysis. The diagnosis is established clinically; however, it is often delayed until multiple episodes have occurred as the patient and clinician fail to recognize the association between symptoms and the onset of menses.

Although a histopathologic documentation of thoracic endometriosis is not required, there have been reports of the diagnosis being established by cytology of pleural fluid, bronchoscopic aspirates, and needle aspiration of lung nodules. Elevated serum CA-125 (elevated CA-125 serum levels are typically seen in ovarian cancer and occasionally in benign disorders such as endometriosis) and endometrial antibody have been reported in patients with thoracic endometriosis. Pleural and pulmonary endometriotic implants have been described on computed tomography and magnetic resonance imaging scans. These scans may be negative unless obtained during menses. Bronchial arteriography has been used to establish the diagnosis in a few women. Regardless of the radiographic modality used, the appearance of thoracic endometriosis is not specific, and the diagnosis remains indirect.

Thoracoscopic findings may consist of diaphragmatic perforations and pleural implants which vary in size depending on the time of menstruation. The diaphragmatic perforations vary in size from 1mm to 2 cm and are found predominately on the right diaphragm. Macroscopically, the endometriotic implants appear as brown–yellow and sometimes red nodules surrounded by neovascularization. Although these ‘gunshot’ lesions are diagnostic of endometrial foci, pleural biopsy should be performed to confirm the diagnosis.


Several theories have been proposed to explain the pathogenesis of endometriosis. The coelomic metaplasia theory proposes that undifferentiated cells within the peritoneal cavity are capable of dedifferentiating into endometrial tissue. The implantation theory states that endometrial tissue from the uterus shed during menstruation is transported by the fallopian tubes and implants on pelvic structures. Another theory involves retrograde menstruation. This theory has been supported by the observation that females with genital tract obstruction have an increased likelihood of tubal reflux; however, the incidence of retrograde menstruation is similar among those with and without endometriosis. Therefore, the development of endometriosis may be dependent upon the quantity of endometrial tissue reaching the peritoneal cavity as well as the capacity of the immune system to recognize and destroy the displaced endometrial cells.

Some authors have speculated that altered immunity may play a role in the development of endometriosis. An inability of cellular immunity to recognize the presence of endometrial tissue outside the uterus may be pivotal. Various cytokines and proinflammatory cells may lead to the proliferation of endometriotic implants.

Endometriosis in locations outside the pelvis is explained by dissemination of endometrial cells through pelvic lymphatics and veins. This microembolization phenomenon explains why endometrial tissue can be found in locations like the skin, central nervous system, vertebrae, and aorta. However, in catamenial pneumothorax, more than 90% of cases occur on the right side, suggesting another plausible mechanism for this form of thoracic endometriosis. Congenital diaphragmatic holes are more commonly found on the right. These holes allow for a direct peritoneal–pleural communication, thereby allowing for endometrial tissue to preferentially migrate into the right pleural space. However, fewer than 25% of patients have visible thoracic implants during thoracoscopy. Others have theorized that air originates from abdominal endometriosis which can then move preferentially through these diaphragmatic defects into the pleural space.

Less-accepted theories involve high circulating levels of prostaglandin F2 during menstruation causing bronchoconstriction and leading to alveolar rupture and pneumothorax. Another speculates that swelling of intrapulmonary endometriotic implants at menses causes airway obstruction resulting in alveolar wall rupture and pneumothorax. Neither of theses theories, however, explains the right-side predilection for catamenial pneumothorax.

Animal Models

Animal models have shown that intravenous injected endometrial tissue remain viable, invade lung parenchyma, and undergo cyclic changes during the menstrual cycle.

Management and Current Therapy

Successful treatment requires both eradication and suppression of existing thoracic endometrial tissue and prevention of reseeding from the pelvis. Medical treatment, the primary modality, consists of suppression of the ectopic endometrium by interfering with ovarian estrogen secretion. This can be accomplished by suppressing ovulation using oral contraceptives, progestins, danazol, or GNRH analogs. We do not routinely recommend hysterectomy with bilateral salpingo-oophorectomy unless the patient is also suffering from disabling endometriosis–pelvic pain.

Progestins cause decidualization of endometrial tissue. Adverse side effects are problematic and include nausea, fluid retention, abnormal uterine bleeding, and depression.

Danazol, an isoxazol derivative of 17a-ethinyltesterone, induces anovulation by attenuating the mid-cycle surge of luteinizing hormone secretion, increasing free testosterone concentration, and inhibiting multiple enzymes in the steroid pathway. The recommended dose of danazol in the treatment of endometriosis is 600–800 mg daily. At these doses, danazol has significant androgenic side effects, and irreversible liver toxicity has been reported.

GNRH agonists decrease the secretion of follicle stimulating hormone and luteinizing hormone, resulting in hypogonadotropic hypogonadism. A persistent hypoestrogenic state ensues resulting in endometrial tissue atrophy and amenorrhea. Adverse effects associated with GNRH agonists include vaginal bleeding, hot flashes, decreased libido, depression, osteoporosis, and sleep disturbance.

All medical regimens appear to be effective in the treatment of endometriosis. Several case reports have shown that GNRH agonists were effective in the treatment of thoracic endometriosis when oral contraceptives had failed. Nevertheless, recurrence rates are as high as 50% with hormonal therapy. This suggests that endometrial implant regression is not complete and/or that recurrent embolization continues.

The management of catamenial pneumothorax is similar to the acute management of other forms of spontaneous pneumothorax. Observation is appropriate for a small, asymptomatic pneumothorax. Catheter aspiration or closed tube thoracostomy should be considered for patients with symptoms or larger pneumothoraces.

With these conservative approaches, pneumothorax recurrence will occur in all who do not receive appropriate estrogen suppression therapy. Moreover, despite estrogen suppression, pneumothorax recurrence approaches 50% over the next 18 months. Therefore, pneumothorax prevention strategies should be considered in those who develop a recurrent pneumothorax.

Chemical pleurodesis (via tube thoracostomy, thoracoscopic talc poudrage, or surgical pleurodesis using pleural abrasion and partial parietal pleurectomy with or without talc poudrage) is highly effective in preventing catamenial pneumothorax and hemothorax. However, patients may continue to experience catamenial chest pain despite these therapeutic interventions. Catamenial chest pain is presumed to be due to cyclic proliferation of pleuropulmonary endometriotic implants in response to ovarian steroids. These symptoms can be relieved with hormonal suppression but recur if estrogen replacement therapy is initiated.

It is well recognized that women suffering from thoracic endometriosis are in their reproductive years and may wish to conceive children. The use of hormonal ablation strategies would inhibit conception. On the other hand, pregnancy would be a protective state in preventing the occurrence of a pneumothorax. In women with thoracic endometriosis who are considering pregnancy, a pleurodesis procedure can be performed prior to discontinuation of estrogen suppression.

Surgical resection of intraparenchymal implants and neodymium–yttrium aluminum garnet (Nd– YAG) laser resection of endobronchial implants have been successful in a limited number of reports. These strategies should be entertained in those with limited disease who have recurrent hemoptysis despite appropriate estrogen suppression.

Further Reading

Elliot DL, Barker AF, and Dixon LM (1985) Catamenial hemoptysis: new methods of diagnosis and therapy. Chest 87: 687–688.

Fonseca P (1998) Catamenial pneumothorax: a multifactorial etiology. Journal of Thoracic and Cardiovascular Surgery 116: 872–873.

Hibbard LT, Schumann WR, and Goldstein GE (1981) Thoracic endometriosis: a review and report of two cases. American Journal of Obstetrics Gynecology 140: 227–232.

Joseph J and Sahn SA (1996) Thoracic endometriosis syndrome: new observations from an analysis of 110 cases. American Journal of Medicine 100: 164–170.

Lillington GA, Mitchell SP, and Wood GA (1972) Catamenial pneumothorax. Journal of the American Medical Association 219: 1328–1332.

Olive DL and Pritts EA (2001) Treatment of endometriosis. New England Journal of Medicine 345: 266–275.

Olive DL and Schwartz LB (1993) Endometriosis. New England Journal of Medicine 328: 1759–1769.

Sahn SA (2004) Thoracic endometriosis. Up to Date American Thoracic Society (electronic media).

Wilkins SB, Bell-Thomson J, and Tyras DH (1985) Hemothorax associated with endometriosis. Journal of Thoracic and Cardiovascular Surgery 89: 636–638.

Wood DJ, Krishnan K, Stocks P, et al. (1993) Catamenial haemoptysis: a rare cause. Thorax 48: 1048–1049.

Yamazaki S, Ogawa J, Koide S, et al. (1980) Catamenial pneumothorax associated with endometriosis of the diaphragm. Chest 77: 107–109.