Background: A significant percentage of pneumothorax in women is due to thoracic endometriosis. Pathophysiologic mechanisms continue to be debated, and pathologic aspects are poorly known.
Methods: Clinical and pathologic records of all consecutive women of reproductive age operated on for pneumothorax between 2000 and 2011 were retrospectively reviewed.
Results: Two hundred twenty-nine women (mean age, 33 years) underwent surgery. One hundred forty-four cases (63%) were right-sided, and pneumothoraces were catamenial for 80 women (35%). Diagnosed pelvic endometriosis was associated in 29 cases. At pathology, thoracic endometriosis was diagnosed in 54 cases (24%). Endometrial glands were observed in 33 of 54 cases and were often cystic (16 of 33). Stroma was observed in 51 of 54 cases and endometrial stroma without glands in 21 cases. Hemosiderin-laden macrophages were observed in 27 of 54 cases. All cases of thoracic endometriosis were positive for progesterone and/or estrogen receptors (intense and nuclear). Catamenial pneumothoraces (n = 80, 34.9%) were endometriosis related in 50% of cases (n = 40, 17% of the whole population). Pneumothoraces were noncatamenial but endometriosis related in 6% of cases (n = 14) and merely idiopathic in 60% of patients (n = 135). Multivariate analysis showed that right side, presence of diaphragmatic abnormalities, relapse after unilateral surgery, and presence of hemosiderin-laden macrophages were independent variables associated with thoracic endometriosis (all, P < .02). Apical emphysema-like changes were found in 184 of the 213 patients (86%) with apical resection and were significantly associated with the absence of thoracic endometriosis (P < .001).
Conclusions: In women with surgically treated pneumothorax, prevalence of catamenial/endometriosis-related pneumothorax is high. Clinicians and pathologists must be aware to recognize such a difficult diagnosis.