Pathological diagnosis of thoracic endometriosis

A 26-year-old woman, who underwent abdominal surgery because of pelvic endometriosis, suffered from upper abdominal pain, fever and dyspnoea 2 days postoperatively. Paralytic ileus and right-sided pneumothorax were revealed. Treatment with a chest drain was not successful and, thus, a video-assisted thoracoscopic surgery was performed, revealing endometriosis-like lesions. Basic histopathology did not confirm the visual diagnosis, but additional immunohistochemical staining for oestrogen and progesterone receptors showed positive reaction in epithelial lung cells, thus proved the diagnosis thoracic endometriosis. A resection of the apex of the right upper lobe and pleurodesis by talc poudrage was performed after which a mesh graft was applied on the diaphragm. After 5 years of follow-up, no recurrent pneumothorax occurred.

pneumothorax, cardiothoracic surgery, pathology, obstetrics and gynaecology

Endometriosis-related pneumothorax, mainly catamenial and, thus, related to the menstrual cycle, is a rare and complex clinical presentation of thoracic endometriosis in absence of a respiratory disease. The classic patient is a woman in the reproductive age with symptoms of coughing, dyspnoea and chest pain associated to the menstrual cycle.1 Pelvic endometriosis is involved in 30%–50% of cases with thoracic endometriosis.2 After surgical removal of the endometriosis lesions in the thorax and pleurodesis, the recurrence rate of pneumothorax is decreased by 20% (33.3% to 12.9%).3 4 Morbidity can be decreased by early diagnosis and therapy.5 As the histological diagnosis in pelvic endometriosis is mainly straightforward, diagnostic difficulties may appear when examining thoracic biopsies, due to a lack of specific features microscopically. Pathological diagnosis of endometriosis is based on the typical presence of endometriosis glands, stroma and haemosiderin-laden macrophages (figure 1).1 2 In thoracic endometriosis, histological diagnosis and interpretation are impaired due to small biopsies with scattered clusters of endometrial stromal cells and the absence of glandular or stromal components.1 2 Furthermore, haemorrhage and menstrual changes leading to histiocyte infiltration can obscure typical characteristics of the stroma.6 Therefore, it is necessary to have additional possibilities to diagnose thoracic endometriosis. In a large cohort of 246 women with spontaneous pneumothorax in reproductive age, only 18 cases (7.3%) were histologically proven to be pneumothorax caused by thoracic endometriosis.1 This incidence in women in reproductive age seems an underestimation in diagnosis as Legras et al stated that 41% of spontaneous pneumothoraxes of women in reproductive age are catamenial and 24% of all pneumothoraxes are caused by thoracic endometriosis.7 The aim of this report is to describe the value of additional oestrogen receptor (OR) and progesterone receptor (PR) staining to confirm the diagnosis of thoracic endometriosis by presenting a case and review of available literature on histopathologic examination and additional staining with OR and PR antibodies. Case presentation A 26-year-old woman with increasing complaints of dysmenorrhea, dyschesia, dyspareunia, dysuria and secondary subfertility for 3 years was seen at the Endometriose in Balans clinic at the Haaglanden Medical Center, a specialised centre for women with endometriosis. A transvaginal ultrasound revealed a so-called frozen pelvis (American Society of Reproductive Medicine (ASRM)stage 4), enlarged ovaries probably due to endometriosis, adenomyosis and signs of deep invasion endometriosis (DIE) invading the muscularis layer of the sigmoid. The MRI scan confirmed the results of the transvaginal ultrasound: DIE with involvement of the sigmoid colon. After a preoperative regimen of oral contraceptives followed by leuprolide for 3 months, the endometriosis was treated by laparoscopic surgery, removing all endometriosis from the left ureter, left uterosacral ligament and sigmoid colon and by performing an appendectomy (Enzian score A0 B0 C1 FU FI FO). The second day postoperatively, the patient suffered from pain in the upper part of the abdomen, fever and dyspnoea. An abdominal CT scan and subsequent chest X-ray revealed a paralytic ileus and right-sided hydropneumothorax. Conservative treatment with a chest drain was not successful. The thoracic surgeon was consulted and a video-assisted thoracoscopic surgery (VATS) was performed. During the procedure, typical endometriosis lesions on the visceral and parietal pleura (figure 2A) and multiple diaphragmatic perforations (figure 2B) were observed. A resection of the apex of the right upper lobe and pleurodesis by talc poudrage was performed. A mesh graft was applied on the diaphragm. Pathological examination of the lung biopsies showed normal alveoli with enlargements of alveolar spaces and chronic inflammation with bleeding spots without histopathologic signs of endometriosis. Finally, immunohistochemical staining for OR and PR showed positive reaction in epithelial lung cells (figure 3) and proved the presence of thoracic endometriosis. After 5 years of follow-up, no recurrent pneumothorax occurred. Investigations In April 2020, we searched in the database PubMed for studies on catamenial pneumothorax or endometriosis-related pneumothorax using the MESH terms and keywords ‘catamenial pneumothorax’, ‘endometriosis’ and ‘pathology’. As search limit, only studies published in English and concerning humans were included. After the literature search, all titles and abstracts were assessed by one observer (SK). Only studies that described the histopathological examination including immunohistochemical staining in catamenial pneumothorax or endometriosis-related pneumothorax were included. Catamenial pneumothorax was defined as spontaneous pneumothorax within 72 hours after menstruation started. The following data were extracted: number of cases and the method of histopathology and immunohistochemical staining. Outcome and follow-up The literature search identified 80 results. After screening of title and abstract, 32 articles remained for full-text review of which 10 studies were eligible to include. Main reasons for exclusion were lack of description of how pathology was performed or absence of immunohistochemical staining (figure 4). The included studies consisted of 408 women of which 365 had catamenial pneumothorax and 43 had non-catamenial but endometriosis-related pneumothorax. In all studies, histopathological analysis included immunohistochemical staining for OR, PR and/or CD10. Overall, a total of 111 (61.7%) women were diagnosed with thoracic endometriosis by histopathology with H&E staining. On the other hand, when additional staining for ER, PR and/or CD10 was performed, respectively, 148 (82.2%), 158 (87.8%) and 92 (51.1%) patients could be diagnosed (table 1). Table 1. | Author | Range of ages in years | Number of catamenial pneumothorax cases | Histopathology (H&E) | Immunohistochemical staining | || | OR | PR | CD10 | |||| | Mataca et al8 | 25–28 | 3/3 | 0 (0%) | 3 (100%) | 3 (100%) | 3 (100%) | | Kawaguchi et al9 | 8–62 | 12/55 | 1 (1.8%) | 37 (67.3%) | 37 (67.3%) | 4 (7.3%) | | Maniglio et al16 | 37 | 1/1 | 1 (100%) | 1 (100%) | – | – | | Furuya et al17 | 22 | 1/1 | 1 (100%) | 1 (100%) | 1 (100%) | 1 (100%) | | Ghigna et al1 | 16–54 | 9/18 | 8 (44.4%) | 18 (100%) | 18 (100%) | – | | Haga et al2 | 24–50 | 84/84 | 84 (100%) | 74 (88.1%) | 84 (100%) | 74 (88.1%) | | Fang et al18 | 45 | 1/1 | 1 (100%) | – | – | 1 (100%) | | Kumakiri et al19 | 29–47 | 11/11 | 9 (81.1%) | 9 (81.1%) | 9 (81.1%) | 9 (81.1%) | | Morcos et al20 | 29 | 1/1 | 1 (100%) | 1 (100%) | 1 (100%) | – | | Flieder et al10 | 33–38 | 5/5 | 5 (100%) | 5 (100%) | 5 (100%) | – | | Total | 180 | 111 (61.7%) | 148 (82.2%) | 158 (87.8%) | 92 (51.1%) | H&E, Hematoxylin and Eosin; OR, Oestrogen Receptor; PR, Progesterone Receptor. In the study from Mataca et al, only the 3 women with catamenial pneumothorax showed positive coexpression of OR, PR and CD10, all the other women with non-catamenial pneumothorax and no endometriosis-related pneumothorax (N=50) were negative for these markers.8 Kawaguchi et al showed that in the group positive for both OR and PR, two different staining patterns were observed, a scattered and aggregated pattern.9 The scattered pattern cases, which comprised 89.2% (N=33) of the OR-positive and PR-positive cases, were all negative for CD10 and the aggregated pattern cases (11.8%, N=4) were positive for CD10. The study also included a control group of 34 men with pneumothorax. Biopsies were stained and revealed 22 positive cases for OR, PR and CD10, which all showed the scattered pattern. The other 12 cases stained negative for OR, PR and CD10 and had the aggregated pattern. Other additional receptors like CK20, CEA and SMA all stained negative in thoracic endometriosis biopsies.10

This case report and review describes the difficulty to diagnose thoracic endometriosis as underlying cause of (recurrent) pneumothorax and the helpful role of OR and PR staining. Most of the included studies used immunohistochemical staining with OR and PR antibodies, but also CD10 was in some cases used to prove thoracic endometriosis. Although CD10 staining seems to be less relevant because of its lack in specificity, it can be useful to diagnose thoracic endometriosis because it stains positive in endometrial stromal cells. So even without the presence of endometrial glands, thoracic endometriosis can still be identified. Staining with other receptors does not seem to be useful in identifying thoracic endometriosis.10 Only the study from Haga et al showed inconsistency between the number of OR-positive and PR-positive stained cases, for which no explanation was provided.2 Dixit et al studied the changes in OR and PR in endometrium and found a significant decreased amount of PR in infertile patients.11 This could possibly be an explanation of the difference in number of OR-positive and PR-positive cases in the study from Haga et al, however, infertility is not mentioned in the article, so this remains unknown. None of the included studies in this review suggested to use only OR or PR staining instead of both receptors. Since most studies showed the same number of positive cases with both OR and PR staining, it could be a possibility to just use one of these markers in the future. Minimal literature exists on the histopathological diagnosis of thoracic endometriosis. Some of the studies we included showed no additional value of immunohistochemical staining besides H&E staining, whereas other studies proved the beneficial utility. The percentage of proved thoracic endometriosis cases varied substantially per study in this review but overall more thoracic endometriosis could be diagnosed with additional staining. This variation in percentage of proved diagnoses between the studies included in this review could be explained by the presence of selection bias as in most studies the selection of women with catamenial pneumothorax was not well described. In addition, the criteria for histopathological diagnosis were not mentioned in all studies. In two studies included in this review, the high percentage of proved diagnoses after OR, PR and CD10 staining in women with non-catamenial pneumothorax suggests that the actual number of thoracic endometriosis cases is even much higher and indicates that thoracic endometriosis should always be considered in women in the reproductive age.1 9 Though, Kawaguchi et al questioned whether the diagnosis can be made with positive OR and PR staining in combination with the scattered staining pattern.9 Kawaguchi et al stated that thoracic endometriosis can only be certainly diagnosed when the biopsy is positive for OR and PR and an aggregated pattern is found, as also a high percentage of men with pneumothorax showed positive OR and PR staining and a scattered pattern.9 On the other hand, Mataca et al showed no positive expression of OR, PR and CD10 in women with non-catamenial or endometriosis-related pneumothorax.8 These are conflicting results and more research on these findings should be carried out. The most helpful method to diagnose thoracic endometriosis is still unclear, but our review showed additional value of OR and PR staining above H&E staining. Staining for OR and PR is particularly helpful when biopsies lack specific features, especially endometrial glands. In our described case, we showed the benefit of immunohistochemical staining as well. Without diagnosis, patients could suffer from recurrent pneumothorax for years with chest pain and dyspnoea consequently. It is important to diagnose thoracic endometriosis, so effective treatment can be introduced. In our patient, the placement of thoracic drains failed as conservative treatment as recurrences occurred repeatedly after clamping the drain. Ghigna et al explained the occurrence of these recurrences by several hypotheses: spontaneous rupture of blebs, shedding of endometrial spots in pleural tissue and transdiaphragmatic passage of air from the genital tract during menstruation.1 In our case, during VATS typical endometriosis, lesions on the visceral and parietal pleura and multiple diaphragmatic perforations were seen. After resection of the apex of the right upper lobe, pleurodesis with talc poudrage and placement of a mesh graft on the diaphragm, no recurrence occurred within 5 years follow-up. Retrospectively, our patient declared to have had several episodes of dyspnoea and chest pain associated to the menstrual cycle before the primary surgery. This suggests that our patient already had catamenial pneumothorax before the surgery. It also emphasises the importance of questioning specific symptoms related to thoracic endometriosis in the workup of women with pelvic endometriosis as these are highly related to each other. Even in our specialised clinic for endometriosis, where only women with endometriosis are seen, we missed this diagnosis in the clinical workup. Until now, immunohistochemistry with OR and PR antibodies is only routinely used for endometrial dating throughout the normal menstrual cycle and detection of impaired endometrial receptivity in women with endometriosis and subfertility.12 13 Furthermore, OR and PR staining are also used in diagnosing breast cancer, and staining with CD10 is used for different kind of malignancies, for example, to differentiate between hepatic malignancies.14 15 OR and PR stainings are not usually incorporated in the diagnostic process of pleural lesions of patients with spontaneous pneumothorax and only described in small case studies. In addition, no literature was found on the use of OR, PR and CD10 in other pulmonary diseases. As far as we know, this is the first review on the additional value of OR and PR staining, still it seems that a lot of cases of thoracic endometriosis are missed. The best approach in diagnosing thoracic endometriosis is still undetermined. Most studies confirmed the beneficial value of staining OR and PR, but more research should be performed on these diagnostic tools. To conclude, our case and earlier studies underline that recurrent pneumothorax, both catamenial and non-catamenial, in women in the reproductive age need to raise suspicion of thoracic endometriosis. Our conclusions emphasise the importance of additional staining with OR and PR antibodies as it has additive value in the histologic diagnosis of thoracic endometriosis when histopathology fails to show the typical presentation. Histologic confirmation is mandatory to counsel patients for postoperative treatment and follow-up. Our recommendation would be to additionally stain for OR and PR in all women with a pneumothorax in the reproductive age. Learning points. Catamenial pneumothorax can be easily missed and remains an underdiagnosed disease. Recurrent pneumothorax, both catamenial and non-catamenial, in women in the reproductive age needs to raise suspicion of thoracic endometriosis. Oestrogen and progesterone receptor stainings have additive value in the histologic diagnosis of thoracic endometriosis. Histologic confirmation of thoracic endometriosis is mandatory to counsel patients for postoperative treatment and follow-up. Footnotes Contributors: SK: drafting the work, contributed to the analysis and interpretation of reviewed articles, approved the final version of the article. WO: surgeon of the patient described in the case, contributed to the conception of the article, critically revised the article and approved the final version of the article. HMH: contributed to the conception of the article, critically revised the article and approved the final version of the article. TM: contributed to the conception of the article, contributed to the analysis and interpretation of reviewed article and approved the final version of the article. Funding: The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors. Competing interests: None declared. Provenance and peer review: Not commissioned; externally peer reviewed. Ethics statements Patient consent for publication Obtained.

- 1.Ghigna M-R, Mercier O, Mussot S, et al. Thoracic endometriosis: clinicopathologic updates and issues about 18 cases from a tertiary referring center. Ann Diagn Pathol 2015;19:320–5. 10.1016/j.anndiagpath.2015.07.001 [DOI] [PubMed] [Google Scholar] - 2.Haga T, Kumasaka T, Kurihara M, et al. Immunohistochemical analysis of thoracic endometriosis. Pathol Int 2013;63:429–34. 10.1111/pin.12089 [DOI] [PubMed] [Google Scholar] - 3.Nezhat C, Lindheim SR, Backhus L, et al. Thoracic endometriosis syndrome: a review of diagnosis and management. JSLS 2019;23:e2019.00029. 10.4293/JSLS.2019.00029 [DOI] [PMC free article] [PubMed] [Google Scholar] - 4.Chen J-S, Tsai K-T, Hsu H-H, et al. Intrapleural minocycline following simple aspiration for initial treatment of primary spontaneous pneumothorax. Respir Med 2008;102:1004–10. 10.1016/j.rmed.2008.02.006 [DOI] [PubMed] [Google Scholar] - 5.Joseph J, Sahn SA. Thoracic endometriosis syndrome: new observations from an analysis of 110 cases. Am J Med 1996;100:164–70. 10.1016/S0002-9343(97)89454-5 [DOI] [PubMed] [Google Scholar] - 6.Clement PB. The pathology of endometriosis: a survey of the many faces of a common disease emphasizing diagnostic pitfalls and unusual and newly appreciated aspects. Adv Anat Pathol 2007;14:241–60. 10.1097/PAP.0b013e3180ca7d7b [DOI] [PubMed] [Google Scholar] - 7.Legras A, Mansuet-Lupo A, Rousset-Jablonski C, et al. Pneumothorax in women of child-bearing age: an update classification based on clinical and pathologic findings. Chest 2014;145:354–60. 10.1378/chest.13-1284 [DOI] [PubMed] [Google Scholar] - 8.Mataca E, Rossi G, Colby TV. The helpful role of CD10 and hormonal receptors co-expression in the histologic diagnosis of catamenial pneumothorax. Int J Surg Pathol 2019;27:593–7. 10.1177/1066896919846386 [DOI] [PubMed] [Google Scholar] - 9.Kawaguchi Y, Hanaoka J, Ohshio Y, et al. Diagnosis of thoracic endometriosis with immunohistochemistry. J Thorac Dis 2018;10:3468–72. 10.21037/jtd.2018.05.121 [DOI] [PMC free article] [PubMed] [Google Scholar] - 10.Flieder DB, Moran CA, Travis WD, et al. Pleuro-pulmonary endometriosis and pulmonary ectopic deciduosis: a clinicopathologic and immunohistochemical study of 10 cases with emphasis on diagnostic pitfalls. Hum Pathol 1998;29:1495–503. 10.1016/S0046-8177(98)90021-1 [DOI] [PubMed] [Google Scholar] - 11.Dixit SG, Ghatak S, Singh P, et al. Estrogen receptor, progesterone receptor and CD8+ expression in endometrium of women of unexplained infertility. J Gynecol Obstet Hum Reprod 2018;47:533–7. 10.1016/j.jogoh.2018.05.006 [DOI] [PubMed] [Google Scholar] - 12.Moberg C, Bourlev V, Ilyasova N, et al. Levels of oestrogen receptor, progesterone receptor and αB-crystallin in eutopic endometrium in relation to pregnancy in women with endometriosis. Hum Fertil 2015;18:30–7. 10.3109/14647273.2014.922705 [DOI] [PubMed] [Google Scholar] - 13.Singha J, Khan K, Chatterjee S. Diagnostic utility of CD10 immunohistochemical staining on cellblock in differentiating hepatocellular carcinoma from secondary malignancies of liver. Indian J Pathol Microbiol 2018;61:510–5. 10.4103/IJPM.IJPM_788_16 [DOI] [PubMed] [Google Scholar] - 14.Hicks D, Dell’Orto P, Falzon M, et al. Immunohistochemical performance of estrogen and progesterone receptor antibodies on the Dako Omnis staining platform: evaluation in multicenter studies. Appl Immunohistochem Mol Morphol 2017;25:313–9. 10.1097/PAI.0000000000000311 [DOI] [PMC free article] [PubMed] [Google Scholar] - 15.Garcia E, Bouchard P, De Brux J. Use of immunocytochemistry of progesterone and estrogen receptors for endometrial dating. Hum Fertil 2015;18:30–7. 10.1210/jcem-67-1-80 [DOI] [PubMed] [Google Scholar] - 16.Maniglio P, Ricciardi E, Meli F, et al. Catamenial pneumothorax caused by thoracic endometriosis. Radiol Case Rep 2018;13:81–5. 10.1016/j.radcr.2017.09.003 [DOI] [PMC free article] [PubMed] [Google Scholar] - 17.Furuya K, Otsuka H, Koezuka S, et al. Resection of pulmonary endometriosis using video-assisted thoracoscopic surgery under preoperative CT-guided marking. Gen Thorac Cardiovasc Surg 2017;65:175–8. 10.1007/s11748-015-0607-y [DOI] [PubMed] [Google Scholar] - 18.Fang H-Y, Jan C-I, Chen C-K, et al. Catamenial pneumothorax due to bilateral pulmonary endometriosis. Respir Care 2012;57:1182–5. 10.4187/respcare.01256 [DOI] [PubMed] [Google Scholar] - 19.Kumakiri J, Kumakiri Y, Miyamoto H, et al. Gynecologic evaluation of catamenial pneumothorax associated with endometriosis. J Minim Invasive Gynecol 2010;17:593–9. 10.1016/j.jmig.2010.04.011 [DOI] [PubMed] [Google Scholar] - 20.Morcos M, Alifano M, Gompel A, et al. Life-Threatening endometriosis-related hemopneumothorax. Ann Thorac Surg 2006;82:726–9. 10.1016/j.athoracsur.2005.10.002 [DOI] [PubMed] [Google Scholar]