Iron overload in the peritoneal cavity of women with pelvic endometriosis

FERTILITY AND STERILITY VOL. 78, NO. 4, OCTOBER 2002 Copyright 2002 American Society for Reproductive Medicine Published by Elsevier Science Inc. Printed on acid-free paper in U.S.A. Iron overload in the peritoneal cavity of women with pelvic endometriosis Anne Van Langendonckt, Ph.D., Françoise Casanas-Roux, Ph.D., and Jacques Donnez, M.D., Ph.D. Department of Gynecology, Université Catholique de Louvain, Brussels, Belgium Objective: To examine the possible involvement of iron in the physiopathology of endometriosis. Design: Prospective study. Setting: Department of gynecology in a university hospital. Patient(s): Seventy patients undergoing laparoscopy. Intervention(s): Collection of peritoneal fluid (n 57), blood samples, and biopsy samples from endometrium (n 62) and from endometriotic (n 33) and normal-appearing peritoneum (n 53). Main Outcome Measure(s): Measurement of iron and ferritin in serum and peritoneal fluid and staining of iron deposits with Prussian blue in tissues. Result(s): Iron and ferritin concentrations were significantly higher in the peritoneal fluid of patients with endometriosis compared with controls during the secretory phase. Higher rates of ferritin and hemosiderin deposits were observed in the peritoneum adjacent to red (100%), black (57%), and white (62%) lesions compared with normal-appearing peritoneum (25%). Deposits were more frequent during the secretory phase than the proliferative phase in healthy peritoneum from controls, whereas they were found throughout the cycle in the vicinity of lesions in patients with endometriosis. Similar rates of iron deposition were observed in the stroma of black and white lesions and in eutopic endometrium from patients with endometriosis. Conclusion(s): Iron overload was observed in the cellular and peritoneal fluid compartments of the peritoneal cavity of women with endometriosis. Iron deposits seem to be related to the presence of lesions, suggesting that iron may be involved in the pathogenesis of endometriosis. (Fertil Steril 2002;78:712 8. 2002 by American Society for Reproductive Medicine.) Key Words: Human, peritoneal endometriosis, peritoneal fluid, iron, ferritin, hemosiderin, mesothelium Received August 1, 2001; revised and accepted April 16, 2002. Supported by a grant from the Fonds National de la Recherche Scientifique de Belgique. Anne Van Langendonckt is a scientific collaborator with the Fonds National de la Recherche Scientifique de Belgique. Reprint requests: Jacques Donnez, M.D., Ph.D., Department of Gynecology, Université Catholique de Louvain, Cliniques Universitaires St. Luc, Avenue Hippocrate 10, 1200 Brussels, Belgium (FAX: 32-2-764.95.07; E-mail: donnez@gyne. ucl.ac.be). 0015-0282/02/$22.00 PII S0015-0282(02)03346-0 During menses, menstrual effluent containing endometrial cells and blood may be transported into the abdominal cavity through the Fallopian tubes. This phenomenon, known as retrograde menstruation, has been shown to occur in most menstruating women, as evidenced by the presence of blood and endometrial tissue in peritoneal fluid (1, 2). Retrograde menstruation was first described in 1924 by Sampson (3), who suggested that it may be the origin of pelvic endometriosis. In this disease, endometrial fragments transplanted into the peritoneal cavity can implant and grow, leading to the development of lesions containing endometrial-like glands and stroma. Several theories have been proposed to explain why ectopic endometrial tissue implants in approximately 10% to 15% of patients, whereas in most women, endometrial cells are resorbed by the peritoneal environment (4). It has been suggested that the quantity of retrograde menstruation may be greater in patients with endometriosis and that peritoneal protective mechanisms may be overwhelmed in case of increased menstrual reflux (2, 5). Higher levels of iron, which is probably released after lysis of erythrocytes, have been found in the peritoneal fluid of patients with endometriosis (6), and the concentration of iron is related to the severity of disease (7). Free iron, which is a strong prooxidant (8, 9), may act as a proinflammatory factor in the peritoneal cavity and promote the development of pelvic endometriosis. However, few studies have examined how the peritoneal environment copes with the presence of erythrocytes and iron. 712

Erythrocytes are probably absorbed and degraded by macrophages, as observed in other tissues (8). Iron metabolism by macrophages seems to be enhanced in case of endometriosis. This is supported by the fact that siderophages, iron-storing macrophages that are heavily laden with hemosiderin, are considered to be indicators of endometriosis (10, 11). In pelvic endometriosis, iron transport also appears to be increased, as expression of transferrin receptors by peritoneal macrophages is higher (12) and transferrin concentrations in peritoneal fluid are increased (13). Iron storage in endometriotic lesions is indicated by the observation of iron pigment, which is more frequently encountered in endometriotic than in nonendometriotic lesions (14). Although pelvic iron deposition is considered to be a typical feature of endometriosis, it is not clear whether the presence of deposits is restricted to endometrial cells or also affects the peritoneum. We further investigated the potential role of iron in the physiopathology of endometriosis. Iron storage in the peritoneal cavity was examined by measuring ferric iron and ferritin, a major iron-storing protein in peritoneal fluid, and the presence of iron-ferritin and iron-hemosiderin deposits in the peritoneum and ectopic endometrium of patients with peritoneal endometriosis and controls. We also examined whether the phase of the ovulatory cycle and the type of endometriotic lesion may influence these variables. MATERIALS AND METHODS Patients and Sample Collection Seventy patients with regular ovulatory cycles undergoing laparoscopy for tubal sterilization, infertility, or pelvic pain were included. No patient had used any form of hormonal treatment for at least 3 months before surgery. Patients were divided into two groups. Controls (n 27; mean [ SD] age, 33.6 6.7 years) had no endometriotic lesions detected at laparoscopy, and women with endometriosis (n 43; mean age, 31.0 5.2 years), peritoneal endometriotic lesions were visible at laparoscopy. Morphologic characteristics of endometriosis were confirmed by the presence of both glandular epithelium and stroma in all peritoneal lesions (15, 16). Blood samples were obtained the day before surgery. Peritoneal fluid was collected from 27 controls and 30 women with endometriosis during laparoscopy, after insertion of the suction probe through the first contraincision. Care was taken to avoid contamination of the sample with blood from the abdominal wall. Samples of endometrium (19 controls and 43 women with endometriosis) and normallooking peritoneum (21 controls and 32 women with endometriosis) were obtained for biopsy. Among women with endometriosis, samples of peritoneal endometriotic lesions (n 33) were also obtained from 16 patients for biopsy and classified according to criteria described elsewhere (16, 17) as black (n 14), red (n 6), or white (n 13) peritoneal endometriotic lesions. Collection of biopsy and peritoneal fluid samples was approved by the ethics review board of the Catholic University of Louvain. Peritoneal fluid was centrifuged at 600 g for 10 minutes to remove cells. The peritoneal fluid supernatant was filtered through a 0.8- m prefilter and a 0.2- m filter (Acrodisc; Gelman Science, Ann Arbor, MI) and stored at 20 C until analysis. Biopsy specimens were rapidly processed and fixed in formaldehyde for histologic examination, dating, and iron deposit staining. Iron deposits in tissues and iron and ferritin concentrations in peritoneal fluid were expressed according to the phase of the menstrual cycle proliferative, secretory, or menstrual by dating the corresponding endometrial biopsy using the criteria of Noyes et al. (18). Peritoneal fluid findings were also expressed according to the macroscopic aspect of the peritoneum at the time of laparoscopy (15). Samples were classified in the mainly red lesion group when mainly red lesions were diagnosed (n 9), the white/ black lesion group when black or black and white lesions were present but no red lesions were detected (n 10), and the only white lesion group when only white lesions and no red or black lesions were visible (n 7). Patients who had several types of lesions were excluded from this classification. Measurement of Iron and Ferritin in Peritoneal Fluid and Blood Iron was dissociated from proteins at an acidic ph and measured by using a colorimetric method. Ferric iron was reduced, and the Fe 2 formed a colored complex with ferrozine. The absorbency was read at 560 nm and was proportional to the total iron concentration (19). Ferritin was measured by using radioimmunoassay, as described by Miles et al. (20). Staining of Iron Deposits Ferric iron deposits were detected by using Prussian blue staining according to Perls reaction (21). In brief, dewaxed sections 6 m thick were stained for 20 minutes in a solution of 1% potassium hexacyanoferrate and 2% HCl (v/v), and nuclei were counterstained with nuclear fast red (22). The presence of iron deposits was assessed in sections of endometrium, normal-looking peritoneum, and peritoneal endometriotic lesions. The presence or absence of deposits was recorded in about 7,500 cells in the various tissue samples. Results were expressed as the percentage of samples showing iron deposits. Different types of deposits were distinguished as described below. FERTILITY & STERILITY 713

TABLE 1 Concentrations of iron and ferritin in the peritoneal fluid of controls and patients with peritoneal endometriosis, by type of lesion. Women with endometriosis Substance Controls Mainly red lesions Black and white lesions White lesions only Iron concentration (mg/l) (n 26) (n 9) (n 10) (n 7) 33.1 (22.8, 78.4) 68.0 (47.8, 89.0) a 71.0 (59.8, 111.0) a 78.0 (63.3, 91.5) a Ferritin concentration (ng/ml) (n 27) (n 10) (n 9) (n 7) 56.0 (46.3, 64.5) 122.0 (44.0, 162.5) a 169.5 (60.0, 569.0) a 87.5 (66.5, 102.5) a Note: Values are medians and 25 th and 75 th percentiles. a Significantly higher than control values (P.05 Kruskal Wallis test) Statistical Analysis Iron and ferritin concentrations in the peritoneal fluid are expressed as medians and interquartile (25% and 75%) ranges. Because iron and ferritin concentrations were not normally distributed, data were analyzed by using nonparametric analysis of variance by ranks (Kruskal Wallis) and the Mann Whitney rank-sum test. Statistical analysis of data on iron and ferritin in the peritoneal fluid was initially performed by dividing the patient samples into four groups according to phase of the cycle (secretory or proliferative) and the presence or absence of endometriosis. Statistical analysis was also performed on data from women with endometriosis according to the three types of lesions detected at laparoscopy. Rates of iron deposition and ratios of iron and ferritin in peritoneal fluid to serum ratios were analyzed by using the 2 test. P.05 was considered significant. RESULTS Iron and Ferritin in Peritoneal Fluid and Blood Total iron and its storage molecule ferritin were measured in the peritoneal fluid of 30 women with endometriosis and 27 controls. The volume of peritoneal fluid was similar in both groups. Figure 1 shows finding by cycle phase. Analysis of differences between groups by using the Kruskal Wallis test demonstrated that iron and ferritin concentrations differed among the four groups (P.007 for iron and P.04 for ferritin). The following groups were analyzed: control during the proliferative phase; control during the secretory phase; endometriosis during the proliferative phase; and endometriosis during the secretory phase. The Mann Whitney test revealed that iron and ferritin concentrations were significantly higher in women with endometriosis than controls during the secretory phase (P.002 for iron and P.008 for ferritin). Iron and ferritin concentrations did not differ significantly for the proliferative and secretory phases in either group. Concentrations of iron (median, 92.0 mg/ml) and ferritin (median, 118.0 ng/ml) were also determined in the peritoneal fluid of six patients with endometriosis during the menstrual phase of the cycle. However, these patients were excluded from statistical analysis because of the lack of corresponding controls. Analysis of the results according to the type of endometriotic lesion revealed that concentrations of iron and ferritin in the peritoneal fluid were significantly higher in women with endometriosis, regardless of the type of lesion detected at laparoscopy, than in controls (P.05, Kruskal Wallis test) (Table 1). Concentrations of iron and ferritin in peritoneal fluid did not differ significantly among patients with red, black, or white lesions. Iron and ferritin concentrations were measured in blood serum and compared with peritoneal fluid values. Iron concentrations were lower in peritoneal fluid than in serum, and the ratio of peritoneal fluid to serum values was similar by phase of cycle and between women with endometriosis and controls (mean ratio, 0.7 0.5 vs. 0.6 0.3). Ferritin concentrations were higher in peritoneal fluid than in serum (ratio, 5.2 3.9 in women with endometriosis and 4.5 2.2 in controls), suggesting that ferritin in the peritoneal fluid stems from local production or release rather than serum transudation. Iron Deposits in Endometrium and Peritoneum General Characteristics Iron deposits, as identified by Prussian blue staining, were found in various forms, such as small extracellular or intracellular granules, in isolation (Fig. 2A), shaped as clusters (Fig. 2A), or pale blue cytoplasmic staining (Fig. 2B) sometimes associated with iron-laden macrophages (Fig. 2C and D). As described elsewhere (9, 23), iron conglomerates 714 Van Langendonckt et al. Iron overload and peritoneal endometriosis Vol. 78, No. 4, October 2002

FIGURE 1 Concentrations of iron and ferritin in the peritoneal fluid of controls and patients with peritoneal endometriosis, according to phase of the menstrual cycle. Horizontal lines represent medians. During the secretory phase, iron and ferritin concentrations in the peritoneal fluid of patients with endometriosis were significantly higher than those in controls (P.01). Iron Deposits in the Peritoneum Significantly more samples with iron deposits came from peritoneum adjacent to lesions than from healthy peritoneum (Table 2), regardless of the type of lesion. Iron deposits more often occurred close to red lesions (100%) than to black (57%; P.05) or white lesions (62%; P.05). Isolated granules or faint blue staining corresponding to soluble ferritin were significantly more frequent near peritoneal lesions than in normal-looking peritoneum. Insoluble and intensely positive-stained conglomerates occurred in about 20% of the peritoneal biopsies, regardless of localization, except in the peritoneum close to red lesions, where their occurrence reached higher (50%) but not significant rates. During the proliferative phase of the cycle, iron deposits were rare in normal-looking peritoneum from controls and women with endometriosis (Table 3), whereas deposits were found during the secretory phase in 50.0% (P.02) and 33.3% (.05 P.10) of participants, respectively. In contrast, in peritoneum close to lesions, a similar proportion of samples with iron deposits was found during the proliferative and the secretory phase of the cycle, regardless of the type of peritoneal lesion. Iron Deposits in the Endometrium Similar rates of iron deposits were found in the stroma of eutopic endometrium of patients with endometriosis (18 of 43), red lesions (4 of 6), black lesions (5 of 14), and white lesions (4 of 13). No difference was observed according to phase of the ovulatory cycle (data not shown). Total iron deposits were significantly lower in the stroma of black and white peritoneal lesions and the endometrium of patients with peritoneal endometriosis than in the stroma of eutopic endometrium of patients without peritoneal endometriosis (13 of 19) (P.05). DISCUSSION correspond to iron stored in hemosiderin, which is located in siderosomes, whereas blue granules and faint blue background cytosolic staining are associated with ferritin. Iron deposits were detected in the stroma of eutopic and ectopic endometrium (Fig. 2), whereas glandular epithelium showed no Prussian blue staining. Positive iron staining was also found in peritoneal cells (Fig. 2D), under the mesothelium lining the peritoneum or deeper into the peritoneal tissue, close to peritoneal endometriotic lesions, and in normal-looking areas. The presence of iron deposits in the peritoneum was analyzed according to the type of peritoneal endometriotic lesion (Table 2) and phase of the ovulatory cycle (Table 3). Our findings confirm that iron concentrations are increased in the peritoneal fluid of patients with endometriosis (6). Furthermore, our results indicate that concentrations of iron in the peritoneal fluid were similar in patients with black and white peritoneal lesions and those with mainly red lesions. Red lesions are likely to be the first stage of peritoneal endometriotic lesions, with higher vascular and metabolic activity, whereas black lesions are probably advanced endometriosis and white lesions are thought to be quiescent or latent lesions (24). This suggests that iron concentrations in the peritoneal fluid remain high during the later stages of the disease. Iron storage was investigated by analyzing soluble ferritin in the peritoneal fluid and detecting soluble and insoluble iron deposits in peritoneal and endometrial tissues. Ferritin was detected in the peritoneal fluid of controls and women FERTILITY & STERILITY 715

FIGURE 2 Iron deposits in endometrium and peritoneum (Prussian blue staining). Various forms of iron deposits are observed. (A), Small extracellular and intracellular granules, isolated deposits (arrows), and deposits shaped as clusters (arrowhead). (B), Faint blue cytoplasmic staining associated (C) or not associated (D) with iron-laden macrophages. Bar 11 m. with endometriosis, and the concentrations of soluble ferritin were higher in peritoneal fluid than in serum, suggesting local production of ferritin in the peritoneal cavity. However, markedly increased ferritin levels were found in the peritoneal fluid of patients with endometriosis, and the pattern according to type of lesion was similar to that observed for iron concentrations in the peritoneal fluid. In both groups, soluble iron deposits that probably corresponded to ferritin-iron were found in peritoneal cells, as indicated by the presence of faint Prussian blue staining in peritoneal cells and numerous positive granules. Iron deposits were observed in 24% of control peritoneal biopsy samples and 25% of normal-looking biopsy samples from patients with endometriosis. However, in patients with endometriosis, iron deposits in peritoneum were more frequently encountered close to a red (100%), black (57%), or white (62%) lesion than in normal-looking peritoneum, suggesting that their occurrence depends on the presence and activity of the peritoneal lesions. Ferritin synthesis is probably induced after lesion bleeding and might also be enhanced by the secretion of cytokines, such as tumor necrosis factor-, interleukin-1, and interleukin-6 (8), which are induced in case of endometriosis (5). In conditions involving excessive storage of iron, ferritin accumulation may lead to formation of an insoluble storage form, hemosiderin (8, 25). In 50% of cases, the peritoneum in the vicinity of red lesions presents the typical feature of iron-overloaded tissues, showing densely stained conglomerates corresponding to insoluble hemosiderin (9, 23). Furthermore, macrophages heavily laden with hemosiderin were often seen in the vicinity of the lesions. These observations suggest that in healthy controls as well as patients with endometriosis, repeated hemorrhage into the peritoneal cavity due to retrograde menstruation and ovulation leads to iron storage by peritoneal cells, as observed in most tissues after hemorrhage. However, in the case of endometriosis, the iron-scavenging system may be overwhelmed by additional bleeding from lesions. Of note, iron-ferritin deposits were still present in the peritoneum 716 Van Langendonckt et al. Iron overload and peritoneal endometriosis Vol. 78, No. 4, October 2002

TABLE 2 Rates of iron deposits in the peritoneum of patients with and without endometriosis. Endometriosis group Staining Control group Normal-looking peritoneum Peritoneum near red lesion Peritoneum near black lesion Peritoneum near white lesion Total iron deposits 5/21 (23.8) 8/32 (25.0) 6/6 (100) a,b 8/14 (57.1) b 8/13 (61.5) b Faint blue staining 4/21 (19.0) 8/32 (25.0) 5/6 (83.3) c 8/14 (57.1) c 8/13 (61.5) c and granules Conglomerates 4/21 (19.1) 5/32 (15.6) 3/6 (50.0) 3/14 (21.4) 2/13 (15.4) Note: Values are number (percentage) of cases. a Significantly more frequent in the peritoneum close to red peritoneal lesions than in the peritoneum close to black or white peritoneal lesions (P.05). b Total iron deposits are significangly more frequent in peritoneum close to red, black and white peritoneal lesions than in normal-looking peritoneum of patients with peritoneal endometriosis (P.001, P.02 and P.02, respectively) and without peritoneal endometriosis (P.001, P.05 and P.0, respectively). c Faint blue and granule staining are significantly more frequent in peritoneum close to red, black and white peritoneal lesions than in normal-looking peritoneum of patients with peritoneal endometriosis (P.01, P.02 and P.02, respectively) and without peritoneal endometriosis (P.001, P.05 and P.05, respectively). around lesions in patients with only white cicatricial lesions, indicating that deposits were not resorbed during the later stages of disease. At the level of the endometriotic lesion, a similar number of samples from ectopic endometrium showed iron deposits compared with eutopic endometrium from patients with endometriosis. As reported elsewhere, iron deposits were restricted to the stroma of the endometrium (26). Iron deposits tended to be less frequent in black or white lesions, which had less vascularization than did red lesions and eutopic endometrium from controls (27). Results were also analyzed according to the phase of the ovulatory cycle. Iron and ferritin levels in peritoneal fluid did not vary according to the phase of the ovulatory cycle in patients with or without endometriosis. The presence of iron TABLE 3 Rates of iron deposits in the peritoneum of patients with and without endometriosis, by phase of the ovulatory cycle. Phase Controls Normal-looking peritoneum Women with endometriosis Peritoneum near lesions Proliferative 0/10 a 1/11 (9.1) 8/12 (66.7) Secretory 4/8 (50.0) 6/18 (33.3) 14/20 (70.0) Menstrual 1/3 (33.3) 1/3 (33.3) 0/1 Note: Values are number (percentage) of cases. a In the normal-looking peritoneum of controls, iron deposits are significantly more frequent during the secretory phase than during the proliferative phase (P.02). in peritoneal fluid throughout the cycle is consistent with the report by Halme et al. (1), who showed that erythrocytes may be found in the peritoneal cavity even during the nonmenstruating phase. Surprisingly, the increase in peritoneal fluid levels of iron and ferritin in patients with endometriosis patients was observed during the secretory phase of the cycle. This may be because processes other than menstrual reflux, such as lesion bleeding (which is not under hormonal control) (28) or ovulation occurring at the end of the proliferative phase, may contribute to the accumulation of iron in the peritoneal fluid. Ferritin-iron deposits in the vicinity of lesions were observed throughout the menstrual cycle, whereas deposits were more frequent during the secretory phase than the proliferative phase in healthy peritoneum. In conclusion, our data suggest that iron homeostasis in the peritoneal cavity may be disrupted in endometriosis. The fluid, macrophages, peritoneal tissue, and endometriotic lesion of the peritoneal cavity appear to be loaded with iron in this disease. Levels of the iron-scavenging molecules ferritin and transferrin, which protect cells against the deleterious effect of free iron, appear to be increased in the peritoneal cavity in endometriosis. However, cellular defenses might be overwhelmed owing to repeated bleeding of lesions, as suggested by the accumulation of ferritin and insoluble hemosiderin deposits in macrophages and in the peritoneum surrounding lesions. Further studies are needed to assess the consequences of iron overload in the peritoneal environment. Acknowledgments: The authors thank Drs. Marie-Madeleine Dolmans and Jean-Paul Van Gossum for assistance with collection of tissues. FERTILITY & STERILITY 717

References 1. Halme J, Hammond MG, Hulka JF, Raj SG, Talbert LM. Retrograde menstruation in healthy women and in patients with endometriosis. Obstet Gynecol 1984;64:151 4. 2. Kruitwagen RF. Menstruation as the pelvic aggressor. Bailliere Clin Obstet Gynaecol 1993;7:687 700. 3. Sampson JA. Peritoneal endometriosis due to the menstrual dissemination of endometrial tissue into the peritoneal cavity. Am J Obstet Gynecol 1927;14:422 69. 4. Healy DL, Rogers PA, Hii L, Wingfield M. Angiogenesis: a new theory for endometriosis. Hum Reprod Update 1998;4:736 40. 5. Vinatier D, Dufour P, Oosterlynck D. Immunological aspects of endometriosis. Hum Reprod Update 1996;2:371 84. 6. Arumugam K. Endometriosis and infertility: raised iron concentration in the peritoneal fluid and its effect on the acrosome reaction. Hum Reprod 1994; 9:1153 7. 7. Arumugam K, Yip YC. De novo formation of adhesions in endometriosis. The role of iron and free radical reactions. Fertil Steril 1995; 64:62 4. 8. Ponka P, Beaumont C, Richardson DR. Function and regulation of transferrin and ferritin. Semin Hematol 1998;35:35 54. 9. Richter GW. The iron-loaded cell. The cytopathology of iron storage. A review. Am J Pathol 1978; 91: 362 404. 10. Gaulier A, Jouret-Mourin A, Marsan C. Peritoneal endometriosis. Report of a case with cytologic, cytochemical and histopathologic study. Acta Cytol 1983;27:446 9. 11. Stowell SB, Wiley CM, Perez-Reyes N, Powers CN. Cytologic diagnosis of peritoneal fluids. Applicability to the laparoscopic diagnosis of endometriosis. Acta Cytol 1997;41:817 22. 12. Martinez-Roman S, Balasch J, Creus M, Fabregues F, Carmona F, Vilella R, et al. Transferrin receptor (CD71) expression in peritoneal macrophages from fertile and infertile women with and without endometriosis. Am J Reprod Immunol 1997;38:413 7. 13. Mathur SP, Lee JH, Jiang H, Arnaud P. Rust PF. Levels of transferrin and alpha 2-HS glycoprotein in women with and without endometriosis. Autoimmunity 1999;29:121 7. 14. Moen MH, Halvorsen TB. Histologic confirmation of endometriosis in different peritoneal lesions. Acta Obstet Gynecol Scand 1992;71:337 42. 15. Nisolle M, Paindaveine B, Bourdon A, Berlière M, Casanas-Roux F, Donnez J. Histologic study of peritoneal endometriosis in infertile women. Fertil Steril 1990;53:984 8. 16. Donnez J, Nisolle M, Casanas-Roux F. Three-dimensional architectures of peritoneal endometriosis. Fertil Steril 1992;57:980 3. 17. Nisolle M, Paindaveine B, Bourdon A, Casanas-Roux F, Donnez J. Endométriose péritonéale: aspects typiques et atypiques. Acta Endosc 1992;22:15 23. 18. Noyes RW, Hertig AT, Rock J. Dating the endometrial biopsy. Fertil Steril 1950;1:3 25. 19. Carter P. Spectrometric determination of serum iron at the submicrogram level with a new reagent (ferrozine). Anal Biochem 1971;40: 450 8. 20. Miles LE, Lipschitz DA, Bieber CP, Cook JD. Measurement of serum ferritin by a 2-site immunoradiometric assay. Anal Chem 1974;61:209 24. 21. Perls M. Nachweis von Eisenoxyd in gewissen Pigmenten. Virchows Archiv 1867;39:42 8. 22. Gabe M. Détection des cations minéraux. In: Gabe M (ed). Techniques histologiques. Paris: Masson et Cie, 1968:324. 23. Iancu TC. Ferritin and hemosiderin in pathological tissues. Electron Microsc Rev 1992;5:209 29. 24. Nisolle M, Donnez J. Peritoneal endometriosis, ovarian endometriosis, and adenomyotic nodules of the rectovaginal septum are three different entities. Fertil Steril 1997;68:585 96. 25. Fischbach FA, Gregory DW, Harrison PM, Hoy TG, Williams JM. On the structure of hemosiderin and its relation to ferritin. J Ultrastruct Res 1971; 37:495 503. 26. McCulloch TA, Wagner B, Duffy S, Barik S, Smith JH. The pathology of hysterectomy specimens following trans-cervical resection of the endometrium. Histopathology 1995;27:541 7. 27. Nisolle M, Casanas-Roux F, Anaf V, Mine JM, Donnez J. Morphometric study of the stromal vascularization in peritoneal endometriosis. Fertil Steril 1993;59:681 4. 28. Nisolle M, Casanas-Roux F, Wyns C, de Menten Y, Mathieu PE, Donnez J. Immunohistochemical analysis of estrogen and progesterone receptors in endometrium and peritoneal endometriosis: a new quantitative method. Fertil Steril 1994;62:751 9. 718 Van Langendonckt et al. Iron overload and peritoneal endometriosis Vol. 78, No. 4, October 2002