A C TA Obstetricia et Gynecologica
AOGS M A I N R E SE A RC H A R TI C LE
Intrafollicular iron and ferritin in women with ovarian endometriomas LAURA BENAGLIA1, ALESSIO PAFFONI1, ALICE MANGIARINI1, LILIANA RESTELLI1, NORA BETTINARDI2, EDGARDO SOMIGLIANA1, PAOLO VERCELLINI1,3 & LUIGI FEDELE1,3 1
Obstetrics-Gynecology Department, Policlinic Maggiore Hospital (Ospedale Maggiore), Milan, 2Analysis Laboratory, Ca’ Granda Foundation, Policlinic Maggiore Hospital (Ospedale Maggiore), Milan, and 3University of Milan, Milan, Italy
Key words Endometriosis, iron, ferritin, endometrioma, follicular fluid Correspondence Laura Benaglia, Infertility Unit, Fondazione Ca’ Granda, Ospedale Maggiore Policlinico, Via M. Fanti 6, 20122 Milan, Italy. E-mail: [email protected] Conflicts of interest The authors have stated explicitly that there are no conflicts of interest in connection with this article. Please cite this article as: Benaglia L, Paffoni A, Mangiarini A, Restelli L, Bettinardi N, Somigliana E, et al. Intrafollicular iron and ferritin in women with ovarian endometriomas. Acta Obstet Gynecol Scand 2015; 94: 646–653. Received: 25 September 2014 Accepted: 30 March 2015
Abstract Objective. To evaluate whether iron contained in ovarian endometriomas can diffuse through the cyst wall and negatively affect ovarian function. Design. Prospective case series. Setting. Infertility unit in an academic setting. Population. Thirty-nine infertile women with unilateral endometriomas who underwent in vitro fertilization. Methods. Iron and ferritin assessments in pools of follicular fluids obtained from affected and contralateral intact gonads. Main outcome measures. Iron and ferritin concentrations. Results. Follicular fluid iron content did not differ between the two gonads. The median [interquartile range (IQR)] follicular concentrations in the affected and unaffected ovaries were 59 (IQR 44–74) and 59 (IQR 47–73) lg/dL, respectively (p = 0.77). Conversely, ferritin concentration was significantly higher in affected gonads. The median (IQR) concentrations of ferritin in the affected and unaffected ovaries were 57 (IQR 31–146) and 33 (IQR 23–67) lg/mL, respectively (p = 0.026). When considering together the 78 studied ovaries, no significant correlations emerged between follicular iron and ferritin and variables reflecting ovarian responsiveness and oocyte developmental competence. Conclusions. Iron may diffuse from ovarian endometriomas into the adjacent ovarian tissue. However, this phenomenon does not appear to markedly affect ovarian function. Some effective biological mechanisms such as ferritin storage may effectively sequester free iron, so limiting its detrimental effects.
DOI: 10.1111/aogs.12647
ICSI, intracytoplasmatic sperm injection; IQR, interquartile range; IVF, in vitro fertilization; ROS, reactive oxygen species.
Abbreviations:
reaction that may be potentially harmful to the surrounding cells (4,5). ROS are indeed highly reactive and become stable by acquiring electrons from nucleic acids,
Introduction Endometriotic cysts contain a huge amount of free iron. The concentration exceeds 100 nmol/L, which is up to 10 000-fold higher than the concentration normally present in serum (0.013–0.027 mmol/L) (1). Accordingly, massive iron staining has also been observed in the stroma of the cyst, mostly located in macrophages (2,3). The elevated concentration of iron in ovarian endometriomas is a matter of concern because nonprotein-bound “free” or “catalytic” iron can mediate the production of reactive oxygen species (ROS) via the Fenton reaction, a
646
Key Message Iron may diffuse from ovarian endometriomas into the adjacent ovarian tissue. Its detrimental effects are, however, lessened by some biological compensatory mechanisms that prevent major effects on ovarian function.
ª 2015 Nordic Federation of Societies of Obstetrics and Gynecology, Acta Obstetricia et Gynecologica Scandinavica 94 (2015) 646–653
L. Benaglia et al.
lipids, proteins, carbohydrates or any nearby molecule, causing a cascade of chain reactions resulting in cellular damage. There is some recent evidence suggesting that the detrimental effect of endometrioma contents is not merely limited to the cells lining the internal layer of the cysts. The amount of oxidative stress affecting the normal ovarian cortex surrounding an endometrioma was shown to be greater than that in other types of cysts (6). Some studies have also demonstrated that granulosa cells from infertile patients with endometriosis exhibit more signs of oxidative stress when compared with those from unaffected patients (7,8). On this basis, it has been hypothesized that factors present in the endometriomas, and iron in particular, may diffuse in the surrounding tissue causing ROS generation, in a similar way to what happens inside the cyst (9–11). This aspect is of biological relevance, considering that oxidative stress was shown to negatively affect ovarian developmental competence (12). In the present study, we tested the above-mentioned hypothesis that iron contained in endometriomas may diffuse through the cyst wall and negatively affect folliculogenesis. With this aim, we recruited women with unilateral endometriomas undergoing in vitro fertilization (IVF) and compared iron and ferritin concentrations in the follicular fluid of the two gonads. Moreover, we correlated iron and ferritin concentrations to oocyte developmental competence and follicular response to controlled ovarian stimulation with gonadotropins.
Material and methods Women selected for IVF–intracytoplasmatic sperm injection (ICSI) cycle at the Infertility Unit of the Fondazione Ca’ Granda, Ospedale Maggiore Policlinic, Milan, Italy, between January 2012 and January 2013 were prospectively evaluated for study entry. Eligible women were those with unilateral ovarian endometriomas. More specifically, inclusion criteria were (i) age 18–42 years, (ii) indication for IVF-ICSI, (iii) presence of one or more unilateral ovarian endometriomas on transvaginal ultrasound, (iv) absence of nonendometriotic ovarian cysts, (v) acceptance to participate. Women entering the study were subsequently excluded in the following circumstances: (i) cancelled oocyte retrieval for any reason, (ii) accidental follicular fluid contamination with endometrioma content, (iii) absent follicular growth in any of the two gonads. Women were enrolled only for their first completed cycle performed during the study period. The study was accepted by the local Institutional Review Board and all participating women signed an informed consent.
Iron, ferritin, and endometriomas
During the IVF-ICSI cycle, women were monitored and managed according to a standardized clinical protocol as reported in detail elsewhere (13). Briefly, the patients underwent baseline transvaginal ultrasound the month preceding the ovarian hyperstimulation. The presence of ovarian endometriomas and previous documentation on their presence were systematically recorded at this time. Ovarian endometrioma was defined as a roundshaped cystic mass with a minimum diameter of 10 mm, with thick walls, regular margins, homogeneous low echogenic fluid content with scattered internal echoes, and without papillary projections (14). Doubtful cases were excluded. To rule out functional cysts, the presence of the endometriomas had also to be documented at least on one previous ultrasound scan performed at least 2 months before the IVF-ICSI cycle. In general, the first assessment was obtained during the standard diagnostic work-up that all infertile women undergo in our unit and that systematically included transvaginal ultrasound. The second assessment was performed in the IVF cycle preparation, hence in the month preceding the ovarian hyperstimulation. The diameter of the endometriomas was calculated as the mean of three perpendicular diameters. The policy of the unit for endometriomas up to a diameter of 4 cm was to advise conservative management if the woman was asymptomatic or if the pain complaints could be effectively treated with medical treatment (15,16). This limit of 4 cm was, however, not stringent. For larger cysts, the woman was informed that the available evidence is limited and a shared decision was taken after an in-depth discussion of the pros and cons of conservative and surgical approaches. The regimen and the dose of gonadotropins were chosen on an individual basis based on age, day 3 serum follicle stimulating hormone, serum anti-M€ ullerian hormone, sonographic assessment of antral follicular count, and results from previous ovarian hyperstimulation cycles. Gonadotropins used were recombinant follitropin-a (Gonal-F; Merck Serono, Rome, Italy) or human menopausal gonadotropin (Meropur; Ferring, Milan, Italy). The maximal dose was 450 IU/day. During ovarian hyperstimulation women underwent serial transvaginal ultrasound and had serum estrogen and progesterone assessments. Human chorionic gonadotropin was administered subcutaneously when the leading follicles had a mean diameter >18 mm. The total number of growing follicles (i.e. those with a mean diameter ≥ 11 mm calculated as the mean of three perpendicular diameters) per ovary was systematically recorded at this time. Oocyte retrieval was performed transvaginally 36 h after human chorionic gonadotropin administration. All follicles with a mean diameter >10 mm were aspirated. All efforts were made to avoid follicular fluid contamina-
ª 2015 Nordic Federation of Societies of Obstetrics and Gynecology, Acta Obstetricia et Gynecologica Scandinavica 94 (2015) 646–653
647
Iron, ferritin, and endometriomas
tion with the endometrioma content. However, if it was necessary to reach follicles located behind the endometrioma, the cyst could be transfixed. As mentioned above, frank contamination with endometrioma content was an exclusion criterion. After completion of oocyte collection in one ovary, the aspiration system was rinsed before moving to the contralateral ovary and the embryologist was warned to separate follicular fluids belonging to different ovaries. Immediately after recovery of oocytes, a pool of all aspirated follicular fluids was collected separately for each ovary into a large Petri dish. A maximum volume of 12 mL of follicular fluid was then transferred from the pool to a sterile tube and successively centrifuged at 350 g for 10 min at room temperature to separate it from cells. For every patient, two samples of the centrifuged follicular fluids for each of the two ovaries were transferred into cryovials and stored at 20°C until iron and ferritin measurement. Cryovials were identified by the embryologist with a label indicating patients’ identification code and “right” or “left” ovary. Both the embryologist and the biologist performing iron and ferritin measurement were masked from the presence or absence of endometriomas in the specific ovary. IVF or ICSI procedures were performed according to standardized criteria (17) and embryo transfer was performed 2–5 days after the oocyte collection. Cycles could be cancelled because of low or hyper-ovarian response. These women were excluded from the present analysis. Clinical pregnancy was defined as the ultrasonographic demonstration of intrauterine viable embryos 4–5 weeks after embryo transfer. Iron and ferritin were measured on an automatic platform (Cobas 8000; Roche Italia, Monza, Italy). Colorimetric and immunochemical methods were used for iron and ferritin, respectively (Cobas Iron gen. 2; Elecsys Ferritin, Roche, Italy). More specifically, ferric iron is unbound from proteins by an acid reaction (pH < 2.0). The ferric ion is then reduced to ferrous iron by ascorbate. Bivalent iron ions form a colored complex with ferrozine, detected at 700/570 nm; color intensity is directly proportional to iron concentration. For ferritin, the reaction is conduct in four steps: the sample is incubated with a biotinylated and ruthenium-labeled monoclonal antibody anti-ferritin; the sandwich complex that is formed binds itself to a solid phase by biotin–streptavidin; after washing, the reaction mixture is aspirated into the measuring cell where the microparticles are magnetically captured onto the surface of the electrode. Application of a voltage to the electrode then induces chemiluminescent emission, which is measured by a photomultiplier. Both methods are certified for serum/plasma and were previously validated in follicular fluids with
648
L. Benaglia et al.
particular regard towards coefficient of variation, linearity, drift, and hemolysis (data not shown).
Statistical analysis Analysis of the data was carried out with the STATISTICS PACKAGE FOR SOCIAL SCIENCES v18.0 (SPSS Inc., Chicago, IL, USA). p values < 0.05 were considered significant. Data are presented as number (%), mean SD or median [interquartile range (IQR)], as appropriate. Given the paucity of available data on follicular iron and ferritin statistical distribution, it was decided to consider them as non-normally distributed and data were analyzed using nonparametric statistics. The sample size was calculated by setting type I and II errors at 0.05 and 0.20 and considering clinically relevant a threefold increase in the proportion of follicular pools with high iron or ferritin in affected ovaries. Given the absence of previous data indicating a threshold to define high levels for both molecules in the follicular fluid, we arbitrarily set these thresholds at the 90th centile of the distribution observed in the intact gonads. In other words, the frequency of elevated iron or ferritin in intact gonads was 10% (per definition) and we claimed as biologically relevant detecting this condition in 25% of the affected gonads (odds ratio ≥ 3). On this basis, we aimed to have a sample size of at least 35 women. A Wilcoxon test for paired data and McNemar test were used to compare affected and unaffected gonads, as appropriate. Correlations were tested using nonparametric statistics and data were presented using the Spearman’s rank correlation coefficient (Spearman’s Rho).
Results Thirty-nine women were recruited. Baseline characteristics of the selected women and IVF-ICSI outcome are shown in Tables 1 and 2, respectively. A high proportion of women had a concomitant male factor of infertility and required ICSI. Ovarian responsiveness in the affected and contralateral intact gonads did not differ. Specifically, the mean SD numbers of developed follicles, oocytes retrieved and suitable oocytes were 5.4 3.1 and 6.6 5.3 (p = 0.12), 3.9 3.1 and 4.5 2.8 (p = 0.29) and 2.5 2.3 and 3.1 2.4 (p = 0.27), respectively. Oocyte developmental competence did not differ. Indeed, the number of embryos and the total number of topquality embryos obtained from the two gonads were 1.6 1.9 and 1.8 1.8 (p = 0.51) and 1.0 1.5 and 0.9 1.4 (p = 0.81), respectively. The follicular concentration of iron and ferritin according to the ovary of origin is illustrated in Figure 1. Specifically, the median (IQR) concentrations of iron in the affected and unaffected ovaries weres 59 (44–74) and 59
ª 2015 Nordic Federation of Societies of Obstetrics and Gynecology, Acta Obstetricia et Gynecologica Scandinavica 94 (2015) 646–653
L. Benaglia et al.
Iron, ferritin, and endometriomas
Table 1. Baseline characteristics of the studied population (n = 39).
Characteristics Age (years) Body mass index (kg/m2) Previous pregnancies Smoking Duration of infertility (years) Concomitant male factor of infertility Previous IVF-ICSI cycles Day 3 serum FSH (IU/mL) AMH (ng/mL) AFC (both ovaries) Previous surgery for ovarian endometriomasa Side of the endometriomas Right Left Number of endometriomas 1 2 Mean diameter of the endometriomas (mm)b CA-125 (IU/mL) Medical treatment before entry into cycle Oral contraceptives Progestins
Number (%), mean SD or median (IQR) 35.0 3.8 21.6 2.7 8 (20%) 7 (18%) 3.5 2.0 24 (62%) 9 (23%) 7.4 2.3 2.3 1.6 10 (5–15) 12 (31%)
21 (54%) 18 (46%) 33 (85%) 6 (15%) 24 10 40 (25–61) 10 (26%) 7 (18%)
€llerian hormone; CA-125, AFC, antral follicle count; AMH, anti-Mu cancer antigen 125; FSH, follicle-stimulating hormone; IQR, interquartile range; IVF-ICSI: in vitro fertilization–intracytoplasmic sperm injection. a Surgery was performed in affected ovary in five cases, in unaffected ovaries in two cases and bilaterally in the remaining five cases. b If more than one endometrioma was present, data refer to the largest one.
(47–73) lg/dL, respectively (p = 0.77). The median (IQR) concentrations of ferritin were 57 (31–146) and 33 (23–67) lg/mL, respectively (p = 0.026). Iron concentration was above the 90th centile of the distribution in unaffected ovaries (90 lg/dL) in three (8%) intact and four (10%) affected gonads (p = 1.00). Ferritin concentration was above the 90th centile of the distribution in unaffected ovaries (132 lg/mL) in three (8%) intact and 11 (28%) affected gonads (p = 0.021). Follicular concentration of iron correlated between the two ovaries (q = 0.76, p < 0.001). A similar figure emerged for follicular ferritin (q = 0.55, p < 0.001). Finally, a significant correlation was documented also when correlating ferritin and iron in the 78 available gonads (q = 0.42, p < 0.001). Twelve women (31%) were previously operated on for ovarian endometriomas (Table 1). Lesions were bilateral in five women and unilateral in the remaining seven, allowing us to evaluate follicular levels of iron and ferritin
Table 2. Characteristics of in vitro fertilization–intracytoplasmatic sperm injection cycles in the studied population (n = 39).
Characteristics Regimen of ovarian hyper-stimulation Long protocol GnRH antagonists Other Duration of stimulation (days) Total dose of FSH administered (IU) E2 at the time of hCG administration (pg/mL) Total number of follicles ≥11 mm Number of oocytes retrieved Number of suitable oocytesa Technique used IVF ICSI Fertilization rate (%) Number of embryos Cleavage rate (%) Number of top-quality embryos Rate of top-quality embryos (%) Number of embryos transferred None 1 2 Clinical pregnancies Embryos implanted (implantation rate)
Number (%), mean SD or median (IQR)
26 (67%) 6 (20%) 5 (13%) 10.4 2.0 2700 (1500–3285) 2013 (1568–2645) 11.8 6.9 8.5 4.7 6.0 3.7 14 (36%) 25 (64%) 75 (43–100) 3.4 2.9 65 (33–100) 2.8 2.5 50 (20–71) 7 16 16 11 11
(18%) (41%) (41%) (28%) (23%)
E2, estradiol; FSH, follicle-stimulating hormone; GnRH, gonadotropinreleasing hormone; hCG, human chorionic gonadotropin; IQR, interquartile range; ICSI, intracytoplasmic sperm injection; IVF, in vitro fertilization. a Suitable oocytes correspond to oocytes used. They were defined as metaphase II oocytes or type 1 cumulus–oocytes complexes according to the ESHRE Istanbul Consensus Conference (2011).
in 17 previously operated ovaries (10 with recurrence). The median (IQR) iron concentrations in operated and nonoperated (n = 61) gonads were 67 (40–77) and 59 (48–72) lg/dL, respectively (p = 0.78). The median (IQR) ferritin concentrations were 69 (24–185) and 41 (23–66) lg/mL, respectively (p = 0.26). The analyses on follicular concentration of iron and ferritin were repeated excluding these 12 previously operated women (n = 27). The median (IQR) concentrations of iron in the affected and unaffected ovaries were 59 (48–72) and 58 (47–73) lg/dL, respectively (p = 0.68). The median (IQR) concentrations of ferritin were 49 (26–90) and 33 (23–66) lg/mL, respectively (p = 0.05). Iron concentration was above the 90th centile in three (11%) affected and three (11%) intact gonads (p = 1.00). Ferritin concentration was above the 90th centile in six (23%) affected and one (4%) intact gonad (p = 0.12).
ª 2015 Nordic Federation of Societies of Obstetrics and Gynecology, Acta Obstetricia et Gynecologica Scandinavica 94 (2015) 646–653
649
Follicular levels (Iron µg/dL; Ferritin µg/mL)
Iron, ferritin, and endometriomas
L. Benaglia et al.
250
Discussion
200
150
100
*
50
0 Iron
Ferritin
Figure 1. Follicular concentration of iron and ferritin in ovaries with and without endometriomas. Box bars of the affected ovaries are colored grey and those from unaffected gonads are white. Concentration of iron did not differ (p = 0.77) whereas a statistically significant difference was documented for ferritin (p = 0.026). *Statistical significant p value.
Correlations of follicular iron and ferritin with variables reflecting ovarian function and oocyte developmental competence are shown in Table 3. Considering follicular iron, none of the correlations were significant. In the affected gonad, no significant correlation emerged between ferritin and any of the variables considered. In the contralateral intact gonads, significant correlations emerged between follicular ferritin and the number of developing follicles, the number of oocytes retrieved, the number of suitable oocytes, the number of pronucleated oocytes and the number of embryos obtained. All of these correlations were lost when considering data from both ovaries together. All the correlations tested in Table 3 were also evaluated in the subgroup of women who did not previously undergo ovarian surgery (n = 27) and none was significant (data not shown).
We demonstrated that levels of ferritin, but not total iron, are enhanced in follicles developing in proximity to ovarian endometriomas. Moreover, we did not observe biologically significant correlations between these two molecules and responsiveness to ovarian hyperstimulation or oocyte developmental competence. In fact, some correlations were significant (mainly between ferritin and ovarian responsiveness), but they contrasted with the hypothesis tested, they were limited to the intact ovaries and they were lost when considering both gonads together. On these bases, we interpreted these significant correlations as being due to a type I error. Our results are partly in line with those from a very recent study from Sanchez et al. who tested the same hypothesis (11). These authors recruited 13 women with unilateral endometriomas and evaluated follicular content in ferritin and iron in individual follicles according to their location. Specifically, they compared follicles in close contact with the endometriomas (n = 35), those located in the affected ovary but not in close contact with the cyst (n = 28) and those from the contralateral gonad (n = 40). The authors overall observed a decreasing gradient of ferritin concentration in the three groups, in line with our findings. In contrast to our data, however, these authors also observed this gradient for iron concentration. This discrepancy is difficult to explain. We hypothesize a type I error in the study from Sanchez et al., considering in particular that these authors analyzed data from only 13 women (11). The use of several follicles from the same woman may have indeed lead to overappreciation of the association. In other words, if diffusion of iron from an endometrioma to the surrounding tissue is a rare but possible event, the inclusion of even a single case contributing for several follicles may have artificially led to detecting a statistically significant difference.
Table 3. Correlation (Spearman’s Rho) between follicular iron and ferritin and variables of ovarian function. Affected gonad
Unaffected gonad
Both gonads
Variables
n
Iron
Ferritin
n
Iron
Ferritin
n
Iron
Ferritin
Number of follicles ≥11 mm Number of oocytes retrieved Number of suitable oocytes Number of pronucleated oocytes Fertilization rate Number of embryos Cleavage stage rate Number of top quality embryos Rate of high quality embryos
39 39 39 33 33 33 33 33 33
0.13 0.08 0.16 0.01 0.12 0.01 0.09 0.09 0.08
0.15 0.11 0.13 0.02 0.07 0.01 0.17 0.07 0.32
39 39 39 30 30 30 30 30 30
0.30 0.25 0.26 0.11 0.20 0.04 0.34 0.06 0.28
0.42* 0.55* 0.47* 0.38* 0.02 0.34* 0.09 0.11 0.17
78 78 78 63 63 63 63 63 63
0.22 0.16 0.20 0.06 0.16 0.03 0.21 0.02 0.08
0.12 0.18 0.18 0.18 0.05 0.18 0.07 0.09 0.12
Reported values refer to Spearman’s rank correlation coefficient (Spearman’s Rho). *p < 0.05.
650
ª 2015 Nordic Federation of Societies of Obstetrics and Gynecology, Acta Obstetricia et Gynecologica Scandinavica 94 (2015) 646–653
L. Benaglia et al.
An alternative explanation is that the spread of iron is spatially limited and can therefore be detected only in follicles that are in close contact with the endometrioma. Sanchez et al. separated follicles of the affected ovary according to their proximity to the endometrioma and could therefore report results separately. In contrast, we evaluated iron concentration in the whole follicular pool. In other words, if the spread of iron is spatially very limited, we may have missed it because of a diluting effect. Finally, even if less likely, it cannot be excluded that the discrepancy between the two studies may also be related to the different measurement system used. Ferritin is a 24-subunit molecule that plays a central role for iron metabolism because it sequesters iron during times of iron excess and releases iron when iron is scarce. The function of ferritin is, however, not limited to the mere regulation of iron storage. Indeed, ferritin mRNA transcription is stimulated by iron but even more significantly by oxidative damage and inflammation. This regulation reflects two different, albeit linked functions. On one hand, modulation of ferritin production aims to maintain free iron at constant concentration. Ferritin increases to counterbalance the excess in free iron and to store it. On the other hand, the enhanced ferritin production secondary to oxidative stress and inflammation aims at a different but relevant function, i.e. minimizing the potential for ROS formation. Decreasing the labile iron pool indeed prevents the cascade of events responsible for ROS formation. We speculate that the enhanced ferritin concentration detected in the follicular fluid of follicles developing in the proximity of ovarian endometriomas may be explained by both regulatory mechanisms. First, iron may diffuse in the surrounding tissue from the endometriotic fluid where it is present in extremely high concentrations. The increased ferritin may reflect this increased burden. Failing to demonstrate any difference in iron concentration between the two gonads does not rule out these pathogenic mechanisms because this is a chronic exposure and the excess in iron may be already stored in the ferritin. Moreover, we measured the total concentration of iron and not exclusively free iron, which is the form of the molecule involved in the regulation of ferritin secretion. The observation of a significant correlation between follicular ferritin and follicular iron when considering together the studied gonads tends to support this interpretation. Second, the observed increased ferritin may reflect an enhanced state of inflammation in ovaries with endometriomas. Tumor necrosis factor-a, interleukin-1 and interleukin-6 were demonstrated to trigger the synthesis of ferritin in relation to the inflammatory response. Studies focusing on the cytokine milieu in the endometriomas are conflicting (10,18–21) but there are data suggesting that at least interleukin-6 is enhanced
Iron, ferritin, and endometriomas
(20,21). Notably, the extra-gonadal milieu may also participate in the regulation of iron and ferritin concentration in the ovarian follicles. The observation that, in our study, both iron and ferritin correlated between the two gonads supports this possibility. Of relevance here is the observation that the concentration of iron, interleukin-6 and tumor necrosis factor-a. is markedly higher in the peritoneal fluid of women with endometriosis (22–25). Together, our results suggest that iron contained in the endometriomas may actually diffuse in the surrounding tissue. However, within the ovary, there are biological mechanisms effectively counterbalancing the enhanced labile free iron load. The higher concentration of ferritin despite a similar concentration of total iron should be interpreted within this view. These mechanisms effectively protect from the deleterious effects of free iron as supported by the lack of biologically relevant correlations between ferritin and variables reflecting oocyte developmental competence. Our observation is in line with the available clinical evidence obtained in women with endometriomas who are undergoing IVF. Indeed, two recent studies on women with unilateral unoperated endometriomas documented a similar responsiveness to ovarian hyperstimulation in the affected and contralateral intact gonads (26,27). Moreover, two studies comparing IVF outcome in women with bilateral endometriomas and in a control group of unaffected women failed to document any detrimental impact on the chances of pregnancy (28,29). Some limitations of our study should be acknowledged. First, we did not attempt to investigate markers of oxidative stress and to identify the cells secreting ferritin within the follicular fluid. This information would have provided a more comprehensive vision of the situation. Further studies are therefore warranted to specifically elucidate these aspects. Second, we lack a histological diagnosis of endometriosis. This is a common but accepted limit of most studies evaluating the impact of ovarian endometriomas on ovarian function. This limitation is, however, of scant relevance given the high accuracy of transvaginal ultrasound. The sensitivity and specificity of this diagnostic tool have been reported to be 84–100% and 90–100%, respectively (14). Moreover, to further minimize the risk of misdiagnosis, women carrying endometriomas with atypical sonographic appearance were excluded and the presence of the endometriomas had to be documented on at least two occasions and at least two menstrual cycles apart. Third, we focused on a very particular population and we cannot exclude some confounders. Of relevance here is that we exclusively recruited women with infertility and an indication to IVF, the dimension of the ovarian endometriomas was generally small and one-third of the women had previously undergone ovarian surgery.
ª 2015 Nordic Federation of Societies of Obstetrics and Gynecology, Acta Obstetricia et Gynecologica Scandinavica 94 (2015) 646–653
651
Iron, ferritin, and endometriomas
Inferences of our findings to the whole population of women with ovarian endometriomas should therefore be made with caution. In conclusion, iron may diffuse from ovarian endometriomas into the adjacent ovarian tissue. However, this phenomenon does not seem to markedly affect ovarian function because of some effective biological mechanisms such as ferritin storage that properly counterbalance the potentially highly detrimental effects of free iron. These results tend to support conservative management of reproductive-age women with ovarian endometriomas. In other words, our data do not support the idea that removing these cysts may prevent damage to ovarian function. Further evidence is needed to definitely disentangle this issue. Studies reporting on larger cysts and studies investigating ovarian function over time in affected gonads would be particularly interesting because we cannot exclude both a role of the dimension of the endometriomas and a time-related effect.
Funding This study received only local institutional funds. References 1. Iizuka M, Igarashi M, Abe Y, Ibuki Y, Koyasu Y, Ikuma K. Chemical assay of iron in ovarian cysts: a new diagnostic method to evaluate endometriotic cysts. Gynecol Obstet Invest. 1998;46:58–60. 2. Slater M, Quagliotto G, Cooper M, Murphy CR. Endometriotic cells exhibit metaplastic change and oxidative DNA damage as well as decreased function, compared to normal endometrium. J Mol Histol. 2005;36:257–63. 3. Kobayashi H, Yamada Y, Kanayama S, Furukawa N, Noguchi T, Haruta S, et al. The role of iron in the pathogenesis of endometriosis. Gynecol Endocrinol. 2009;25:39–52. 4. Jomova K, Valko M. Importance of iron chelation in free radical-induced oxidative stress and human disease. Curr Pharm Des. 2011;17:3460–73. 5. Vercellini P, Crosignani P, Somigliana E, Vigan o P, Buggio L, Bolis G, et al. The ‘incessant menstruation’ hypothesis: a mechanistic ovarian cancer model with implications for prevention. Hum Reprod. 2011;26:2262–73. 6. Matsuzaki S, Schubert B. Oxidative stress status in normal ovarian cortex surrounding ovarian endometriosis. Fertil Steril. 2010;93:2431–2. 7. Seino T, Saito H, Kaneko T, Takahashi T, Kawachiya S, Kurachi H. Eight-hydroxy-1393 20 -deoxyguanosine in granulosa cells is correlated with the quality of oocytes and embryos in an in vitro fertilization-embryo transfer program. Fertil Steril. 2002;77:1184–90.
652
L. Benaglia et al.
8. Karuputhula NB, Chattopadhyay R, Chakravarty B, Chaudhury K. Oxidative status in granulosa cells of infertile women undergoing IVF. Syst Biol Reprod Med. 2013;59:91–8. 9. Bryan N, Ahswin H, Smart N, Bayon Y, Wohlert S, Hunt JA. Reactive oxygen species (ROS) – a family of fate deciding molecules pivotal in constructive inflammation and wound healing. Eur Cell Mater. 2012;24:249–65. 10. Sanchez AM, Vigan o P, Somigliana E, Panina-Bordignon P, Vercellini P, Candiani M. The distinguishing cellular and molecular features of the endometriotic ovarian cyst: from pathophysiology to the potential endometriomamediated damage to the ovary. Hum Reprod Update. 2014;20:217–30. 11. Sanchez AM, Papaleo E, Corti L, Santambrogio P, Levi S, Vigan o P, et al. Iron availability is increased in individual human ovarian follicles in close proximity to an endometrioma compared with distal ones. Hum Reprod. 2014;29:577–83. 12. Jana SK, K NB, Chattopadhyay R, Chakravarty B, Chaudhury K. Upper control limit of reactive oxygen species in follicular fluid beyond which viable embryo formation is not favorable. Reprod Toxicol. 2010;29:447–51. 13. Somigliana E, Paffoni A, Busnelli A, Cardellicchio L, Leonardi M, Filippi F, et al. IVF outcome in poor responders failing to produce viable embryos in the preceding cycle. Reprod Biomed Online. 2013;26:569–76. 14. Savelli L. Transvaginal sonography for the assessment of ovarian and pelvic endometriosis: how deep is our understanding? Ultrasound Obstet Gynecol. 2009;33:497– 501. 15. Practice Committee of the American Society for Reproductive Medicine. Endometriosis and infertility: a committee opinion. Fertil Steril. 2012;98:591–8. 16. Dunselman GA, Vermeulen N, Becker C, Calhaz-Jorge C, D’Hooghe T, De Bie B, et al. ESHRE guideline: management of women with endometriosis. Hum Reprod. 2014;29:400–12. 17. Ragni G, Somigliana E, Benedetti F, Paffoni A, Vegetti W, Restelli L, et al. Damage to ovarian reserve associated with laparoscopic excision of endometriomas: a quantitative rather than a qualitative injury. Am J Obstet Gynecol. 2005;193:1908–14. 18. Fasciani A, D’Ambrogio G, Bocci G, Monti M, Genazzani AR, Artini PG. High concentrations of the vascular endothelial growth factor and interleukin-8 in ovarian endometriomata. Mol Hum Reprod. 2000;6:50–4. 19. Darai E, Detchev R, Hugol D, Quang NT. Serum and cyst fluid levels of interleukin (IL) -6, IL-8 and tumour necrosis factor-alpha in women with endometriomas and benign and malignant cystic ovarian tumours. Hum Reprod. 2003;18:1681–5. 20. Velasco I, Acien P, Campos A, Acien MI, Ruiz-Macia E. Interleukin-6 and other soluble factors in peritoneal fluid
ª 2015 Nordic Federation of Societies of Obstetrics and Gynecology, Acta Obstetricia et Gynecologica Scandinavica 94 (2015) 646–653
L. Benaglia et al.
21.
22.
23.
24.
and endometriomas and their relation to pain and aromatase expression. J Reprod Immunol. 2010;84:199– 205. Opøien HK, Fedorcsak P, Polec A, Stensen MH, Abyholm T, Tanbo T. Do endometriomas induce an inflammatory reaction in nearby follicles? Hum Reprod. 2013;28:1837– 45. Van Langendonckt A, Casanas-Roux F, Donnez J. Iron overload in the peritoneal cavity of women with pelvic endometriosis. Fertil Steril. 2002;78:712–8. Kyama CM, Overbergh L, Debrock S, Valckx D, Vander Perre S, Meuleman C, et al. Increased peritoneal and endometrial gene expression of biologically relevant cytokines and growth factors during the menstrual phase in women with endometriosis. Fertil Steril. 2006;85:1667– 75. Lousse JC, Defrere S, Van Langendonckt A, Gras J, Gonzalez-Ramos R, Colette S, et al. Iron storage is significantly increased in peritoneal macrophages of endometriosis patients and correlates with iron overload in peritoneal fluid. Fertil Steril. 2009;91:1668–75.
Iron, ferritin, and endometriomas
25. Birt JA, Nabli H, Stilley JA, Windham EA, Frazier SR, Sharpe-Timms KL. Elevated peritoneal fluid TNF-a incites ovarian early growth response factor 1 expression and downstream protease mediators: a correlation with ovulatory dysfunction in endometriosis. Reprod Sci. 2013;20:514–23. 26. Almog B, Shehata F, Sheizaf B, Tan SL, Tulandi T. Effects of ovarian endometrioma on the number of oocytes retrieved for in vitro fertilization. Fertil Steril. 2011;95:525–7. 27. Benaglia L, Pasin R, Somigliana E, Vercellini P, Ragni G, Fedele L. Unoperated ovarian endometriomas and responsiveness to hyperstimulation. Hum Reprod. 2011;26:1356–61. 28. Reinblatt SL, Ishai L, Shehata F, Son WY, Tulandi T, Almog B. Effects of ovarian endometrioma on embryo quality. Fertil Steril. 2011;30:2700–2. 29. Benaglia L, Bermejo A, Somigliana E, Faulisi S, Ragni G, Fedele L, et al. In vitro fertilization outcome in women with unoperated bilateral endometriomas. Fertil Steril. 2013;99:1714–9.
ª 2015 Nordic Federation of Societies of Obstetrics and Gynecology, Acta Obstetricia et Gynecologica Scandinavica 94 (2015) 646–653
653
Copyright of Acta Obstetricia et Gynecologica Scandinavica is the property of WileyBlackwell and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.
AOGS M A I N R E SE A RC H A R TI C LE
Intrafollicular iron and ferritin in women with ovarian endometriomas LAURA BENAGLIA1, ALESSIO PAFFONI1, ALICE MANGIARINI1, LILIANA RESTELLI1, NORA BETTINARDI2, EDGARDO SOMIGLIANA1, PAOLO VERCELLINI1,3 & LUIGI FEDELE1,3 1
Obstetrics-Gynecology Department, Policlinic Maggiore Hospital (Ospedale Maggiore), Milan, 2Analysis Laboratory, Ca’ Granda Foundation, Policlinic Maggiore Hospital (Ospedale Maggiore), Milan, and 3University of Milan, Milan, Italy
Key words Endometriosis, iron, ferritin, endometrioma, follicular fluid Correspondence Laura Benaglia, Infertility Unit, Fondazione Ca’ Granda, Ospedale Maggiore Policlinico, Via M. Fanti 6, 20122 Milan, Italy. E-mail: [email protected] Conflicts of interest The authors have stated explicitly that there are no conflicts of interest in connection with this article. Please cite this article as: Benaglia L, Paffoni A, Mangiarini A, Restelli L, Bettinardi N, Somigliana E, et al. Intrafollicular iron and ferritin in women with ovarian endometriomas. Acta Obstet Gynecol Scand 2015; 94: 646–653. Received: 25 September 2014 Accepted: 30 March 2015
Abstract Objective. To evaluate whether iron contained in ovarian endometriomas can diffuse through the cyst wall and negatively affect ovarian function. Design. Prospective case series. Setting. Infertility unit in an academic setting. Population. Thirty-nine infertile women with unilateral endometriomas who underwent in vitro fertilization. Methods. Iron and ferritin assessments in pools of follicular fluids obtained from affected and contralateral intact gonads. Main outcome measures. Iron and ferritin concentrations. Results. Follicular fluid iron content did not differ between the two gonads. The median [interquartile range (IQR)] follicular concentrations in the affected and unaffected ovaries were 59 (IQR 44–74) and 59 (IQR 47–73) lg/dL, respectively (p = 0.77). Conversely, ferritin concentration was significantly higher in affected gonads. The median (IQR) concentrations of ferritin in the affected and unaffected ovaries were 57 (IQR 31–146) and 33 (IQR 23–67) lg/mL, respectively (p = 0.026). When considering together the 78 studied ovaries, no significant correlations emerged between follicular iron and ferritin and variables reflecting ovarian responsiveness and oocyte developmental competence. Conclusions. Iron may diffuse from ovarian endometriomas into the adjacent ovarian tissue. However, this phenomenon does not appear to markedly affect ovarian function. Some effective biological mechanisms such as ferritin storage may effectively sequester free iron, so limiting its detrimental effects.
DOI: 10.1111/aogs.12647
ICSI, intracytoplasmatic sperm injection; IQR, interquartile range; IVF, in vitro fertilization; ROS, reactive oxygen species.
Abbreviations:
reaction that may be potentially harmful to the surrounding cells (4,5). ROS are indeed highly reactive and become stable by acquiring electrons from nucleic acids,
Introduction Endometriotic cysts contain a huge amount of free iron. The concentration exceeds 100 nmol/L, which is up to 10 000-fold higher than the concentration normally present in serum (0.013–0.027 mmol/L) (1). Accordingly, massive iron staining has also been observed in the stroma of the cyst, mostly located in macrophages (2,3). The elevated concentration of iron in ovarian endometriomas is a matter of concern because nonprotein-bound “free” or “catalytic” iron can mediate the production of reactive oxygen species (ROS) via the Fenton reaction, a
646
Key Message Iron may diffuse from ovarian endometriomas into the adjacent ovarian tissue. Its detrimental effects are, however, lessened by some biological compensatory mechanisms that prevent major effects on ovarian function.
ª 2015 Nordic Federation of Societies of Obstetrics and Gynecology, Acta Obstetricia et Gynecologica Scandinavica 94 (2015) 646–653
L. Benaglia et al.
lipids, proteins, carbohydrates or any nearby molecule, causing a cascade of chain reactions resulting in cellular damage. There is some recent evidence suggesting that the detrimental effect of endometrioma contents is not merely limited to the cells lining the internal layer of the cysts. The amount of oxidative stress affecting the normal ovarian cortex surrounding an endometrioma was shown to be greater than that in other types of cysts (6). Some studies have also demonstrated that granulosa cells from infertile patients with endometriosis exhibit more signs of oxidative stress when compared with those from unaffected patients (7,8). On this basis, it has been hypothesized that factors present in the endometriomas, and iron in particular, may diffuse in the surrounding tissue causing ROS generation, in a similar way to what happens inside the cyst (9–11). This aspect is of biological relevance, considering that oxidative stress was shown to negatively affect ovarian developmental competence (12). In the present study, we tested the above-mentioned hypothesis that iron contained in endometriomas may diffuse through the cyst wall and negatively affect folliculogenesis. With this aim, we recruited women with unilateral endometriomas undergoing in vitro fertilization (IVF) and compared iron and ferritin concentrations in the follicular fluid of the two gonads. Moreover, we correlated iron and ferritin concentrations to oocyte developmental competence and follicular response to controlled ovarian stimulation with gonadotropins.
Material and methods Women selected for IVF–intracytoplasmatic sperm injection (ICSI) cycle at the Infertility Unit of the Fondazione Ca’ Granda, Ospedale Maggiore Policlinic, Milan, Italy, between January 2012 and January 2013 were prospectively evaluated for study entry. Eligible women were those with unilateral ovarian endometriomas. More specifically, inclusion criteria were (i) age 18–42 years, (ii) indication for IVF-ICSI, (iii) presence of one or more unilateral ovarian endometriomas on transvaginal ultrasound, (iv) absence of nonendometriotic ovarian cysts, (v) acceptance to participate. Women entering the study were subsequently excluded in the following circumstances: (i) cancelled oocyte retrieval for any reason, (ii) accidental follicular fluid contamination with endometrioma content, (iii) absent follicular growth in any of the two gonads. Women were enrolled only for their first completed cycle performed during the study period. The study was accepted by the local Institutional Review Board and all participating women signed an informed consent.
Iron, ferritin, and endometriomas
During the IVF-ICSI cycle, women were monitored and managed according to a standardized clinical protocol as reported in detail elsewhere (13). Briefly, the patients underwent baseline transvaginal ultrasound the month preceding the ovarian hyperstimulation. The presence of ovarian endometriomas and previous documentation on their presence were systematically recorded at this time. Ovarian endometrioma was defined as a roundshaped cystic mass with a minimum diameter of 10 mm, with thick walls, regular margins, homogeneous low echogenic fluid content with scattered internal echoes, and without papillary projections (14). Doubtful cases were excluded. To rule out functional cysts, the presence of the endometriomas had also to be documented at least on one previous ultrasound scan performed at least 2 months before the IVF-ICSI cycle. In general, the first assessment was obtained during the standard diagnostic work-up that all infertile women undergo in our unit and that systematically included transvaginal ultrasound. The second assessment was performed in the IVF cycle preparation, hence in the month preceding the ovarian hyperstimulation. The diameter of the endometriomas was calculated as the mean of three perpendicular diameters. The policy of the unit for endometriomas up to a diameter of 4 cm was to advise conservative management if the woman was asymptomatic or if the pain complaints could be effectively treated with medical treatment (15,16). This limit of 4 cm was, however, not stringent. For larger cysts, the woman was informed that the available evidence is limited and a shared decision was taken after an in-depth discussion of the pros and cons of conservative and surgical approaches. The regimen and the dose of gonadotropins were chosen on an individual basis based on age, day 3 serum follicle stimulating hormone, serum anti-M€ ullerian hormone, sonographic assessment of antral follicular count, and results from previous ovarian hyperstimulation cycles. Gonadotropins used were recombinant follitropin-a (Gonal-F; Merck Serono, Rome, Italy) or human menopausal gonadotropin (Meropur; Ferring, Milan, Italy). The maximal dose was 450 IU/day. During ovarian hyperstimulation women underwent serial transvaginal ultrasound and had serum estrogen and progesterone assessments. Human chorionic gonadotropin was administered subcutaneously when the leading follicles had a mean diameter >18 mm. The total number of growing follicles (i.e. those with a mean diameter ≥ 11 mm calculated as the mean of three perpendicular diameters) per ovary was systematically recorded at this time. Oocyte retrieval was performed transvaginally 36 h after human chorionic gonadotropin administration. All follicles with a mean diameter >10 mm were aspirated. All efforts were made to avoid follicular fluid contamina-
ª 2015 Nordic Federation of Societies of Obstetrics and Gynecology, Acta Obstetricia et Gynecologica Scandinavica 94 (2015) 646–653
647
Iron, ferritin, and endometriomas
tion with the endometrioma content. However, if it was necessary to reach follicles located behind the endometrioma, the cyst could be transfixed. As mentioned above, frank contamination with endometrioma content was an exclusion criterion. After completion of oocyte collection in one ovary, the aspiration system was rinsed before moving to the contralateral ovary and the embryologist was warned to separate follicular fluids belonging to different ovaries. Immediately after recovery of oocytes, a pool of all aspirated follicular fluids was collected separately for each ovary into a large Petri dish. A maximum volume of 12 mL of follicular fluid was then transferred from the pool to a sterile tube and successively centrifuged at 350 g for 10 min at room temperature to separate it from cells. For every patient, two samples of the centrifuged follicular fluids for each of the two ovaries were transferred into cryovials and stored at 20°C until iron and ferritin measurement. Cryovials were identified by the embryologist with a label indicating patients’ identification code and “right” or “left” ovary. Both the embryologist and the biologist performing iron and ferritin measurement were masked from the presence or absence of endometriomas in the specific ovary. IVF or ICSI procedures were performed according to standardized criteria (17) and embryo transfer was performed 2–5 days after the oocyte collection. Cycles could be cancelled because of low or hyper-ovarian response. These women were excluded from the present analysis. Clinical pregnancy was defined as the ultrasonographic demonstration of intrauterine viable embryos 4–5 weeks after embryo transfer. Iron and ferritin were measured on an automatic platform (Cobas 8000; Roche Italia, Monza, Italy). Colorimetric and immunochemical methods were used for iron and ferritin, respectively (Cobas Iron gen. 2; Elecsys Ferritin, Roche, Italy). More specifically, ferric iron is unbound from proteins by an acid reaction (pH < 2.0). The ferric ion is then reduced to ferrous iron by ascorbate. Bivalent iron ions form a colored complex with ferrozine, detected at 700/570 nm; color intensity is directly proportional to iron concentration. For ferritin, the reaction is conduct in four steps: the sample is incubated with a biotinylated and ruthenium-labeled monoclonal antibody anti-ferritin; the sandwich complex that is formed binds itself to a solid phase by biotin–streptavidin; after washing, the reaction mixture is aspirated into the measuring cell where the microparticles are magnetically captured onto the surface of the electrode. Application of a voltage to the electrode then induces chemiluminescent emission, which is measured by a photomultiplier. Both methods are certified for serum/plasma and were previously validated in follicular fluids with
648
L. Benaglia et al.
particular regard towards coefficient of variation, linearity, drift, and hemolysis (data not shown).
Statistical analysis Analysis of the data was carried out with the STATISTICS PACKAGE FOR SOCIAL SCIENCES v18.0 (SPSS Inc., Chicago, IL, USA). p values < 0.05 were considered significant. Data are presented as number (%), mean SD or median [interquartile range (IQR)], as appropriate. Given the paucity of available data on follicular iron and ferritin statistical distribution, it was decided to consider them as non-normally distributed and data were analyzed using nonparametric statistics. The sample size was calculated by setting type I and II errors at 0.05 and 0.20 and considering clinically relevant a threefold increase in the proportion of follicular pools with high iron or ferritin in affected ovaries. Given the absence of previous data indicating a threshold to define high levels for both molecules in the follicular fluid, we arbitrarily set these thresholds at the 90th centile of the distribution observed in the intact gonads. In other words, the frequency of elevated iron or ferritin in intact gonads was 10% (per definition) and we claimed as biologically relevant detecting this condition in 25% of the affected gonads (odds ratio ≥ 3). On this basis, we aimed to have a sample size of at least 35 women. A Wilcoxon test for paired data and McNemar test were used to compare affected and unaffected gonads, as appropriate. Correlations were tested using nonparametric statistics and data were presented using the Spearman’s rank correlation coefficient (Spearman’s Rho).
Results Thirty-nine women were recruited. Baseline characteristics of the selected women and IVF-ICSI outcome are shown in Tables 1 and 2, respectively. A high proportion of women had a concomitant male factor of infertility and required ICSI. Ovarian responsiveness in the affected and contralateral intact gonads did not differ. Specifically, the mean SD numbers of developed follicles, oocytes retrieved and suitable oocytes were 5.4 3.1 and 6.6 5.3 (p = 0.12), 3.9 3.1 and 4.5 2.8 (p = 0.29) and 2.5 2.3 and 3.1 2.4 (p = 0.27), respectively. Oocyte developmental competence did not differ. Indeed, the number of embryos and the total number of topquality embryos obtained from the two gonads were 1.6 1.9 and 1.8 1.8 (p = 0.51) and 1.0 1.5 and 0.9 1.4 (p = 0.81), respectively. The follicular concentration of iron and ferritin according to the ovary of origin is illustrated in Figure 1. Specifically, the median (IQR) concentrations of iron in the affected and unaffected ovaries weres 59 (44–74) and 59
ª 2015 Nordic Federation of Societies of Obstetrics and Gynecology, Acta Obstetricia et Gynecologica Scandinavica 94 (2015) 646–653
L. Benaglia et al.
Iron, ferritin, and endometriomas
Table 1. Baseline characteristics of the studied population (n = 39).
Characteristics Age (years) Body mass index (kg/m2) Previous pregnancies Smoking Duration of infertility (years) Concomitant male factor of infertility Previous IVF-ICSI cycles Day 3 serum FSH (IU/mL) AMH (ng/mL) AFC (both ovaries) Previous surgery for ovarian endometriomasa Side of the endometriomas Right Left Number of endometriomas 1 2 Mean diameter of the endometriomas (mm)b CA-125 (IU/mL) Medical treatment before entry into cycle Oral contraceptives Progestins
Number (%), mean SD or median (IQR) 35.0 3.8 21.6 2.7 8 (20%) 7 (18%) 3.5 2.0 24 (62%) 9 (23%) 7.4 2.3 2.3 1.6 10 (5–15) 12 (31%)
21 (54%) 18 (46%) 33 (85%) 6 (15%) 24 10 40 (25–61) 10 (26%) 7 (18%)
€llerian hormone; CA-125, AFC, antral follicle count; AMH, anti-Mu cancer antigen 125; FSH, follicle-stimulating hormone; IQR, interquartile range; IVF-ICSI: in vitro fertilization–intracytoplasmic sperm injection. a Surgery was performed in affected ovary in five cases, in unaffected ovaries in two cases and bilaterally in the remaining five cases. b If more than one endometrioma was present, data refer to the largest one.
(47–73) lg/dL, respectively (p = 0.77). The median (IQR) concentrations of ferritin were 57 (31–146) and 33 (23–67) lg/mL, respectively (p = 0.026). Iron concentration was above the 90th centile of the distribution in unaffected ovaries (90 lg/dL) in three (8%) intact and four (10%) affected gonads (p = 1.00). Ferritin concentration was above the 90th centile of the distribution in unaffected ovaries (132 lg/mL) in three (8%) intact and 11 (28%) affected gonads (p = 0.021). Follicular concentration of iron correlated between the two ovaries (q = 0.76, p < 0.001). A similar figure emerged for follicular ferritin (q = 0.55, p < 0.001). Finally, a significant correlation was documented also when correlating ferritin and iron in the 78 available gonads (q = 0.42, p < 0.001). Twelve women (31%) were previously operated on for ovarian endometriomas (Table 1). Lesions were bilateral in five women and unilateral in the remaining seven, allowing us to evaluate follicular levels of iron and ferritin
Table 2. Characteristics of in vitro fertilization–intracytoplasmatic sperm injection cycles in the studied population (n = 39).
Characteristics Regimen of ovarian hyper-stimulation Long protocol GnRH antagonists Other Duration of stimulation (days) Total dose of FSH administered (IU) E2 at the time of hCG administration (pg/mL) Total number of follicles ≥11 mm Number of oocytes retrieved Number of suitable oocytesa Technique used IVF ICSI Fertilization rate (%) Number of embryos Cleavage rate (%) Number of top-quality embryos Rate of top-quality embryos (%) Number of embryos transferred None 1 2 Clinical pregnancies Embryos implanted (implantation rate)
Number (%), mean SD or median (IQR)
26 (67%) 6 (20%) 5 (13%) 10.4 2.0 2700 (1500–3285) 2013 (1568–2645) 11.8 6.9 8.5 4.7 6.0 3.7 14 (36%) 25 (64%) 75 (43–100) 3.4 2.9 65 (33–100) 2.8 2.5 50 (20–71) 7 16 16 11 11
(18%) (41%) (41%) (28%) (23%)
E2, estradiol; FSH, follicle-stimulating hormone; GnRH, gonadotropinreleasing hormone; hCG, human chorionic gonadotropin; IQR, interquartile range; ICSI, intracytoplasmic sperm injection; IVF, in vitro fertilization. a Suitable oocytes correspond to oocytes used. They were defined as metaphase II oocytes or type 1 cumulus–oocytes complexes according to the ESHRE Istanbul Consensus Conference (2011).
in 17 previously operated ovaries (10 with recurrence). The median (IQR) iron concentrations in operated and nonoperated (n = 61) gonads were 67 (40–77) and 59 (48–72) lg/dL, respectively (p = 0.78). The median (IQR) ferritin concentrations were 69 (24–185) and 41 (23–66) lg/mL, respectively (p = 0.26). The analyses on follicular concentration of iron and ferritin were repeated excluding these 12 previously operated women (n = 27). The median (IQR) concentrations of iron in the affected and unaffected ovaries were 59 (48–72) and 58 (47–73) lg/dL, respectively (p = 0.68). The median (IQR) concentrations of ferritin were 49 (26–90) and 33 (23–66) lg/mL, respectively (p = 0.05). Iron concentration was above the 90th centile in three (11%) affected and three (11%) intact gonads (p = 1.00). Ferritin concentration was above the 90th centile in six (23%) affected and one (4%) intact gonad (p = 0.12).
ª 2015 Nordic Federation of Societies of Obstetrics and Gynecology, Acta Obstetricia et Gynecologica Scandinavica 94 (2015) 646–653
649
Follicular levels (Iron µg/dL; Ferritin µg/mL)
Iron, ferritin, and endometriomas
L. Benaglia et al.
250
Discussion
200
150
100
*
50
0 Iron
Ferritin
Figure 1. Follicular concentration of iron and ferritin in ovaries with and without endometriomas. Box bars of the affected ovaries are colored grey and those from unaffected gonads are white. Concentration of iron did not differ (p = 0.77) whereas a statistically significant difference was documented for ferritin (p = 0.026). *Statistical significant p value.
Correlations of follicular iron and ferritin with variables reflecting ovarian function and oocyte developmental competence are shown in Table 3. Considering follicular iron, none of the correlations were significant. In the affected gonad, no significant correlation emerged between ferritin and any of the variables considered. In the contralateral intact gonads, significant correlations emerged between follicular ferritin and the number of developing follicles, the number of oocytes retrieved, the number of suitable oocytes, the number of pronucleated oocytes and the number of embryos obtained. All of these correlations were lost when considering data from both ovaries together. All the correlations tested in Table 3 were also evaluated in the subgroup of women who did not previously undergo ovarian surgery (n = 27) and none was significant (data not shown).
We demonstrated that levels of ferritin, but not total iron, are enhanced in follicles developing in proximity to ovarian endometriomas. Moreover, we did not observe biologically significant correlations between these two molecules and responsiveness to ovarian hyperstimulation or oocyte developmental competence. In fact, some correlations were significant (mainly between ferritin and ovarian responsiveness), but they contrasted with the hypothesis tested, they were limited to the intact ovaries and they were lost when considering both gonads together. On these bases, we interpreted these significant correlations as being due to a type I error. Our results are partly in line with those from a very recent study from Sanchez et al. who tested the same hypothesis (11). These authors recruited 13 women with unilateral endometriomas and evaluated follicular content in ferritin and iron in individual follicles according to their location. Specifically, they compared follicles in close contact with the endometriomas (n = 35), those located in the affected ovary but not in close contact with the cyst (n = 28) and those from the contralateral gonad (n = 40). The authors overall observed a decreasing gradient of ferritin concentration in the three groups, in line with our findings. In contrast to our data, however, these authors also observed this gradient for iron concentration. This discrepancy is difficult to explain. We hypothesize a type I error in the study from Sanchez et al., considering in particular that these authors analyzed data from only 13 women (11). The use of several follicles from the same woman may have indeed lead to overappreciation of the association. In other words, if diffusion of iron from an endometrioma to the surrounding tissue is a rare but possible event, the inclusion of even a single case contributing for several follicles may have artificially led to detecting a statistically significant difference.
Table 3. Correlation (Spearman’s Rho) between follicular iron and ferritin and variables of ovarian function. Affected gonad
Unaffected gonad
Both gonads
Variables
n
Iron
Ferritin
n
Iron
Ferritin
n
Iron
Ferritin
Number of follicles ≥11 mm Number of oocytes retrieved Number of suitable oocytes Number of pronucleated oocytes Fertilization rate Number of embryos Cleavage stage rate Number of top quality embryos Rate of high quality embryos
39 39 39 33 33 33 33 33 33
0.13 0.08 0.16 0.01 0.12 0.01 0.09 0.09 0.08
0.15 0.11 0.13 0.02 0.07 0.01 0.17 0.07 0.32
39 39 39 30 30 30 30 30 30
0.30 0.25 0.26 0.11 0.20 0.04 0.34 0.06 0.28
0.42* 0.55* 0.47* 0.38* 0.02 0.34* 0.09 0.11 0.17
78 78 78 63 63 63 63 63 63
0.22 0.16 0.20 0.06 0.16 0.03 0.21 0.02 0.08
0.12 0.18 0.18 0.18 0.05 0.18 0.07 0.09 0.12
Reported values refer to Spearman’s rank correlation coefficient (Spearman’s Rho). *p < 0.05.
650
ª 2015 Nordic Federation of Societies of Obstetrics and Gynecology, Acta Obstetricia et Gynecologica Scandinavica 94 (2015) 646–653
L. Benaglia et al.
An alternative explanation is that the spread of iron is spatially limited and can therefore be detected only in follicles that are in close contact with the endometrioma. Sanchez et al. separated follicles of the affected ovary according to their proximity to the endometrioma and could therefore report results separately. In contrast, we evaluated iron concentration in the whole follicular pool. In other words, if the spread of iron is spatially very limited, we may have missed it because of a diluting effect. Finally, even if less likely, it cannot be excluded that the discrepancy between the two studies may also be related to the different measurement system used. Ferritin is a 24-subunit molecule that plays a central role for iron metabolism because it sequesters iron during times of iron excess and releases iron when iron is scarce. The function of ferritin is, however, not limited to the mere regulation of iron storage. Indeed, ferritin mRNA transcription is stimulated by iron but even more significantly by oxidative damage and inflammation. This regulation reflects two different, albeit linked functions. On one hand, modulation of ferritin production aims to maintain free iron at constant concentration. Ferritin increases to counterbalance the excess in free iron and to store it. On the other hand, the enhanced ferritin production secondary to oxidative stress and inflammation aims at a different but relevant function, i.e. minimizing the potential for ROS formation. Decreasing the labile iron pool indeed prevents the cascade of events responsible for ROS formation. We speculate that the enhanced ferritin concentration detected in the follicular fluid of follicles developing in the proximity of ovarian endometriomas may be explained by both regulatory mechanisms. First, iron may diffuse in the surrounding tissue from the endometriotic fluid where it is present in extremely high concentrations. The increased ferritin may reflect this increased burden. Failing to demonstrate any difference in iron concentration between the two gonads does not rule out these pathogenic mechanisms because this is a chronic exposure and the excess in iron may be already stored in the ferritin. Moreover, we measured the total concentration of iron and not exclusively free iron, which is the form of the molecule involved in the regulation of ferritin secretion. The observation of a significant correlation between follicular ferritin and follicular iron when considering together the studied gonads tends to support this interpretation. Second, the observed increased ferritin may reflect an enhanced state of inflammation in ovaries with endometriomas. Tumor necrosis factor-a, interleukin-1 and interleukin-6 were demonstrated to trigger the synthesis of ferritin in relation to the inflammatory response. Studies focusing on the cytokine milieu in the endometriomas are conflicting (10,18–21) but there are data suggesting that at least interleukin-6 is enhanced
Iron, ferritin, and endometriomas
(20,21). Notably, the extra-gonadal milieu may also participate in the regulation of iron and ferritin concentration in the ovarian follicles. The observation that, in our study, both iron and ferritin correlated between the two gonads supports this possibility. Of relevance here is the observation that the concentration of iron, interleukin-6 and tumor necrosis factor-a. is markedly higher in the peritoneal fluid of women with endometriosis (22–25). Together, our results suggest that iron contained in the endometriomas may actually diffuse in the surrounding tissue. However, within the ovary, there are biological mechanisms effectively counterbalancing the enhanced labile free iron load. The higher concentration of ferritin despite a similar concentration of total iron should be interpreted within this view. These mechanisms effectively protect from the deleterious effects of free iron as supported by the lack of biologically relevant correlations between ferritin and variables reflecting oocyte developmental competence. Our observation is in line with the available clinical evidence obtained in women with endometriomas who are undergoing IVF. Indeed, two recent studies on women with unilateral unoperated endometriomas documented a similar responsiveness to ovarian hyperstimulation in the affected and contralateral intact gonads (26,27). Moreover, two studies comparing IVF outcome in women with bilateral endometriomas and in a control group of unaffected women failed to document any detrimental impact on the chances of pregnancy (28,29). Some limitations of our study should be acknowledged. First, we did not attempt to investigate markers of oxidative stress and to identify the cells secreting ferritin within the follicular fluid. This information would have provided a more comprehensive vision of the situation. Further studies are therefore warranted to specifically elucidate these aspects. Second, we lack a histological diagnosis of endometriosis. This is a common but accepted limit of most studies evaluating the impact of ovarian endometriomas on ovarian function. This limitation is, however, of scant relevance given the high accuracy of transvaginal ultrasound. The sensitivity and specificity of this diagnostic tool have been reported to be 84–100% and 90–100%, respectively (14). Moreover, to further minimize the risk of misdiagnosis, women carrying endometriomas with atypical sonographic appearance were excluded and the presence of the endometriomas had to be documented on at least two occasions and at least two menstrual cycles apart. Third, we focused on a very particular population and we cannot exclude some confounders. Of relevance here is that we exclusively recruited women with infertility and an indication to IVF, the dimension of the ovarian endometriomas was generally small and one-third of the women had previously undergone ovarian surgery.
ª 2015 Nordic Federation of Societies of Obstetrics and Gynecology, Acta Obstetricia et Gynecologica Scandinavica 94 (2015) 646–653
651
Iron, ferritin, and endometriomas
Inferences of our findings to the whole population of women with ovarian endometriomas should therefore be made with caution. In conclusion, iron may diffuse from ovarian endometriomas into the adjacent ovarian tissue. However, this phenomenon does not seem to markedly affect ovarian function because of some effective biological mechanisms such as ferritin storage that properly counterbalance the potentially highly detrimental effects of free iron. These results tend to support conservative management of reproductive-age women with ovarian endometriomas. In other words, our data do not support the idea that removing these cysts may prevent damage to ovarian function. Further evidence is needed to definitely disentangle this issue. Studies reporting on larger cysts and studies investigating ovarian function over time in affected gonads would be particularly interesting because we cannot exclude both a role of the dimension of the endometriomas and a time-related effect.
Funding This study received only local institutional funds. References 1. Iizuka M, Igarashi M, Abe Y, Ibuki Y, Koyasu Y, Ikuma K. Chemical assay of iron in ovarian cysts: a new diagnostic method to evaluate endometriotic cysts. Gynecol Obstet Invest. 1998;46:58–60. 2. Slater M, Quagliotto G, Cooper M, Murphy CR. Endometriotic cells exhibit metaplastic change and oxidative DNA damage as well as decreased function, compared to normal endometrium. J Mol Histol. 2005;36:257–63. 3. Kobayashi H, Yamada Y, Kanayama S, Furukawa N, Noguchi T, Haruta S, et al. The role of iron in the pathogenesis of endometriosis. Gynecol Endocrinol. 2009;25:39–52. 4. Jomova K, Valko M. Importance of iron chelation in free radical-induced oxidative stress and human disease. Curr Pharm Des. 2011;17:3460–73. 5. Vercellini P, Crosignani P, Somigliana E, Vigan o P, Buggio L, Bolis G, et al. The ‘incessant menstruation’ hypothesis: a mechanistic ovarian cancer model with implications for prevention. Hum Reprod. 2011;26:2262–73. 6. Matsuzaki S, Schubert B. Oxidative stress status in normal ovarian cortex surrounding ovarian endometriosis. Fertil Steril. 2010;93:2431–2. 7. Seino T, Saito H, Kaneko T, Takahashi T, Kawachiya S, Kurachi H. Eight-hydroxy-1393 20 -deoxyguanosine in granulosa cells is correlated with the quality of oocytes and embryos in an in vitro fertilization-embryo transfer program. Fertil Steril. 2002;77:1184–90.
652
L. Benaglia et al.
8. Karuputhula NB, Chattopadhyay R, Chakravarty B, Chaudhury K. Oxidative status in granulosa cells of infertile women undergoing IVF. Syst Biol Reprod Med. 2013;59:91–8. 9. Bryan N, Ahswin H, Smart N, Bayon Y, Wohlert S, Hunt JA. Reactive oxygen species (ROS) – a family of fate deciding molecules pivotal in constructive inflammation and wound healing. Eur Cell Mater. 2012;24:249–65. 10. Sanchez AM, Vigan o P, Somigliana E, Panina-Bordignon P, Vercellini P, Candiani M. The distinguishing cellular and molecular features of the endometriotic ovarian cyst: from pathophysiology to the potential endometriomamediated damage to the ovary. Hum Reprod Update. 2014;20:217–30. 11. Sanchez AM, Papaleo E, Corti L, Santambrogio P, Levi S, Vigan o P, et al. Iron availability is increased in individual human ovarian follicles in close proximity to an endometrioma compared with distal ones. Hum Reprod. 2014;29:577–83. 12. Jana SK, K NB, Chattopadhyay R, Chakravarty B, Chaudhury K. Upper control limit of reactive oxygen species in follicular fluid beyond which viable embryo formation is not favorable. Reprod Toxicol. 2010;29:447–51. 13. Somigliana E, Paffoni A, Busnelli A, Cardellicchio L, Leonardi M, Filippi F, et al. IVF outcome in poor responders failing to produce viable embryos in the preceding cycle. Reprod Biomed Online. 2013;26:569–76. 14. Savelli L. Transvaginal sonography for the assessment of ovarian and pelvic endometriosis: how deep is our understanding? Ultrasound Obstet Gynecol. 2009;33:497– 501. 15. Practice Committee of the American Society for Reproductive Medicine. Endometriosis and infertility: a committee opinion. Fertil Steril. 2012;98:591–8. 16. Dunselman GA, Vermeulen N, Becker C, Calhaz-Jorge C, D’Hooghe T, De Bie B, et al. ESHRE guideline: management of women with endometriosis. Hum Reprod. 2014;29:400–12. 17. Ragni G, Somigliana E, Benedetti F, Paffoni A, Vegetti W, Restelli L, et al. Damage to ovarian reserve associated with laparoscopic excision of endometriomas: a quantitative rather than a qualitative injury. Am J Obstet Gynecol. 2005;193:1908–14. 18. Fasciani A, D’Ambrogio G, Bocci G, Monti M, Genazzani AR, Artini PG. High concentrations of the vascular endothelial growth factor and interleukin-8 in ovarian endometriomata. Mol Hum Reprod. 2000;6:50–4. 19. Darai E, Detchev R, Hugol D, Quang NT. Serum and cyst fluid levels of interleukin (IL) -6, IL-8 and tumour necrosis factor-alpha in women with endometriomas and benign and malignant cystic ovarian tumours. Hum Reprod. 2003;18:1681–5. 20. Velasco I, Acien P, Campos A, Acien MI, Ruiz-Macia E. Interleukin-6 and other soluble factors in peritoneal fluid
ª 2015 Nordic Federation of Societies of Obstetrics and Gynecology, Acta Obstetricia et Gynecologica Scandinavica 94 (2015) 646–653
L. Benaglia et al.
21.
22.
23.
24.
and endometriomas and their relation to pain and aromatase expression. J Reprod Immunol. 2010;84:199– 205. Opøien HK, Fedorcsak P, Polec A, Stensen MH, Abyholm T, Tanbo T. Do endometriomas induce an inflammatory reaction in nearby follicles? Hum Reprod. 2013;28:1837– 45. Van Langendonckt A, Casanas-Roux F, Donnez J. Iron overload in the peritoneal cavity of women with pelvic endometriosis. Fertil Steril. 2002;78:712–8. Kyama CM, Overbergh L, Debrock S, Valckx D, Vander Perre S, Meuleman C, et al. Increased peritoneal and endometrial gene expression of biologically relevant cytokines and growth factors during the menstrual phase in women with endometriosis. Fertil Steril. 2006;85:1667– 75. Lousse JC, Defrere S, Van Langendonckt A, Gras J, Gonzalez-Ramos R, Colette S, et al. Iron storage is significantly increased in peritoneal macrophages of endometriosis patients and correlates with iron overload in peritoneal fluid. Fertil Steril. 2009;91:1668–75.
Iron, ferritin, and endometriomas
25. Birt JA, Nabli H, Stilley JA, Windham EA, Frazier SR, Sharpe-Timms KL. Elevated peritoneal fluid TNF-a incites ovarian early growth response factor 1 expression and downstream protease mediators: a correlation with ovulatory dysfunction in endometriosis. Reprod Sci. 2013;20:514–23. 26. Almog B, Shehata F, Sheizaf B, Tan SL, Tulandi T. Effects of ovarian endometrioma on the number of oocytes retrieved for in vitro fertilization. Fertil Steril. 2011;95:525–7. 27. Benaglia L, Pasin R, Somigliana E, Vercellini P, Ragni G, Fedele L. Unoperated ovarian endometriomas and responsiveness to hyperstimulation. Hum Reprod. 2011;26:1356–61. 28. Reinblatt SL, Ishai L, Shehata F, Son WY, Tulandi T, Almog B. Effects of ovarian endometrioma on embryo quality. Fertil Steril. 2011;30:2700–2. 29. Benaglia L, Bermejo A, Somigliana E, Faulisi S, Ragni G, Fedele L, et al. In vitro fertilization outcome in women with unoperated bilateral endometriomas. Fertil Steril. 2013;99:1714–9.
ª 2015 Nordic Federation of Societies of Obstetrics and Gynecology, Acta Obstetricia et Gynecologica Scandinavica 94 (2015) 646–653
653
Copyright of Acta Obstetricia et Gynecologica Scandinavica is the property of WileyBlackwell and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.
