Impact of endometrioma on iron levels and oxidative stress in the follicular fluid in women with endometriosis: a cross-sectional study

This is a cross-sectional analytical study among women who were undergoing treatment for infertility in the Department of Reproductive Medicine and Surgery at Christian Medical College, Vellore, during the period July 2022 to July 2023. The study was approved by the Institutional Review Board (IRB Min No.14504 dated 23.02.2022) and conformed to the principles of the Declaration of Helsinki 1975 and subsequent amendments. Informed consent was obtained from all the participants. We recruited infertile women with laparoscopically diagnosed endometriosis with ( n  = 14) or without endometrioma ( n  = 7). Among those with endometrioma, 8 had unilateral endometrioma, while 6 had bilateral endometrioma. As controls ( n  = 14), we included women with infertility due to tubal factors, male factors, or those with unexplained infertility. We excluded those who had known causes of infertility, such as severe pelvic adhesions, polycystic ovarian syndrome (PCOS), untreated thyroid disorders (defined as TSH  4.78 µIU/ml) and uncontrolled diabetes mellitus (defined as HbA1c > 8%). Follicular fluid (FF) was collected during oocyte retrieval for ART (~ 5–6 mL). The tubes were immediately centrifuged at 800 g for 10 min and the supernatant was stored at −70 °C. Depending on the study group, when we were comparing follicular fluid from women with unilateral endometrioma (affected and not affected), we aspirated from both ovaries. For women with bilateral endometrioma also, we aspirated one mature follicle from both the ovaries and same for control to maintain uniformity. Each follicle aspiration yielded approximately 2–3 ml of fluid with two yielding approximately 5–6 ml. Iron was estimated using a ferrozine-based colorimetric assay on Roche Cobas c702 auto-analyzer (Roche Diagnostics), while transferrin (Roche Cobas c502 using immunoturbidimetric method) and interleukin-6 was estimated on Roche Cobas e801using electrochemiluminescence-based immunoassay in the clinical biochemistry laboratory in the institution. Malondialdehyde (MDA) is a well-established and widely utilized marker of lipid peroxidation 41 , 42 . Measuring MDA in follicular fluid directly assesses the oxidative environment that developing oocytes are exposed to. This measurement serves as a localized and relevant biomarker, reflecting the cumulative oxidative damage occurring within the affected ovary. MDA was estimated in the FF using High-Performance Liquid Chromatography (HPLC) as previously described 43 . Briefly, 100 µL of FF was mixed with 400 µL of 0.1% O-phosphoric acid and 100 µL of 0.6% 2-thiobarbituric acid. The mixture was incubated at 90 °C for 30 min, after which the reaction was terminated by placing the tubes on ice. Subsequently, 10 µL of the reaction mixture was injected into the HPLC system for analysis. The HPLC system used was a Shimadzu platform comprising an LC-20AD pump, CTO-20 A column oven, SIL-20AC autosampler, and RF-20 A fluorescence detector. Data acquisition and peak integration were performed using LabSolutions software. Chromatographic separation was achieved using a LiChrospher ® RP-18 column (5 μm particle size, 15 cm × 3.2 mm I.D.). The mobile phase consisted of 50 mM potassium dihydrogen phosphate (pH adjusted to 6.8 with 5.0 mM KOH) and methanol, mixed in a 60:40 (v/v) ratio. The mobile phase was filtered through a 0.22 μm Millipore membrane filter under vacuum. The flow rate was maintained at 1.0 mL/min. The fluorescence detector was set at an excitation wavelength (λex) of 527 nm and an emission wavelength (λem) of 551 nm. Follicular fluid samples are processed to extract non-transferrin-bound iron using an iron chelator nitrilotriacetic acid (NTA) 44 . Briefly, 180 µL of follicular fluid samples were incubated with 40 µL of tris-carbonatocobaltate (III) to block the unoccupied sites of transferrin and then treated with 24.44 µL of 800 mM nitrilotriacetic acid (NTA), which is a weak Ligand that selectively binds with non-transferrin bound iron to form Fe-NTA complex. The solution was ultrafiltered with 30 kDa filters at 14,000 g for 60 min. The filtered solution containing the Fe-NTA complex was then analyzed by inductively coupled mass spectrometry (ICP-MS) to quantify the iron present in the sample to determine the NTBI levels. Statistical analyses were carried out using the SPSS software version 21. The Kolmogorov-Smirnov test was done to look for normality of the distribution of data prior to statistical analysis. Data distribution was skewed in the majority of cases. Therefore, appropriate non-parametric tests were chosen for statistical analysis. Continuous data were represented as median and interquartile range. Wilcoxson sign-rank test was used for paired data. Kruskal Wallis test and Tukey’s post-hoc test (with Bonferroni correction) was used for comparison of multiple groups. Spearman correlation was used to look for a correlation between parameters of oxidative stress and iron/inflammation and the Chi-square test for discrete variables. A p value of < 0.05 was considered statistically significant in all cases. We did not exclude any data as outliers in the analysis.

We recruited 21 women with endometriosis and 14 controls. Among those with endometriosis, 8 had unilateral endometrioma, 6 had bilateral endometrioma and 7 had endometriosis without endometrioma. The clinical characteristics are shown in Table  1 . Table 1 Clinical characteristics of the participants. Parameters Controls ( n  = 14) 1 Endometriosis ( n  = 21) 1 P value 2 Participant Age (years) 33 (32, 36) 29 (28, 34) 0.06 BMI (kg/m 2 ) 26.4 (22.7, 30.4) 25.8 (23.9, 27.4) 0.80 Duration of infertility (years) 8.0 (6.5, 13.0) 7.0 (5.0, 10.0) 0.14 Primary infertility 11 (73%) 16 (76%) > 0.9 Secondary infertility 4 (27%) 5 (24%) > 0.9 Number of antral follicles 10 (5, 12.5) 8 (7, 12) 0.80 Number of follicles aspirated 10 (7, 13) 9 (7, 12) > 0.9 Number of retrieved oocytes 9 (5, 12) 8 (6, 12) > 0.9 Number of good-quality embryos 5 (3.5, 6) 3 (2, 8) 0.66 1 Median (IQR); n (%). 2 Mann-Whitney U test. Clinical characteristics of the participants. 1 Median (IQR); n (%). 2 Mann-Whitney U test. Fig. 1 Oxidative stress, iron and inflammatory markers in the follicular fluid (FF) of women with endometriosis and controls. MDA (oxidative stress marker) ( A ), iron ( B ), transferrin ( C ), transferrin saturation ( D ), interleukin-6 ( E ) in endometriosis ( n  = 21) and controls ( n  = 14) and NTBI ( F ) levels in endometriosis ( n  = 15) and controls ( n  = 12). Data displayed as box and whisker plots and analysed using Mann-Whitney U test. P  < 0.05 was considered statistically significant in all cases. Five outliers are not shown in E. Oxidative stress, iron and inflammatory markers in the follicular fluid (FF) of women with endometriosis and controls. MDA (oxidative stress marker) ( A ), iron ( B ), transferrin ( C ), transferrin saturation ( D ), interleukin-6 ( E ) in endometriosis ( n  = 21) and controls ( n  = 14) and NTBI ( F ) levels in endometriosis ( n  = 15) and controls ( n  = 12). Data displayed as box and whisker plots and analysed using Mann-Whitney U test. P  < 0.05 was considered statistically significant in all cases. Five outliers are not shown in E. There were significantly higher levels of MDA in the endometriosis group ( n  = 21) than in the control group ( n  = 14) [Fig.  1 A]. However, no significant differences were noted in follicular fluid iron levels [Fig.  1 B], transferrin [Fig.  1 C], transferrin saturation [Fig.  1 D], the inflammatory marker (IL-6) [Fig.  1 E], and NTBI [Fig.  1 F] in the 2 groups. Fig. 2 Oxidative stress, iron and inflammatory markers in the FF (FF) of women with and without endometrioma compared to controls. MDA (oxidative stress marker) ( A ), iron ( B ), transferrin ( C ), transferrin saturation ( D ) and interleukin-6 ( E ) levels in women with unilateral/bilateral endometrioma ( n  = 14), endometriosis without endometrioma ( n  = 7) and controls ( n  = 14). Data displayed as box and whisker plots. Kruskal-Wallis statistical test was used for overall significance. Pair-wise comparisons were done using Mann-Whitney U test. P  < 0.05 was considered statistically significant in all cases. Oxidative stress, iron and inflammatory markers in the FF (FF) of women with and without endometrioma compared to controls. MDA (oxidative stress marker) ( A ), iron ( B ), transferrin ( C ), transferrin saturation ( D ) and interleukin-6 ( E ) levels in women with unilateral/bilateral endometrioma ( n  = 14), endometriosis without endometrioma ( n  = 7) and controls ( n  = 14). Data displayed as box and whisker plots. Kruskal-Wallis statistical test was used for overall significance. Pair-wise comparisons were done using Mann-Whitney U test. P  < 0.05 was considered statistically significant in all cases. Compared to controls, MDA levels were elevated in FF in those with endometriosis, whether or not an endometrioma was present [Fig.  2 A]. Interestingly, MDA levels were higher in women without endometrioma ( n  = 7) compared to those with endometrioma (unilateral or bilateral) ( n  = 14) [Fig.  2 A]. Iron, transferrin, and transferrin saturation were not significantly different between the 3 groups [Fig.  2 B-D]. Interleukin-6 levels were significantly higher in the FF of women with endometrioma when compared to those without endometrioma and controls ( n  = 14) [Fig.  2 E]. Fig. 3 Oxidative stress, iron and inflammatory markers in the FF obtained from affected and unaffected ovaries of women with unilateral ovarian endometrioma and controls. MDA (oxidative stress marker) ( A ), iron ( B ), transferrin ( C ), transferrin saturation ( D ) and interleukin-6 ( E ) levels in affected and unaffected ovaries of women with unilateral endometrioma ( n  = 8) and controls ( n  = 14). Data displayed as box and whisker plots. Kruskal-Wallis statistical test was used for overall significance. Pair-wise comparisons were done using Mann-Whitney U test. P  < 0.05 was considered statistically significant in all cases. Oxidative stress, iron and inflammatory markers in the FF obtained from affected and unaffected ovaries of women with unilateral ovarian endometrioma and controls. MDA (oxidative stress marker) ( A ), iron ( B ), transferrin ( C ), transferrin saturation ( D ) and interleukin-6 ( E ) levels in affected and unaffected ovaries of women with unilateral endometrioma ( n  = 8) and controls ( n  = 14). Data displayed as box and whisker plots. Kruskal-Wallis statistical test was used for overall significance. Pair-wise comparisons were done using Mann-Whitney U test. P  < 0.05 was considered statistically significant in all cases. In women with unilateral endometrioma ( n  = 8), MDA levels in the FF from both ovaries (irrespective of whether endometrioma was present or not) were significantly higher than those in controls ( n  = 14). Interestingly, MDA levels in the ovary with endometrioma were not significantly different from those in the contralateral ovary without endometrioma [Fig.  3 A]. In fact, MDA levels in the FF of ovaries from women with non-ovarian endometriosis were significantly higher than those of women with ovarian endometriomas ( p  = 0.038) [Fig.  2 A]. The presence of endometrioma was also not associated with higher iron levels, transferrin saturation, or IL-6 levels [Fig.  3 B-E]. Fig. 4 Oxidative stress, iron and inflammatory markers in the FF obtained from women with bilateral endometrioma ( n  = 6), endometriosis without endometrioma ( n  = 7) and control ( n  = 14). MDA (oxidative stress marker) ( A ), iron ( B ), transferrin ( C ), transferrin saturation ( D ) and interleukin-6 ( E ) levels in women with bilateral endometrioma ( n  = 6), endometriosis without endometrioma ( n  = 7) and controls ( n  = 14). Data displayed as box and whisker plots. Kruskal-Wallis statistical test was used for overall significance. Pair-wise comparisons were done using Mann-Whitney U test. P  < 0.05 was considered statistically significant in all cases. Oxidative stress, iron and inflammatory markers in the FF obtained from women with bilateral endometrioma ( n  = 6), endometriosis without endometrioma ( n  = 7) and control ( n  = 14). MDA (oxidative stress marker) ( A ), iron ( B ), transferrin ( C ), transferrin saturation ( D ) and interleukin-6 ( E ) levels in women with bilateral endometrioma ( n  = 6), endometriosis without endometrioma ( n  = 7) and controls ( n  = 14). Data displayed as box and whisker plots. Kruskal-Wallis statistical test was used for overall significance. Pair-wise comparisons were done using Mann-Whitney U test. P  < 0.05 was considered statistically significant in all cases. Similarly, in women with bilateral endometrioma ( n  = 6), MDA levels were higher compared to controls but not significantly different from those with non-ovarian endometriosis (those without endometrioma) ( n  = 7) [Fig.  4 A]. Again, no significant differences were noted in the levels of iron-related parameters [Fig.  4 B-D]. However, IL-6 levels tended to be higher in the bilateral endometrioma group when compared to the non-ovarian endometriosis and control groups; however, this was not statistically significant [Fig.  4 E] (Kruskal-Wallis p  = 0.079). To further explore the association between oxidative stress, iron, and inflammation, we did a correlation analysis between these parameters (Table  2 ). No significant correlations were noted between MDA and iron-related parameters or IL-6 (Table  2 ). Table 2 Bivariate correlation analysis for MDA. Total ( n  = 35) Controls ( n  = 14) Endometriosis ( n  = 21) MDA Vs. r P r P r P Iron 0.140 0.421 −0.029 0.923 0.263 0.249 Transferrin −0.276 0.109 −0.273 0.345 −0.225 0.328 Transferrin saturation 0.252 0.144 0.187 0.523 0.364 0.321 Interleukin-6 0.594 0.957 0.271 0.349 −0.439 0.156 1 Spearman correlation coefficient. 2 ‘p’ value of the correlation. Bivariate correlation analysis for MDA. 1 Spearman correlation coefficient. 2 ‘p’ value of the correlation.

In the current study, we found higher levels of the oxidative stress marker, MDA, in the FF of the endometriosis group when compared to the controls. The MDA levels were significantly higher in women with non-ovarian endometriosis compared to those women with ovarian endometrioma. Furthermore, the MDA levels were not significantly higher in the ovary with endometrioma versus the contralateral ovary without endometrioma. On the other hand, the inflammatory marker interleukin-6 was found to be higher only in those women with endometrioma compared to those without endometrioma and controls. We did not find any association between iron levels in the FF and the presence/absence of endometrioma. Similar to the results of our study, two other studies have also observed higher MDA levels in FF in the endometriosis group when compared to controls 16 , 21 . Another study has found higher levels of 8-hydroxy-2’-deoxyguanosine (a marker of oxidative DNA damage) in the FF of women with endometriosis 8 . Additionally, elevated MDA in the serum samples of women with endometriosis has also been reported in one of the two studies 16 . Taken together, elevated oxidative stress in the FF seems to be a characteristic feature of endometriosis, and this was reinforced by the findings in the current study. Some earlier studies have reported elevated iron levels in the FF of women with endometriosis 17 , 21 , 33 . In the present study, iron levels, as well as transferrin saturation, were similar in those with endometriosis and controls. One possible reason for this may be related to the method used to estimate iron. Singh et al., 2013 have used atomic absorption spectrophotometry 21 . This method will measure both heme and non-heme iron. Therefore, contamination of FF with blood during FF aspiration may have resulted in elevated iron levels due to the presence of hemoglobin. In our study, we used a colorimetric kit-based assay, which will only estimate non-heme iron and not heme iron present in hemoglobin. Supporting this, Benaglia et al., 2015 also used a colorimetric method and did not find significant differences between the iron levels in affected and unaffected ovaries in women with unilateral endometrioma 20 . On the other hand, Li et al., 2020 17 and Ni et al., 2022 33 using similar kit-based colorimetric assays found elevated iron levels in FF in those with endometriosis when compared to controls. Overall, it is unclear whether iron levels are elevated or not in FF in endometriosis, with different studies giving conflicting results. Further work with larger sample sizes is required to clarify this. Supporting our finding of the absence of elevated iron in FF of women with endometriosis is the fact that NTBI levels were also not elevated in these women. NTBI represents the redox active form of iron that is not bound to transferrin, and therefore catalyses the formation of highly reactive hydroxyl free radicals 34 . Multiple studies have shown that oxidative tissue damage associated with iron is mediated by excess levels of NTBI 35 , 36 . The absence of elevated NTBI in the FF of women with endometriosis supports our conclusion that iron does not play a major role in the induction of oxidative stress. To the best of our knowledge, this is the first study that has estimated NTBI in FF in the context of endometriosis. It has been postulated that there exists a link between inflammation and endometriosis-associated infertility 37 . Interleukin-6 (IL-6) was raised in the peritoneal fluid of women with endometriosis when compared to women with non-endometriotic benign ovarian cysts or leiomyomas 38 . To the best of our knowledge, our study is the first to compare the levels of IL-6 in the FF of women with endometriosis with or without endometrioma. We found significantly elevated IL-6 levels in women with unilateral/bilateral endometrioma when compared to those without endometrioma and controls. This agrees with the literature that there could be a link between inflammation and endometriosis-related infertility, at least in the context of ovarian endometrioma 39 , 40 . The presence of endometrioma is presumed to adversely affect growing follicles by altering their microenvironment 20 . Although we found elevated levels of MDA, a marker of lipid peroxidation, in FF, suggesting the presence of oxidative stress in women with endometriosis, the presence of endometrioma per se neither increased local iron levels nor exacerbated oxidative stress [Fig. 2 ]. MDA levels were similar in the affected and unaffected ovaries in those with unilateral endometrioma [Fig. 3 A]. These findings suggest that the presence of endometriosis, and not ovarian endometrioma per se, increases oxidative stress in the FF. Therefore, the systemic effect of endometriosis, and not the local effect of endometrioma, may underlie oxidative stress. Considering that FF is an ultrafiltrate of plasma, a pertinent question in this context is whether MDA levels were elevated in the blood in women with endometriosis. Unfortunately, we do not have serum samples to estimate MDA, and this is a limitation of this study. Not all studies have shown consistent results regarding systemic oxidative stress 15 , 16 . One study, for example, found no significant difference in serum malondialdehyde (MDA) levels between women with endometriosis and age-matched controls 15 . This suggests that oxidative stress in endometriosis may be more pronounced locally in the follicular environment rather than systemically. Current clinical guidelines for managing ovarian endometriomas in women undergoing ART recommend surgical excision for cysts exceeding 3 cm 32 . This recommendation is largely based on concerns about the impact of endometrioma on accessibility, procedural risk during oocyte retrieval, and possibly the oocyte quality. However, the exact impact of surgical excision on oocyte quality remains a topic of significant debate, with existing literature presenting mixed evidence 32 . Our study offers a nuanced perspective that challenges the assumption that routine excision of endometrioma is necessary to improve oocyte quality during IVF. While we observed higher levels of interleukin-6, indicating localized inflammation in women with endometriomas, this inflammatory response did not correlate with an increase in oxidative stress within the ovarian microenvironment (Table 2 ). This finding is important because both inflammation and oxidative stress are well-recognized detrimental factors for oocyte development and competence 12 . If endometriomas do not significantly contribute to these negative microenvironmental conditions, the justification for their excision to improve oocyte quality becomes less compelling. Consequently, our findings suggest that the primary justifications for surgical excision of endometriomas before ART should be re-evaluated and potentially narrowed. Avoiding unnecessary surgery can help preserve ovarian reserve and reduce risks, especially for women with endometriosis undergoing ART treatment. The strength of our study lies in our classification of women with endometriosis into those with endometriosis and those without, and our systematic assessment of oxidative stress and iron markers in these groups. We employed the ferrozine-based colorimetric method to estimate iron levels, which specifically measures non-heme iron. Since this assay does not detect iron bound within hemoglobin or other heme-containing compounds, it reduces the likelihood of overestimation due to blood contamination of FF during oocyte retrieval. This provides a more reliable assessment of the iron levels in the follicular fluid. Third, we also estimated NTBI levels in FF. To the best of our knowledge, this is the first study to have done so. Nevertheless, our study has limitations. The sample size is low; a study with a larger sample size will be required to make definitive conclusions. Second, as mentioned above, our study does not have data on the serum levels of iron, transferrin, IL-6, and MDA to correlate with the FF levels. Third, confounding effects may arise from the impact of controlled ovarian stimulation, as the only opportunity to obtain follicular fluid is during IVF treatment. While our study highlights the presence of oxidative stress in FF in patients with endometriosis, we have not explored specific pathways that may play a role in its etiopathogenesis. Future studies should adopt a longitudinal design that includes parallel measurements of serum and follicular fluid biomarkers to understand better systemic versus local effects of endometriosis. In conclusion, we found that oxidative stress in follicular fluid is significantly elevated in women with endometriosis, irrespective of the presence of endometriomas. Notably, endometriomas did not further increase iron levels or exacerbate oxidative stress locally. Therefore, we suggest that the decision to surgically remove an endometrioma before ART should be based on clinical considerations, such as addressing technical challenges during oocyte retrieval.

Endometriosis is a common gynecological disorder that affects 5–10% of women of reproductive age across the world 1 . Endometriosis is frequently associated with infertility and has a prevalence of 25–40% among infertile women 2 , 3 . Although various mechanisms have been proposed, it is still unclear how endometriosis affects fertility 4 . One of the suggested mechanisms is the exposure of oocytes to a hostile pelvic environment, which adversely impacts oocyte quality, leading to a reduction in the implantation potential of the embryo 2 , 5 , 6 . Increased oxidative stress is considered one of the hallmarks of this altered follicular environment and has been linked to the pathogenesis of infertility 5 , 7 – 10 . The follicular fluid (FF) creates a suitable microenvironment for the developing oocyte 11 . Studies suggest that the composition and content of the FF is altered in women with endometriosis, which in turn affects the oocyte quality 12 – 14 . A few studies have also documented increased reactive oxygen species (ROS) and decreased antioxidants in the FF of women with endometriosis 15 , 16 . In addition, the oocyte maturation rate was lower after exposure of mouse oocytes to FF derived from women with endometriosis 17 . Excess iron acts as a prooxidant, leading to oxidative stress by generating reactive oxygen species (ROS) through Fenton reactions 18 . In endometriosis, the breakdown of retrograde menstrual blood increases iron levels in peritoneal fluid 19 . Additionally, iron from ovarian endometriomas, originating from degraded blood, may contribute to higher iron levels in follicular fluid (FF) 20 . Studies have found that ferritin levels are elevated in the FF of ovaries with endometriomas compared to those without, with increased ROS and iron levels in the FF of women with endometriosis versus those with tubal infertility 21 . Both iron and oxidative stress are linked to chronic inflammatory processes in endometriosis 5 , 22 . Iron overload activates the pro-inflammatory transcription factor, nuclear factor kappa B (NF-κB), which, in turn, increases the synthesis of inflammatory cytokines 23 – 27 . Increased levels of inflammatory cytokines, such as interleukin- 6 (IL-6) and tumor necrosis factor- α (TNF-α), were observed in the FF of women with endometriosis 28 . Inflammation can lead to fibrosis within the ovaries and consequently negatively affect the ovarian reserve in these women with endometriosis 29 . Among women with endometriosis, there is ambiguity about whether the presence of ovarian endometrioma is an independent factor that affects the oocyte quality and thereby the ART treatment outcomes 30 . However, the surgical removal of these cysts can negatively affect ovarian reserve and result in lower success following in vitro fertilization (IVF) 31 . Hence, the current guidelines recommend proceeding with IVF without removal of the endometrial cyst when the size of the cyst is less than 3 cm 32 . The current study was planned to explore associations between iron content and markers of oxidative stress and inflammation in the FF of infertile women with endometriosis. We hypothesized that the presence of endometrioma increases iron levels and oxidative stress in the follicular fluid, thereby impacting oocyte quality.