This observational study has a cross-sectional design in which women were recruited in the obstetrics and gynecology service within the specialized unit for the follow-up and control of women with endometriosis at Hospital Universitario La Paz (HULP, Madrid, Spain). All women between 25 and 55 years of age, without amenorrhea and associated comorbidities (cancer, mood or emotional disorders, hypertension, obesity) and harmful habits (alcohol consumption or drug abuse) who had been diagnosed with ovarian endometriosis (OE) under pharmacological treatment were invited to participate. These women were categorized according to whether they had undergone surgery (OE-S) or not (OE-NS). In addition, women who had been diagnosed with deep endometriosis (DE) and had undergone ovarian or bowel resection surgery were invited to participate. The women in the control group were recruited from regular gynecological follow-up visits, none of whom had been diagnosed with endometriosis, cancer or had undergone surgery. The recruitment period lasted from March 2023 to October 2024. Finally, a total of 40 women signed the informed consent form and were included in the study (Figure 1). As this was an exploratory study, no sample size calculation techniques were used.

The women were electronically sent a variable collection booklet consisting of sociodemographic questions (ad-hoc) including, nationality, educational level, civil status, work situation, monthly income, smoking habits and tobacco index, and following some type of diet. In addition, health and quality of life variables related to endometriosis were collected. The response to the complete booklet did not exceed 25 min. The booklet was handed out at successive clinical follow-up visits.

In addition, the women were scheduled to take a 4 ml blood sample by venipuncture. This sample was used to determine hematological parameters for clinical medicine laboratory of HULP and to test their systemically health at the time of the study.

This design has the approval of the Research Ethical Committee of HULP (PI-5435; approved on 02 December 2022).

The analysis was performed by R software version 4.4.1 (R Core Team 2022, Vienna, Austria; https://www.R-project.org/) with the RStudio interface (version 2023.06.0 + 421 for Windows; Boston, MA, USA). The packages used were rio, dplyr, compareGroups, tidyverse, ggplot2, ggpubr, patchwork, corrplot, and pheatmap.

Considering the sample size of the study, robust nonparametric techniques were used to avoid bias from atypical data. The data was summarized by sample size (n) and relative frequency (%) in the categorical variables and median and interquartile range [Q1; Q3] in the quantitative variables. The contrast by groups was performed by Fisheŕs exact test in the categorical variables and Kruskal-Walli's test in the quantitative variables. The post-hoc analysis was verified by Tukeýs Honestly significant differences (HSD) test. Effect size (Cohen's d) was calculated to estimate the magnitude (absolute value) of significant group differences (d ≥ |0.8| indicate large effect, being substantially important). These metrics were included to enhance the interpretability of the findings. The correlations between global health and quality of life variables were tested by Spearman's Rho coefficient. In addition, heatmap analysis was performed to cluster the groups by similarities in global health and quality of life using Euclidean distance between variables. This visual approach allows an intuitive understanding of the relationships among the groups based on their psychological profiles, highlighting the distinct clusters within the data. Therefore, the data were typed by min and max value. Additionally, fold-change was represented by linear regression model (LRM). To calculate the fold-change for each quality of life variable, the median of OE or DE group was subtracted to the median of control group. This difference was divided by median value of control, and then, the fold-change values were plotted. For LRM, each quality of life dependent variable was included and the endometriosis group (OE or DE) served as the independent variable. The control group was consistently used as the reference, allowing for comparisons between each endometriosis group and control (i.e., OE vs. control, DE vs. control). From LRM, the P-value (P) of the standardized coefficient was extracted to classify fold-change as significant. No missing data imputation techniques were used in this study. The P < 0.05 was considered statistically significant in all analyses.

The women had 36.0 [30.0; 44.0] years old, being Spanish the 91.9%. There were no differences between groups in the social variables. No significant differences were detected between groups in hematological and coagulation parameters (Table 1).

Physical and social functions, emotional role, body pain and global mental component did not show statical differences between groups (Figures 2A–E). Physical role was significantly lower in OE-NS (d = 1.73) and DE (d = 1.36) compared to control women; OE-S group showed no significant differences with the control group (Figure 2F). General health was significantly lower in OE-NS (d = 1.65) and DE (d = 1.89) groups compared to the control. OE-S showed higher score than DE with no significance compared to control (Figure 2G). Vitality (d = 1.25) and physical component (d = 1.53) were significantly lower in DE compared to control. Women with OE showed intermediate scores between DE and control groups (Figures 2H,I). Mental health was significantly lower in women with OE-NS compared to control (d = 1.37), with no alteration in the other groups (Figure 2J). Pain sensitivity was higher in OE-NS (d = 2.43) and DE (d = 1.18) compared to control and women with OE-S did no show differences with control (Figure 2K). The perception of social support was lower in OE compared to control, with non-differences with DE (Figure 2L).

The core and modular health profile were significantly lower in DE and OE-NS groups compared to control. OE-S did not show significant differences with DE or control groups (Figures 3A,B). Psychological well-being and support for endometriosis were significantly lower in all groups of women with endometriosis compared to control (Figures 3C,D). Similar to control, vitality, sexual and occupational functions were higher in OE-S compared to OE-NS and DE women (Figures 3E–G). Reproductive function was significantly lower in OE-NS compared to OE-S (d = 2.61) and control (d = 3.13) (Figure 3H). The menstrual characteristics of women with endometriosis were significantly lower compared to control, but did not show differences with OE women (Figure 3I). Finally, the perception of failures in life benefits was significantly lower in DE compared to control (d = 1.89), while OE groups showed no significant differences (Figure 3J).

The correlation matrices showed that OE-S had intermediate correlations patterns between control and women with DE, while women with OE-NS presented similar correlations to DE group (Figure 4). The heatmap between quality of life and health profile variables showed that women in OE-S group were close to control while OE-NS was close to women with DE (Figure 4E).

Quality of life variables showed negative fold-change in women with endometriosis compared to control. Exceptionally, women with endometriosis had higher fold-change than control in pain sensitivity and core module of health, being significantly different for pain in OE-NS and DE and in core health for OE-S. Compared to control, women with OE-NS had 15 significantly altered quality of life variables, while women with OE-S had 11. Women with DE had 16 modified variables (Figure 5).

Treatment planning should consider endometriosis subtype and severity, as different presentations may require specific therapeutic approaches. The reduced social support perception in women with endometriosis highlights the need to provide psychological resources and support groups to improve overall well-being. Interventions should incorporate regular assessments for psychological distress and referrals to suitable mental health and social support services. In addition, a structured psychological intervention plan can enhance quality of life and help women with ovarian cysts cope with physical and emotional challenges before and after surgery. Key components include: (1) preoperative support with counseling, education, and relaxation techniques to reduce anxiety; (2) cognitive-behavioral therapy to address negative thoughts, develop coping strategies, and improve problem-solving; (3) pain management training using biofeedback and relaxation techniques; (4) social and emotional support through peer groups and family counseling; and (5) postoperative recovery support with reintegration strategies, stress management, and follow-up psychological care. Given the limited resources available to address psychological distress in this population, women should receive care in a specialist center (44). Routine distress screening and timely referrals to psychological and social services can help enhance the quality of endometriosis life. In addition, given the sensitivity of discussing reproductive health, especially among women with endometriosis, women should be given the opportunity to discuss the emotional distress it causes them. Provide supportive social and health care environment for women.

While this study provides valuable insights, several limitations should be acknowledged. One of them would be the relatively small sample size. Additionally, the uneven distribution of participants across groups could introduce bias. These factors may also constrain the generalizability of the findings to broader populations. The limited and unbalanced sample was due to contextual challenge inherent to the recruitment process, particularly given the clinical nature of the population studied. Furthermore, it should be noted that surgery could also have a negative impact on quality of life, especially in the immediate after it. Although this has not been considered in this study, it would be highly relevant to explore this aspect in future research. Therefore, longitudinal studies should explore mechanisms underlying quality of life differences across endometriosis subtypes and evaluate targeted interventions and effect of surgical processes. Additionally, due to the cross-sectional and observational design of this study, causal relationships cannot be established, limiting the interpretation of associations found. Additionally, it would be important to consider recording, and therefore modulation, of fold-change for other factors, such as comorbidities, since endometriosis can coexist with other health issues that affect women's quality of life.