Exploring the link: Systemic immune-inflammation index as a marker in endometriosis-Insights from the NHANES 2001-2006 cross-sectional study

Endometriosis, a chronic ailment affecting women of reproductive age, is characterized by ectopic endometrial tissue growth in non-uterine sites, such as ovaries, the pelvic cavity, and the peritoneum [ 1 ]. This disorder not only triggers severe pain and irregular menstruation but can also lead to infertility [ 2 ]. Recent studies have posited a potential link between endometriosis and several factors, including hormonal imbalances, immune system dysfunctions, and genetic predispositions [ 3 ]. Raimondo D et al. [ 4 ] highlight central sensitization as a key factor, noting its significant prevalence among women with endometriosis. Treatment options typically encompass medications, hormone therapy, and surgical interventions. Given the disease’s recurrent and chronic nature, management often demands a tailored, long-term strategy [ 5 ]. Additionally, research has illuminated a notable association between endometriosis and mental health complications, thereby accentuating the necessity for a comprehensive treatment paradigm [ 6 , 7 ]. Enhanced understanding and effective management of endometriosis are crucial in advancing women’s health outcomes. Endometriosis, characterized as both an inflammatory and immune-related disease, presents a pro-inflammatory state within the endometrial immune environment, potentially impacting embryo implantation and disease progression [ 8 ]. The role of immune cells in the development of endometriosis, especially the influence of endocrine-immunological interactions on its progression, has become a focal point of research [ 9 ]. Recent studies emphasize the link between systemic immune-inflammatory states and endometriosis, noting a higher incidence in young women with autoimmune or inflammatory diseases [ 10 ], and an elevated allergy prevalence among endometriosis patients [ 11 ]. Immune cell counts, reflecting systemic immune and inflammatory status, appear to correlate with endometriosis risk. For example, endometriosis patients exhibit a reduction in the levels of classical and intermediate monocytes, while the levels of plasmacytoid dendritic cells and non-classical monocyte increase in the blood [ 10 ]. Additionally, lower iTreg and Treg/Th17 ratios and higher Th17 (inflammatory) levels have been observed in patients not receiving hormone therapy compared to those who are [ 12 ]. These findings underscore the importance of exploring new inflammation-based or immune cell count-related markers for endometriosis risk assessment, crucial for prevention strategies. Introduced by Hu et al. in 2014 [ 13 ], the systemic immune-inflammation index (SII) consolidates counts of neutrophils, lymphocytes, and platelets into a formula for assessing systemic immune-inflammatory status. SII has emerged as a simple, cost-effective, and reliable clinical index, useful in diagnosing and managing various diseases. Notably, higher SII values in hepatocellular carcinoma patients have been linked to poorer overall and disease-free survival [ 14 ], with similar patterns observed in other solid tumors like esophageal, ovarian, and pancreatic cancers [ 15 – 17 ]. Beyond its applications in oncology, the SII has also been validated for cardiovascular disease risk assessment [ 18 ] and in the context of autoimmune diseases and chronic inflammatory such as chronic obstructive pulmonary disease (COPD) [ 19 , 20 ]. Despite its widespread clinical utility, the relationship between SII and endometriosis remains relatively unexplored. This study aims to elucidate the correlation between SII levels and endometriosis among participants from the National Health and Nutrition Examination Survey (NHANES) in the United States, with the objective of offering valuable insights for endometriosis prevention.

Table 1 presents an overview of the initial characteristics of the 3,390 individuals participating in the study. Among these, 3,063 were not diagnosed with endometriosis, whereas 327 were. On average, participants were 35.72 years old. The average age among those diagnosed with endometriosis was 38.60 years, in contrast to 35.41 years in those without, suggesting an increased incidence in older individuals. Moreover, endometriosis prevalence was higher among married, widowed, divorced, or separated individuals. In the group with endometriosis, the BMI averaged 28.61 kg/m 2 , while it was slightly higher at 29.89 kg/m 2 in the group without endometriosis. However, with a P -value above 0.05, there was no discernible significant link between BMI and the occurrence of endometriosis. Similar trends were noted in alcohol consumption and pregnancy history. Significant associations with endometriosis were found for race, education, household income poverty ratio, and age of menarche. More specifically, non-Hispanic whites, individuals with at least a high school education, those with higher household income poverty ratios, and those with a lower mean age of menarche, were more prevalent among endometriosis cases. Blood analysis showed higher mean neutrophil and platelet counts, but lower lymphocyte counts in endometriosis patients, with a statistically significant P -value of less than 0.05. A multifactorial logistic regression approach was employed to investigate the association between SII and the incidence of endometriosis. As presented in Table 2 , the stepwise construction of three models established a statistically significant positive correlation between SII levels and endometriosis prevalence in Model 3. Specifically, for each unit increase in SII within the uppermost quartile (Q4), there was a 2.14% rise in the probability of having endometriosis relative to the bottom quartile (Q1). The regression curve in Fig 2 , demonstrating an "S" shape, revealed a positive correlation between increased SII levels and endometriosis risk. Through threshold effect analysis, a critical point an SII value of 1105.76 was identified; beneath this benchmark ( K 1105.76), a significant positive correlation was noted. Note: In this study, endometriosis was examined as the dependent outcome, with the Systemic Immune-Inflammation Index (SII) concentration being analyzed as an independent factor. To investigate the relationship between SII and endometriosis, multivariate logistic regression analyses were conducted, considering a range of potential confounding factors. Model 1, the initial model, did not include adjustments for any potential confounders. Model 2 adjusted for variables such as participant age, marital status, poverty-to-income ratio, and race. Model 3 further refined these adjustments to include all covariates, offering a comprehensive analysis. Subgroup analyses and interaction tests were conducted to assess the consistency of the association between SII and endometriosis across various demographic characteristics. Stratification was based on age, race, education, marital status, poverty income ratio, alcohol consumption, BMI, age of menarche, and pregnancy history. As shown in Fig 3 , there were no significant interactions between SII and marital status ( P = 0.3935), poverty income ratio ( P = 0.1780), BMI ( P = 0.1233), alcohol consumption ( P = 0.9783), and age of menarche ( P = 0.5569). However, age ( P < 0.001), race ( P = 0.0067), educational level ( P = 0.0108), and pregnancy history ( P = 0.0125) significantly influenced the relationship between SII and endometriosis.

The results of this research indicate a significant link between heightened SII levels and an elevated risk of endometriosis, highlighting the potential role of SII as a predictive biomarker for assessing the risk of developing endometriosis. However, to validate these preliminary observations, further research involving extensive, prospective studies is imperative.

The data utilized in this study were sourced from the NHANES database, an extensive and nationally representative survey initiative launched by the National Center for Health Statistics (NCHS) in 1999. NHANES gathers a wide array of health-related information using methods including questionnaire-based interviews, physical examinations, and laboratory tests. It employs a sophisticated multistage probability sampling design to amass diverse health-related data. Participants are interviewed at their homes and subsequently undergo physical examinations and laboratory evaluations at a Mobile Examination Center (MEC), making NHANES a crucial resource for assessing the health and nutritional status of the U.S. population and supporting epidemiological and health science research. Ethical approval for NHANES research protocols was granted by the Research Ethics Review Board of the National Center for Health Statistics (NCHS), under Protocol #2005–06, securing written informed consent from every participant. For this study, all analyses were meticulously conducted in accordance with NHANES guidelines and regulations. Through an exhaustive search and screening of the NHANES database from 2001 to 2006, we selected female participants aged 20–55 years, excluding males and individuals with incomplete information on ’self-reported endometriosis’, ’SII’, and ’covariates’, resulting in a final cohort of 3,390 participants. Fig 1 illustrates the screening process flowchart. In this study, data on endometriosis were obtained from the Reproductive Health Questionnaire (RHQ360) and the Reproductive Health Diagnosis (RHD361) self-report sections of the NHANES, where participants were queried on whether they had been diagnosed with endometriosis by a healthcare professional. Individuals confirming such a diagnosis were classified into a group representing self-identified cases of endometriosis, whereas those who negated the diagnosis formed the control group, indicating an absence of self-reported endometriosis. The primary independent variable, the SII, is a novel inflammation marker calculated by multiplying platelet and neutrophil counts and then dividing by the lymphocyte count. To derive the SII, data from NHANES’s "L25" and "Complete Blood Count (CBC)" segments were utilized, specifically employing lymphocyte (LBDLYMNO), neutrophil (LBDNENO), and platelet (LBXPLTSI) counts. These values were then used to calculate the SII for each participant, which was further categorized into four quartiles (Q1 to Q4) for analysis. Variables including age, ethnic background, marital status, educational attainment, and economic status, determined via interviews in households, were incorporated into the study’s analysis. Age was stratified into three groups: 20 to 29 years (Group 1), 30 to 39 years (Group 2), and 40 to 55 years (Group 3). Marital status categories were married, divorced, widowed, separated, never married, or living with a partner. Educational attainment was classified as either below high school or high school and above. Economic status were gauged using the poverty income ratio (PIR), with classifications being low (PIR 3.0). Alcohol consumption was classified based on annual drink count: non-drinker (0 drinks), light drinker (1–30 drinks), or heavy drinker (>30 drinks). Disease-related covariates included age of menarche and pregnancy history, sourced from RHQ010 and RHQ031. Body mass index (BMI) classifications were established as underweight (BMI 30 kg/m 2 ). The baseline characteristics of participants were detailed utilizing descriptive statistical methods: mean values and standard deviations were used for continuous variables, and categorical variables were described using percentages. The relationship between SII and endometriosis was analyzed using multifactorial logistic regression. Three models were employed in the analysis: Model 1, which was unadjusted, Model 2, which took into account adjustments for age, ethnicity, educational background, and poverty-income ratios, and Model 3, which included comprehensive adjustments for all covariates. Weighted one-way logistic regression assessed the impact of covariates on the SII-endometriosis association. Statistical analyses were performed with R software, version 4.3.2 (accessible at http://www.R-project.org ). In these analyses, a two-tailed P -value below 0.05 was considered statistically significant.

(XLS)