Endometriosis and hypertriglyceridemia: Why do we care about severity and typology?

The present study utilized data from the Endometriosis, Natural History, Diagnosis and Outcomes (ENDO) study (2007–2009). 2 The ENDO study was designed to 1) examine the scope and magnitude of endometriosis by diagnostic method and choice of comparison group; and 2) investigate the relationship between endocrine disrupting chemicals (EDCs) and occurrence of gynecologic pathology, including endometriosis. Eligibility criteria for enrollment in the ENDO study includes: 18–44 years of age, currently menstruating, no breast feeding ≥ 6 months, no cancer history other than nonmelanoma skin cancer, and no injectable hormonal treatment within the past 2 years. Participants included 473 premenopausal women undergoing diagnostic or therapeutic laparoscopy or laparotomy regardless of clinical indication (pelvic pain, pelvic mass, menstrual irregularities, suspected fibroids, tubal ligation, and infertility) from clinical centers in Salt Lake City, UT and San Francisco, CA. The University of Utah Institutional Review Board (IRB) approved the study (IRB #00021614), and all participants gave written informed consent before enrollment and data collection. Standardized data collection included a baseline personal interview, collection of biospecimens (blood, urine, peritoneal fluid, omental fat, and endometriosis implants when available), and in-person anthropometric assessment. 2 The baseline questionnaire, anthropometric assessment, and specimen collection occurred before surgery. Socio-demographics, reproductive and medical history, and lifestyle information were self-reported. For the anthropometric assessment, height was measured with a portable stadiometer and weight with electronic scales. Body Mass Index (BMI) was calculated from height and weight measurements (weight [kg] / height [m2]). BMI was categorized according to the Centers for Disease Control and Prevention classification <18.5 kg/m2; 8.5–24.99 kg/m2; 25.0–29.99 kg/m2; and ≥ 30.0 kg/m2). Serum cotinine (ng/mL) was measured and used to evaluate smoking. Hypertriglyceridemia was defined as non-fasting triglyceride concentration level ≥175mg/dL, as this has been determined in prior research to be the optimal non-fasting cut-point for mid-life women. 16 A sensitivity analysis was conducted using the standard cut-point of ≥150 mg/dL. 17 The exposure of interest was visualized endometriosis diagnosis (yes/no) along with endometriosis stage and typology. For endometriosis, all surgeons participating in the ENDO study were trained in the diagnosis and staging of endometriosis. The surgeons completed a standardized operative report immediately after surgery to capture the degree of endometriosis and gynecologic and pelvic pathology. 2 , 18 Endometriosis staging was categorized using the revised American Society for Reproductive Medicine (rASRM) disease stage. The rASRM uses a weighted point score to categorize endometriosis staging: stage I, minimal (scores 1–5); stage II, mild (scores 6–15); stage III, moderate (scores 16–40); and stage IV, severe (scores >40). 19 Endometriosis typology was assessed using the rASRM form. Women with only superficial lesions on the ovary or peritoneum were considered to have superficial endometriosis (SE), women with deep lesions (>5mm invasion) noted in the peritoneum or with obliteration of the posterior cul-de-sac were considered to have deep infiltrating endometriosis (DE), and women with deep lesion of any size noted in the ovary were considered to have ovarian endometrioma (OE). Women who had deep ovarian and peritoneal lesions were considered to have OE and DE. 20 , 21 Due to sample size limitations, endometriosis staging was reclassified into minimal to mild (stage I/II) and moderate to severe (stage III/IV) endometriosis. Typology was also classified into SE only, OE or DE, and OE and DE. “No endometriosis” served as the comparison for diagnosis, staging, and typology. Women with “No endometriosis” is also defined as women with no endometriosis but with other gynecological conditions and women with no endometriosis with normal pelvis. Confounders were determined via literature review and the use of directed acyclic graphs. 22 Our primary models adjusted for baseline age (continuous), race and ethnicity (Hispanic vs non-Hispanic; non-white vs white), marital status (not-married vs married), BMI (continuous, kg/m2), income (Below poverty Line, ≤180% of poverty Line, vs > 180% of poverty line) and serum cotinine (continuous, ng/ml). Additional covariates considered included level of physical activity (low, moderate, high), history of alcohol consumption (yes, no), and average daily caffeine consumption (continuous). We also include models adjusted for self-reported parity, prior to laparoscopy/laparotomy. While endometriosis impacts infertility and consequently parity, parity is known to impact endometriosis. We used modified Poisson regression with robust standard errors to calculate adjusted prevalence ratios (aPRs) and 95% CIs for the association between endometriosis diagnosis, staging, and typology and hypertriglyceridemia. 23 , 24 Given that triglyceride levels have been shown to vary by stage of menstrual cycle (follicular and luteal), 25 stratified results were presented. Geometric mean concentrations of triglycerides by endometriosis diagnosis, stage, and typology were generated overall and by menstrual cycle phase. We formally assessed whether the association between endometriosis and hypertriglyceridemia may be modified by menstrual cycle phase via the Wald test. We assessed potential non-linearity of triglycerides in relation to endometriosis using restricted cubic splines, adjusting for age, race/ethnicity, marital, categorical BMI, and log serum cotinine variables. Tests for non-linearity used the likelihood ratio test, comparing the adjusted model with only the linear term to the adjusted model with the linear and the cubic spline terms. Further, we conducted an additional analysis using the standard hypertriglyceridemia cutpoint of ≥150 mg/dL vs less for comparability with past and future studies. 17 Finally, we considered alternative models adjusting for additional potential confounders including physical activity, caffeine and alcohol consumption, and parity. Study participants were remunerated for their time and travel. Full human subjects approval was obtained for the conduct of this study; each of the women provided informed consent before any data collection. 2

Among 473 women in the study, 108 (22.8%) had hypertriglyceridemia (≥175mg/dL) and 190 (40.2%) had endometriosis. Women with endometriosis were more likely to be younger, >180% of the poverty line, and have a lower BMI compared to women without endometriosis ( Table 1 ). Women with hypertriglyceridemia had a higher BMI compared to those without hypertriglyceridemia. Further, women with moderate to severe endometriosis had a higher prevalence of hypertriglyceridemia (35.2%) than women with minimal to mild endometriosis (18.7%) and those with no endometriosis diagnosis (22.6%). Differences were also observed by endometriosis subtype, and women with combined OE and DE endometriosis had the highest prevalence of hypertriglyceridemia (50.0%) ( Table 1 ). In addition, women with combined OE and DE endometriosis had higher mean (169 ± 81 mg/dl) and median [25%, 75%] (179 [99, 215] mg/dl) triglycerides compared to women with no endometriosis (mean=147 ± 101; median [25%, 75%] 121 [90, 166] mg/dl). Overall, women with diagnosed endometriosis had a 1.24 (95% CI 0.87, 1.77) higher prevalence of hypertriglyceridemia after adjusting for age at baseline, race/ethnicity, marital status, BMI, income, and serum cotinine compared to women without endometriosis ( Table 2 ). This pattern varied by endometriosis severity and typology. Compared to women with no endometriosis, women with moderate to severe endometriosis had a 1.74 (95% CI 1.03, 2.95) higher adjusted prevalence of hypertriglyceridemia; while there was no association observed with minimal to mild (Stage I/II) endometriosis (aPR: 1.06, 95% CI: 0.70, 1.59). When we investigated endometriosis typology, we observed that women with combined OE and DE experienced a 3.59 (95% CI 2.33, 5.54) higher adjusted prevalence of hypertriglyceridemia compared to those without endometriosis. On average, among everyone in the cohort, there was little difference in hypertriglyceridemia prevalence between women assessed in the follicular phase (22.0%) versus the luteal phase (21.9%) ( Table 1 ). However, among women diagnosed with endometriosis, hypertriglyceridemia prevalence was higher in the follicular phase (26.7%) versus the luteal phase (20.6%). ( Table 3 ). While there was no clear pattern for staging ( Table 3 ), we found the highest prevalence of hypertriglyceridemia among women with OE and DE in the follicular phase (70.0%) versus the luteal phase (33.3%) ( Table 3 ). Indeed, after adjustment, women with OE and DE were 4.38 times more likely to have hypertriglyceridemia (95% CI: 2.36, 8.12) in the follicular phase compared to 2.92 (95% CI: 1.06, 8.04) in the luteal phase ( Table 3 ). Continuous assessment of geometric mean concentrations of serum triglyceride indicated a pattern of higher concentrations in the follicular phase, versus luteal phases, for women with endometriosis, severe staging, and most notably typology as evidenced by the non-overlapping 95% CIs ( Supplementary Table 2 ). Restricted cubic splines revealed a linear relationship between endometriosis and serum triglycerides; however, precision was low due to limited power in our study ( Supplementary Figure 1 ). Results were slightly attenuated when evaluating hypertriglyceridemia at the level of ≥150 mg/dL vs less ( Supplementary Table 1 ). Results did not appreciably change when considering other confounding factors ( Supplementary Table 3 ).

Endometriosis is a chronic gynecologic condition without an established causal pathway. 1 It is considered a major reproductive health concern in the US, severely impacting more than 6.5 million women ages 15–44. 2 , 3 Endometriosis may also lead to debilitating symptoms that can severely impact quality of life such as chronic pelvic pain, dysmenorrhea, and infertility. 4 , 5 Prior research has suggested that endometriosis may be linked to cardiovascular disease event. In particular, women with endometriosis are at an increased risk of coronary heart disease, cerebrovascular disease, and and stroke. 6 Therefore, the impact of endometriosis on lipid profiles such as triglycerides is of great interest. This is partly due to conditions such as elevated triglycerides also regarded as a risk factor for heart disease and stroke. 7 Research on the associations between endometriosis and elevated triglyceride levels is currently limited, with very few studies exploring hypertriglyceridemia (defined as triglyceride levels of 150 mg/dl or higher). In the limited existing research, studies have been inconsistent. 8 – 13 Melo et al. 8 observed that triglyceride levels were higher in women with endometriosis (105.3 mg/dl) compared to women without endometriosis (83 mg/dl). Similarly, Li et al. 12 reported higher mean triglyceride levels in women with endometriosis (1.87 mmol/L) compared to a control group (1.38 mmol/L). In contrast to Melo et al. 8 and Li et al. 12 research, Zheng et al. 9 observed higher triglycerides levels in women with no endometriosis (1.23 mmol/L) compared to 1.12 mmol/L in women with endometriosis. This conflict could be attributed to its use of a study sample from a single health center, having a predominantly Asian participant demographic, and study participants with no endometriosis being older and having a higher Body Mass Index (BMI). However, Zheng et al. 9 did observe in women with endometriosis, higher staging was associated with higher triglyceride concentrations (moderate 0.96 mmol/L vs severe endometriosis 1.26 mmol/L). 8 The pattern of higher concentrations among higher staged endometriosis was also observed by Verit et al. 11 However, most studies have not been able to include data on stage and no prior studies we are aware of have considered endometriosis typology. In addition to not assessing severity or type of endometriosis, prior studies have not accounted for triglyceride fasting status and phase of menstrual cycle (follicular vs luteal). Epidemiological research has suggested that non-fasting vs fasting triglycerides better predict health outcomes, such as cardiovascular disease. 14 Additionally, detailed within-in woman assessments have shown high variability in lipid levels by menstrual cycle phase, with higher levels documented in the follicular phase. 15 Further, current research has not accounted for multiple confounding factors that can impact the relationship between endometriosis and hypertriglyceridemia. In particular, the role of physical activity, alcohol and caffeine consumption, and smoking. The present study aims to address prior gaps in the literature by assessing the association between endometriosis diagnosis, staging or severity, typology, and the occurrence of hypertriglyceridemia, considering fasting status, menstrual cycle phase, and multiple confounding factors assessed prior to incident endometriosis diagnosis.

We observed that moderate to severe (III/IV) endometriosis was more strongly associated with hypertriglyceridemia than minimal to mild (I/II) endometriosis. Prior studies have reported associations between elevated triglyceride levels and endometriosis staging. 8 , 9 , 11 , 12 To illustrate, Verit et al. 11 and Zheng et al. 9 both reported increasing triglycerides levels by increasing endometriosis severity. In the Verit et al. study, mean triglyceride levels were 135.0 mg/dL in women with no endometriosis, 157.3 mg/dL in women with minimal to mild endometriosis, and 189.0 mg/dL in women with moderate to severe endometriosis. Importantly, our study is also unique in that it is the only one that includes a multivariable assessment on the risk of hypertriglyceridemia or triglyceride levels ≥ 175 mg/dL. Poor lipid metabolism in women with endometriosis is a potential clinical pathway for hypertriglyceridemia occurrence. 12 Sphingomyelin, a class of phospholipids and a critical component of the plasma membranes, is an important determinant of lipoprotein metabolism. 27 , 28 While sphingomyelin is essential for cellular integrity, its accumulation—potentially exacerbated by endometriosis—can disrupt lipid metabolism and signaling pathways, contributing to elevated lipid levels in the bloodstream and tissues. Beyond lipid handling, sphingolipids more broadly are implicated in generalized inflammatory processes, which may explain why chronic inflammation from endometriosis, or indeed from other conditions, can contribute to increased cardiovascular risk. 27 Notably, research has associated higher-stage endometriosis with elevated sphingomyelin levels, supporting our finding that hypertriglyceridemia risk was highest in women with moderate to severe endometriosis compared to those with minimal to mild disease or no endometriosis. Thus, both lipid-specific and inflammation-mediated pathways may be relevant in linking endometriosis to cardiometabolic risk . 27 , 28 The present study also showed an association between endometriosis typology and hypertriglyceridemia. We observed that combined ovarian and deep infiltrating endometriosis (OE and DE) had a higher risk of hypertriglyceridemia than SE or OE alone. Importantly, we found no study that has directly assessed hypertriglyceridemia risk by endometriosis typology. The study of Vouk et al., 29 however, documented elevated concentrations of sphingomyelins in women with ovarian endometriosis. As previously described, elevated levels of sphingomyelin disrupt lipid metabolism, which likely results in an abnormal increase in lipid uptake in the cells and tissue. Further, the study of Bedin et al. 30 examined lipid nanoparticle concentrations, an analogue to low density lipoprotein (LDL) receptors, in deep endometriotic tissues. The results suggest increased LDL uptake by endometriotic tissue. Although the study focuses on LDL, it is worth noting that LDL and triglycerides share similar clinical implications. When we investigated whether cycle phase influenced the association between endometriosis and triglycerides, we found that the prevalence of hypertriglyceridemia and geometric mean triglyceride concentrations in women with endometriosis was higher in the follicular phase than in the luteal phase. Estrogen concentration, also associated with lipid metabolism, is known to rise with ovulation, possibly leading to higher lipid levels. 15 The effect is more pronounced in the follicular phase as estrogen levels peak. The fact that we only found triglycerides to be higher in the follicular versus luteal phase among women with endometriosis may be related to women with endometriosis having higher estrogen levels, 31 however further research is needed to confirm these novel findings. Additionally, while our findings collaborate the positive association between estrogen and lipoprotein cholesterol, 15 , 32 more research looking specifically at triglycerides is needed.

Hypertriglyceridemia has far-reaching consequences for women’s health, requiring increased attention. The condition is associated with an increased risk of all-cause mortality and incident cardiovascular disease events. 32 There are also documented associations with pregnancy-related complications such as preeclampsia, gestational diabetes, and fetal macrosomia. 33 , 34 It may also be important for health professionals to inquire about the menstrual cycle phase during routine metabolic panels, as there is variability in triglyceride levels observed across different phases of the menstrual cycle, which can impact the interpretation of the test results. More research is also needed on the impact of endometriosis severity and localization endometriosis (staging vs typology) on triglycerides across the menstrual cycle phase. In summary, while overall we found that women with, compared to without endometriosis, had a null association with hypertriglyceridemia, when looking at typology, we found women with ovarian endometrioma combined with deep infiltrating had a four-fold higher prevalence of hypertriglyceridemia. Additionally, women with moderate to severe staging had a two-fold higher prevalence. We also observed differences by menstrual cycle phase. Although research is extensive on the effects of endometriosis on women’s health, few studies exist assessing hypertriglyceridemia risk as an outcome, and even fewer studies specifically focusing on endometriosis staging and typology.