Endometriosis has been linked to cardiometabolic alterations, but whether these associations vary by disease severity or phenotype is unclear. We examined lipid profiles across endometriosis diagnosis, stage, and typology.

Data came from 476 women in the NICHD ENDO cohort. Endometriosis was confirmed laparoscopically and staged using the rASRM criteria (I−IV). Typology was categorized as superficial endometriosis (SE), ovarian endometrioma (OE), deep infiltrating endometriosis (DE), and OE+DE. We compared endometriosis status, stage (I/II vs III/IV), and typology to no endometriosis using adverse lipid thresholds (total cholesterol ≥200 mg/dL, HDL <50 mg/dL, LDL ≥100 mg/dL, triglycerides ≥175 mg/dL, non-HDL ≥130 mg/dL, VLDL ≥30 mg/dL, ApoA1 <125 mg/dL, and ApoB ≥120 mg/dL). Adjusted prevalence ratios (aPR) and 95% CIs were estimated via generalized linear models, controlling for age, race/ethnicity, BMI, income, marital status, and serum cotinine.

Endometriosis diagnosis alone was not associated with adverse lipid profiles. In contrast, moderate/severe disease showed higher prevalence of elevated triglycerides (aPR= 2.27; 95% CI: 1.18,4.35) and VLDL (aPR= 2.41; 95% CI: 1.50, 3.85). Typology revealed stronger patterns: OE and OE+DE were associated with adverse profiles across multiple markers (aPRs 1.59–4.09), particularly ApoB and triglycerides. Minimal/mild disease and SE were not associated.

The metabolic signal was phenotype-driven rather than diagnosis-driven, with severe stage and OE/OE+DE showing clear associations with adverse lipid profiles. These findings suggest lipid profiles may serve as markers of phenotype severity or shared biological milieu. Replication in larger cohorts is needed.