Development and Validation of the Riemma Inflammation Score for Endometriosis (RISE): A Preoperative Tool to Predict Disease Presence and Deep Infiltrating Phenotype

Background: Endometriosis is a chronic, inflammatory gynaecological disease characterised by ectopic endometrial tissue and systemic immune activation. Despite advances in imaging, the diagnosis of endometriosis—particularly deep infiltrating endometriosis (DIE)—remains delayed and frequently requires surgery for confirmation. The RISE score (Riemma Inflammation Score for Endometriosis) was designed as a pragmatic, low-cost, and biologically grounded diagnostic tool integrating inflammatory, biochemical, and clinical parameters to identify women with endometriosis and to discriminate between superficial and deep phenotypes. Methods: This prospective two-step study enrolled consecutive women undergoing surgical assessment for suspected endometriosis. In Step 1, candidate variables including CA-125, neutrophil, platelet and monocyte counts, lymphocyte- to-monocyte ratio (LMR), platelet-to-lymphocyte ratio (PLR), albumin, and pain scores were recorded prior to surgery. A multivariate predictive model was developed using logistic and LASSO regression to identify women with histologically confirmed endometriosis versus controls. In Step 2, a refined model was derived to distinguish deep from superficial disease among affected women. Each model underwent internal validation and calibration testing. Results: In Step 1, the RISE model achieved excellent discrimination for endometriosis (AUC = 0.904, 95 % CI 0.871–0.939), with sensitivity = 0.86 and specificity = 0.81 at an optimal cut-off of 0.418. Calibration was almost perfect (slope = 1.02; intercept = –0.05; Hosmer–Lemeshow p = 0.976). Key predictors included monocyte count, LMR, serum albumin, and dysmenorrhoea severity. 4 In Step 2, the RISE model successfully differentiated deep from superficial endometriosis (AUC = 0.742, 95 % CI 0.653–0.830; sensitivity = 0.64; specificity = 0.76; cut-off = 0.381; Hosmer–Lemeshow p = 0.368). Low albumin and reduced LMR were the strongest discriminants of deep infiltrating lesions. Discussion: The RISE project demonstrated that systemic inflammatory and biochemical alterations can accurately predict both the presence and the depth of endometriotic disease. The two-step approach, prospectively validated, mirrors clinical reasoning: first recognising disease presence, then defining its severity. The RISE framework thereby bridges laboratory and clinical diagnostics, offering a reproducible, inexpensive, and accessible adjunct to imaging-based evaluation. Conclusions: The RISE score represents an innovative and biologically plausible tool that quantifies systemic inflammation to predict endometriosis presence and phenotype. Its strong diagnostic performance, prospective validation, and feasibility suggest potential for widespread clinical implementation and integration into modern diagnostic algorithms. External multicentric validation and incorporation with imaging or molecular data are warranted to confirm its clinical and prognostic utility.