Research Article: A preoperative nomogram integrating transvaginal ultrasound endometrial thickness and nutritional, inflammatory, and metabolic indicators for predicting lymph node metastasis in endometrial cancer
Abstract:
Preoperative assessment of lymph node metastasis (LNM) risk is critical for surgical planning in endometrial cancer (EC). Transvaginal ultrasound (TVU) endometrial thickness (ET) is a routinely available morphological parameter, whereas nutritional and metabolic indicators reflect the systemic host-tumor interaction. Although prior studies have incorporated either imaging or biomarker-based predictors individually, a unified preoperative framework integrating TVU morphology with a comprehensive panel of nutritional, inflammatory, and metabolic biomarkers for LNM prediction has not been established.
This retrospective cohort study enrolled 203 patients with histopathologically confirmed EC who underwent definitive surgical staging at our institution between January 2021 and December 2025. Preoperative TVU-measured ET, tumor size, depth of myometrial invasion (MI), and cervical stromal invasion were recorded, alongside nutritional, inflammatory, and metabolic biomarkers including CA-125, HE4, albumin, neutrophil-to-lymphocyte ratio (NLR), platelet-to-lymphocyte ratio (PLR), prognostic nutritional index (PNI), fibrinogen, fasting blood glucose, and triglyceride-glucose (TyG) index. Spearman correlation analysis quantified associations between ET and biomarkers. Univariate and multivariate logistic regression identified independent LNM predictors. Receiver operating characteristic (ROC) curve analysis with DeLong tests evaluated discriminative performance, and a nomogram was constructed and internally validated.
LNM was confirmed in 35 patients (17.2%). ET was significantly higher in LNM-positive patients (34.7?±?11.6?mm vs. 22.4?±?8.1?mm, p <?0.001) and demonstrated moderate-to-strong positive correlations with CA-125 ( ? =?0.512), HE4 ( ? =?0.468), NLR ( ? =?0.398), fibrinogen ( ? =?0.426), and TyG index ( ? =?0.233), and negative correlations with albumin ( ? =??0.374) and PNI ( ? =??0.441) (all p <?0.001). Multivariate analysis identified ET (adjusted OR?=?1.092 per mm, 95% CI: 1.043–1.143), deep MI???50% (OR?=?4.827), CA-125???35?U/mL (OR?=?3.614), NLR???3.0 (OR?=?3.082), PNI?<?50 (OR?=?2.947), fibrinogen ?4.0?g/L (OR?=?2.614), grade 3 histology (OR?=?3.408), and non-endometrioid type (OR?=?2.891) as independent predictors of LNM (all p <?0.05). The combined nomogram achieved an AUC of 0.941 (95% CI: 0.895–0.987), significantly outperforming ET alone (AUC?=?0.812, p <?0.001).
Preoperative TVU-ET is significantly correlated with nutritional, inflammatory, and metabolic indicators and independently predicts LNM in EC. A nomogram integrating ET with CA-125, NLR, PNI, fibrinogen, tumor grade, and histological type provides promising, good internally validated discrimination for LNM and may guide individualized surgical planning. External validation in independent cohorts remains necessary before clinical implementation.
Introduction:
Preoperative assessment of lymph node metastasis (LNM) risk is critical for surgical planning in endometrial cancer (EC). Transvaginal ultrasound (TVU) endometrial thickness (ET) is a routinely available morphological parameter, whereas nutritional and metabolic indicators reflect the systemic host-tumor interaction. Although prior studies have incorporated either imaging or biomarker-based predictors individually, a unified preoperative framework integrating TVU morphology with a comprehensive panel of…
Read more