Predictive value of pan-immune-inflammation value and ultrasonographic parameters in cervical lymphadenopathy:...

Accurate preoperative differentiation of cervical lymphadenopathy (LAP) is essential for optimal surgical planning. This study aimed to evaluate the contribution of preoperative pan-immune-inflammation value (PIV), platelet-to-lymphocyte ratio (PLR), demographic characteristics, and ultrasonographic findings in predicting malignant histopathology in patients undergoing excisional biopsy for cervical LAP, and to develop a practical clinical risk score.

This retrospective cohort study was conducted at a tertiary referral center between January 2014 and December 2025. Patients who underwent excisional cervical lymph node biopsy with available preoperative ultrasonography and complete blood count data were included. The primary outcome was malignant versus non-malignant (reactive or granulomatous) histopathology. Candidate predictors included age, sex, neck level, ultrasonographic features (hilum loss, conglomeration, roundness index, and longest diameter), and inflammatory indices (PIV and PLR). Independent predictors were identified using multivariable logistic regression analysis, while discriminatory performance and optimal cut-off values were assessed using receiver operating characteristic (ROC) analyses.

Of the 481 patients included, 184 (38.2%) had reactive, 91 (18.9%) granulomatous, and 206 (42.8%) malignant histopathology. Patients with malignancy were older (median age, 51.5 years) and more frequently male (65.5%) (both p < 0.001). Multivariable analysis demonstrated that conglomeration (OR, 6.18), male sex (OR, 3.51), Level V localization (OR, 7.77), increased roundness index (OR, 2.33), age (OR per year, 1.04), and PIV (OR per unit, 1.001) were independently associated with malignancy. Although PIV showed moderate discriminatory ability (AUC, 0.624), it achieved high specificity (83.3%) at a cut-off value > 567.1. A six-variable, points-based clinical scoring system (range, 0-11) effectively stratified malignancy risk (7.4% in the low-risk group and 83.1% in the high-risk group).

Malignancy risk in cervical lymphadenopathy can be reliably predicted through the combined assessment of demographic characteristics, ultrasonographic patterns, and the Pan-Immune-Inflammation Value. Conglomeration, Level V localization, advanced age, and elevated PIV emerged as the strongest independent predictors of malignancy. However, multicenter external validation is required before the proposed risk score can be implemented in routine clinical practice.