DIAGNOSTIC ACCURACY OF ULTRASONOGRAPHY FOR DIAGNOSIS OF MALIGNANCY OF LYMPH NODES IN NEWLY DIAGNOSED CASES OF CARCINOMAS OF THE HEAD AND NECK, TAKING HISTOPATHOLOGY AS THE GOLD STANDARD

Abstract

Background: Cervical lymph node involvement remains one of the most decisive prognostic indicators in head and neck carcinoma, directly influencing staging, treatment decisions, and overall survival. Early and accurate identification of malignant nodes is therefore essential for optimal management. Ultrasonography (USG), being non-invasive, widely accessible, and free of radiation exposure, offers real-time assessment of nodal morphology and vascularity. However, its diagnostic performance varies across populations, underscoring the need for continued evaluation to strengthen evidence for its clinical application.

Objective: To determine the diagnostic accuracy of ultrasonography in detecting malignant cervical lymph nodes using histopathology as the gold standard and to identify ultrasonographic features that independently predict malignancy.

Methods: A cross-sectional diagnostic accuracy study was conducted on 100 newly diagnosed head and neck carcinoma patients. Demographic and clinical variables—including age, gender, smoking and alcohol use, and primary tumor site—were recorded. High-resolution USG was performed to evaluate lymph node size, shape, borders, hilum presence, echogenicity, and Doppler vascularity. Ultrasonographic impressions were compared with histopathological results obtained through fine-needle aspiration or excisional biopsy. Sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), accuracy, p-values, and multivariate logistic regression were calculated to assess diagnostic performance and independent predictors.

Results: Histopathology confirmed malignant nodes in 48% of patients. USG correctly identified 40 malignant nodes and 47 benign nodes, with 8 false negatives and 5 false positives. Sensitivity was 83.3%, specificity 90.4%, PPV 88.9%, NPV 85.5%, and overall accuracy 87% (p < 0.001). Malignant nodes were round in 72.9%, had irregular borders in 79.2%, absent hilum in 87.5%, hypoechogenicity in 93.8%, and peripheral vascularity in 81.2% (all p < 0.001). Logistic regression identified lymph node size >2 cm, smoking, alcohol use, and USG features—round shape, irregular borders, absent hilum, hypoechogenicity, and peripheral vascularity—as significant independent predictors of malignancy.

Conclusion: Ultrasonography demonstrated high diagnostic accuracy and reliably differentiated malignant from benign cervical lymph nodes. Distinct ultrasonographic features served as strong predictors of malignancy and can support targeted biopsy, improve staging precision, and refine treatment planning in head and neck carcinoma.