Diagnostic value of high-frame-rate contrast-enhanced ultrasound and nomogram model in lymphomas

Lymphomas typically present as painless enlarged lymph nodes. Current imaging modalities demonstrate limited diagnostic value in their evaluation. Our study aims to investigate the clinical utility of high-frame-rate contrast-enhanced ultrasound(H-CEUS) in characterizing superficially enlarged lymph nodes and develop a nomogram integrating clinical information with ultrasonographic characteristics to predict the risk of occurrence and aggressiveness of superficial lymphomas involving the subcutaneous lymphoid tissue.

Based on histopathological findings, 288 patients were categorized into the benign group (n = 71) and malignant group [n = 217, metastatic subgroup (n = 150), lymphoma subgroup (n = 67)]. The lymphoma subgroup subdivided into aggressive lymphoma (n = 32) subgroup and indolent lymphoma (n = 35) subgroup. For comparative analysis, 67 patients were randomly selected from the benign and metastatic groups to constitute the non-lymphoma subgroup. Univariate analysis was performed to compare clinical information and ultrasonographic characteristics across groups. The independent risk factors of lymphoma screening were analyzed by multivariate logistic regression, based on which a nomogram was constructed. The diagnostic efficacies of H-CEUS, and the nomogram in lymphoma evaluation were compared using receiver operating characteristic (ROC)curve analysis.

With pathological results as the "gold standard", the sensitivity, specificity, positive and negative predictive values, accuracy and the area under the ROC curve of H-CEUS in diagnosing benign and malignant lymph nodes were higher than those of conventional contrast-enhanced ultrasound (C-CEUS). Compared to indolent subgroup, aggressive subgroup exhibits a higher heterogeneous enhancement pattern on H-CEUS and higher lactate dehydrogenase (LDH) levels (all p < .001). The multivariate logistic regression analysis revealed that age > 59 years, striped/reticular hyperechoic patterns in the lymph node cortex, and the H-CEUS centrifugal "fireworks "pattern were all independent risk factors for lymphoma (all p < .05). In contrast, the US + Age + H-CEUS combination exhibited the most optimized performance. Nomogram model demonstrated strong discriminative ability, with a concordance index (C-index) of 0.874 (95% CI: 0.811-0.938).

H-CEUS can better identify the properties of superficial enlarged lymph nodes than C-CEUS. In H-CEUS, its specific performance of centrifugal "fireworks" pattern provides a more reliable diagnostic basis for lymphoma. By combining clinical data, B-mode ultrasound and H-CEUS, the diagnostic efficiency of superficial lymphoma is greatly improved. Nomogram and H-CEUS enhancement pattern can be used to predict the risk and invasiveness of lymphoma, which has broad clinical application prospects.