Differentiation of clear cell and non clear cell renal cell carcinoma through high-frame-rate contrast-enhanced ultrasound

Abstract

To evaluate the value of high-frame-rate contrast-enhanced ultrasound (H-CEUS) in differentiating between clear cell renal cell carcinoma (ccRCC) and non-clear cell renal cell carcinoma (nccRCC). A retrospective study was conducted on 84 patients diagnosed with CCRCC and 28 patients diagnosed with nccRCC, all patients underwent conventional ultrasound (US), conventional contrast-enhanced ultrasound (C-CEUS), and High-frame-rate contrast-enhanced ultrasound (H-CEUS) at the First Affiliated Hospital of Nanchang University between October 2022 and April 2024. Dynamic image sequences were acquired for each patient. Clinical, US, C-CEUS, and H-CEUS parameters were compared between the two groups using t -tests, χ² tests, Fisher’s exact tests, and U -tests as appropriate. Binary logistic regression was employed to identify independent risk factors associated with ccRCC and nccRCC, combining clinical, US, C-CEUS, and H-CEUS data. Receiver operating characteristic (ROC) curve analysis was performed to assess the diagnostic accuracy of C-CEUS and H-CEUS. Nomograms were constructed using R to visualize the probability of ccRCC based on significant parameters. Significant differences were observed between ccRCC and nccRCC groups in color Doppler flow imaging (CDFI) patterns ( χ² =10.967, p = 0.012). Analysis of contrast-enhanced ultrasound (CEUS) characteristics revealed significant differences between the two groups in terms of enhancement mode, regression mode, peak intensity, homogeneity, and presence of no enhancement for both H-CEUS and C-CEUS (all p < 0.05). Time-intensity curve (TIC) parameter analysis showed that time to peak (TTP) was shorter in the ccRCC group compared to the nccRCC group for both H-CEUS and C-CEUS ( Z = 3.434, 2.355; p = 0.001, = 0.019). Conversely, time to decline (DT/2) and mean transit time (MTT) were longer in the ccRCC group for both H-CEUS and C-CEUS ( Z = 3.749, 3.649; p < 0.001, < 0.001 in H-CEUS; Z = 3.407, 3.239; p = 0.001, = 0.001 in C-CEUS). Peak intensity (PI) and area under the curve (AUC) were also higher in the ccRCC group for both H-CEUS and C-CEUS ( Z = 3.007, 4.173; p = 0.003, < 0.001 in H-CEUS; Z = 3.897, 3.911; p < 0.001, < 0.001 in C-CEUS). Binary logistic regression analysis identified regression mode, homogeneity, and TTP of H-CEUS as independent risk factors for differentiating ccRCC from nccRCC. ROC curve analysis demonstrated superior diagnostic performance of H-CEUS compared to C-CEUS, with a sensitivity of 81.0%, specificity of 89.3%, and an area under the curve (AUC) of 0.935. For C-CEUS, peak intensity, homogeneity, and PI were identified as independent risk factors. ROC curve analysis showed a sensitivity of 69.1%, a specificity of 87.5%, and an AUC of 0.784. Statistical analysis confirmed that H-CEUS had significantly higher diagnostic efficiency than C-CEUS (Z=-2.892, p = 0.004). Nomograms were constructed using R language to visually represent the probability of ccRCC based on significant C-CEUS and H-CEUS parameters. H-CEUS, with its improved time resolution, shows potential for enhancing diagnostic differentiation between ccRCC and nccRCC. This technology may provide a valuable tool for the precise diagnosis of renal cell carcinoma.