Mytsyk Yulian
Voxel Medical Diagnostic Centre, Katowice, Poland.
Department of Urology, Department of Radiology, Danylo Halytsky Lviv National Medical University, Lviv, Ukraine.
Pol J Radiol. 2025 May 21;90:e239-e252. doi: 10.5114/pjr/202588. eCollection 2025.
This study aimed to assess the diagnostic performance of multiphase contrast-enhanced computed tomography (MCECT) in differentiating benign and malignant solid and cystic small renal masses (SRMs), predicting histologic subtypes, and grading, using signal intensity (SI) and tumour-to-cortex signal intensity (TCSI) ratio.
A retrospective analysis was conducted on 181 patients with solid and cystic SRMs (≤ 4 cm). MCECT imaging across 4 phases (non-contrast, corticomedullary, nephrographic, and excretory) was performed. SI and TCSI values were measured, and their diagnostic performance was evaluated using receiver operating characteristic (ROC) analysis. Solid, Bosniak IIF, III, and IV SRMs underwent histopathological confirmation.
Among solid SRMs, excretory phase SI achieved an area under the curve (AUC) of 0.848 for differentiating RCC from other SRMs, with 100% sensitivity and 61.3% specificity. For distinguishing renal cell carcinoma (RCC) from benign SRMs, the most effective parameter was the TCSI ratio obtained from computed tomography excretory phase (88.6% sensitivity, 52.4% specificity, 0.763 AUC). For Bosniak IIF cysts, the corticomedullary phase SI provided an AUC of 0.902, with 93% sensitivity and 87.5% specificity. RCC subtyping showed distinct SI characteristics across phases, particularly for clear cell RCC. Nephrographic phase SI differentiated low- versus high-grade RCC, with an AUC of 0.901, 90.2% sensitivity, and 86.4% specificity.
MCECT-derived imaging biomarkers, particularly SI and TCSI, are effective non-invasive tools for characterising SRMs, aiding in the differentiation of benign and malignant lesions, histological subtypes, and tumour grades. Their integration with advanced radiomics could further enhance diagnostic accuracy.
本研究旨在评估多期对比增强计算机断层扫描（MCECT）在鉴别实性和囊性小肾肿块（SRM）的良恶性、预测组织学亚型及分级方面的诊断性能，采用信号强度（SI）和肿瘤与皮质信号强度（TCSI）比值进行评估。
对181例实性和囊性SRM（≤4 cm）患者进行回顾性分析。进行了4期（非增强、皮质髓质期、肾实质期和排泄期）的MCECT成像。测量了SI和TCSI值，并使用受试者操作特征（ROC）分析评估其诊断性能。实性、Bosniak IIF、III和IV级SRM均进行了组织病理学确认。
在实性SRM中，排泄期SI在鉴别肾细胞癌（RCC）与其他SRM时曲线下面积（AUC）为0.848，敏感性为100%，特异性为61.3%。对于区分肾细胞癌（RCC）与良性SRM，最有效的参数是计算机断层扫描排泄期获得的TCSI比值（敏感性88.6%，特异性52.4%，AUC 0.763）。对于Bosniak IIF级囊肿，皮质髓质期SI的AUC为0.902，敏感性为93%，特异性为87.5%。RCC亚型在各期显示出不同的SI特征，尤其是透明细胞RCC。肾实质期SI可区分低级别与高级别RCC，AUC为0.901，敏感性为90.2%，特异性为86.4%。
MCECT衍生的成像生物标志物，尤其是SI和TCSI，是表征SRM、辅助鉴别良恶性病变、组织学亚型和肿瘤分级的有效非侵入性工具。它们与先进的放射组学相结合可进一步提高诊断准确性。
