Chen Ziman, Wang Yingli, Gunda Simon Takadiyi, Han Xinyang, Su Zhongzhen, Ying Michael Tin Cheung
Department of Health Technology and Informatics, The Hong Kong Polytechnic University, Kowloon, Hong Kong, China.
Department of Ultrasound, EDAN Instruments, Inc., Shenzhen, China.
Quant Imaging Med Surg. 2024 Feb 1;14(2):1766-1777. doi: 10.21037/qims-23-962. Epub 2024 Jan 2.
Assessing renal fibrosis non-invasively in patients with chronic kidney disease (CKD) remains a considerable clinical challenge. This study aimed to investigate the diagnostic efficacy of different approaches that combine shear wave elastography (SWE) and estimated glomerular filtration rate (eGFR) in distinguishing between mild fibrosis and moderate-to-severe fibrosis in CKD patients.
In this prospective study, 162 patients underwent renal SWE examinations and renal biopsies. Using SWE, the right renal cortex stiffness was measured, and the corresponding SWE value was recorded. Four diagnostic patterns were used to combine eGFR and SWE value: in isolation, in series, in parallel, and in integration. The receiver operating characteristic (ROC) curve was established, and the area under the ROC curve (AUC) was calculated to quantify diagnostic performance. Sensitivity, specificity, and accuracy were computed.
The eGFR demonstrated sensitivity of 68.2% and specificity of 83.8%, whereas the SWE value displayed sensitivity of 84.1% and specificity of 62.2%, yielding a similar AUC (78.2% and 77.8%, respectively). Combining in series improved specificity to 97.3%, superior to other diagnostic patterns (all P values <0.01), but compromised sensitivity to 58.0%. When combined in parallel, the sensitivity increased to 94.3%, exceeding any other strategies (all P values <0.05), but the specificity dropped to 48.7%. The integrated strategy, incorporating eGFR with SWE value via the logistic regression algorithm, exhibited an AUC of 85.8%, outperforming all existing approaches (all P values <0.01), with balanced sensitivity, specificity, and accuracy of 86.4%, 74.3%, and 80.9%, respectively.
Using an integrated strategy to combine eGFR and SWE value could improve diagnostic performance in distinguishing between mild renal fibrosis and moderate-to-severe renal fibrosis in patients with CKD, thereby helping clinicians perform a more accurate clinical diagnosis.
对慢性肾脏病(CKD)患者进行肾纤维化的无创评估仍然是一项重大的临床挑战。本研究旨在探讨结合剪切波弹性成像(SWE)和估计肾小球滤过率(eGFR)的不同方法在区分CKD患者轻度纤维化和中重度纤维化方面的诊断效能。
在这项前瞻性研究中,162例患者接受了肾脏SWE检查和肾活检。使用SWE测量右肾皮质硬度,并记录相应的SWE值。采用四种诊断模式将eGFR和SWE值相结合:单独使用、串联、并联和整合。建立受试者工作特征(ROC)曲线,并计算ROC曲线下面积(AUC)以量化诊断性能。计算敏感性、特异性和准确性。
eGFR的敏感性为68.2%,特异性为83.8%,而SWE值的敏感性为84.1%,特异性为62.2%,AUC相似(分别为78.2%和77.8%)。串联结合可将特异性提高到97.3%,优于其他诊断模式(所有P值<0.01),但敏感性降至58.0%。并联结合时,敏感性增加到94.3%,超过任何其他策略(所有P值<0.05),但特异性降至48.7%。通过逻辑回归算法将eGFR与SWE值相结合的整合策略,AUC为85.8%,优于所有现有方法(所有P值<0.01),敏感性、特异性和准确性分别为86.4%、74.3%和80.9%,较为均衡。
采用整合策略将eGFR和SWE值相结合可提高CKD患者轻度肾纤维化与中重度肾纤维化鉴别的诊断性能,从而帮助临床医生进行更准确的临床诊断。