Academic Department of Gastroenterology, Medical School of National & Kapodistrian University of Athens, "Laiko" General Hospital, Athens, Greece.
Academic Department of Gastroenterology, Medical School of National & Kapodistrian University of Athens, "Laiko" General Hospital, Athens, Greece.
Eur J Radiol. 2024 Jun;175:111475. doi: 10.1016/j.ejrad.2024.111475. Epub 2024 Apr 18.
To systematically review the accuracy of spleen stiffness measurement (SSM) by 2D- Shear Wave Elastography (2D-SWE) in predicting high risk for bleeding varices (HRV) in cirrhotic patients.
PubMed, Embase, Web of Science, Medline, Cochrane, and Google Scholar databases were searched up to 31/05/2023 for all human studies using 2D-SWE to estimate SSM and endoscopy to detect HRV. Meta-analysis was performed using a generalized linear mixed model. Publication bias was evaluated using the funnel plot asymmetry test. The Area Under the Summarized Receiver Operating Characteristic curve (AUSROC) was estimated using the "mada" package.
A total of 13 studies and 1970 patients were included. Of them, 27.8 % had HRV. The pooled sensitivity and polled specificity of SSM in detecting HRV were 90 % (95 %CI:87-92 %) and 68 % (95 %CI:58-77 %), respectively, with an AUSROC at 0.86 (95 %CI:0.82-0.90). The median cutoff value of SSM in detecting HRV was 34.2 kPa. In studies including exclusively HBV cirrhotic patients, SSM's polled sensitivity and specificity in predicting HRV was 88 % (95 %CI:82-92 %) and 73 % (95 %CI:68-78 %), respectively. The AUSROC was 0.84 (95 %CI:0.81-0.87). The number of repeated measurements per patient (<5 or ≥ 5) did not affect the method's capability. Using Aixplorer to evaluate SSM had a higher sensitivity in ruling out HRV than other 2D-SWE devices.
Our meta-analysis supports that SSM by 2D-SWE has a good diagnostic performance for ruling out HRV in cirrhosis.
系统评价二维剪切波弹性成像(2D-SWE)测量脾脏硬度(SSM)预测肝硬化患者高危出血静脉曲张(HRV)的准确性。
检索PubMed、Embase、Web of Science、Medline、Cochrane 和 Google Scholar 数据库,截至 2023 年 5 月 31 日,收集所有使用 2D-SWE 估计 SSM 和内镜检测 HRV 的人体研究。使用广义线性混合模型进行荟萃分析。使用漏斗图不对称检验评估发表偏倚。使用“mada”包估计汇总受试者工作特征曲线下面积(AUSROC)。
共纳入 13 项研究和 1970 例患者,其中 27.8%有 HRV。SSM 检测 HRV 的 pooled 敏感性和 pooled 特异性分别为 90%(95%CI:87-92%)和 68%(95%CI:58-77%),AUSROC 为 0.86(95%CI:0.82-0.90)。SSM 检测 HRV 的中位数截断值为 34.2 kPa。在仅包括乙型肝炎肝硬化患者的研究中,SSM 预测 HRV 的 pooled 敏感性和特异性分别为 88%(95%CI:82-92%)和 73%(95%CI:68-78%),AUSROC 为 0.84(95%CI:0.81-0.87)。每位患者的重复测量次数(<5 次或≥5 次)不影响该方法的能力。使用 Aixplorer 评估 SSM 对排除 HRV 的敏感性高于其他 2D-SWE 设备。
本荟萃分析支持 2D-SWE 测量 SSM 对肝硬化患者 HRV 具有良好的诊断性能。