Letter: comparative diagnostic accuracy of magnetic resonance elastography vs. eight clinical prediction rules for noninvasive diagnosis of advanced fibrosis in biopsy‐proven NAFLD
Letter: comparative diagnostic accuracy of magnetic resonance elastography vs. eight clinical prediction rules for noninvasive diagnosis of advanced fibrosis in biopsy‐proven NAFLD
复制标题
信件：磁共振弹性成像与八种临床预测规则对活检证实的 NAFLD 晚期纤维化无创诊断的诊断准确性比较
DOI：
10.1111/apt.13871
复制
发表时间：
2017
影响因子：
7.6
通讯作者：
M. Shao
中科院分区：
文献类型：
--
作者：
H. Shi;L. Liu;M. Shao
SIRS，我们饶有兴趣地阅读了Cui等人的文章。近年来，非酒精性脂肪性肝病（NAFLD）的患病率呈逐年上升趋势。NAFLD的范围从良性肝脏脂肪堆积到非酒精性脂肪性肝炎，非酒精性脂肪性肝炎是一种进行性疾病，可能导致纤维化、肝硬化和肝细胞癌。然而，在临床实践中，晚期纤维化的识别是复杂的。在前瞻性文章中，作者提出了一种新的指标，二维磁共振弹性成像（2D-MRE），这是一种利用剪切波的先进磁共振成像模式。结果显示，与其他临床预测规则相比，2D-MRE为NAFLD患者中晚期纤维化的诊断提供了显着更高的准确性。然而，我们对这项研究仍有一些问题。Jeffrey研究的目的是找到一种区分晚期纤维化和其他疾病的指标。晚期纤维化的公认定义是F3-F4。在这篇文章中，48，26，9，13和6例患者分别患有0，1，2，3和4期纤维化。因此，仅19例（18.6%）患者为晚期纤维化（3-4期纤维化）（表1）。阳性病例比例相对较低。因此，我们认为本研究的样本量设计存在小缺陷。Hajian-Tilaki为不同条件下诊断试验准确性和试验结果的研究提供了样本量计算的概念框架。根据本文，我们计算了上述前瞻性研究的样本量，发现不少于100。在本研究中，总样本量仅为102个，实验样本的比例仅占总样本的18.6%。因此，如果样本量更大，结果可能更有说服力。此外，在发展中国家，地雷危险教育的成本效益是需要解决的另一个问题。与磁共振成像相比，瞬态弹性成像（Fibroscan，Echosens，巴黎，法国）更便宜，更快速，更方便。许多研究表明，Fibroscan在诊断NAFLD中的肝纤维化方面显示出非常有前途的结果，8特别是对于NAFLD患者的纵向随访。因此，需要进一步的研究来比较MRE与弹性成像技术。如果Fibroscan与MRE相比有更多的优势，将大大降低检查成本，更有利于随访。总之，2D-MRE是有前途的，有效的，可用于放射学诊断晚期纤维化，但我们认为需要更多的数据来证明这一论点。需要进一步的研究来评估2D-MRE的优点和缺点。这些技术的比较和组合有待进一步研究。
SIRS, We read the article by Cui et al. with interest. As we know, the prevalence rate of nonalcoholic fatty liver disease (NAFLD) has increased year by year recently. NAFLD ranges from benign hepatic fat accumulation to nonalcoholic steatohepatitis, which is a progressive form of disease that may lead to fibrosis, cirrhosis, and hepatocellular carcinoma. However, the identification of advanced fibrosis is complicated in clinical practice. In the prospective article, the authers propose a new indicator, two-dimensional magnetic resonance elastography (2D-MRE), which is an advanced magnetic resonance imaging modality utilising shear waves. The results have revealed that 2D-MRE had provided a significantly higher accuracy for diagnosis of advanced fibrosis in NAFLD patients compared to other clinical prediction rules. However, we still have some questions concerning this study. The aim of Jeffrey’s study was to find an indicator to differentiate advanced fibrosis from others. A wellaccepted definition of advanced fibrosis is F3–F4. In this article, 48, 26, 9, 13 and 6 patients suffered stage 0, 1, 2, 3 and 4 fibrosis, respectively. Thus, only 19 (18.6%) cases of the patients were advanced fibrosis (stage 3–4 fibrosis) (table 1). The proportion of positive cases was relatively low. Thus, we think that design of the sample size about this study has small flaws. Hajian-Tilaki provided a conceptual framework of sample size calculations in the studies of diagnostic test accuracy in various conditions and test outcomes. According to this article, we have calculated the sample size of the prospective study above and found it is not less than 100. In this research, the total sample size is only 102, and the proportion of the experimental samples only accounts for 18.6% of the total sample. Therefore, the result could be more convincing, if the sample size is larger. In addition, in developing countries, the cost-effectiveness of MRE is another aspect which needs to be addressed. Compared with MRE, transient elastography (Fibroscan, Echosens, Paris, France) is cheaper, faster and more convenient. Many studies have shown that Fibroscan has demonstrated very promising results for the diagnosis of liver fibrosis in NAFLD, 8 especially for longitudinal follow-up of patients with NAFLD. Thus, further studies are needed to compare MRE with elastography techniques. If Fibroscan has more advantages compared with MRE, it will greatly reduce the costs of inspection, and is more conducive for follow-up. In conclusion, 2D-MRE is promising, effective, and available radiographically to diagnose advanced fibrosis, but we think more data are required to prove the arguments. Further studies are needed to evaluate the advantages and disadvantages of 2D-MRE. Also, the comparation and combination of these technologies awaits further study.