Ultrasound for Distal Forearm Fracture: A Systematic Review and Diagnostic Meta-Analysis

Study Objective
To determine the diagnostic accuracy of ultrasound for detecting distal forearm fractures.
Methods
A systematic review and diagnostic meta-analysis was performed according to the PRISMA statement. We searched MEDLINE, Web of Science and the Cochrane Library from inception to September 2015. All prospective studies of the diagnostic accuracy of ultrasound versus radiography as the reference standard were included. We excluded studies with a retrospective design and those with evidence of verification bias. We assessed the methodological quality of the included studies with the QUADAS-2 tool. We performed a meta-analysis of studies evaluating ultrasound to calculate the pooled sensitivity and specificity with 95% confidence intervals (CI95%) using a bivariate model with random effects. Subgroup and sensitivity analysis were used to examine the effect of methodological differences and other study characteristics.
Results
Out of 867 publications we included 16 studies with 1,204 patients and 641 fractures. The pooled test characteristics for ultrasound were: sensitivity 97% (CI95% 93–99%), specificity 95% (CI95% 89–98%), positive likelihood ratio (LR) 20.0 (8.5–47.2) and negative LR 0.03 (0.01–0.08). The corresponding pooled diagnostic odds ratio (DOR) was 667 (142–3,133). Apparent differences were shown for method of viewing, with the 6-view method showing higher specificity, positive LR, and DOR, compared to the 4-view method.
Conclusion
The present meta-analysis showed that ultrasound has a high accuracy for the diagnosis of distal forearm fractures in children when used by proper viewing method. Based on this, ultrasound should be considered a reliable alternative, which has the advantages of being radiation free.

研究目的
本研究旨在评估超声检测前臂远端骨折的诊断准确性。

研究方法
本研究依据PRISMA声明开展系统评价及诊断性荟萃分析。我们检索了MEDLINE、Web of Science及Cochrane图书馆自建库至2015年9月的相关文献，纳入以X线摄影作为金标准、评估超声诊断准确性的所有前瞻性研究，排除回顾性设计研究及存在验证偏倚的研究。采用QUADAS-2工具对纳入研究的方法学质量进行评价。通过带有随机效应的双变量模型，对评估超声诊断效能的研究进行荟萃分析，以计算合并敏感度、合并特异度及其95%置信区间（CI95%）；并通过亚组分析及敏感性分析，探究方法学差异及其他研究特征对结果的影响。

研究结果
本次检索共获取867篇文献，最终纳入16项研究，涉及1204名患者及641例骨折病例。超声的合并检测特征如下：敏感度97%（CI95% 93%~99%）、特异度95%（CI95% 89%~98%）、阳性似然比（LR）20.0（8.5~47.2）、阴性似然比0.03（0.01~0.08）；对应的合并诊断比值比（DOR）为667（142~3133）。不同扫查视图方式存在显著差异：与4视图法相比，6视图法的特异度、阳性似然比及诊断比值比均更高。

结论
本项荟萃分析结果显示，采用规范扫查视图方式时，超声对儿童前臂远端骨折具有较高的诊断准确性。据此，超声可被视为一种可靠的替代检查手段，其优势在于无电离辐射。