The diagnostic accuracy of magnetic resonance imaging for anterior cruciate ligament injury in comparison to arthroscopy: a meta-analysis

We performed this meta-analysis to examine the diagnostic accuracy of MRI for the diagnosis of anterior cruciate ligament (ACL) injury in comparison to arthroscopy. We also compared the diagnostic accuracy of MRI with magnetic field intensities (MFI) greater than or equal to 1.5T with those below 1....

Full description

Saved in:
Bibliographic Details
Published in:Scientific reports 2017-08, Vol.7 (1), p.7583-10, Article 7583
Main Authors: Li, Kun, Du, Jun, Huang, Li-Xin, Ni, Li, Liu, Tao, Yang, Hui-Lin
Format: Article
Language:English
Subjects:
59/57
692/699/578
Accuracy
Anterior cruciate ligament
Arthroscopy
Humanities and Social Sciences
Knee
Magnetic fields
Magnetic resonance imaging
Meta-analysis
multidisciplinary
NMR
Nuclear magnetic resonance
Science
Science (multidisciplinary)
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:We performed this meta-analysis to examine the diagnostic accuracy of MRI for the diagnosis of anterior cruciate ligament (ACL) injury in comparison to arthroscopy. We also compared the diagnostic accuracy of MRI with magnetic field intensities (MFI) greater than or equal to 1.5T with those below 1.5T, in addition to different MRI sequences. Studies relevant to the diagnosis of ACL injury by MRI and arthroscopy were analyzed. Computer and manual retrieval were carried out on studies published between January 1, 2006 and May 31, 2016. Twenty-one papers were included. Neither threshold nor non-threshold effects were present ( p  = 0.40, p  = 0.06). The pooled sensitivity (SE), specificity (SP), positive likelihood ratio (LR+), negative likelihood ratio (LR−) and diagnostic odds ratio (DOR) with 95% confidence interval (CI) were 87% (84–90%), 90% (88–92%), 6.78 (4.87–9.44), 0.16 (0.13–0.20) and 44.70 (32.34–61.79), respectively. The area under the curve (AUC) was 0.93. The risk of publication bias was negligible ( p =  0.75). In conclusion, examination by MRI is able to provide appreciable diagnostic performance. However, the principle, which states that the higher the MFI, the better the diagnostic accuracy, could not be verified. Additionally, conventional sequences (CSs) associated with proton density-weighted imaging (PDWI) are only slightly better than CSs alone, but not statistically different.
ISSN:2045-2322
2045-2322
DOI:10.1038/s41598-017-08133-4