Keywords:
Observer performance, Imaging sequences, Diagnostic procedure, MR, Musculoskeletal joint, Management, Bones, Athletic injuries, Trauma
Authors:
U. Khanzada1, M. Hussain2, S. A. Khan1, K. Masood1, H. Adil1, S. O. Adil1; 1Karachi/PK, 2Karachi, Si/PK
DOI:
10.1594/ecr2018/C-1417
Methods and materials
The cross-sectional study was conducted at Radiology department of a tertiary care hospital from Feb 2017 to Apr 2017.
A total of 233 patients of any age or gender presenting for routine MRI knee with any complaint were included through non-probability convenient sampling.
While patients with MRI claustrophobia or with any contraindication to MRI such as any metallic device or implant was excluded.
MRI was performed on 1.5 Tesla or 0.4 Tesla MRI machines.
Axial and sagittal T2 weighted images,
sagittal proton density images and coronal STIR images were obtained.
Two sets of the images were made. One set labeled limited protocol included sagittal PD and coronal STIR images and other set included complete protocol MRI knee.
Two radiologists blinded to each other’s result interpreted MRI.
One interpreted the full protocol and record findings and other interpreted limited protocol and record findings.
The findings of limited protocol were compared with the full protocol for calculation of sensitivity,
specificity,
PPV,
NPV and diagnostic accuracy.
SPSS version 21 was used for statistical analysis.
Mean ±standard deviation (SD) was calculated for the quantitative variable like the age while frequency and percentages were calculated for qualitative variables like gender,
ACL tear,
PCL tear,
median meniscus tear and lateral meniscus tear on both full protocol MRI and limited protocol MRI.
Diagnostic accuracy including sensitivity,
specificity,
positive predictive value (PPV) and negative predictive value (NPV) of limited protocol MRI was calculated using full protocol MRI as the gold standard.