Patients
MR imaging examinations performed from February 2010 to June 2010 of 30 patients (13 women and 17 men with a mean age of 48.8 years [range.
11-75 years]) .
Dynamic contrast-enhanced MRI (3.0T) was performed in 12 patients with infectious condition and 18 patients with non-infectious condition.
In infectious group,
cellulitis in 3 patients,
myonecrosis with myositis in 3 patients,
synovitis in 2 patients,
osteomyelitis in 2 patients and septic arthritis in 2 patients.
Examination Technique
Bone dynamic MR images of the joints were obtained using a 3.0 T MR imager (Intera Achieva,
Philips healthcare,
The Netherlands).
The maximum slew rate was 200mT/m/sec and the maximum gradient strength was 80mT/m.
Variation in matrix size and field of view for different infection sites ranged from 160x160 to 256 x 256,
respectively.
Bone dynamic MR images were obtained with a THRIVE-SPAIR sequence (T1 High Resolution Isotropic Volume Examination for dynamic using SPAIR fat suppression) [TR/TE,
3.8/1.91 ms; slice thickness/gap,
3 mm/0 mm; field of view,
200 × 200 mm; SENSE factor,
1; NSA,
1; flip angle,
10°; band width,
±89.4kHz; dynamic number,
20; scan time,
3min 47sec].
20 dynamic MR images were acquired for each slice location.
After dynamic MR image acquisition was started,
0.1mmol/kg Body Weight of Gadobutrol 1.0 Molar (Gadovist®,
Bayer Schering Pharma AG,
Berlin,
Germany) was rapidly injected in the antecubital vein and was followed by a 20-mL saline flush.
Before an administration of intravenous injection of gadolinium,
4 images were acquired for baseline and additional 16 images were acquired (interval: 10 seconds,
total: 200 seconds).
Total one hundred sixty images were obtained for dynamic contrast-enhanced MRI.
Regions of interest (ROIs) were placed over the region of synovium for each patient.
The mean signal intensity-time curves were computed by averaging across voxels within ROI for each dynamic phase and then were normalized using baseline signal intensity.
Statistical analysis was performed to test if there are significant differences of dynamic parameters between patients with infectious and non-infectious condition.
Data Analysis
All MR images were analyzed by two radiologists who were unaware of the clinical findings.
DCEMRI(3.0T) was performed in 30 patients.
After an injection of gadolinium,
20 images were acquired(interval: 10 seconds,
total: 200 seconds).
Total one hundred sixty images were obtained for DCEMRI.
Regions of interest(ROI) with a standardized size of 2mm(3.14mm2) were placed over the synovium for each patient.
The mean signal intensity-time curves were computed by averaging across voxels within ROI for each dynamic phase and then were normalized using baseline signal intensity.
The normalized signal intensity-time curves were fitted to a sigmoid equation,
and then 3 parameters (maximal slope,
time of half rising and peak value of enhancement) were computed.
Statistical analysis was performed with dynamic parameters between infectious and non-infectious groups.
Semi-quantitative analysis of the signal intensity (SI)–time curve was performed with the region-of-interest technique.
The highest amplitude of synovial enhancement was used to determine the SI-time curves for MR imaging enhancement.
Statistical analysis
Statistical analysis was performed using the SPSS Statistics 17 package (SPSS Inc.,
Chicago,
IL).
To assess differences between the ROI values for Semi-quantitative Analysis and Mann-Whitney tests were used.
A p-values less than 0.05 was indicated significant results.