| [Objective]The patients with long-term knee pain and discomfort were evaluated on a 3T MRI scanner, used T1WI, T2WI, T1rho and T2 mapping sequences. Classification of meniscus injury or degeneration, measurement T1rho and T2 relaxation time with each meniscus. To assess and compare the relationship between T1rho and T2 relaxation time on the meniscus of different stages.[Materials and Methods]1. SubjectsThe 34 patients (14 male,20 female; median age=60 years; age ranged from 40 to 78 years) with long-term knee pain and discomfort, in Zhujiang Hospital of the Southern Medical University, during January 2011-October 2013.Inclusion criteria:1. One year or more history of knee pain and discomfort.2. Age≥40 years, BMI index [(20±5) kg/m2].3. Non-immune diseases and immune-related arthritis history.4. MRI examination without contraindications disease. Exclusion criteria:1. With previous history of knee surgery.2. Severe deformity of the knee can not be straightened or horizontal position, or braking time <40 minutes.3. Meniscal malformations (such as discoid meniscus etc).4. Image quality is lower than the diagnostic criteria.The experiment by the hospital ethics committee to discuss and agree, that all experiment data were collected in Department of Radiology, Zhujiang Hospital of the Southern Medical University. All patients informed consent.2. Imaging hardware and Imaging protocolAll MRI experiments were performed on a 3.0T clinical MR scanner (Philips Achieva TX 3.0T MR scanner, Koninklijke Philips Electronics N.V., Eindhoven, The Netherlands). Using SENSE-Flex-M 8-channel transmit-receive phased-array knee coil for the imaging measurements. Supine, left or right knee scan, advanced foot, straight position, the lower edge of the patella to scan center, scan range including tibial plateau and femoral condyle. Conventional sagittal T1/T2 FSE sequence, then sagittal T2 mapping and T1rho sequence, total scanning time of approximately 34 minutes. MRI scans were performed in one continuous session without removing the subject from the scanner. Measurements were conducted in the evening between 8 and 10 p.m. All patients in the week before the examination without strenuous exercise, resting half an hour before the scan.Scan parameters:sagittal Ti-weighted fast spin echo (FSE) imaging were: repetition time (TR)/echo time (TE)= shortest/62 ms; field of view (FOV)=14 cm; matrix=256x256; slice thickness= 5mm, slice interval= 5mm; and number of excitations (NEX)= 2.Sagittal T2-weighted fast spin echo (FSE) with fat saturation sequenceimaging were:TR 700ms、TE 20ms; field of view (FOV)= 14 cm; matrix=256x256; slice thickness= 5mm, slice interval= 5mm; and number of excitations (NEX)= 2.Sagittal T1rho-weighted images were obtained using the spin-lock technique and spiral image acquisition. Acquisition parameters for 3D-balanced-TFE were: TR=5.95ms, TE=2.95 ms; FOV= 14 cm, matrix= 256x256; slice thickness= 4mm, slice interval=1mm; flip angle=50 degrees, time of spin-lock (TSL)= 1 ms,10 ms, 20 ms,30 ms,40 ms; NEX= 2. spectral presaturation of inversion recovery (SPIR) fat saturation with aspin-lock frequency= 759.5 Hz/pixel.Parameters for sagittal T2 mapping were:TR= 2700 ms; TE= 16 ms,32 ms,48 ms,64 ms,80 ms,96 ms,112ms, and 128 ms, FOV= 14cm, matrix=256x256; thickness= 4.0 mm, slice interval= 0 mm; NEX= 2.maps and T2 maps of menisci were reconstructed by fitting a series of the T1rho-weighted/T2-weighted image intensity pixel-by-pixel with post processing software comes with The MRI machine.3. MR images analysis and processingThe clinical evaluation of menisci was performed using 2D-sagittal T1-weighted, T2-weighted, Tirho mapping and T2 mapping by two experienced (more than 8 years) radiologist. The radiologist was blinded to subjects’ specific information, measurement T1rho and T2 relaxation date, and Classification used Stoller’s grading. Four different compartments of the meniscus were segmented (medial anterior, medial posterior, lateral anterior, lateral posterior). Meniscal classification was performed using the Stoller grading (normal meniscus, Ⅰ,Ⅱ, Ⅲ).Meniscus segmentation and measure was performed by using in-house software ImageJ 1.48v (developed by National Institutes of Health, USA). Freehand selections (a tool in ImageJ 1.48v) was used to simplify the drawing of splines delineating meniscal areas. In each lateral and medial meniscus, distinct regions were defined and segmentation was performed in the individual T2 and T1rho maps separately. Each meniscus compartments to measure only one dimension value, Select the level of the lesion shows the most clear. As for normal meniscal the measurement area most large, most uniform signal level. In order to avoid interference synovial fluid surrounding meniscus, ROI edge line is located slightly medial meniscus edge. If the tear of the meniscus reach the surface near articular, the ROI line across the tear line.4. StatisticalUsing Microsoft Office Excel 2013 version and IBM SPSS 19.0 version to entry and analyze data. Measurement results are displayed as (mean ± standard deviation, X±S).4.1 Single factor analysis of variance (ANOVA) was use to analyze the difference between the different grades by T1rho relaxation time of the meniscal. Multiple comparisons method (Dunnett T3) was use to analyze the differences between the different grade. Test standards P<0.05. In line graph displays the trend in different grade T1rho relaxation time.4.2 Single factor analysis of variance (ANOVA) was use to analyze the difference between the different grades by T2 relaxation time of the meniscal. Multiple comparisons method (Dunnett T3) was use to analyze the differences between the different grade. Test standards P<0.05. In line graph displays the trend in different grade T2 relaxation time.4.3 Pearson correlation analysis was use to analyze the correlation between meniscal’s T1rho and T2 relaxation time.[Results]All patients completed the whole examination. Of which 3 patients were excluded from the poor image quality because the inspection process jitter. Of which 31 patients were enrolled to analyze with good image quality. The enrolled patients include 17 female,14 male, a total of 124 meniscus. Of which excluding 5 meniscus with fragmentation and no clear boundary, and ultimately included in the study 119 meniscus.Classification by T1rho relaxation time pseudo-color images, we have 22 normal meniscis, accounting for 19%, T1rho relaxation time (21.08± 3.01) ms,29 meniscus for Ⅰ degree, accounting for 24%, T1rho relaxation time (24.53± 2.56) ms, 36 meniscus for Ⅱ degree, accounting for 30%, T1rho relaxation time (26.60±1.60) ms,32 meniscus for Ⅲ degree, accounting for 27%, T1rho relaxation time (30.68 ±2.81)ms.T1rho relaxation time for the meniscus was increasing significantly along with the Stoller’s grade increasing, meniscus T1rho relaxation time is differences between the different Stoller’s grade, the difference is remarkable, with a statistically significant (P<0.05).Classification by T2 relaxation time pseudo-color images, we have 15 normal meniscis, accounting for 13%, T2 relaxation time (0.48 ± 0.50) ms,18 meniscus for I degree, accounting for 15%, T2 relaxation time (8.86 ± 1.70) ms,42 meniscus for Ⅱ degree, accounting for 35%, T2 relaxation time (16.01 ± 4.26) ms,44 meniscus for Ⅲ degree, accounting for 37%, T2 relaxation time (38.23 ± 13.61ms).T2 relaxation time for the meniscus was increasing significantly along with the Stoller’s grade increasing, meniscus T2 relaxation time is differences between the different Stoller’s grade, the difference is remarkable, with a statistically significant (P<0.05).Using Spearman (correlation analysis method by spss) to analyze the relationship between meniscal’s Tirho and T2 relaxation time. Calculating the correlation coefficient R=0.729 (P<0.01). Can be think there is a positive correlation between the meniscus’s Tirho and T2 relaxation time, and the correlation is more significant.(Conclusion]1. Articular cartilage proteoglycan content is the main factor influence Tirho relaxation times, although the meniscus cartilage proteoglycan content is relatively low, but as the meniscus injury Tirho relaxation time also will be increased, the performance is similar to articular cartilage, Tirho techniques can be used as a meniscal’s quantitative analysis tool.2. The main factor affecting of meniscars T2 relaxation time is the arrangement direction of the collagen fibers, the integrity of collagen fibers and anisotropy. As the meniscus injury T2 relaxation time also will be increased, T2 relaxation time can be used as a meniscal’s quantitative analysis tool.3. Tirho and T2 mapping can be used to quantitatively study meniscus injury in osteoarthritis, and there was a positive correlation between the two method. The two method reflect different material composition in the meniscus, so its the sensitivity is different. May be T2 mapping is more sensitive than Tirho in osteoarthritis meniscus damage analysis.4. Tirho and T2 mapping can be used as meniscal’s noninvasive quantitative analysis method in Osteoarthritis, but two methods can not be used as independent standards diagnostic tool for knee meniscus injury, it should be combined with conventional morphology (T1 WI and T2WI) for meniscus injury diagnostic. The specific principles and reasons For further study, its has a huge space for development. |
PDF Full Text Request