|
Acrucial prerequisite for adequate long-term diarthrodial joint function is the integrity of its articular cartilage.1 Many previous studies2-9 have demonstrated that the prevalence of articular chondral trauma is high, and cartilage damage is generally acknowledged as an early factor in the process of irreversible joint degeneration. The recent development of new surgical procedures and tissue engineering,2 such as autologous chondrocyte implantation (ACI), have promising results of forming repair hyaline or hyalinelike cartilage. Therefore, there is a strong need for a quick noninvasive diagnostic technique that can give a valid estimate of the status of the cartilage reliably, discriminating intact cartilage from various grades of impaired cartilage.
A variety of magnetic resonance imaging (MRI) sequences of articular cartilage have attracted intense interest and been the subject of numerous research studies over the past years.2,3,10-25 They can be broadly divided into physiologic and morphologic techniques.3 Since clinical evaluation of cartilage has primarily focused on morphologic evaluation, we attempted to compare several available MRI sequences in the morphologic assessment of articular cartilage of patients with traumatic injury on the knees, by using surgical examination as the reference standard.
METHODS
Patients
This study was approved by the institutional review board of our hospital and written informed consent was obtained from each patient. Inclusion criteria were as follows: (a) patient had definite history of external injury on the knee and (b) subsequently referred for open knee surgery or arthroscopy of the knee, and (c) the surgery record of the condition of the articular cartilage was available. Exclusion criteria were as follows: (a) surgery that did not allow one to review articular cartilage, (b) prior knee surgery, (c) unavailability of the MR image, (d) and contraindications for MR imaging.
One hundred and thirty-eight knees of 137 patients satisfied the inclusion criteria between March and November, 2008, and one of these patients had been imaged by MR and operated on both knees. The age of the 137 patients was 19–62 years (mean of 31 years). Eight-nine (65%) of the patients were male and 49 (35%) patients were female. Eighty-one patients had traffic accidents, including directed impact on their knees (in 72 patients), five patients were wounded in a fall during activity (in 4 patients) or from high (3 meters approximately) place (in 1 patients), forty-six had sprained their knees, and six had trauma by heavy blows with certain objects. All of the external injury of the knees occurred within six months before MRI examination.
MRI
MRI scans had been performed between March and November, 2008. The conventional MRI of the knee, spin-echo T1-weighted (SE T1WI) and dual echo fast spin-echo Proton-Density/ T2-weighted (FSE PDWI/T2WI), had been used routinely in the past decade at our hospital. In order to evaluate the articular cartilage of the knee, fat-suppressed three-dimensional spoiled gradient echo (fat-suppressed 3D SPGR) and three-dimensional fast imaging employing steadystate acquisition (3D FIESTA) sequences were added for the scanning of the knees of the involved patients. All the MRI protocols were performed on a 1.5T GE GENESIS_SIGNA GEMSOW scanner (General Electric Medical Systems, Milwaukee, WI, USA). An extremity coil was used for all images, with the following parameters of the above four sequences: (a) coronal SE T1WI: TR/TE=360 ms/9 ms, the bandwidth=20.8 kHz, the echo train length=1/2, field of view (FOV)=18 cm×18 cm, slice thickness=5.0 mm, interslice gap=1.0 mm, the matrix size=320×192, the number of excitations (NEX)=3; (b) coronal and sagittal dual-echo FSE PDWI/T2WI: TR/TEPD/TET2=3660 ms/10.9 ms/119.5 ms, the bandwidth = 25 kHz, the echo train length=1/2, other parameters the same as SE T1WI; (c) sagittal fat-suppressed 3D SPGR: TR/TE=60 ms/5.0 ms, the bandwidth=15.6 kHz, the echo train length=1/1, field of view (FOV)=18 cm×13.5 cm, slice thickness=1.5 mm, interslice gap=0 mm, the matrix size=256×192, the NEX =0.75, flip angle=40. The NEX was reduced to 0.75, so the imaging time of SPGR can be shorten from more than 10 minutes (NEX=1) to about 5 minutes, which was very important for the patients with wounded knees; (d) sagittal 3D FIESTA: TR/TE=5.8 ms/1.8 ms, the bandwidth=41.7 kHz, the echo train length=1/1, field of view (FOV)=20 cm × 20 cm, slice thickness=1.6 mm, interslice gap= –0.8mm, the matrix size=320×320, the NEX=3, flip angle=30.
Imaging time (in minutes:seconds) were as follows: T1WI coronal sequences=3:34, PD/T2 WI coronal or sagittal sequences both are 3:47, 3D SPGR sequences = 4:55–5:59, and 3D FIESTA sequences = 4:41–5:17. The time ranges of SPGR and FIESTA sequences were according to different number of slices, which between 46 and 64, depending on different size of knee.
When suspecting patellar fracture or injuries on the retinaculum patellae, an axial scan of SE T1WI and FSE PD/T2WI sequences were added. Multiplanar reconstructions of 3D SPGR and 3D FIESTA were performed in the workstation according to standards.
Image analysis
All 138 knees of 137 patients had their MR examinations retrospectively reviewed by consensus reading of two experienced musculoskeletal radiologists. The readers were blinded to the patient's specific history or mechanism of injury, and without knowledge of the results of arthroscopy or open knee surgery at the time of review. First, the image of 3D SPGR was evaluated by two radiologists for evidence of chondral injuries. Second, two weeks later, the same two radiologists interpreted the image of 3D FIESTA of each patient. Finally, after another two weeks, a combination of the conventional sequences (SE T1WI, FSE PD/T2WI) and 3D SPGR was available for the interpretation of injury by the same two radiologists. The diagnosis allowed assessment of interobserver agreement for the detection of these lesions. Disagreements were settled by consensus.
The following six positions of the articular surface of knee were evaluated: the lateral femoral condyle (LFC), medial femoral condyle (MFC), lateral tibial plateau (LTP), medial tibial plateau (MTP), trochlea and patella. Chondral injuries were assessed using a modified version of the Noyes classification system of articular cartilage defects of the knee joint:10,11,25 grade 0, intact cartilage with normal signal and uniform thickness; grade 1, focal abnormal signal without surface abnormalities; grade 2, superficial ulceration or fissuring, with a depth of not more than 50% of cartilage thickness; grade 3, deep ulceration or fissuring of more than 50% but less than 100% of cartilage thickness; grade 4, full-thickness cartilage defect with normal or erosion of subchondral bone. If different grades of cartilage lesions were present within the same articular surface, the worst lesion was used for grading.
Surgery
One hundred and twenty-eight patients had undergone arthroscopy, including one on both knees. The other nine patients had open knee surgery. All arthroscopies or open knee surgeries were performed within 7 days of MR examination (minimum time post MR exam=1 day, maximum time post MR exam=7 days, mean time post MR exam=3 days). The operation records were correlated with MRIs with respect to the presence and location of articular injuries.
For cartilage evaluation during surgery, the grading of the cartilage lesions was performed in correspondence with the MR imaging, with minor adaptations required according to the surgical findings of abnormalities:10,11,25 grade 0, normal cartilage; grade 1, cartilage softening or discoloration; grade 2, erosions or fissures with a depth of not more than 50% of the cartilage thickness; grade 3, erosions or fissures with a depth of greater than 50% but less than 100%; and grade 4, full thickness cartilage defect, or osteochondral injury with separation of osteochondral fragment. Surgical grading served as the standard of reference.
Statistical analysis
The articular cartilage was evaluated on a surface-by- surface basis in the same manner as assessment by arthroscopy. We selected the diagnostic cut-offs between grade 1 and grade 2, i.e., grade 0 and 1 (intact cartilage surface) were presumed as negative, and grades 2, 3, and 4 (cartilage defects) were defined as positive. Sensitivity, specificity, and accuracy of 3D SPGR, 3D FIESTA, and the combination of the conventional MRI and 3D SPGR imaging were calculated (Excel, version 11.1.1, Microsoft, Redmond, Wash). The Wilson score method was used to determine the 95% confidence intervals for sensitivity and specificity. The two-tailed exact form of the McNemar statistic was used as the test of significance comparing the grading results of the different MR images (SPSS 11.0, Chicago, III). Significance was set at P <0.05.
RESULTS
According to the findings from arthroscopy or open knee surgery, 489 articular surfaces were normal (grade 0), 156 cartilage surfaces were focal softening or discoloration (grade 1). There were 132, 25 and 26 grade 2, grade 3 and grade 4 cartilage abnormalities, respectively (Figure 1). The incidence of cartilage changes (grade 1 to 4) was 41% (339/828) and the incidence of cartilage defects (grade 2 to 4) was 22% (183/828). Grade 4 and grade 3 lesions were absent at MTP. Generally, the more severity of the grade of cartilage abnormalities, the less the numbers of cartilage abnormalities at each articular surface (Figure 2).
 view in a new window |
Figure 1. The surgical grading evaluation of the 828 articular surfaces of 138 knees. |
 view in a new window |
Figure 2. The distributions of the surgical grading results at different articular surfaces. LFC: lateral femoral condyle. MFC: medial femoral condyle. LTP: lateral tibial plateau. MTP: medial tibial plateau. |
In the evaluation of 3D SPGR, 329 articular surfaces were judged as normal (grade 0) and 253 surfaces were grade 1 (Figure 3). The numbers of grade 2, 3 and 4 lesions were 193, 30 and 23, respectively (Figures 4–7). The rate of exact match between 3D SPGR and surgical grade was 49% (P >0.05). During the interpretation of 3D FIESTA images, 309 positions of the articular cartilage were grade 0 and 280 positions were grade 1 (Figure 3). There were 186 surfaces interpreted as grade 2 (Figure 4), 25 grade 3 (Figure 5), and 28 grade 4 (Figures 4, 6, and 7). The rate of exact match between 3D FIESTA and surgical grade was 61% (P >0.05), which was higher than that of 3D SPGR, however, there was no significant difference between them (P >0.05). With the combination of 3D SPGR and conventional MRI sequences, the rate of exact match between MR and surgical results was highest (82%), though there was no significant difference among the three MRI evaluation procedures (P >0.05). The numbers of grade 0 and grade 1 (Figure 3) lesions were 434 and 207, respectively. The numbers of grade 2, 3 and 4 lesions were 137, 23 and 27, respectively (Figures 4–7). The comprehensive data are shown in Table 1.
 view in a new window |
Figure 3. Grade 1 lesion in a 27-year-old male with history of pain after a fall. A: Sagittal proton density-weighted image (3660/10.9) shows localized cartilage hyperintensity (arrow) of the medial femoral condyle. The patient also has patellar fracture (shows on other conventional MR images). B: Sagittal fat-suppressed 3D SPGR image (60/5, 40º flip angle) shows distinct cartilage hypointensity (white arrow). The streak-like hypointensity (black arrow) of the tibial plateau is artifact. C: Sagittal 3D FIESTA image (5.8/1.8, 30º flip angle) shows mild irregular margin of cartilage. Both the two layers, light grey inside and dark grey outside, represents the articular cartilage. Hematocele in suprapatellar bursa is also shown.
Figure 4. Grade 2 lesion on patella and Grade 4 lesion on femoral condyle in a 62-year-old male with history of knee trauma and positive grinding test. A: Sagittal proton density-weighted image (3660/10.9) suspects of localized cartilage defect (white arrowhead) of patella, and shows focal cartilage defect (black arrowhead) and subchondral bruising (white arrows) of medial femoral condyle. The patient also has tear on anterior horn of lateral meniscus (shows on other conventional MR images). B: Sagittal fat-suppressed 3D SPGR image (60/5, 40º flip angle) shows focal cartilage defect (white arrowhead) that involved no more than 50% of the cartilage thickness of patella, and focal cartilage thinning and defect that involves almost the entire cartilage thickness (black arrowhead) of femoral condyle. C: Sagittal 3D FIESTA image (5.8/1.8, 30º flip angle) shows the morphological and signal change of the above two sites of cartilage, and the areas of bone bruise (white arrows) of femoral condyle is clear.
Figure 5. Grade 3 lesion in a 28-year-old male with swollen left knee and limited range of motion after traffic accident. A: Sagittal proton density-weighted image (3660/10.9) shows femoral bruise near patellofemoral joint (pentagon), and fibular head fracture (black arrow). B: Sagittal fat-suppressed 3D SPGR image (60/5, 40º flip angle) shows almost full-thickness fissure on patellar cartilage (white arrow). The fracture line on fibular head (black arrow). C: Sagittal 3D FIESTA image (5.8/1.8, 30º flip angle) shows intensity change of patellar cartilage (white arrow), while the femoral bruise (pentagon) and fracture line on fibular head (black arrow) are clear. |
 view in a new window |
Figure 6. Grade 4 lesion in a 33-year-old female complained persistent knee pain after traffic accident. A: Sagittal proton density-weighted image (3660/10.9) shows stread-like high fluid signal (arrow) in midportion of lateral femoral condyle. The patient also has marrow edema on lateral femoral condyle (shows on other conventional MR images). B: Sagittal fat-suppressed 3D SPGR image (60/5, 40º flip angle) shows areas of full-thickness cartilage defect (white arrow) of lateral femoral condyle. The lineate hypointensity (black arrowheads) through the cartilage of patella is artifact. C: Sagittal 3D FIESTA image (5.8/1.8, 30º flip angle) shows strip-like vacancy of articular cartilage, which is filled with synovial fluid (arrow). D: Coronal reconstruction of 3D SPGR, the up ending of the dot line shows the same cartilage defect.
Figure 7. Grade 4 lesion in a 41-year-old male after badminton injury. Sagittal proton density-weighted image (3660/10.9). A: and Sagittal T2-weighted image (3660/119.5). B: show localized hyperintensity (arrowhead) at the posterial edge of lateral femoral condyle and areas of bone bruise (pentagon). Sagittal fat-suppressed 3D SPGR image (60/5, 40º flip angle). C: Axial reconstruction of 3D SPGR. D: and Sagittal 3D FIESTA image (5.8/1.8, 30º flip angle). E: Focal loss of cartilage (white arrow) and subchondral bone (black arrow) in posterial aspect of lateral femoral condyle. |
 view in a new window |
Table 1. Comparison of grading between surgical examination and MR |
Sensitivity, specificity and accuracy of the three appraisal steps were calculated at all six positions of the articular surfaces of the knee. In the first step, i.e., the inter- pretation of 3D SPGR images, the range of the sensitivity, specificity and accuracy at different positions were 42%–77%, 80%–92%, and 70%–86%, respectively (Table 2). In the second step, i.e. the interpretation of 3D FIESTA images, the range of the sensitivity, specificity and accuracy at different positions were 50%–69%, 83%–95%, and 72%–89%, respectively (Table 3). In the third step, i.e. the judgment of the combination of 3D SPGR and conventional MR images, the overall range of the sensitivity, specificity and accuracy were highest, at 55%–81%, 93%–100%, and 82%–93%, respectively (Table 4). On the above surface-by-surface basis, the positive and negative predictive values of the overall knees were calculated, as well as sensitivity and specificity. The positive predictive value of 3D SPGR was 71% (174/246), and the negative predictive value was 83% (485/582). For 3D FIESTA, the positive and negative predictive values were 74% (178/239) and 79% (465/589), respectively. And similarly, highest positive and negative predictive values (both were 90%, 168/187 and 574/641) occurred when interpreting 3D SPGR and conventional MR images at the same time.
 view in a new window |
Table 2. Diagnostic performance of 3D SPGR MR image |
 view in a new window |
Table 3. Diagnostic performance of 3D FIESTA MR image |
 view in a new window |
Table 4. Diagnostic performance of the combination of 3D SPGR and conventional MR image |
DISCUSSION
Because of the human bipedal nature, the knee is one of the most commonly involved joints in external injuries, including acute and chronic trauma as well as repetitive activities.5 The complex structure and function of the articular cartilage can be disrupted by even minor injuries.7,8 Wa et al8 retrospectively reviewed 31 516 knee arthroscopies of patients in all age groups and reported chondral lesions in 19 827 (63%) of patients, with a mean of 2.7 articular cartilage injuries per knee. The incidence of grade 3 lesions was 41% and grade 4 lesions was 19%. A review of 1000 arthroscopies by Hjelle et al8 also reported an incidence of 5% for grade 3 and 4 chondral lesions. In our study, the incidence of cartilage changes of all the evaluated surfaces was 41% (339/828), cartilage defects was 22% (183/828), grade 3 lesions was 3% (25/828), and grade 4 lesions was 3% (26/828). All the incidences of cartilage lesions of different grades in our study were lower than that of the previous studies, however, these differences may be partly because of the relatively smaller sample size and precluding asymptomatic and osteoarthritis or other chronic degeneration knees in our research.
The early detection of the articular cartilage abnormities was vital for the early intervention to prevent further degeneration. 3D SPGR MR imaging is one of the most commonly used sequences for cartilage evaluation in clinical practice.3,14,18,19 In this study, chondral lesions of the knee were shown to be accurately determined with 3D SPGR, with 64% sensitivity, 87% specificity and 80% accuracy for detection. Compared with 85%–95% sensitivity in David et al study,18 the value of sensitivity in this study is lower. This difference may partly be because we decreased the number of excitations (NEX) to 0.75, which can shorten the imaging time from 10–12 minutes to about 5 minutes, sacrificing some resolution. However, the twofold decrease in imaging time was particularly important for increasing the endurance of involved patients and reducing the opportunity of motion artifact, especially for severely injured patients. In order to redeem the loss of signal-to-noise ratios (SNR), we adopted a TR of 60 mseconds, a TE of 5 mseconds, and a flip angle of 40º to maximize the contrast among articular cartilage, fluid, and marrow.19 Since fat-suppressed 3D SPGR essentially suppresses all stationary tissue, it is not useful in evaluating the fibrocartilage, ligaments, or soft tissues of the knee,19 and not sensitive for marrow edema, which is often an indication for a defect of the overlying cartilage.2,25 The lamination sign is often observed on 3D SPGR, especially in the patellofemoral compartment. It was manifested on 3D SPGR images as a trilaminar appearance consisting of a hyperintense superficial lamina, hypointense intermediate lamina and hyperintense deep lamina. It cannot be confused with the hyaline cartilage lamination lamellar construction or cartilage lesion. It is at least partly due to truncation artifacts, which results from insufficient sampling of high spatial frequencies and appears as signal-intensity fluctuations near sharp boundaries between tissues with strong contrast.18
FIESTA is a derivation of steady-state free precession (SSFP), which was first described in 1986 as fast imaging with steady-state precession (FISP).22 Although previous work with fluctuating equilibrium MR imaging has been limited by low spatial resolution, improvements in automated shimming routines and gradient technology have allowed its use for detecting cartilage defects in the knee.16,20 In our study, hyaline cartilage lesions of the knee on 3D FIESTA images were shown to accurately match with that of surgical examination (P >0.05), with 59% sensitivity, 88% specificity and 78% accuracy for detection, similar to that of 3D SPGR. Compared with 3D SPGR, 3D FIESTA MR imaging provides contrast between synovial fluid and cartilage while preserving signal from the cartilage itself and may be more useful in the detection of cartilage defects.16 In our study, some superficial or cartilage defects were impalpable on 3D FIESTA images, which may be partly due to lack of fat suppression. We also noted, though, that FEISTA MR imaging was more sensitive to bone edama or bruise in injured knees than that of SPGR, which can make some contribution in detecting and evaluating the cartilage lesion for proper grading. However, the use of this sequence in the diagnosis of meniscal, tendon, or ligament disease may not be as useful as fast SE MR imaging and will be the subject of future studies.16
In the assessment of the combination of 3D SPGR and conventional MR imaging (2D T1-weighted SE and dual-echo fast SE Proton-Density/T2-weighted sequences), the sensitivity, specificity and accuracy were 71%, 97% and 90%, respectively. Both the positive and negative predictive values were 90%. All of the evaluation indices reached the highest values compared to that of simple 3D SPGR or simple 3D FIESTA, but there was no significant difference among them (P >0.05). Nevertheless, fat suppressed proton-density fast SE imaging allows demonstration of defects and gross morphologic changes of articular cartilage. Furthermore, the most important purpose of employing the conventional MR sequences was dedicated by their ability of depiction in bone fracture, bone marrow bruising, menisci tear, ligament interruption and muscles edema, which can be frequently observed on traumatic knees. Detecting associated alterations in subchondral marrow signal, including bone bruising, bone edema, or subchondral fracture on conventional MR images, may be a help in delineating areas of overlying cartilage injury on 3D SPGR images accordingly. The finding of a focal signal change in the subchondral bone marrow should encourage careful evaluation of possible overlying hyaline articular cartilage injury or disease.2 Lastly, the all-round grasp of a traumatic knee can guide the clinician in preparing the patient toward the expectations of the additional rehabilitation time necessary for a cartilage resurfacing technique, compared with simple arthroscopic hyaline cartilage repair.6
In conclusion, our study indicated that less than half of the total articular surfaces of the involved traumatic knee had morphological changes (grade 1 to 4) on hyaline cartilage, and one fifth had cartilage defects (grade 2 to 4). Both 3D SPGR and 3D FIESTA imaging had a similar performance in detecting these chondral lesions in a reasonable time, and the comprehensive information of neighboring structures and more accurate evaluation in cartilage grading results can be obtained when combining conventional MRI of the knee.
REFERENCES
1. Christian G. New techniques for cartilage imaging: T2 relaxation time and diffusion-weighted MR imaging. Radiol Clin N Am 2005; 43: 641-653.
2. Michael PR, Douglas WG, Carl SW, Lawrence MW. MRI of articular cartilage: revisiting current status and future directions. AJR 2005; 185: 899-914.
3. David BS, Jeff M, Jean HB, Scott BR. High-resolution 3D cartilage imaging with IDEAL–SPGR at 3 T. AJR 2007; 189: 1510-1515.
4. Ding C, Cicuttini F, Scott F, Glisson M, Jones G. Sex differences in knee cartilage volume in adults: role of body and bone size, age and physical activity. Rheumatology 2003; 42: 1317-1323.
5. John AC, Mark ES. Imaging of sports-related knee injuries. Radiol Clin N Am 2002; 40: 181-202.
6. Hollis GP, Li FF. Magnetic resonance imaging of articular cartilage: trauma, degeneration, and repair. Am J Sports Med 2006; 34: 661-677.
7. Zhou GD, Wang XY, Miao CL, Liu TY, Zhu L, Liu DL, et al. Repairing porcine knee joint osteochondral defects at non-weight bearing area by autologous BMSC. Natl Med J China (Chin) 2004; 84: 925-931.
8. Wa J, Brian JC. Cartilage restoration, part 1: basic science, historical perspective, patient evaluation, and treatment options. Am J Sports Med 2005; 33: 295-306.
9. Anna IV, Jukka SJ, Lars P, Llkka K. Arthroscopic Cartilage indentation and cartilage lesions of anterior cruciate ligament-deficient knees. Am J Sports Med 2005; 33: 408-414.
10. Duc SR, Koch P, Schmid MR, Horger W, Hodler J, Pfirrmann CW. Diagnosis of articular cartilage abnormalities of the knee: prospective clinical evaluation of a 3D Water-Excitation True FISP Sequence. Radiology 2007; 243: 475-482.
11. Achel SO, Susan AC, Jenny TB, Diego J. Acute injury of the articular cartilage and subchondral bone: a common but unrecognized lesion in the immature knee. AJR 2004; 182: 111-117.
12. Hiroshi Y, Kathryn S, Mark G, Michael FD, Philipp L. Articular cartilage of knee: normal patterns at MR imaging that mimic disease in healthy subjects and patients with osteoarthritis. Radiology 2004; 231: 31-38.
13. David GD, Thomas RM, Carl RW. Detection of knee hyaline cartilage defects using fat-suppressed three-dimensional spoiled gradient-echo MR imaging: comparison with standard MR imaging and correlation with arthroscopy. AJR 1995; 165: 377-382.
14. Garry G, Thomas RM, Martha G, David GD. What′s new in cartilage? Radio Graphics 2003; 23: 1227-1242.
15. Kim S, Huh YM, Song HT, Lee SA, Lee JW, Lee JE, et al. Chronic tibiofibular syndesmosis injury of ankle: evaluation with contrast- enhanced fat-suppressed 3D Fast Spoiled Gradient-recalled acquisition in the steady state MR imaging. Radiology 2007; 242: 225-235.
16. Gold GE, Hargreaves BA, Vasanawala SS, Webb JD, Shimakawa AS, Brittain JH, et al. Articular cartilage of the knee: evaluation with fluctuating equilibrium MR imaging — Initial experience in healthy volunteers. Radiology 2006; 238: 712-718.
17. Bauer JS, Krause SJ, Ross CJ, Krug R, Carballido-Gamio J, Ozhinsky E, et al. Volumetric cartilage measurements of porcine knee at 1.5-T and 3.0-T MR imaging: evaluation of precision and accuracy. Radiology 2006; 241: 399-406.
18. David GD. Fat-suppressed three-dimensional spoiled gradient- recalled MR imaging: assessment of articular and physeal hyaline cartilage. AJR 1997; 169: 1117-1123.
19. Disler DG, Peters TL, Muscoreil SJ, Ratner LM, Wagle WA, Cousins JP, et al. Fat-suppressed spoiled GRASS imaging of knee hyaline cartilage: technique optimization and comparison with conventional MR imaging. AJR 1994; 163: 887-892.
20. Scott BR, Norbert JP, Marcus TA, Garry EG. Rapid MR. imaging of articular cartilage with steady-state free precession and multipoint fat–water separation. AJR 2003; 180: 357-362.
21. Shreyas SV, Brian AH, John MP, Dwight GN, Christopher FB, Garry EG. Rapid musculoskeletal MRI with phase-sensitive steady-state free precession: comparison with routine knee MRI. AJR 2005; 184: 1450-1455.
22. George R, Gregory JM, James LF, Evelyn EB, Paul TW, Robert ER, et al. MR-guided gatheter navigation of the intracranial subarachnoid space. Am J Neuroradiol 2003; 24: 626-629.
23. John IL, Heidi W, Robert JW, Matt AB, Colin LW. 3-T imaging of the cochlear nerve and labyrinth in cochlear-implant candidates: 3D fast recovery fast spin-echo versus 3D constructive interference in the steady state techniques. Am J Neuroradiol 2004; 25: 618-622.
24. Witte RJ, Lane JI, Driscoll CL, Lundy LB, Bernstein MA, Kotsenas AL, et al. Pediatric and adult cochlear implantation. Radiographics 2003; 23: 1185-1200.
25. Richard K, Paul S, Jason F, Arthur DS. Subchondral bone marrow edema in patients with degeneration of the articular cartilage of the knee joint. Radiology 2006; 238: 943-949.
|
PDF(982K)
Rapid Response |