MRI evaluation of knee cartilage injury: Comparison of 3D FIESTA with the fat suppressed 3DSPGR imaging

 

Madan Mohan Babu¹, Raja Kollu^2*, Seema U³, Charudutt Sambhaji⁴, Ramprakash H V⁵, Anees Dudekula⁶

 

Abstract              Background: To compare the sensitivity, specificity and accuracy of 3-dimentional fast imaging employing steady state acquisition (3DFIESTA) sequence with that of 3 - dimensional spoiled gradient echo (3DSPGR) sequence in the assessment of articular cartilage in patients with multiple ligament knee injury using arthroscopy as the reference standard. Materials and methods A total of 300 surfaces of 50 knees were evaluated by using sagittal 3D FIESTA and sagittal fat-suppressed 3D SPGR sequences. Articular cartilage lesions were graded on MRI and arthroscopy using a modified Noyes scoring system. Sensitivity, specificity, and accuracy were assessed. An interobserver agreement was determined with rho(r) values. Results: The performance of 3D FIESTA sequences, sensitivity, specificity, and accuracy were ranged from 54-80%, 88-98%, and 86-92%, respectively was similar to that of fat-suppressed 3D SPGR sequences, sensitivity, specificity, and accuracy were 62-98%, 90-98% and 88 to 98%, respectively in the detection of knee articular cartilage lesions. The interobserver agreement varied from good to excellent (rho values ranged from 0.63-0.99). Conclusion: 3D FIESTA has good diagnostic performance, comparable with fat-suppressed 3D SPGR in the evaluation of knee cartilage injury, and can be integrated into routine knee MRI protocols due to the short acquisition time.

Key words: MRI: Magnetic resonance imaging, 3D FIESTA:3- Dimensional fast imaging employing steady state acquisition, 3D SPGR:3-Dimensional spoiled gradient echo sequence, ACL: Autologous chondrocyte implantation.

 

INTRODUCTION

Detection and characterization of joint cartilage lesions have become increasingly important in routine work due to the high prevalence of osteoarthritis and post-traumatic cartilage abnormalities. Traumatic cartilage injuries of the knee are common in physically active individuals; however, the accurate numbers and localizations are not well known. Magnetic resonance imaging (MRI) of the knee joint is routinely used to detect traumatic cartilage injuries. MRI has good sensitivity and specificity in the examination of menisci and ligament injuries to the knee. However, for cartilage injuries, past literature (1,2,3) has produced varying results compared to arthroscopy, with a sensitivity ranging from 60 to 100%, a specificity ranging from 70 to 100 percent and an accuracy ranging from 81 to 98%. The integrity of its articular cartilage is the most crucial prerequisite for an adequate long-term joint function. Many studies (1,2,3,4) have shown that the prevalence of joint chondral trauma is high and cartilage damage is generally recognized as an early factor in the irreversible joint degeneration process. Recent developments in new surgical procedures and tissue engineering, such as autologous chondrocyte implantation (ACL) have had promising results in the formation of hyaline cartilage repairs. MRI is a widely accepted noninvasive technique for the assessment of cartilage lesions. A variety of MRI sequences for the assessment of articular cartilage have attracted considerable interest and have been the subject of numerous research studies in recent years(2,3). With conventional spin-echo sequences, sensitivities for the detection of hyaline cartilage lesions have been reported as low as 29 percent to 53 percent (2,3). Better results have been published using cartilage-specific sequences including T2 or intermediate weighted fast spin-echo sequences (4,5), 3-Dimensional spoiled gradient-echo sequence ( 3D SPGR), or 3- Dimensional fast imaging employing steady-state acquisition (3D FIESTA) sequences. The objective of the present study is to compare the sensitivity, specificity, and accuracy of 3-dimensional fast imaging employing steady-state acquisition (3DFIESTA) sequence with that of 3 - dimensional spoiled gradient echo (3DSPGR) sequence in the assessment of articular cartilage in patients with multiple ligament knee injury using arthroscopy as the reference standard.

 

MATERIALS AND METHODS

This is a prospective study conducted in the Department of Radiodiagnosis and Imaging, Kasturba Medical College, Manipal, from October 2010 to September 2012. The research procedure was accepted by our Institutional Ethics Board and informed consent were obtained from all patients before the MRI examination. The study was approved by the Ethics Committee of the institute, EC Reg NO: IEC423/2011 and taken permission from an appropriate authority. Informed consent was obtained from all patients before the MRI study. All patients with traumatic knee injury referred for MR evaluation were scanned on a 1.5T GE HDxt MRI system using a dedicated quadrature phased array knee coil with 3- dimensional fast imaging employing steady-state acquisition (3DFIESTA) and 3-dimensional spoiled gradient echo (3DSPGR) sequences. Imaging parameters for sequences performed during the MRI examination are listed in Table 1

Table 1: Imaging parameters for sequences performed during the MRI examination

Patients with a definite history of external trauma to knee and Knee with a multiligamental injury were included. Patients with Previous h/o knee surgery, Contraindications for MR imaging, Knee with isolated meniscal or cruciate ligament injury, and Non-availability of arthroscopy record were excluded from the study. 300 surfaces of 50 knees have been tested for cartilage injuries. The following six articular surfaces were tested for cartilage signal and morphology. Medial patellar facet, Lateral patellar facet, Medial femoral condyle, Medial tibial condyle, Lateral femoral condyle, and Lateral tibial condyles using a modified version of the Modified Noyes classification system (6,7) (Table 2) for articular cartilage defects. All patients in this study underwent conventional knee arthroscopy by an orthopedic surgeon who was blinded to the MRI grading results of cartilage lesions. All MRI scans were evaluated by a single senior radiologist who is blind to the arthroscopy results.

 

Table 2: Modified Noyes classification system for articular cartilage defects

All patients in this study underwent conventional knee arthroscopy by an orthopedic surgeon who was blinded to the MRI grading results of cartilage lesions. All MRI scans were evaluated by a single senior radiologist, who is blind to the arthroscopy results. Correlation of grades between arthroscopy and MR imaging for different articular surfaces were calculated by using spearman’s rho correlation (r). Spearman’s rho correlation ranges from -1 to + 1. -1 –Negative correlation,0 - No correlation, +1-Positive correlation. The more positive the r, the stronger the correlation.

RESULTS AND OBSERVATIONS

The study group consists of 50 subjects, of whom 38 had a history of RTA, 6 had a history of fall and the remaining 6 had a history of blunt trauma. All patients had been injured in the last six months before the MRI examination.43 (86%) were below 50 years of age and 17 (14%) were above 50 years of age (graph 1)

 

Graph 1: Age distribution of the subjects in the study

 

Grading and distribution of cartilage defects: A total of 300 surfaces of 50 knees were evaluated, of which 221(72%) were normal(Fig 1) surfaces and 79(27%) had cartilage defects.9(3%) were grade 1 lesions(Fig 2) 24(8%) were grade 2A(Fig3),10(3%)were grade 2B (Fig 4)and 36(12%) were grade 3(Fig 5) arthroscopic lesions. Chondral lesions were most often seen on the medial patellar surface and medial femoral surface (Graph 2). The distribution of grading of articular lesions for different articular surfaces was calculated and compared with arthroscopic grading. More than 85% of lesions showed a maximum difference of one grade between arthroscopy and MR imaging for different articular surfaces (Graph 3).

Because fat-suppressed 3DSPGR effectively suppresses all stationary tissue, it is not useful to test the ligaments, or soft tissues of the knee (2,14) and is not sensitive for marrow edema, which is often a sign of overlying cartilage defects (2). The main drawback of the 3DSPGR sequence in clinical application is its long scanning time. With increased scan time, the potential for motion artifact also increases, which is deleterious to the interpretation of the image. Another drawback of this sequence is the weak contrast between the cartilage and synovial fluid, which can reduce the conspicuousness of superficial cartilage lesions (15,16). More recently, SSFP imaging, with a relatively short period of acquisition, has been applied to experiments and clinical practice (17,18). This high SNR technique has also been referred to as "true fast imaging with SSFP' (TrueFISP), "balanced fast field echo' (Balanced FFE) and "fast imaging with the steady-state acquisition (FIESTA), which generates signal contrast based on the ratio of T2 to T1 in tissues (19).In 3D FIESTA imaging, the signal intensity of the cartilage is intermediate to low and that of the joint fluid is bright, which can provide an arthrographic effect that shows the subtle defects of articular cartilage(15).In the present study, cartilage lesions of the knee detected on 3D FIESTA images were confirmed on surgical examination with sensitivity ranging from 54-96%, the specificity of 88%, and accuracy of 90%, which is almost similar to that of 3D SPGR (Table 4). X. Li et al.. (15) reported that the performance of 3DFIESTA sequences (sensitivity, specificity, and accuracy were 80, 94, and 92 percent, respectively)was close to that of fat-suppressed 3D GSPR sequences (sensitivity, specificity, and accuracy were 82, 92, and 90 percent, respectively)for the identification of knee joint cartilage lesions. Interobserver agreements ranged from fair to good to strong (kappa values from 0.43 - 0.83).Hargreaves et al. (19) reported that SSFP sequences have a higher SNR and contrast between the cartilage and synovial fluid compared with SPGR. Duc et al. (17) found a sensitivity ranging from 52 to 65%, a specificity varying from 80 to 94%, and an accuracy from 71to 76% with 3D water-excitation True FISP sequences.

 

Table 4: Sensitivity, specificity, and accuracy of 3D FIESTA imaging among different studies.

 

Compared with 3D SPGR, 3D FIESTA MR imaging provides a contrast between synovial fluid and cartilage while preserving signal from the cartilage itself and maybe more useful in the detection of cartilage defects. In the present study, some superficial cartilage defects were undetectable on 3D FIESTA images, which may be partly due to a lack of fat suppression. We also noticed that 3D FIESTA imaging was more sensitive to bone edema or bruise in wounded knees compared to 3D SPGR, and this advantage allowed some contribution to the identification and assessment of cartilage lesion for proper grading. In the present study, the acquisition time of 3D FIESTA was 4 to 5 min, which is decreased significantly compared to 6.5 to 7 min of 3D SPGR. However, the use of this sequence in the diagnosis of meniscal, tendon, or ligament disease may not be as effective as FSE MR imaging and will be the focus of future studies. The main drawback of SSFP is the extreme banding artifacts caused by local inhomogeneity of the magnetic field when the TR is long. With the development of MRI gradient hardware, which allows lower TRs, the probability of banding artifacts caused by field inhomogeneity has decreased. In the present study, TR was kept minimum (4 to 6 ms) to reduce banding artifacts.Magnetic resonance imaging showed a relatively low sensitivity to the detection of chondral lesions of the lateral tibial plateau; partially due to the relatively small number of chondral lesions that were identified in this region in arthroscopy. It should be noted, that the Disler et al.. (3,4) also described the lateral tibial plateau as a difficult region to identify chondral lesions. This difficulty may also be due to the convex surface of the plateau, which, when subjected to sectioning into tomographic coronal and sagittal images, may cause more partial volume effects and imaging artifacts. Magnetic resonance imaging, on the other hand, showed a higher sensitivity (> 90%) to detect defects involving patellar facets and medial femoral and tibial condyles that are relatively thick and smooth, and this result was consistent with those of other studies (3,4,15).

 

Limitations

There were some drawbacks in the present study. The first was that the orthopaedists had access to the clinical MRI results, which was a possible source of bias in grading during arthroscopy. Second, the size of the sample was fairly small. Third, arthroscopy was an incomplete reference norm since it is based on the operator. Orthopedicians visualized only cartilage surfaces unless the subchondral bone was exposed. Fat saturation was not possible with the 3D FIESTA series, which affected the dynamic range of images. At present, more modern devices with higher field and gradient strengths have been commonly used in the clinical setups (20). Imaging time and image quality could also be enhanced by the updated SSFP sequences used in the recent study (21).

 

CONCLUSION

Both 3D FIESTA and three 3D gradient echo (3D SPGR) imaging had a similar performance in detecting the chondral lesions. In conclusion,3D FIESTA is a sequence with a good diagnostic performance that is comparable to the 3D SPGR sequence for the assessment of knee cartilage lesions. It is proposed that due to its short acquisition time, it to be incorporated into standard knee MRI protocols.

 

BIBLIOGRAPHY

1. Paavo-Ilari Kuikka, Martti J. Kiuru, Maria H. Niva et al.Sensitivity of Routine 1.0-Tesla Magnetic Resonance Imaging Versus Arthroscopy as Gold Standard in Fresh Traumatic Chondral Lesions of the Knee in Young Adults The Journal of Arthroscopic and Related Surgery, Vol 22, No 10 (October), 2006:1033-1039.
2. LI Xiao-ming, Peng Wen-jia, WU Hua, Kacher Daniel, XIA Li-ming, AIFei et al. MRI findings in injured articular cartilage of the knee correlated with surgical findings. Chin Med J 2009;122(21):2624-2630.
3. David G. Disler, Thomas R. McCauley, Chad, Kelman Mark D, Fuchs Lee M, Ratner Carl R Wirth. Fat-Suppressed Three-Dimensional Spoiled Gradient Echo MR Imaging of Hyaline Cartilage Defects in the Knee: Comparison with Standard MR Imaging and Arthroscopy AJR1996;61:121-132.
4. David G. Disler, Thomas R. McCauley Carl R. Wirth Marc D. Fuchs. 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:165, August 1995
5. Dr.Gonzalo Delgado P. Assessment of articular cartilage using magnetic resonance imaging. Magnetic Resonance Unit, Clinica Avansalud Providencia, Santiago, Chile.
6. Kijowski R, Blankenbaker DG, Davis KW, Shinki K, Kaplan LD, Smet AAD. Comparison of 1.5- and 3.0-T MR Imaging for Evaluating the Articular Cartilage of the Knee Joint1. Radiology. 2009 Mar 1;250(3):839–48.2.
7. Crema MD, Roemer FW, Marra MD, Burstein D, Gold GE, Eckstein F, et al. Articular Cartilage in the Knee: Current MR Imaging Techniques and Applications in Clinical Practice and Research. Radiographics. 2011 Jan 1;31(1):37–61.
8. John AC, Mark ES. Imaging of sports-related knee injuries. Radiol Clin N Am 2002; 40: 181-202.
9. 17Zhou 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.
10. 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.
11. Andrew H. Sonin, Raymond A. Pensy, Michael E. Mulligan Stephen Hatem. Grading Articular Cartilage of the Knee Using Fast Spin-Echo Proton Density–Weighted MR Imaging Without Fat Suppression. AJR 2002;179:1159–1166.
12. Hollis G. Potter, James M. Linklater, Answorth a. Allen, Jo A. Hannafin, and Steven b. Haas. Magnetic resonance imaging Of articular cartilage in the knee: An evaluation with use of fast-spin-echo imaging. The journal of bone and joint surgery september 1998 vol. 80-a, no. 9, 1276-1284.
13. Sonin AH, Pensy RA, Mulligan ME, Hatem S. Grading articular cartilage of the knee using fast spin-echo proton density weighted MR imaging without fat suppression. AJR Am J Roentgenol 2002;179:1159—66.
14. 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.
15. Li X, Yu C, Wu H, Daniel K, Hu D, Xia L, et al. Prospective comparison of 3D FIESTA versus fat-suppressed 3D SPGR MRI in evaluating knee cartilage lesions. Clin Radiol . 2009;64(10):1000–8.
16. Hargreaves BA, Gold GE, Beaulieu CF, et al. Comparison of new sequences for high-resolution cartilage imaging. MagnReson Med 2003;49:700-9.
17. Duc SR, Pfirrmann CW, Schmid MR, et al. Articular cartilage defects detected with 3D water-excitation true FISP: prospective comparison with sequences commonly used for knee imaging. Radiology 2007;245:216-23.
18. Kornaat PR, Reeder SB, Koo S, et al. MR imaging of articular cartilage at 1.5 T and 3.0 T: comparison of SPGR and SSFP sequences. Osteoarthr Cartil 2005;13:338-44.
19. Hargreaves BA, Gold GE, Beaulieu CF, et al. Comparison of new sequences for high-resolution cartilage imaging. Magn Reson Med 2003;49:700-9.
20. Von Engelhardt LV, Kraft CN, Pennekamp PH, et al. The evaluation of articular cartilage lesions of the knee with a 3-Tesla magnet. Arthroscopy 2007;23:496-502.
21. Duc SR, Pfirrmann CW, Koch PP, et al. Internal knee de- rangement assessed with 3-minute three-dimensional iso- voxel true FISP MR sequence: preliminary study. Radiology 2008;246:526-35.
    

 

 

 

 

 

Height, IVDL-Intervertebral Disc Length