JOURNAL OF MAGNETIC RESONANCE IMAGING 13:412 416 (2001) Original Research Evaluation of Chondromalacia of the Patella with Axial Inversion Recovery Fast Spin-Echo Imaging Sang Hoon Lee, MD, 1 Jin-Suck Suh, MD, 1 * Jaemin Cho, MD, 1 Seong Joon Kim, MD, 1 and Sung Jae Kim, MD 2 The purpose of our study was to assess the accuracy of inversion recovery fast spin-echo (IR-FSE) imaging for the evaluation of chondromalacia of the patella. Eightysix patients were included, they underwent magnetic resonance (MR) examination and subsequent knee arthroscopy. Medial and lateral facets of the patella were evaluated separately. Axial images were obtained by using IR-FSE (TR/TE/TI 3000/25/150 msec; echo train length, 8; 4-mm thickness; 12-cm field of view; 512 256 matrix; two, number of excitations) with a 1.5-T MR machine. MR interpretation of chondromalacia was made on the basis of the arthroscopic grading system. Of a total of 172 facets graded, arthroscopy revealed chondromalacia in 14 facets with various grades (G0, 158; G1, 1; G2, 3; G3, 6; G4, 4). Sensitivity, specificity, and accuracy in the chondromalacia grades were 57.1%, 93.0%, and 90.1%, respectively. There was one false-negative case (G4) and 11 false-positive cases (G1, eight; G2, two; G3, one). Sensitivity and specificity corrected by one grade difference were improved to 85.7% and 98.1%, respectively. When cartilage changes were grouped into early (corresponding to grade 1 and 2) and advanced (grade 3 and 4) diseases, sensitivity and specificity of the early and advanced diseases were 75% and 94% and 80% and 99%, respectively. IR-FSE imaging of the knee revealed high specificity but low sensitivity for the evaluation of chondromalacia of the patella. J. Magn. Reson. Imaging 2001;13:412 416. 2001 Wiley-Liss, Inc. CHONDROMALACIA OF THE PATELLA implies degeneration of the cartilage of the patella. It is known to be the most frequent cause of internal derangement of the knee (1). The signs and symptoms of chondromalacia of the patella are nonspecific, although some patients have a dull, aching discomfort that is well localized to the anterior part of the knee (2). In some patients this may be accompanied by patellar tracking abnormalities. 1 Department of Radiology, Research Institute of Radiological Science, Yonsei University College of Medicine, Seoul, Korea. 2 Department of Orthopedic Surgery, Yonsei University College of Medicine, Seoul, Korea. *Address reprint requests to: J-S.S., Department of Diagnostic Radiology, Yonsei University Colloge Medicine, #134 Shinchon-dong, Seodaemun-ku, Seoul 120-752, Korea. E-mail: jss@yumc.yonsei.ac.kr Received June 5, 2000; Accepted August 30, 2000. Many investigators have addressed chondromalacia of the patella by using various magnetic resonance (MR) imaging techniques, but inversion recovery (IR) has not been used because of its long acquisition time. However, since the development of the fast spin echo (FSE), the IR method combined with FSE can provide good contrast between the cartilage and the joint fluid, subchondral bone, and bone marrow with a reasonable acquisition time. Thus we evaluated the accuracy of IR-FSE in detecting cartilage lesions of the patella. MATERIALS AND METHODS From March 1997 to February 1998, 579 patients underwent MR examination because of suspected internal derangement of the knee. Eighty-six of them (age, 12 66 years) were included in this study, all of whom underwent MR examination and subsequent knee arthroscopy after 55 days (6 days to 5 months). MR imaging was performed with a 1.5-T imaging system (Signa, General Electric, Milwaukee, WI) and an extremity coil. Axial IR-FSE imaging ([TR/TE/TI 3000/25/150 msec]; echo train length, 8; 3-mm thickness with no interslice gap; 12-cm field of view; 512 256 matrix; number of excitations, two) was included in our routine knee MR protocol. Two musculoskeletal radiologists independently reviewed the MR images without knowledge of the arthroscopic diagnosis. The differences were corrected by consensus between two reviewers. Chondromalacia was defined to be present when cartilage showed focal high signal intensity compared with the adjacent cartilage. The MR interpretations were compared with arthroscopic results (Table 1). MR grading of chondromalacia was made based on the arthroscopic grading system (3): 0, normal; 1, softening or blister by high signal intensity and swelling of the cartilage at MR image; 2, fragmentation and fissuring in an area one-half inch or less in diameter; 3, fragmentation and fissuring in an area more than one-half inch in diameter; and 4, fullthickness fissure and exposed bone. Cartilage at the medial and lateral facets was evaluated separately. If a facet showed various grades at multiple areas, the facet was assigned the highest appropriate grade. 2001 Wiley-Liss, Inc. 412
IR-FSE Imaging for Chondromalacia of the Patella 413 total of 172 facets evaluated, arthroscopy revealed chondromalacia in 14 facets with various grades (G0, 158; G1, 1; G2, 3; G3, 6; G4, 4). Grade 1 chondromalacia showed a focal globular high signal, whereas chondromalacia grade 2 to grade 4 showed a fissurelike high signal (Figs. 2 and 3). Sensitivity, specificity, and accuracy of IR-FSE imaging in terms of making a correct grading of the chondromalacia were 57.1%, 93.0%, and 90.1%, respectively. There was one falsenegative case (G4) and 11 false-positive cases with normal arthroscopy (G1, eight; G2, two; G3, one; Fig. 4). Sensitivity and specificity corrected by one grade improved to 85.7% and 98.1%, respectively. When cartilage changes were grouped into early (corresponding to grades 1 and 2) and advanced (grades 3 and 4) disease, sensitivity and specificity of early disease were 75% and 94%, while for advanced disease they were 80% and 99%, respectively (Table 2). Figure 1. Normal appearance of patellar cartilages. A 23-yearold man with an acute knee injury. An axial inversion recovery FSE image (TR/TE/TI 3000/25/150 msec, 512 256 matrix, NEX 2) shows articular cartilage with intermediate signal intensity. No focally high signal intensity, fissure, or thinning of the cartilage is apparent. RESULTS On IR-FSE images, cartilage appeared to have an intermediate signal compared with the bright joint fluid and the dark subchondral cortex and fat-suppressed marrow, and the signal intensity gradually decreased from the superficial to the deep zone of cartilage (Fig. 1). Of a DISCUSSION The term chondromalacia, which means softening of the articular cartilage, has been used synonymously with disorders of the patella since 1924 (4). Chondromalacia probably should be considered as a descriptive term rather than a diagnosis. Indeed there are softening changes in articular cartilage during the earliest stages of osteoarthritis, but transient, localized, posttraumatic, and precocious soft cartilage lesions do occur that are unrelated to arthritis or joint degeneration. Such lesions differ in etiology and natural history, but all are characterized, at least in the early stages, as chondromalacia (4). In the first stage, as described by Shahriaree (4), the disruption of collagen fibers leads to cartilage softening. Figure 2. A: Grade 1 chondromalacia patellae in a 21-year-old man with chronic knee pain. The axial inversion recovery FSE image (TR/TE/TI 3000/25/150 msec, 512 256 matrix, NEX 2) shows a lesion with focal high signal intensity at the medial facet (arrow), which proved to be due to a softening of the cartilage (grade 1) at arthroscopy. B: Grade 2 chondromalacia patellae in a 22-year-old man with a previous knee injury. The axial inversion recovery FSE image shows disruption of superficial cartilage line and high-signal-intensity lesion of the medial facet, but the fissure did not reach one-half inch in diameter (arrow), which was confirmed as a superficial fissure by arthrography (grade 2).
414 Lee et al. Figure 3. A: A grade 3 chondromalacia patellae in a 63-year-old woman with chronic knee pain. The axial inversion recovery-fse image (TR/TE/TI 3000/25/150 msec, 512 256 matrix, NEX 2) shows a fissure extending more than half the thickness of the cartilage at the medial facet (arrow), which was confirmed as grade 3 chondromalacia at arthroscopy. B: A grade 4 chondromalacia patellae in a 61-year-old man with chronic knee pain. The axial inversion recovery FSE image shows a deep fissure down to the subchondral bone (arrowhead) at the lateral facet, again proven arthroscopically. As the stresses continue, further disruption of collagen fibers occurs, leading to separation of the more superficial layers of cartilage from the deeper layers, and producing a blister that may contain gelatinous material. When the blister ruptures, surface ulceration and fragmentation are produced (5). Hyaline cartilage is histologically divided into four arbitrary zones that are defined by both microanatomic and ultrastructural features (6): the superficial (tangential) zone, where the collagen fibers are horizontally arranged parallel to the articular surface; the intermediate (transitional) zone, where collagen fibers are oriented obliquely to the articular surface; the deep (radial) zone, in which the collagen fibers run vertically toward the surface; and the calcified cartilage zone, located between the radial zone and the subchondral bone. The correlation between signal intensity and microscopic features of the hyaline cartilage has been investigated but not fully established (6 8). In this study, the signal intensity of the patellar cartilage in IR-FSE was intermediate (i.e., between the high signal intensity of the joint fluid and the low signal intensity of the marrow), and it gradually decreased when measured on moving from the surface to the deeper zones. When short tau inversion recovery (STIR) sequence is used for musculoskeletal imaging, it can provide optimal contrast for tumor detection (9), and in the case of chondromalacia, it produces an elevated signal intensity because of the increase in both the T1 and T2 relaxation times of the chondromalacic lesions (10). The STIR sequence has the advantage of yielding homogeneous suppression of the fat signal in contrast to the frequency-selective fat-suppression pulse sequence. This is because the STIR sequence is less dependent on B0 inhomogeneity for effective fat suppression (11 13), but STIR imaging does have some limitations: the signal-to-noise ratio is not sufficiently high to generate high spatial resolution images, a long imaging time is required, and fewer sections are obtainable (9). Concerning the long imaging time, it was found that it could be reduced to less than one fourth of the imaging time required for the conventional STIR sequence by using IR-FSE as a combination pulse sequence of STIR and FSE, and image quality was comparable with that of the STIR sequence (14). In FSE, a relatively long effective TE is generally chosen to minimize both image blurring and the loss of small-object contrast (15 17). However, we decided to choose short TE in the case of IR-FSE to improve the signal-to-noise ratio, and this did not cause any problems in terms of MR image interpretation. IR-FSE images are often thought to have an undesirably high level of tissue contrast, rendering some injuries more conspicuous (18). In this study, it is conceivable that a high false-positive rate resulted from an exaggerated signal-intensity change in the early stage of chondromalacia. As has been illustrated, magic angle and partial volume effects also are possible causes of false-positive results (6). Axial, chemical saturated fat-suppression MR imaging has high specificity in excluding chondromalacia (19), but it also has intermediate sensitivity (72%). Similar sensitivity of early and advanced chondromalacia, 74% and 85%, respectively, was demonstrated by another FS-FSE study (20). We also found that sensitivity had increased to 75% and that specificity was similar after being corrected for early and advanced disease. With regard to sensitivity and specificity, similar results have been demonstrated for IR-FSE and fat-suppressed (FS)-FSE, but it has not been established which one is
IR-FSE Imaging for Chondromalacia of the Patella 415 Figure 4. False-positive cases. A: Axial IR-FSE image (TR/TE/TI 3000/25/150 msec, 512 256 matrix, NEX 2) shows a focal lesion with high signal intensity (grade 1) at the medial facet (arrow) in a 23-year-old man with chronic knee trauma. B: Superficial fissure is seen at the medial facet (arrowhead, grade 2) in a 64-year-old woman with chronic knee pain. C: Deep fissure is suspected at the ridge separating the medial and odd facet (double arrow, grade 3) in a 45-year-old woman with previous knee trauma. the better, because direct comparison was not done. Recht et al (21) reported high sensitivity (96%) and specificity (95%) of an FS SPGR image of chondromalacia, but their results did not include grade details. Table 1 Comparison of MR and Arthroscopic Results MR grade Arthroscopic grade 0 1 2 3 4 Total 0 147 0 0 0 1 148 1 8 1 1 0 0 10 2 2 0 1 1 0 4 3 1 0 0 3 0 4 4 0 0 1 2 3 6 Total 158 1 3 6 4 172 MR arthrography was found to be more sensitive for the detection of chondromalacia than T1-weighted, proton density-weighted, and SPGR with fat saturation MR imaging, but it was not significantly better than T2- weighted imaging (22). It also is an invasive and more expensive procedure. Table 2 Sensitivity and Specificity of Early and Advanced Grades Grade Sensitivity (%) Specificity (%) Accuracy (%) Early (grade 1, 2) 75 (3/4) 94 (158/168) 94 (161/172) Advanced 80 (8/10) 99 (160/162) 98 (168/172) (grade 3, 4)
416 Lee et al. Table 3 Comparison of Both Facets Medial Lateral MR 18 6 Arthroscopy 10 4 Sensitivity (%) 50 75 False positive 9 2 False negative 1 0 This study used arthroscopy as a standard reference. The arthroscopic appearance of healthy cartilage is bluish white, smooth, glistening, and resilient. The earliest change in damaged cartilage is softening. Probing the affected area and comparing it with the surrounding articular cartilage is the best way of finding this anomaly. By pressing a probe gently over the damaged site, a soft indentation can be felt (4), but the accuracy of arthroscopy as the standard of reference in MR evaluation of early cartilaginous disorders has been questioned (23). In particular, softened cartilage may be overlooked. For stage I chondromalacia, the interobserver variability of the arthroscopic evaluation may be very high because the evaluation of cartilaginous softening can be subjective (19). Consequently, in falsepositive cases with a focal signal change (MR grade 1 chondromalacia), grade 1 chondromalacia is still suspected. One case was false negative with grade 4 by arthroscopy at the medial facet, but the IR-FSE image was not evaluated properly because of motion artifact. Eleven cases were false positive (Table 3); nine of them were located at the medial facet, and the remaining two were at the lateral facet. In young patients, softening or blistering typically tends to occur at the medial facet (8). The most common site for grade 1 lesions is along the ridge that separates the medial and odd facet; grade 3 and 4 lesions are more common on the lateral facet (24). In our study, all grade 1 and 2 cases proven during arthroscopy were found at the medial facet, as were most false-positive cases. Our results imply that very careful examinations should be performed when an early stage of chondromalacia is suspected in the medial facet. In summary, IR-FSE imaging of the knee revealed high specificity but low sensitivity for the evaluation of chondromalacia of the patella. REFERENCES 1. Ficat RP, Philippe J, Hungerford DS. Chondromalacia patellae: a system of classification. Clin Orthop 1979;144:55 62. 2. Canale ST. Campbell s operative orthopaedics. ninth ed. St. Louis: Mosby, 1996. p 1273 1279. 3. Outerbridge RE. The etiology of chondromalacia patellae. J Bone Joint Surg 1961;43:752 757. 4. Shahriaree H. Chondromalacia. Contemp Orthop 1985;11:27 39. 5. Yulish BS, Montanez J, Goodfellow DB, Bryan PJ, Mulopulos GP, Modic MT. Chondromalacia patellae: assessment with MR imaging. Radiology 1987;164:763 766. 6. Rubenstein JD, Kim JK, Morava-Protzner I, Stanchev PL, Henkelman RM. Effects of collagen orientation on MR imaging characteristics of bovine articular cartilage. Radiology 1993;188:219 226. 7. Paul PK, Jasani MK, Sebok D, Rakhit A, Dunton AW, Douglas FL. Variation in MR signal intensity across normal human knee cartilage. J Magn Reson Imaging 1993;3:569 574. 8. Waldschmidt JG, Rilling RJ, Kajacsy-Balla AA, Boynton MD, Erickson SJ. In vitro and in vivo MR imaging of hyaline cartilage: zonal anatomy, imaging pitfalls, and pathologic conditions. Radiographics 1997;17:1387 1402. 9. Shuman WP, Patten RM, Baron RL, Liddell RM, Conrad EU, Richardson ML. Comparison of STIR and spin-echo MR imaging at 1.5 T in 45 suspected extremity tumors: lesion conspicuity and extent. Radiology 1991;179:247 252. 10. Smith RC, Constable T, Reinhold C, McCauley T, Lange RC, Mc- Carthy S. Fast spin echo STIR imaging. J Comput Assist Tomogr 1994;18:209 213. 11. Weisskoff RM, Edelman RR. Clinical magnetic resonance imaging, 2nd ed. Philadelphia: Saunders, 1996. p 35 36. 12. Pui MH, Chang SK. Comparison of inversion recovery fast spinecho (FSE) with T2-weighted fat-saturated FSE and T1-weighted MR imaging in bone marrow lesion detection. Skeletal Radiol 1996; 25:149 152. 13. Hilfiker P, Zanetti M, Debatin JF, Mckinnon G, Hodler J. Fast spin-echo inversion-recovery imaging versus fast T2-weighted spin-echo imaging in bone marrow abnormalities. Invest Radiol 1995;30:110 114. 14. Weinberger E, Shaw DWW, White KS, et al. Nontraumatic pediatric musculoskeletal MR imaging: comparison of conventional and fastspin-echo short inversion time inversion-recovery technique. Radiology 1995;194:721 726. 15. Panush D, Fulbright R, Sze G, Smith RC, Constable RT. Inversionrecovery fast spin-echo MR imaging: efficacy in the evaluation of head and neck lesions. Radiology 1993;187:421 426. 16. Constable RT, Anderson AW, Zhong J, Gore JC. Factors influencing contrast in fast spin-echo MR imaging. Magn Res Imaging 1992; 10:497 511. 17. Constable RT, Gore JC. The loss of small objects in variable TE imaging: implications for FSE, RARE, and EPI. Magn Res Med 1992;28:9 24. 18. Arndt WF III, Truax AL, Barnett FM, Simmons GE, Brown DC. MR diagnosis of bone contusions of the knee: comparison of coronal T2-weighted fast spin-echo with fat saturation and fast spin-echo STIR images with conventional STIR images. AJR Am J Roentgenol 1996;166:119 124. 19. Rose PM, Demlow TA, Szumowski J, Quinn SF. Chondromalacia patellae: fat-suppressed MR imaging. Radiology 1994;193:437 440. 20. Bredella MA, Tirman PFJ, Peterfy CG, et al. Accuracy of T2- weighted fast spin-echo MR imaging with fat saturation in detecting cartilage defects in the knee: comparison with arthroscopy in 130 patients. AJR Am J Roentgenol 1999;172:1073 1080. 21. Recht MP, Kramer J, Marcelis S, et al. Abnormalities of articular cartilage in the knee: analysis of available MR techniques. Radiology 1993;187:473 478. 22. Gagliardi JA, Chung EM, Chandnani VP, et al. Detection and staging of chondromalacia patellae: relative efficacies of conventional MR imaging, MR arthrography, and CT arthrography. AJR Am J Roentgenol 1994;163:629 636. 23. Hodler J, Resnick D. Chondromalacia patellae (commentary). AJR Am J Roentgenol 1992;158:106 107. 24. Hayes CW, Sawyer RW, Conway WF. Patellar cartilage lesions: in vitro detection and staging with MR imaging and pathologic correlation. Radiology 1990;176:479 483.