| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1
Department of Radiology, University of Pittsburgh Medical Center, 200 Lothrop
St., Pittsburgh, PA 15213.
2
Present address: Mallinckrodt Institute of Radiology, 510 S. Kingshighway, St.
Louis, MO 63110.
3
Department of Orthopedic Surgery, Center for Sports Medicine and
Rehabilitation, University of Pittsburgh Medical Center, Baum Blvd. at Craig
St., Pittsburgh, PA 15213.
Received April 23, 1998;
accepted after revision September 16, 1999.
Address correspondence to D. A. Rubin.
Abstract
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
MATERIALS AND METHODS. Two musculoskeletal radiologists retrospectively reviewed the MR studies of 18 knees with arthroscopically proven treatable cartilage infractions, noting articular surface defects and associated subchondral bone edema; subchondral edema was defined as focal regions of high signal intensity in the bone immediately underlying an articular surface defect on a T2-weighted or short inversion time inversion recovery (STIR) image.
RESULTS. The first observer saw focal subchondral edema deep relative to a cartilage surface defect in 15 (83%) of the 18 cases; in two additional cases a surface defect was seen without underlying edema. The second observer identified 13 knees (72%) with surface defects and associated subchondral edema and three with chondral surface defects and no associated edema. Subchondral edema was seen more frequently on fat-suppressed images and on STIR images than non-fat-suppressed images.
CONCLUSION. Focal subchondral edema is commonly visible on MR images of treatable, traumatic cartilage defects in the knee; this MR finding may prove to be an important clue to assist in the detection of these traumatic chondral lesions.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
|
|
|
|
Chondral flaps and fractures represent a distinct and important subset of cartilage disorders in the knee [5,8,9,14]. These are traumatic injuries, often associated with ligament tears, and are usually produced by a twisting force during athletics or by a direct blow [1, 2, 5, 6, 14]. These injuries are very difficult to diagnose clinically because the history and physical findings frequently mimic those of meniscal tears [2, 5, 8, 14], which are more common. Radiographically, the lesions are occult, being composed entirely of articular cartilage, unlike osteochondral fractures and osteochondritis dissecans lesions [2,3,5,8,14]. Most important, specific operative treatment for cartilage fractures and flaps now exists [10], ranging from abrasion and microfracture of the subchondral bone [2, 4, 7, 14, 15] to biologic osteoarticular grafting and chondral transplantation [16,17,18,19]. Ideally, a non-invasive imaging technique would enable radiologists to distinguish these lesions from the more common meniscal tears because the treatment and prognosis of chondral fractures differs from that for meniscal tears.
In our daily MR imaging practice, we do not routinely use any cartilage-specific pulse sequences. Rather, we rely on a combination of spin-echo, fat-suppressed fast spin-echo, and short inversion time inversion recovery (STIR) images to enable diagnosis of meniscal, ligamentous, and musculotendinous abnormalities in the knee. Recognizing articular cartilage injuries on these images can be challenging. We have noticed that the bone immediately deep relative to a traumatic chondral defect often shows increased signal intensity on T2-weighted and STIR images, and we use this sign to alert us to critically examine the overlying cartilage surface for surface defects. We performed this study to further investigate the association of focal subchondral edema with treatable articular cartilage injuries.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
From October 1994 through September 1997, arthroscopy for a suspected internal derangement was performed on a total of 578 knees that had also undergone MR imaging. Eighteen of these patients (3%) met our inclusion criteria: 14 men and four women. Patients ranged in age from 18 to 59 years (mean, 31 years). Twelve of the patients (67%) were athletes (college, n = 7; recreational, n = 4; professional, n = 1). American football was the sport in which most patients (n = 6) participated; two patients participated in multiple sports, two played basketball, and one each participated in soccer and running. Eleven patients (61%) recalled specific inciting injuries. Knee pain was present in all but two patients. Eight patients complained of swelling and five had locking or "catching" of the knee. A meniscal tear was the initial suspected clinical diagnosis in 16 patients (89%). The referring physician questioned whether a cartilage injury was present in only two patients before MR imaging.
Arthroscopy was performed by one of four fellowship-trained sports medicine orthopedists, but two of the surgeons who had a special interest in the knee performed 15 (83%) of the 18 procedures. Three of the injuries were chondral flaps (Fig. 1A,1B,1C,1D,1E), and the remaining 15 were chondral fractures (separations). Lesions involved the medial femoral condyle in eight knees, the lateral femoral condyle in eight, and the femoral trochlea in two. All the lesions were of full thickness. The size of the chondral defect was not noted in the arthroscopy report of four knees. In the remaining 14 knees, the defect ranged from 3 x 4 mm to 30 x 30 mm. In nine patients the defect measured 1 cm or less in the largest diameter. The mean area of the measured defects was 2.0 cm2. Fifteen cases (medial femoral condyle, n = 8; lateral femoral condyle, n = 5; trochlea, n = 2) were treated with microfracture. In two lateral condyle lesions, the surgeon performed mosaicplasty, a technique involving multiple osteoarticular autografts [17]. A single lesion was believed to be potentially treatable but because of its small size (3 x 4 mm), the surgeon elected not to.
Six (38%) of the 16 cartilage injuries on the femoral condylar surfaces also had abnormalities of the associated meniscus, either a meniscal tear or a prior partial menisectomy. Other associated lesions found at surgery included tear of the anterior cruciate ligament (n = 5), medial collateral ligament tear (n = 1), cyclops lesion (n = 1) (arthrofibrosis after anterior cruciate ligament reconstruction), and meniscal tear in the compartment opposite the one involved with the cartilage injury (n = 1) (e.g., a lateral meniscal tear in a knee with a medial femoral condyle cartilage defect). In 10 (56%) of the 18 patients, the cartilage injuries were isolated arthroscopic lesions; that is, these 10 patients had no other lesion that required arthroscopy.
The time between MR imaging and subsequent arthroscopy varied from 5 to 93 days (median, 15 days). We did not routinely record the time interval between the injury (if any) and MR imaging. Fifteen patients underwent surgery within 6 weeks of the MR imaging study. Imaging was performed using a 1.5-T system (Signa; General Electric Medical Systems, Milwaukee, WI) and a whole-volume send-receive coil. All patients underwent a standard examination that we use to screen for common knee disorders. A transaxial fast spin-echo localizer (TR range/TEeff range, 2100-5250/38-80) was followed by a sagittal dual-echo conventional spin-echo sequence (2000-2950/15-17, 70-75). In addition, the first four patients were examined with a coronal fast spin-echo sequence (2167-2716/23-24) and a coronal fast inversion-recovery sequence (2666-3833/66-76; inversion time, 155 msec). The last 14 patients were examined with a coronal spin-echo sequence (483-800/14-18), a coronal fast spin-echo sequence with chemical fat suppression (3533-5133/80-96), and a sagittal fast inversion-recovery sequence (2900-3800/66-76; inversion time = 155-166 msec). The following additional scanning parameters were used: field of view, 14-16 cm; slice thickness, 3-4 mm (spin echo and fast spin echo) or 5-6 mm (inversion recovery); interslice gap, 1 mm; matrix size, 256 x 192-256; and number of excitations, 1-2. The echo train length was eight for the T2-weighted fast spin-echo sequences and four for the STIR sequences. Typical imaging times for the MR sequences were 22 min in the first four patients and 26 min for the final 14 patients, excluding the time necessary for setup, patient positioning, and prescanning.
Two radiologists with fellowship training in musculoskeletal imaging reviewed the MR examinations. Coronal and sagittal T2-weighted spin-echo and fast spin-echo sequences and STIR sequences were initially viewed separately, and then each examination was reviewed in its entirety. The two observers worked independently. This study was designed primarily to investigate the association between subchondral bone changes and treatable chondral defects; therefore, for each case the radiologists were told which surface showed the abnormality at surgery. The observers were otherwise unaware of the details of the arthroscopic findings (e.g., depth and diameter of the lesions, location of the defects along the articular surface, type of treatment used) and of the original prospective MR interpretation. For each pulse sequence, each observer noted whether a defect was visible within the articular surface. Subchondral edema was defined as increased signal within the trabecular bone immediately deep relative to a visible cartilage defect, compared with the remainder of the epiphyseal bone marrow, for each pulse sequence. Only areas of marrow edema adjacent to chondral defects were recorded.
Each observer established a final diagnostic category for each knee—no defect, defect without edema, or defect with edema—after viewing all the pulse sequences based on the images that showed the most prominent findings. Interrater agreement was assessed using the kappa statistic. The original MR imaging reports were reviewed to determine how frequently cartilage injuries had been initially suggested.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
=
0.71) and fast spin-echo images (86%; = 0.71). Concordance was lower
for the STIR images (72%; = 0.32).
|
|
|
|
|
|
When a final diagnosis was made using the information from al the pulse sequences combined, observer 1 identified subchondral edema associated with 15 of the lesions (83%) and observer 2 identified 13 articular defects (72%) with associated subchondral edema (Table 2). Despite being told which surface showed a cartilage injury at arthroscopy, neither observer could identify one lesion of the lateral femoral condyle, even in retrospect. This defect was the smallest in the study, measuring 3 x 4 mm, and was not treated at arthroscopy because of its small size. Observer 2 also failed to see one trochlear lesion, which was identified by observer 1 (Fig. 4A,4B). We expected that the surface defects would be easy to see on the MR images, given that they were all of full thickness and that the observers were told which surface was involved. However, both observers noted that the actual defects were easily overlooked and that subchondral edema was often recognized first and was used as a guide to identify the overlying surface abnormality. Because of the study design, the observers did not record the presence of subchondral edema if it was not associated with a visible cartilage defect; thus, we cannot comment on the specificity of this finding.
|
|
|
The three chondral flap lesions showed a characteristic appearance on the MR images, with an articular surface crater containing an in situ chondral fragment, which was attached at one end to the adjacent normal cartilage (Fig. 5A,5B). Although there was subchondral edema in the underlying bone in all three patients, no edema was seen in the detached chondral fragment.
|
|
Review of the prospective radiologic interpretations showed that the cartilage surface was prospectively called abnormal in 14 of the 18 cases, although a specific diagnosis of chondral flap or chondral fracture was not specified in every case. In one lesion of the lateral femoral condyle, the differential diagnosis of early avascular necrosis was offered. Three of the four defects not prospectively diagnosed were visible retrospectively by both observers. In two of these cases, the lack of subchondral edema may have initially decreased the conspicuity of the lesion (Fig. 6A,6B). Additionally, three of the four knees containing prospectively missed lesions were examined using our older imaging protocol, which did not include sagittal STIR images.
|
|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The pain and swelling that occur in patients with chondral injuries mimic the symptoms of a torn meniscus, which is frequently the initial clinical diagnosis [4,5,6, 8, 14]. Chondral injuries may coexist with meniscal lesions or may be isolated arthroscopic findings [6, 14], as was the case in 56% of the patients in our series. The initial clinical diagnosis was a meniscal tear in 89%; a chondral injury was considered in only two of the 18 knees before MR imaging. This trend was not surprising because isolated cartilage injuries are difficult to diagnose clinically [2] and are relatively infrequent compared with meniscal tears [8, 21, 22]. Although cartilage injuries can be shown on conventional arthrography [14], arthroscopy is currently the preferred method of diagnosis [5, 7, 9, 14, 22].
Surgical treatment for chondral injuries is evolving. Older, nonspecific interventions such as joint lavage and débridement have produced disappointing long-term results, with either incomplete healing or production of fibrocartilage, which lacks many of the mechanical properties of hyaline cartilage and eventually degenerates [10, 15, 23]. Specific treatment of focal cartilage defects now involves drilling, abrasion, or microfracture to perforate the tidemark and subchondral plate, inducing bleeding into the defect from the subchondral bone [2, 4, 7, 14, 15]. This procedure is followed by a period of non-weight-bearing and continuous passive motion, which enhances the ability of the reparative tissue to differentiate into hyalinelike cartilage [24]. New therapies include application of various biologic grafts such as periosteal and perichondral autografts [10], osteochondral allografts [16, 17], osteochondral autografts [19], and autologous cartilage transplants [18]. Indications for these procedures and their associated outcomes are currently being investigated, but the more aggressive grafting and transplantation techniques require significant preoperative planning [16,17,18,19], emphasizing the need for a reliable method of noninvasive diagnosis.
In our practice, we routinely use two-dimensional spin-echo, fat-suppressed fast spin-echo, and STIR imaging to reveal meniscal, ligamentous, and musculotendinous disorders in the knee. We found that we were able to retrospectively identify treatable chondral defects on these images. The cartilage injuries appeared as defects in the articular surface with vertical margins, usually sharply marginated from the adjacent normal cartilage, similar to their arthroscopic appearance [4, 6, 8]. Prospectively, even though most of the MR studies were requested to evaluate the knee for a potential meniscal tear, an articular cartilage abnormality was identified or suggested in 78% of the cases (14/18). This figure likely represents an overestimate of our prospective sensitivity because any abnormality noted in the articular surface was counted as positive; the radiologists who prospectively interpreted the MR studies did not, as a rule, try to distinguish traumatic chondral lesions from other disorders of the chondral surface. Speer et al. [21] studied the MR appearance of acute chondral injuries in the knee using conventional two-dimensional pulse sequences. Their sensitivity for detecting these lesions prospectively was 31% and retrospectively was 71%. However, Speer et al. did not use fat-suppression techniques to enhance the visualization of subchondral edema, which may have limited their ability to identify some of the lesions [25]. Although the prospective reports in our study were often not detailed enough to determine whether radiologists used subchondral edema as a clue in their evaluation of the chondral surfaces, we suspect that having worked with these sequences on a routine basis, the radiologists did use this information, either knowingly or unknowingly. This fact may in part explain the relatively high percentage of lesions found prospectively compared with the sensitivity reported by Speer et al. Additionally, our strict inclusion criteria—that all lesions had to be potentially treatable—may have biased the sample set to more obvious lesions than those included in prior studies.
In our study, subchondral edema was identified to be associated with chondral surface defects in 83% (observer 1) and 72% (observer 2) of the cases (Table 2). We postulate three possible mechanisms for the generation of this marrow edema: The injury to the subchondral bone can precede the articular cartilage injury, can occur at the same time as the cartilage injury, or can follow the cartilage injury. Support for the first possibility comes from animal studies of experimentally created chondral injuries that show injury to the subchondral trabeculae can precede injury to the overlying cartilage by several weeks [26,27,28,29,30]. Marrow edema shown on MR images is thought to reflect the initial injury to the subchondral bone in these cases [28].
Support for the second possibility is that the initial force responsible for the cartilage fracture produces a transient depression of the articular surface that is transmitted to the subchondral plate. In this instance, the subchondral marrow edema would represent a direct contusion, or true bone "bruise." Support for the third hypothesis is that the initial insult produces a cartilage defect large enough to expose the underlying bone to direct compression against the opposing articular surface once joint loading recommences [10]. Of course, several of the mechanisms proposed may work in concert to produce the MR finding of subchondral edema.
Our study is limited by its retrospective design in which observers were told which compartment contained the chondral injury, so they could specifically direct their attention to the underlying bone. We expected it would be fairly easy to identify the surface defects, knowing their location. Thus, we were surprised when both observers noted that it was frequently the signal changes in the subchondral bone that were first recognized and that these subchondral bone changes often directed the observers' attention to the surface defects in the overlying cartilage. We believe that a familiarity with the appearance of these lesions, especially the importance of subchondral edema beneath cartilage infractions, should increase the radiologist's ability to detect these lesions prospectively. Interestingly, a recent study using fat-suppressed T2-weighted fast spin-echo sequences to evaluate articular cartilage illustrated two knees in which subchondral edema was associated with degenerative chondrosis [31]. The authors of that study did not indicate how frequently they saw subchondral bone changes. Nor did they comment on whether subchondral edema assisted in the search for surface defects, but we suspect that subchondral edema may be useful in identifying some degenerative cartilage lesions as well as traumatic cartilage injuries.
A second limitation of our study is the small number of total lesions, in part because we used very strict inclusion criteria to maintain a uniform patient group. All knees had to have normal findings on radiographs. Additionally, for each lesion the arthroscopy report had to state that the chondral defect was at least potentially treatable. We were hindered by the variable terminology used by the arthroscopists, which meant that some potential cases may have been excluded because the arthroscopy report was not sufficiently detailed to determine whether a given lesion was degenerative or traumatic. The small number of cases also reflects the relative rarity of treatable cartilage defects [5, 8, 22].
Despite their relative infrequency, we contend that cartilage flaps and fractures form an important subset of knee disorders. These injuries clinically can mimic meniscal tears and are radiographically occult. Specific operative interventions for chondral injury have recently been introduced and are currently undergoing clinical evaluation. On routine MR images, chondral injuries appear as sharply demarcated defects in the articular surface. Characteristic subchondral edema revealed on fat-suppressed or STIR images may alert the radiologist to the presence of a defect in the overlying chondral surface that may have been otherwise overlooked.
Acknowledgments
We thank Eric Jablonowski for artistic assistance with Figure
1A,1B,1C,1D,1E.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  K.-P. Jungius, M. R. Schmid, M. Zanetti, J. Hodler, P. Koch, and C. W. A. Pfirrmann Cartilaginous Defects of the Femorotibial Joint: Accuracy of Coronal Short Inversion Time Inversion-Recovery MR Sequence Radiology, August 1, 2006; 240(2): 482 - 488. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. D. Kendell, C. A. Helms, J. W. Rampton, W. E. Garrett, and L. D. Higgins MRI Appearance of Chondral Delamination Injuries of the Knee Am. J. Roentgenol., May 1, 2005; 184(5): 1486 - 1489. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. Y. El-Khoury, K. J. Alliman, H. J. Lundberg, M. J. Rudert, T. D. Brown, and C. L. Saltzman Cartilage Thickness in Cadaveric Ankles: Measurement with Double-Contrast Multi-Detector Row CT Arthrography versus MR Imaging Radiology, December 1, 2004; 233(3): 768 - 773. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Brittberg and C. S. Winalski Evaluation of Cartilage Injuries and Repair J. Bone Joint Surg. Am., April 28, 2003; 85(90002): 58 - 69. [Full Text]
|
|
|
|
|
|
S. B. Reeder, N. J. Pelc, M. T. Alley, and G. E. Gold Rapid MR Imaging of Articular Cartilage with Steady-State Free Precession and Multipoint Fat-Water Separation Am. J. Roentgenol., February 1, 2003; 180(2): 357 - 362. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. A. Helms The Meniscus: Recent Advances in MR Imaging of the Knee Am. J. Roentgenol., November 1, 2002; 179(5): 1115 - 1122. [Full Text] [PDF]
|
|
|
|
|
|
D. A. Turner, D. A. Rubin, and J. M. Costello Subchondral Bone Marrow Edema in Degenerative Chondrosis Am. J. Roentgenol., December 1, 2000; 175 (6): 1749 - 1750. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH |
