**2.1 Conventional radiographic technique**

Conventional radiographic techniques have proved to be of limited value in the imaging of articular cartilage as such techniques only allows the indirect assessment of cartilage (Lund, 1980; cited by Recht et al, 1993). Plain radiographs, in general, significantly underestimate the extent of cartilage damage (Blackburn et al, 1994). However, plain X-ray may reveal osteochondral lesions, including osteochondritis dissecans and loose bodies (Morelli et al, 2002).

Scintigraphy and computerized tomographic evaluations are limited because of their lack of sensitivity and requirement for ionizing radiation (Blackburn et al, 1994). Computed tomography combined with arthrography improves both the visualization of cartilage and the detection of abnormalities, but this method is relatively insensitive in the delineation of small cartilage lesions (Handelberg et al, 1990). Bone scans may indicate osteochondral injuries, but is not specific and does not necessarily indicate pure chondral lesions or their size (Morelli, et al. 2002).

#### **2.2 Arthroscopy**

188 Modern Arthroscopy

Fig. 5. ICRS classification (www.cartilage.org)

The most accurate diagnostic modality for traumatic knee articular cartilage injury is arthroscopy (DeHaven, 1980). However, even with this "Gold Standard" modality the posterior tibial and femoral lesions can be difficult to identify and may be missed (Terry, 1988; cited by Speer 1991). Although arthroscopy treatment can be performed on a chondral fracture discovered unexpectedly, it would be advantageous to know before arthroscopy whether a chondral injury is present. The surgeon then could advise the patient before surgery about treatment options and expected outcome, and decide on the type and timing of surgery (Rubin, 1998).

#### **2.3 Magnetic resonance imaging**

The MR imaging appearance of chondral fractures is analogous to their arthroscopic appearance (Rubin et al, 1997). Chondral separations manifest as a defect in the articular surface extending down to the subchondral plate, with vertically oriented walls and sharp demarcation from the surrounding cartilage. When a flap is present a fragment will be seen, attached on one side (Rubin et al, 1997). Lesion conspicuity can be increased by performing MR arthrography, especially if the patient is examined some time after the acute insult (Rubin, 1998). The presence of a joint effusion could alleviate the need for iatrogenic introduction of intra-articular contrast agents. The theory is that this would offer an

Traumatic Chondral Lesions of the Knee Diagnosis and Treatment 191

Fat-suppressed 3-D SPGR imagining has several advantages over conventional sequences. It generates positive contrast between cartilage and adjustment structures, making joint

Fat-suppression maximizes the contrast between cartilage and adjacent marrow, an improvement over T2–weighted spin-echo imagining, and minimizes chemical- shift artifact. Three-dimensional acquisition can generate very thin slices without loss of

In summary, fat-suppressed 3-D SPGR imagining of the knee provides a striking positive contrast between hyaline cartilage and adjacent structures, and may improve the accuracy of

The sensitivity of fat-suppressed 3-D SPGR imaging was compared with that of standard MR imaging for detecting hyaline cartilage defects of the knee, using arthroscopy as the standard of reference. Disler et al (1994) assessed 114 consecutive patients for hyaline cartilage defects of the knee with both standard MR imaging sequences and a sagittal fatsuppressed 3-D SPGR sequences. Forty eight patients with meniscal or ligament injury or persistent symptoms underwent subsequent arthroscopy. The standard MR and SPGR images of these 48 patients were then retrospectively analyzed for articular defects in a blinded fashion by two independent observers. Sensitivity, specificity, and intra-observer and inter-observer agreement were determined for the different imaging techniques. A quarter of the patients who went on to arthroscopy were found to have isolated hyaline cartilage lesions. The SPGR imaging sequences had a significantly higher sensitivity than the standard MR imaging sequences for detecting hyaline cartilage defects (75 to 85 % and 29 to 38 % respectively, p < 0.001, for each component). Significant differences in sensitivity were found for each surface except the trochlear and lateral tibial surfaces. No difference in specificity were found (97 % and 97 % respectively, p > 0. 99). Combined evaluation of standard MR and SPGR images gave no added diagnostic advantage (sensitivity 86 %; specificity 97 %; p > 0.42). Except for the lateral tibial surface, reproducibility among readings and between readers was excellent. The conclusion from the study was that fatsuppressed 3-D SPGR imaging is more sensitive than standard MR imaging for the detection of hyaline cartilage defects of the knee (Disler et al, 1996). In day-to-day practice a routine clinical MRI scan has low sensitivity in diagnosing chondral damage when

Levy et al (1996) reported that preoperative MRI scans correctly identified 21 percent of the chondral lesions seen at arthroscopic examination. However, since 1996 the new awareness of the significance of chondral problems, due to extensive laboratory and clinical research, and various attempts to repair hyaline articular surface, has resulted in an increased interest

 Development of refined MRI techniques and recent advantages in MRI technology appear to be very promising. Magnetic resonance imaging has the potential to replace the more conventional invasive techniques, like arthroscopy and biopsy, in the evaluation of articular

Rubin et al (2000) retrospectively reviewed the MR studies of 18 knees with arthroscopically proven acute articular cartilage defects, noting the associated subchondral oedema.

in magnetic resonance imaging as a diagnostic and evaluation tool (Bobic, 2005).

infusion unnecessary to show the cartilage margin (Kornaat et al, 2005).

MR diagnosis of hyaline cartilage abnormalities (Disler et al, 1994).

compared with arthroscopic findings (Bobic, 2005).

cartilage damage and repair (Bobic, 2005).

**2.8 Significant of focal subchondral oedema** 

**2.7 Advantages of 3-D SPGR** 

information.

"arthrogram" effect, and allow indirect visualization of chondral lesions as well (Beltran, 1980; Spritzer 1988; cited by Speer 1991).

In a comparison study between MR and anatomic section Hodler et al (1992) concluded that standard MRI does not consistently allow detection of focal articular cartilage defects. Commonly used MRI sequences are not reliable enough to be effective in the diagnostic evaluation of degenerative changes of articular cartilage.

#### **2.4 Can MRI replace arthroscopy in diagnosis?**

The role of MRI for the diagnosis of chondral lesions of the knee joint is still unclear, and the sensitivity of the method ranges from 15 to 96 percent (Friemert et al, 2003). In a prospective study by Friemert et al (2003) of how MRI can replace arthroscopy in the routine diagnosis of cartilage damage, they found that the MRI cartilage specific sequences have a sensitivity of 33 percent and specificity of 99 percent and positive and negative prediction values of 75 and 98 percent respectively. With gadolinium enhanced MRI the sensitivity was 53 percent and the specificity was 98 percent. The positive prediction value was 48 percent and the negative prediction value was 98 percent. They concluded that the MRI examination techniques recommended in the literature are not able to replace arthroscopy for the diagnosis of cartilage damage of the knee joint, and in view of the high specificity (97 to 98 %) MRI is suitable for identifying cartilage lesions. In view of the low sensitivity of MRI to cartilage injury, a cautious attitude towards an operative cartilage treatment is justified. Because that MRI can not replace arthroscopy for the diagnosis of cartilage damage and so arthroscopy still has to be seen as the method of choice for the evaluation of cartilage damage (Friemert et al, 2003).

#### **2.5 Best MRI sequences to visualize articular cartilage**

Currently the most widely used techniques for articular cartilage imagining by MR are fat suppressed proton-density weighted fast spin-echo sequences, and fat suppressed spoiled gradient recalled echo (SPGR) sequences (Kornaat et al, 2005). SPGR sequences are often chosen for cartilage volume and thickness estimation because the 3D acquisition, along with higher intensity cartilage signal, provides robust visualization of cartilage and detection of cartilage pathology. However, new MR imaging pulse sequences, specifically steady-state free precession (SSFP), have recently attracted attention for their optimal visualization of cartilage. The new sequences give greater cartilage intensity, increased cartilage and contrast-to-noise ratio and reduced imaging time than conventional pulse sequences (Kornaat et al, 2005).

#### **2.6 Comparison between 3-D SPGR and conventional MR imaging**

Fat suppressed 3-D spoiled gradient recalled acquisition in the steady state (SPGR) MRI technique was compared with 2-D SPGR images and conventional T1 and T2 weighted spinecho and multiplanar by Disler et al (1994). They studied ten healthy volunteers and concluded that fat-suppressed 3-D SPGR imagining is an improvement over fat-suppressed spin-echo imagining because the fluid signal is diminished relative to cartilage. As the sequence essentially suppresses all stationary tissue, it is not useful in evaluating the fibrocartilage, ligaments or soft tissues of the knee. However the technique shows cartilage as an object of high signal intensity relative to adjacent tissues, giving the technique great potential for evaluating this structure.
