Diagnosis of Osteochondral Lesions by MRI

Fig. 3.1

Comparison of image quality using 1.5 and 3 T MRI. Osteochondral lesion at the talar dome in both images (fat saturated intermediate-weighted fast spin echo sequence). The 3T image (b) shows better delineation of the cartilage, more detail and is less blurry than the 1.5T image (a). Differences are due to the higher signal-to-noise ratio at 3T

In addition to the hardware, the choice of adequate imaging sequences is critical. Usually spin-echo sequences are used; these include fluid-sensitive intermediate-weighted fast spin-echo sequences as well as non-fat-saturated T1-weighted and proton-density-weighted sequences. Fat-saturated intermediate-weighted fast spin-echo sequences provide information on the cartilage layer, the bone marrow, the tendons, and the ligaments at the same time. The advantage of fat saturation includes better visualization of the bone marrow edema pattern and less chemical shift artifacts at the interface between the cartilage and bone marrow. The workhorse sequences are 2D fast spin-echo sequences and they are usually the main part of a standard routine imaging protocol [19, 21]. Table 3.1 shows representative sequences used for clinical imaging of the ankle at 1.5 and 3.0 T.

Table 3.1

Standard clinical sequences and sequence parameters for ankle imaging

Sequence	Field strength	TR (ms)	TE (ms)	Flip angle	NEX	ETL	Matrix (pixels)	FOV (cm)	BW (kHz)	ST (mm)
axT1	3.0 T	675	15.7	90	2	5	384 × 256	12	31.25	3
axT1	1.5 T	600	10	90	2	3	256 × 192	12	31.25	3
axT2	3.0 T	4,500	42	90	2	16	512 × 256	12	31.25	3
axT2	1.5 T	4,000	40	90	2	12	320 × 224	12	16.67	3
sagT1	3.0 T	675	15.4	90	2	4	384 × 256	12	31.25	3
sagT1	1.5 T	625	23.5	90	2	4	384 × 224	12	16.67	3
sagIR	3.0 T	3,700	68	90	2	15	320 × 160	12	31.25	3
sagIR	1.5 T	3,400	68	90	2	8	256 × 192	12	16.67	3
corIM	3.0 T	4,000	16.7	90	4	9	384 × 256	10 × 8	31.25	2
corIM	1.5 T	4,000	15.5	90	3	12	384 × 224	10 × 8	16.67	2

In addition, thin section 3D sequences have been introduced to allow for better visualization of the cartilage layer. Among these, 3D fast spin-echo sequences have been found to be particularly useful [12, 28, 29] (Fig. 3.2). Using 3D fast spin-echo sequences provides isotropic datasets of the ankle, which can be reconstructed in any imaging plane, e.g., from a sagittal source image dataset, coronal and axial sequences can be generated. The advantage over standard 2D fast spin-echo sequences is the decrease of partial volume effects, allowing better depiction of subtle cartilage defects. A number of other 3D sequences based on gradient echoes have also been developed, such as balanced steady-state free precession (bSSFP), iterative decomposition of water and fat with echo asymmetry, and least-squares estimation combined with spoiled gradient echo (IDEAL-SPGR) and multiecho in steady-state acquisition (MENSA) sequences. A recent study, however, found that 3D fast spin-echo sequences may be superior to those in visualizing cartilage and associated bone marrow changes [7].

Fig. 3.2

Standard fat saturated intermediate-weighted fast spin echo sequence (a) and thin Section 3 D fast spin echo CUBE sequence (b). Note higher detail in the CUBE sequence, which better depicts full thickness cartilage defect at the medial talar dome (arrows)

Short-tau inversion recovery (STIR) sequences have also been used at the ankle as they are very fluid sensitive and provide excellent depiction of bone marrow abnormalities. In addition, they reduce magic angle effects, thus optimizing evaluation of the ankle tendons [31]. Contrast media are usually not required for imaging of the ankle but have been suggested previously to improve evaluation of the viability of osteochondral lesions and osteochondral autograft transfer systems [18].

3.3 MR Imaging Findings in Osteochondral Lesions

Common etiologies for osteochondral lesions of the talus are acute or chronic intra-articular injuries, and most frequently they are related to sports injuries. MRI is usually performed after an ankle sprain, which does not improve over time or if locking or catching occurs. Standard radiographs not infrequently are normal at the time of the injury, and they may also be negative on subsequent studies. Radiographic findings, which are suspicious for osteochondral injury, may be subchondral lucency or a small fracture fragment. CT and MRI are second-line imaging techniques. While CT has a high spatial resolution and is excellent for identifying small bony lesions, MRI has the advantage of directly visualizing cartilage and of identifying bone bruises and microfractures, which may not be visualized with CT. MRI provides information on cartilage defects and bone marrow abnormalities, but because of the limited cartilage thickness, MRI is challenging and the MRI technique needs to be adequately chosen as outlined above.

The initial classification of osteochondral lesions was based on radiographs and developed by Berndt and Harty in 1959 [3]. This is still widely used, and additional MRI-based classifications have been developed [9, 22, 32]. The original Berndt and Harty Stage I represents an area of osteochondral compression, Stage II a partially loose fragment, Stage III a completely detached fragment without displacement, and Stage IV a completely detached and displaced fragment. A grade 0 has been added, which is an x-ray-negative but MRI-positive lesion [4]. Scranton and others have added a Stage V to describe lesions with deep cystic changes [30].

In 2003, Mintz et al. proposed an MRI grading system of osteochondral lesions [22], which represents a modification of the arthroscopic grading system of the ankle proposed by Cheng et al. [8]. This system differentiates 6 grades: grade 0 is normal; grade 1 represents a hyperintense but morphologically intact cartilage surface (Fig. 3.3); grade 2, a fibrillation or fissures not extending to bone (Fig. 3.4); grade 3, a flap or exposed bone (Fig. 3.5); grade 4, a loose undisplaced fragment (Fig. 3.6), and grade 5, a displaced fragment (Fig. 3.7).

Fig. 3.3

Coronal fat-saturated intermediate weighted fast spin echo sequence demonstrating an osteochondral lesion at the medial talar dome (arrow). There is increase in signal of the cartilage and irregularity of the underlying bone, but the cartilage surface appears intact and there are no defects

Fig. 3.4

Sagittal fat-saturated dual echo steady state (DESS) sequence demonstrating an osteochondral injury at the talus. There is cartilage fissuring (arrow) with underlying bone marrow edema pattern (bone bruise)

Fig. 3.5

Sagittal fat-saturated intermediate weighted fast spin echo sequence showing an osteochondral lesion with a cartilage flap, a partially separated layer of cartilage with delamination (arrow) and underlying mild bone marrow edema pattern

Fig. 3.6

Coronal (a) and sagittal (b) fat-saturated intermediate weighted fast spin echo sequence showing an osteochondral lesion at the medial talar dome, which consists of a loose fragment, but the fragment is not displaced (arrows). Fluid between the bony fragment and the adjacent bone and adjacent bone marrow edema pattern is also depicted

Fig. 3.7

Coronal fat-saturated dual echo steady state (DESS) sequence demonstrates an osteochondral injury at the lateral aspect of the talar dome with a mildly displaced osteochondral fragment (arrow)

Only gold members can continue reading. Log In or Register to continue