Literature
Home医源资料库在线期刊放射学杂志2003年1月第226卷第2期

Assessment of Knee Cartilage in Cadavers with Dual-Detector Spiral CT Arthrography and MR Imaging1

来源:放射学杂志
摘要:Indexterms:Knee,CT,4521。Knee,ligaments,menisci,andcartilage,4521。Knee,MR,4521。Knee,arthritis,4521。...

点击显示 收起

1 From the Departments of Radiology (B.C.V.B., F.E.L., R.M., B.M., J.M.), Orthopaedic Surgery (P.P.), and Anatomy (J.J., B.L.), Cliniques Universitaires St Luc, Université Catholique de Louvain, 10 av Hippocrate, 1200 Brussels, Belgium. Received March 13, 2001; revision requested April 3; revision received June 29; accepted July 25.

     ABSTRACT

Top
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES

 
PURPOSE: To assess dual-detector spiral CT arthrography in the evaluation of the entire knee cartilage obtained from cadavers.

MATERIALS AND METHODS: Two independent observers characterized articular cartilage in 12 cadaver knees in which MR imaging and dual-detector spiral CT arthrography were performed and compared their findings to those found during macroscopic assessment. The sensitivity and specificity of MR imaging and spiral CT arthrography for detecting grade 2A or higher and grade 2B or higher cartilage lesions, the Spearman correlation coefficient between arthrographic and macroscopic grading, and statistics for assessing interobserver reproducibility were determined.

RESULTS: At spiral CT arthrography, sensitivities and specificities ranged between 80% and 88% for the detection of grade 2A or higher cartilage lesions and ranged between 85% and 94% for the detection of grade 2B or higher cartilage lesions. At MR imaging, sensitivities and specificities ranged between 78% and 86% and between 76% and 91% for the detection of grade 2A or higher and grade 2B or higher cartilage lesions, respectively. Spearman correlation coefficients between spiral CT arthrography or MR imaging and macroscopic grading of articular surfaces were 0.797 and 0.702, respectively.

CONCLUSION: Dual-detector spiral CT arthrography of the knee is a valuable method for the assessment of open cartilage lesions of the entire knee.

 

Index terms: Knee, CT, 4521.12115 • Knee, ligaments, menisci, and cartilage, 4521.311, 4521.321 • Knee, MR, 4521.121411 • Knee, arthritis, 4521.77


     INTRODUCTION

Top
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES
 
Magnetic resonance (MR) is the imaging modality of choice for the noninvasive assessment of articular cartilage; however, cartilage imaging remains one of the most difficult and controversial areas in musculoskeletal MR imaging (14). Conventional computed tomography (CT) of the patella (511) performed after knee arthrography enables accurate depiction of open cartilage lesions, which show substance loss at least in the most superficial aspect of the articular cartilage. Developments in spiral CT technology with multiple-detector arrays that enable submillimeter spatial resolution (12,13) raised the question of the potential use of dual-detector spiral CT arthrography of the knee for the assessment of the hyaline cartilage of the entire joint (14). The purpose of this study was to assess dual-detector spiral CT arthrography of cadaver knees in the evaluation of the entire knee cartilage.


     MATERIALS AND METHODS

Top
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES
 
Between January and March 2000, 12 knees sectioned across the distal third of the femur and the proximal third of the tibia were obtained from six embalmed cadavers (three women, three men; age range at time of death, 52–79 years; mean, 61 years). These were six consecutive cadavers without previous knee surgery. Radiographs of the knee specimens showed normal femorotibial joints in all specimens and degenerative changes of the femoropatellar joint in two cadavers (four knees). Authorization to perform imaging in cadaver knee specimens was obtained from the ethics committee of our institution.

Two radiologists (F.E.L., R.M.) working together performed MR and spiral CT imaging of one knee at a time. MR images of knee specimens were obtained at room temperature with a 1.5-T magnet (Gyroscan Intera; Philips Medical Systems, Best, the Netherlands) and the use of a quadrature knee coil supplied by the manufacturer. Transverse, coronal, and sagittal images were obtained by using a fat-saturated intermediate-weighted fast spin-echo sequence. Section thickness was 3 mm with 0.3-mm intersection gaps. Repetition times ranged between 3,015 and 3,525 msec with an echo time of 29 msec, which was adapted to provide an optimal signal contrast between the hyaline cartilage and the articular fluid. The echo train length was five. MR images were obtained with an 18-cm field of view, a 256 x 512 matrix, and three signals acquired. All images were stored on optic disks.

Immediately after MR imaging, a 10-mL volume of ionic contrast material (320 mg of iodine per milliliter; meglumine ioxalate and sodium ioxalate; Hexabrix 320, Guerbet, Aulnay-sous-bois, France) was injected with fluoroscopic control through a 20-gauge needle placed in the suprapatellar pouch of the knee. Before CT was performed, the injected knees were manipulated to provide a homogeneous coating of the articular surfaces. CT was performed by using a dual-detector helical unit (Twin RTS, Real Time Scanning; Marconi, Cleveland, Ohio) with the spiral-scanning mode. All knees were placed in the supine position. After a lateral projection scout image, a 65- to 85-second scanning was performed to image the area between the upper pole of the patella and the tibial plateaus. Spiral scanning was performed at 140 kVp and 135 mAs, with a focal spot size of 0.75 mm. The collimation beam was 1 mm. A dynamic oscillating spot was used (15). The field of view at acquisition was 250 mm. The table speed was 0.75 mm/sec (effective pitch of 0.75), and the effective section thickness was 0.55 mm. For reconstruction, a 360° linear interpolation algorithm, a high frequency kernel, an increment of 0.3 mm (40% section overlap), and a zoom factor of 1.94 were used. Images were reconstructed on a 512 x 512 matrix, and in-plane resolution was 0.43 mm. Longitudinal resolution was 0.3 mm because with a pitch value of 0.75 and a reconstruction increment of 60% of the nominal section width, longitudinal resolution equaled the reconstruction increment of 0.3 mm (16,17). All reconstructed images were prospectively stored on erasable optic disks. One radiologist (F.E.L.) obtained sagittal, coronal, and transverse reformations by using a workstation (Omnipro; Silicon Graphics, Mountain View, Calif), with imaging planes identical to those obtained at MR imaging of the corresponding knee. Twenty coronal and 40 sagittal reformations with a 0.45-mm thickness and 1.6-mm interval were obtained. Twenty 0.45-mm-thick transverse reformations were obtained with a 2.6-mm interval. All reformations were printed on film by using bone settings (window width, 1,900 HU; window level, 450 HU), with a zoom factor of 1.8.

After MR and CT, an orthopedic surgeon (P.P.) who was unaware of the imaging findings disarticulated each knee specimen. Specimens were rinsed, and the articular surfaces were painted by using a paintbrush filled with undiluted green dye (Wak-Chemie Medical, Bad Boden, Germany) (18). Once the articular surfaces were completely colored with green dye, the specimens were rinsed with saline. Dye covering normal hyaline cartilage was rinsed away. Areas of abnormal hyaline cartilage showed dark green markings against a pale gray or yellow background of normal hyaline cartilage when examined en face. Specimens were photographed. Sites of complete cartilage ulceration with subchondral bone eburnation showed the yellow color of subchondral bone.

For decalcification, each set of three bone pieces was immersed for 10 days in a solution of 25% chlorhydric acid solution (DC3; Solvay, Brussels, Belgium). Decalcification did not alter hyaline cartilage coloration. After decalcification, the specimens were sectioned with a commercially available rotating saw (MAS 9100; Bosch, Munich, Germany) into 1.5-mm-thick sections by a musculoskeletal radiologist (B.C.V.B.). The femur and tibia were sectioned in a sagittal plane and the patella in a transverse plane. A total of 678 anatomic sections were obtained.

Immediately after sectioning, each anatomic section was numbered and photographed. Hyaline cartilage status was graded by using visual inspection by an orthopedic surgeon (P.P.) and a radiologist (B.C.V.B.) working in consensus. Both ignored CT and MR findings. The system used to grade articular cartilage at macroscopic inspection was a modification of the Noyes classification system (19), which was used at arthroscopy (Table 1). The Noyes system is based on the following four parameters: integrity of the articular surface, depth of involvement, location of the lesion, and diameter of the lesion. In this study, the first two parameters were used. Grading of articular surfaces was performed in 12 anatomic areas of each knee including the medial facet, the crest and the lateral facet of the patella, the meniscus-covered and the free surface of both tibial plateaus, the trochlea, and the anterior and posterior halves of both femoral condyles. The most severe lesion observed in any section of each of the 12 anatomic regions was used to grade the articular surface of that region. A grade 0 surface was defined by intact cartilage surface without green dye on the surface. A grade 1 surface was defined by cartilage blistering or fibrillation and cartilage softening at palpation. The articular surface was colored green (abnormal surface), but there was no substance loss on corresponding anatomic sections. Grade 2A, 2B, and 3 surfaces corresponded to cartilage areas with open lesions that were defined by surface changes on the en-face view and substance loss on the anatomic section, with penetration of the dye in the lesion. Depth of substance loss was measured by means of a magnifying glass (Peak Loop, Tokyo, Japan) with 10 times enlargement. This loupe, applied to the anatomic sections, had a scale marked in increments of 0.1 mm. Depth of cartilage loss was measured and expressed as percentage of cartilage thickness of adjacent normal cartilage. Grade 2A surface showed a cartilage loss of less than 50% of cartilage thickness; grade 2B surface showed cartilage loss of at least 50% but not down to bone; grade 3 surface showed substance loss down to bone.


fig.ommitted TABLE 1. Grading System for Macroscopic Inspection of Articular Cartilage Lesions and Interpretation of MR and CT Arthrographic Images

 

 
Image Analysis
Six months after performing MR and CT of the specimens and macroscopic examinations of the anatomic sections, two musculoskeletal radiologists (F.E.L., B.C.V.B.) separately analyzed the CT and MR images that had been printed on film. The sets of MR images of six knees and the sets of CT images of the six other knees were viewed first and, 3 weeks later, the six remaining MR and six remaining CT image sets were analyzed. For each imaging modality, transverse, coronal, and sagittal images were viewed simultaneously. Both observers were blinded to the identity of the specimens. For the 12 anatomic regions of each knee that were identical to those defined for the macroscopic analysis, the grade of the articular surface was determined by taking into account the deepest lesion observed in that region. After a 2-month delay, the first observer repeated the blinded retrospective interpretation of all examinations to determine intraobserver reproducibility.

At MR imaging, a grade 0 surface was defined by a homogeneous intermediate signal intensity of the hyaline cartilage, without surface irregularity (Table 1). Grade 1 surface was defined by diffuse signal intensity changes in the most superficial area of the cartilage without evident substance loss. Grade 2A surface was defined by high signal intensity cartilage areas that clearly reached the upper surface of the cartilage and involved less than half of the cartilage thickness. Grade 2B surface was defined by signal intensity changes that involved at least the superficial half of the cartilage thickness but did not reach the subchondral bone. Grade 3 surface was defined by a complete loss of normal cartilage, with abnormal signal intensity changes reaching the subchondral bone. When surface grading differed between the imaging planes, the most severe grade was reported.

At spiral CT arthrography, grade 0 surface was defined by a sharp line of contrast material on the cartilage surface, without substance loss (Table 1). Grade 1 surface was defined by a loss of the sharp and smooth contour of the cartilage surface with the appearance of subtle undulation and unsharpness of the cartilage surface. There was no contrast material within the cartilage. Grade 2A surface was defined by the penetration of contrast material within the cartilage that involved less than half of the thickness of the normal adjacent cartilage. Grade 2B surface was defined by the penetration of contrast material within the cartilage that involved at least the superficial half of the cartilage thickness but did not reach the subchondral bone. Grade 3 surface was defined by the complete loss of normal cartilage, with contrast material reaching the subchondral bone. When surface grading differed between the imaging planes, the most severe grade was elected.

Statistical Analysis
Statistical analysis was performed by a statistician (J.J.). Sensitivity and specificity for detecting grade 2A or higher cartilage lesions were calculated by considering as test-positive areas, all areas in which at least grade 2A lesions were present at spiral CT arthrography or MR imaging. Sensitivity and specificity for detecting lesions with at least 50% substance loss were calculated by considering as test-positive areas, all areas in which at least grade 2B lesions were present at spiral CT arthrography or MR imaging. Correlations between grades determined at spiral CT arthrography or MR imaging and those determined by visual inspection of anatomic sections were assessed by using the Spearman rank coefficient. Interobserver and intraobserver agreement were assessed by using statistics. As suggested by Landis and Koch (20), values less than 0 were considered to be poor agreement; between 0 and 0.2, slight agreement; between 0.21 and 0.40, fair agreement; between 0.41 and 0.60, moderate agreement; between 0.61 and 0.80, substantial agreement; and between 0.81 and 1.0, almost perfect agreement.

Sensitivities and specificities of dual-detector spiral CT arthrography and MR imaging were compared by using the McNemar test (21). Spearman rank coefficients, from findings observed at spiral CT arthrography and MR imaging, were compared by using the Hotelling test (22). A P value less than .05 was considered to indicate a statistically significant difference.


     RESULTS

Top
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES
 
Macroscopic Analysis
At macroscopic examination, 25 of 144 articular surfaces (12 knees x 12 surfaces each) were intact (grade 0); 53 surfaces were classified grade 1 (Fig 1); 32 surfaces, grade 2A (Fig 2); 14 surfaces, grade 2B (Fig 3); and 20 surfaces, grade 3 (Fig 4).


fig.ommitted Figure 1a. (a) Sagittal reformation of the lateral femorotibial joint reconstructed after spiral CT arthrography of a cadaver knee shows normal appearance of the cartilage surface (arrow) of the lateral tibial plateau. (b) The cartilage (arrow) was also considered to be normal (grade 0) on the corresponding fat-saturated intermediate-weighted sagittal MR image (3,425/29 [repetition time msec/echo time msec]). (c) Close-up photograph of the corresponding anatomic section of the posterior half of the lateral tibial plateau depicts green dye on the meniscus-covered area of the tibial plateau (arrows). This area was considered to be grade 1 at macroscopic examination. Superficial cartilage changes were underestimated at spiral CT arthrography and MR imaging.

 

 

fig.ommitted
 
Figure 1b. (a) Sagittal reformation of the lateral femorotibial joint reconstructed after spiral CT arthrography of a cadaver knee shows normal appearance of the cartilage surface (arrow) of the lateral tibial plateau. (b) The cartilage (arrow) was also considered to be normal (grade 0) on the corresponding fat-saturated intermediate-weighted sagittal MR image (3,425/29 [repetition time msec/echo time msec]). (c) Close-up photograph of the corresponding anatomic section of the posterior half of the lateral tibial plateau depicts green dye on the meniscus-covered area of the tibial plateau (arrows). This area was considered to be grade 1 at macroscopic examination. Superficial cartilage changes were underestimated at spiral CT arthrography and MR imaging.

 

 

fig.ommitted Figure 1c. (a) Sagittal reformation of the lateral femorotibial joint reconstructed after spiral CT arthrography of a cadaver knee shows normal appearance of the cartilage surface (arrow) of the lateral tibial plateau. (b) The cartilage (arrow) was also considered to be normal (grade 0) on the corresponding fat-saturated intermediate-weighted sagittal MR image (3,425/29 [repetition time msec/echo time msec]). (c) Close-up photograph of the corresponding anatomic section of the posterior half of the lateral tibial plateau depicts green dye on the meniscus-covered area of the tibial plateau (arrows). This area was considered to be grade 1 at macroscopic examination. Superficial cartilage changes were underestimated at spiral CT arthrography and MR imaging.

 

 

fig.ommitted Figure 2a. (a) Coronal reformation of the femorotibial joints reconstructed after spiral CT arthrography of a cadaver knee shows contrast material (arrow) in the lateral aspect of the medial condyle cartilage. The anterior half of the medial femoral condyle was considered to be grade 2A because substance loss was inferior to 50% of normal cartilage. (b) Corresponding coronal fat-suppressed intermediate-weighted MR image (3,085/29) shows a focal area of increased signal intensity (arrow) in the same area that involves less than 50% of cartilage thickness (grade 2A surface). (c) Close-up photograph of the most lateral anatomic section of the medial femoral condyle shows a defect (arrows) that involves less than 50% of the cartilage thickness. The anterior half of the medial femoral condyle was considered to be grade 2A.

 

 

fig.ommitted Figure 2b. (a) Coronal reformation of the femorotibial joints reconstructed after spiral CT arthrography of a cadaver knee shows contrast material (arrow) in the lateral aspect of the medial condyle cartilage. The anterior half of the medial femoral condyle was considered to be grade 2A because substance loss was inferior to 50% of normal cartilage. (b) Corresponding coronal fat-suppressed intermediate-weighted MR image (3,085/29) shows a focal area of increased signal intensity (arrow) in the same area that involves less than 50% of cartilage thickness (grade 2A surface). (c) Close-up photograph of the most lateral anatomic section of the medial femoral condyle shows a defect (arrows) that involves less than 50% of the cartilage thickness. The anterior half of the medial femoral condyle was considered to be grade 2A.

 

 

fig.ommitted Figure 2c. (a) Coronal reformation of the femorotibial joints reconstructed after spiral CT arthrography of a cadaver knee shows contrast material (arrow) in the lateral aspect of the medial condyle cartilage. The anterior half of the medial femoral condyle was considered to be grade 2A because substance loss was inferior to 50% of normal cartilage. (b) Corresponding coronal fat-suppressed intermediate-weighted MR image (3,085/29) shows a focal area of increased signal intensity (arrow) in the same area that involves less than 50% of cartilage thickness (grade 2A surface). (c) Close-up photograph of the most lateral anatomic section of the medial femoral condyle shows a defect (arrows) that involves less than 50% of the cartilage thickness. The anterior half of the medial femoral condyle was considered to be grade 2A.

 

 

fig.ommitted Figure 3a. (a) Sagittal reformation of the lateral femorotibial joint reconstructed after spiral CT arthrography of a cadaver knee shows a focal defect (arrow) in the meniscus-covered area of the lateral tibial plateau. This surface was considered to be grade 2B because the defect involved more than 50% of cartilage thickness but did not appear to reach the subchondral bone. (b) Corresponding fat-saturated intermediate-weighted sagittal MR image (3,425/29) shows the defect (arrow) in a similar manner, with residual cartilage at the bottom of the lesion indicative of a grade 2B surface, a finding that was confirmed at macroscopic examination (image not shown).

 

 

fig.ommitted Figure 3b. (a) Sagittal reformation of the lateral femorotibial joint reconstructed after spiral CT arthrography of a cadaver knee shows a focal defect (arrow) in the meniscus-covered area of the lateral tibial plateau. This surface was considered to be grade 2B because the defect involved more than 50% of cartilage thickness but did not appear to reach the subchondral bone. (b) Corresponding fat-saturated intermediate-weighted sagittal MR image (3,425/29) shows the defect (arrow) in a similar manner, with residual cartilage at the bottom of the lesion indicative of a grade 2B surface, a finding that was confirmed at macroscopic examination (image not shown).

 

 

fig.ommitted Figure 4a. (a) Coronal reformation of the femorotibial joints reconstructed after spiral CT arthrography of a cadaver knee shows a focal defect (arrow) in the middle third of the lateral femoral condyle. The anterior half of the lateral femoral condyle was considered to be grade 3 because the lesions reached the subchondral bone. (b) On the corresponding fat-saturated intermediate-weighted coronal MR image (3,085/29), the defect (arrow) reaches the subchondral bone and indicates a down to bone lesion (grade 3 surface) that was confirmed at inspection of anatomic section (not shown).

 

 

fig.ommitted Figure 4b. (a) Coronal reformation of the femorotibial joints reconstructed after spiral CT arthrography of a cadaver knee shows a focal defect (arrow) in the middle third of the lateral femoral condyle. The anterior half of the lateral femoral condyle was considered to be grade 3 because the lesions reached the subchondral bone. (b) On the corresponding fat-saturated intermediate-weighted coronal MR image (3,085/29), the defect (arrow) reaches the subchondral bone and indicates a down to bone lesion (grade 3 surface) that was confirmed at inspection of anatomic section (not shown).

 

 
Detection of Grade 2A or Higher and Grade 2B or Higher Cartilage Lesions
Sensitivity and specificity values are listed in Table 2. At spiral CT arthrography, sensitivities for the detection of grade 2A or higher cartilage lesions were 80% and 86% for observers 1 and 2, respectively; specificities were 88% and 86% for observers 1 and 2, respectively. At MR imaging, sensitivities for the detection of grade 2A or higher cartilage lesions were 83% and 86% for observers 1 and 2, respectively; specificities were 80% and 78% for observers 1 and 2, respectively.


fig.ommitted TABLE 2. Sensitivity and Specificity Values for Detection of Cartilage Lesions

 

 
At spiral CT arthrography, sensitivities for the detection of grade 2B or higher cartilage lesions were 85% and 94% for observers 1 and 2, respectively, and specificity was 94% for both observers. At MR imaging, sensitivities for the detection of grade 2B or higher cartilage lesions were 76% and 91% for observers 1 and 2, respectively, and specificity was 84% for both observers.

Grading of Articular Surfaces
At spiral CT arthrography, 77 (53%) of the 144 regions were graded identically to macroscopic examination by observer 1. The 67 regions that were graded differently at spiral CT arthrography and macroscopic examination differed by one grade in 61 regions and by two grades in six. A total of 96% of the regions were graded within one grade and 100% were within two grades (Table 3). Results obtained by observer 2 are in Table 3. There was substantial agreement between grading at macroscopic examination and grading at spiral CT arthrography (Spearman correlation coefficient of 0.797 and 0.794 for observers 1 and 2, respectively) (Table 4).


fig.ommitted
 
TABLE 3. Results of Grading 144 Cartilage Surfaces in 12 Cadaver Knees by Two Independent Observers

 

 

fig.ommitted TABLE 4. Comparison of Spearman Correlation Coefficients

 

 
At MR imaging, 54 (38%) of the 144 regions were graded identically to macroscopic examination by observer 1. The 90 regions that were graded differently at MR imaging and macroscopic examination differed by one grade in 71 regions and by two grades in 16 (Table 3). There was substantial agreement between grading at macroscopic examination and grading at MR imaging (Spearman correlation coefficient of 0.702 and 0.688 for observers 1 and 2, respectively) (Table 4).

Interobserver and Intraobserver Reproducibility
At spiral CT arthrography, interobserver agreement was moderate for grading articular surfaces and substantial for detecting grade 2A or higher and grade 2B or higher cartilage lesions (Table 5). Intraobserver agreement was substantial for grading articular surfaces and almost perfect for detecting grade 2A or higher and grade 2B or higher cartilage lesions (Table 5). The values observed at MR imaging are shown in Table 5.


fig.ommitted TABLE 5. Inter- and Intraobserver Reproducibility Statistics

 

 
Comparison between Spiral CT Arthrography and MR Imaging
There was no statistically significant difference between spiral CT arthrography and MR imaging in sensitivity and specificity for the detection of grade 2A or higher cartilage lesions and in sensitivity for the detection of grade 2B or higher cartilage lesions (Table 2). Specificity for the detection of grade 2B or higher cartilage lesions was significantly higher at spiral CT arthrography than at MR imaging for both observers (P = .008 and .005, for observers 1 and 2, respectively) (Table 2). Spearman correlation coefficients, from spiral CT arthrography findings, for grading articular surfaces with respect to macroscopic analysis were significantly higher than those obtained from MR imaging findings for both observers (P = .013) (Table 4).


     DISCUSSION

Top
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES
 
Findings of this study demonstrated that dual-detector spiral CT arthrography of the knee enabled accurate assessment of the articular cartilage of the entire knee. Sensitivities and specificities of spiral CT arthrography ranged between 80% and 88% for the detection of grade 2A or higher cartilage lesions and between 85% and 94% for the detection of grade 2B or higher cartilage lesions. Furthermore, grading of articular surfaces by means of spiral CT arthrography satisfactorily correlated with grading of anatomic sections, with Spearman correlation coefficients of 0.797 and 0.794 for observers 1 and 2, respectively. Findings of previous studies (911) with conventional CT performed after arthrography showed that conventional CT enabled accurate detection of patellar cartilage lesions. Administration of an intraarticular positive contrast agent enables delineation of the cartilage surface with subsequent excellent lesion conspicuity, because open (grade 2A or higher) cartilage lesions filled with contrast material show high-attenuation density, whereas intact cartilage shows low-attenuation density (5,7,9). We can only postulate as to why spiral CT arthrography enabled assessment of the articular cartilage of the entire knee joint. Most likely, the value of spiral CT arthrography is derived from its multiplanar capacity and spatial resolution. Sagittal and coronal images reconstructed after spiral CT arthrography enable excellent depiction of the femorotibial cartilage because these images are perpendicular to the articular surfaces. The spatial resolution achieved by means of spiral CT arthrography, with 0.43-mm in-plane resolution and 0.3-mm longitudinal resolution, also contributed to the results. Stair-step artifacts that can be observed on reconstructed images on surfaces inclined with respect to the longitudinal axis were not observed because the detector collimation and table increment were much lower than the longitudinal dimensions of the articular surface that were parallel to the x-ray beam (23).

Additionally, interobserver agreement was moderate for grading articular surfaces and substantial for detecting grade 2A or higher and grade 2B or higher cartilage lesions. This level of agreement achieved by means of spiral CT arthrography parallels that of Disler et al (24), who obtained by means of conventional MR images values ranging between 0.43 and 0.83, depending on the articular surfaces. Our results are less promising than those of Daenen et al and Potter et al, who observed excellent interobserver reproducibility in the detection of patellar cartilage lesions by means of conventional CT arthrography (10) and conventional MR imaging (25), respectively. These differences may be related to the number of response categories between the current study and previous studies.

Finally, comparison of the results obtained by means of spiral CT arthrography or fat-saturated intermediate-weighted MR imaging with respect to macroscopic analysis demonstrated the absence of significant differences in sensitivity for the detection of grade 2A or higher cartilage lesions in the entire knee between the two techniques, as previously demonstrated for the patellar cartilage (10,11). However, our results demonstrated that spiral CT arthrography might be superior to fat-saturated intermediate-weighted MR imaging in the determination of the depth of the substance loss, thus providing a better measurement of the depth of cartilage loss in grade 2B or higher cartilage lesions. Actually, Spearman correlation coefficients for grading articular surfaces and specificity for the detection of grade 2B or higher cartilage lesions were significantly higher at spiral CT arthrography than at MR imaging for the two observers. Limited performance of MR imaging in the determination of substance loss depth has been reported, with possible overgrading of deep cartilage defects (2530), but the MR sequence used in the current study was accurate and clinically contributive in studies (27,31,32) in which results similar to ours were obtained. However, the comparison of the CT and MR results was limited by the sequence and section thickness that were used in our MR imaging protocol. Results of spiral CT arthrography remain comparable with those obtained by means of more advanced MR techniques, such as three-dimensional imaging with thin sections (3336) and, possibly, intraarticular injection of gadolinium derivatives (37).

Spiral CT arthrography definitely has several intrinsic disadvantages in comparison with MR imaging. First, spiral CT arthrography is invasive because it requires intraarticular contrast material injection and ionizing radiation. Second, it is inherently limited because of its poor contrast resolution and the lack of attenuation variability within the zones of the hyaline cartilage. Consequently, it is limited to the depiction of the cartilage surface, and closed cartilage lesions remain occult (1,10,11).

Our study had several limitations mainly due to the use of cadaver knees (32). First, the use of knee from cadavers may interfere with the results because of the lack of normal articular fluid at MR imaging and the lack of active mobilization of the knee after intraarticular injection of contrast material before CT arthrography. Second, the use of knees from elderly subjects may also influence the results because of different cartilage lesion patterns among elderly and young patients (38). Third, the standard was established with the findings of anatomic specimens by an orthopedic surgeon and a radiologist who also participated in the analysis of the images. The possibility that this observer (B.C.V.B.) was influenced by his analysis of the 678 anatomic sections was probably eliminated by the delay of 4 months between analysis of the sections and that of the CT and MR images. Both observers were blinded to the cadaver identification number of the specimens. The fact that both observers had similar results suggests that this bias did not occur. Finally, observer awareness of cartilage lesions was higher in the present study than it would be in clinical practice, which may have led to improved sensitivity for detecting lesions.

In conclusion, assessment of the entire articular cartilage in cadaver knees by means of dual-detector spiral CT arthrography demonstrated a sensitivity and specificity range between 80% and 88% for detection of all open cartilage lesion and between 85% and 94% for detection of lesions with at least 50% substance loss.

 

     REFERENCES

Top
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES
 

  1. Hodler J, Resnick D. Current status of imaging of articular cartilage. Skeletal Radiol 1996; 25:703-709.

  2. McCauley TR, Disler DG. MR imaging of articular cartilage. Radiology 1998; 209:629-640.

  3. Peterfy CG, Genant HK. Emerging applications of magnetic resonance imaging in the evaluation of articular cartilage. Radiol Clin North Am 1996; 34:195-213.

  4. Kneeland B. Magnetic resonance imaging of articular cartilage. Semin Roentgenol 2000; 35:249-255.

  5. Boven F, Bellemans M, Geurts J, De Boeck H, Potvliege R. The value of computed tomography scanning in chondromalacia patellae. Skeletal Radiol 1982; 8:183-185.

  6. Handelberg F, Shahabpour M, Casteleyn PP. Chondral lesions of the patella evaluated with computed tomography, magnetic resonance imaging, and arthroscopy. Arthroscopy 1990; 6:24-29.

  7. Ihara H. Double-contrast CT arthrography of the cartilage of the patellofemoral joint. Clin Orthop 1985; 198:50-55.

  8. Conway WF, Hayes CW, Loughran T, et al. Cross-sectional imaging of the patellofemoral joint and surroundings structures. RadioGraphics 1991; 11:195-217.

  9. 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.

  10. Daenen BR, Ferrara MA, Marcelis S, Dondelinger RF. Evaluation of patellar cartilage surface lesions: comparison of CT arthrography and fat-suppressed FLASH 3D MR imaging. Eur Radiol 1998; 8:981-985.

  11. Rand T, Brossman J, Pedowitz R, Ahn JM, Haghigi P, Resnick D. Analysis of patellar cartilage: comparison of conventional MR imaging and MR and CT arthrography in cadavers. Acta Radiol 2000; 41:492-497.

  12. Berland LL, Smith JK. Multidetector-array CT: once again, technology creates new opportunities. Radiology 1998; 209:327-329.

  13. Liang Y, Kruger RA. Dual-slice spiral versus single-slice spiral scanning: comparison of the physical performance of two computed tomography scanners. Med Phys 1996; 23:205-220.

  14. Vande Berg BC, Lecouvet FE, Poilvache P, et al. Dual-detector spiral CT arthrography of the knee: accuracy for detection of meniscal abnormalities and unstable meniscal tears. Radiology 2000; 216:851-857.

  15. Berland LL, ed. Practical CT technology In: Routine scan factors. New York, NY: Raven, 1987; 56-72.

  16. Brink JA, Heiken JP, Wang G, McEnery KW, Schlueter FJ, Vannier MW. Spiral (helical) CT: principles and technical considerations. RadioGraphics 1994; 14:887-893.

  17. Brink JA, Davros WJ, Pennington J, Englis MR, eds. Helical/spiral CT: a practical approach In: Helical/spiral CT: technical principles. New York, NY: McGraw-Hill, 1995; 1-26.

  18. Meachim G, Fergie IA. Morphological patterns of articular cartilage fibrillation. J Pathol 1975; 115:231-240.

  19. Noyes FR, Stabler Craig L. A system for grading articular cartilage lesions at arthroscopy. Am J Sports Med 1989; 17:505-513.

  20. Landis JR, Koch GG. The measurement of observer agreement for categorical data. Biometrics 1977; 33:159-174.

  21. Dwyer AJ. Matchmaking and McNemar in the comparison of diagnostic modalities. Radiology 1991; 178:328-330.

  22. Cohen A. Comparison of correlated correlations. Stat Med 1989; 8:1485-1495.

  23. Wang G, Vannier MW. Stair-step artifacts in three-dimensional helical CT: an experimental study. Radiology 1994; 191:79-83.

  24. Disler DG, McCauley TR, Kelman CG, et al. Fat-suppressed three-dimensional spoiled gradient-echo MR imaging of hyaline cartilage defects in the knee: comparison with standard MR imaging and arthroscopy. AJR Am J Roentgenol 1996; 167:127-132.

  25. Potter HG, Linklater JM, Allen AA, Hannafin JA, Haas SB. Magnetic resonance imaging of articular cartilage in the knee. J Bone Joint Surg Am 1998; 80:1276-1284.

  26. van Leersum M, Schweitzer ME, Gannon F, Vinitski S, Finkel G, Mitchell DG. Thickness of patellofemoral articular cartilage as measured on MR imaging: sequence comparison of accuracy, reproducibility, and interobserver variation. Skeletal Radiol 1995; 24:431-435.

  27. Bredella MA, Tirman PF, 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.

  28. Kramer J, Recht MP, Imhof H, Stiglbauer R, Engel A. Postcontrast MR arthrography in assessment of cartilage lesions. J Comput Assist Tomogr 1994; 18:218-224.

  29. Trattnig S, Huber M, Breitenseher MJ, et al. Imaging articular cartilage defects with 3D fat-suppressed echo planar imaging: comparison with conventional 3D fat-suppressed gradient echo sequence and correlation with histology. J Comput Assist Tomogr 1998; 22:8-14.

  30. van Leersum M, Schweitzer ME, Gannon F, Finkel G, Vinitski S, Mitchell DG. Chondromalacia patellae: an in vitro study. Skeletal Radiol 1996; 25:727-732.

  31. Rose PM, Demlow TA, Szumowski J, Quinn SF. Chondromalacia patellae: fat-suppressed MR imaging. Radiology 1994; 193:437-440.

  32. Ahn JM, Kwak SM, Kang HS, et al. Evaluation of patellar cartilage in cadavers with a low-field-strength extremity-only magnet: comparison of MR imaging sequences, with macroscopic findings as the standard. Radiology 1998; 208:57-62.

  33. Rubenstein JD, Li JG, Majumdar S, Henkelman RM. Image resolution and signal-to-noise ratio requirements for MR imaging of degenerative cartilage. AJR Am J Roentgenol 1997; 169:1089-1096.

  34. Brossmann J, Frank LR, Pauly JM, et al. Short echo time projection reconstruction MR imaging of cartilage: comparison with fat-suppressed spoiled GRASS and magnetization transfer contrast MR imaging. Radiology 1997; 203:501-507.

  35. Uhl M, Allmann KH, Tauer U, et al. Comparison of MR sequences in quantifying in vitro cartilage degeneration in osteoarthritis of the knee. Br J Radiol 1998; 71:291-296.

  36. Seo GS, Aoki J, Moriya H, et al. Hyaline cartilage: in vivo and in vitro assessment with magnetization transfer imaging. Radiology 1996; 201:525-530.

  37. Chandnani VP, Ho C, Chu P, Trudell D, Resnick D. Knee hyaline cartilage evaluated with MR imaging: a cadaveric study involving multiple imaging sequences and intraarticular injection of gadolinium and saline solution. Radiology 1991; 178:557-561.

  38. Recht MP, Piraino DW, Paletta GA, Schils JP, Belhobek GH. Accuracy of fat-suppressed three-dimensional spoiled gradient-echo flash MR imaging in the detection of patellofemoral articular cartilage abnormalities. Radiology 1996; 198:209-212.
作者: Bruno C. Vande Berg MD PhD Frédéric E. Lecouv 2007-5-12
医学百科App—中西医基础知识学习工具
  • 相关内容
  • 近期更新
  • 热文榜
  • 医学百科App—健康测试工具