| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||
ABSTRACT |
Thursday, May 10, 1:30 PM–3:30 PM
Abstracts 273-284
Moderator(s): William Morrison and Brian Sabb
1:30 PM
273. The Incidence of Degenerative Joint Disease and Arthroscopically Treatable Lesions Detected by MRI and Stratified by Age
Salsamendi J.*; Umans H.; Engelsohn E.; Chang S.; DiFelice G. Jacobi Medical Center, Albert Einstein College of Medicine, Bronx, NY
Address correspondence to J. Salsamendi (jasonsalsamendi{at}gmail.com)
Objective: To determine: 1) the incidence of degenerative joint disease (DJD) based on knee MRI as it relates to advancing age, and 2) the incidence of lesions potentially treatable by arthroscopy focusing on meniscal tear, chondral flaps and loose bodies with and without DJD in the population as stratified by age group.
Materials and Methods: All knee MRI performed in two 1.5-T facilities from 11/2002–5/2006 (9-90 years of age, 653 women, 365 men) were retrospectively reviewed for DJD (none-mild vs. moderate-severe), chondral flaps, loose bodies and meniscal tear. Meniscal tears were categorized as degenerative, radial, bucket handle, oblique longitudinal or complex. All imaging included coronal T1, FSEIR, sagittal and proton-density fat suppressed sequences. Exclusion criteria included fracture and infection. Due to the retrospective nature of this study, correlation to clinical exam was not possible. The incidence of meniscal tear by type, chondral flaps or loose bodies with and without DJD was determined by age, stratified in 5-year groups from <40 to >70. The percentage of these lesions deemed potentially treatable by arthroscopy in a given age group with and without DJD was compared to: a) the total number of patients in that age group, and to b) the matched number of patients in each age group either with or without DJD. In this way we assessed the yield of arthroscopically treatable lesions as stratified by age in all-comers as it compared to individuals either known to have DJD or not. The incidence of DJD was calculated for each age group. Pearson correlation and paired t-test analysis was performed to compare the incidence of arthroscopically treatable lesions with respect to age both with and without DJD.
Results: The incidence of DJD is 7.4% in the group <40 and increases to 64.4% in the group >70. The incidence of arthroscopically treatable lesions in patients without DJD ranges from 29% in the 50-55 group to 11% in the >70 group. In patients with DJD, the incidence of findings potentially treatable by arthroscopy ranges from 23% to 42% with no correlation to patient age, as demonstrated by Pearson correlation coefficient = 0.12. Paired t-test analysis demonstrates a statistically significant (p < 0.001) higher mean incidence of findings potentially amenable to arthroscopy in patients with DJD as compared to patients without DJD.
Conclusion: The incidence of DJD increases with advancing age. Without DJD, the incidence of lesions potentially amenable to arthroscopic treatment appears to peak between 50-55. MRI identifies these lesions with relatively increased incidence in the context of DJD independent of age.
274. Atraumatic Lateral Collateral Ligament Complex Signal abnormalities by MRI in Patients with Lateral Compartment Knee Osteoarthrosis
Chen Y.1*; Fayad L. M.1; Raman S. P.1; Morrison W. B.2; Carrino J. A.1 1. Johns Hopkins Hospital, Baltimore, MD; 2. Thomas Jefferson University Hospital, Philadelphia, PA
Address correspondence to Y. Chen (ychen97{at}jhmi.edu)
Objective: To determine the features and prevalence of atraumatic lateral collateral ligament complex (LCLC) signal abnormalities by MRI in patients with lateral compartment knee osteoarthrosis (OA).
Materials and Methods: At 1.5 T, 95 patients (52 women, 43 men), 51 with OA (mean age 59, range 32-77) and 44 patients with atraumatic knee pain and no history of OA (mean age 46, range 9-78) underwent MRI. Imaging was retrospectively reviewed by 2 observers who graded the severity of OA on radiographs (Kellgren and Lawrence classification) and severity of signal abnormalities of the LCLC components on MRI (0 = no signal abnormality, 1 = abnormal signal adjacent to the component, 2 = partial abnormal signal within the component, 3 = abnormal signal and complete discontinuity of the component). A chi square analysis was used to test for an association of signal abnormalities for each LCLC component (iliotibial band (ITB), fibular collateral ligament (FCL), bicep femoris tendon (BFT) and popliteus tendon (PLT)) with OA. The measure of effect was assessed with an odds ratio (OR) using a 95% confidence interval (CI). The degree of severity of LCLC signal abnormalities was compared with the degree of severity of OA using the Mantel-Haenszel test for trend.
Results: LCLC signal abnormalities were identified in 84% (43/51) of patients with OA (at the ITB [n = 21/51], FCL [n = 43/51], BFT [n = 25/51] and PLT [n = 18/51]) compared to only 25% (11/44) of patients without OA. Severity of the LCLC signal abnormalities was mild to moderate in nature (for FCL, 21/43 had a severity of 1 and 22/43 had a severity of 2). Lateral compartment OA was significantly associated with signal abnormalities in the FCL (OR 7.8, CI [3.0–20], p < 0.0001), BFT (OR 3.1, CI [1.3-7.5], p = 0.01), and PLT (3.4, CI [1.2-9.2], p = 0.018). The severity of lateral compartment OA was significantly associated with severity of signal abnormalities in the FCL (p < 0.0001) and PLT (p = 0.02).
Conclusion: LCLC signal abnormalities are common in patients with OA in the absence of trauma. When identified, caution should be given to the interpretation of LCLC signal abnormalities, as they may be considered a feature of OA and should not be incorrectly attributed to an acute traumatic injury.
275. Semimembranosus Strain Injury: Another Sign of Knee Instability
Burns J.*; Sperling K. E.; Haramati N.; Levy M. Montefi ore Medical Center, Bronx, NY
Address correspondence to J. Burns (judahburns{at}hotmail.com)
Objective: To assess the validity of using distal semimembranosus tendon signal as a secondary sign of posteromedial knee instability in cases of anterior cruciate ligament (ACL) tear on MR.
Materials and Methods: A case-controlled, retrospective review of available MR images in a consecutive series of 34 surgically proven ACL tears was conducted to identify abnormal high signal within the substance of the distal semimembranosis tendon. Age and sex matched controls were identified using our RIS database (20 men, median age 31 range 19-46; 14 women, median age 30 range 14-46 years). Associated findings, including contusion to the posteromedial tibial plateau, posterolateral compartment injury, and meniscal injury were also documented. Pre-surgical charts, when available, were reviewed to assess chronicity of injury. Statistical significance was tested using chi-square and Fisher exact tests.
Results: Abnormal MR signal within the distal semimembranosus tendon was demonstrated in 28 of 34 (82%) cases of surgically documented ACL tear. These included both cases of acute and chronic injury. This finding was strongly associated with ACL injury (p < 0.0001). The finding of distal semimembranosus signal on MR is 71% sensitive and 79% specific for ACL tear. Abnormal MR signal was associated with ACL tear even in cases where contusion of the posteromedial tibial plateau was not demonstrated (p < 0.005).
Conclusion: Semimembranosus strain injury represents a distinct injury to the posteromedial knee in cases of traumatic ACL tear. While contusion of the posteromedial tibial plateau has been previously associated with such injuries, intra-tendon signal, separate from the tibial insertion site, is highly associated with posteromedial knee instability. Secondary signs of instability are invaluable in assessing problematic cases, and we feel that this semimembranosus sign is another valid secondary sign.
276. Segond Fracture: A Harbinger of Posterolateral Corner Injuries?
Malone W. J.1*; Gopez A. G.2; Zoga A. C.2; Morrison W. B.2 1. Geisinger Medical Center, Danville, PA; 2. Thomas Jefferson University, Philadelphia, PA
Address correspondence to W. Malone (wjmalone{at}geisinger.edu)
Objective: Segond fractures are uncommon, but when present the injury suggests accompanying ACL tear. Posterolateral corner (PLC) injuries can be difficult to identify clinically, as well as with traditional MR imaging planes. However, they are an important entity to diagnose because depending on their severity, instability from missed PLC injuries has been implicated in complications from failed cruciate reconstruction to early osteoarthritis. We sought to determine the prevalence of PLC injuries in the setting of Segond fracture.
Materials and Methods: We performed a database search over a 5-year period for MRI and radiographic reports with keyword "Segond fracture." Images were reviewed by 2 radiologists in consensus for Segond fracture as well as meniscal and ligamentous injuries-including the major stabilizing structures of the posterolateral corner (biceps femoris tendon, fibular collateral ligament, popliteus tendon, poplitofibular ligament, origin of the lateral gastrocnemius muscle as well as the arcuate/capsule complex).
Results: Of 13 patients (M:F = 10:3, mean age 31, range 15-59), 13 (100%) had complete cruciate ligament tears, 11 ACL and 2 PCL tears. Seven of 13 (54%) had medial collateral ligament injuries with 7 of 13 (54%) also demonstrating meniscal tears (5 medial, 2 lateral). All 13 (100%) demonstrated posterolateral corner injury (9 fibular collateral ligament, 3 biceps femoris, 9 popliteus, 11 popliteofibular ligament, 5 arcuate/capsule complex, and 3 lateral gastrocnemius tendon/muscle).
Conclusion: Our preliminary data supports that the diagnosis of a Segond fracture should raise suspicion not only for anterior cruciate ligament injury, but also for injury to the posterolateral corner structures.
277. MRI Sensitivity of Meniscal Tear Detection in the Setting of Acute ACL Tears: Has Detection of Lateral Meniscal Tears Improved?
Blackmon G. B.*; Garrett W. E.; Helms C. A. Duke University Medical Center, Durham, NC
Address correspondence to G. Blackmon (garyunblackmon{at}hotmail.com)
Objective: It has been shown that sensitivity for meniscal tears (particularly in the posterior horn of the lateral meniscus) decreases when a concomitant ACL tear is present. We postulated that knowing this and having an increased suspicion for these tears would lead to a higher sensitivity, even approaching the sensitivity for meniscal tears in a knee without an ACL tear. We investigated whether or not the sensitivity of detecting a lateral meniscal tear, particularly in the posterior horn, continues to remain decreased in the setting of an acute ACL tear.
Materials and Methods: A retrospective review of our database from January 2005 through July 2006 identified 85 patients who had a knee MRI demonstrating an acute ACL tear and arthroscopy on the affected knee within 6 months following the MRI. The sensitivity and specificity of meniscal tear detection was performed on the medial and lateral menisci. Additionally, the distribution of the missed tears in the lateral menisci was plotted. All studies were performed on a 1.5-T magnet using a standard imaging protocol and interpreted by fellowship-trained musculoskeletal radiologists. Surgery was performed by multiple board certified orthopedic surgeons at a single institution.
Results: The sensitivity and specificity for lateral meniscal tear detection was 76% and 98%, respectively. The sensitivity and specificity for medial meniscal tear detection was 95% and 96%, respectively. There were a total of 10 missed lateral meniscal tears and the distribution was as follows: 3 in the anterior horn, 3 in the body of the meniscus, 4 in the posterior horn, and the location of 1 of the tears was not clearly detailed in the operative report.
Conclusion: Despite heightened awareness of potential missed tears in the lateral meniscus, and particularly with the posterior horn lateral meniscus, in the setting of an acute ACL tear, the sensitivity for lateral meniscal tears remains significantly lower than medial meniscal tears.
278. Association of Peripheral Vertical Meniscal Tears With ACL Tears
Gage J. A.*; Lacy J. N.; Vinson E. N.; Cothran R. L. Duke University, Durham, NC
Address correspondence to J. Gage (jeff.gage{at}gmail.com)
Objective: The purpose of this study is to describe a type of meniscal tear seen on magnetic resonance imaging (MRI), the peripheral vertical tear, and to determine the incidence of ACL tears in patients with this type of meniscal tear.
Materials and Methods: Two hundred consecutive knee MRI examinations with meniscal tears were retrospectively reviewed by two musculoskeletal-trained radiologists in consensus to assess the location and type/morphology of the meniscal tear(s). The status of the ACL (intact, torn/deficient, or reconstructed) was also assessed. Operative notes of those patients who underwent surgery following their MRI were reviewed to determine the status of the menisci and ACL. The incidence of peripheral vertical tears among meniscal tears was determined. The incidence of ACL tears in patients with peripheral vertical meniscal tears was compared with that in knees with other types of meniscal tears.
Results: A total of 246 meniscal tears were classified in our patient population (mean age 49.1 years). Nineteen peripheral vertical tears were identified in 17 patients. In 16 of these 19 tears (84.2%), there was also evidence of an ACL tear, including 9 tears coexisting with acute ACL tears, 2 with chronic ACL deficiency, and 5 with ACL reconstruction. Ten of the peripheral vertical tears in patients with ACL tear/deficiency or reconstruction involved the posterior horn medial meniscus, while 6 involved the posterior horn lateral meniscus. None of these tears involved an anterior horn or body. Of the three patients with a peripheral vertical tear and an intact ACL, one involved the anterior horn of the lateral meniscus and two involved the posterior horn of the medial meniscus. Of the 227 meniscal tears not of the peripheral vertical type, there was evidence of a torn, deficient, or reconstructed ACL in only 17 (7.5%), which represents a statistically significant difference (p < 0.001).
Conclusion: Peripheral meniscal tears have been shown to be more common in the unstable knee. It has also been previously shown that there is an increased incidence of posterior horn peripheral meniscal tears in the ACL deficient knee, and that these tears may be difficult to detect in the setting of an acute ACL injury. Our findings indicate that peripheral vertical meniscal tears, particularly when involving the posterior horns, are highly associated with ACL tear, deficiency or reconstruction. Therefore, the finding of a peripheral vertical meniscal tear on MRI should prompt close inspection of the ACL for evidence of acute or chronic injury.
279. Injury to the Infrapatellar Fat Pad in Acute ACL Injury as a Potential Cause of Fat Pad Scarring
Toomayan G. A.*; Robertson F.; Major N. M. Duke University Medical Center, Durham, NC
Address correspondence to G. Toomayan (glen.toomayan{at}duke.edu)
Objective: Prior research has focused on anterior cruciate ligament (ACL) reconstructive surgery as a stimulus for the development of arthrofibrosis of the knee. We hypothesized that there is an association between injury to the infrapatellar fat pad at the time of ACL rupture and later development of scarring in the fat pad, as may be seen in patients who develop arthrofibrosis.
Materials and Methods: IRB approval was obtained and HIPAA compliance was maintained throughout this retrospective study. Review of our MRI database identified 53 knees over a 10-year period with findings of ACL rupture on initial MRI and at least one subsequent MRI of the same knee. Knees were divided into two groups: (1) ACL reconstruction with bone-patellar tendon-bone or hamstring graft between initial and follow-up MRI and (2) non-operative treatment. Sagittal fast spin echo T2-weighted images (TR 4000-6000, TE 60-80) were assessed for linear or globular high intensity signal in the infrapatellar fat pad on initial MRI considered to represent fat pad injury and linear or globular low intensity signal in the infrapatellar fat pad on subsequent MRI considered to represent scarring or fibrosis.
Results: Fifty three knees were imaged (43 ACL reconstruction, 10 non-operative treatment) with an average interval between initial and follow-up MRI of 762 days. Initial MRI revealed fat pad injury in 24 knees and follow-up MRI demonstrated fat pad scarring in 23 knees. Logistic regression demonstrated that the association of injury to the fat pad with later scarring was statistically significant (p < 0.01), even after adjusting for treatment group. Odds ratio for the effect of fat pad injury on scarring while controlling for surgery was 5.5. ACL reconstruction as an independent variable was not associated with later scarring (p = 0.19).
Conclusion: Imaging findings of injury to the infrapatellar fat pad are common in patients with acutely ruptured ACL. There is an association between these findings and later development of imaging evidence of fat pad scarring which is independent of treatment (ACL reconstruction or non-operative). A patient with injury to the fat pad is approximately six times more likely to later develop fat pad scarring than a patient without fat pad injury. While the diagnosis of arthrofibrosis remains a clinical one, we believe that injury to the infrapatellar fat pad at the time of ACL injury contributes to the later development of fat pad scarring.
280. MR Imaging Following Arthroscopic Repair of Medial Meniscal Root Tear
Shim J.*; Kim J.; Gwon D.; Lee G.; Kim H. Inje University Seoul Paik Hospital, Seoul, South Korea
Address correspondence to J. Shim (promind{at}lycos.co.kr)
Objective: Medial meniscal root tear can be repaired arthroscopically. The purpose of this study was to evaluate postoperative MR imaging findings in cases of arthroscopically repaired medial meniscal root tear.
Materials and Methods: We retrospectively reviewed MR images of the knees of 17 patients performed from March 2005 to September 2006, including three men and fourteen women with an average age of 58 years (range, 44-65 years). MR imaging was performed between three and twenty months after arthroscopic repair. For MR imaging analysis of the degree of root tear healing, we divided the root into three zones according to Cooper's classification. We also analyzed the signal intensity at repair sites and the presence of medial meniscal extrusion. Follow-up arthroscopy was done in three patients between one day and six months after postoperative MRI.
Results: The incidence of root continuity was 100% (17/17) in zone 1, 53% (9/17) in zone 2, and 29% (5/17) in zone 3, revealing decreased healing from zone 1 to zone 3. All 17 patients showed increased signal intensity at sites of repair, and medial meniscal extrusion was seen in 12 of 17 patients (71%). Follow-up arthroscopic findings done for three knees corresponded with MR findings.
Conclusion: We postulate that poor root healing in zone 3 is associated with avascularity. MR imaging following arthroscopic repair of medial meniscal root tears demonstrated a high incidence of increased signal intensity at sites of repair and a high frequency of medial meniscal extrusion.
281. Internal Derangement Criteria for Meniscal Tear of the Knee at Low Field Strength: Correlation of Findings on a 0.3-T Open MRI with Arthroscopy
Bhagat N. N.*; Zoga A. C.; Morrison W. B.; Bergin D. Thomas Jefferson University, Philadelphia, PA
Address correspondence to N. Bhagat (nitesh.bhagat{at}mail.tju.edu)
Objective: Standard criteria for meniscal tear were defined at 1.5 T. We sought to document criteria for meniscal tear on a low field (0.3-T) MRI system and to evaluate the usefulness of secondary signs.
Materials and Methods: Two musculoskeletal radiologists independently reviewed 61 0.3-T MR exams of the knee for meniscal tear (location, size, type, and number of slices with surfacing signal) as well as secondary signs (subchondral edema, periligamentous fluid, extrusion, and overlying cartilaginous defects. Results were compared to arthroscopy.
Results: In the 61 exams (M:F = 23:38, mean age 50, range 16-73), 43 medial meniscus tears (MMT) and 24 lateral meniscus tears (LMT) were found at arthroscopy. Most tears involved the posterior horn (80%) and average number of segments involved was 1.5; the majority of tears were complex (40%). Overall sensitivity/specificity of reader 1 vs. reader 2 for MMT was 95/76% vs. 79/88%; for LMT, 67/91% vs. 71//91%. Sensitivity/specificity for MMT on PD, T1, STIR and T2 for abnormal surfacing signal on at least one slice was 88/69%, 72/69%, 74/81%, and 40/94%. For abnormal surfacing signal visualized on two or more slices, the sensitivity/specificity was 65/94%, 63/88%, 49/88%, and 18/100%. For LMT sensitivity/specificity of PD, T1, STIR and T2 was 83/74%, 71/91%, 67/97% and 38/97% for abnormal surfacing signal on at least 1 slice and 23/94%, 54/97%, 42/97% and 21/100% for two or more slices. Regarding secondary signs, sensitivity/specificity of adjacent subchondral bone marrow edema, periligamentous fluid and meniscal extrusion was 67/94%, 63/88% and 40/94% for MMT, and 25/97%, 8/100%, and 4/97% for LMT.
Conclusion: Criteria for meniscal tear may need to be altered for exams acquired on low field MRI (e.g., surfacing signal on one slice = tear) to achieve clinically acceptable sensitivity. Secondary signs may improve specificity if used in conjunction with primary criteria.
282. MRI of the Painful Knee Using GRE In- and Out- of Phase Sequence
Huang B.2*; Pierre-Jerome C.1; Tidland M.1; Gothlin J.1; Monu J.2 1. Sahlgrenska-Molndal University Hospital, Gothenburg, Sweden; 2. University of Rochester Medical Center, Rochester, NY
Address correspondence to B. Huang (brady_huang{at}urmc.rochester.edu)
Objective: MRI using gradient recalled echo in and out of phase (GRE I-OP) sequence has been shown to be useful in imaging the spine. The current study assesses the usefulness of the GRE I-O P technique in evaluating the knee.
Materials and Methods: MRI examination was performed on 115 symptomatic knees (65 men, 50 women) using GRE I-OP, T1W and fat-saturated T2W spin echo (SE) sequences. The images were analyzed for presence and extent of osteochondral injuries. The lesions were first stratified into three categories based on location including osseous, chondral, and osteochondral varieties. Then the SE and the GRE I-OP images were compared for lesion conspicuity and image acquisition times.
Results: The patients were aged between 18 and 85 years. The GRE I-OP images were acquired in less than 3 minutes. The GRE I-OP images were superior in showing the chondral and osteochondral lesions. The chondral lesions appeared as: a) chondral defect (25 cases), b) chondral fissure (11 cases) and c) chondral thinning (34 cases). The osteochondral lesions were either: a) concave osteochondral fracture (8 cases), b) in form of osteochondritis dissecans (5 cases). Osseous lesions with edema were better visualized with the fat-saturated spin echo T2W images. Five categories of osseous lesions were detected: a) lesions with edema only (55 knees), b) lesions with edema and fracture line (32 knees), c) lesions with fracture line without edema (25 knees), d) cystic lesions with edema (18 cases), e) cystic lesions without edema (9 cases).
Conclusion: The GRE in- and out- of phase is a fast and reliable sequence. It improves visualization of osteochondral lesions and can be safely added to the repertoire of MRI sequences for knee examination.
283. Intraosseous Ganglia of the Knee: MR Imaging Characteristics
Kirkpatrick C. T.*; Morrison W. B.; Zoga A. C.; Gopez A.; Bergin D.; Deely D. M. Thomas Jefferson University Hospital, Philadelphia, PA
Address correspondence to C. Kirkpatrick (ctkirkpatrick92{at}hotmail.com)
Objective: To determine the prevalence and location of intraosseous ganglion cysts of the knee and identify associated pathology in adjacent structures.
Materials and Methods: A keyword search of knee MRI reports for "Intraosseous ganglion" or "Intraosseous cyst" was performed. Images were evaluated for: location, size, lobulation, surrounding marrow edema, contiguous soft tissue component, joint effusion, Baker's cyst, and pathology in adjacent structures. Knees with extensive osteoarthritis, SONK, or discrete cartilage defect overlying the cyst were excluded.
Results: 105 cysts were found in 95 knees (M/F = 52/43; mean age 49, range 28–76). The most common locations were medial to the PCL insertion (37%), the anterior tibia centrally at the ACL insertion (21%), and lateral to the ACL insertion (9.5%). Other locations included the lateral femoral notch at the ACL origin (8.6), proximal tibiofibular joint (6.7%), semimembranosus insertion (3.8%), PCL insertion lateral to midline (2.9%), ACL insertion medial to midline (1.9%), and origin of the medial head of the gastrocnemius tendon (1.9%). Mean size was 7 x 9 x 11 mm. Lobulation was: none (21%), mild (43%), moderate (29%), marked (7%). Complexity was mild (42%), moderate (47%), marked (11%). Marrow edema was noted around 60 ganglia (57%) extending an average of 5mm. 27 (26%) had soft tissue extension, on average 7 x 13 x 19 mm. Joint effusion was absent in 30%, small in 47%, moderate in 16%, and large in 7%. Baker's cyst was absent in 49%, small in 44%, and moderate in 7%. When ganglia were adjacent to the PCL, it was normal in 64%, degenerated in 32%, and torn in 4%. When ganglia were adjacent to the ACL, it was normal in 28%, degenerated in 70%, torn in 2% and associated with an ACL ganglion in 23%. The medial meniscus was torn in 54% when adjacent to intraosseous ganglia. The lateral meniscus was torn in 38% when adjacent to intraosseous ganglia.
Conclusion: The most common location for intraosseous ganglion cysts of the knee is the posterior tibial plateau medial to the PCL insertion. Intraosseous ganglia are frequently associated with adjacent ligament degeneration, meniscal tear, or soft tissue ganglion.
284. Incidental Enchondromas of the Knee
Walden M. J.3*; Murphey M. D.1 1. Armed Forces Institute of Pathology, Washington, DC; 2. Uniformed Services University of Health Sciences, Bethesda, MD; 3. Walter Reed Army Medical Center, Washington, DC
Address correspondence to M. Walden (mandj02{at}comcast.net)
Objective: To describe the incidence of enchondromas on routine knee MR in a large patient series.
Materials and Methods: We retrospectively reviewed 449 consecutive routine knee MR examinations for identification of enchondromas. MR was considered positive for enchondroma upon identification of focal marrow replacement with lobular margins on T1-weighted images and corresponding hyperintensity on T2-weighted images. Lesions with similar characteristics that were subchondral in location or had associated overlying cartilaginous defects were excluded. Patients with lesions that met these imaging criterion were further evaluated for patient demographics, evidence of endosteal scalloping, lesion size, concurrent findings of internal derangement, relationship to the physeal plate, axial location in the marrow and radiographic correlation (if available).
Results: We found 14 (3.1%) enchondromas about the knee in 13 (2.9%) patients (1 patient had 2 enchondromas, both in the distal femur). There were 9 cases in the femur (2.0%), 3 in the tibia (0.7%) and 1 in the fibula (0.2%). Size range varied from 0.3 x 0.3 x 0.4 cm to 4.0 x 2.2 x 2.6 cm (average 1.9 x 1.2 x 1.3 cm). The lesions were located in the metaphysis in 71% (n = 10), in the epiphysis in 7% (n = 1) and in the diaphysis in 21% (n = 3). The lesions were located centrally in the medullary canal in 57% (n = 8) and eccentrically in 43% (n = 6). The lesions abutted the physeal plate in 43% of cases. The remaining 57% of lesions were not adjacent to the physeal plate, with 50% within 1.2 cm and 50% greater than 1.2 cm from the growth plate. Radiographs were available for correlation in 3 cases. In 1 case, typical ring and arc chondroid matrix was seen and in the other 2, the chondroid rests were too small to be identified. No aggressive imaging characteristics to suggest chondrosarcoma were seen. All patients with enchondromas had accompanying abnormalities found on MR that corresponded to the patient's complaints and accounted for the imaging performed.
Conclusion: Incidental enchondromas can be identified in 2.9% of routine MR knee examinations. These are most frequently encountered in the distal femur (2.0%), a figure which is higher than seen in autopsy series. This may be due to higher sensitivity of MR for finding small cartilage rests. The frequency of these lesions about the knee is important to recognize to avoid confusion with other pathologic entities.
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
