Presentation Format: Oral presentation.
Purpose or Learning Objective: Diagnosis of joint invasion (JI) in patients with malignant bone tumors: value and
reproducibility of direct and indirect signs on magnetic resonance imaging (MRI).
Methods or Background: For this retrospective study, MR images of patients with malignant bone tumors with
(n = 24) and without (n = 24) JI, who had undergone surgical resection at our institution, were analyzed
by three radiologists independently. Readers assessed direct (intrasynovial tumor
tissue [ITT], intra-articular destruction of cartilage/bone, and invasion of capsular/ligamentous
insertions) and indirect (tumor size, bone marrow replacement/edema of epiphyseal/transarticular
bone, synovial contrast enhancement, and joint effusion) signs of JI. Histopathologic
assessment of resected specimens was used as the standard of reference. Odds ratios
(ORs), sensitivity, specificity, positive predictive value (PPV) and negative predictive
value (NPV), and inter- and intraobserver reproducibility (Fleiss's and Cohen's κ)
were calculated for each feature. Diagnostic performance of combinations of direct
features was also assessed.
Results or Findings: A total of 48 patients (28.7 ± 21.4 years of age; 26 men) were evaluated. Seven tumors
were located at the shoulder, 14 at the hip, and 27 at the knee. Osteosarcomas (n = 30) were the most frequent entity, followed by chondrosarcomas (n = 11). All readers assessed the presence of JI with a high sensitivity (92–100%)
and specificity (88–100%). Best predictors for JI were direct visualization of ITT
(OR: 186.23–228.84; p < 0.001) and destruction of intra-articular bone (OR: 68.82–323.91; p < 0.001). Direct visualization of ITT was also highly reliable in assessing JI (sensitivity,
specificity, PPV, NPV: 92–100%, respectively), with excellent interobserver reproducibility
(κ = 0.83). Synovial contrast enhancement and epiphyseal bone marrow replacement were
the most sensitive indirect signs (87–96%) but lacked specificity (29–54%). By combining
highly specific direct signs (visualization of ITT, capsular and ligamentous invasion),
sensitivity was increased (96%) and specificity (100%) was maintained.
Conclusion: Joint invasion by malignant bone tumors can be assessed reliably on preoperative
MR images with excellent sensitivity, specificity, and reproducibility. In particular,
direct visualization of intrasynovial tumor tissue and destruction of intra-articular
bone was most valuable, whereas indirect signs were less predictive and specific.
Combining multiple signs with high specificities may further increase sensitivity
while maintaining high specificity. These results will be essential for orthopaedic
surgery because they may help prevent inadequate resections and postoperative complications.
