Medizinische Universität Graz Austria/Österreich - Forschungsportal - Medical University of Graz
Gewählte Publikation:
Brückmann, C.
Do we need bone scans for cartilage tumors?
Humanmedizin; [ Diplomarbeit ] Graz Medical University; 2018. pp. 66
[OPEN ACCESS]
FullText
- Autor*innen der Med Uni Graz:
- Gilg Magdalena Maria
- Betreuer*innen:
- Leithner Andreas
- Altmetrics:
- Abstract:
- INTRODUCTION: Cartilage tumors represent the largest bone sarcoma group and the differentiation between enchondromas (EC) and low-grade chondrosarcomas (CS) remains a diagnostic challenge. It has been described that bone scans provide further information about the malignant potential of cartilage tumors. We aimed to identify the diagnostic benefit of bone scans in cartilage tumors. MATERIAL AND METHODS: We retrospectively analyzed 461 patients with a cartilage tumor (321 EC, 54 CS, 74 osteochondromas (OC), 4 chondromyxoidfibromas (CMF), 4 chondroblastomas (CB), 3 Ollier disease) diagnosed at a mean age of 45.8 years (range 6 – 84) in whom a Tc99m bone scan (3 phasic) was performed. Bone scan data, x-rays, magnetic resonance imaging (MRI) and histopathological records were evaluated. For EC and CS diagnostic accuracy of bone scans was assessed with receiver operator characteristic curves (ROC) to evaluate the area under the curve (AUC). RESULTS: Of 375 patients suffering from EC or CS, 65 (17%) underwent biopsy and 176 (46%) surgery, while in 134 cases (35%) a conservative treatment and observation in regular intervals was chosen. The most common localization of tumors were the long bones (n=245, 65%), followed by hand and foot (n=99, 26%) and trunk (n=20, 5%). The mean tracer uptake for EC was 3.8 (range 0.5 – 68.0), for CS grade 1 4.47 (range 0,5 – 12,3), grade 2 8.2 (range 3,3 – 13,5) and grade 3 7.5 (range 6,3 – 8,9). The mean tracer uptake of EC was significantly lower than for all CS (p< 0.001), for low- grade CS (p< 0.001) respectively. In 153/321 lesions (48%) the radiologist identified EC correctly, whereas 3/54 CS (6%) were correctly identified with the bone scan compared to the definitive diagnosis. In 145 patients (39%) the bone scan was inconclusive and in 54 (14%) patients the bone scan did not show any pathological finding. On multivariate analysis tracer uptake was significantly increased with pathological fracture (OR 3.9, 95% CI 3-5.2, p<0.001) and peripheral localization of tumors (OR 2.8, 95% CI 1.5-5-2, p<0.001). The AUC for all cartilage tumors increased from 0.70 (Sp 80%, Sen. 50%) to 0.80 (Sp. 80%, Sen. 64%) when adding bone scan information. The AUC for EC and CS grade 1 increased from 0.67 (Sp 80%, Sen 46%) to 0.73 (Sp 80%, Sen 52%). DISCUSSION: We could see a statistically significant increase of tracer uptake in CS (all CS vs. EC; Grade 1 vs. EC) compared to EC. Localization and the presence of pathological fracture were significant independent predictors for higher tracer uptake. However, assessment of diagnostic performance showed, that bone scans only provide minimal additional information to select between EC and CS, low-grade CS respectively. Taking into account costs and resources to perform a bone scan as well as radiation exposure for patients, its use in the group of cartilage tumors should be reconsidered in the future.