Imaging of soft tissue tumors

Abstract

The aim of this project on soft tissue tumors, was to evaluate existing imaging methods and test new Magnetic Resonance Imaging (MRI) sequences for diagnosis and assessment of cytotoxic therapy and to relate the imaging studies to cytology to explore the limitations of each procedure.

Patients and method: The studies were based on patients with a soft tissue lesion diagnosed and treated at the Orthopedic Tumor Service at the Karolinska Hospital 1990-2003. In 175 patients with lipomatous tumors the radiological findings were related to cytological and histological diagnoses. In 33 patients who had performed Dynamic MRI, three enhancement features attributed to malignancy were evaluated and related to cytological and histological diagnoses. Diffusion MRI performed in 29 patients was compared to diagnosis and therapy. In 36 patients with a soft tissue sarcoma the contrast enhancement after radiotherapy was related to histological necrosis. Finally, in 18 patients and 4 healthy volunteers, MR examinations acquired before and after fine needle aspiration biopsy (FNAB) were compared to determine if needle biopsy influences the imaging of soft tissues.

Results and conclusion: The new MRI methods, Dynamic and Diffusion MRI, were not found to have a role in the initial diagnostic workup of soft tissue tumors. Although dynamic imaging was mostly correct regarding malignancy/benignity, 3 of 20 sarcomas would have been missed. However, dynamic imaging may provide additional information when the cytological diagnosis is inconclusive. The diffusion values of benign lesions overlapped with the sarcomas and could not be used for diagnostic purposes. However, the diffusion increased in all sarcomas examined after radiotherapy, which warrants further studies of therapy assessment. MRI with static registration of contrast enhancement had a limited value for this purpose since several tumors with good therapy response enhanced extensively.

Cytology is highly accurate for the diagnosis of lipomatous tumors, but the diagnostic accuracy can be further improved if compliancy with imaging is assured. In tumors with less than 75 % fat, liposarcoma is the most likely diagnosis. A cytological diagnosis of atypical lipomatous tumor (ALT) or lipoma should not be considered compliant with radiology unless the FNAB has been acquired from the least fatty part of the tumor. For lesions with 75-95% fat, liposarcoma is unlikely, but biopsy is still indicated for safety. In lesions with radiological feature similar to subcutaneous fat, biopsy is only indicated if the differentiation between lipoma and ALT influences the treatment strategy.

The risk of peritumoral bleeding after FNAB appears small and manageable. Provided there is close cooperation between the orthopedic surgeon, radiologist and cytologist, FNAB can be performed before MR examination without adverse consequences. However, it is important that the needle path is chosen in co- operation with the orthopedic surgeon who will perform the definitive surgery. Furthermore, the cytological diagnosis must articulate with clinical and MRI findings. If not, a repeat FNAB or other biopsy procedures are indicated. If MRI shows tumor heterogeneity such as represented by lipomatous tumors it is especially important to assure that the different parts of the lesion are adequately sampled.