Peer review improves MSK tumor diagnosis

Researchers in Belgium have found that cooperation and consensus between radiologists and pathologists bolsters the diagnosis of musculoskeletal soft-tissue tumors when MR and histology results are contradictory.

Belgium's incidence of benign and malignant MSK soft-tissue tumors such as intramuscular myxomas or myxofibrosarcomas is low, preventing investigators from gaining insight into these conditions. Research studies are scarce, and diagnosis, characterization, and staging methods are not standardized. In addition, little exchange between local research institutions exists, said principal investigator Dr. Jan L. M. Gielen, a radiologist at Antwerp University Hospital (UZA). Gielen spoke at a scientific session of the European Congress of Radiology meeting in March.

The Belgian Soft Tissue Neoplasm Registry (BSTNR) was created in 2001 by the UZA's radiology department and all MRI centers throughout the country. It provided the first 170 cases of lesions diagnosed as malignant by MRI despite having inconclusive histology. The BSTNR also assembled a peer-review group of six pathologists led by Dr. Raf Sciot from the University of Leuven.

Gielen and colleagues analyzed the benefit of having radiologists work closely with the peer-review group in all cases in which MRI and pathology results were discrepant. They found that the multidisciplinary approach achieved the highest level of accuracy for unresolved diagnosis of these tumors.

The group confirmed 73% of diagnoses and changed 27% of them. Peer-review results provided sensitivity, specificity, positive- and negative-predictive value of 88%, 96%, 93.5%, and 92.5%, respectively. All findings were statistically significant (p

Cases originally diagnosed as benign that changed to malignant were relevant because they had the potential to worsen prognosis. False negatives, however, did not bear such significance, Gielen said.

Cooperation between pathologists and radiologists reinforces the diagnosis of soft-tissue tumors, a trend several researchers mentioned in regard to bone tumor protocols. The findings suggest that all cases with MR and pathology discrepancies should be presented to a peer-review committee, which can provide a weighty second opinion to pathologists and help strengthen the scientific value of the existing literature, Gielen said.

For more information from the Diagnostic Imaging archives:

MR surpasses CT for detecting bone metastases

Radiologists look over their shoulders, knees, and hips

Low-dose multislice CT makes gains in bone marrow diagnosis

Musculoskeletal tumors call for team approach