BackgroundDespite the overall diagnostic utility of core needle biopsy (CNB) comparable to incisional biopsy, increased diagnostic errors have been suggested of CNB for myxoid soft tissue tumors. This study compared the diagnostic performance of CNB between myxoid and non-myxoid soft tissue tumors. Methods369 patients who underwent ultrasound-guided CNB prior to resection for soft tissue tumors were classified into two groups according to resection pathology; myxoid group (n = 75) and non-myxoid group (n = 294). One-hundred and ninety-three patients were male and the median age of the patients was 40 years. Two-hundred and sixty-three tumors were malignant. ResultsCNB correctly diagnosed malignancy in 84% (58 of 69) for the myxoid group and 95% (184 of 194) for the non-myxoid group. For diagnosing histologic grade of soft tissue sarcoma, CNB correctly identified high grade in 78% (18 of 23) for the myxoid group and 74% (94 of 128) for the non-myxoid group. Correct diagnosis rate of histological type was significantly lower in the myxoid group (63% [47 of 75] in the myxoid group and 83% [242 of 294] in the non-myxoid group, p = 0.013). ConclusionOur study suggests that CNB is useful for myxoid soft tissue tumors of the extremity, with regard to diagnosing malignancy and histologic grade. However, CNB was less useful for identifying histologic subtype in myxoid tumors than in non-myxoid tumors.