PURPOSE: To differentiate benign from malignant bone tumors by analyzing the vascular distribution within bone tumors with dynamic contrast-enhanced MRI. METHODS: We studied dynamic contrast-enhanced MRI for 49 bone tumors (22 malignant and 27 benign tumors). Seven small regions of interest (ROI) were set inside the largest portion of each tumor. Four ROI were placed evenly on the periphery and three ROI were placed evenly on the line of the longest breadth within the tumor. The slope of the curve (%Slope) was calculated on the time-intensity curves of the whole tumor and of each ROI. The variance values for the %Slope of the ROI were calculated to assess the dispersion of the intensity change at each ROI within the tumor. RESULTS: Mean value of the %Slopes of whole tumor regions for malignant bone tumors (70.4 +/- 60.3%) was significantly higher than that for benign bone tumors (37.6 +/- 52.9%) (P = 0.015), although giant cell tumor (GCT), a locally aggressive tumor, had a relatively higher %Slope. Mean value of the variance of %Slopes for malignant bone tumors (3485.9 +/- 5942.5) was significantly higher than that for all benign tumors (470.4 +/- 583.9) (P = 0.012), indicating that the %Slope values of seven ROI within malignant bone tumors varied more widely compared with the ROI inside benign bone tumors. GCT also demonstrated a lower value. CONCLUSION: Our method of analyzing the signal intensity change at seven separate regions that evaluates the vascular distribution within a tumor could be a useful tool for differentiating between benign and malignant bone tumors.
PURPOSE: To differentiate benign from malignant bone tumors by analyzing the vascular distribution within bone tumors with dynamic contrast-enhanced MRI. METHODS: We studied dynamic contrast-enhanced MRI for 49 bone tumors (22 malignant and 27 benign tumors). Seven small regions of interest (ROI) were set inside the largest portion of each tumor. Four ROI were placed evenly on the periphery and three ROI were placed evenly on the line of the longest breadth within the tumor. The slope of the curve (%Slope) was calculated on the time-intensity curves of the whole tumor and of each ROI. The variance values for the %Slope of the ROI were calculated to assess the dispersion of the intensity change at each ROI within the tumor. RESULTS: Mean value of the %Slopes of whole tumor regions for malignant bone tumors (70.4 +/- 60.3%) was significantly higher than that for benign bone tumors (37.6 +/- 52.9%) (P = 0.015), although giant cell tumor (GCT), a locally aggressive tumor, had a relatively higher %Slope. Mean value of the variance of %Slopes for malignant bone tumors (3485.9 +/- 5942.5) was significantly higher than that for all benign tumors (470.4 +/- 583.9) (P = 0.012), indicating that the %Slope values of seven ROI within malignant bone tumors varied more widely compared with the ROI inside benign bone tumors. GCT also demonstrated a lower value. CONCLUSION: Our method of analyzing the signal intensity change at seven separate regions that evaluates the vascular distribution within a tumor could be a useful tool for differentiating between benign and malignant bone tumors.
Authors: R Erlemann; M F Reiser; P E Peters; P Vasallo; B Nommensen; C R Kusnierz-Glaz; J Ritter; A Roessner Journal: Radiology Date: 1989-06 Impact factor: 11.105
Authors: K L Verstraete; A Dierick; Y De Deene; D Uyttendaele; F Vandamme; H Roels; M Kunnen Journal: Magn Reson Imaging Date: 1994 Impact factor: 2.546
Authors: G Pan; A K Raymond; C H Carrasco; S Wallace; E E Kim; A Shirkhoda; N Jaffe; J A Murray; R S Benjamin Journal: Radiology Date: 1990-02 Impact factor: 11.105
Authors: W D Zimmer; T H Berquist; R A McLeod; F H Sim; D J Pritchard; T C Shives; L E Wold; G R May Journal: Radiology Date: 1985-06 Impact factor: 11.105
Authors: P Lang; G Honda; T Roberts; M Vahlensieck; J O Johnston; W Rosenau; A Mathur; C Peterfy; C A Gooding; H K Genant Journal: Radiology Date: 1995-12 Impact factor: 11.105