Diffusion-Weighted Magnetic Resonance Imaging for the Evaluation of Musculoskeletal Tumors




Conventional MR imaging provides low specificity in the differential diagnosis of musculoskeletal (MSK) tumors and is unable to offer information about the extent of tumoral necrosis and the presence of viable cells, information crucial to assess treatment response and prognosis. Therefore, diffusion-weighted imaging (DWI) is now used with conventional MR imaging to improve diagnostic accuracy and treatment evaluation. This article discusses the technical aspects of DWI, particularly the quantitative and qualitative interpretation of images in MSK tumors. The clinical application of DWI for tumor detection, characterization, differentiation of tumor tissue from others, and assessment of treatment response are emphasized.


MR imaging has become the diagnostic method of choice for preoperative and posttreatment staging of musculoskeletal (MSK) tumors because of the high resolution, tissue contrast, and multiplanar capability of this technique.


In addition, MR imaging offers several advantages when compared with other imaging methods in the evaluation and staging of soft tissue tumors. Several studies have demonstrated morphologic parameters such as size, margin demarcation, involvement of adjacent vital structures, homogeneity in signal intensity, and measurement of relaxation time as criteria to evaluate soft tissue tumors. In accordance with these criteria, malignancy can be predicted with the following parameters ( Fig. 1 ) :




  • Heterogeneous signal intensity in a T1 scan



  • Tumor necrosis



  • Bone or neurovascular involvement



  • Mean diameter of more than 66 mm.




Fig. 1


The evaluation of soft tissue tumors with conventional MR imaging using morphologic parameters. ( A ) Axial T2-weighted image. A huge tumor inside the muscle with a necrotic center and many solid nodules on the wall ( arrows ), which suggests a malignant tumor. This is a synovial sarcoma. ( B ) Axial T2-weighted image. ( C ) Axial T1-weighted image. A solid tumor hyperintense on T2-weighed image ( B , arrow ) and isointense on T1-weighted image ( C , arrow ) inside the muscle, with fat surrounding it, which is highly suggestive of a hemangioma. However, in several cases, conventional MR imaging presents low specificity in the differential diagnosis because many of the lesions present nonspecific characteristics.


However, conventional MR imaging provides low specificity in the differential diagnosis of several MSK tumors because many of the lesions exhibit nonspecific characteristics. As a result, a correct histologic diagnosis is possible in only a quarter to one-third of cases. Conventional MR imaging is unable to offer information about the extent of tumoral necrosis and the presence of viable cells, information that is crucial for the assessment of treatment response and prognosis. Therefore, advanced MR imaging techniques, such as diffusion-weighted imaging (DWI), are now used in association with conventional MR imaging with the objective of improving diagnostic accuracy and treatment evaluation. DWI allows quantitative and qualitative analyses of tissue cellularity and cell membrane integrity and has been widely used for tumor detection and characterization and to monitor treatment response ( Fig. 2 ).




Fig. 2


Tumor characterization. There are two different solid tumors in the shoulder. ( A ) Coronal T1 fat suppression contrast-enhanced image. The bigger tumor has heterogeneous contrast enhancement with edema surrounding it ( arrow ). ( B ) Axial diffusion-weighted image. ( C ) In the axial ADC map, the tumor tissue presents facilitated diffusion ( arrow ) and PIDC value = 1.52 × 10 −3 mm 2 /s, suggesting benign tumor (desmoid tumor). ( D ) Coronal T1 fat suppression contrast-enhanced image. The smaller tumor has homogeneous contrast enhancement ( arrow ). ( E ) Axial diffusion-weighted image. ( F ) In the axial ADC map the tumor presents restricted diffusion ( arrow) with a PIDC = 0.80 × 10 −3 mm 2 /s, suggesting malignant tumor (leiomyosarcoma grade 2).


The tumor tissue is usually more cellular when compared with other tissues and tends to appear at high signal intensities (restricted diffusion) when DWI is used ( Fig. 3 ).




Fig. 3


Tumor detection. A grade 2 periosteal chondrosarcoma in the humeral shaft. ( A ) Axial T2-weighted image before surgery. There is a small lesion hyperintense on the image ( circle ) in the posterior periosteum of the humeral shaft. ( B ) Axial diffusion-weighted image. The tumor is hyperintense on DWI sequence ( circle ), easier to detect because of the high tissue contrast. ( C ) Axial T2-weighted image 6 months after the surgery. There is a very small recidive of the tumor in the surgical site ( circle ), which is difficult to find on conventional MR imaging. ( D ) Axial diffusion-weighted image. The tumoral recidive shows hyperintensity on diffusion-weighted image sequence ( circle ), which facilitates the detection.


The tissue contrast obtained using DWI is different from that obtained using conventional MR imaging, which facilitates the detection of soft tissue and bone tumors, particularly bone metastasis. In fact, previous studies have concluded that DWI is an extremely sensitive method for identifying bone metastases and is superior to both positron emission tomography (PET) and scintigraphy in terms of detection capability. The detection of bone metastasis is important for cancer staging and in the determination of treatment strategy, and some reports have demonstrated whole body DWI to be highly sensitive and efficient for this purpose.


Tumors differ in cellularity characteristics, and this difference is useful in determining their histologic composition. It has been reported that DWI can differentiate benign from malignant soft tissue tumors ( Fig. 4 ).




Fig. 4


An indeterminate palpable soft tissue mass for 2 months located inside the muscle in the right arm in a 16-year-old girl. ( A ) Axial T1-weighted fat suppression postcontrast image. The tumor presents diffuse enhancement encasing the neurovascular bundle ( red arrow ) and invading the periosteal surface with edema in the medullar region ( blue arrow ). ( B ) Axial ADC map. The tumor tissue has facilitated diffusion on the ADC map ( arrow ) with PIDC = 1.9 × 10 −3 mm 2 /s, suggesting a benign tumor (ossificans myositis).


The malignant tumors have more cellularity than benign tumors and tend to have a more restricted diffusion ( Fig. 5 ).




Fig. 5


The qualitative analysis of DWI. ( A ) Axial T2-weighted image. Non–Hodgkin lymphoma in the posterior ribs in a 52-year-old man, hyperintense on T2-weighted image. ( B ) Axial diffusion-weighted image. Tumor has high cellularity (non–Hodgkin lymphoma) and less extracellular space and presents very high signal intensity on diffusion-weighted image (b = 600 s/mm 2 ) sequence. ( C ) Very low signal intensity on the ADC map (restricted diffusion). ( D ) Axial T2 fat suppression image. A desmoid tumor inside the posterior muscle compartment of the thigh, hyperintense on T2-weighted fat suppression image. ( E ) Axial diffusion-weighted image. Tumor with less cellularity and more extracellular space shows high signal intensity on diffusion-weighted image (b = 600 s/mm 2 ) sequence. ( F ) Intermediate to high signal intensity on the ADC map (facilitated diffusion).


In accordance with this finding, perfusion-corrected DWI has demonstrated potential in differentiating benign from malignant soft tissue masses. In addition, DWI is also used for differentiating between chronic expanding hematomas (CEHs) and malignant soft tissue tumors. CEHs are frequently misdiagnosed as malignant soft tissue tumors because of their morphologic characteristics, which include large size, slow progressive enlargement, and heterogeneous signal intensity on conventional MR imaging. DWI has also been shown to be an additional tool for differentiating vertebral fracture caused by osteoporotic collapse with bone marrow edema as well as pathologic collapse caused by tumor infiltration or metastatic disease.


On the other hand, some investigators have reported overlapping apparent diffusion coefficient (ADC) values in benign and malignant soft tissue tumors, which consequently could not be used to differentiate them. This overlapping is likely because of the fact that ADC values can be affected by cellularity and the extracellular matrix. For example, myxoid matrix is widely seen in the interstitial spaces in many soft tissue tumors, and this presence can influence the ADC values ( Fig. 6 ).




Fig. 6


A huge myxoid liposarcoma in the posterior muscular compartment of the thigh. ( A ) Sagittal short tau inversion recovery (STIR) image. The tumor looks like a cystic lesion on the STIR sequence, but ( B ) axial T1-weighted fat suppression postcontrast image shows contrast enhancement. ( C ) Axial ADC map. The tumor tissue has facilitated diffusion on the ADC map, with PIDC = 2.56 × 10 −3 mm 2 /s, probably because of the myxoid matrix in the extracellular space. These types of tumors will have significantly higher ADC and PIDC values than nonmyxoid malignant solid tumors. ( D ) On histologic evaluation, the final diagnosis was a myxoid liposarcoma (Hematoxylin-eosin [H&E], original magnification ×100).


As a result, myxoid tumors will have significantly higher ADC values than nonmyxoid tumors. It makes no difference if the tumor is benign or malignant.


DWI can also be used to monitor tumor response to treatment, most likely because effective anticancer therapy results in changes in the tumor microenvironment, resulting in an increase in the diffusion of water molecules and a consequent increase in the ADC value ( Fig. 7 ).




Fig. 7


Evaluation of response to mesenchymal chondrosarcoma using DWI. ( A ) Axial T1-weighted fat suppression postcontrast image. There is a big mass with heterogeneous contrast enhancement located inside the muscle in the thigh. ( B ) Axial ADC map (before treatment). The tumor presents very low signal intensity on the ADC map (restricted diffusion) and PIDC value = 0.89 × 10 −3 mm 2 /s. ( C ) Axial ADC map (3 months later). The tumor presents areas of intermediate to high signal intensity (facilitated diffusion) inside the tumor ( red arrow ) and a few areas of low signal intensity (restricted diffusion) ( blue arrow ) suggesting viable cells. ( D ) Axial ADC map (9 months later). The tumor presents high signal intensity on the ADC map (facilitated diffusion) and PIDC value = 2.8 × 10 −3 mm 2 /s suggesting good response to the treatment.


Furthermore, DWI has been used to provide information regarding cellular changes related to cytotoxic treatment in soft tissue sarcomas. Some investigators have suggested that it could be possible to evaluate the response of osteosarcoma to chemotherapy using DWI, considering that the ADC values of viable tumor tissue and tumor necrosis differ significantly. This information is a crucial prognostic factor for patients with osteosarcoma.


This article provides a short discussion of the technical aspects of DWI, particularly the quantitative and qualitative interpretation of diffusion-weighted (DW) images in MSK tumors. The clinical application of DWI for tumor detection, characterization, differentiation of tumor tissue from nontumor tissue, and assessment of treatment response are emphasized.


Technical aspects


Herein, the authors briefly discuss some important concepts regarding the specificity of DWI in the MSK system. However, for a detailed explanation of the physics of DWI see the article elsewhere in this issue. See the article by Figueiredo and colleagues elsewhere in this issues for further exploration of this topic.


DWI exploits the random motion of water molecules in the body, which is classically called the Brownian motion. In biologic tissues, the movement of water molecules is restricted because their motion is modified and limited by their interactions with cellular membranes and macromolecules. The DWI signal in vivo is therefore derived from the motion of water protons in extracellular, intracellular, and intravascular spaces.


DWI yields qualitative and quantitative information that reflects tissue cellularity and cell membrane integrity, which complements the morphologic information obtained by conventional MR imaging. Thus, the data obtained from DWI must be interpreted using qualitative and quantitative approaches.


Qualitative analysis is achieved via visual assessment of the relative tissue signal attenuation of both the DW image and the ADC parametric map. The visual assessment of DW image enables tissue characterization based on differences in water diffusion and is performed by observing the relative attenuation of the signal intensity of images obtained at different b values. In a heterogeneous tumor, for instance, the more cystic or necrotic fraction of the tumor will show greater signal attenuation on high–b value images because water diffusion is less restricted, whereas the more cellular solid tumor areas will continue to show a relatively high signal intensity ( Fig. 8 ).




Fig. 8


( A–E ) Signal attenuation of a heterogeneous tumor, with necrotic portion on axial diffusion images with different b values. The more cystic or necrotic fraction of the tumor ( long arrows ) shows greater signal attenuation on high–b value images because water diffusion is less restricted, whereas the more cellular solid tumor areas ( arrowhead ) will continue to show relatively high signal intensity.


By contrast, on the ADC parametric map, visual assessment reveals a trend opposite to that of DW images: areas of restricted diffusion in highly cellular areas appear as low signal intensity areas compared with less cellular areas, which have a higher signal intensity (see Fig. 5 ). Quantitative analysis is performed by calculating the conventional ADC value and/or perfusion-insensitive diffusion coefficient (PIDC) value. The conventional ADC value is calculated using a biexponential function from DWI, which includes low b values (b = 0–600 s/mm 2 ), or can be obtained alternatively by drawing regions of interest (ROIs) on the ADC map. However, an exponential function fitted only through the high b values (b = 300, 450, and 600s/mm 2 ) can be used to describe the PIDC value. This measurement excludes the initial reduction of signal intensity that is probably caused by vascular capillary perfusion. Consequently, for large b values, perfusion effects tend to be canceled out. The PIDC map may provide more accurate information about tumor tissue cellularity by minimizing vascular contributions, which are higher in malignant tumors ( Fig. 9 ).


Sep 18, 2017 | Posted by in MAGNETIC RESONANCE IMAGING | Comments Off on Diffusion-Weighted Magnetic Resonance Imaging for the Evaluation of Musculoskeletal Tumors

Full access? Get Clinical Tree

Get Clinical Tree app for offline access