Clear separation of fat and water signals is essential in musculoskeletal MRI because both dominate the appearance of bone marrow, soft tissues and fluid-sensitive abnormalities. Until recently, technical constraints limited routine fat-water separation despite its recognised diagnostic value. Advances in sequence design and reconstruction have enabled the Dixon method to be combined with fast or turbo spin-echo imaging, which is widely used in clinical practice. This approach provides uniform fat suppression, multiple image contrasts from a single acquisition and access to fat-specific information. As a result, Dixon imaging is increasingly used in bone marrow and spine imaging and is expanding into whole-body MRI, rheumatology, neuromuscular disorders and imaging near metallic implants.
Integration With Routine Spin-Echo Imaging
Spin-echo sequences remain central to musculoskeletal MRI because they provide reliable tissue contrast and are relatively insensitive to magnetic field inhomogeneity. Fat suppression is often required to highlight increased water content associated with many pathological processes, while preserved fat signal remains important for lesion characterisation. Common approaches include chemical shift selective techniques and inversion recovery methods, each with practical limitations.
The Dixon method separates fat and water by acquiring images at echo times where the two signals are in-phase and out-of-phase. These data are then reconstructed into water-only and fat-only images. Unlike suppression-based methods, Dixon resolves both signals within each voxel, allowing direct assessment of tissue composition. Although originally used with gradient-echo imaging, Dixon has been adapted to fast spin-echo sequences, making it suitable for routine musculoskeletal protocols.
In clinical use, Dixon provides more consistent fat suppression than chemical shift selective methods, especially over large fields of view or in areas affected by field inhomogeneity. It can also reduce radiofrequency energy deposition because additional fat saturation pulses are not required. Compared with inversion recovery techniques, Dixon may offer higher signal-to-noise ratio by retaining both fat and water signal.
Protocol Efficiency and Fat-Specific Information
A single Dixon acquisition produces four coregistered image sets: in-phase, out-of-phase, water-only and fat-only images. When applied to fluid-sensitive sequences, this allows simultaneous access to anatomical information, fat-suppressed images and fat-specific contrast. This versatility supports protocol simplification by reducing the need for multiple separate sequences.
Must Read: High-Yield Musculoskeletal Ultrasound for Emergency Care
In spine imaging, a single T2-weighted Dixon sequence can provide the contrasts often obtained from several sagittal acquisitions. In-phase images support anatomical assessment, while fat-only images can replace conventional T1-weighted imaging in selected situations. This approach has been used in degenerative spine assessment to reduce scan time while maintaining diagnostic coverage.
Fat-only images are specific to fat signal rather than simply fat-sensitive. This is particularly important in bone marrow imaging, where identifying residual intralesional fat helps differentiate marrow-replacing from non–marrow-replacing lesions. Lesions lacking fat signal appear dark on fat-only images, while fat-containing lesions remain bright. Quantitative assessment of fat content, using fat fraction or signal drop between in-phase and out-of-phase images, can support interpretation in equivocal cases but is considered complementary rather than definitive.
Clinical Uses and Limitations
Bone marrow lesion assessment is a key application, including evaluation of vertebral compression fractures. A single T2-weighted Dixon sequence can support morphological assessment and provide information on fat content that contributes to distinguishing benign from malignant fractures.
In rheumatologic imaging, Dixon supports evaluation of inflammatory and structural changes. Water-only images highlight active inflammatory signal, while fat-only images show fat replacement and structural abnormalities. Similar benefits have been described in hand MRI for rheumatoid arthritis, where reliable fat suppression supports scoring systems and shorter protocols.
Whole-body MRI is well suited to Dixon because large field-of-view examinations often require multiple sequences. A single T2-weighted Dixon acquisition can replace combinations of T1-weighted and fat-suppressed sequences, reducing examination time. In neuromuscular disorders, water-only images show muscle signal abnormalities, fat-only images depict fatty replacement and in-phase images support muscle volume assessment.
Dixon also supports musculoskeletal tumour imaging, particularly with gadolinium administration. T1-weighted Dixon sequences provide consistent precontrast and postcontrast datasets, and fat-only images remain unaffected by contrast, aiding tissue characterisation.
Limitations include fat-water swapping artefacts, which can misassign fat and water signals, and India ink artefact on out-of-phase images, which may mimic pathology. Acquisition times are typically longer than standard fast spin-echo imaging, limiting use where very high spatial resolution is required. Quantitative fat assessment is sensitive to technical factors and lacks standardisation, so results should be interpreted cautiously.
Dixon-based fast spin-echo imaging has become an important tool in musculoskeletal MRI by combining uniform fat suppression, multiple coregistered contrasts and fat-specific information in a single acquisition. These features support more efficient protocols and improved lesion assessment, particularly in bone marrow and spine imaging, with expanding applications in whole-body, rheumatologic and neuromuscular imaging. Awareness of artefacts, scan time and quantitative limitations remains essential for appropriate clinical use.
Source: Radiology
Image Credit: iStock
