MRI Spine: Normal Anatomy & Sequences

Why MRI?

Standard Routine Protocol

Individual Sequences Explained

T1 Weighted Imaging

• Parameters: short TR (300-700 ms), short TE (<30 ms), spin-echo (SE)
• Bright (high signal): fat, fatty bone marrow, subacute blood
• Dark (low signal): CSF, water, oedema, most tumours
• Best for: bone marrow, disc anatomy, osseous structures, soft-tissue planes
• Normal bone marrow is bright due to its fat content; a dark vertebral body on T1 should raise concern for infiltrative disease

T2 Weighted Imaging

• Parameters: long TR (2000-3000 ms), long TE (80-120 ms); fast spin-echo (FSE/TSE) is used in practice
• Bright: CSF (the hallmark), normal hydrated discs, free water, oedema
• Dark: cortical bone, ligaments, dehydrated/degenerate discs, fibrocartilage
• The spinal cord and nerve roots appear at intermediate signal, creating maximum contrast against the bright CSF - this "myelographic effect" is the reason T2 is so valuable
• Note: on FSE sequences fat remains relatively bright, which can mask vertebral metastases

STIR (Short Tau Inversion Recovery)

• Fat-suppression technique that nulls fat signal by choosing the appropriate inversion time (TI)
• Synergistic effect: prolonged T1 and T2 in abnormal tissue both increase signal, giving very high lesion-to-background contrast
• Excellent for: Modic changes, facet joint degeneration, vertebral metastases, occult fractures, acute vs. chronic fractures, multiple sclerosis plaques, ligamentous injury
• Advantage: works in low-field-strength magnets, insensitive to field inhomogeneity, suitable for large FOV spine coverage
• Disadvantage: noisier images, longer acquisition (more motion artefact in thoracic spine), cannot be combined with Gd (would suppress enhancing lesions)

Frequency-Selective Saturation (Dixon / FS T2)

• Radiofrequency pulse tuned to lipid resonant frequency suppresses fat specifically
• More complete fat suppression than STIR but requires high field strength and uniform field - less reliable at the periphery of large FOV
• Can be added to T1 (pre-Gd) to detect occult vertebral fractures

SPAIR (Spectral Presaturation Attenuated Inversion Recovery)

• Combines elements of frequency-selective sat and STIR
• Better image quality and SNR than STIR with shorter acquisition time
• Preferred over STIR for spine at 1.5T and 3T when field homogeneity is reliable

Gradient-Echo (GRE / GRE T2*)

• Shorter TE available compared to SE - reduces CSF pulsation artefacts
• Particularly useful in the cervical and thoracic spine
• Especially sensitive to haemosiderin/haemorrhage - a prognostic marker after cord injury
• Limitations: lower contrast for intramedullary lesions than FSE; susceptible to metal artefact (avoid near implants)

Gadolinium-Enhanced T1

• Standard dose: 0.1 mmol/kg IV
• Enhancement appears as T1 bright signal where the blood-brain barrier is disrupted or where tissue is hypervascular
• Key uses:
  • Distinguish postoperative scar (enhances immediately, uniformly) from recurrent disc (peripheral/delayed enhancement)
  • Tumour and infection delineation
  • Leptomeningeal disease
• Pre- and post-contrast T1 images over the same region are essential for comparison

DWI (Diffusion-Weighted Imaging)

• Less established in the spine than brain due to field inhomogeneity, small cord diameter, and respiratory/cardiac motion
• Applications: spinal cord infarction, active MS lesions, cord compression assessment, drop metastases (in children), post-traumatic injury
• DTI (Diffusion Tensor Imaging) can evaluate structural integrity of white matter tracts - still largely a research tool

FLAIR

• Suppresses CSF signal (CSF appears black) - eliminates CSF pulsation artefacts
• Most lesions remain bright
• At 3T, T1-FLAIR has been shown to better delineate tissue interfaces (disc/CSF, soft tissue/bone) than standard T1 FSE in the lumbar spine

3D Imaging

• Uses thin slices (1-2 mm) with GRE or FSE; allows multiplanar reconstructions
• 3D FSE reduces CSF flow void artefacts
• Increasingly used for nerve root and foraminal assessment

Normal Anatomy: Signal Characteristics at a Glance

Coil & Positioning

• Use a dedicated spine surface coil (or phased-array coil) for maximum SNR and resolution
• A spatial presaturation slab is placed over the larynx/carotid (cervical) or anterior chest/abdomen (thoraco-lumbar) to reduce vascular and respiratory artefacts
• Acquisition starts with a localiser (multidirectional low-resolution sequence) to plan optimal sagittal angulation
• Axial slices are then planned from high-resolution sagittal images, targeting regions of interest

Metal Artefact