MR Neurography of the Lumbosacral Plexus
The lumbosacral plexus (LSP) consists of lumbar and sacral plexus. The lumbar plexus is formed by the ventral rami of TH12, L1, L2, L3 and L4 nerve roots from the level of the L2 through L5 transverse processes, whereas the sacral plexus includes the ventral rami of L4, L5, S1, S2 and S3 nerve roots. Lumbosacral plexus provides both motor and sensory innervation to pelvis and lower extremity.
Diagnosing lumbrosacral plexopathy was mainly based on medical history, clinical findings and electromyography (EMG) testing, while Computed Tomography (CT) was used to assess tumours and guide biopsies.
MR Neurography is a non-invasive imaging technique for the dedicated assessment of peripheral nerves. Because of its excellent soft-tissue contrast, MRN offers anatomical information that is not obtainable with other modalities (fig. 1).
Nowadays, MRN plays a prominent role in the evaluation of lumbosacral plexus and its related pathology, thanks to the recent advances in hardware, software and development of new imaging techniques.
Nevertheless, MRI examinations of the lumbosacral plexus remain challenging due to technical limitations, such as magnetic field inhomogeneities, big FOV and strong blood signal which can obscure or mimic pathology.
Fig. 1: Coronal thin-section MIP 3D DESS (A) and sagittal thin-section MIP 3D DESS (B). These reconstruction planes clearly demonstrate the LSP and its components, providing great anatomical detail. Image dataset acquired at 3 Tesla. Images courtesy of Bac Nguyen.
The MR Neurography of the lumbosacral plexus can be very useful in the following clinical indications:
Surgical and radiation treatment planning
Leg pain, sacral or coccyx pain
Unknown neuropathy or plexopathy
LSP evaluation in patients with indeterminate results at lumbar spine MRI
Differentiation between lumbrosacral plexopathy and spine-related abnormalities (fig. 2)
Planning for MRI–guided administration of pain medicatiom
Fig. 2: Axial T2 FSE (A), sagittal T2 FSE (B), thin-section coronal MIP 3D DESS (C) and curved planar reconstruction 3D DESS (D). 3D DESS apparently points out that the lesion is not an LSP pathology but is related to the lumbar spine. Image dataset acquired at 1.5 Tesla. Images courtesy of Dimitris Priovolos
3.1 Field strength
3T scanners are preferred because they provide greater SNR compared to 1.5T, which can result in higher resolution images and/or faster acquisition times.
1.5T should be used when scanning patients with metallic implants.
3.2 Imaging plane
Traditional LSP protocols include all three imaging planes (axial, coronal, and sagittal).
Coronal is the most important plane because it depicts all nerves together and demonstrates their longitudinal extent.
The lumbosacral plexus is better examined with a conjunction of morphological and functional imaging techniques (multiparametric imaging).
3.3.1 Gold standards
Conventional 2D T1 FSE and T2 FSE with and without fat suppression are routinely used in MRN of the LSP. STIR or Dixon/IDEAL techniques provide robust fat suppression, whereas chemical fat suppression technique often leads to imperfect fat saturation due to big FOV and magnetic field inhomogeneities.
3D fat-suppressed T1 GRE is a good alternative for thin-slice imaging, before and after the administration of contrast (Gadolinium is mainly used for the assessment of tumours and inflammations).
Modified 3D FSE sequences enables the acquisition of high-resolution, isotropic datasets (the term isotropic means that the voxel is uniform in all directions) with contrast similar to the conventional 2D FSE sequences, within clinically acceptable acquisition times and without SAR limitations (fig. 3).
Fig. 3: Coronal MIP 3D STIR (A) and multiplanar reconstruction (MPR) in axial (B) and sagittal (C) plane. Three-dimensional modified STIR acquisition enables the acquisition of high-resolution, 3D datasets of the LSP with image contrast similar to the conventional 2D FSE sequences, within clinically acceptable acquisition times and without SAR limitations. Unfortunately though, they suffer from strong perineural signal which can lead to interpretation pitfalls. Image dataset acquired at 1.5 Tesla. Images courtesy of Bac Nguyen.
3.3.2 Additional sequences
Three-dimensional Double Echo Steady State sequence with water excitation technique (3D DESS WE) can provide high-quality, isotropic images of the LSP (fig. 4). 3D DESS WE offers the advantage of nulled signal from the adjacent vessels, cerebrospinal fluid (CSF) and fat, which can greatly help in the differentiation between lumbrosacral plexopathy and spine-related abnormalities (fig. 2). Unfortunately, DESS is not available from all MR vendors, therefor its applications are limited.
3D SSFP-echo DWI is a steady-state GRE sequence. Water excitation technique and low b value (80-90 s/mm2) are used in order to null the signal from the fat and vessels. The main advantage of 3D SSFP-echo DWI is the easy generation of high-quality, high-contrast isotropic images of the LSP, in clinically acceptable acquisition times.
Three-dimensional nerve-SHeath signal increased with INKed rest-tissue rarE Imaging (3D SHINKEI) enables nerve-selective images with increased depiction of smaller lumbosacral plexus branches. Many papers establish that 3D SHINKEI offers better nerve SNR and CNR compared to modified 3D STIR. Unfortunately, 3D SHINKEI is not available from all MR vendors, therefore its applications are limited.
Fig. 4: Coronal MIP 3D DESS (A) and multiplanar reconstruction (MPR) in sagittal (B) and axial (C) plane. 3D DESS sequence with water excitation technique provides high-resolution, isotropic imaging of the lumbosacral plexus, with excellent visualization of the nerves. Image dataset acquired at 1.5 Tesla. Images courtesy of Bac Nguyen
3.3.3 Functional techniques
Functional sequences, such as DWI and DTI, can also be helpful in the MR Neurography of the lumbosacral plexus.
Diffusion-Weighted Imaging (DWI) is used to evaluate random, thermally-induced, microscopic water motion within a tissue, well known as Brownian motion. Diffusion-weighted imaging (DWI) offers the advantage of suppressed background structures and flowing blood in vessels, thereby selectively highlights the nerves with high signal intensity (fig. 5). STIR technique is mainly used to provide robust fat suppression.
Diffusion Tensor Imaging (DTI) is an extension of DWI that takes into account and measures the anisotropic diffusion (biological tissues are highly anisotropic). DTI can provide useful structural information about nerves (fig. 6).
Nonetheless, DWI and DTI have a relatively low spatial resolution, can be time consuming and may suffer from image distortions, limitations that cannot be neglected in imaging of fine structures such as peripheral nerves.
Fig. 5: Coronal, sagittal and axial thin-section MIPs of a Diffusion-weighted Imaging (DWI) STIR acquisition, obtained with a b value of 800 s/mm2. DWI STIR offers the advantage of suppressed background structures (fat, muscles) and flowing blood in vessels, thereby selectively highlights the nerves with high signal intensity. Image dataset acquired at 1.5 Tesla.
Fig. 6: DTI (fiber-tracking map) of the sacral plexus (left) and sciatic nerve (right). DTI depicts the anisotropic diffusion and can provide useful structural information about the nerves. Image dataset acquired at 3.0 Tesla. Images courtesy of Bac Nguyen.
3.4 Image Post-processing
Due to the obliquity of plexus branches, the longitudinal extent of LSP and its related pathology is difficult to be determined on direct coronal, sagittal or axial images. Multiplanar reconstruction (MPR), curved planar reconstruction (CPR) and maximum intensity projection (MIP) are used to highlight the peripheral nerves in their entirety. For MPR and CPR, 3D isotropic (or nearly isotropic) acquisitions should be obtained.
LIMITATIONS AND SOLUTIONS
Artifacts and pitfalls are common in the MRN of the LSP, resulting in degraded image quality and incorrect image interpretation. Specifically, MRN can suffer from:
Increased perineural signal, originated from the adjacent vessels. Modified 3D FSE acquisitions are mainly affected by strong vessels signal. The use of additional sequences (DESS and/or DWI) can address this disadvantage (fig. 7).
Inhomogeneous fat suppression, when chemical fat saturation techniques are used. STIR and Dixon techniques can resolve this issue, offering robust fat suppression.
Susceptibility artifacts, especially at 3T. FSE acquisitions with the lowest possible TE and high bandwidth (rBW) can reduce this type of artifacts. Patients with metallic implants should be scanned at 1.5T.
Magic angle effect. This effect can be avoided with higher TE values or patient re-positioning.
Chemical shift artifacts, especially at 3T. FSE acquisitions, in-phase TEs, high rBW and fat suppression should be selected in order to overcome chemical shift artifacts.
Fig. 7: Coronal MIP 3D DESS (A) and coronal MIP 3D STIR (B). 3D DESS provides better visualization of the lumbosacral plexus, while the strong vessels signal can obscure or mimic pathology in 3D STIR imaging. Image dataset acquired at 1.5 Tesla. Images courtesy of Bac Nguyen.
MR Neurography is a great modality for the high-quality, high-resolution imaging of the lumbosacral plexus.
Anatomical and functional sequences can be used for the assessment of the LSP and its pathology, providing great anatomical information and evaluation of lesions' extent.
Finally, MRN is a valuable adjunct to clinical examination and EMG but Radiographers and Radiologists should be aware of potential artifacts and pitfalls which can lead to degraded image quality and incorrect image interpretations.
1. Mürtz P, Kaschner M, Lakghomi A, Gieseke J, Willinek WA, Schild HH, Thomas D. Diffusion-weighted MR neurography of the brachial and lumbosacralplexus: 3.0 T versus 1.5 T imaging. Eur J Radiol. 2015; 84(4): 696-702.
2. Delaney H, Bencardino J, Rosenberg ZS. Magnetic Resonance Neurography of the Pelvis and Lumbosacral Plexus. Neuroimaging Clin N Am. 2014; 24(1): 127-50.
3. Chhabra A, Subhawong TK, Bizzell C, Flammang A, Soldatos T. 3T MR neurography using three-dimensional diffusion-weighted PSIF: technical issues and advantages. Skeletal Radiol. 2011 Oct;40(10):1355-60.
4. Cho Sims G, Boothe E, Joodi R, Chhabra A. 3D MR Neurography of the Lumbosacral Plexus: Obtaining Optimal Images for Selective Longitudinal Nerve Depiction. AJNR Am J Neuroradiol. 2016
5. Soldatos T, Andreisek G, Thawait GK, Guggenberger R, Williams EH, Carrino JA, Chhabra A. High-Resolution 3-T MR Neurography of the Lumbosacral Plexus. Radiographics 2013; 33(4): 967-87.
6. Chhabra A, Faridian-Aragh N. High-Resolution 3-T MR Neurography of Femoral Neuropathy. AJR Am J Roentgenol. 2012; 198(1): 3-10.
7. Zhang ZW, Song LJ, Meng QF, Li ZP, Luo BN, Yang YH, Pei Z. High-Resolution Diffusion-Weighted MR Imaging of the Human Lumbosacral Plexus and Its Branches Based on a Steady-State Free Precession Imaging Technique at 3T. AJNR Am J Neuroradiol. 2008; 29(6): 1092-4.
8. Kasper JM, Wadhwa V, Scott KM, Rozen S, Xi Y, Chhabra A. SHINKEI - a novel 3D isotropic MR neurography technique: technical advantages over 3DIRTSE-based imaging. Eur Radiol. 2015; 25(6): 1672-7.
9. Cejas C, Escobar I, Serra M, Barroso F. High resolution neurography of the lumbosacral plexus on 3 T magnetic resonance imaging. Radiologia 2015; 57(1): 22-34.