Magic Angle Effects in MR Neurography
Karyn E. Chappella,
Matthew D. Robsonc,
Amanda Stonebridge-Fostera,
Alan Glovera,
Joanna M. Allsopd,
Andreanna D. Williamsd,
Amy H. Herlihyd,
Jill Mossb,
Philip Gishena and
Graeme M. Bydder
a Department of Imaging, Hammersmith Hospital NHS Trust, London, England
b Department of Histopathology, Hammersmith Hospital NHS Trust, London, England
c Oxford University Centre for Clinical Magnetic Resonance Research, MRS Unit, John Radcliffe Hospital, Oxford, England
d The Robert Steiner Magnetic Resonance Unit, Imaging Sciences Department, MRC Clinical Sciences Centre, Imperial College School of Medicine, London, England and University of San Diego, CA
View larger version (95K):
[in a new window]
|
FIG 1. A and B, Transverse STIR (2500/30/160 TR/TEeffective/TI) MR images of the median nerve at 0° (A) and 55° (B). The median nerve (arrow) has a higher signal intensity in B. There is no apparent change in signal intensity in the flexor tendons (see also Fig 3).
| |
View larger version (17K):
[in a new window]
|
FIG 2. Plot of signal intensity against TE at 55° ( ) and 0° (•) on STIR images (1500/22, 33, 44, 55, 66/107 TR/TE/TI) with a monoexponential fit, in an adult volunteer. The signal intensity is higher at 55°. The mean T2 was also longer at 55° than at 0° (65.8 msec versus 47.2 msec).
| |
View larger version (23K):
[in a new window]
|
FIG 3. Plots of signal intensity against orientation for the median nerve and hypothenar muscle and for an adjacent flexor tendon. Same subject as in Fig 1.
A, The median nerve shows a 98% increase in signal intensity, which peaks at about 60° and decreases as the angulation is increased to 90°. Muscle shows no significant change in signal intensity.
B, The adjacent flexor tendon, examined with the same sequence and plotted on the same scale, follows the same pattern but has a lower initial signal intensity and shows less change.
| |
View larger version (100K):
[in a new window]
|
FIG 4. A and B, STIR images of the brachial plexus (A) and nerves entering upper arm (B). The components of the brachial plexus (arrows in A) have a higher signal intensity than that of skeletal muscle, whereas nerves in the upper arm (arrows in B), emerging from the brachial plexus and nearly parallel to B0, are isointense or slightly hyperintense to muscle.
| |
View larger version (82K):
[in a new window]
|
FIG 5. Sagittal MR images of the ulnar nerve with the elbow flexed to 125°.
A, This most medial section shows that the nerve is isointense to muscle in the upper arm, but at the level of the condyle (arrow) signal intensity increases as the nerve rotates toward 55°.
B, Middle section shows the nerve (arrow) perpendicular to B0 where it is isointense to muscle.
C, Lateral section shows the nerve (arrows) at 125° to B0 where it is hyperintense to muscle.
| |
View larger version (148K):
[in a new window]
|
FIG 6. Sagittal MR image of the median nerve with the wrist flexed to 55°. The nerve is parallel to B0 in the upper forearm and is isointense with muscle, but where the nerve is flexed toward 55° it shows increased signal intensity (arrow). This may simulate disease.
| |
View larger version (131K):
[in a new window]
|
FIG 7. A and B, Transverse MR images of the sciatic nerve orientated at 0° to B0 (A) and 55° to B0 (B). The signal intensity in the nerve (arrow) in B is higher than that in A.
| |
View larger version (162K):
[in a new window]
|
FIG 8. Electron micrograph of a human peripheral nerve (stain was aqueous uranyl nitrate followed by Reynolds lead citrate; original magnification, X 14,000). The numerous small dots are collagen fibers seen in cross section.
| |
View larger version (63K):
[in a new window]
|
FIG 9. Diagram of brachial plexus. A black line at 55° to the body axis and B0 is superimposed. The components of the plexus are generally orientated parallel to this line, which is at the magic angle. (Reprinted from reference 33 with permission of Icon Learnings Systems.)
| |
View larger version (15K):
[in a new window]
|
FIG 10. Fit of median nerve, muscle, and flexor tendon data to the model described in the Appendix. There is a close fit for median nerve and flexor tendon, assuming a significant fraction of tissue with dipolar interactions. The muscle data fit without need for significant dipolar interactions.
| |
