Nondiagnostic Lumbar Spine MRI from Titanium Hardware Artifact
A 62-year-old woman with a two-level lumbar fusion (L4–L5) using titanium pedicle screws and interbody cages is referred for MRI because of new right L5 radicular pain. The fusion was 18 months ago. At an outside hospital, the radiologist reported that the sagittal T2-weighted images were 'nondiagnostic at L4–L5' because of a large signal void obscuring the thecal sac and right foramen. She now presents to your 1.5 T site. The ordered protocol is a routine lumbar spine without contrast: sagittal T1, sagittal T2, axial T2, and STIR, 25 minutes total. The referring surgeon wants to know whether there is recurrent disc herniation, stenosis, or hardware complication. The patient has no contraindications, but she is in pain and cannot lie still for more than about 20 minutes without moving. The scanner has standard spine coils but no dedicated metal-artifact-reduction software package; however, you can change bandwidth, echo time, echo train length, slice thickness, and matrix.
1. Why does the artifact occur, and which sequence parameter changes can reduce it on this scanner?
Read the worked answer
Titanium is less ferromagnetic than stainless steel, but it still distorts the local magnetic field. That distortion causes frequency shifts, signal void, and bright pile-up around the screws, obscuring the thecal sac and foramen. The effect is worse with gradient-echo sequences, long echo times, low bandwidth, and thicker slices.
Practical changes on a 1.5 T scanner without dedicated metal-artifact software:
- Increase receive bandwidth to reduce frequency-encoding displacement.
- Shorten echo time.
- Use fast spin echo rather than gradient echo.
- Increase echo train length to keep scan time acceptable.
- Use thinner slices, such as 3 mm, and a higher matrix to improve spatial resolution.
- Avoid unnecessary STIR if it is degraded; a fat-saturated FSE may be more useful.
- If available, use view-angle tilting or a metal-artifact-reduction sequence, but do not assume it is present.
Do not simply enlarge the field of view; that lowers resolution without correcting the distortion. At 1.5 T the artifact is generally less severe than at 3 T, which is an advantage here.
2. How would you prioritize the protocol and manage the patient's comfort and motion?
Read the worked answer
Run the most diagnostically important sequences first while she is most comfortable: a high-bandwidth sagittal T2 FSE through the lumbar spine, then focused axial T2 slices at L4–L5, then sagittal T1. If time remains, add a fluid-sensitive sequence with the same artifact-reduction settings. Use 3 mm slices, a matrix appropriate for the coil, and the shortest echo time that preserves contrast.
Position her with a pillow under the knees to reduce lumbar lordosis and pain. Explain that the scan is broken into short runs and that she can signal if she needs a break. Check in between sequences, but keep breaks brief to avoid losing the slot. If she moves, repeat only the essential sequence.
Tell the radiologist in advance that metal-artifact-reduction parameters were used and that residual distortion may limit evaluation at the hardware level. Ask whether the surgeon's question is primarily neural compression, hardware complication, or adjacent-segment disease, because that determines which sequences matter most. Document the parameters changed and any residual limitation.
3. If the artifact remains nondiagnostic, what alternatives should be discussed?
Read the worked answer
Do not keep repeating the same sequence with the same parameters; that adds time and motion without new information. If the optimized FSE images still obscure the canal and foramen, discuss alternatives with the radiologist and surgeon:
- A focused CT of the lumbar spine to assess screw position, loosening, and bony anatomy; CT is less affected by titanium artifact when appropriate techniques are used.
- CT myelography if neural compression must be assessed and MRI remains nondiagnostic, recognizing that it is invasive.
- Referral to a site with dedicated metal-artifact-reduction sequences such as SEMAC or MAVRIC if those are clinically necessary.
- Clinical correlation and further evaluation for hardware infection or loosening, including laboratory tests and imaging as directed.
Document that the study was limited by susceptibility artifact despite parameter optimization, and specify which levels were obscured. The goal is not to produce a perfect image at any cost, but to give the surgeon reliable information or a clear next step.