DOI:10.2214/AJR.07.2033
AJR 2007; 189:1175-1178
© American Roentgen Ray Society
Axial Compression Frame for MRI of Thoracolumbar Spine
Daniel S. J. Choy1,
Gian Paolo Tassi2 and
Richard B. Libes1
1 Laser Spine Center and Columbia University, 66 E 80th St., Suite 1A, New York,
NY 10021.
2 Casa di Cura Villa Anna, San Benedetto del Tronto, Pescara, Italy.
Received February 8, 2007;
accepted after revision May 18, 2007.
Address correspondence to D. S. J. Choy
(info{at}laserspinecenter.com).
Abstract
OBJECTIVE. Our objective was to present a method of performing
thoracolumbar MRI with intervertebral disk pressure at 150 kPa without the
patient being seated.
CONCLUSION. Spine MRI with compression is more physiologic and will
produce a higher yield than standard supine MRI.
Keywords: compression • lumbar spine • MRI • percutaneous laser disk decompression • spine
Introduction
All MRI is performed on patients with suspected herniated disk disease with
the patient in the supine position and is therefore nonphysiologic. Patients
are generally most comfortable in the supine position and most uncomfortable
in the standing or sitting position. Nachemson and Morris
[1] showed intradiskal pressure
of the lumbar spine to average 15–20, 100, and 150 kPa in the supine,
standing, and sitting positions, respectively. Ideally, then, patients with
suspected herniated disk disease should be imaged sitting, when the
intradiskal pressure is highest. In 1996, Jolesz showed augmentation of disk
protrusion during MRI in the sitting position compared with one in the supine
position (Jolesz F, presented at the 1996 annual meeting of the Laser
Association of Neurosurgeons International) (Figs.
1,
2,
3). However, at the time, only
two MRI sitting scanners were available worldwide, and they cost $5 million
each.
View larger version (137K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 1 —MR image in patient having symptoms compatible with
L5–S1 disk herniation shows slight bulge (arrow) of L5–S1
disk. (Reprinted with permission from Choy DSJ. Percutaneous laser disc
decompression: a practical guide. New York, NY: Springer-Verlag,
2003:126–127 [5])
|
|
View larger version (188K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 2 —MR image of patient in
Figure 1 obtained 5 minutes
later. Note increased protrusion of disk (arrow). (Reprinted with
permission from Choy DSJ. Percutaneous laser disc decompression: a
practical guide. New York, NY: Springer-Verlag, 2003:126–127
[5])
|
|
View larger version (94K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 3 —Sitting MRI scanner (Flexview 8800, GE Healthcare) with which
image in Figure 2 was produced.
(Reprinted with permission from Choy DSJ. Percutaneous laser disc
decompression: a practical guide. New York, NY: Springer-Verlag,
2003:126–127 [5])
|
|
To obtain the same physics benefits without the expense, one of the authors
designed and built for $75 a wood compression frame that, with a patient lying
on the frame, would fit in a standard MRI bore
(Fig. 4). Its use is described
in the Materials and Methods section.
View larger version (7K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 4 —First axial compression frame built of hard marine-grade
plywood, with shoulder restraints and a movable footboard with hardwood dowel
fixation. (Reprinted with permission from Choy DSJ. Percutaneous laser
disc decompression: a practical guide. New York, NY: Springer-Verlag,
2003:126–127 [5])
|
|
After 9 years of use, a more advanced, professionally manufactured aluminum
frame with a pressure gauge was created
(Fig. 5A). The purpose of this
article is to describe both frames and their applications.
Materials and Methods
In the wood frame, the patient lies on the frame with the knees flexed so
that the axial dimension from the shoulders to the feet is shortened by 4
inches (10 cm). The footplate is then fixed with hardwood dowels. The patient
and frame are then positioned in the MRI bore. MRI is performed with the
patient's knees flexed (no compression), then with the knees fully extended
(axial compression). This frame was used for more than 10 years in 143
patients.
In the aluminum frame, the patient lies on the frame with the legs fully
extended and the feet in contact with the footplate. A shoulder harness
spreads the pressure over the shoulders and trapezius and is connected to the
footplate with nondistensible straps. Precompression MRI is performed. Imaging
under compression is achieved by a pump mechanism shortening the distance
between the footplate and shoulder harness to raise lumbar intradiskal
pressure to 150 kPa.
MRI was performed on a 1.5-T unit (Signa, GE Healthcare) using a standard
surface coil. Sagittal T2 fast spin-echo sequences were obtained with and
without compression. The imaging parameters were as follows: TR/TE, 4,000/170;
256 x 256 matrix; 4 signals averaged; 30-cm rectangular field of view;
and 5-mm slice thickness with a 1-mm gap. After applying compression, a
sagittal fast spin-echo T1 sequence was performed to determine the superior
offset required because of the patient's change in position. All other
parameters remained the same.
A radiologist identified any bulging or herniated disks and compared
qualitatively the degree of protrusion between compressed and noncompressed
images. In addition, any change in the patient's symptoms was recorded after
the compression sequence.
Ten patients about to undergo percutaneous laser disk decompression had
18-gauge needles inserted into their L4–L5 disks under aseptic
conditions with local anesthesia and C-arm monitoring. The patient
demographics were seven men, age range, 27–64 years; and three women,
age range, 39–75 years. There were disk herniations of L4–L5 in
seven, L5–S1 in two, and L3–L4 in one. The patients' heights
ranged from 1.6 to 1.8 m, and weights ranged from 58 to 73 kg.
With the patient in the frame, a needle was filled with sterile saline and
connected to an IC912/VI pressure gauge (Eliwell-Invensys) with nondistensible
plastic tubing 4 mm in diameter (Rilsan PA 11 DIN 74324 ATM 66; 2.8 mm in
lumen diameter, Omnexus). Data were obtained with ascending and then
descending readings of intradiskal pressures at intervals of 30 kPa from 20 to
230 kPa, and corresponding footplate pressures were expressed in psi. Thus, it
was possible to obtain two curves with each patient.
One of the authors (height, 162.6 cm; weight, 59 kg) volunteered for the
first compression in the aluminum frame with 189 lb (85.91 kg) footplate
pressure (equivalent to 150 kPa in L4–L5).
Results
Two models were fit to obtain estimates for footplate pressure as a
function of intradiskal pressure. In the first model, footplate pressure was
fit as a quadratic function of the kPa value. In the second model, the
log10 of footplate pressure was fit as a quadratic function of kPa
value. For the second model, estimates of the footplate pressure at the 150
kPa value were back-transformed.
The following estimates were obtained for kPa equal to 150. For the area of
interest, the quadratic model appeared to fit somewhat closer to the observed
means than the log10 model: The quadratic model estimate was 189.4
and the 95% CI was 178.4–200.3. The log10 model estimate was
199.0 and the 95% CI was 187.5–211.2. The composite curves derived from
the intradiskal and footplate pressure study are seen in
Figure 6.
View larger version (9K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 6 —Graph shows composite curves for observed means of
intradiskal pressure (kPa) and corresponding footplate pressure (kg). In
lumbar disk, 150 kPa corresponds to 189 lb (85.91 kg) of foot pressure. Black
line = raw means, short dashed line = quadratic model, long dashed line = log
model.
|
|
Representative MR images of patients with and without axial compression
with the wood frame are seen in Figures
7A,
7B and
8A,
8B. The control
(noncompression) and compression MR images are shown. An increase of disk
bulging with compression can be seen. Testing the aluminum frame in a GE
Healthcare MRI machine (Flexview 8800) at 1.5 T produced no effects on the T2
image.
View larger version (62K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 7A —Representative MRI of lumbosacral spine. Sagittal T2 images
obtained with wood frame. A was obtained without compression and
B was obtained with compression. Marks indicate disk bulges.
|
|
View larger version (71K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 7B —Representative MRI of lumbosacral spine. Sagittal T2 images
obtained with wood frame. A was obtained without compression and
B was obtained with compression. Marks indicate disk bulges.
|
|
View larger version (70K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 8A —Representative MRI of lumbosacral spine. Sagittal images
obtained with aluminum frame. A was obtained without compression and
B was obtained with compression. Marks indicate disk bulges.
|
|
View larger version (88K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 8B —Representative MRI of lumbosacral spine. Sagittal images
obtained with aluminum frame. A was obtained without compression and
B was obtained with compression. Marks indicate disk bulges.
|
|
With the wood frame (n = 143 patients), 70 (49%) patients reported
exacerbation of back or sciatic pain, and in 32 (22%), there was observable
augmentation of disk herniation. In 48 (30%), there was an increase of
symptoms without a change in imaging. In 10 (7%), there was an increase in
imaging but no change in symptoms, In 22 (15%), there were both an increase in
symptoms and an increase in imaging. Six patients could not be compressed
because of body habitus. In the 10 years of performing spine MRI under
compression, no neurologic complications occurred.
At 189-lb (85.91-kg) footplate pressure, our volunteer's height (162.56 cm)
shortened by 1.3 cm (0.8%). No symptoms were reported. T1 and T2 images showed
no change.
Discussion
When herniated disk disease is suspected in a patient, it is obvious that
the optimal MRI should be performed under conditions when the patient
experiences the most pain, and this is either in the sitting or standing
position. MRI performed in the supine position, when the patient is most
comfortable and the intradiskal pressure is lowest, is nonphysiologic.
It has been our experience that many equivocal MRI examinations can be
converted to positive examinations with axial compression. Moreover, there is
a bonus in that many patients report exacerbation of their sciatic pain. In
this respect, the frame confirms the origin of the patient's pain as
diskogenic.
Although the aluminum frame is new and no extensive experience has been
obtained with it, it is superior to the wood frame in that, based on in vivo
data, we can obtain targeted lumbar intradiskal pressures of 150 kPa. We
expect to obtain similar if not superior data with this frame. The intradiskal
pressures can be achieved and are reproducible.
Shortening our volunteer's height by 1.3 cm with 189-lb (85.91-kg)
footplate compression with the aluminum frame represents a change of 0.8%.
This is within the range reported by Kimura et al.
[2] in patients axially loaded
with 50% body weight. Probably contributing to this shortening are compressive
changes in the knee, hip, and sacroiliac joints. It can be extrapolated that
intervertebral disk compression contributes to total change. Axial
compression, by augmenting the image of disk protrusion and reproducing the
patient's pain pattern, can provide the spine surgeon with additional data to
justify an interventional procedure.
Our central thesis that axial spine compression during MRI can contribute
to the overall evaluation of the patient with suspected disk herniation
disease is confirmed by prior work by Danielson et al.
[3] and Hargens et al.
[4].
We dispute the use by Kimura et al.
[2] of 50% of the patient's
body weight based on cadaver studies as a clinically meaningful compression of
the lumbar disks. We believe that our actual measure of disk pressure in kPa
confers a greater degree of accuracy.
The second-generation aluminum compression frame with a pressure gauge has
obvious advantages over the original wood frame and will now serve as our
instrument of choice for both thoracolumbar spine MRI and CT in patients with
suspected herniated disk disease.
In conclusion, in our experience over a period of 10 years, we have found
axial compression MRI of the thoracolumbar spine in cases of suspected
herniated disk disease to be useful in generating more meaningful diagnostic
data in terms of augmentation of disk bulge and reproducing pain patterns. It
has been completely safe. There are sufficient scientific and clinical bases
for these results. The new aluminum frame does not affect the MR images. We
believe all MRI and CT of the spine in suspected herniated disk disease should
be performed with axial compression.
References
- Nachemson A, Morris JM. In vivo measurements of intradiskal
pressure: discometry, a method for determination of pressure in the lower
lumbar discs. J Bone Joint Surg Am 1964;46
:1077
–1092[Abstract/Free Full Text]
- Kimura S, Steinbach GC, Watenpaugh DE, Hargens AR. Lumbar spine
disc height and curvature responses to an axial load generated by a
compression device compatible with magnetic resonance imaging.
Spine 2001; 26:2596
–2600[CrossRef][Medline]
- Danielson BI, Willen J, Gaulitz A, Niklason T, Hansson TH. Axial
loading of the spine during CT and MR in patients with suspected lumbar spinal
stenosis. Acta Radiol 1998;39
: 604–611[Medline]
- Hargens AR, Hutchinson KJ, Ballard RE, Murthy G. Intervertebral
disc: loaded on earth and unloaded in space. In: Reed R, Rubin K, eds.Connective tissue biology, vol. 7, Integration and
reductionism
. London, United Kingdom: Portland Press,1998
- Choy DSJ. Percautaneous laser disc decompression: a
practical guide. New York, NY: Springer-Verlag,2003
: 126–127
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
Top
Abstract
Introduction
