Disc degeneration
I would like to explore more deeply the reason why
I think a muscle approach must be considered when a patient's films
show disc degeneration. The clinician has the choice to view that
evidence of degeneration in accordance with one of two models.
Model A: Patient presents with pain. An X-ray
or MRI shows a disc problem. The clinician can conclude that pressure
on the nerve root is causing the pain.
Here's a diagram of this model:
Clinical approach: NSAIDs, cortisone injection, sometimes
surgery, "don't do what causes you pain."
Yet Rene Caillet, M.D. wrote about a study of degenerative
thoracic discs and pain: "Only 29% of asymptomatic patients had
normal MRI scanning of the thoracic spine, whereas 60% of these
asymptomatic patients were manifesting bulging or herniated thoracic
disks with 54% showing evidence of spinal cord indenture, with 24%
showing frank disk herniations, and with 46% having annual tears.
These authors concluded that disk herniations are part of the normal
aging process of the spine. Of symptomatic patients, 12% had normal
results on MRI scan."(7)
[Since this presentation, researchers at the Stanford
University Medical Center reported on a study of "96 patients with
known risk factors for disc degeneration. People whose discs had
high intensity zones were only slightly more likely to experience
back pain during normal activity than those without obvious disc
problems. Additionally, high intensity zones were found in 25 percent
of people who-despite their known degenerative disc disease-had
no corresponding symptoms of low back pain." This research was presented
April 12, 2000 at the annual meeting of the International Society
for the Study of the Lumbar Spine and earned the Volvo Corporation
Award for the Clinical Study of Low Back Pain. It was to be published
in the December 2000 issue of Spine.(8)]
Such studies call into question the Model A practice
of blaming pain on degenerative discs.
Model B: Patient presents with pain. Regardless
of whether an X-ray or MRI shows a disc problem, the clinician can
search for a muscle contracture that could be causing the pain.
From an X-ray or MRI, the clinician can visualize which muscles
might be contracted to create a particular joint dysfunction. If
releasing the muscle contracture eliminates the pain, consider the
possibility that the muscle contracture was putting pressure on
disc, leading to the degeneration. If the muscle contracture does
not cross a vertebral joint, consider that the afflicted distal
muscle may be affecting the segmental nerve, and the segmental nerve
is contracting the muscles along the spine and other muscles innervated
by that segmental nerve. This is the theory espoused by Chan Gunn,
who writes: "Ordinarily, when several of the most painful shortened
muscles in a region have been treated [with dry needling], pain
is alleviated in that region. Relaxation and relief in one region
often spreads to the entire segment, to the opposite side, and to
paraspinal muscles. These observations suggest that needling has
produced more than local changes-a reflex neural mechanism involving
spinal modulatory system mechanisms, opioid, or non-opioid, may
have been activated."(9)
In short, distal muscle contracture may cause contracture
of the paraspinal muscles, possibly leading to disc degeneration.
I always apply Model B when a patient presents with
pain and a diagnosis of a disc problem. I search for a muscle contracture
that might be causing the pain, and usually I find it.
If a muscle contracture leads to pain and disc degeneration,
then the approach is to treat the muscle contracture.
Here is a diagram of this model:
Approach: Treat the muscle contracture
Omitted graphic: When muscles across a disc shorten,
they compress it and at the same time, cause arthralgia in the facet
joints. Source: Gunn CG: The Gunn Approach to the Treatment of
Chronic Pain, Churchill Livingstone, New York, 1996, (p. 8).
[Permission could not be obtained for reproduction of this graphic
on a website.]
Especially suspect muscle contracture involvement
when stretching and/or massage improves the problem. If massage
makes the problem worse, there also may be muscle contracture involvement,
but the clinician may be massaging and releasing an antagonist which
is splinting against the muscle in contracture.
Case: Female, 61, a frequent recipient of Tandem
Point therapy for assorted pain patterns. She presented on July
5, 1999 with pain on the lateral aspect of the left thigh and leg
and pain in the posterior gluteal region. She reported that the
pain began after hiking 10 miles in one day, and since had "come
and gone on its own schedule." She identified the Travell & Simons
pain referral pattern for the anterior division of the gluteus minimus
as her pain pattern. Palpation found the anterior division of the
gluteus minimus to be unusually hard. Tandem Point therapy was applied
to the gluteus minimus trigger points plus Sp 6 and Sp 8, and immediately
afterward the patient reported that her pain was gone. The patient
then reported that the previously described pain pattern was the
same pain pattern she had suffered before she had two back surgeries
ten years earlier. Before those surgeries she had received no physical
therapy, her physicians had not palpated her muscles, nor had they
ordered an EMG to determine any muscle involvement. Follow-up: This
patient received some further work on her gluteus minimus in two
of her four sessions between July 5, 1999 and February, 2000. On
February 14, 2000, she had no pain in the lateral aspect of her
left thigh or leg.
HOME
| TABLE OF CONTENTS | PREVIOUS
| NEXT
P.O. Box 1023, Haddonfield, NJ 08033
|
|