Treating a Pinched Nerve in the Neck
Gary E Cordingley, MD, PhD
There are numerous treatments for a pinched spinal nerve in the neck.
However, rigorous evidence of their benefit is scant.
Let's suppose that you have been diagnosed as having a pinched nerve in your neck, also known as
cervical radiculopathy. If so, you probably have pain in the neck and one shoulder. The pain might radiate
into your arm and you might have weakness or numbness in the arm as well. Moving your neck in certain
positions probably worsens the pain.
If you're a younger adult, the pinch could be due to a herniated (slipped) disc. Discs are the soft spacers
that separate each pair of stacked neck-bones (vertebrae). If you're an older adult, the pinch is more likely
due to a bony spur (spondylosis). In either case, you're in good company. A survey in Sicily showed 3.5
active cases at any one time of cervical radiculopathy per population of 100,000. In Rochester, Minnesota,
another survey showed 85 new cases each year of cervical radiculopathy per population of 100,000.
Let's say that your doctor has evaluated you thoroughly by taking a history of your symptoms and
performing a physical examination. Perhaps with the additional help of an MRI of your cervical spine (neck)
and electrical tests of nerve and muscle function (nerve conduction studies and electromyography) the
diagnosis of cervical radiculopathy is deemed definite. Furthermore, there is no sign that the spinal cord
itself is pinched. Now what?
Now what, indeed! Choosing a treatment for this condition is far from straightforward. Out of hundreds of
published medical reports concerning treatment of cervical radiculopathy, most are case reports or case
series. A "case series" translates roughly as: "We gave six patients in a row the same treatment and five of
them got better." What can be concluded from a study of this kind? Did the treatment make the patients
better or would they have improved anyway? We don't know.
The missing ingredient here is a comparison group of untreated or differently treated individuals known as
a control group. The other mark of a quality study is that the chosen treatment is randomized, meaning
that the research subjects agreed in advance to be assigned to one treatment group or another based on
the equivalent of a coin-toss. So out of the hundreds of published studies involving treatment of this
common condition, how many were randomized controlled trials? Unfortunately, the answer is just one.
Liselott Persson, Carl-Axel Carlsson and Jane Carlsson at the University Hospital of Lund, Sweden,
randomly allocated 81 patients who had symptoms of cervical radiculopathy present for at least three
months to any of three treatments -- surgery, physical therapy or a cervical collar. The patients ranged
from 28 to 64 years old and 54% of them were male. The surgeons used the so-called Cloward procedure,
removing fragments of protruding discs and spurs through an incision in the front of the neck, and then
fusing two neck-bones together by means of a bone-graft. Physical therapy involved 15 sessions over a
span of three months and consisted of whatever the physical therapist considered appropriate, variously
including any of the following: heat application, cold application, electrical stimulation, ultrasound,
massage, manipulation, exercise and education. In the cervical collar group, patients wore rigid,
shoulder-resting collars every day for three months. Additionally, some of the subjects wore soft collars
How did the study turn out? Three of the subjects who were assigned to surgery refused the procedure
because they had already improved on their own. For statistical purposes their outcomes were included
with those who actually received the operation. After three months the surgery and physical therapy
groups reported, on average, less pain. After an additional 12 months patients in all three groups had less
pain than at the beginning of the study and the outcomes of each treatment were statistically alike.
Measurements of mood and overall function following treatment were likewise equal among the groups.
So, over the long haul, no treatment was better than the others. Of course, within each group some
patients did better or worse than others and this spread of outcomes was not reflected in the overall
averages. In fact, five patients in the collar group and one patient in the physical therapy group went on to
receive surgery owing to lack of satisfactory improvement. In addition, eight patients in the surgery group
underwent a second operation that in one case was due to a complication of the first operation.
With this Swedish study representing the only rigorous investigation of treatment outcomes in cervical
radiculopathy, there are a number of unanswered questions. For example, what are the effects on cervical
radiculopathy of painkillers, anti-inflammatory drugs, local injections, systematic traction or other forms of
surgery? We don't know. What happens if there is no treatment whatsoever? We don't know the answer to
that question either.
Thus, in the care of individual patients there is a yin-yang balancing act between the medical edict of
"Above all, do no harm" and the practical dictum of "Do what you have to do." This balancing act usually
means starting with less intrusive treatments like drugs and physical therapy. If symptoms fail to improve or
become unbearable, an operation may be helpful.
(C) 2006 by Gary Cordingley