LETTER TO THE EDITOR
Letter to the Editor concerning ‘‘C5 palsy following anterior
decompression and spinal fusion for cervical degenerative
diseases’’ by Hashimoto M et al. (2010) Eur Spine J 19:1702–1710
Hiroyuki Yoshihara
Received: 2 November 2010 / Accepted: 19 December 2010 / Published online: 4 January 2011
Ó Springer-Verlag 2010
Dear Editor,
With interest I read the article by Hashimoto et al. [1]
describing C5 palsy following anterior decompression and
spinal fusion for cervical degenerative diseases. C5 palsy is
still a controversial entity for which we do not have a clear
understanding of the pathophysiology. There are several
potential hypotheses, which have been discussed for more
than a decade. The two most-commonly proposed expla-
nations are tethering of nerve roots and localized reperfu-
sion spinal cord injury. Many articles, which have
described C5 palsy are from Japan leading some spine
surgeons in other countries to regard C5 palsy as a com-
plication of laminoplasty. Please allow me to communicate
my unique viewpoint regarding C5 palsy.
I have been trained in spinal surgery both in Japan and
the US and noticed several important differences in the
treatment of cervical spine disease. Anterior cervical
decompression and fusion (ACDF) is much more popular
than laminoplasty in the US and ACDF cases for treatment
of cervical radiculopathy is much more common in the US.
One principle difference in postoperative care between
Japan and the US is the length of the hospital stay. In the
US, patients are usually discharged 1 or 2 days after sur-
gery compared with around 7 days in Japan. After post-
operative day 1 or 2, muscle strength is not usually tested
by surgeons in the US. On the other hand, patients have
motor strength testing by their surgeons every day for
1 week in Japan. C5 palsy often occurs 2–7 days after
surgery. Shoulder elevation may not be important to
patients during short-term recovery period after ACDF and
so patients in the US may never notice weakness of the
deltoid muscle. The C5 palsy often recovers quickly
enough that strength can also recover without notice. I
believe early hospital discharge contributes to the rare
reports of C5 palsy in the US.
I feel laminoplasty is not favored in the US partially due
to the perceived increased risk of C5 palsy. But, as
Hashimoto et al. [1] mentioned, the incidence of C5 palsy
after ACDF was 4.3% on average and ranged from 1.6 to
12.1%, similar to the incidence after laminoplasty although
the number of reports describing anterior surgery was
smaller than for posterior surgery. If tethering of nerve
roots is the cause of C5 palsy, this cannot explain C5 palsy
after ACDF because the spinal cord does not shift poste-
riorly after surgery. Imagama et al. [2] recently studied the
clinical features and radiological findings of C5 palsy in
patients after cervical laminoplasty and reported that a
group of patients who developed C5 palsy had significant
narrowing of the intervertebral foramen of C5 after lam-
inoplasty. But ACDF surgery usually opens up the foramen
via indirect decompression, which also does not explain
why patients have a similar rate of C5 palsy after ACDF.
Chiba et al. [3] studied C5 palsy patients after expansive
open-door laminoplasty using magnetic resonance imaging
(MRI) and suggested that a certain impairment in the gray
matter of the spinal cord may play an important role in C5
palsy. Hasegawa et al. [4] studied C5 palsy patients with
chronic cervical cord compressive lesions who underwent
decompression surgery and compared the incidence of C5
palsy among anterior and posterior procedures. They found
no differences among procedures and concluded that C5
palsy might result from a transient and localized spinal
cord lesion caused by reperfusion after decompression of a
chronic compressive lesion. Hashimoto et al. [1] did not
H. Yoshihara (&)
University of Colorado Denver, 12631 E. 17th Avenue,
Academic Office 1, Room 4501, Aurora, CO 80045, USA
e-mail: hiroyoshihara55@yahoo.co.jp;
hiroyuki.yoshihara@ucdenver.edu
123
Eur Spine J (2011) 20:1188–1189
DOI 10.1007/s00586-010-1674-5