Abstract
Background: Whether lumbar arthrodesis can relieve
isolated low-back pain in the absence of focal neurological findings
or instability is unclear. The results of published studies are
also inconsistent with regard to whether temporary back-pain relief
with external spinal skeletal fixation can predict lasting back-pain relief
after arthrodesis. This report presents the results, with regard
to clinical benefit and complications, of more than 100 external
spinal skeletal fixation procedures undertaken as a prelude to lumbar
arthrodesis.
Methods: The records of all patients who underwent
external spinal skeletal fixation between 1989 and 1999 were reviewed
with attention to perioperative complications, pain relief from
the test procedure, the clinical benefit from a subsequent arthrodesis,
and the functional status after the arthrodesis. Analyzed data included
the frequency of neurological complications and infections and the
benefit (Prolo score) after staged spinal arthrodesis in patients
who underwent arthrodesis after temporarily experiencing pain relief
with the test procedure.
Results: A total of 103 external spinal skeletal
fixation procedures were undertaken. Neurological complications
occurred in two procedures (2%); one resulted in permanent
sciatica. Infections occurred in five patients (5%). Sixty
patients experienced pain relief during the external fixation test,
but only twenty-seven of forty-nine patients who went on to have
an arthrodesis and had sufficient follow-up reported that they were
doing well at a minimum of one year later. In no case did the external
spinal skeletal fixation procedure cause a permanent increase in
low-back pain.
Conclusions: On the basis of this analysis, external
spinal skeletal fixation should not be used as a predictor of pain
relief after lumbar arthrodesis.
Back pain, particularly in the absence of radicular signs
or symptoms, remains an enigma to orthopaedic surgeons. Spinal fusion
is frequently undertaken to relieve it, but review articles suggest
that pain relief is achieved after only approximately 70% of
these arthrodeses1-3. The problem
seems to be more one of "whom to fuse" rather than "how
to fuse" because many surgical techniques consistently
achieve a solid fusion but pain relief may not follow.
As early as 1977 external spinal skeletal fixation was performed
as an alternative for fracture care4,
and in 1986 it was first reported to be effective in relieving back
pain through temporary immobilization of the spine5. However, studies examining the ability
of this technique to predict the clinical success of spinal fusion5-10 have included only 482 cases,
to our knowledge, and the findings have been alarmingly inconsistent.Some authors have reported that temporary
pain relief with external spinal skeletal fixation is a good predictor
of pain relief after spinal arthrodesis, whereas others have not5,7. The reported frequency of neurological
complications (0% to 9%) and infectious complications
(10% to 36%) varies greatly as well.
Mechanical studies have suggested that external spinal skeletal
fixation constructs provide substantial support to the spine and
that this is a basis for relief of mechanical symptoms11-15.
The purpose of the current study was to report the results of external
spinal skeletal fixation used to predict pain relief at a minimum
of one year after subsequent spinal arthrodesis.
This study was approved by the hospital’s Institutional Review
Board and began as a prospective, randomized clinical trial, the
first thirty-five cases of which have been previously reported12. On the basis of the positive results
of that trial, external spinal skeletal fixation was adopted into
clinical practice as a regular but not routine procedure in my regional
spine practice. It was used for patients with back pain without
sciatica or thecal sac compression as assessed with neuroimaging
studies (computed tomography, magnetic resonance imaging, or myelography);
when the results of alternative diagnostic tests (discography and
facet blocks) were equivocal; when clinical factors such as disability
support, narcotic dependency, somatic amplification, or litigant
status suggested a poor prognosis; and/or when the patient
demonstrated severe anxiety at the possibility of spinal arthrodesis.
My concerns regarding the seemingly poor clinical results of arthrodeses
following the 100 external spinal skeletal fixation procedures that
I had performed prompted this data review. I had ceased my practice
of using external spinal skeletal fixation as a diagnostic predictor
of the success of arthrodesis.
The records of 103 patients who underwent external spinal skeletal
fixation were reviewed with attention to perioperative complications,
pain relief from the test procedure, the clinical benefit from a
subsequent arthrodesis, and the functional status after the arthrodesis.
Analyzed data included the frequency of neurological complications
and infections and the benefit (Prolo score) after staged spinal
arthrodesis in patients who underwent arthrodesis after temporarily
experiencing pain relief with the test procedure.
The patients included fifty-eight men with a mean age of forty-four
years (range, twenty-seven to sixty-eight years) and forty-five
women with a mean age of forty-six years (range, twenty-five to
seventy-three years) at the time of spinal fixation. Seventy-four
of the 103 patients were receiving income-replacement benefits from
disability insurance (forty-six were clients of the Workplace Safety
and Insurance Board of Ontario, and the other twenty-eight had privately-funded
disability insurance). Forty-one patients had undergone a total
of sixty-four previous operations (mean, 1.5; range, one to four) prior
to the use of external spinal skeletal fixation.
Instrumentation levels were determined on the basis of the results
of concordantly painful discograms in fifty-five patients, temporary
pain relief after facet blocks in thirty-eight, severe disc degeneration
in six, spondylolisthesis in three, and previous fracture in one.
The external fixation included the fourth lumbar to first sacral
levels in fifty-two patients, the fifth lumbar to first sacral levels
in sixteen, the fourth and fifth lumbar levels in eight, the third
lumbar to first sacral levels in twelve, and other levels in fifteen.
Instrumentation extended as far cephalad as the eleventh thoracic
level.
All of the external spinal skeletal fixation procedures performed
by the author in the study period are reported. Standard surgical
consent as well as informed consent to enrollment in the trial was
required of all patients. The devices were inserted and assembled
with the patient under general anesthesia on an inpatient basis.
Because we had no intraoperative neurological monitoring system
(such as somatosensory evoked potentials or motor evoked potentials)
at our hospital, anesthesia was administered without muscle relaxation
so that a "twitch" might be observed if an external
fixation screw irritated a nerve root. All patients received a single intravenous
dose of a first-generation cephalosporin antibiotic preoperatively.
Under fluoroscopic control, Schanz screws with a 6-mm shaft and
5-mm thread diameter (Synthes Canada, Mississauga, Ontario, Canada)
were inserted down the pedicles of the vertebrae at the extremes
of the proposed arthrodesis. Four screws were inserted in each patient;
for example, for a proposed fourth lumbar-to-first sacral arthrodesis,
two screws each were inserted at the fourth lumbar vertebra and
at the first sacral vertebra. The screws were driven across the
vertebral body to engage the anterior vertebral cortex for optimal
purchase. Radiolucent components from the AO large tubular external
fixator set (Synthes Canada) were coupled to them and assembled
as a quadrilateral frame (Fig. 1). The Schanz pin interfaces with
the skin were sealed with collodion (pyroxylin; Glaxo, Canada, Mississauga,
Ontario, Canada) prior to application of a dry gauze dressing.
Patients were discharged from the hospital as soon as they were
independently mobile, usually on the morning after fixator application.
The outcome parameters of the external spinal skeletal fixation
test consisted of the patient’s subjective report of pain
relief during routine daily activity, as assessed with an interview
at the two-week postoperative clinic visit, and scores determined
with the system of Prolo et al.16 prior
to external spinal skeletal fixation and at the time of the last
follow-up.
Routine pin-site care was not performed postoperatively other than
for a single change of the dry gauze operative dressing, if bleeding
had soiled it, at the time of discharge from the hospital. Patients
were asked to keep the fixator dry, and showering and immersion
in water were forbidden.
For the first sixty-three patients, parenteral antibiotic therapy was
continued for twenty-four hours and then no additional prophylaxis
was given. For the subsequent forty patients, a change in protocol
was prompted by the alarming results reported by Soini and Seitsalo17 and my own experience with a case
of a screw-track abscess. Oral antistaphylococcal prophylaxis was
begun immediately after the surgical procedure in these patients
and was continued for the duration of the external spinal skeletal
fixation test. Also, the pins were sealed with antibacterial ointment (Polysporin [polymyxin
B sulfate and bacitracin zinc]). Implants were removed
with the patient under local or no anesthesia on an outpatient basis
at two weeks.
Patients who did not experience pain relief or who had temporarily
increased pain during the external spinal skeletal fixation test
were assumed to have nonstructural back pain and were discharged
from surgical care to regional chronic pain clinics. Subsequent
posterolateral arthrodesis (with iliac wing autograft and translaminar
or pedicle screw and rod instrumentation) was planned for all patients
who experienced pain relief during the external spinal skeletal
fixation test. The operation was deferred for a minimum of six weeks
to allow for complete pin-track healing.
Fusions were considered solid when continuous trabeculated bone
was present between the transverse processes and independent reviewers
on the radiology staff confirmed the presence of stable implants
without lucency or fracture. All patients who did not have fusion
underwent a reoperation for pseudarthrosis repair. The benefit of
the arthrodesis was analyzed with use of the Prolo functional scores16 at baseline and at the time of the
most recent follow-up.
In the period from January 1989 through September 1999, 103 external
spinal skeletal fixation procedures were attempted, and 100 were
successfully completed. Three procedures could not be accomplished
for technical reasons. In two patients, the pedicles at the fourth
lumbar level were too small to accept the 5-mm threaded portion
of the Schanz screws, and, in another, the hypoplastic fifth lumbar
pedicles could not be visualized fluoroscopically. The duration
of follow-up after the external spinal skeletal fixation test averaged 24.5
months (range, one to 118 months). Forty patients in whom the test
did not relieve pain were followed for only the one-month period
required to confirm uncomplicated primary healing of the pin sites,
and then they were discharged to nonsurgical care and not studied
further. Sixty surgical candidates remained. At the time of writing,
eleven patients who had experienced temporary pain relief from the
external spinal skeletal fixation procedure and thus were thought
to qualify for arthrodesis had either not yet or only just recently
undergone arthrodesis. The remaining forty-nine patients were observed
for an average of forty-one months (range, thirteen to 118 months)
after arthrodesis.
Neurological complications occurred in only two external spinal
skeletal fixation procedures (2%). In one case, a twitch
of the buttock (suggesting lumbosacral nerve root irritation) was observed
during sacral screw insertion, but this patient was neurologically
normal and asymptomatic postoperatively. Another patient awoke with
unilateral radicular symptoms (no neurological deficit) after one
side of the external spinal skeletal fixation frame was inadvertently
assembled in compression. These symptoms were incompletely relieved
by removal of the device. A subsequent computed tomography scan
confirmed that all of the screw tracks were well centered in the pedicles
and completely intraosseous. This patient’s back pain had
been relieved by the external spinal skeletal fixation, and he had
definitive pain relief after a successful fusion. However, residual,
nondisabling radicular symptoms were still present at the time of
the latest follow-up, at nineteen months.
Of the forty-nine patients undergoing arthrodesis, only twenty-seven
(55%) experienced sufficient relief of symptoms to have
meaningful functional benefit after a mean duration of follow-up
of forty-seven months (range, twelve to 101 months). Their mean
Prolo score increased from 5 to 7. Eighteen patients (37%)
had no change in their Prolo score after the arthrodesis, and four
patients (8%) had a decrease in the mean score, from 6
to 4.
Twenty-nine patients were not covered by disability insurance.
Of these, fifteen did not experience pain relief during the external
spinal skeletal fixation test, and the test was aborted in one patient
when a fourth lumbar pedicle was found to be hypoplastic. Thirteen
patients who were not covered by disability insurance and had pain
relief with external spinal skeletal fixation had a successful spinal
fusion. At a mean of forty-five months (range, twelve to 105 months), only
ten of these thirteen patients had notable pain relief, as evidenced
by an increase in their mean Prolo score from 4 at baseline to 8
at the final review. The Prolo scores of the remaining three patients
did not change.
There were infections in four of the initial sixty-three patients who
had parenteral antibiotic therapy continued for only twenty-four
hours. Screw-track infections developed in three patients; the infections
responded to dressing changes and oral antibiotics prescribed on
an ambulatory basis. A subcutaneous abscess developed in one patient,
who required hospital admission for incision and drainage in addition
to parenteral antibiotics. There was only a single superficial screw-track infection
among the forty patients who received more comprehensive prophylaxis,
but this reduction in the rate of infectious complications was not
significant (p < 0.4, Fisher exact test).
Asingle surgeon with experience with the technique performed
all of the 103 external spinal skeletal fixation procedures; therefore,
the series does not contain the so-called "toe" of
the investigator’s learning curve. Nonetheless, in three
patients, anatomical constraints precluded successful percutaneous
Schanz-screw insertion.
The successful external spinal skeletal fixation procedures provided
temporary pain relief in 60% (sixty) of 100 patients with
chronic back pain in the present study. However, despite that pain
relief, only 55% (twenty-seven) of forty-nine patients
with sufficient follow-up had definitive benefit from a subsequent
anatomically successful fusion with bone graft and posterior instrumentation.
Even in the potentially optimal candidate subgroup of uninsured
patients, only ten of thirteen patients had pain relief after the
arthrodesis.
These results are similar to but certainly not superior to those reported
in the literature1-3. Data pooled
from similar clinical series of patients reported on in sufficient
detail to allow analysis (excluding my earlier report12) demonstrated definitive long-term
pain relief in only 36% (173) of 482 patients who had an
anatomically successful spinal arthrodesis after experiencing temporary
pain relief with external spinal skeletal fixation6-10.
The infrequent occurrence of neurological complications in this
series (two of 100 patients) is consistent with that of other reports,
in which neurological complications occurred in twenty-six (5%)
of 514 instrumentation procedures4-10,17,18.
Anatomical studies have shown that fluoroscopic guidance does not
always lead to optimal or safe percutaneous pedicle screw placement19. The rate of infection in this series
was relatively low (5%) and similar to the rates reported
by most previous authors4-10,13,14,18.
External spinal skeletal fixation was conceived as a diagnostic tool
to increase accuracy in the selection of candidates for relief of
low-back pain by lumbar arthrodesis. The results of this study demonstrate
that the procedure is not without serious complications and that
subsequent arthrodeses are no more successful than those following
more orthodox and less invasive diagnostic procedures. Because of
its poor predictive value for back pain relief, external spinal
skeletal fixation should not be used as a diagnostic test to identify
candidates for lumbar arthrodesis.
Note: The author acknowledges the assistance of Constantina Nanos
in assembling database information and Dr. Mohit Bhandari for statistical
analysis.
Frymoyer JW. Back pain and sciatica. N Engl J Med,1988;318: 291-30. 318291
1988
[PubMed][CrossRef]
Hanley EN Jr, David SM. Current concepts review. Lumbar arthrodesis for the treatment of
back pain. J Bone Joint Surg Am,1999;81: 716-30. 81716
1999
[PubMed]
Turner JA, Ersek M, Herron
L, Haselkorn J, Kent D, Ciol MA, Deyo R. Patient outcomes after lumbar spinal fusions. JAMA,1992;268: 907-11. 268907
1992
[PubMed][CrossRef]
Magerl FP. Stabilization of the lower thoracic and lumbar spine with external
skeletal fixation.. Clin Orthop,1984;189: 125-41. 189125
1984
[PubMed]
Olerud S, Sjostrom L, Karlstrom
G, Hamberg M. Spontaneous effect of increased stability of the lower
lumbar spine in cases of severe chronic back pain. The answer of
an external transpeduncular fixation test.. Clin Orthop,1986;203: 67-7. 20367
1986
[PubMed]
Esses SI, Botsford DJ, Kostuik
JP. The role of external spinal skeletal fixation in the assessment of
low-back disorders. Spine,1989;14: 594-601. 14594
1989
[PubMed][CrossRef]
Faraj AA, Akasha K, Mulholland
RC. Temporary external fixation for low back pain: is it worth doing?. Eur Spine J,1997;6: 187-90. 6187
1997
[PubMed][CrossRef]
Jeanneret B, Jovanovic M, Magerl
F. Percutaneous diagnostic stabilization for low back pain.
Correlation with results after fusion operations. Clin Orthop,1994;304: 130-8. 304130
1994
[PubMed]
Soini J, Slatis P, Kannisto
M, Sandelin J. External transpedicular fixation test of the lumbar spine
correlates with the outcome of subsequent lumbar fusion. Clin Orthop,1993;293: 89-96. 29389
1993
[PubMed]
van der Schaaf DB, van Limbeek
J, Pavlov PW. Temporary external transpedicular fixation of the lumbosacral spine. Spine,1999;24: 481-5. 24481
1999
[PubMed][CrossRef]
Axelsson P, Johnsson R, Stromqvist
B. Mechanics of the external fixation test in the lumbar
spine. A roentgen stereophotogrammetric analysis. Spine,1996;21: 330-3. 21330
1996
[PubMed][CrossRef]
Bednar DA, Raducan V. External spinal skeletal fixation in the management of
back pain. Clin Orthop,1996;322: 131-9.. 322131
1996
[PubMed]
Sahni IK, Hipp JA, Kirking
BC, Alexander JW, Esses SI. Use of percutaneous transpedicular external fixation pins
to measure intervertebral motion. Spine,1999;24: 1890-3. 241890
1999
[PubMed][CrossRef]
Schläpfer F, Wörsdörfer
O, Magerl F, Perren SM. Stabilization of the lower thoracic
and lumbar spine: comparative in vitro investigation of an external
skeletal and various internal fixation devices. In: Uhthoff HK,
Stahl E, editors. Current concepts of external fixation
of fractures. New York: Springer; 1982. p 367.
Edwards AG, McNally DS, Mulholland
RC, Goodship AE. The effects of posterior fixation on internal intervertebral
disc mechanics. J Bone Joint Surg Br,1997;79: 154-60. 79154
1997
[PubMed][CrossRef]
Prolo DJ, Oklund SA, Butcher
M. Toward uniformity in evaluating results of lumbar spine
operations. A paradigm applied to posterior interbody fusions. Spine,1986;11: 601-6. 11601
1986
[PubMed][CrossRef]
Soini JR, Seitsalo SK. The external fixation test of the lumbar spine. 30 complications
in 25 of 100 consecutive patients. Acta Orthop Scand,1993;64: 147-9. 64147
1993
[PubMed][CrossRef]
Axelsson P, Johnsson R, Stromqvist
B, Andreasson H. External pedicular fixation of the lumbar spine: outcome
evaluation by functional tests. J Spinal Disord,1999;12: 147-50. 12147
1999
[PubMed][CrossRef]
Wiesner L, Kothe R, Ruther
W. Anatomic evaluation of two different techniques for the
percutaneous insertion of pedicle screws in the lumbar spine. Spine,1999;24: 1599-603. 241599
1999
[PubMed][CrossRef]