The pain and disability associated with lumbar spinal stenosis can
interfere with patients' lifestyles, leading them to seek medical attention.
Typical symptoms of spinal stenosis (neurogenic claudication) include pain,
numbness, and paresthesias in the posterolateral aspects of the thighs and
legs. These symptoms usually radiate distally and are aggravated by prolonged
walking or standing. Activities in which the lumbar spine is in flexion, such
as walking uphill, leaning forward on a walker or shopping cart, or riding a
bicycle, are usually better tolerated by patients with neurogenic
claudication. It is crucial to distinguish between neurogenic and vascular
claudication when establishing the diagnosis of spinal
stenosis1.
Claudication of vascular etiology is usually associated with calf pain, which
may extend proximally. In peripheral vascular disease, lower-extremity pulses
are diminished and the symptoms are aggravated by activity in any position and
are relieved by rest in any position. Only after a clear diagnosis of
neurogenic claudication has been made and a vascular etiology has been ruled
out should treatment of spinal stenosis commence. Treatment options for
lowback pain and neurogenic claudication related to lumbar spinal stenosis
include nonoperative modalities, surgical decompression alone, or surgical
decompression and fusion with or without instrumentation. The treatment
approach is guided by the severity and duration of the symptoms, the type and
extent of the stenosis, any associated instability or deformity, the degree of
disc degeneration, and the presence of coexisting medical conditions.
Look for this and other related articles inInstructional
Course Lectures,Volume 54, which will be published by the
American Academy of Orthopaedic Surgeons in February 2005:• "Spinal Stenosis: Pathophysiology, Clinical and Radiologic
Classification," by Eeric Truumees, MD
Look for this and other related articles inInstructional
Course Lectures,Volume 54, which will be published by the
American Academy of Orthopaedic Surgeons in February 2005:
• "Spinal Stenosis: Pathophysiology, Clinical and Radiologic
Classification," by Eeric Truumees, MD
The onset of lumbar spinal stenosis is usually insidious, and functional
loss is typically
slow2. An initial
course of nonsurgical therapy is therefore recommended. Delaying surgery for a
trial of nonsurgical treatment even in patients with severe stenosis presents
little danger and minimal detrimental effects on the surgical
outcome3. However,
the results of surgical decompression may be worse if leg pain has been
present for more than a
year4. Conservative
management is certainly appropriate for patients with mild-tomoderate symptoms
of neurogenic claudication, as it has often controlled or prevented the
progression of
symptoms2. A
comprehensive nonsurgical program, with the patient playing an active role, is
recommended as the initial treatment for most individuals with lumbar
stenosis5.
The first phase of nonsurgical treatment is pain reduction with passive
modalities6.
Activity modification and relative rest are important for decreasing the
severity of symptoms. Strict bed rest is no longer advocated, and patients are
encouraged to become active as soon as they are able to tolerate it. Patients
are instructed to avoid aggravating activities, such as heavy lifting and
excessive trunk extension, that decrease the anterior-posterior diameter of
the spinal canal7.
Rigid braces can extend the lumbar spine and may thus prove to be more harmful
than helpful by aggravating stenosis symptoms. An elastic lumbar binder may
provide benefit by reducing loads across the lumbar spine, but it should be
worn only for a short period of time in order to avoid deconditioning of the
paraspinal muscles8.
Oral medications, particularly salicylates and nonsteroidal antiinflammatory
drugs, are recommended initially. Muscle relaxants may be utilized, in
addition, for relief of muscle spasms. Tricyclic antidepressants are effective
in some patients with chronic numbness and dysesthetic nerve
pain9. Oral
corticosteroids can be beneficial for acute flare-ups by reducing nerve root
irritation, but they should be used only briefly because long-term treatment
can potentially have adverse
effects9. Narcotic
medications, which cause constipation and are potentially habit-forming,
should be prescribed sparingly and for only brief periods for patients with
incapacitating pain who cannot tolerate anti-inflammatory
medications8. Salmon
calcitonin has provided pain relief and improved walking tolerance, most
likely through its action on the central opioid receptors, in some patients
with only mild symptoms of lumbar
stenosis10.
Calcitonin can be particularly effective for the treatment of stenosis
resulting from Paget disease. Other nonoperative treatment modalities include
cryotherapy, hot packs, manual therapy (massage), acupuncture, ultrasound,
transcutaneous electrical nerve stimulation, traction, and chiropractic
treatment; however, we are not aware of any prospective, randomized studies
proving their
benefit11.
Chiropractic manipulation for patients with lumbar stenosis has been shown
to be most effective when the symptoms are posture-dependent, there is no
evidence of segmental instability, and the patient has the cognitive and
physical abilities to participate in the treatment. In a case report of
multilevel stenosis, flexion-distraction manipulation decreased the intensity
and frequency of leg pain and even led to the resolution of back
pain12. It has been
speculated that spinal manipulation exerts a powerful placebo effect that
produces a specific but short-term
benefit13. Others
believe that manipulative therapy reduces local ischemia and mechanical
compression of chronically irritated nerve roots. Although we do not recommend
chiropractic treatment to all of our patients, if it is sought by a patient,
we caution against extension manipulation.
Epidural steroid injections, although controversial, can reduce the
radicular pain associated with acute exacerbations of neurogenic claudication
by providing analgesic and local anti-inflammatory effects that decrease
spinal stiffness and can facilitate eventual progression to the active phase
of
therapy14,15.
In a study limited to patients with symptomatic lumbar stenosis, Hoogmartens
and Morelle reported that less than half of their patients treated with
epidural steroid injections demonstrated functional improvement approximately
two years after treatment and the outcomes were only slightly better than
those associated with a
placebo16. In a
randomized, prospective, double-blind study, Cuckler et al. found no benefit
from interlaminar injection of methylprednisolone acetate compared with the
effect of physiologic saline solution at twentyfour hours or at approximately
twenty months after
treatment17. This
study was somewhat limited by the authors' definition of a successful result
as a 75% or greater subjective decrease in preinjection symptoms and by their
failure to evaluate outcomes during the interval between a day and a year
after injection. Interlaminar and caudal injections were most commonly
utilized in the past and still may be best for multilevel pathology. Although
the interlaminar and caudal routes of epidural injection are technically
easier than injection through the arthritic posterior elements in elderly
patients with spinal
stenosis18, acute
radicular pain in a specific nerve-root distribution is best treated with a
transforaminal selective nerve-root injection (corticosteroid and bupivacaine)
performed under fluoroscopic guidance. Although selective nerve-root blocks
have provided better outcomes in patients with radicular pain resulting from a
herniated nucleus pulposus, in one series 72% of patients with stenosis
experienced a short-term benefit but only 28% had long-term success from the
injection19. In a
prospective, randomized, controlled, double-blind study, Riew et al. reported
that 71% of patients who had initially requested surgical intervention decided
not to have surgery after selective nerve-root injection with bupivacaine and
betamethasone20.
Since a large proportion of patients for whom all other nonsurgical treatment
methods have failed may decide against surgery after a selective nerve-root
block, patients who have a lumbar radicular pain component to the stenosis
should be considered for a selective nerve-root block. In addition, a
selective nerveroot block is a good prognosticator of surgical outcome, as
patients who obtain >50% relief of leg pain for at least one week tend to
have =50% relief of leg pain, compared with the preoperative intensity,
within one month after the surgery and lasting at least six months
postoperatively4.
Derby et al. also found a strong correlation among the preoperative duration
of leg pain, the response to selective nerve-root blocks, and relief of leg
pain after surgical
decompression4. In
their series, all patients who had had leg pain for less than a year and a
good response to a selective nerve-root block had a successful outcome from
surgical decompression and 95% of patients who had had leg pain for greater
than a year and no response to a selective nerve-root block had a poor
surgical outcome.
The active phase of nonsurgical treatment consists of formal functional
physical therapy, which lays the foundation for faster recovery and decreased
disability and
pain21. It is
important to tailor the program to each individual in order to accommodate the
limitations imposed by coexisting medical conditions. Flexion-based exercises
are the mainstay of physical therapy for patients with lumbar
stenosis22. Studies
of lumbar spine biomechanics have shown that flexion of the spine dynamically
increases, while extension dynamically decreases, the cross-sectional area of
the spinal
canal23,24.
As a result of the increased anterior-posterior diameter of the spinal canal
in flexion (Fig. 1), there is
more space for the neural elements and microcirculation is improved, allowing
patients to tolerate the exercise program
better7. For this
reason, conditioning exercises both on a stationary bicycle and on an inclined
treadmill are effective for patients with lumbar
stenosis21.
Exercise is encouraged because it leads to weight loss, improved
cardiovascular fitness, and the release of endorphins. Aquatic therapy;
stretching of the hip flexors, hamstrings, and paraspinal muscles; and
strengthening of the abdominal and trunk muscles can also provide benefits.
Patient education, including training regarding posture and activities of
daily living, should also be included in every patient's treatment
regimen.
Outcomes of Nonsurgical Treatment
Nonsurgical treatment can minimize the progression of symptoms related to
lumbar stenosis but is unlikely to affect the underlying
pathoanatomy8. In a
study of the natural course of spinal stenosis in thirty-two patients who had
been treated nonoperatively for four years, the pain level was unchanged in
the majority (70%) of the patients, decreased in 15%, and increased in
15%25. Therefore,
the authors recommended expectant observation as an alternative to surgery. In
another series, only 52% of patients treated nonoperatively reported a
decrease in their predominant type of pain (back or leg), and even fewer (42%)
were satisfied with the
outcome26. Onel et
al. reported the best results from an intensive nonoperative treatment
regimen, with a substantial decrease in symptoms in 70% of patients and a mild
decrease in 23%27.
However, Onel et al. did not report the duration of follow-up or long-term
outcomes. We are not sure which components of nonoperative therapy actually
provide the benefit in patients who show improvement with such treatment. The
decision to provide conservative care is further complicated by the number of
different treatment strategies available and the fact that few modalities have
been analyzed in repeated or well-done prospective, randomized studies. In the
majority of patients who are treated nonsurgically, symptoms remain stable
over time25.
Although some investigators have found conservative measures to be of little
long-term benefit, most have recommended nonsurgical therapy as the first line
of treatment for lumbar stenosis, reserving surgery for intolerable pain, a
progressive neurologic deficit, cauda equina syndrome (which is rare), and
patients for whom conservative measures have failed.
Indications
Proper selection of patients for surgical treatment of lumbar stenosis is
crucial to achieving better outcomes. Patients who are functionally limited in
terms of both walking tolerance and activities of daily living are potential
surgical candidates, but the ultimate decision to proceed with surgery should
be made by the
patient28.
Intractable pain, especially neurogenic claudication (leg or buttock pain),
that has not responded to nonsurgical treatment is also a reason to consider
operative intervention. Isolated back pain is not the best indication for
surgery because of the unpredictability of the outcomes of surgical
intervention for this
symptom29,30.
The majority of patients with spinal stenosis can be treated on an elective
basis. Urgent surgical decompression is recommended only for patients with a
rapidly progressive neurologic deficit and/or cauda equina syndrome (bladder
and bowel dysfunction), both of which are rare clinical presentations in
patients with spinal stenosis. The perfect surgical candidate has severe leg
symptoms of a neurogenic claudicatory nature and corresponding stenosis on
imaging studies (Figs. 2-A and
2-B), no or minimal axial back pain, no or minimal neurologic
deficit, no evidence of vascular claudication, and no medical
comorbidities31.
Principles of Surgical Management
When surgical intervention is indicated in patients with lumbar spinal
stenosis, the obvious clinical goals are to improve function, decrease pain,
and decrease or prevent further neurologic deficits. In order to achieve these
objectives, thorough decompression of the neural elements, usually extending
from pedicle to pedicle, is required. The extent of decompression that is
necessary is determined by the symptoms and by the findings on the imaging
studies (magnetic resonance imaging or enhanced computed tomography) and is
modified according to the pathological findings at the time of the surgery.
Even when symptoms are unilateral, bilateral decompression should be performed
if there is radiographic evidence of bilateral stenosis because contralateral
symptoms will soon
develop18. Another
important goal of surgical treatment of stenosis is the maintenance of
stability of the spinal column, which can be facilitated during the
decompression by clear identification and preservation of the pars
interarticularis and by undercutting and preserving at least the lateral 50%
of the facet joints. It is important to avoid removing too much bone laterally
in the area of the pars interarticularis, as this can lead to fracture, pain,
and segmental
instability32.
Fusion with or without instrumentation is recommended to maintain stability in
patients with complex stenosis associated with degenerative spondylolisthesis
or degenerative scoliosis (a curve of >30°) or if >50% of the facet
joints are removed bilaterally during the
decompression33.
Techniques of Surgical Decompression
Laminectomy is the gold standard for decompression in the treatment of
lumbar spinal stenosis. Loupe magnification and headlights are essential for
adequate visualization and a safer operation. After induction of anesthesia,
the patient is placed prone in a kneeling position with the abdomen dependent
on a laminectomy or Andrews frame (Orthopedic Systems, Union City, California)
to minimize intraoperative bleeding from the epidural venous
plexus34. If
internal fixation is planned, a more lordotic position on the Jackson table
may be preferred, but more blood loss is to be expected. After the skin is
injected with 0.25% Marcaine (bupivacaine) with epinephrine, a standard
midline incision is made and dissection is carried out down to the lumbodorsal
fascia (Fig. 3). The fascial
layer should be clearly exposed with a Cobb elevator to make the closure
easier at the end of the procedure. Subperiosteal dissection with the Cobb
elevator and electrocautery is then performed, exposing the spinous processes
and lamina and maintaining the fascial envelope surrounding the paraspinous
muscles (Fig.
4)35.
Violation of the muscle can result in extensive bleeding from perforating
arteries and can decrease visualization. It is important to maintain
hemostasis along each step of the operation. A localizing lateral radiograph
of the lumbar spine should be made to identify the correct levels for
decompression. The dissection is then carried out laterally superficial to the
facet joints, with care taken to preserve the facet joint capsules and to
identify the pars interarticularis (Fig.
5). The parafacetal arteries that lie medial and lateral to the
facet joints may cause additional hemorrhage during the lateral dissection
(Fig. 6), but the bleeding can
be easily controlled with bipolar or Bovie
electrocautery35.
If a posterolateral fusion is planned, the lateral dissection needs to extend
to the transverse processes bilaterally, while the intertransverse membrane is
maintained. The inferior half of the spinous process at the top of the
decompression and the superior half of the spinous process of the inferior
level to be decompressed are removed with a Leksell rongeur. A Horsley bone
biter is used to remove the intervening spinous processes
(Fig. 7). The Leksell rongeur
is used again to remove any remaining soft tissue and to thin the lamina. Bone
wax applied to the bleeding bone allows maintenance of a dry surgical field.
The ligamentum flavum ("yellow ligament") is identified, and a
curet is used to gently dissect the insertion of the ligament from the
undersurface of the inferior edge of the most caudal lamina, where a central
decompression begins. Removal of the laminae should always start centrally,
since the midline is the last area to become stenotic and thus the safest
place to begin a
dissection35.
In an L3 to L5 decompression, we recommend beginning at the inferior aspect
of the L4 lamina and later removing the superior laminar edge of L5, exposing
the lateral recess of L4-L5 and the entry zone to the L5 neuroforamen. Working
in a caudal-to-cephalad direction, the surgeon utilizes Kerrison punches to
remove the remaining lamina and ligamentum flavum in the midline, exposing the
thecal sac (Fig. 8). The
midline trough that is created makes up the central decompression
(Fig. 9). A ball-tip probe or
angled dural elevator is helpful for protecting the dura as the decompression
is extended laterally to the pedicles, preserving the lateral 50% of the facet
joints if possible. Medial facetectomy and removal of any osteophytic ridge
adjacent to the intervertebral disc space is usually sufficient to decompress
the entrance
zone36. Adequate
visualization is paramount and can be obtained with Frazier-tip suction and by
packing the areas surrounding the location where the surgeon's attention is
focused. It is important to dig in with the foot of the Kerrison punch,
pulling up gently as the bite is taken and to angle the instrument along the
course of the nerve root to minimize the chance for iatrogenic nerve-root
injury or durotomy.
The midzone, which is located anterior to the pars interarticularis and
inferior to the pedicle, can be decompressed with careful undercutting of the
hypertrophic facet joints and the area under the pars interarticularis with a
small Kerrison rongeur (Fig.
10)36.
Total facetectomy should be avoided if possible to minimize the risk of
iatrogenic instability. Preservation of bone in the area of the pars
interarticularis is also crucial to maintain spinal stability. A small curet
can be used to smooth any bone spikes left by the Kerrison rongeur, keeping
its smooth surface against the dura. The nerve roots are identified, and the
neuroforamen is probed to make sure it is adequately patent. The probe should
be passed in an inferolateral direction parallel to the course of the nerve
root (Fig. 11). In the
presence of exit-zone stenosis, thorough decompression is best performed by
trimming the medial, lateral, and superior margins of the involved superior
articular
process36. It may
be best to approach the exit zone from the lateral aspect of the facet joint,
making sure to cauterize the posterior perforating branch of the lumbar
artery, which runs immediately adjacent to the facet joint. Since there is a
small chance of disc herniation in the presence of spinal
stenosis37, a
discectomy is performed only if necessary to further decompress both the
exiting and the traversing nerve roots. We advocate discectomy only for
extruded soft disc material or a free fragment, since violation of
degenerated, well-contained discs may lead to unnecessary destabilization of
the anterior column.
Bilateral laminotomies have been suggested as an alternative to complete
laminectomy in properly selected patients with mainly lateral recess stenosis.
The laminotomies are essentially performed through a microdiscectomylike
approach at multiple levels and bilaterally (if necessary), guided by the
location of the pathology. Advocates argue that the potential for
postoperative iatrogenic spondylolisthesis is reduced by the performance of
laminotomies, which preserve the midline stabilizing structures. Others
believe that patients have less back pain but a higher chance of neural
complications after
laminotomies38.
Bilateral laminotomy, also known as the "port-hole" technique,
provided relief of back pain in 80% of patients, relief of leg symptoms in
96%, and a good or excellent outcome in 88% after four years of follow-up in
one series39.
Furthermore, since the procedure is relatively minimally destabilizing in
patients with degenerative spondylolisthesis and stenosis, nearly 90% of the
patients in that series showed no progression of the slip at the four-year
follow-up examination.
In a retrospective study of twentyone patients with single or two-level
spinal stenosis and degenerative spondylolisthesis, McCulloch described a
microscopic decompression and fusion, without instrumentation, performed
through bilateral fascial incisions with preservation of the spinous processes
and interspinous-supraspinous ligament
complex40. At an
average of thirtyeight months postoperatively, 76% of the patients had a
satisfactory outcome, a result that is comparable with those of laminectomy.
Additionally, there was no evidence of slip progression despite the
performance of the fusion without instrumentation.
A fenestration technique has also been reported for the treatment of
isolated lumbar nerve-root foraminal
stenosis41. The
decompression is performed through a 5-mm drill hole in the lamina immediately
inferior to the superior facet. In 1998, Mackay and Wheelwright reported that,
of fifty patients with spinal stenosis treated with unilateral decompression
with fenestration restricted to the clinically relevant level and side, 60%
rated their outcome as excellent or good and 28% considered it to be fair at a
mean of thirty-two months
postoperatively42.
The authors did not discuss long-term results or the development of
contralateral symptoms and the potential need for a second procedure. This
method preserves spinal stability even if it is performed at multiple levels.
Fenestration also allows early patient mobilization and shorter hospital stays
according to the authors who described it, but central stenosis cannot be
decompressed in this way and the potential for neural injury is probably
higher.
Distraction laminoplasty is another recently described technique that
allows a central and lateral decompression while maintaining the osseous
architecture
posteriorly43. The
interspinous ligaments are removed with a rongeur just as in a laminectomy,
and the interlaminar space is exposed. After adequate distraction of the
interlaminar space with a lamina spreader placed against the spinous
processes, an osteotomy of the medial 20% of the facet joints and the inner
one-third of the lamina is performed and that bone is removed. Additional
investigations of the long-term outcomes of distraction laminoplasty and
studies directly comparing it with laminectomy, which is the current gold
standard, are needed.
Surgical Outcomes
The reported success rates of decompressive surgery for spinal stenosis
vary. Patients considering surgery should be advised that, although the
majority will improve, a quarter or more will not and some who do better
initially will have a recurrence of symptoms with
time26.
Good-to-excellent outcomes have been reported in as many as 80% of patients
treated with decompression for lumbar spinal
stenosis44-46.
Most studies have shown that patients do well in the short term, better than
those treated nonoperatively, but the results tend to deteriorate over time.
In the Maine Lumbar Spine Study, 55% of patients treated surgically and only
28% of those treated nonsurgically reported definite relief of their
predominant symptoms at one
year47. The
patients placed in the surgically treated group tended to have severe
symptoms, but they still had better outcomes. The rates of surgical and
nonsurgical treatment were similar for the patients with moderate symptoms,
and the outcomes of the surgical treatment were better. After four years of
follow-up in the Maine study, 63% of the surgically treated patients were
satisfied and 70% reported that the predominant pain had decreased. In another
series, at seven to ten years after decompression surgery for spinal stenosis,
75% of the patients were somewhat or very satisfied, 23% had required revision
surgery, and a third complained of severe back or buttock
pain48. A
metaanalysis of the literature revealed that good-to-excellent long-term
results had been reported in 64% of patients who had undergone decompressive
surgery49.
Regardless of the treatment, unsatisfactory outcomes remain relatively
common. About 25% to 40% of patients treated surgically and 50% to 60% of
those treated nonsurgically were not satisfied with the outcomes at four
years26. The
surgical outcomes may depend on the duration of the symptoms. A meta-analysis
of more than 1600 cases of degenerative spinal stenosis revealed that the best
results were obtained when surgical intervention had been carried out within
the first few years after the onset of the
disease50. Many
believe that prolonged structural compromise of spinal nerve roots can lead to
chronic and irreversible damage that surgical decompression cannot alter.
According to Derby et al., the results of surgical decompression are usually
better if the procedure is performed within a year after the onset of
symptoms4.
There is a role for arthrodesis in addition to decompression for the
treatment of certain cases of stenosis. The indications for fusion include
instability at the involved motion segment, degenerative scoliosis (especially
with curve progression or a curve magnitude of >30°), revision
decompression at the same level, resection of >50% of the facets
bilaterally, and degenerative
spondylolisthesis33,38.
Decompressive surgery alone, through laminotomies if possible, can be
indicated in a few elderly patients who present with mild scoliosis and little
or no back pain. Although decompression alone has achieved good results in
>80% of patients with spinal stenosis and associated degenerative
spondylolisthesis51,
several studies have shown that better results are obtained with a concomitant
fusion
procedure37,52,53.
Postoperative Care
Regardless of the surgical procedure, patients are encouraged to get out of
bed the day of the surgery and begin physical therapy. Early mobilization is
facilitated by the injection of Marcaine at the time of wound closure, which
reduces postoperative incisional pain. A lumbar binder or lumbosacral orthosis
is rarely prescribed for patients who have undergone decompression alone and
is usually used during walking for the first six weeks after fusion procedures
with or without instrumentation. Patients are advised to avoid bending,
lifting, or twisting for six to twelve weeks postoperatively. Patients are
usually seen for follow-up at two to three weeks for a first check and
radiographs and to begin outpatient physiotherapy. Patients are then seen at
twelve weeks and one year after the surgery if no fusion was performed.
Patients who underwent a fusion procedure, especially with instrumentation,
require more frequent and longer follow-up. They are also seen first at two to
three weeks postoperatively, followed by appointments, which include
radiographic examination, at six weeks, three months, six months, one year,
yearly up to five years, and every five years thereafter.
Complications
The risks associated with surgical treatment of lumbar stenosis without
fusion are no more serious or life-threatening than are the risks of total hip
replacement in elderly
patients54. In a
retrospective study of patients over the age of seventy years who had
undergone decompressive lumbar procedures, Benz et al. found serious
complications that could affect the quality of life in only 12% of patients
and reported an early mortality rate of
<2%55. Ragab et
al. retrospectively reviewed the results of surgical treatment of lumbar
stenosis in 118 patients who were seventy years of age or older and concluded
that advanced age did not increase the morbidity associated with the
surgery56. Possible
complications of lumbar decompressive surgery include epidural hematoma,
thromboembolic events, dural tears (especially in revisions), infection,
instability following wide decompression, nerve root injury, nonunion or
hardware failure following fusion, and adjacent segment degeneration.
The reported prevalence of vascular complications after decompressive
lumbar laminectomy is
<0.02%57.
Although major vascular injury is infrequent, it is still important to monitor
the blood pressure, hematocrit, and peripheral perfusion to the lower
extremities postoperatively, especially in elderly patients. West and Anderson
reported a 14% rate of deep venous thrombosis in a group of patients treated
with spine surgery, 85% of whom had undergone a posterior-only
procedure58. Since
patients with a spinal cord injury were included in that series, the actual
risk of elective surgery for lumbar stenosis is probably much lower. In
another series, the prevalence of symptomatic pulmonary embolism after
posterior decompression and fusion was only
0.5%59. Persistent
postoperative spinal fluid leaks and pseudomeningoceles are rare because most
dural tears are identified at the time of surgery and are immediately repaired
with a water-tight suture closure. The prevalence of dural tears secondary to
operations on the lumbar spine has been reported to be anywhere from <1% to
almost
15%60-62.
Despite antibiotic prophylaxis, postoperative infection remains a risk,
especially in the setting of instrumentation. In a large series reported on by
Deyo et al., postoperative infection occurred after only 0.5% of lumbar spine
procedures63. A
historical cohort study of patients with symptomatic stenosis and degenerative
spondylolisthesis who were treated with lumbar decompression and arthrodesis,
with or without instrumentation, revealed an infection rate of 2% to
3%64. In the same
study, the prevalence of nerve root injury from pedicle screw placement was
0.4%64. Nerve root
injury can result from the use of pedicle screw instrumentation or the
manipulation of a severely compressed nerve.
Another potential complication of surgery for lumbar stenosis is recurrence
of symptoms. Substantial back and leg symptoms develop in up to 10% to 15% of
patients who have undergone an adequate lumbar
decompression65.
Substantial osseous regrowth after decompression may be the reason symptoms
recur and can decrease patient
satisfaction66.
Degeneration with stenosis at segments adjacent to previously treated segments
can also cause leg pain and claudicatory symptoms to
recur67.