Cauda equina syndrome in patients with low lumbar fractures I T

Transcription

Cauda equina syndrome in patients with low lumbar fractures I T
Neurosurg Focus 16 (6):e6, 2004, Click here to return to Table of Contents
Cauda equina syndrome in patients
with low lumbar fractures
ISSADA THONGTRANGAN, M.D., HOANG LE, M.D., JON PARK, MD., F.R.C.S.(C),
AND DANIEL H. KIM, M.D.
Department of Neurosurgery, Stanford University Medical Center, Stanford, California
Object. Symptoms of cauda equina syndrome (CES) can include low-back pain, sciatica, lower-extremity weakness,
sensory deficit, perineal hypesthesia or anesthesia, and loss of bowel or bladder function. Several causes of the syndrome are recognized, but its optimal treatment remains controversial and has been broadly based on data gathered
from series involving herniated discs. Information on the treatment of CES caused by low lumbar traumatic injuries
has not been well documented.
Methods. Between January 2000 and December 2003, 17 consecutive cases of CES caused by low lumbar traumatic
injuries at L2–5 were identified. The traumatic injuries consisted of gun shot wound in two cases, motor vehicle accident in 11, and a fall from height in four. Conus medullaris injuries causing CES were excluded from this review.
Presenting symptoms, mechanisms of injury, radiographic images, timing of surgery, surgical approaches, and neurological status at the final follow up were documented. All patients underwent follow up of at least 12 months.
Fourteen of 17 patients had satisfactory outcomes. Despite undergoing surgery within the first 24 hours postinjury,
three patients had what was classified as a poor outcome given their residual deficits and included two cases with gunshot injuries. Recovery of leg weakness occurred within 4 months, whereas bladder and bowel function recovered
within 3 months. All patients in this series underwent decompression within less than 48 hours after syndrome onset.
Overall, the authors found no difference regarding timing of surgery between patients in the satisfactory outcome
group and those in the poor outcome group.
Conclusions. Based on the evidence in this study, the severity of a patient’s condition on initial presentation is the
most crucial factor in predicting outcome following CES due to low lumbar injuries. Although the matter of the timing of surgery remains controversial, the authors of this study recommend that surgery be performed within 48 hours
of syndrome onset.
KEY WORDS • cauda equina syndrome • burst fracture • lumbar spine • spinal injury
In an adult, the conus medullaris ends at the L1–2 intervertebral disc level and continues to taper finally to form
the terminal filum. The nerves of the cauda equina provide
sensory and motor function to the lower extremities and
sensation to the perineum and genitalia as well as innervate the pelvic viscera. They also provide necessary pathways for coordinated activities such as micturition, defecation, and sexual function.
Injury to the conus medullaris or the nerves of the cauda
equina can result in CES, which in its triad can include
lower extremity weakness, perineal sensory loss, and
bowel and bladder dysfunction. Several causes of the syndrome have been reported in the literature and include
acute traumatic injury,6,9,19 herniated disc,2,4,13,19,20 and metastatic lesion.16 Although reports of CES caused by large
central disc herniations have been well documented, the
literature on CES caused by traumatic lumbar fractures is
sparse. Timing of treatment and recovery from the synAbbreviations used in this paper: CES = cauda equina syndrome;
CT = computerized tomography; GSW = gunshot wound; MVA =
motor vehicle accident.
Neurosurg. Focus / Volume 16 / June, 2004
drome have been based mainly on information gathered
from clinical series involving disc herniations.
In this report, we reviewed a clinical series of patients
with CES caused by acute traumatic lumbar fractures
(L2–5) and analyzed outcomes based on surgical treatment.
CLINICAL MATERIAL AND METHODS
We retrospectively identified 17 patients (12 male and
five female) with a mean age of 38.5 years (range 17–80
years) who had been diagnosed with low lumbar traumatic injuries from the L-2 to L-5 vertebrae and CES on
arrival to Stanford University Medical Center between
January 2000 and December 2003. We defined “CES” as
“a complex of saddle hypesthesia, motor weakness in the
lower extremities, and bowel or bladder dysfunction.” At
least two of the three criteria were necessary for diagnosis
and inclusion into our review.
Mechanisms of injury and neurological status at presentation were documented. Radiographic studies including plain x-ray films and CT scans of the lumbar spine had
been obtained in all patients. We excluded any fracture
28
I. Thongtrangan, et al.
above L-2 to exclude CES from a conus medullaris injury. We defined “early decompression” as “treatment
within 48 hours of the development of the full syndrome”
and “late decompression” as treatment any time following this.
All patients had undergone posterior stabilization with
pedicle screw/rod instrumentation with or without an additional anterior procedure. The anterior procedure had
been considered mainly for decompression purposes and
performed via a left mini-retroperitoneal approach. Standard corpectomy and anterior column reconstruction by using a structural allograft or an expandable cage with or
without anterior instrumentation had been performed in
selected cases.
All patients had been followed up for at least 12 months
postoperatively, and the recovery of neurological function
was documented. Radiographic images were also monitored for any instrumentation-related complications.
We defined “satisfactory outcome” as “an absence of
bowel and/or bladder dysfunction and an ability to ambulate independently by the time of the last follow up,”
whereas “poor outcome” was “a residual neurological deficit in any subcategory.”
RESULTS
Traumatic injuries included GSW in two cases, MVA in
11, and fall from a height in four. The level of fracture was
at L-2 (seven cases), L-3 (seven cases), and L-4 (three
cases). No patient had involvement of L-5. Demographic
and clinical details are featured in Table 1.
Of the 17 patients, 14 were determined to have a satisfactory outcome at the final follow up. The remaining
three were classified as having a poor outcome because of
residual neurological deficit including paraplegia and
bowel and/or bladder dysfunction. A CT scan demonstrated spinal canal compromise in all cases (range 55–90%
compromise). Three cases also had fracture of the lamina
and dural tearing. All patients underwent surgery for decompression and stabilization. An anterior procedure was
performed in only one case, a posterior procedure in two
cases, and a posterior–anterior approach in the remaining
14 cases in (Table 1). The mean follow up was 13.5
months (range 12–17 months).
Timing of Surgery
All patients underwent surgery within 48 hours of presentation, with a mean time to decompression and stabilization of 20.8 hours (range 8–48 hours). Fourteen patients underwent surgery within 24 hours, whereas the
remaining three patients underwent surgery within 48
hours. Eleven of 14 patients who had undergone surgery
within the first 24 hours had satisfactory outcomes; the
other three patients with satisfactory outcomes underwent
surgery within 48 hours.
TABLE 1
Demographic and clinical data*
Fracture
of
Lamina/
Dural
Tearing
Leg
Weakness
Saddle
Paresthesia
Bladder
Dysfunction
Age
(yrs),
Sex
Mechanism of
Injury
Level
of
Injury
% Spinal
Canal
Involvement
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
55, M
43, F
21, F
74, F
35, F
36, M
46, M
31, M
80, M
17, M
45, M
19, F
20, M
54, M
33, M
MVA
MVA
MVA
fall
MVA
MVA
MVA
MVA
fall
fall
MVA
fall
MVA
MVA
MVA
L-3
L-3
L-2
L-4
L-2
L-3
L-4
L-2
L-3
L-3
L-2
L-2
L-3
L-2
L-4
55
70
60
80
70
65
75
65
55
60
75
60
65
80
90
no
no
no
no
no
no
no
no
no
no
no
no
no
no
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
no
yes
yes
no
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
no
yes
no
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
no
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
no
yes
no
yes
yes
yes
yes
20
16
30
48
24
12
16
20
24
16
14
8
36
16
24
AP
AP
AP
ant
AP
AP
AP
AP
AP
AP
AP
AP
AP
AP
AP
17
12
15
12
14
14
12
13
15
12
16
12
14
14
13
16
20, M
GSW
L-2
retained
bullet
yes
yes
yes
yes
yes
12
pst
12
17
25, M
GSW
L-3
retained
bullet
yes
yes
yes
yes
yes
18
pst
13
Case
No.
Decreased
Rectal
Tone
Timing
of
Surgery
(hrs)
Surgical
Approach
F/U
(mos)
Residual
Bowel/
Bladder
Dysfunction
Ability
to
Ambulate
no
no
no
no
no
no
no
no
no
no
no
no
no
no
partial
recovery
of bladder
dysfunction
paraplegia,
complete
bowel/
bladder
dysfunction
paraplegia,
complete
bowel/
bladder
dysfunction
* Ant = anterior; AP = anterior–posterior; F/U = follow up; pst = posterior.
29
Neurosurg. Focus / Volume 16 / June, 2004
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
no
no
Cauda equina syndrome in patients with low lumbar fractures
Interestingly, all cases with a poor outcome in this
series involved surgery performed within the first 24
hours. Among the three cases with poor outcomes, two
patients suffered from GSWs with severe destruction of
the posterior spinal elements and tearing of the dura mater.
The remaining patient had an L-4 burst fracture in conjunction with a lamina fracture and dural tearing.
The mean timing of surgery in the satisfactory outcome
group was 21.4 hours, whereas that in the poor outcome
group was 18.2 hours. Overall, we could not make a consistent conclusion in regard to timing of surgery and clinical outcome. Nonetheless, it appears that patients with
GSWs and CES will have a worse overall clinical outcome compared with that in patients with CES from a
traumatic fracture injury.
Motor Weakness
Fifteen of 17 patients in this series had leg weakness on
initial presentation. The two patients with no leg weakness
were younger than 20 years old. One had fallen from a
height and the other had been involve in an MVA. Interestingly, in both cases there was an L-3 burst fracture with
spinal canal involvement.
Thirteen of 15 patients with leg weakness on presentation recovered to the point of being able to ambulate independently. Such recovery began at 8 weeks after surgery,
and no further recovery was observed after 4 months. In
the two cases involving GSWs the patients presented with
complete paralysis of the lower extremities, which remained the same despite early decompression and stabilization within 20 hours.
Bladder Function
Sixteen of 17 cases involved bladder dysfunction, that
is, urinary retention requiring catheterization. In the entire
series, only one patient presented with normal bladder
function. Among patients in the satisfactory outcome
group (14 of 17 patients), there was no residual bladder
dysfunction at the final follow up. Bladder function improved a mean of 9 days after surgery (range 3 days–3
months). No bladder recovery occurred after 3 months.
Three of the 17 patients were classified as having a poor
outcome based on residual bladder dysfunction at the final
follow up. The two patients who suffered GSWs did not
recover bladder function. The patient with an L-4 burst
fracture experienced partial recovery with residual urinary
retention at the final follow up.
Bowel Function
Of the 17 patients identified for inclusion in our review, 15 had decreased anal tone on presentation. Normal
anal tone was recovered within a mean of 10 days after
surgery in the group with a good outcome (range 5 days–3
months). No further recovery occurred after 3 months.
Of the three patients with a poor outcome, two had significant residual anal sphincter dysfunction. Not surprisingly, these two had suffered GSWs, which required regular bowel evacuations either with the aid of laxatives or
manually.
Saddle Paresthesia
Saddle paresthesia was documented in 15 cases on presentation, including those with GSWs. Recovery occurred
Neurosurg. Focus / Volume 16 / June, 2004
in all cases except those involving GSW. Saddle paresthesia resolved a mean of 4 days after decompression and stabilization (range 24 hours–8 days).
ILLUSTRATIVE CASES
Case 1
This 33-year old man suffered an L-4 burst fracture
after having been in an MVA during which the vehicle
rolled over. Physical examination revealed weakness of
the quadriceps muscles bilaterally, perineal hypesthesia,
and urinary retention. Radiographic images (Fig. 1A) and
CT scans (Fig. 1B and C) of the lumbar spine revealed an
L-4 burst fracture with 90% canal involvement as well as
fracture of the lamina. A combined posterior–anterior decompression and stabilization procedures were performed
within 24 hours. A dural tear was noted at the time of the
surgery. Postoperative x-ray films (Fig. 2) demonstrated
posterior stabilization with pedicle screw/rod instrumentation, and expandable cage placement for anterior column reconstruction after corpectomy. He regained perineal sensation 72 hours after surgery, complete motor
strength in the lower extremities within the first 2 months,
and partial recovery of bladder function by the final follow up. He demonstrated postmicturition residual urinary
retention at the final follow up.
Case 2
This 20-year-old man was shot in the back. Physical examination revealed paraplegia with complete loss of sensation to light touch and pinprick below the thighs, no rectal tone, saddle anesthesia, and bladder incontinence.
Radiographic images (Fig. 3A) and CT scans (Fig. 3B and
C) revealed severe destruction of the L-3 posterior elements and an intraspinally retained bullet at the level of L3. The anterior and middle columns were intact. The patient underwent a posterior laminectomy, removal of the
bullet, posterior stabilization, and fusion by using a pedicle screw/rod system. Not surprisingly, dural laceration
was noted intraoperatively with severe injury to the cauda
equina. Postoperatively, he has not recovered neurological
function despite having undergone early surgery within 18
hours. As of the final follow up, he remains an L-3 paraplegic with bowel and bladder incontinence.
DISCUSSION
The incidence of CES ranges from 1 to 5% depending
on the origin of the disease.2,5,8,19,20 Central disc herniations
are the most common origins reported in the published literature.2,3,5,8,13,14 Acute trauma of the lower lumbar spine as
a cause of CES is uncommon. In our series of patients
with CES caused by traumatic lower lumbar injuries, 14
(82%) of 17 patients achieved satisfactory outcomes following surgical decompression.
Timing of decompression in the treatment of CES
remains controversial despite the traditional neurosurgical
teaching of early surgical intervention. In several published papers, authors support the role of early decompression with good to excellent outcome.8,10,15,19 In addition, in
a recent metaanalysis,1 researchers asserted that patients
undergoing surgery less than 48 hours after onset of symp30
I. Thongtrangan, et al.
Fig. 2. Case 2. Postoperative x-ray films demonstrating posterior stabilization performed using a pedicle screw/rod system and
expandable cage placement following anterior and middle column
reconstruction. The alignment was well maintained with satisfactory decompression.
Fig. 1. Anteroposterior (left) and lateral (right) x-ray films
demonstrating an L-4 burst fracture. B: Sagittal CT scans demonstrating an L-4 burst fracture with segmental deformity and bone
retropulsion. C: Axial CT scans revealing an L-4 burst fracture
with significant canal compromise and lamina fracture posteriorly.
31
toms of CES had better outcomes than those in whom
surgery had been undertaken later. Despite this, other investigators have questioned the role of early decompression. Kostuik, et al.,13 retrospectively reviewed 31 patients
with CES caused by lumbar disc herniations who demonstrated excellent outcomes and recovery of motor function
with late decompression up to 3 days postonset. They
found no correlation between the length of time from the
onset of symptoms to surgery and the extent of neurological recovery. Delamarter, et al.,7 in an animal model, also
found similar evidence. These investigators developed a
canine model of CES and evaluated neurological recovery
following immediate, early, and delayed decompression.
No differences in somatosensory evoked potentials were
noted between early and delayed decompression. In our
series, the timing of surgery did not correlate well with
neurological recovery. Nonetheless, all of the patients in
our study had undergone decompression and stabilization
within 48 hours.
Factors affecting recovery may correlate more with
the type and severity of initial injury. In our review,
three patients had a poor outcome because of residual neurological impairment including motor weakness, bladder
dysfunction, and bowel dysfunction. Of these three, two
suffered a GSW to the lower lumbar area.
The role of bullet removal in neurological recovery has
been controversial. Analysis of recent data from a study
performed by the Regional Model Spinal Cord Injury Systems21 indicates that laminectomy and bullet removal
plays no role in a spinal cord level lesion, but is a worthwhile procedure in treating injury at the level of, as well
as caudad to, the conus medullaris. This is probably based
Neurosurg. Focus / Volume 16 / June, 2004
Cauda equina syndrome in patients with low lumbar fractures
pressure. The concept that compressed neural tissue below
a threshold pressure may retain the ability to recover indefinitely may describe the excellent recovery rates described in both clinical and experimental studies involving
delayed decompression. In addition, potential peripheral
nerve regeneration or repair plays a major role in recovery
of the cauda equina. Once compression pressure exceeds
the threshold, however, permanent damage might occur
and the time to full recovery prolonged or unlikely.
Data from most series indicate that motor recovery is
complete by 1 year, but there are reports of improvement
in bladder function after a considerably longer timeframe11,17 even up to 5 years later.4,12
In our study, 15 of 17 patients recovered motor strength
in the lower extremities to the point that they were able to
ambulate independently by 4 months. Bowel and bladder
function fully recovered within three months. No further
improvements were noted after 3 months. Our data demonstrate the potential recovery following surgical decompression in patients with CES caused by low lumbar traumatic injuries. The natural history of recovery from CES
in patients treated with conservative therapy is unknown
because the preferred treatment option remains surgical in
nature. Note, however, that the initial mechanism of injury
may be an important factor in predicting recovery, especially given the poor outcome in the patients with GSWs.
Also, based on our experience, the severity of initial
traumatic injury appears to play a major role in recovery
rather than timing of surgery. Regardless, given the morbidity associated with continual bladder dysfunction, we
remain advocates of early surgical decompression for
CES caused by traumatic injuries. Full decompression
may require the anterior approach for the most optimal
outcome.
Fig. 3. Case 2. Anteroposterior (left) and lateral (right) x-ray
films revealing gunshot injuries, with the retained bullet at L-3.
B: Sagittal CT scans revealing destruction of posterior spinal elements, with retained bullet in the spinal canal. C: Axial CT scans
demonstrating severe fracture and destruction of lamina and posterior elements, with retained bullet in the spinal canal.
on the regional anatomy of the spinal cord and the greater
susceptibility of the spinal cord to injury compared with
these factors in the spinal nerve roots. It is probable that
root level lesions are associated with greater clinical improvement after decompression because of the relatively
large number of lumbar and sacral roots that can be compressed by a bullet at this level and the potential for axons to regenerate. Nonetheless, the patients in the present
study with complete CES failed to show any neurological
improvements despite early decompression and stabilization. Thus, these results may indicate that the severity of
the initial injury plays an important role in recovery. In the
case of an L-4 burst fracture with associated lamina fracture, dural tear, and near complete obliteration of the spinal canal from the retropulsed bone fragment, the patient
had severe compression of the cauda equina in both the
dorsal and ventral directions at the time of injury. Rydevik, et al.,18 demonstrated the importance of threshold
Neurosurg. Focus / Volume 16 / June, 2004
CONCLUSIONS
In this paper we focus on the neurological recovery
from CES following low lumbar traumatic injuries and
show that direct decompression and stabilization within
48 hours offers a favorable chance of recovery. Most patients with lower extremity weakness recovered to the
point that they were able to ambulate independently, except those with GSWs. Bowel and bladder function can be
expected to recover within 3 months; thereafter recovery
is unlikely. Severity of the initial injury appears to be the
most important factor for recovery.
References
1. Ahn UM, Ahn NU, Buchowski JM, et al: Cauda equina syndrome secondary to lumbar disc herniation: a meta-analysis of
surgical outcomes. Spine 25:1515–1522, 2000
2. Buchner M, Schiltenwolf M: Cauda equina syndrome caused by
intervertebral lumbar disk prolapse: mid-term results of 22 patients and literature review. Orthopedics 25:727–731, 2002
3. Busse JW, Hsu WS: Rapid progression of acute sciatica to
cauda equina syndrome. J Manipulative Physiol Ther 24:
350–355, 2001
4. Chang HS, Nakagawa H, Mizuno J: Lumbar herniated disc presenting with cauda equina syndrome. Long-term follow-up of
four cases. Surg Neurol 53:100–105, 2000
5. Choudhury AR, Taylor JC: Cauda equina syndrome in lumbar
disc disease. Acta Orthop Scand 51:493–499, 1980
32
I. Thongtrangan, et al.
6. Cybulski GR, Stone JL, Kant R: Outcome of laminectomy for
civilian gunshot injuries of the terminal spinal cord and cauda
equina: review of 88 cases. Neurosurgery 24:392–397, 1989
7. Delamarter RB, Sherman JE, Carr JB: 1991 Volvo Award in experimental studies. Cauda equina syndrome: neurologic recovery following immediate, early, or late decompression. Spine
16:1022–1029, 1991
8. Dinning TA, Schaeffer HR: Discogenic compression of the cauda equina: a surgical emergency. Aust N Z J Surg 63:927–934,
1993
9. Gertzbein SD, Court-Brown CM, Marks P, et al: The neurological outcome following surgery for spinal fractures. Spine 13:
641–644, 1988
10. Gleave JR, Macfarlane R: Cauda equina syndrome: what is the
relationship between timing of surgery and outcome? Br J
Neurosurg 16:325–328, 2002
11. Hellstrom P, Kortelainen P, Kontturi M: Late urodynamic finding after surgery for cauda equina syndrome caused by a prolapsed lumbar intervetebral disk. J Urol 135:308–312, 1986
12. Jennett WB: A study of 25 cases of compression of the cauda
equina by prolapsed intervertebral discs. J Neurochem 19:
109–116, 1956
13. Kostuik JP, Harrington I, Alexander D, et al: Cauda equina syndrome and lumbar disc herniation. J Bone Joint Surg Am 68:
386–391, 1986
14. Lavyne MH: Cauda equina syndrome secondary to lumbar disc
herniation. Neurosurgery 34:561, 1994
33
15. McLaren AC, Bailey SI: Cauda equina syndrome: a complication of lumbar discectomy. Clin Orthop 204:143–149, 1986
16. Nather A, Bose K: The results of decompression of cord or
cauda equina compression from metastatic extradural tumors.
Clin Orthop 169:103–108, 1982
17. Nielsen B, de Nully M, Schmidt K, et al: A urodynamic study
of cauda equina syndrome due to lumbar disc herniation. Urol
Int 35:167–170, 1980
18. Rydevik BL, Pedowitz R, Hargens AR, et al: Effects of acute,
graded compression on spinal nerve root function and structure.
An experimental study of the pig cauda equina. Spine 16:
487–493, 1991
19. Shapiro S: Cauda equina syndrome secondary to lumbar disc
herniation. Neurosurgery 32:743–747, 1993
20. Shapiro S: Medical realities of cauda equina syndrome secondary to lumbar disc herniation. Spine 25:348–352, 2000
21. Waters RL, Sie IH: Spinal cord injuries from gunshot wounds
to the spine. Clin Orthop 408:120–125, 2003
Manuscript received April 16, 2004.
Accepted in final form May 2, 2004.
Address reprint requests to: Daniel H. Kim, M.D., Department of
Neurosurgery, Stanford University Medical Center, Room R-201,
Edwards Building, 300 Pasteur Drive, Stanford, California 943055327. email: [email protected].
Neurosurg. Focus / Volume 16 / June, 2004