Keywords
gunshot injury to spine - gunshot wound - spinal injury - complete spinal cord injury
- penetrating spinal injury
Introduction
Gunshot injury to the spine, which was earlier associated predominantly with military
population, is now seen increasingly in civilians due to easy availability of firearms
and increase in violence in the society. With the contribution of 13 to 17%[1] to spinal injuries, it is now the third most common cause[2] of spinal injury after road traffic accident and fall. Gunshot wound (GSW) to the
spine is a complex injury, with its ideal management being a matter of controversy.
Medical literature is replete with studies in favor of conservative management, thus
implying that surgical intervention does not have much effect on the ultimate neurologic
outcome. However, the catch lies in the fact that most of the studies are based on
military population where there is a frequent use of firearms with a relatively higher
muzzle velocity. Velocity of impact leading to the kinetic energy imparted by it to
the surrounding tissues is the single most important factor that determines the destructive
ability of projectile. Gunshot injuries are generally caused by high-velocity assault
weapons of the range 2,800 to 3,800 ft/s.[3] Thus, GSWs received on battlefield are more destructive and carry a worse prognosis
than their civilian counterparts.[4] Spinal cord injury (SCI) from spinal GSWs frequently presents as complete lesion
with a decreased potential for neurologic recovery than with closed trauma.[5] Bullet damages tissues by the following mechanisms: (1) direct impact of bullet
along its path, (2) pressure or shock waves created by bullet, and (3) temporary cavitation.
SCI may therefore occur even without any obvious injury to the spinal canal and is
termed as spinal cord concussion.
Commonly accepted indications for surgery include persistent cerebrospinal fluid (CSF)
fistula, arteriovenous fistula, evidence of overt cervical instability, infection,
and neurologic deterioration (particularly for incomplete injury) with persistent
spinal cord impingement.
Current neurosurgical practice guidelines have not laid any protocol for the management
of penetrating spine injuries. The most important question that a neurosurgeon encounters
is whether a decompressive laminectomy can escalate the chance of regaining the neurologic
function that is already lost or at least arrest the progression of the same.
Aims and Objectives
The aim of this study was to analyze the outcome of surgical intervention in patients
with firearm injury to the spine with respect to the neurologic status.
Methods
It is a retrospective study conducted at Department of Neurosurgery, Jai Prakash Narayan
Apex Trauma Centre (JPCATC), All India Institutes of Medical Sciences (AIIMS), New
Delhi. Case records of patients with gunshot injury to the spine from January 2013
to March 2018, in whom surgical intervention was planned, were initially retrieved.
These records were then analyzed on the basis of various clinical, radiologic, and
operative parameters. At follow-up, patients were clinically examined, and radiologic
status was compared with that of the preoperative variables. Assessment in the neurologic
status was done using Nurick grade.
All the results were analyzed in SPSS version 22.0 (IBM Inc.). The analysis included
frequency table, bar diagram, and pie chart. All quantitative variables were estimated
using measures of central location (mean) and measures of dispersion (standard deviation).
Pearson chi-square test and Student t-test were used for qualitative variables and continuous variables, respectively.
For all analysis, p < 0.05 was considered statistically significant.
Results
Case records of 20 patients were analyzed. Two patients were excluded due to early
mortality during the period of stabilization, and one patient refused surgery. Out
of 17 patients who underwent surgery, 1 died during the hospital stay, hence was also
excluded.
All the patients included in the study were males with the mean age of 29.93 ± 2.95
years ranging from 16 to 56 years. Fourteen (87.5%) out of 16 patients suffered a
civilian gunshot injury whereas the mode of injury was military in the remaining 2
patients.
Entry wound was most common on the back (56.25%), with the thoracic vertebra being
the most common segment involved (68.75%) followed by cervical (18.75%) and lumbar
(12.5%). Exit wound was present in only 2 of the 16 patients.
Nine (56.25%) out of 16 patients had complete SCI at presentation (ASIA-A). Canal
compromise was present in seven (43.75%) patients. Out of 11 patients with involvement
of the thoracic vertebra, 5 showed canal compromise, and 4 of these 5 patients were
ASIA-A at presentation. None of the three patients with cervical spinal injury had
canal compromise, and none of them had complete SCI at presentation.
Median time from gunshot injury to surgery was 25 days. All 16 patients underwent
decompressive procedures with 3 undergoing concurrent stabilization procedure as well.
Bullet removal was done in 8 (50%) out of 16 patients. Dural tear was seen in 13 (81.25%)
patients. Overall mean duration of postoperative hospital stay was 7 days whereas
it was prolonged to 8 days in patients with dural tear.
Median period of follow-up was 3.3 years. Overall 9 (56.25%) out of 16 patients showed
improvement in neurologic status at follow-up after surgery. Out of seven patients
with complete SCI who underwent surgery, only one patient showed improvement in neurologic
status at follow-up.
Out of seven patients with canal compromise who underwent surgery, neurologic improvement
at follow-up was observed in two patients.
Out of 16 patients undergoing decompressive procedure, 8 showed improvement whereas
1 out of 3 patients undergoing combined approach showed improvement in neurologic
status at follow-up.
Six out of eight patients undergoing bullet removal showed improvement in neurologic
status at follow-up.
All patients with involvement of the cervical and lumbar spine showed improvement
in neurologic status at follow-up. However, only 6 out of 11 patients with thoracic
spine involvement showed improvement in neurologic outcome.
Discussion
The patient demographics and patterns of injury in this study correspond closely to
those of prior studies. Most gunshot injuries to the spine occurred in young men with
a mean age of 29.93 ± 2.95 years in our population. All except two patients were victims
of a handgun, involved in civilian violence. [Fig. 1] depicts the frequency distribution of gunshots according to various vertebral levels
suffering ballistic injury, indicating a predominance of injury to the thoracic region.
These data correspond well with findings that injuries to the thoracic spine are more
common than lumbar and cervical injuries, presumably from the tendency of GSWs to
congregate around the center of the body mass.[6]
[7]
Fig. 1 The frequency distribution of patients according to various vertebral segments.
In the thoracic vertebra, the canal-to-cord ratio is less compared with that in the
cervical and lumbar regions. Hence, canal compromise and complete cord injury are
seen more frequently in gunshot injuries to the thoracic vertebra.[8] Thus the level of vertebral injury is an important factor in the long-term outcome
in cases with penetrating spinal injury. One patient with thoracic vertebra involvement
and without canal compromise also showed complete SCI, thus suggesting spinal cord
concussion as the reason behind such a clinical presentation.
All patients underwent decompressive procedures, and three of them also required additional
stabilization procedure. One of these three patients had D10 body and pedicle fracture;
thus, the patient underwent pedicle screw rod fixation along with laminectomy. The
other two patients had comminuted L3 body fracture and thus required additional stabilization
procedure. Civilian gunshot injuries are less likely to be unstable when compared
with their military counterparts, owing to the relatively less impact velocity. We
performed stabilization procedure in patients with injury to all the three columns.
In cases with involvement of only two columns, especially when pedicles are involved,
additional stabilization/fixation procedures should be contemplated.[4]
[9] [Figs. 2]
[3] show the images of two cases for which surgical intervention was undertaken.
Fig. 2 (A, B) Preoperative CT scans of a patient showing fracture of left pedicle of D1 with the
bullet in situ. (C) Healed entry wound. (D) Healed surgical wound.
Fig. 3 (A, B) The entry (left flank) and exit (right flank) wounds, respectively. (C, D) Preoperative images of L3 vertebral body fracture. (E, F) Position of the screws that were placed during the stabilization procedure.
Aarabi et al and Hammoud et al recruited sufficient number of operative and nonoperative
patients in each of the complete and incomplete neurologic deficit group. On analysis
they did not find surgery to be associated with improved neurologic status.[10]
[11] Cybulski et al[12] only evaluated patients with injuries to the cauda equina and concluded that surgery
was of benefit whether done early (< 72 hours) or late (> 72 hours), but there were
no control data (nonoperative) to support their conclusion. Benzel et al[13] concluded that surgery was indicated in those patients with incomplete injury and
evidence of continued compression of neurologic structures within the spinal canal.
In a multicenter study, Waters and Adkins[14] found that bullet removal benefited those with complete or incomplete lesions between
T12 and L4, but there was no effect in those between T1 and T11. [Table 1] shows in a nutshell the surgical outcome in the aforementioned studies.
Table 1
Surgical outcomes in various studies in nutshell
|
Author
|
Year
|
No. of patients
|
No. of surgeries
|
Percentage of improved outcome
|
|
Benzel et al[13]
|
1987
|
35
|
13
|
13 (100%)
|
|
Cybuiski et al[12]
|
1989
|
88
|
61
|
29 (47.5%)
|
|
Hammoud et al[11]
|
1995
|
64
|
23
|
4 (17.3%)
|
|
Aarabi et al[10]
|
1996
|
145
|
87
|
42 (48.27%)
|
In our study, at follow-up, 8 out of the 16 patients who had undergone decompressive
surgery showed improvement in the neurologic status by at least one Nurick grade.
Eight out of 16 patients got the bullet removed, and 6 of them showed improvement
in neurologic status at follow-up. Bullet removal during surgery was seen to be associated
with better outcome. Civilian gunshot injury, involvement of the cervical region,
lack of dural breach, and minimal intraoperative blood loss were associated with relatively
greater percentage of patients getting benefitted from surgical intervention. However,
the impact of these factors was not statistically significant enough to consider them
as predictors for a better surgical outcome. The only factor that was statistically
significant enough to influence the ultimate outcome was the neurologic status at
the time of presentation ([Table 2]).
Table 2
The degree of association of various factors on the improvement of surgical outcome
at follow-up.
|
Parameters
|
No. of patients
|
Improvement
|
p-value
|
|
Mode of Injury
|
|
|
0.88
|
|
Civilian
|
14
|
8
|
|
|
Military
|
2
|
1
|
|
|
Exit Wound
|
|
|
0.88
|
|
Yes
|
2
|
0
|
|
|
No
|
14
|
9
|
|
|
Complete SCI at presentation
|
|
|
0.006
|
|
Present
|
7
|
1
|
|
|
Absent
|
9
|
8
|
|
|
Vertebral level of injury
|
|
|
|
|
Cervical
|
3
|
3
|
0.15
|
|
Lumbar
|
2
|
1
|
0.88
|
|
Thoracic
|
11
|
6
|
0.63
|
|
Canal compromise
|
|
|
0.07
|
|
Present
|
7
|
2
|
|
|
Absent
|
9
|
7
|
|
|
Associated vascular injury
|
|
|
0.44
|
|
Present
|
1
|
1
|
|
|
Absent
|
15
|
8
|
|
|
Early surgery
|
|
|
|
|
Yes
|
7
|
4
|
0.67
|
|
No
|
9
|
5
|
|
|
Bullet removal
|
|
|
0.15
|
|
Yes
|
8
|
6
|
|
|
No
|
8
|
3
|
|
|
Minimal Intraoperative blood loss
|
|
|
0.15
|
|
Yes
|
8
|
6
|
|
|
No
|
8
|
31
|
|
|
Dural tear
|
|
|
0.60
|
|
Yes
|
13
|
7
|
|
|
No
|
3
|
2
|
|
Treatment of spinal GSW requires a multidisciplinary approach with the goal of maintaining
or restoring spinal stability and neurologic function and minimizing complications.
Neurologic deterioration in the presence of persistent spinal cord impingement is
an obvious and emergency indicator for surgical intervention. In this setting, if
emergency surgery can be performed to eliminate spinal cord compression, it should
be performed, particularly for epidural hematoma, empyema, bone or foreign-body compressive
intrusion, or other remediable mass lesion. If mechanical forces from unstable spinal
elements cause persistent compression and neurologic deterioration, cervical realignment
and fusion may also be necessary.
Conclusion
Surgical intervention, timing of surgery, amount of intra-operative blood loss, and
dural breach had no significant impact on the overall surgical outcome. Mode of injury,
cervical level of injury, and bullet removal were seen to be associated with improved
outcome. Neurologic status at the time of presentation is the single most important
factor that determines the ultimate outcome. Indication and type of surgical intervention
is still an ambiguous topic. Owing to lack of conclusive evidence, we believe that
there is role for surgery when it is done with an intention to restore the neurologic
function.
