- Email: [email protected]
Magnetic Resonance Imaging to Evaluate Cervical Spinal Cord Injury from Gunshot Wounds from Handguns
Accepted Manuscript Magnetic Resonance Imaging to Evaluate Cervical Spinal Cord Injury from Gunshot Wounds from Handguns Justin Slavin, M.D., Narlin Beaty, M.D., Prashant Raghavan, M.B.B.S., Charles Sansur, M.D., M.H.Sc., Bizhan Aarabi, M.D. PII:
S1878-8750(15)01049-9
DOI:
10.1016/j.wneu.2015.08.033
Reference:
WNEU 3137
To appear in:
World Neurosurgery
Received Date: 3 May 2015 Revised Date:
12 August 2015
Accepted Date: 14 August 2015
Please cite this article as: Slavin J, Beaty N, Raghavan P, Sansur C, Aarabi B, Magnetic Resonance Imaging to Evaluate Cervical Spinal Cord Injury from Gunshot Wounds from Handguns, World Neurosurgery (2015), doi: 10.1016/j.wneu.2015.08.033. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT
Magnetic Resonance Imaging to Evaluate Cervical Spinal Cord Injury from Gunshot Wounds from Handguns Justin Slavin, M.D., 1 Narlin Beaty, M.D.,1 Prashant Raghavan, M.B.B.S.2, Charles Sansur, M.D., M.H.Sc. 1 Bizhan Aarabi M.D. 1, **
RI PT
1. University of Maryland, Department of Neurosurgery, 22 South Greene St. Suite S12-D, Baltimore, MD 21201
AC C
EP
TE D
M AN U
** Address Correspondence to: Bizhan Aarabi M.D. 22 South Greene St. Suite S-12-D Baltimore, MD 21201 Telephone: 410-328-6034 Fax: 410-328- 0756 Email: [email protected]
SC
2. University of Maryland, Department of Diagnostic Radiology and Nuclear Medicine, 22 South Greene St., Baltimore, MD 21201
ACCEPTED MANUSCRIPT
ABSTRACT
Background and Purpose: Patients presenting with gunshot wounds (GSWs) to the neck
RI PT
are difficult to assess due to often severe injuries that are incompletely evaluated by CT alone. Our institution treats hundreds of GSW patients each year and we present our experience using MRI in the evaluation of cervical GSWs.
SC
Materials and Methods: From August 2000 to July 2012 all GSWs to the cervical spine treated at our institution were cataloged. Seventeen patients had one or more MRI studies
M AN U
of the cervical spine. Informed consent was obtained prior to MRI indicating the risks of retained metal fragments in the setting of high magnetic fields. CT scans were obtained before and after MRI to document any possible migration of metal fragments. We documented patients’ neurologic examination results before and after MRI and at follow
TE D
up.
Results: Patients’ age range was 18 to 56 (mean 29.8). Eleven of 17 patients had
EP
retained metal fragments seen on CT scan including 3 patients with fragments within the spinal canal. No patient experienced a decline in neurologic function after MRI. No
AC C
migration of retained fragments was observed. Fifteen of 17 patients returned for follow up examinations, with an average follow up interval of 39.1 weeks (range: 1.3 to 202.3 weeks; median: 8 weeks).
Conclusion: For carefully selected patients, MRI can be an effective tool in assessing GSWs to the neck and it can significantly improve the evaluation and management of this cohort. No patient in our series suffered a complication related to MRI.
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
Abbreviation Key: GSW=gunshot wound; SCI=spinal cord injury
ACCEPTED MANUSCRIPT
INTRODUCTION
The use of MR in the setting of cervical gunshot wounds (GSWs) remains controversial.
RI PT
CT angiogram is generally the first-line imaging modality to evaluate for vascular and osseous injury. However, CT does not reliably assess the integrity of ligamentous
structures or the status of the spinal cord and canal. In the setting of retained bullet
SC
fragments, CT scans may be impaired by severe metal artifact that can obscure adjacent structures. MRI offers distinct benefits over CT in assessing injury to the cervical spine
M AN U
after gunshot injury but often remains underutilized. Although studies5, 13 show that it is possible to perform MRI safely in the setting of retained bullets or fragments, the procedure must be undertaken with caution and the presence of retained bullet fragments remains a relative contraindication for MRI.2, 6, 7, 13
TE D
Most commercially available ammunition commonly seen in civilian ballistic trauma is lead-based, exhibits no ferromagnetic properties, and has no notable susceptibility to magnetic forces causing movement or rotation, heating, or even image distortion.6, 7, 12, 14
EP
However, studies did not test imported ammunition with potential for contamination of lead bullets with ferromagnetic material. Furthermore, higher velocity and armor-piercing
AC C
gun and rifle bullets are more likely to have ferromagnetic coating while shotgun pellets are increasingly produced from steel due to environmental concerns. Although there is an inherent risk of encountering an undetected or unknown ferromagnetic object, there is benefit to using MRI to better evaluate the extent of injury in certain patients. In this review of patients with GSW to the cervical spine, we obtained 22 MRIs on 17 patients. The dramatic reduction in metal artifact around retained fragments allows for evaluation
ACCEPTED MANUSCRIPT
of important vascular and neurologic structures. We show that MRI is an effective clinical tool with value above and beyond CT angiography for evaluation of injury and prognosis in the setting of cervical GSWs. Our data is novel in its incorporation of
RI PT
American Spinal Injury Association motor exams before and after MRI to illustrate that retained bullet fragments did notcause progression of neurological deficit.
SC
METHODS
From August 2000 to July 2012 all gunshot wounds to the cervical spine at our institution
M AN U
were cataloged.3 This retrospective review of imaging and electronic medical records was approved by the institutional review board. Of the 144 gunshot wounds to the cervical spine, forty patients sustained cervical spinal cord injury (SCI). The protocol for diagnostic imaging in the setting of penetrating cervical trauma at our institution is CT
TE D
angiography of the neck which is routinely performed within 1 hour of admission. Sixteen of these patients also underwent MRI during their initial hospitalization and one patient had an MRI on a secondary admission after experiencing a decline in neurologic
EP
exam in the weeks after discharge. Four patients had multiple MRIs performed during their hospitalization with 3 patients having a repeat MRI and one patient having 3 total
AC C
scans.CT imaging was routinely performed immediately after MRI to evaluate for migration of any retained fragments.
All MRI studies were performed with a 1.5T magnet. MRI protocol for cervical traumatic injury at our institution is T1, T2, and STIR sequences with axial and sagittal reconstructions. T1 with gadolinium sequences are performed when indicated (i.e. concern for infection). T2-weighted spin echo sequences could be added to the study if
ACCEPTED MANUSCRIPT
artifact on T2 sequences continued to obstruct the area of interest as this sequence reduces the impact of susceptibility artifact.
RI PT
American Spinal Injury Association scale assessments were obtained on admission as well as during follow up neurosurgical examinations and compared to radiographic
findings. Two of the 17 patients who had MRI were lost to follow up; both patients were
SC
high functioning with deficits corresponding to mild central cord syndrome.
MRI was performed on patients after discussion between the neurosurgery service,
M AN U
trauma surgery service, radiology, and the patient when CT imaging was determined to be inadequate to guide treatment. Discussion with patients regarding the safety of MRI and supplemental informed consent for the procedure focused on the risk of high magnetic fields on retained metal of unknown composition. We counseled patients that
TE D
published data indicated that lead-based ammunition, the majority of domestic ammunition for handguns, did not exhibit dangerous ferromagnetic qualities but that exceptions were certainly possible. We also advised that consequences of retained
EP
ferromagnetic material in high magnetic fields were unpredictable and could include migration and heating of the fragments which could cause new neurologic deficits or
AC C
local tissue or vascular injury. In the case of repeat MRI, we continued to advise patients that there was still risk to the study even if no problems had occurred in a prior study.
RESULTS
Seventeen patients had MR imaging of the cervical spine performed using a 1.5T magnet after gunshot injuries. The age range of the patients was 18 to 56 (mean 29.8). All of the
ACCEPTED MANUSCRIPT
patients sustained injuries with handguns consisting of 16 assaults and one self-inflicted wound. Eleven patients were found to have complete SCI (AIS A) with a clear neurologic level corresponding to pathology. One patient had complete motor deficit but retained
RI PT
sensory function below the level of injury (AIS B). Five patients presented with
incomplete injury consistent with central cord (AIS C/D). Fifteen of 17 patients returned for follow up examinations. Excluding those without follow up, average follow up
SC
interval was 39.1 weeks (range of 1.3 to 202.3 weeks, median 8 weeks). These data are
M AN U
shown in Table 1.
The average time between admission and first MRI was 42 hours (range 3.5 hours to 7 days). We assessed the T2 and STIR sequences for each patient to evaluate for evidence of SCI and ascending edema.1 T2-weighted spin echo sequences were not required in any patient as susceptibility artifacts were minimal on T2-weighted sequences. All patients
TE D
with ascending edema at least one vertebral level above the site of primary SCI, except for patient 1 (Table 1), had complete neurologic deficit at time of presentation and all of
EP
these patients remained complete at the time of follow up.
No patient experienced a change or decline in neurologic exam during or following MRI.
AC C
Eleven patients received MRI after GSW with retained metal fragments. Three had retained fragments directly adjacent to or within the cervical spinal cord. Furthermore, each of the 11 patients with retained fragments received additional CT scan immediately following MRI which confirmed no migration of the metallic fragment. Additionally, T2weighted images did not suffer any significant susceptibility artifact from retained metal suggesting the absence of ferromagnetic properties in the retained metal.
ACCEPTED MANUSCRIPT
Police records, bullet manufacturing/caliber when available, and patient testimony were used to confirm weapon type. Bullet trajectories were analyzed from the admission CT scan to determine whether the trajectory included the spinal canal (intracanalicular) or
RI PT
excluded the spinal canal (extracanalicular). Entry and exit wounds recorded on physical examination were also used to identify the bullet’s trajectory. Three representative cases
relevant prognostic and diagnostic information.
M AN U
Representative Cases
SC
have been selected to illustrate the ability of MRI in the setting of bullet injury to obtain
Patient 1 (Table 1) experienced a penetrating injury to the cervical spine and presented with bilateral arm and hand weakness (ASIA motor score 59 and AIS grade D). CT scan showed a bullet trajectory passing through the spinous process of C3, 6mm displaced
TE D
from the spinal canal. Although there were retained metal fragments in the paraspinous muscles, we performed an MRI of the cervical spine to better determine the presence of SCI. On MRI, we did see evidence of injury to the spinal cord with significant ascending
EP
edema but did not observe evidence of ligamentous instability (Figure 1). As a result of MRI correlation with physical exam, we continued mean arterial pressure elevation
AC C
greater than 80mmHg for 48 hours per the standard protocol for blunt cervical trauma at our institution. Further, the patient was managed with a cervical orthotic device only as MRI demonstrated no instability necessitating surgical fusion or canal compromise requiring decompression.
Patient 14 presented with severe neurologic deficit (ASIA 0A) after GSW to the cervical spine. CT scan showed a bullet trajectory with the bullet impacting the lateral mass ,
ACCEPTED MANUSCRIPT
deflected laterally into the deep cervical soft tissues. Severe metal artifact precluded identification of cord injury or compressive epidural hematoma. We performed an MRI specifically to evaluate the integrity of the spinal cord to see if, despite the severe
RI PT
deficits, there was any potential for recovery of neurologic function (Figure 2).
Unfortunately, MRI did confirm severe SCI allowing us to offer the family an accurate
prognostic picture for this patient and identified the lack of need for surgical intervention.
SC
Despite bullet fragments within the spinal canal, we were able to visualize all relevant structures in full detail and observed no change in position in these fragments in
M AN U
subsequent CT scans.
Patient 17 (Table 1) represented a diagnostic problem with persistent fevers after injury. Infections are common complications after GSW, particularly in the neck when injuries to the esophagus and trachea are not infrequent. On workup for fever, this patient was
TE D
noted to have enlarging deep cervical collections in line with the bullet’s trajectory seen on CT with contrast. Injuries that impact or traverse the spinal canal can result in
EP
pseudomeningoceles which are indistinguishable from seromas and abscesses on CT. These collections were also difficult to fully evaluate due to metal artifact impairing clear
AC C
views of the potentially enhancing margins. MRI with contrast was performed which definitively revealed collections with no mural enhancement, containing fluid isointense to CSF, contiguous with the spinal canal consistent with pseudomeningocele without external cerebrospinal fluid fistula (Figure 3). This spared the patient an exploratory surgery for evacuation of potential abscess which would likely have resulted in a CSF fistula through the operative site.
ACCEPTED MANUSCRIPT
DISCUSSION
Safety of MRI in GSW
RI PT
In the setting of GSW with retained bullet fragments, MRI has been demonstrated to be safe in some cases6, 11, 13. In-vitro studies of MRI and ammunition show that the only
factors in formation of artifact and dispersion of heat of gunshot bullets were composition
SC
and metal purity of the embedded slug.8 In several studies examining the ferromagnetic
properties of ammunition and safety within strong magnetic fields of up to 1.5T, the most
M AN U
consistent factor causing susceptibility to magnetic forces was the presence of steel in the bullet’s manufacture.4, 8-10, 12 For most commercial handgun ammunition, the presence of steel is rare in bullets with the notable exception of armor-piercing ordinance.4, 8, 10, 12 Steel ammunition is more commonly found in high velocity rifle ammunition, especially
TE D
rounds for AK-47 and M-16 rifles, as well as many shotgun pellets used for hunting (to limit environmental impact of lead rounds).4, 8, 12 Of the studies available, the only handgun bullet with primarily lead composition to exhibit ferromagnetic properties is the
EP
Century Arm 7.38mm Mauser ammunition.12 Dedini et al. even tested ammunition at higher fields up to 7T with no evidence of magnetic forces or susceptibility artifact in any
AC C
ammunition determined to be safe at 1.5T (all ammunition without steel tested as safe in all field strengths).4
Two in vivo studies also argue the relative safety of performing MRI with retained bullet fragments after civilian assault injuries. Smugar et al and Finitsis et al evaluated a total of 38 gunshot injury patients with retained bullet fragments using 1.5T MRI. No patients demonstrated any new neurologic deficits after MRI nor was there any observed bullet
ACCEPTED MANUSCRIPT
rotation or migration on subsequent radiographic imaging.6, 13 Our case series supports the above findings in that, although the specific types of ammunition were unknown at the time of decision to proceed with MRI, no patients who underwent MRI with retained
RI PT
bullets including intracanalicular fragmentssuffered neurologic deterioration or other
complications. Six patients demonstrated improvement in neurologic exam after follow up. Although 2 patients did experience slight declines in exam overall (Patient 14,
SC
8A→6A; Patient 3, 30A→27A), these declines were noted over a week after any
exposure to MRI and the patients had stable exams immediately following the imaging
M AN U
procedure. CT scans performed immediately after MRI also confirmed the absence of bullet migration. No patient noted any complaint of discomfort or sensation of warmth during the scan. The absence of metal artifact on T2-weighted imaging further suggests the absence of ferromagnetic properties in any retained fragments.
TE D
Published data does support that most conventional handgun ammunition is safe for patients undergoing strong magnetic fields. Nevertheless, when possible, it is prudent to obtain spent cartridges or samples of the ammunition used to determine ferromagnetic
EP
properties (more likely in accidental or self-inflicted injuries as opposed to assault).7, 14 If
AC C
ammunition samples are not available, both the surgeon and radiologist need to consider the proximity of a metallic fragment to patent vascular structures or functional neurologic structures. One study suggests that ferromagnetic detection systems would be adequate to determine if a specific retained fragment exhibited ferromagnetic properties that would preclude MRI.10 Even if the determination is made that an MRI scan is relatively low risk, in this series we counseled the patient and family extensively and obtained informed consent for potential interaction between magnetic fields and retained metal fragments.
ACCEPTED MANUSCRIPT
Rifle and shotgun ammunition has a much higher likelihood of containing ferromagnetic steel and should be considered a contraindication to undergoing MRI unless that
RI PT
ammunition is proven to be safe.
Utility of MRI in GSW
MRI offers benefits over conventional imaging modalities to justify its use including
SC
improved soft tissue visualization and reduced metal artifact. MRI has a major role in the assessment of blunt trauma to the cervical spine because of its unique ability to show the
M AN U
integrity of soft tissue structures. These include the spinal cord, spinal ligaments, cerebrospinal fluid, paraspinous musculature and cervical vasculature. This is especially important for penetrating trauma because the pathology cannot be completely predicted by bony injury and bullet trajectory alone. There are numerous potential clinical
TE D
indications where MRI is valuable; two prime examples are concussive SCI and persistent spinal cord compression from soft tissue injury. Patient 1, as described previously, had significant concussive SCI despite no penetrating trauma to the cord and
EP
no destabilizing injury. Persistent spinal cord compression, e.g., epidural hematoma, is an indication for urgent operative decompression and can only be shown by MRI if metal
AC C
streak artifact obscures the area of interest.3
Non-ferromagnetic and ferromagnetic projectile fragments, by virtue of their increased density may produce an unacceptable level of streak artifact on CT and CT angiography resulting in obscuration of adjacent osseous, vascular and soft tissue structures. Non feeromagnetic materials have no significant impact on MR images. Ferromagnetic fragments on the other hand may result in susceptibility artifacts on MRI. However, cord
ACCEPTED MANUSCRIPT
signal abnormalities are best depicted on T2 or T2W spin echo sequences, which typically suffer less degradation from such artifacts than those sequences acquired with a gradient echo technique. Figure 1 shows the difference in imaging clarity between CT
RI PT
and MRI in the setting of retained metal. Patient 17 (Table 1) is a clear example of this benefit in whom streak artifact precluded differentiation between abscess and
pseudomeningocele. Though data suggests non-ferromagnetic metal does not cause
SC
increased susceptibility artifact at higher magnetic field strengths, theconsequences of
M AN U
imaging patients at higher field strengths than 1.5T may be different from our experience.
Based on our experience with MRI in the setting of GSW to the cervical spine, we identify several clinical situations where MRI may be beneficial at guiding medical and surgical care. Though CT imaging (plain CT, angiography, myelogram) is adequate in many cases to obtain relevant diagnostic information, metal streak artifact can obscure
TE D
the area of interest in which case MRI can offer additional information. Patients with incomplete neurologic injury are of particular concern in obtaining adequate diagnostic
EP
information as these patients have higher potential for recovery of function with prompt treatment. However, even patients with complete neurologic injury merit evaluation for
AC C
compressive lesions to see if surgical decompression might offer improvement in function.
1) A patient with neurologic deficits in the setting of a bullet trajectory that is distant from the spinal canal. Concussive injury from the bullet’s energy transfer to surrounding tissues is sometimes enough to cause injury to the spinal cord. While it is unknown how best to manage these injuries medically, MRI can be useful in
ACCEPTED MANUSCRIPT
confirming the absence of a lesion requiring surgical decompression. Furthermore, MRI is uniquely suited to confirm the presence of a noncompressive spinal cord injury which we manage medically in similar fashion as a
RI PT
patient with central cord syndrome.
2) A patient with neurologic deficit in the setting of a bullet trajectory closely
SC
approximating the spinal canal but not traversing the canal. These patients are at high risk for compressive lesions with potentially salvageable neurologic
M AN U
function. MRI imaging can confirm the presence or absence of surgical lesions as well as provide prognostic information of SCI severity.
3) A patient with collections along the bullet’s tract concerning for infection or abscess. Although any collection along a penetrating injury tract is concerning for
TE D
infection, there is particular concern in the setting of fever or leukocytosis without another obvious source. The trauma population is often complicated by multiple
EP
injuries to multiple organ systems that could be responsible for fever or leukocytosis. In addition to avoiding metal streak artifact seen on CT, MRI is
AC C
more sensitive than CT at diagnosing soft tissue infections as well as diagnosing a CSF fistula.
CONCLUSION
Our experience with MRI in the setting of GSW has significantly improved the evaluation and management of our patient population. No patient in our series suffered
ACCEPTED MANUSCRIPT
an adverse event related to MRI. We were also able to obtain diagnostic information which was incompletely assessed or unavailable using CT imaging alone. For carefully selected patients with appropriate counseling on the risk and benefit of the study, MRI
AC C
EP
TE D
M AN U
SC
RI PT
can an effective tool in assessing and managing trauma related to GSW to the neck.
ACCEPTED MANUSCRIPT
REFERENCES
RI PT
Reference List 1. Aarabi B, Simard JM, Kufera JA, et al: Intramedullary lesion expansion on
magnetic resonance imaging in patients with motor complete cervical spinal cord
SC
injury. J Neurosurg Spine 17:243-250, 2012
M AN U
2. Bashir EF, Cybulski GR, Chaudhri K, et al: Magnetic resonance imaging and computed tomography in the evaluation of penetrating gunshot injury of the spine. Case report. Spine (Phila Pa 1976 ) 18:772-773, 1993
3. Beaty N, Slavin J, Diaz C, et al: Cervical spine injury from gunshot wounds. J
TE D
Neurosurg Spine 21:442-449, 2014
4. Dedini RD, Karacozoff AM, Shellock FG, et al: MRI issues for ballistic objects:
EP
information obtained at 1.5-, 3- and 7-Tesla. Spine J 13:815-822, 2013 5. Eshed I, Kushnir T, Shabshin N, et al: Is magnetic resonance imaging safe for
AC C
patients with retained metal fragments from combat and terrorist attacks? Acta Radiol 51:170-174, 2010
6. Finitsis SN, Falcone S, Green BA: MR of the spine in the presence of metallic bullet fragments: is the benefit worth the risk? AJNR Am J Neuroradiol 20:354356, 1999
ACCEPTED MANUSCRIPT
7. Hammoud MA, Haddad FS, Moufarrij NA: Spinal cord missile injuries during the Lebanese civil war. Surg Neurol 43:432-437, 1995
RI PT
8. Hess U, Harms J, Schneider A, et al: Assessment of gunshot bullet injuries with the use of magnetic resonance imaging. J Trauma 49:704-709, 2000
9. Hollerman JJ, Fackler ML, Coldwell DM, et al: Gunshot wounds: 1. Bullets,
SC
ballistics, and mechanisms of injury. AJR Am J Roentgenol 155:685-690, 1990
M AN U
10. Karacozoff AM, Pekmezci M, Shellock FG: Armor-piercing bullet: 3-T MRI findings and identification by a ferromagnetic detection system. Mil Med 178:e380-e385, 2013
11. Martinez-del-Campo E, Rangel-Castilla L, Soriano-Baron H, et al: Magnetic
TE D
resonance imaging in lumbar gunshot wounds: an absolute contraindication? Neurosurg Focus 37:E13, 2014
12. Smith AS, Hurst GC, Duerk JL, et al: MR of ballistic materials: imaging artifacts
EP
and potential hazards. AJNR Am J Neuroradiol 12:567-572, 1991
AC C
13. Smugar SS, Schweitzer ME, Hume E: MRI in patients with intraspinal bullets. J Magn Reson Imaging 9:151-153, 1999
14. Teitelbaum GP, Yee CA, Van Horn DD, et al: Metallic ballistic fragments: MR imaging safety and artifacts. Radiology 175:855-859, 1990
ACCEPTED MANUSCRIPT
Table 1 Patient demographics and relevant clinical and radiographic findings Level of SCI focus
Bony Injury
on MRI
Trajectory through
Deepest retained
Admissio
Follow up ASIA
Time to MRI
Ascending
spinal canal
fragment
n ASIA
(time of follow
(hours)
edema (cm)
RI PT
#/gender/age
up weeks)
none
C3 spinous process
No
Paraspinal Muscle
59D
none
87
2.13
R C3 lateral mass
No
none
75D
85D (1.3)
6
3/M/56
0
C7
C6-7 spinous process
No
none
30A
27A (1.6)
10
0
4/M/52
C3
L C3/4 lateral mass/lamina
No
Deep Subcutaneous
98D
100D (36.6)
25
0
5/M/31
C7
Total destruction of C6-7 vertebral
No
Vertebral Body
0A
15A (4.0)
47
9.76
Subarachnoid
14A
24A (3.1)
5, 64
4.47, 7.5
none
70D
none
3.5
0
none
8A
8A (8.0)
43
7.73
Subarachnoid
0A
0A (17.9)
31, 10.5
8.0, 8.0
bodies and L lateral masses 6/M/20
C7
7/M/22
none
8/M/42
C7
B C7-T1 lateral mass and lamina
Yes
9/M/21
C5
B C4-5 pedicle and lamina
No
C5
Yes No
B C4-5 lamina and spinous process
11/M/22
C7
C5 vertebral body and L C6-7
12/M/30
C7
C7-T1 vertebral body
13/M/30
C5
C4-5 lamina
14/M/21
C6
C6-7 vertebral body
15/M/18
C7
T1 vertebral body
16/M/33
C6
C6-7 vertebral body and B lateral
17/M/23
C5
AC C
masses
none
0A
0A (2.9)
184
5.34
No
None
3A
8A (32.9)
48
4.17
No
Vertebral Body
100D
81D (18.4)
46 days
0
Yes
R Lateral Mass
0A
0A (202)
119
9.43
Yes
Intramedullary
8A
6A (3.4)
34
5.73
Yes
Deep Fascial
6A
26A (138)
7
6.04
No
none
NT/B*
NT/B* (72)
15, 12 days
0, 1.65
No
Paraspinal
0A
0A (7.1)
9, (11, 27
6.86, 7.5,
days)
3.58
EP
pedicle and lamina
R C4-5 lamina and lateral mass
*NT – non testable ASIA exam because of concomitant traumatic head injur
days
No
TE D
10/M/44
R C7 lateral mass and L C7 lamina C4 spinous process
SC
C3
2/M/19
M AN U
1/M/22
Musculature
ACCEPTED MANUSCRIPT
LEGENDS:
Figure 1 (A) and (B) Axial and sagittal CT scans without contrast show ballistic
RI PT
fragments in a trajectory through the spinous process and paraspinous musculature
without impingement of the spinal canal. (C) Sagittal T2 MRI sequence demonstrates cord signal hyperintensity at the level of bony injury.
SC
Figure 2 (A) Axial CT angiogram shows retained bullet fragments in the spinal canal and paraspinous musculature. Severe metal streak artifact limits visualization of the contents
M AN U
of the spinal canal and evaluation of bony injury. (B) Axial T2 MRI reveals minimal metal artifact from the retained fragments and confirms spinal cord injury. (C) Sagittal CT angiogram shows bullet fragments with similar limitations from streak artifact regarding the contents of the spinal canal. (D) Sagittal T2 MRI shows significant spinal
TE D
cord injury with cord contusion and ascending edema.
Figure 3 (A) Axial CT angiogram from admission shows fractured posterior elements and retained metal fragments. (B) Axial CT with contrast to evaluate for suspected
EP
abscess in the setting of fever and leukocytosis. New fluid collection is seen in paraspinous tissues in the bullet trajectory concerning for abscess vs. pseudomeningocele.
AC C
(C) and (D) Sagittal and axial T2 MRI images shows collection to be in direct communication with subarachnoid space indicative of pseudomeningocele.
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT
Highlights
•
AC C
EP
TE D
M AN U
SC
•
MRI offers improved visualization over CT scan of soft tissue structures of the spine after GSW including the spinal cord, ligamentous injury, and epidural collections with reduced metal streak artifact. Using 1.5T magnets, no patient in our series suffered injury as a result of undergoing MRI of the cervical spine. Frank discussions regarding the risk of potential complications weighed against the potential benefits of the study should be part of a detailed informed consent process.
RI PT
•
S1878-8750(15)01049-9
DOI:
10.1016/j.wneu.2015.08.033
Reference:
WNEU 3137
To appear in:
World Neurosurgery
Received Date: 3 May 2015 Revised Date:
12 August 2015
Accepted Date: 14 August 2015
Please cite this article as: Slavin J, Beaty N, Raghavan P, Sansur C, Aarabi B, Magnetic Resonance Imaging to Evaluate Cervical Spinal Cord Injury from Gunshot Wounds from Handguns, World Neurosurgery (2015), doi: 10.1016/j.wneu.2015.08.033. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT
Magnetic Resonance Imaging to Evaluate Cervical Spinal Cord Injury from Gunshot Wounds from Handguns Justin Slavin, M.D., 1 Narlin Beaty, M.D.,1 Prashant Raghavan, M.B.B.S.2, Charles Sansur, M.D., M.H.Sc. 1 Bizhan Aarabi M.D. 1, **
RI PT
1. University of Maryland, Department of Neurosurgery, 22 South Greene St. Suite S12-D, Baltimore, MD 21201
AC C
EP
TE D
M AN U
** Address Correspondence to: Bizhan Aarabi M.D. 22 South Greene St. Suite S-12-D Baltimore, MD 21201 Telephone: 410-328-6034 Fax: 410-328- 0756 Email: [email protected]
SC
2. University of Maryland, Department of Diagnostic Radiology and Nuclear Medicine, 22 South Greene St., Baltimore, MD 21201
ACCEPTED MANUSCRIPT
ABSTRACT
Background and Purpose: Patients presenting with gunshot wounds (GSWs) to the neck
RI PT
are difficult to assess due to often severe injuries that are incompletely evaluated by CT alone. Our institution treats hundreds of GSW patients each year and we present our experience using MRI in the evaluation of cervical GSWs.
SC
Materials and Methods: From August 2000 to July 2012 all GSWs to the cervical spine treated at our institution were cataloged. Seventeen patients had one or more MRI studies
M AN U
of the cervical spine. Informed consent was obtained prior to MRI indicating the risks of retained metal fragments in the setting of high magnetic fields. CT scans were obtained before and after MRI to document any possible migration of metal fragments. We documented patients’ neurologic examination results before and after MRI and at follow
TE D
up.
Results: Patients’ age range was 18 to 56 (mean 29.8). Eleven of 17 patients had
EP
retained metal fragments seen on CT scan including 3 patients with fragments within the spinal canal. No patient experienced a decline in neurologic function after MRI. No
AC C
migration of retained fragments was observed. Fifteen of 17 patients returned for follow up examinations, with an average follow up interval of 39.1 weeks (range: 1.3 to 202.3 weeks; median: 8 weeks).
Conclusion: For carefully selected patients, MRI can be an effective tool in assessing GSWs to the neck and it can significantly improve the evaluation and management of this cohort. No patient in our series suffered a complication related to MRI.
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
Abbreviation Key: GSW=gunshot wound; SCI=spinal cord injury
ACCEPTED MANUSCRIPT
INTRODUCTION
The use of MR in the setting of cervical gunshot wounds (GSWs) remains controversial.
RI PT
CT angiogram is generally the first-line imaging modality to evaluate for vascular and osseous injury. However, CT does not reliably assess the integrity of ligamentous
structures or the status of the spinal cord and canal. In the setting of retained bullet
SC
fragments, CT scans may be impaired by severe metal artifact that can obscure adjacent structures. MRI offers distinct benefits over CT in assessing injury to the cervical spine
M AN U
after gunshot injury but often remains underutilized. Although studies5, 13 show that it is possible to perform MRI safely in the setting of retained bullets or fragments, the procedure must be undertaken with caution and the presence of retained bullet fragments remains a relative contraindication for MRI.2, 6, 7, 13
TE D
Most commercially available ammunition commonly seen in civilian ballistic trauma is lead-based, exhibits no ferromagnetic properties, and has no notable susceptibility to magnetic forces causing movement or rotation, heating, or even image distortion.6, 7, 12, 14
EP
However, studies did not test imported ammunition with potential for contamination of lead bullets with ferromagnetic material. Furthermore, higher velocity and armor-piercing
AC C
gun and rifle bullets are more likely to have ferromagnetic coating while shotgun pellets are increasingly produced from steel due to environmental concerns. Although there is an inherent risk of encountering an undetected or unknown ferromagnetic object, there is benefit to using MRI to better evaluate the extent of injury in certain patients. In this review of patients with GSW to the cervical spine, we obtained 22 MRIs on 17 patients. The dramatic reduction in metal artifact around retained fragments allows for evaluation
ACCEPTED MANUSCRIPT
of important vascular and neurologic structures. We show that MRI is an effective clinical tool with value above and beyond CT angiography for evaluation of injury and prognosis in the setting of cervical GSWs. Our data is novel in its incorporation of
RI PT
American Spinal Injury Association motor exams before and after MRI to illustrate that retained bullet fragments did notcause progression of neurological deficit.
SC
METHODS
From August 2000 to July 2012 all gunshot wounds to the cervical spine at our institution
M AN U
were cataloged.3 This retrospective review of imaging and electronic medical records was approved by the institutional review board. Of the 144 gunshot wounds to the cervical spine, forty patients sustained cervical spinal cord injury (SCI). The protocol for diagnostic imaging in the setting of penetrating cervical trauma at our institution is CT
TE D
angiography of the neck which is routinely performed within 1 hour of admission. Sixteen of these patients also underwent MRI during their initial hospitalization and one patient had an MRI on a secondary admission after experiencing a decline in neurologic
EP
exam in the weeks after discharge. Four patients had multiple MRIs performed during their hospitalization with 3 patients having a repeat MRI and one patient having 3 total
AC C
scans.CT imaging was routinely performed immediately after MRI to evaluate for migration of any retained fragments.
All MRI studies were performed with a 1.5T magnet. MRI protocol for cervical traumatic injury at our institution is T1, T2, and STIR sequences with axial and sagittal reconstructions. T1 with gadolinium sequences are performed when indicated (i.e. concern for infection). T2-weighted spin echo sequences could be added to the study if
ACCEPTED MANUSCRIPT
artifact on T2 sequences continued to obstruct the area of interest as this sequence reduces the impact of susceptibility artifact.
RI PT
American Spinal Injury Association scale assessments were obtained on admission as well as during follow up neurosurgical examinations and compared to radiographic
findings. Two of the 17 patients who had MRI were lost to follow up; both patients were
SC
high functioning with deficits corresponding to mild central cord syndrome.
MRI was performed on patients after discussion between the neurosurgery service,
M AN U
trauma surgery service, radiology, and the patient when CT imaging was determined to be inadequate to guide treatment. Discussion with patients regarding the safety of MRI and supplemental informed consent for the procedure focused on the risk of high magnetic fields on retained metal of unknown composition. We counseled patients that
TE D
published data indicated that lead-based ammunition, the majority of domestic ammunition for handguns, did not exhibit dangerous ferromagnetic qualities but that exceptions were certainly possible. We also advised that consequences of retained
EP
ferromagnetic material in high magnetic fields were unpredictable and could include migration and heating of the fragments which could cause new neurologic deficits or
AC C
local tissue or vascular injury. In the case of repeat MRI, we continued to advise patients that there was still risk to the study even if no problems had occurred in a prior study.
RESULTS
Seventeen patients had MR imaging of the cervical spine performed using a 1.5T magnet after gunshot injuries. The age range of the patients was 18 to 56 (mean 29.8). All of the
ACCEPTED MANUSCRIPT
patients sustained injuries with handguns consisting of 16 assaults and one self-inflicted wound. Eleven patients were found to have complete SCI (AIS A) with a clear neurologic level corresponding to pathology. One patient had complete motor deficit but retained
RI PT
sensory function below the level of injury (AIS B). Five patients presented with
incomplete injury consistent with central cord (AIS C/D). Fifteen of 17 patients returned for follow up examinations. Excluding those without follow up, average follow up
SC
interval was 39.1 weeks (range of 1.3 to 202.3 weeks, median 8 weeks). These data are
M AN U
shown in Table 1.
The average time between admission and first MRI was 42 hours (range 3.5 hours to 7 days). We assessed the T2 and STIR sequences for each patient to evaluate for evidence of SCI and ascending edema.1 T2-weighted spin echo sequences were not required in any patient as susceptibility artifacts were minimal on T2-weighted sequences. All patients
TE D
with ascending edema at least one vertebral level above the site of primary SCI, except for patient 1 (Table 1), had complete neurologic deficit at time of presentation and all of
EP
these patients remained complete at the time of follow up.
No patient experienced a change or decline in neurologic exam during or following MRI.
AC C
Eleven patients received MRI after GSW with retained metal fragments. Three had retained fragments directly adjacent to or within the cervical spinal cord. Furthermore, each of the 11 patients with retained fragments received additional CT scan immediately following MRI which confirmed no migration of the metallic fragment. Additionally, T2weighted images did not suffer any significant susceptibility artifact from retained metal suggesting the absence of ferromagnetic properties in the retained metal.
ACCEPTED MANUSCRIPT
Police records, bullet manufacturing/caliber when available, and patient testimony were used to confirm weapon type. Bullet trajectories were analyzed from the admission CT scan to determine whether the trajectory included the spinal canal (intracanalicular) or
RI PT
excluded the spinal canal (extracanalicular). Entry and exit wounds recorded on physical examination were also used to identify the bullet’s trajectory. Three representative cases
relevant prognostic and diagnostic information.
M AN U
Representative Cases
SC
have been selected to illustrate the ability of MRI in the setting of bullet injury to obtain
Patient 1 (Table 1) experienced a penetrating injury to the cervical spine and presented with bilateral arm and hand weakness (ASIA motor score 59 and AIS grade D). CT scan showed a bullet trajectory passing through the spinous process of C3, 6mm displaced
TE D
from the spinal canal. Although there were retained metal fragments in the paraspinous muscles, we performed an MRI of the cervical spine to better determine the presence of SCI. On MRI, we did see evidence of injury to the spinal cord with significant ascending
EP
edema but did not observe evidence of ligamentous instability (Figure 1). As a result of MRI correlation with physical exam, we continued mean arterial pressure elevation
AC C
greater than 80mmHg for 48 hours per the standard protocol for blunt cervical trauma at our institution. Further, the patient was managed with a cervical orthotic device only as MRI demonstrated no instability necessitating surgical fusion or canal compromise requiring decompression.
Patient 14 presented with severe neurologic deficit (ASIA 0A) after GSW to the cervical spine. CT scan showed a bullet trajectory with the bullet impacting the lateral mass ,
ACCEPTED MANUSCRIPT
deflected laterally into the deep cervical soft tissues. Severe metal artifact precluded identification of cord injury or compressive epidural hematoma. We performed an MRI specifically to evaluate the integrity of the spinal cord to see if, despite the severe
RI PT
deficits, there was any potential for recovery of neurologic function (Figure 2).
Unfortunately, MRI did confirm severe SCI allowing us to offer the family an accurate
prognostic picture for this patient and identified the lack of need for surgical intervention.
SC
Despite bullet fragments within the spinal canal, we were able to visualize all relevant structures in full detail and observed no change in position in these fragments in
M AN U
subsequent CT scans.
Patient 17 (Table 1) represented a diagnostic problem with persistent fevers after injury. Infections are common complications after GSW, particularly in the neck when injuries to the esophagus and trachea are not infrequent. On workup for fever, this patient was
TE D
noted to have enlarging deep cervical collections in line with the bullet’s trajectory seen on CT with contrast. Injuries that impact or traverse the spinal canal can result in
EP
pseudomeningoceles which are indistinguishable from seromas and abscesses on CT. These collections were also difficult to fully evaluate due to metal artifact impairing clear
AC C
views of the potentially enhancing margins. MRI with contrast was performed which definitively revealed collections with no mural enhancement, containing fluid isointense to CSF, contiguous with the spinal canal consistent with pseudomeningocele without external cerebrospinal fluid fistula (Figure 3). This spared the patient an exploratory surgery for evacuation of potential abscess which would likely have resulted in a CSF fistula through the operative site.
ACCEPTED MANUSCRIPT
DISCUSSION
Safety of MRI in GSW
RI PT
In the setting of GSW with retained bullet fragments, MRI has been demonstrated to be safe in some cases6, 11, 13. In-vitro studies of MRI and ammunition show that the only
factors in formation of artifact and dispersion of heat of gunshot bullets were composition
SC
and metal purity of the embedded slug.8 In several studies examining the ferromagnetic
properties of ammunition and safety within strong magnetic fields of up to 1.5T, the most
M AN U
consistent factor causing susceptibility to magnetic forces was the presence of steel in the bullet’s manufacture.4, 8-10, 12 For most commercial handgun ammunition, the presence of steel is rare in bullets with the notable exception of armor-piercing ordinance.4, 8, 10, 12 Steel ammunition is more commonly found in high velocity rifle ammunition, especially
TE D
rounds for AK-47 and M-16 rifles, as well as many shotgun pellets used for hunting (to limit environmental impact of lead rounds).4, 8, 12 Of the studies available, the only handgun bullet with primarily lead composition to exhibit ferromagnetic properties is the
EP
Century Arm 7.38mm Mauser ammunition.12 Dedini et al. even tested ammunition at higher fields up to 7T with no evidence of magnetic forces or susceptibility artifact in any
AC C
ammunition determined to be safe at 1.5T (all ammunition without steel tested as safe in all field strengths).4
Two in vivo studies also argue the relative safety of performing MRI with retained bullet fragments after civilian assault injuries. Smugar et al and Finitsis et al evaluated a total of 38 gunshot injury patients with retained bullet fragments using 1.5T MRI. No patients demonstrated any new neurologic deficits after MRI nor was there any observed bullet
ACCEPTED MANUSCRIPT
rotation or migration on subsequent radiographic imaging.6, 13 Our case series supports the above findings in that, although the specific types of ammunition were unknown at the time of decision to proceed with MRI, no patients who underwent MRI with retained
RI PT
bullets including intracanalicular fragmentssuffered neurologic deterioration or other
complications. Six patients demonstrated improvement in neurologic exam after follow up. Although 2 patients did experience slight declines in exam overall (Patient 14,
SC
8A→6A; Patient 3, 30A→27A), these declines were noted over a week after any
exposure to MRI and the patients had stable exams immediately following the imaging
M AN U
procedure. CT scans performed immediately after MRI also confirmed the absence of bullet migration. No patient noted any complaint of discomfort or sensation of warmth during the scan. The absence of metal artifact on T2-weighted imaging further suggests the absence of ferromagnetic properties in any retained fragments.
TE D
Published data does support that most conventional handgun ammunition is safe for patients undergoing strong magnetic fields. Nevertheless, when possible, it is prudent to obtain spent cartridges or samples of the ammunition used to determine ferromagnetic
EP
properties (more likely in accidental or self-inflicted injuries as opposed to assault).7, 14 If
AC C
ammunition samples are not available, both the surgeon and radiologist need to consider the proximity of a metallic fragment to patent vascular structures or functional neurologic structures. One study suggests that ferromagnetic detection systems would be adequate to determine if a specific retained fragment exhibited ferromagnetic properties that would preclude MRI.10 Even if the determination is made that an MRI scan is relatively low risk, in this series we counseled the patient and family extensively and obtained informed consent for potential interaction between magnetic fields and retained metal fragments.
ACCEPTED MANUSCRIPT
Rifle and shotgun ammunition has a much higher likelihood of containing ferromagnetic steel and should be considered a contraindication to undergoing MRI unless that
RI PT
ammunition is proven to be safe.
Utility of MRI in GSW
MRI offers benefits over conventional imaging modalities to justify its use including
SC
improved soft tissue visualization and reduced metal artifact. MRI has a major role in the assessment of blunt trauma to the cervical spine because of its unique ability to show the
M AN U
integrity of soft tissue structures. These include the spinal cord, spinal ligaments, cerebrospinal fluid, paraspinous musculature and cervical vasculature. This is especially important for penetrating trauma because the pathology cannot be completely predicted by bony injury and bullet trajectory alone. There are numerous potential clinical
TE D
indications where MRI is valuable; two prime examples are concussive SCI and persistent spinal cord compression from soft tissue injury. Patient 1, as described previously, had significant concussive SCI despite no penetrating trauma to the cord and
EP
no destabilizing injury. Persistent spinal cord compression, e.g., epidural hematoma, is an indication for urgent operative decompression and can only be shown by MRI if metal
AC C
streak artifact obscures the area of interest.3
Non-ferromagnetic and ferromagnetic projectile fragments, by virtue of their increased density may produce an unacceptable level of streak artifact on CT and CT angiography resulting in obscuration of adjacent osseous, vascular and soft tissue structures. Non feeromagnetic materials have no significant impact on MR images. Ferromagnetic fragments on the other hand may result in susceptibility artifacts on MRI. However, cord
ACCEPTED MANUSCRIPT
signal abnormalities are best depicted on T2 or T2W spin echo sequences, which typically suffer less degradation from such artifacts than those sequences acquired with a gradient echo technique. Figure 1 shows the difference in imaging clarity between CT
RI PT
and MRI in the setting of retained metal. Patient 17 (Table 1) is a clear example of this benefit in whom streak artifact precluded differentiation between abscess and
pseudomeningocele. Though data suggests non-ferromagnetic metal does not cause
SC
increased susceptibility artifact at higher magnetic field strengths, theconsequences of
M AN U
imaging patients at higher field strengths than 1.5T may be different from our experience.
Based on our experience with MRI in the setting of GSW to the cervical spine, we identify several clinical situations where MRI may be beneficial at guiding medical and surgical care. Though CT imaging (plain CT, angiography, myelogram) is adequate in many cases to obtain relevant diagnostic information, metal streak artifact can obscure
TE D
the area of interest in which case MRI can offer additional information. Patients with incomplete neurologic injury are of particular concern in obtaining adequate diagnostic
EP
information as these patients have higher potential for recovery of function with prompt treatment. However, even patients with complete neurologic injury merit evaluation for
AC C
compressive lesions to see if surgical decompression might offer improvement in function.
1) A patient with neurologic deficits in the setting of a bullet trajectory that is distant from the spinal canal. Concussive injury from the bullet’s energy transfer to surrounding tissues is sometimes enough to cause injury to the spinal cord. While it is unknown how best to manage these injuries medically, MRI can be useful in
ACCEPTED MANUSCRIPT
confirming the absence of a lesion requiring surgical decompression. Furthermore, MRI is uniquely suited to confirm the presence of a noncompressive spinal cord injury which we manage medically in similar fashion as a
RI PT
patient with central cord syndrome.
2) A patient with neurologic deficit in the setting of a bullet trajectory closely
SC
approximating the spinal canal but not traversing the canal. These patients are at high risk for compressive lesions with potentially salvageable neurologic
M AN U
function. MRI imaging can confirm the presence or absence of surgical lesions as well as provide prognostic information of SCI severity.
3) A patient with collections along the bullet’s tract concerning for infection or abscess. Although any collection along a penetrating injury tract is concerning for
TE D
infection, there is particular concern in the setting of fever or leukocytosis without another obvious source. The trauma population is often complicated by multiple
EP
injuries to multiple organ systems that could be responsible for fever or leukocytosis. In addition to avoiding metal streak artifact seen on CT, MRI is
AC C
more sensitive than CT at diagnosing soft tissue infections as well as diagnosing a CSF fistula.
CONCLUSION
Our experience with MRI in the setting of GSW has significantly improved the evaluation and management of our patient population. No patient in our series suffered
ACCEPTED MANUSCRIPT
an adverse event related to MRI. We were also able to obtain diagnostic information which was incompletely assessed or unavailable using CT imaging alone. For carefully selected patients with appropriate counseling on the risk and benefit of the study, MRI
AC C
EP
TE D
M AN U
SC
RI PT
can an effective tool in assessing and managing trauma related to GSW to the neck.
ACCEPTED MANUSCRIPT
REFERENCES
RI PT
Reference List 1. Aarabi B, Simard JM, Kufera JA, et al: Intramedullary lesion expansion on
magnetic resonance imaging in patients with motor complete cervical spinal cord
SC
injury. J Neurosurg Spine 17:243-250, 2012
M AN U
2. Bashir EF, Cybulski GR, Chaudhri K, et al: Magnetic resonance imaging and computed tomography in the evaluation of penetrating gunshot injury of the spine. Case report. Spine (Phila Pa 1976 ) 18:772-773, 1993
3. Beaty N, Slavin J, Diaz C, et al: Cervical spine injury from gunshot wounds. J
TE D
Neurosurg Spine 21:442-449, 2014
4. Dedini RD, Karacozoff AM, Shellock FG, et al: MRI issues for ballistic objects:
EP
information obtained at 1.5-, 3- and 7-Tesla. Spine J 13:815-822, 2013 5. Eshed I, Kushnir T, Shabshin N, et al: Is magnetic resonance imaging safe for
AC C
patients with retained metal fragments from combat and terrorist attacks? Acta Radiol 51:170-174, 2010
6. Finitsis SN, Falcone S, Green BA: MR of the spine in the presence of metallic bullet fragments: is the benefit worth the risk? AJNR Am J Neuroradiol 20:354356, 1999
ACCEPTED MANUSCRIPT
7. Hammoud MA, Haddad FS, Moufarrij NA: Spinal cord missile injuries during the Lebanese civil war. Surg Neurol 43:432-437, 1995
RI PT
8. Hess U, Harms J, Schneider A, et al: Assessment of gunshot bullet injuries with the use of magnetic resonance imaging. J Trauma 49:704-709, 2000
9. Hollerman JJ, Fackler ML, Coldwell DM, et al: Gunshot wounds: 1. Bullets,
SC
ballistics, and mechanisms of injury. AJR Am J Roentgenol 155:685-690, 1990
M AN U
10. Karacozoff AM, Pekmezci M, Shellock FG: Armor-piercing bullet: 3-T MRI findings and identification by a ferromagnetic detection system. Mil Med 178:e380-e385, 2013
11. Martinez-del-Campo E, Rangel-Castilla L, Soriano-Baron H, et al: Magnetic
TE D
resonance imaging in lumbar gunshot wounds: an absolute contraindication? Neurosurg Focus 37:E13, 2014
12. Smith AS, Hurst GC, Duerk JL, et al: MR of ballistic materials: imaging artifacts
EP
and potential hazards. AJNR Am J Neuroradiol 12:567-572, 1991
AC C
13. Smugar SS, Schweitzer ME, Hume E: MRI in patients with intraspinal bullets. J Magn Reson Imaging 9:151-153, 1999
14. Teitelbaum GP, Yee CA, Van Horn DD, et al: Metallic ballistic fragments: MR imaging safety and artifacts. Radiology 175:855-859, 1990
ACCEPTED MANUSCRIPT
Table 1 Patient demographics and relevant clinical and radiographic findings Level of SCI focus
Bony Injury
on MRI
Trajectory through
Deepest retained
Admissio
Follow up ASIA
Time to MRI
Ascending
spinal canal
fragment
n ASIA
(time of follow
(hours)
edema (cm)
RI PT
#/gender/age
up weeks)
none
C3 spinous process
No
Paraspinal Muscle
59D
none
87
2.13
R C3 lateral mass
No
none
75D
85D (1.3)
6
3/M/56
0
C7
C6-7 spinous process
No
none
30A
27A (1.6)
10
0
4/M/52
C3
L C3/4 lateral mass/lamina
No
Deep Subcutaneous
98D
100D (36.6)
25
0
5/M/31
C7
Total destruction of C6-7 vertebral
No
Vertebral Body
0A
15A (4.0)
47
9.76
Subarachnoid
14A
24A (3.1)
5, 64
4.47, 7.5
none
70D
none
3.5
0
none
8A
8A (8.0)
43
7.73
Subarachnoid
0A
0A (17.9)
31, 10.5
8.0, 8.0
bodies and L lateral masses 6/M/20
C7
7/M/22
none
8/M/42
C7
B C7-T1 lateral mass and lamina
Yes
9/M/21
C5
B C4-5 pedicle and lamina
No
C5
Yes No
B C4-5 lamina and spinous process
11/M/22
C7
C5 vertebral body and L C6-7
12/M/30
C7
C7-T1 vertebral body
13/M/30
C5
C4-5 lamina
14/M/21
C6
C6-7 vertebral body
15/M/18
C7
T1 vertebral body
16/M/33
C6
C6-7 vertebral body and B lateral
17/M/23
C5
AC C
masses
none
0A
0A (2.9)
184
5.34
No
None
3A
8A (32.9)
48
4.17
No
Vertebral Body
100D
81D (18.4)
46 days
0
Yes
R Lateral Mass
0A
0A (202)
119
9.43
Yes
Intramedullary
8A
6A (3.4)
34
5.73
Yes
Deep Fascial
6A
26A (138)
7
6.04
No
none
NT/B*
NT/B* (72)
15, 12 days
0, 1.65
No
Paraspinal
0A
0A (7.1)
9, (11, 27
6.86, 7.5,
days)
3.58
EP
pedicle and lamina
R C4-5 lamina and lateral mass
*NT – non testable ASIA exam because of concomitant traumatic head injur
days
No
TE D
10/M/44
R C7 lateral mass and L C7 lamina C4 spinous process
SC
C3
2/M/19
M AN U
1/M/22
Musculature
ACCEPTED MANUSCRIPT
LEGENDS:
Figure 1 (A) and (B) Axial and sagittal CT scans without contrast show ballistic
RI PT
fragments in a trajectory through the spinous process and paraspinous musculature
without impingement of the spinal canal. (C) Sagittal T2 MRI sequence demonstrates cord signal hyperintensity at the level of bony injury.
SC
Figure 2 (A) Axial CT angiogram shows retained bullet fragments in the spinal canal and paraspinous musculature. Severe metal streak artifact limits visualization of the contents
M AN U
of the spinal canal and evaluation of bony injury. (B) Axial T2 MRI reveals minimal metal artifact from the retained fragments and confirms spinal cord injury. (C) Sagittal CT angiogram shows bullet fragments with similar limitations from streak artifact regarding the contents of the spinal canal. (D) Sagittal T2 MRI shows significant spinal
TE D
cord injury with cord contusion and ascending edema.
Figure 3 (A) Axial CT angiogram from admission shows fractured posterior elements and retained metal fragments. (B) Axial CT with contrast to evaluate for suspected
EP
abscess in the setting of fever and leukocytosis. New fluid collection is seen in paraspinous tissues in the bullet trajectory concerning for abscess vs. pseudomeningocele.
AC C
(C) and (D) Sagittal and axial T2 MRI images shows collection to be in direct communication with subarachnoid space indicative of pseudomeningocele.
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT
Highlights
•
AC C
EP
TE D
M AN U
SC
•
MRI offers improved visualization over CT scan of soft tissue structures of the spine after GSW including the spinal cord, ligamentous injury, and epidural collections with reduced metal streak artifact. Using 1.5T magnets, no patient in our series suffered injury as a result of undergoing MRI of the cervical spine. Frank discussions regarding the risk of potential complications weighed against the potential benefits of the study should be part of a detailed informed consent process.
RI PT
•