Abstract
Background: Whether or not to remove bullets
or bullet fragments from the spinal column of a neurologically intact
patient has been a subject of continual debate. The controversy
is due in part to a lack of information about the long-term effects
of bullet fragments on spinal cord tissue. Although many studies
have demonstrated the toxic effects of metal fragments on brain
tissue, to our knowledge no one has evaluated the effects of the
metals contained in commercially available bullets on spinal cord
tissue.
Methods: Copper, aluminum, and lead fragments from
three commercially available bullet cartridges were implanted in
intradural and extradural locations in seventeen New Zealand White
rabbits. At an average of 9.8 months, the metal content of specimens of
blood, cerebrospinal fluid, and liver were determined. The spinal
cords were harvested and examined histologically.
Results: There was a significant increase in the
copper level of blood from the rabbits with an implanted copper fragment
compared with that of the control animals (p = 0.007).
Concentrations of copper and lead were not elevated, compared with
the control values, in the serum or liver. Histological examination of
the spinal cords revealed major destruction of both the axons and
the myelin of the dorsal column adjacent to the intradural copper
fragments. Intradural fragments of lead caused similar destruction of
myelin and axons in the dorsal column, but to a lesser degree. Minimal
spinal cord or meningeal histological changes were noted around
the aluminum intradural fragments, and no pathological changes were
found near any fragments placed in an extradural location.
Conclusions: The results of this study show that
certain metals contained in commercially available bullets can cause
varying degrees of neural destruction independent of the initial
mechanical injury caused by implantation. Of the three metals tested,
copper fragments consistently caused a substantial localized area
of neural injury within the spinal cord.
Clinical Relevance: In our study, copper fragments
caused local neural toxicity involving as much as 10% of
the spinal cord area, suggesting that there may be a scientific basis
for removal of copper fragments lodged in the spinal cord, even
in the absence of a neurological deficit.
The decision to leave or remove a bullet fragment lodged
in the spinal canal is dependent on many factors. Absolute indications
include neurological deterioration, infection, and lead toxicity.
However, in the absence of these absolute indications, whether to
remove bullets from the spine has been a subject of continual controversy.
On the basis of wartime observations, some surgeons have advocated
the removal of all bullets and metal fragments from the spine1-9. In contrast, many recent studies
have supported a more conservative approach10-18.
Clearly, with regard to patients who exhibit no progressive neurological
deficit, infection, or systemic toxicity, an understanding of the
possible long-term effects of metal fragments on spinal cord tissue
would give the clinician a rationale on which to base an optimal
treatment plan for each patient.
The results of many studies show severe local and systemic toxicity
from metallic fragments implanted in brain tissue19-28.
However, we are not aware of any histological studies that evaluate
the effect of the metals contained in commercially available bullets,
such as copper, lead, and aluminum, when they are embedded in or
near spinal cord tissue1-18.
We evaluated the gross and histological changes caused by copper,
lead, and aluminum fragments placed in intradural and extradural
sites in neurologically intact rabbits.
Eighteen New Zealand White rabbits (approximately six months
of age), weighing 2.2 kg to 3.7 kg, underwent implantation of metallic
fragments (except for one who was given a sham operation). The animals
were killed at an average of 9.8 months (range, 6.9 to 11.5 months).
A rabbit model was chosen because the spinal canals were of sufficient
size to accommodate accurate intradural and extradural implantation
of the metallic fragments.
Bullet Fragments
We used three commercially available bullet cartridges supplied
by a local gun shop: the copper jacket from a lead-core bullet,
an uncovered lead-core bullet, and the aluminum jacket from a lead-core
bullet. We cut pieces of relatively uniform size from the bullet
cores with a utility knife and sterilized them before implantation.
The copper-jacket fragments had an average mass of 0.04 g (range, 0.03
to 0.05 g), and their copper content was nearly 99.9% pure29. The average mass of the lead-core
fragments was 0.035 g (range, 0.03 to 0.04 g). The chemical composition
of the lead alloy in the fragments was approximately 92.4% lead,
6.2% antimony, 0.45% tin, and trace impurities29. The aluminum-jacket fragments had
an average mass of 0.006 g (range, 0.005 to 0.007 g) and were composed
of 98.5% aluminum, 1.2% manganese, and 0.12% copper,
with other trace impurities30.
Differences in density between the alloys account for the variation
in weight of the fragments described above. The bullet fragments
fit easily within the rabbit spinal canal without causing compression
on the spinal cord.
Surgical Technique
The committee on animal research at our institution approved
all of the surgical procedures. Eighteen rabbits underwent sterile
posterior surgical exposure of the upper lumbar spine. This level
was chosen because it contains spinal cord and has the largest spinal
canal diameter; thus, it could accommodate a foreign body with the
least chance of causing spinal cord compression. Seventeen rabbits had
a metal fragment implanted at one or more levels. Surgical exposure
was performed in one rabbit, but no metal fragment was implanted.
Nine rabbits had one or more fragments implanted extradurally, and
eight rabbits had a fragment implanted within the dura of the spinal
cord. The number of animals in each experimental group is shown
in Table I.
The rabbits were anesthetized with intravenous ketamine (43 mg/mL),
xylazine (8.6 mg/mL), and acepromazine (1.4 mg/mL).
Cefazolin (30 mg/kg) was given intramuscularly prior to
the procedure and on the first and second postoperative days.
A standard posterior approach to the lumbar spine was performed
with a laminectomy over the area of fragment implantation. Fragments
implanted in an extradural location were placed in situ.
For fragments intended for an intradural location, the dura was
incised longitudinally and the fragment was placed within the arachnoid.
Because the dural edges could not be apposed over the fragment without
severely compressing it against the underlying spinal cord, no attempt
was made to repair the dural defect.
Postoperatively, none of the animals showed signs of neurological
compromise. One rabbit died from unknown causes in the early perioperative
period and was not included in this study.
Radiographic Evaluation
Anteroposterior and lateral radiographs were made one week after
implantation of the fragments and again just before the animals
were killed.
Fluid and Biopsy Acquisition and Analysis
A liver biopsy sample and an intracardiac blood sample were obtained
immediately before the rabbits were killed. With the animals under
anesthesia, cerebrospinal fluid was obtained with a 22-gauge spinal
needle through a posterior occiput-first cervical approach. Three
rabbits had respiratory arrest after administration of
the test dose of anesthetic, and this complication precluded the
acquisition of cerebrospinal fluid from these animals. All blood, cerebrospinal
fluid, and liver biopsy samples were analyzed with a PerkinElmer
model-3030 atomic absorption spectrometer (Norwalk, Connecticut) with
deuterium arc background correction to determine lead levels and
with an HGA-400 graphite furnace (PerkinElmer) to determine copper
levels.
Perfusion
Each rabbit was perfused with two liters of 4% neutral
buffered formaldehyde. The entire spinal column was then excised en
bloc and fixed in 4% neutral buffered formaldehyde solution
for two weeks.
Histological Examination
The spinal cords were examined grossly and histologically. Histological
specimens were obtained and stained for general histological examination (hematoxylin
and eosin), for myelin (solochrome-cyanine), for axons (Luxol fast
blue and Sevier-Munger), and for connective-tissue components (trichrome).
Representative sections from the eight spinal cords that had
an intradurally placed metal fragment were examined. The areas of
gliosis caused by the three types of metal fragments were measured
and expressed as a percentage of the entire cross-sectional area
of the remaining spinal cord. The histomorphometric measurements
were performed by two blinded observers using an Axioplan 2/Axiophot
2 Universal microscope (Carl Zeiss, Thornwood, New York) and MetaMorph
imaging software (version 3.6; Universal Imaging, West Chester,
Pennsylvania). The three groups were then compared with use of a
two-tailed Student t test.
Statistical Methods
Blood, cerebrospinal fluid, and liver samples were analyzed for
lead and copper content at the time of spinal cord harvest. Although
aluminum was a large component of the Silvertip alloy of the jacket from
the lead-core bullet, it was not assayed in this study due to financial
constraints.
Statistical analysis was carried out to compare samples from
the rabbits in which a particular type of metal fragment (copper
or lead) had been implanted with those from the rabbits that did
not have that particular type of fragment. In this fashion, animals that
did not have a particular type of implanted metal fragment (copper
or lead) served as a control for the animals that did. No attempt
was made to compare the group that had a fragment implanted intradurally
with the group that had one or more fragments implanted extradurally
because the number of animals in each group was too small for statistically
valid comparisons.
The level of significance of the difference in the metallic content
of the blood, cerebrospinal fluid, and liver of animals with and
without a particular fragment was determined with use of a two-sample Student
t test.
Seventeen rabbits underwent surgical implantation of one or more
metal fragments, and one rabbit underwent a sham operation. The
animals had no clinical signs of spinal cord injury or heavy-metal toxicity
for the duration of the experiment.
Blood, Cerebrospinal Fluid, and Liver Analysis
The average serum, cerebrospinal fluid, and liver metal concentrations
are shown in Table II.
Only the blood of the rabbits with a copper fragment had a significant
(although small) increase in the level of copper as compared with
that of the controls (p = 0.007); these animals did not
have a significant increase in the level of copper in the cerebrospinal
fluid or liver (p = 0.096). No significant increases in
the lead level were detected in the blood, cerebrospinal fluid,
or liver of animals in which a lead fragment had been implanted.
Radiographic Results
Radiographs made one week after fragment implantation and again
just before the animals were killed confirmed that the copper and
lead fragments had not migrated from their original positions over the
course of the experiment. The aluminum fragments were radiolucent
and could not be identified.
Visual Inspection
The fragments implanted in an extradural location provoked a
minor reaction in the dura, spinal cord, or surrounding tissues.
A small area of granulation tissue was noted around the fragments,
but this was not pronounced. A small area of greenish deposit was
noted in the tissues surrounding the copper fragments, minor gray-black
pigmentation of the tissue was seen around the lead fragments, and
no color changes were noted around the aluminum fragments.
There were obvious gross alterations in the morphological character
of the spinal cord surrounding the fragments implanted in an intradural
location. A well-demarcated area of spinal cord tissue that was stained
dark green and brown and that extended circumferentially for several
millimeters surrounded the copper fragments. There were similar
changes surrounding the lead intradural fragments, but to a lesser
degree. The aluminum fragments did not stain the surrounding spinal
cord tissue.
Histological Results
Extradural Fragments
The three different types of extradural fragments (lead, copper,
and aluminum) were associated with similar histological changes
in the spinal cord; therefore, the changes will be described together. No
pathological findings were seen within the parenchyma of the spinal
cord near the extradural fragments (Fig. 1, a). Some spinal
cords appeared to have a slightly indented dorsal column at the
level of the fragment. However, the dura was noted to be intact,
and there was no pia-arachnoid thickening. No pathological changes
were noted within the substance of the dorsal column adjacent to
the indentation. The axons appeared normal in both quantity and
spacing. There was no loss of myelin sheath surrounding these axons,
and the astrocytes appeared normal in size. There were no inflammatory
cells seen near the area of indentation.
Intradural Copper Fragments
The spinal cords with a copper intradural fragment revealed a
number of striking histological changes in the dorsal column and
the gray matter (Fig. 1). The trichrome stain revealed
this area to be composed largely of connective tissue suggestive
of excessive fibrosis in the pia-arachnoid layer (Fig. 1, e).
Within the parenchyma of the spinal cord, there was a loss of
the normal anteroposterior depth of the dorsal column. In the dorsal-column
tissue that remained, there appeared to be a gradient of tissue damage
at the site where the copper fragment was located. The most affected
area was that closest to the fragment, which showed tissue necrosis
and a loss of myelin and axons. Occasional macrophages were seen
in this area. Moving in a ventral direction, further into the dorsal
column, an increasing number of axons were present, but there was
a substantial loss of myelin. Only a few inflammatory cells were
seen in this region. A marginal increase in vascularity was also
noted laterally in this area. In the most ventral part of the dorsal
column, there was a shallow area of preservation of both axons and
myelin. The underlying gray matter appeared well preserved.
Histological cross sections were made 1.0 cm cephalad and 1.0
cm caudad to the metal fragment. In these sections, the architecture
of the dorsal column and the gray matter appeared normal. The number and
size of the axons also appeared normal. This finding suggests a
perimeter of damage extending centrifugally £1.0 cm from
the fragment. Cross sections cut >1.0 cm from the fragment,
both caudally and cephalad, also revealed no pathological findings.
Intradural Lead Fragments
The notable histological findings associated with lead fragments
consisted of thickening in the pia-arachnoid region (Fig. 2). The increase
in connective tissue is consistent with fibrosis adjacent to the
metal fragment. The fragment and a large portion of this layer were peeled
off at the time of fragment excision. We believe that this fibrotic
layer was thicker than shown. Some vacuolar changes were noted in
the lateral columns of the white matter; these changes were unilateral
and not contiguous with the site of injury. They may be indicative
of a compressive event, perhaps at the time of surgery.
Intradural Aluminum Fragments
Spinal cords with an intradural aluminum fragment showed none
of the findings seen in those with an intradural copper or lead
fragment (Fig. 3).
Despite substantial compression from one of these fragments, there
was no evidence of pia-arachnoid thickening, no axon or myelin loss,
and a normal overall anteroposterior dimension of the dorsal column.
Determination of Average Lesion Size
Each of the three metals caused different degrees of local neural
destruction within the remaining spinal cord tissue. Lesions caused
by the copper fragments occupied, on the average, 8.96% ± 1.2% of the remaining spinal cord tissue
(as determined in three animals). In contrast, both aluminum and
lead fragments created significantly smaller zones of gliosis (p = 0.0003
and 0.003, respectively; two-tailed Student t test) occupying
0.027% 0.047% (as determined in three rabbits)
and 0.377% 0.533% (as determined in two rabbits), respectively,
of the remaining spinal cord tissue. The difference in the areas
of gliosis between the lead and aluminum groups was not significant
(p = 0.3).
The results of this study show that reactions to metal fragments
from commercially available bullet cartridges in the rabbit spinal
cord can differ profoundly, depending on both the type and the location
of the metal fragment. Bullet fragments located in extradural locations
did not appear to have an effect on spinal cord tissue, which was
apparently protected by the meninges. Intradural implantation elicited
a more robust local reaction. Both copper and lead alloys elicited
a fibrotic reaction in the pia-arachnoid layer. Copper was seen
to destroy axons and myelin, creating a significant area of gliosis within
the remaining spinal cord tissue. Lead caused a similar but less
severe local response. Intradural aluminum alloy fragments were
relatively inert.
Our findings are consistent with those of similar investigations
of the effects of metal foreign bodies on brain tissue. Sights and
Bye28 described the histopathological
reaction of brain tissue to surgically implanted shotgun pellets
in cats. Severe histological reactions, neurological deficits, and
migration of the copper-coated pellets were seen, whereas lead shot
produced a moderate reaction and nickel-coated shot produced a minor reaction.
On the basis of findings from corrosion studies, McFadden24 showed that copper was not an acceptable
metal for implantation within brain tissue. Chusid and Kopeloff20 placed copper powder on the cortex
of monkeys and noted a severe foreign-body reaction that resulted
in death of the animal. Copper appears to exert a toxic effect,
and other authors have recognized the toxicity of copper in brain
tissue19,21-23,25.
Lead is a toxic metal that is not essential for nutrition31. It has been shown to compete with
calcium, inhibiting the release of neurotransmitters, and it also interferes
with the regulation of cell metabolism32-34.
Local and systemic toxicity may occur when lead leaches out of a
bullet located elsewhere in the human body, such as an area bathed
in synovial fluid and the intervertebral disc35-41.
We did not detect blood, cerebrospinal fluid, or liver absorption
of lead in our rabbits.
To our knowledge, no report of systemic copper or aluminum toxicity
from bullets has been published. In our study, the rabbits with
an implanted copper fragment had significantly higher serum copper
levels than did controls. Although, to our knowledge, this finding
has not been reported in human subjects, it may prove useful in
distinguishing copper bullets from the other types of bullets in
use. We were unable to demonstrate a significant difference in levels
of copper in cerebrospinal fluid between our study group and the
control animals. This puzzling result could be due to the fact that
an insufficient number of animals were evaluated for copper content
in the cerebrospinal fluid.
Bullet projectiles vary in terms of weight, velocity, alloy composition,
and presence of a jacket or plated covering. We used fragments from
three different commercially available bullets, reflecting only
a small sample of the different types of metals used in bullet projectiles
manufactured in this country. Our study was not intended to evaluate
the injury caused by the impact of a bullet to the spinal cord,
nor was it designed to compare the effects of jacketed bullets with
those of unjacketed bullets.
The results of our study show that the rabbit spinal cord can
have profoundly different reactions to metal fragments from commercially
available bullet cartridges, depending on both their type and their location.
Extradural placement did not result in any sign of local neurotoxicity.
Intradural copper bullet fragments resulted in an intense area of
local neural destruction. Intradural lead fragments were associated
with similar findings but caused a smaller zone of injury. Implantation
of intradural aluminum fragments did not lead to any local neural
destruction.
Haynes WG: Acute war wounds of the spinal cord. . Am J Surg,1946.72: 424-33, 72424
1946
[PubMed]
Jacobs GB, and Berg RA: Treatment of acute spinal cord injuries in a war zone. . J Neurosurg,1971.34: 164-7, 34164
1971
[PubMed]
McCravey A: War wounds of the spinal cord. A plea for exploration
of spinal cord and cauda equina injuries [abstract]. . JAMA,1945.129: 152-3, 129152
1945
Meirowsky AM: Penetrating wounds of the spinal cord. Problems of paraplegia
and notes on autonomic hyperreflexia and sympathetic blockade. . Clin Orthop,1963.27: 90-107, 2790
1963
[PubMed]
Nino HE; Leppik IE; Lai C; and Martin S: Progressive sensory loss one year after bullet injury of
spinal cord. . JAMA,1978.240: 1173-4, 2401173
1978
[PubMed]
Pool JL: Gunshot wounds of the spineObservations from an evacuation
hospital. . Surg Gynec Obstet,1945.81: 617-22, 81617
1945
[PubMed]
Schneider RC, and Kahn EA: Chronic neurological sequelae of acute trauma to the spine
and spinal cord. Part IV. Problems pertaining to the treatment of
retained foreign bodies. . J Bone Joint Surg Am,1959.41: 457-62, 41457
1959
[PubMed]
Tinsley M: Compound injuries of the spinal cord. . J Neurosurg,1946.3: 306-9, 3306
1946
[PubMed]
Wannamaker GT: Spinal cord injuries. A review of the early treatment in
300 consecutive cases during the Korean conflict. . J Neurosurg,1954.11: 517-24, 11517
1954
[PubMed]
Benzel EC; Hadden TA; and Coleman JE: Civilian gunshot wounds to the spinal cord and cauda equina. . J Neurosurg,1987.20: 281-5, 20281
1987
Heiden JS; Weiss MH; Rosenberg AW; Kurze T; and Apuzzo ML: Penetrating gunshot wounds of the cervical spine in civilians.
Review of 38 cases. . J Neurosurg,1975.42: 575-9, 42575
1975
[PubMed]
Messer HD: Foreign objects in the spinal canal [letter]. . JAMA,1979.241: 2701, 2412701
1979
[PubMed]
Richards JS; Stover SL; and Jaworski T: Effect of bullet removal on subsequent pain in persons
with spinal cord injury secondary to gunshot wound. . J Neurosurg,1990.73: 401-4, 73401
1990
[PubMed]
Roffi RP; Waters RL; and Adkins RH: Gunshot wounds to the spine associated with a perforated
viscus. . Spine,1989.14: 808-11, 14808
1989
[PubMed]
Stauffer ES; Wood RW; and Kelly EG: Gunshot wounds of the spine: the effects of laminectomy. . J Bone Joint Surg Am,1979.61: 389-92, 61389
1979
[PubMed]
Waters RL: Gunshot wounds to the spine: the effects of bullet fragments
in the spinal canal. . J Am Paraplegia Soc,1984.7: 30-3, 730
1984
[PubMed]
Waters RL, and Adkins RH: The effects of removal of bullet fragments retained in
the spinal canal. A collaborative study by the National Spinal Cord
Injury Model Systems. . Spine,1991.16: 934-9, 16934
1991
[PubMed]
Yashon D; Jane JA; and White RJ: Prognosis and management of spinal cord and cauda equina
bullet injuries in sixty-five civilians. . J Neurosurg,1970.32: 163-70, 32163
1970
[PubMed]
Chusid JG, and Kopeloff LM: Epileptogenic effects of metal powder implants in motor
cortex of monkeys. . Int J Neuropsychiat,1967.3: 24-8, 324
1967
Chusid JG, and Kopeloff LM: Epileptogenic effects of pure metals implanted in motor
cortex of monkeys. . J Appl Physiol,1962.17: 697-700, 17697
1962
[PubMed]
Collias JC, and Manuelidis EE: Histopathological changes produced by implanted electrodes
in cat brains. Comparison with histopathological changes in human
and experimental puncture wounds. . J Neurosurg,1957.14: 302-28, 14302
1957
[PubMed]
Dymond AM; Kaechele LE; Jurist JM; and Crandall PH: Brain tissue reaction to some chronically implanted metals. . J Neurosurg,1970.33: 574-80, 33574
1970
[PubMed]
Fischer G; Sayre GP; and Bickford RG: Histologic changes in the cat’s brain after introduction
of metallic and plastic coated wire used in electroencephalography. . Proc Mayo Clin,1957.32: 14-21, 3214
1957
McFadden JT: Metallurgical principles in neurosurgery. . J Neurosurg,1969.31: 373-85, 31373
1969
[PubMed]
Messer HD, and Cerza PF: Copper jacketed bullets in the central nervous system. . Neuroradiology,1976.12: 121-9, 12121
1976
[PubMed]
Ott K; Tarlov E; Crowell R; and Papadakis N: Retained intracranial metallic foreign bodies. Report
of two cases. . J Neurosurg,1976.44: 80-3, 4480
1976
[PubMed]
Sherman IJ: Brass foreign body in the brain stem. A case report. . J Neurosurg,1960.17: 483-5, 17483
1960
[PubMed]
Sights WP, and Bye RJ: The fate of retained intracerebral shotgun pellets. An
experimental study. . J Neurosurg,1970.33: 646-53, 33646
1970
[PubMed]
Federal Cartridge Corporation,
Anoka, MN: Personal communication, 2000.
Winchester/Olin Corporation,
East Alton, IL: Personal communication, 2000
Goyer RA: Nutrition and metal toxicity. . Am J Clin Nutr,1995.61: 646-50, 61646
1995
Kostial K, and Vouk VB: Lead ions and synaptic transmission in the superior cervical
ganglion of the cat. . Br J Pharmacol,1957.12: 219-22, 12219
1957
Simons TJ: Cellular interactions between lead and calcium. . Br Med Bull,1986.42: 431-4, 42431
1986
[PubMed]
Spence I; Drew C; Johnston GA; and Lodge D: Acute effects of lead at central synapses in vitro. . Brain Res,1985.333: 103-9, 333103
1985
[PubMed]
Bolanos AA; Demizio JP; Vigorita VJ; and Bryk E: Lead poisoning from an intra-articular shotgun
pellet in the knee treated with arthroscopic extraction and chelation
therapy. A case report. . J Bone Joint Surg Am,1996.78: 422-6, 78422
1996
[PubMed]
Cagin CR; Diloy-Puray M; and Westerman MP: Bullets, lead poisoning and thyrotoxicosis. . Ann Intern Med,1978.89: 509-11, 89509
1978
[PubMed]
Grogan DP, and Bucholz RW: Acute lead intoxication from a bullet in an intervertebral
disc space. A case report. . J Bone Joint Surg Am,1981.63: 1180-2, 631180
1981
[PubMed]
Leonard MH: The solution of lead by synovial fluid. . Clin Orthop,1969.64: 255-61, 64255
1969
[PubMed]
Primm DD Jr: Lead arthropathyæprogressive destruction of a
joint by a retained bullet. Case report. . J Bone Joint Surg Am,1984.66: 292-4, 66292
1984
[PubMed]
Switz DM; Elmorshidy ME; and Deyerle WM: Bullets, joints, and lead intoxication. A remarkable and
instructive case. . Arch Intern Med,1976.136: 939-41, 136939
1976
[PubMed]
Windler EC; Smith RB; Bryan WJ; and Woods GW: Lead intoxication and traumatic arthritis of the hip secondary
to retained bullet fragments. A case report. . J Bone Joint Surg Am,1978.60: 254-5, 60254
1978
[PubMed]