PEDIATRIC/CRANIOFACIAL

Survival following Self-Inflicted Gunshots

to the Face
Adekunle Elegbede, M.D.,
Ph.D.
Philip J. Wasicek, M.D.
Sara Mermulla, M.D.
Ryan Dunlow, B.S.
Yvonne M. Rasko, M.D.
Bizhan Aarabi, M.D.
Fan Liang, M.D.
Michael P. Grant, M.D.,
Downloaded from http://journals.lww.com/plasreconsurg by BhDMf5ePHKbH4TTImqenVIu1/FN/J4kOf/a3O+pRQdv7el34Wr1+VP9chjT7R2KF1rUn4n4Fmko= on 08/10/2019
Background: Self-inflicted gunshot wounds involving the face are highly mor-
bid. However, there is a paucity of objective estimates of mortality. This study
aims to provide prognostic guidance to clinicians that encounter this uncom-
mon injury.
Methods: A retrospective review of patients presenting to R Adams Cowley
Shock Trauma Center (a Level I trauma center) with self-inflicted gunshot
wounds to the face from 2007 to 2016. Isolated gunshot wounds to the calvaria
or neck were excluded. The data were analyzed to determine predictors of
survival.

Ph.D. Results: Of the 69 patients that met inclusion criteria, 90 percent were male
Arthur J. Nam, M.D., M.S. and 80 percent were Caucasian, with an age range of 21 to 85 years. The most
Baltimore, Md. frequently seen injury patterns showed submental (57 percent), intraoral (22
percent), and temporal (12 percent) entry sites. Fewer than half (41 percent)
of the cohort sustained penetrative brain injury. Overall, there were 18 deaths
(overall mortality, 26 percent), 17 of which were secondary to brain injury.
Independent predictors of death included penetrative brain injury (OR, 17;
p < 0.0001) and age. Mortality was 17 percent among patients younger than 65
years, compared with 73 percent for those aged 65 years or older (p = 0.0001).
Gastrostomy placement was independently associated with 25 percent reduc-
tion in length of hospitalization (p = 0.0003).
Conclusions: Despite tremendous morbidity, the overwhelming majority of
patients who present with facial self-inflicted gunshot wounds will survive, es-
pecially if they are young and have no penetrative brain injury. These findings
should help guide clinical decisions for this devastating injury.  (Plast. Reconstr.
Surg. 144: 415, 2019.)
CLINICAL QUESTION/LEVEL OF EVIDENCE: Risk, III.

T
he incidence of deaths from suicide in the
United States is increasing, with firearms
being the most common mechanism.1 Nev-
ertheless, self-inflicted gunshots are still quite rare
(8.3 per 100,000 annually).2 When compared with
gunshot wounds sustained from assault or acci-
dental discharge, self-inflicted gunshot wounds
represent a far more homogeneous cohort,
with predictable patterns of injury. In general,
From the Department of Plastic and Reconstructive Surgery,
The Johns Hopkins Hospital; the Department of Surgery, the
Division of Plastic and Reconstructive Surgery, and the De-
partment of Neurosurgery, University of Maryland School
of Medicine; and the Division of Plastic, Reconstructive,
and Maxillofacial Surgery, R Adams Cowley Shock Trauma
Center.
Received for publication July 17, 2018; accepted January
10, 2019.
Copyright © 2019 by the American Society of Plastic Surgeons
DOI: 10.1097/PRS.0000000000005842
self-inflicted gunshots are closer in range, have
more consistent trajectories,3–5 and are more
lethal.6–8 The self-inflicted gunshot wound popu-
lation presents with a high incidence of ballistic
injury to the brain; a constellation of complex
craniofacial fractures; significant disfigurement;
and functional limitations to the eyes, nose, and
mouth. These severe morbidities, poor odds of
neurologic recovery, and the potential for reat-
tempt at suicide can influence the management
teams toward a pessimistic outlook of the overall
outcome.9 In contrast, recidivism rates in survivors
are low,3,6 and early surgical intervention includ-
ing débridement and rigid skeletal fixation allows
for improved aesthetic and functional results,
Disclosure: The authors have no financial
­disclosures or other conflict of interest to disclose. No
funding was received for this article.

www.PRSJournal.com 415
Copyright © 2019 American Society of Plastic Surgeons. Unauthorized reproduction of this article is prohibited.
 Plastic and Reconstructive Surgery • August 2019
potentially mitigating the associated psychologi-
cal morbidity.10–12
Several studies have examined firearm inju-
ries to the craniofacial skeleton, including self-
inflicted gunshots wounds.7,10,13 However, there
is limited literature that is focused on injury and
mortality from self-inflicted gunshot wounds
to the face. Rather, facial self-inflicted gunshot
wounds are frequently included in studies of
gunshot wounds to the head, even though the
calvarial gunshot wounds differ with respect to
mortality and reconstructive needs. At our center,
we manage a relatively high number of patients
with self-inflicted gunshot wounds involving the
face. We reviewed our cohort with the objective of
providing prognostic and management guidance
for clinicians that treat this uncommon injury. In
this article, we describe the mortality rates, pre-
dictors of survival, and key aspects of the multi-
disciplinary management for this injury. To our
knowledge, this represents the largest single-insti-
tution longitudinal study of facial self-inflicted
gunshots wounds. We hope that our experience
will provide important insights and clinical guide-
lines for treatment teams that encounter these
patients.
PATIENTS AND METHODS
This study was conducted at the R Adams
Cowley Shock Trauma Center of the University
of Maryland Medical Center. After institutional
review board approval, patients with self-inflicted
gunshot wounds were identified through the hos-
pital’s trauma registry. Demographics and hos-
pital course data were collected retrospectively
on all trauma patients aged 18 years and older
admitted between January of 2007 and Decem-
ber of 2016 with firearm injury involving the face.
We defined the face anatomically as the region
bounded by the external auditory meatus later-
ally, the hairline superiorly, and the chin inferi-
orly.7,14,15 Patients that sustained isolated calvarial
injuries were excluded because their surgical and
rehabilitative requirements and outcomes are dif-
ferent.7 We also excluded patients for whom the
manner of injury was not self-harm or suicide.
To estimate prehospital mortality, we queried the
trauma registry for individuals that died en route
to the hospital and those that were pronounced
dead before admission.
Abstracted patient variables and data
included age, sex, mechanism of injury, bony
and soft-tissue injury patterns, brain involve-
ment, primary and secondary management,
416
Copyright © 2019 American Society of Plastic Surgeons. Unauthorized reproduction of this article is prohibited.
discharge disposition, tracheostomy, complica-
tions, and mortality. All admission craniomax-
illofacial computed tomographic scans were
reviewed by craniofacial surgeons and corrobo-
rated with radiology reports to determine the
bullet trajectory and involved structures. In-hos-
pital mortality was defined as the primary end-
point. Secondary endpoints included hospital
length of stay and death after initial hospitaliza-
tion. The duration of follow-up was calculated
as the interval between the admission date and
the date of the most recent recorded clinical
encounter at our institution.
Study data were collected and managed
using REDCap electronic data capture tools
hosted at the University of Maryland, Balti-
more.16 Clinical variables were summarized
using counts and percentages for categorical
variables. For continuous variables, the mean
± SD or the median and interquartile ranges
were used. Comparisons were performed using
the t test or the Wilcoxon rank sum test for con-
tinuous variables, and the Wald chi-square or
Fisher’s exact test for categorical variables as
appropriate. All tests were two-sided. A value of
p < 0.05 was taken to be suggestive of statistical
significance. Binary logistic regression was used
to identify covariates associated with mortality.
Variables identified to be statistically significant
were analyzed for colinearity. Colinear variables
were excluded from the multivariable regres-
sion model. The receiver operating character-
istic curve was constructed and the area under
the curve was calculated to examine the per-
formance of the model. Data analysis was com-
pleted using statistical software (JMP 12.2l; SAS
Institute, Inc., Cary, N.C.).
RESULTS
Cohort Inclusion/Exclusion
Patients presenting with self-inflicted gunshot
wounds to the face were not commonly encoun-
tered: of 63,861 trauma admissions during the
10-year study period, 417 had sustained firearm
injuries involving the head, face, and/ or neck.
The majority (78 percent) of these patients’ inju-
ries were sustained from assaults and interper-
sonal violence. Only 69 patients (0.1 percent) met
our inclusion criteria. Figure 1 details results from
the trauma registry query. Again, patients with
isolated injury to the calvaria and/or neck were
excluded, as were those whose trauma was not
caused by self-inflicted gunshot.
 Volume 144, Number 2 • Self-Inflicted Facial Gunshots
Fig. 1. Cohort inclusion and exclusion flow diagram. SIGSW, self-inflicted gunshot
wound; EMS, emergency medical services.
Prehospital Mortality
Over the 10-year study period, 39 individu-
als were transported to our center following
facial self-inflicted gunshot wounds but were pro-
nounced dead before admission. Although these
39 individuals did not meet our study inclusion
criteria, they are described briefly for context.
Thiry-eight of the 39 individuals (97 percent)
sustained traumatic brain injury; 19 (49 percent)
were pronounced dead before arrival, and 20 (51
percent) received cardiopulmonary resuscitation/
Advanced Trauma Life Support after arrival, but
were pronounced dead before admission. Given
that 69 patients were admitted alive, we estimate
that 38 percent of individuals that are transported
to our center following self-inflicted gunshot
wounds are pronounced dead before admis-
sion. It should be noted that patients who were
pronounced dead at the scene and those whose
remains were not discovered are not included in
this estimate.
Copyright © 2019 American Society of Plastic Surgeons. Unauthorized reproduction of this article is prohibited.
Demographics
Victims were predominantly male (90 per-
cent) and Caucasian (80 percent), with an age
range from 21 to 85 years (Table 1). Handguns
were the weapon used in the majority of cases (72
percent). Thirty-five percent and 51 percent of
patients had documented histories of psychiatric
and substance abuse disorders, respectively. Half
of the cohort tested positive for alcohol at arrival
(49 percent), and two-thirds (62 percent) tested
positive for at least one controlled substance.
Injury Characteristics
As shown in Table 1, the majority (59 per-
cent) of patients presented with a Glasgow Coma
Scale score of 8, although it is unclear how many
of these patients received analgesia or sedatives
before arrival. The Injury Severity Score ranged
from 4 to 43, reflecting a heterogeneous popula-
tion with regard to trauma severity. Most patients
did not have neck involvement (84 percent with
417
 Plastic and Reconstructive Surgery • August 2019

Table 1.  Patient and Injury Characteristics an Abbreviated Injury Scale score of the neck of
0). One patient was shot more than once, with the
Characteristic Value (%)
second shot being delivered to the chest. Thirty-
Age, yr nine patients (57 percent) were shot through a
 Median 44
 IQR 30–57 submental entry site with cephalad orientation
Sex (Fig. 2); 15 patients (22 percent) had shots deliv-
 Male 62 (90) ered intraorally and eight patients (12 percent)
 Female 7 (10)
Race were shot in the temple.
 White 55 (80) The Abbreviated Injury Scale score of the
 Black 11 (16) brain ranged from 0 to 5, with 64 percent hav-
 Other 3 (4)
Weapon type ing severe brain injury, defined as an Abbrevi-
 Handgun 50 (72) ated Injury Scale score of greater than or equal
 Shotgun 13 (19) to 3. The dura was violated in 32 patients, four of
 Rifle 4 (6)
 Unknown 2 (3) whom had no other evidence of significant brain
Injury descriptors injury. Penetrative brain injury was characterized
 GCS score > 8 28 (41) by frontal (n = 25), parietal (n = 8), and temporal
 RTS
  Median 5.97 lobe (n = 5) penetration by the projectile(s). Four
  IQR 4.09–7.84 patients sustained through-and-through penetra-
 ISS tive brain injury with both entry and exit wounds,
  Median 20
  IQR 14–26 compared with 24 patients that had retained pro-
 AIS score jectiles (no exit wound). Four patients in whom
  Face there was no evidence of penetrative brain injury
   Median 3
   IQR 2–3 sustained intracranial bleeds, with associated cal-
  Brain varial fractures.
   Median 4
   IQR 0–4
  Neck Management
   Median 0 Initial care conformed to the principles of
   IQR 0–0
Medical history the Advanced Trauma Life Support protocol. Two
 Psychiatric history 35 (51) patients (3 percent) required emergent surgical
 Substance abuse history 24 (35) neck exploration. Three patients required a cri-
Positive on toxicology screen
 Alcohol 34 (49) cothyroidotomy, all of which were performed in
 Benzodiazepines 12 (17) the emergency department. Approximately half
 Marijuana 7 (10) of the cohort (48 percent) ultimately underwent
 Opiates 6 (9)
 Cocaine 4 (6) a tracheostomy. Thirteen patients (19 percent)
AIS, Abbreviated Injury Scale; RTS, Revised Trauma Score; IQR, underwent conventional angiography of the
interquartile range; ISS, Injury Severity Score. neck, three of whom required angioembolization

Fig. 2. Self-inflicted shotgun injury delivered cephalad through a submental site producing dramatic injuries, including extensive
mid and lower face fractures and avulsed soft tissue (left). Injuries appear less disfiguring after initial débridement and local tissue
rearrangement (center). The shotgun pellets penetrate the cribriform plate and dura, but cerebral involvement is minimal (right).

418
Copyright © 2019 American Society of Plastic Surgeons. Unauthorized reproduction of this article is prohibited.
 Volume 144, Number 2 • Self-Inflicted Facial Gunshots
Table 2.  Multispecialty Surgical Management
Died (n = 18)
Patients (%)
Tracheostomy 6 (33)
Gastrostomy 2 (11)
Neurosurgery 7 (39)
Ophthalmology 1 (5.5)
Facial reconstructive 5 (28)
Total 12 (67)
for high-grade vessel injury. Formal neurosurgical
and ophthalmologic evaluation was performed
for 35 (51 percent) and 28 patients (41 percent),
respectively.
Table 2 summarizes the surgical procedures
performed for patients who survived and those
that did not survive. The most frequent neurosur-
gical procedures were ventriculoperitoneal shunt
placements, craniotomies, cranioplasties, and
duraplasties. The most frequent ophthalmologic
procedures were for repair of ruptured globes
and enucleation. Among the patients that under-
went reconstructive craniofacial surgical proce-
dures, the number of operations ranged from
one to 16 (mean ± SD, 3.9 ± 4.1), with soft-tissue,
bony débridement, and open reduction and inter-
nal fixation being the most common procedures.
Nine patients required 12 free flaps for bony and
soft-tissue reconstruction. Sixty-one patients (88
percent) required intensive care unit admission,
for which the average length of stay was 8.2 days.
Primary Outcome/Endpoint: Injury/Mortality
Analysis
Overall, 18 patients (26 percent) who sus-
tained self-inflicted gunshot wounds died. Six of
the 18 deaths (33 percent) occurred within 24
hours of admission. All but one of the 18 patients
that died had some form of brain injury; 15 sus-
tained penetrative brain injury, whereas the
other two did not. Of the subset of patients who
died without direct penetrative brain injury, one
patient died from anoxic brain injury, secondary
Table 3.  Univariable Association with Mortality
Variable OR
Age, per year 1.06
ISS, per unit 1.09
AIS score brain, per unit 1.80
Positive toxicology 0.39
Submental entry site 0.27
Vertical trajectory 0.28
Penetrative brain injury 6.87
Upper face fracture 2.58
AIS, Abbreviated Injury Scale; ISS, Injury Severity Score.
Copyright © 2019 American Society of Plastic Surgeons. Unauthorized reproduction of this article is prohibited.
Survived (n = 51)
Procedures Patients (%) Procedures
6 27 (53) 27
2 24 (47) 24
8 11 (22) 15
1 9 (18) 14
8 34 (67) 141
25 44 (86) 221
to uncontrolled hemorrhage and shock. The sec-
ond patient sustained subarachnoid hemorrhage
and pneumocephalus secondary to skull base frac-
tures, but had no evidence of dural violation. The
single non–brain injury death occurred because of
a myocardial infarction in a patient who had also
overdosed on acetaminophen before admission.
Predictors of Mortality
Table 3 shows the variables that were identi-
fied by binomial logistic regression to be signifi-
cantly associated with mortality. The presence of
penetrative brain injury was associated with the
largest odds of death (OR, 7, versus absence of
penetrative brain injury). When detailed by entry
site, mortality was highest for the temporal (63
percent) versus oral (27 percent) and submental
(15 percent) sites. The following variables that
were found to be colinear with penetrative brain
injury were subsequently excluded from the mul-
tivariable regression model: Injury Severity Score,
brain Abbreviated Injury Scale score, submental
entry site, vertical trajectory, and upper face frac-
tures. Perhaps nonintuitively, positive toxicology
was associated with a lower likelihood of death.
The following variables were not significantly dif-
ferent for the survivors versus the patients that
died: sex, race, Glasgow Coma Scale score, face
Abbreviated Injury Scale score, Revised Trauma
Score, weapon type, time to tracheostomy, time on
ventilator, conventional angiography, and num-
ber of operations. Of the 39 individuals that died
before admission, one individual (3 percent) had
sustained injuries from a shotgun, whereas 38 (97
percent) individuals had injuries from handguns.
Among the cohort of 69 patients that were admit-
p ted alive, mortality was 31 percent from shotguns
0.0016 versus 26 percent from handguns (p = 0.72).
0.0051
0.0006 On multivariable regression, the variables
0.0075 that were independently associated with mortal-
0.0250 ity were (1) presence of penetrative brain injury
0.0150
0.0008 (OR, 17; p < 0.0001) and (2) age (OR, 1.09 per
0.0057 year of advancing age; p < 0.0001). The area under
the receiver operating characteristic curve was
419
 Plastic and Reconstructive Surgery • August 2019

88 percent, indicating good precision. Mortality DISCUSSION

was significantly less in patients aged 65 years or
younger (17 percent), compared with those aged
65 years and older (73 percent). Figure 3 displays
the probability of death obtained from the mul-
tiple binomial regression model as a function of
increasing age.
Secondary Outcomes
Overall hospital length of stay ranged from 0.1
to 60 days. Mean length of stay among patients
that survived was 16 ± 12 days, versus 5 ± 5 days
for the patients that died (p < 0.0001). Among the
survivors, multivariable linear regression identi-
fied the following as associated with length of stay:
gastrostomy placement (4 days shorter hospital-
ization for patients that underwent gastrostomy
placement; p = 0.0003), and prolonged ventilator
dependence (1 additional day of hospitalization
per day on the ventilator; p < 0.0001). The major-
ity of the patients that survived were discharged
to a hospital inpatient psychiatry service (32 per-
cent) or an inpatient rehabilitation facility (25
percent). Of note, eight patients (12 percent)
were discharged directly to home following psy-
chiatric consultation and recommendations. For
the patients that survived, median follow-up was
16 months (range, 1 day to 10 years). There were
no known repeated attempts at suicide recorded
in the medical records.
The incidence of self-inflicted gunshot wounds
to the face is rare. At our Level I trauma center,
fewer than 0.1 percent of all trauma admissions
met inclusion criteria. The natural history and
hospital course of our cohort of 69 patients repre-
sents the largest single-center cohort study of facial
self-inflicted gunshot wounds in the literature.
The key observations from our study are as
follows. First, 38 percent of individuals that are
transported to our center following self-inflicted
gunshot wounds are pronounced dead before
admission. Second, 74 percent of the patients that
are admitted alive survived. The odds of survival
were much higher if the patient was 65 years of
age or younger (83 percent). Third, approxi-
mately one-third of those who died did so within
the first 24 hours of admission. Of those who died,
94 percent died because of brain injury. Fourth,
penetrative brain injury and advanced age are
independently associated with mortality. Fifth,
gastrostomy placement is associated with shorter
hospital stay among survivors. Sixth, survivors
require complex multispecialty surgical and criti-
cal care and prolonged hospitalization.
In reviewing the literature on self-inflicted
gunshot wounds to the face, there is a paucity of
high-volume studies, perhaps because of the infre-
quency of this injury. Rather, most published stud-
ies also include injuries that were not self-inflicted,

Fig. 3. Independent predictors of mortality. The probability of death was calculated

from the odds ratio for death/survival, which was obtained from the multiple binomial
regression model. The probability of death is higher for patients that sustain penetra-
tive brain injury (open squares) versus patients that do not (filled circles) (p < 0.0001). The
probability of death also increases with increasing patient age, independent of penetra-
tive brain injury (p < 0.0001). The area under the receiver operating characteristic curve
was calculated to be 88 percent.

420
Copyright © 2019 American Society of Plastic Surgeons. Unauthorized reproduction of this article is prohibited.
 Volume 144, Number 2 • Self-Inflicted Facial Gunshots
which tend to be less lethal,17 or isolated calvarial
injuries, which tend to be more lethal.5,8,18–20 For
example, the mortality from civilian gunshot
wounds to the head has been reported to exceed
90 percent, with 76 percent dying at the scene,13
and autopsy studies of fatal self-inflicted gunshot
wounds demonstrate predominance of temporal
and frontal entry wounds.21 In contrast, among
the patients that survive transport to the hospital
following craniofacial gunshot wounds, the sub-
mentum is a more prevalent entry site, and over-
all mortality rates are as low as 35 percent.9,19,22
Mortality rates from the submental trajectory are
particularly low (15 percent in our cohort), possi-
bly because of neck hyperextension at the instant
of firing. The bullet ends up traversing anteriorly,
primarily through the facial skeleton, sparing
major vessels and intracranial structures.9,18,22
Although our data show lower mortality
among patients with a submental entry site, when
penetrative brain injury is included in the regres-
sion model, the entry site is no longer indepen-
dently associated with mortality. This finding
suggests that although the entry site may affect the
odds, survival is ultimately determined by whether
or not there is penetrative brain injury. Similarly,
we found other injury variables that have been
previously shown to be associated with gunshot
wound mortality (i.e., fractures in the upper face,
higher Injury Severity Score, nonvertical trajec-
tory of gunshot)7,22 to be colinear with penetra-
tive brain injury but not independently associated
with mortality. Conversely, positive toxicology was
associated with lower likelihood of death, but was
not independently associated with lower mortal-
ity. This finding suggests that sober individuals are
more likely to be successful at delivering lethal,
brain-penetrating gunshots.
Although advanced age has also been shown
to be associated with mortality from gunshot inju-
ries,7 our data provide objective estimates for prob-
ability of dying (Fig. 3) as a function of age and
whether or not there is penetrative brain injury.
Furthermore, the contrast between mortality for
patients aged up to 65 years and those older than
65 years (17 percent versus 73 percent) is striking.
This simple statistic should be kept in mind when
making clinical decisions about patients with self-
inflicted gunshot wounds to the face, particularly
because this population is generally older than
the non–self-harm gunshot wound victims.8,23
Our finding that gastrostomy placement is
independently associated with a 4-day shorter (25
percent reduction of the 16-day mean) hospital
stay among the survivors is significant. By reducing
Copyright © 2019 American Society of Plastic Surgeons. Unauthorized reproduction of this article is prohibited.
the length of hospitalization, gastrostomy place-
ment may reduce overall medical costs and permit
earlier rehabilitation.
This study has several limitations. It is a retro-
spective study of a single-institution cohort, does
not provide population-based information, and
excludes victims that die before arrival. Further-
more, the small sample size limits the number of
variables that can be meaningfully included in our
analyses. Notably, we are unable to further char-
acterize any associations of location or extent of
penetrative brain injury with mortality. Finally,
follow-up is poor (median, 16 months), as with
many trauma cohorts, limiting our assessment of
reconstructive outcomes and recidivism.
CONCLUSIONS
Despite tremendous morbidity, the over-
whelming percentage of patients who present
with facial self-inflicted gunshot wounds will sur-
vive their injuries, especially if they are young and
do not have penetrative brain injury. These find-
ings should help guide clinical decision-making
for this uncommon yet devastating injury.
Arthur J. Nam, M.D., M.S.
Division of Plastic Surgery
R Adams Cowley Shock Trauma Center
110 South Paca Street, Suite 4S-125
Baltimore, Md. 21201
anam@umm.edu
REFERENCES
1. Hedegaard H, Warner M, Curtin SC. Increase in suicide in
the United States, 1999–2014. NCHS Data Brief. 2016;241:1–8.
2. U.S. Centers for Disease Control and Prevention 2016. Web-
based Injury Statistics and Query and Reporting System
(WISQARS): WISQARS nonfatal injury reports, 2000 - 2017.
Available at: https://webappa.cdc.gov/sasweb/ncipc/nfi-
rates.html. Accessed April 4, 2018.
3. Williams ST, Kores RC, Currier JM. Survivors of self-inflicted
gunshot wounds: A 20-year chart review. Psychosomatics
2011;52:34–40.
4. Demirci S, Dogan KH, Deniz I, Erkol Z. Evaluation of shot-
gun suicides in Konya, Turkey between 2000 and 2007. Am J
Forensic Med Pathol. 2014;35:45–49.
5. Gross BW, Cook AD, Rinehart CD, et al. An epidemio-
logic overview of 13 years of firearm hospitalizations in
Pennsylvania. J Surg Res. 2017;210:188–195.
6. Shuck LW, Orgel MG, Vogel AV. Self-inflicted gunshot wounds
to the face: A review of 18 cases. J Trauma 1980;20:370–377.
7. Shackford SR, Kahl JE, Calvo RY, et al. Gunshot wounds and
blast injuries to the face are associated with significant morbid-
ity and mortality: Results of an 11-year multi-institutional study
of 720 patients. J Trauma Acute Care Surg. 2014;76:347–352.
8. Mathews EM, Woodward CJ, Musso MW, Jones GN. Suicide
attempts presenting to trauma centers: Trends across age
groups using the National Trauma Data Bank. Am J Emerg
Med. 2016;34:1620–1624.
421
 Plastic and Reconstructive Surgery • August 2019
9. Calhoun KH, Li S, Clark WD, Stiernberg CM, Quinn FB Jr.
Surgical care of submental gunshot wounds. Arch Otolaryngol
Head Neck Surg. 1988;114:513–519.
10. Clark N, Birely B, Manson PN, et al. High-energy ballistic
and avulsive facial injuries: Classification, patterns, and an
algorithm for primary reconstruction. Plast Reconstr Surg.
1996;98:583–601.
11. Denny AD, Sanger JR, Matloub HS, Yousif NJ. Self-inflicted
midline facial gunshot wounds: The case for a combined cra-
niofacial and microvascular team approach. Ann Plast Surg.
1992;29:564–570.
12. Gruss JS, Antonyshyn O, Phillips JH. Early definitive bone
and soft-tissue reconstruction of major gunshot wounds of
the face. Plast Reconstr Surg. 1991;87:436–450.
13. Aarabi B, Tofighi B, Kufera JA, et al. Predictors of out-
come in civilian gunshot wounds to the head. J Neurosurg.
2014;120:1138–1146.
14. Gussack GS, Jurkovich GJ. Penetrating facial trauma: A man-
agement plan. South Med J. 1988;81:297–302.
15. Gant TD, Epstein LI. Low-velocity gunshot wounds to the
maxillofacial complex. J Trauma 1979;19:674–677.
16. Harris PA, Taylor R, Thielke R, Payne J, Gonzalez N, Conde
JG. Research electronic data capture (REDCap): A metadata-
driven methodology and workflow process for providing
422
Copyright © 2019 American Society of Plastic Surgeons. Unauthorized reproduction of this article is prohibited.
translational research informatics support. J Biomed Inform.
2009;42:377–381.
17. Demetriades D, Chahwan S, Gomez H, Falabella A, Velmahos
G, Yamashita D. Initial evaluation and management of gun-
shot wounds to the face. J Trauma 1998;45:39–41.
18. Murphy JA, Lee MT, Liu X, Warburton G. Factors affect-
ing survival following self-inflicted head and neck gunshot
wounds: A single-centre retrospective review. Int J Oral
Maxillofac Surg. 2016;45:513–516.
19. Tholpady SS, DeMoss P, Murage KP, Havlik RJ, Flores RL.
Epidemiology, demographics, and outcomes of cranio-
maxillofacial gunshot wounds in a level I trauma center. J
Craniomaxillofac Surg. 2014;42:403–411.
20. Hadjizacharia P, Brown CV, Teixeira PG, et al. Traumatic suicide
attempts at a level I trauma center. J Emerg Med. 2010;39:411–418.
21. Nikolić S, Živković V, Babić D, Juković F. Suicidal single gun-
shot injury to the head. Am J Forensic Med Pathol. 2012;33:43–46.
22. Johnson J, Markiewicz MR, Bell RB, Potter BE, Dierks EJ.
Gun orientation in self-inflicted craniomaxillofacial gun-
shot wounds: Risk factors associated with fatality. Int J Oral
Maxillofac Surg. 2012;41:895–901.
23. Gani F, Sakran JV, Canner JK. Emergency department vis-
its for firearm-related injuries in the United States, 2006-14.
Health Aff (Millwood) 2017;36:1729–1738.