The Journal of Emergency Medicine, Vol. 57, No. 1, pp.

21–28, 2019
Ó 2019 Elsevier Inc. All rights reserved.
0736-4679/$ - see front matter

https://doi.org/10.1016/j.jemermed.2019.03.013

Brief
Report

FACTORS ASSOCIATED WITH POOR OUTCOME IN PEDIATRIC

NEAR-HANGING INJURIES

Stephanie La Count, MD,* Marlina E. Lovett, MD,*† Songzhu Zhao, MS,‡ David Kline, PHD,‡
Nicole F. O’Brien, MD,*† Mark W. Hall, MD,*† and Eric A. Sribnick, MD, PHD§k
*Department of Pediatrics, Nationwide Children’s Hospital, The Ohio State University College of Medicine, Columbus, Ohio, †Division of
Critical Care Medicine, Nationwide Children’s Hospital, Columbus, Ohio, ‡Center for Biostatistics, The Ohio State University, Columbus, Ohio,
§Division of Neurosurgery, Nationwide Children’s Hospital, The Ohio State University, Columbus, Ohio, and kDepartment of Neurosurgery,
The Ohio State University, Columbus, Ohio
Corresponding Address: Eric Sribnick, MD, PHD, Nationwide Children’s Hospital, 700 Children’s Drive, Columbus, OH 43205

, Abstract—Background: Hanging injury is the most com- an initial GCS score of 3T and prehospital cardiac arrest had
mon method of suicide among children 5 to 11 years of age a favorable neurologic outcome. Conclusions: This is the
and near-hangings commonly occur. Adult studies in near- largest single-center study of children with near-hanging
hanging injury have shown that need for cardiopulmonary injury. An initial GCS score of 3T and prehospital cardiac ar-
resuscitation, initial blood gas, and poor mental status are rest was uniformly associated with poor neurologic out-
associated with poor prognosis. The literature for similar fac- come. Ó 2019 Elsevier Inc. All rights reserved.
tors in children is lacking. Objectives: This retrospective,
single-center study was performed to identify the clinical fac- , Keywords—children; near-hanging; neurologic outcome
tors associated with neurologic outcome in children after
near-hanging. Methods: Inclusion criteria included <18 years
of age and a diagnosis of near-hanging or strangulation. All INTRODUCTION
physician documentation was reviewed, and incidences of res-
piratory complications, seizure, and multiorgan failure were
Over the last 2 decades, the annual suicide rate by suffo-
noted. Pediatric cerebral performance category score was
based on information at discharge and was defined as favor- cation has been steadily rising, increasing from 2.7 per
able (score of 1–4) or unfavorable (score of 5–6). Compari- 100,000 population from 1999 to 2007 to 3.7 per
sons were made between outcome groups and suspected 100,000 population in 2008 to 2015 (1). Data from the
clinical factors. Results: The median age was 11.5 years National Violent Death Reporting System in 16 states
with a median initial Glasgow Coma Scale (GCS) score of from 2005 to 2014 show that 90.7% of asphyxial suicides
10. Of all patients, 25% had a prehospital cardiac arrest, involved hanging, translating to 22,931 lives (2). While
and 51% were admitted to the intensive care unit. Patients most violent suicide attempts involve adults, a recent
with unfavorable outcomes had a lower initial pH (6.9 vs. study noted that 10.5% of attempts involved children 10
7.3) and initial GCS score (3T vs. 14). Patients with an unfa- to 19 years of age (3). Similarly, while the majority of
vorable outcome had significantly higher rates of intensive
hanging and near-hanging events involve adults, self-
care unit admission, respiratory complications, anoxic brain
strangulation (hanging/suffocation) is the most common
injury, and multiorgan failure. No patient who presented with
method of suicide among children, accounting for
78.2% of total suicide deaths of children 5 to 11 years
Reprints are not available from the authors. of age from 1993 to 2012 in the United States (4).

RECEIVED: 17 December 2018; FINAL SUBMISSION RECEIVED: 3 March 2019;

ACCEPTED: 6 March 2019

21
22 S. La Count et al.

Hanging is defined as ‘‘death due to external pressure presentation, radiographic imaging, hospital course, and
on the neck when a ligature is applied to the neck of a disposition. The initial blood gas data presented was ob-
wholly or partly suspended individual (5).’’ The term tained within 1 h of hospital presentation. Because of the
‘‘near-hanging’’ refers to individuals who survive the evolving diagnostic criteria for the following diagnoses
initial insult long enough to reach the hospital (5,6). A over the past 2 decades, the presence of complications,
variety of factors have been proposed to guide such as pulmonary complications (pulmonary edema or
clinicians in predicting neurologic outcomes in adults acute respiratory distress syndrome [ARDS]), seizure,
after near-hanging events, but similar work in pediatric pneumonia, and multiorgan failure were obtained from
patients is lacking (7–9). In adults, factors associated critical care or subspecialist documentation in the medi-
with a poor neurologic outcome after near-hanging cal record during the hospital admission. The diagnosis
include the use of cardiopulmonary resuscitation, initial of anoxic brain injury was obtained by reviewing all neu-
arterial blood gas, and initial mental status (either an roimaging reports and interpretations. Within the institu-
initial Glasgow Coma Scale [GCS] score of 3 at the tion, all neuroimaging is reviewed by a pediatric
time of discovery or comatose status at the time of arrival radiologist. Neurologic examination on the day of
to the emergency department) (10,11). Similarly, in a discharge was used to assign a Pediatric Cerebral Perfor-
small, single-center study of 41 children, those who mance Category (PCPC) score for each patient, which is a
were pulseless at the time of discovery were at high validated measure of short-term neurologic outcome in
risk for death or severe disability (10). An important children (13). The PCPC score ranges from 1 to 6;
prognostic factor that has been described in adult studies 1 = normal (‘‘at age-appropriate level, school-age child
is the duration of hanging, which is often not known in pe- attending regular school classroom’’); 2 = mild disability
diatric strangulation (12). (‘‘conscious, alert, and able to interact at age-appropriate
Given the paucity of data regarding clinical factors level; school-age child attending regular school class-
that can aid in neurologic prognostication of children af- room but grade perhaps not appropriate for age, possibil-
ter near-hanging injury, it is imperative to advance ity of mild neurologic deficit’’); 3 = moderate disability
research in this area so that clinicians can better counsel (‘‘conscious, sufficient cerebral function for age-
families during this difficult time. Therefore, this large, appropriate independent activities of daily life; school-
single-center, retrospective study was performed to iden- age child attending special education classroom or
tify clinical factors that were associated with neurologic learning deficit present’’); 4 = severe disability
outcome in children after near-hanging. (‘‘conscious; dependent on others for daily support
because of impaired brain function’’); 5 = coma or vege-
tative state (‘‘any degree of coma without the presence of
MATERIALS AND METHODS
all brain death criteria; unawareness, even if awake in
Patient Selection appearance, without interaction with environment; cere-
bral unresponsiveness and no evidence of cortex function
Data for this study were obtained retrospectively from a [not aroused by verbal stimuli]; possibility of some re-
state-mandated trauma database (Central Trauma Regis- flexive response, spontaneous eye-opening, and sleep-
try) that is maintained by dedicated data entry personnel wake cycles’’); and 6 = brain death (‘‘apnea, areflexia,
under the supervision of nurse practitioners and physi- and/or electroencephalographic silence’’) (13).
cians. Patients were treated at a level 1 pediatric trauma
center between June 1992 and September 2015. This Statistical Analysis
study was reviewed by the institutional review board.
Informed consent was not required for admission into Descriptive continuous data were reported as median
this study given its retrospective nature. with first and third quartiles, and categorical data were re-
Patients were identified by description of injury in the ported as percentage. The Fisher exact test was used for
trauma registry. Inclusion criteria included patients the comparison of categorical variables and the Wilcoxon
<18 years of age who were admitted with a diagnosis of test was used for the comparison of continuous variables.
near-hanging or strangulation. There were no exclusion All statistical analyses were conducted using SAS soft-
criteria to this study. ware for Windows (version 9.4; SAS Institute Inc.,
Cary, NC). Statistical significance was defined as
Data Collection p # 0.05. To examine outcome, pediatric cerebral perfor-
mance scores were grouped a priori into two categories:
Information extracted from the medical record included: those with a more favorable neurologic outcome (PCPC
patient demographics, mechanism of injury, medical score 1–4, from normal to severe disability) and those
history, vital signs at presentation, laboratory results at with an unfavorable neurologic outcome (PCPC score
Pediatric Near-Hanging and Outcome 23

5–6, those in a comatose/vegetative state or dead). Preho- Table 2. Inpatient Clinical Characteristics
spital characteristics (limited to the presence or absence Total, N 84
of prehospital cardiac arrest) and in-hospital characteris- PICU admission, n (%) 43 (51.2)
tics of the study population were summarized for all sub- Cervical fracture, n 0
Seizure detected, n (%) 17 (20.2)
jects and also stratified by neurologic outcome group. To Pulmonary complications,*n (%) 16 (19.1)
further evaluate the natural course of the disease process Pneumonia, n (%) 8 (9.5)
in children who died, characteristics were compared be- Multiorgan failure, n (%) 10 (11.9)
Anoxic brain injury, n (%) 16 (19.3)
tween children who had withdrawal of life support vs. PICU length of stay, days, median (Q1, Q3) 1 (0, 2)
those who did not. Ventilator days, median (Q1, Q3) 1 (1, 3)
Hospital length of stay, days, median (Q1, Q3) 2 (1, 3)

RESULTS PICU = pediatric intensive care unit; Q1 = first quartile; Q3 = third

quartile.
Patient Demographics and Clinical Characteristics * Denotes both acute respiratory distress syndrome and pulmo-
nary edema.

Eighty-four patients met the inclusion criteria, and their

demographics and clinical characteristics at presentation having either a favorable neurologic outcome (PCPC
are shown in Table 1. Patients ranged from <1 to 17 years score of 1–4) or an unfavorable neurologic outcome
of age (median 11.5 years). After review of the medical (PCPC score of 5 or 6 [i.e., comatose state or death,
record, a general cause of injury could be obtained in respectively]). In our cohort, 90% of patients had either
79 cases (94%). Accidental ligature in an infant (1 month a PCPC score of 1 or 6 at the time of hospital discharge
to 1 year of age) or toddler (1–3 years of age) was noted in (Figure 1).
19 cases (24% of the patients with known causes of Data obtained either in the prehospital setting or at the
injury), and accidental ligature in an older child (>3 years time of admission were examined to determine if any
of age) was noted in 9 cases (11.4%). Voluntary ligature observed findings correlated with patient outcome
without suicidal intent (e.g., ‘‘choking game’’) was noted (Table 3). The unfavorable neurologic outcome group
in 8 cases (10.1% of patients with known causes of (PCPC score 5–6) was younger than those with a favor-
injury). However, most patients (43 cases) presented able neurologic outcome (8.5 vs. 12 years, p = 0.02).
with a history suggesting voluntary ligature with suicidal All patients in the unfavorable neurologic outcome group
ideation (54.4%). had a prehospital cardiac arrest compared with only 1 in
Clinical findings among the cohort of patients are out- the favorable neurologic outcome group (p < 0.0001).
lined in Table 2. Of the 84 total patients, 43 (51%) were The unfavorable neurologic outcome group also had a
initially admitted to the pediatric intensive care unit lower pH on presentation (6.9 vs. 7.3, p < 0.001) as
(PICU). The median PICU length of stay was 1 day (in- well as a lower initial GCS score (3T vs. 14,
terquartile range 0–2 days), and the average hospital p < 0.001). All 20 patients with an unfavorable neurologic
length of stay was 2 days (interquartile range 1–3 days). outcome (PCPC score 5–6) had both a prehospital cardiac
No cervical fractures or dislocations were discovered arrest and an initial GCS score of 3T.
on review of radiographic imaging. There was evidence Of the 64 patients who ultimately had a favorable
of anoxic brain injury in 16 patients (19.3% of the patient neurologic outcome (PCPC score 1–4) at the time of
cohort). discharge, 6 had a GCS of 3T at initial hospital evaluation.
To determine whether initial or prehospital character-
istics correlated with outcome, patient outcome was
measured by PCPC score, and patients were labeled as

Table 1. Prehospital and Admission Patient Characteristics

Total, N 84
Male gender, n (%) 54 (64.3)
Prehospital cardiac arrest, n (%) 21 (25)
Age, y, median (Q1, Q3) 11.5 (3, 14)
Arterial blood gas pH,* median (Q1, Q3) 7.2 (7.0, 7.3)
Arterial blood gas CO2,* median (Q1, Q3) 42 (31, 50)
Initial GCS score, median (Q1, Q3) 10 (3, 15)
Injury severity Score, median (Q1, Q3) 1 (1, 10)

GCS = Glasgow coma scale; Q1 = first quartile; Q3 = third quar-

tile. Figure 1. Bar graph showing the Pediatric Cerebral Perfor-
* Arterial blood gas data were available for 48 of 84 patients. mance Category score distribution for the entire cohort.
24 S. La Count et al.

Table 3. Prehospital and Admission Characteristics, Compared by Patient Outcome

PCPC Score 1–4 PCPC Score 5–6 p Value

Total, n (%) 64 (76.2) 20 (23.8)

Male gender, n (%) 40 (62.5) 14 (70.0) 0.6
Prehospital cardiac arrest, n (%) 1 (1.6) 20 (100.0) <0.0001
Age, y, median (Q1, Q3) 12 (4.5, 14) 8.5 (1.0, 12.5) 0.02
Arterial blood gas pH,* median (Q1, Q3) 7.3 (7.2, 7.4); n = 33/64 (52%) 6.9 (6.7, 7.0); n = 15 (75%) <0.0001
Arterial blood gas CO2,* median (Q1, Q3) 41 (35, 48) 51.5 (18, 79) 0.3
Initial GCS score, median (Q1, Q3) 14 (7, 15) 3 (3, 3) <0.0001
ISS, median (Q1, Q3) 1 (1, 10) 1 (1, 25.5) 0.9

ARDS = acute respiratory distress syndrome; GCS = Glasgow Coma Scale; ISS = Injury Severity Score; PCPC = Pediatric Cerebral Per-
formance Category; PICU = pediatric intensive care unit; Q1 = first quartile; Q3 = third quartile.
* Arterial blood gas data were available for 48 of 84 patients.

Of those 6 patients, none had a prehospital cardiac arrest, and hospital lengths of stay (p = 0.02 for both), but no
and at the time of discharge, 5 patients had a PCPC score other metric (pH, partial pressure of carbon dioxide,
of 1 (normal), and 1 patient had a PCPC score of 2 (mild initial GCS score, and Injury Severity Score) examined
disability). One patient who was ultimately discharged showed a significant difference between groups
with a PCPC score of 1 did have a prehospital cardiac ar- (Table 5).
rest; however, this patient had a reported GCS score of 11
at the initial hospital evaluation. DISCUSSION
Data obtained during hospitalization, including PICU
admission/length of stay and complications of near- To the best of our knowledge, this study represents the
hanging, were also examined to determine if these factors largest single-center cohort of children after near-
correlated with outcome (Table 4). The unfavorable hanging. Our findings contribute to the growing literature
neurologic outcome group had a higher incidence of on neurologic outcome from near-hanging injury. Inten-
PICU admission (p = 0.02), a longer PICU length of tional self-strangulation with suicidal intent was the
stay (p = 0.001), and higher rates of pneumonia most common known reason for near-hanging in this
(p = 0.02) as well as pulmonary complications (pulmo- cohort, and this finding parallels previous research
nary edema or ARDS, p = 0.002). The presence of multi- (4,14). Previous adult studies have noted an association
organ failure and anoxic brain injury were significantly with neurologic outcome and multiple clinical,
different (p < 0.0001) between the two groups. laboratory, and imaging features. Specifically, poor
Nineteen children died during their hospital admis- outcome has been associated with a clinical history of
sion. Nine children underwent withdrawal of life sup- cardiac arrest hypotension at presentation, and an
port for presumed poor neurologic outcome after elevated Injury Severity Score (7–9,11,12,15,16). An
counseling from multiple services (PICU, neurology, initial acidotic blood gas has been associated with poor
or neurosurgery). Ten children died without withdrawal neurologic outcome, as has the presence of anoxic brain
of life support. Patients who had life support withdrawn injury on initial imaging (8,11,12). Similarly, the
were more likely to have had significantly longer PICU limited pediatric evidence has noted an association of

Table 4. Inpatient Clinical Characteristics, Compared by Patient Outcome

PCPC Score 1–4 PCPC Score 5–6 p Value

Total, n (%) 64 (76.2) 20 (23.8)

PICU admission, n (%) 28 (43.7) 15 (75.0) 0.02
Seizure, n (%) 11 (17.2) 6 (30.0) 0.2
Pulmonary edema, n (%) 7 (10.9) 9 (45.0) 0.002
Pneumonia, n (%) 3 (4.7) 5 (25.0) 0.02
ARDS, n (%) 0 (0.0) 4 (20.0) 0.002
Multiorgan failure, n (%) 0 (0.0) 10 (50.0) <.0001
Anoxic brain injury, n (%) 5 (7.8) 11 (57.9) <.0001
PICU length of stay, days, median (Q1, Q3) 0 (0, 1) 2 (0.5, 4) 0.001
Ventilator days, median (Q1, Q3) 1 (1, 2) 2.5 (1, 4) 0.07
Hospital length of stay, days, median (Q1, Q3) 2 (1, 3) 2 (0.5, 4) 0.8

ARDS = acute respiratory distress syndrome; PCPC = Pediatric Cerebral Performance Category; PICU = pediatric intensive care unit;
Q1 = first quartile; Q3 = third quartile.
Pediatric Near-Hanging and Outcome 25

Table 5. Prehospital and Inpatient Characteristics, Compared by End-of-Life Care

Life Support Withdrawn Mortality Without Life Support Withdrawn p Value

Total, N 9 10
Male gender, n (%) 6 (66.7) 7 (70.0) 1
Prehospital cardiac arrest, n (%) 9 (100) 10 (100) N/A
Seizures, n (%) 4 (44.4) 1 (10.0) 0.1
Pulmonary edema, n (%) 5 (55.6) 4 (40.0) 0.7
Pneumonia, n (%) 2 (22.2) 3 (30.0) 1
ARDS, n (%) 3 (33.3) 1 (10.0) 0.3
Multiorgan failure, n (%) 6 (66.7) 4 (40.0) 0.4
Anoxic brain injury, n (%) 7 (77.8) 3 (33.3) 0.2
Age, y, median (Q1, Q3) 5 (1, 9) 11 (0.8, 13.0) 0.3
Arterial blood gas pH, median (Q1, Q3) 6.9 (6.6, 7.0) 6.8 (6.8, 6.9) 0.7
Arterial blood gas CO2, median (Q1, Q3) 47.5 (18.0, 56.0) 79 (58, 87) 0.1
Initial GCS score, median (Q1, Q3) 3 (3, 3) 3 (3, 3) 1
Injury Severity Score, median (Q1, Q3) 1 (1, 26) 1 (1, 25) 0.8
PICU length of stay, days, median (Q1, Q3) 4 (2, 6) 0.5 (0, 3) 0.02
Hospital length of stay, median (Q1, Q3), days 4 (2, 6) 0.5 (0, 3) 0.02

ARDS = acute respiratory distress syndrome; GCS = Glasgow Coma Scale; PICU = pediatric intensive care unit; Q1 = first quartile;
Q3 = third quartile.

pulseless at the time of discovery with death and severe proposed to potentially risk-stratify outcome; however,
disability (10). such information is rarely known in pediatric hangings,
Consistent with a previous pediatric study of near- as noted by our findings and those of others (10,12). In
hanging, the majority of patients in our cohort were fact, hanging duration was not reported in this study
male, which is also similar in the adult literature because it was seldom evident in the medical record.
(7,10,17,18). The median age of our cohort was The presenting GCS score alone has been previously
11.5 years, slightly lower than previous studies in offered as an indicator, and a low GCS score does have
which the median ages were adolescents (13–14 years an association with poor outcome, as seen in our data,
of age) (3,10,14). In our study, the median PICU length other single-center pediatric cohorts, and in analysis of
of stay was 1 day, which is shorter than the mean of national databases (9,10). However, our study and
4.4 days in a small cohort of 16 children reported by others have shown that near-hanging patients with poor
Hackett et al. (14). The difference in PICU length of initial GCS scores can still recover well (10,19).
stay is potentially related to differences in illness severity Therefore, to more accurately predict outcome, multiple
because 63% of their cohort required PICU admission, factors need to be considered. Additional prehospital
compared with 51% in our cohort, and their median dura- data may be insightful to emergency physicians who
tion of intubation was 2.2 days vs. 1 day in our cohort. are actively working to stabilize the child. The
The rate of prehospital cardiopulmonary arrest was prehospital data presented in this article were limited to
25% within our cohort. This is lower than a previously the presence or absence of cardiac arrest. However,
published pediatric study in which 46% were found to additional prehospital data should be evaluated in future
be pulseless upon the arrival of emergency medical ser- studies and could include rates of endotracheal
vices (10). Our retrospective data show that several find- intubation, the presence or absence of hypoxemia, the
ings were associated with poor outcome: a prehospital presence or absence of hypotension, the first recorded
history of cardiac arrest, an initial GCS score of 3T, and heart rhythm, the presence or absence of bystander
a lower initial pH; however, every patient with an unfa- cardiopulmonary resuscitation, and relevant scene times.
vorable neurologic outcome (PCPC score of 5–6) pre- Multiple complications associated with near-hanging
sented with a GCS score of 3T and a history of injuries have been previously described. Complications
prehospital cardiac arrest. No patients who ultimately seen in our cohort of patients that have been reported else-
had a favorable neurologic outcome (PCPC score of 1– where include pneumonia, seizures, anoxic brain injury,
4) presented with both GCS of 3T and prehospital cardiac ARDS, and postobstructive pulmonary edema (8,9,20).
arrest, and there were several patients who presented with In our cohort, PICU admission, PICU length of stay,
one but not both of those findings who did have a good seizure, pulmonary complications (pulmonary edema or
outcome at the time of discharge. ARDS), pneumonia, multiorgan failure, and anoxic
Previous studies have used injury details to prognosti- brain injury were all noted significantly more frequently
cate outcome. For instance, hanging time has been in patients with worse outcome. The rate of anoxic
26 S. La Count et al.

brain injury in our cohort was 19%, similar to an adult gesting that early withdrawal of life support was not a
study reporting a rate of 13% (8). However, multiorgan confounding factor.
failure was seen exclusively in patients with a worse
outcome. Limitations
An analysis of the U.S. National Trauma Data Bank
found that 7% of patients presenting with hanging injury There were several limitations to our study, notably the
had a vertebral fracture and that there was a 3% incidence retrospective, single-center nature of the study. Although
of spinal cord injury (9). Notably, cervical spine injury we tried to perform extraction of the medical record in a
was not seen in our patients, consistent with a smaller systematic manner, it is possible that some data were
study by Davies et al. in which no child who presented omitted. During the period encompassed by this project,
with a hanging injury had neither a cervical vertebral our institution transitioned to an electronic medical re-
fracture nor cervical spinal cord injury (10). This differ- cord. Notably, some data were extrapolated from medical
ence may be attributable to a smaller drop height involved documentation, where there is the possibility of subjec-
in pediatric near-hangings. In addition, the anatomy of tivity in diagnosis of related conditions. During the 20-
pediatric patients may make them less prone to cervical year time period of this study, medicine has evolved,
spine fractures after near-hanging or hanging injuries. and it is possible that identification and treatment strate-
In an autopsy study of 307 hanging victims, age was gies for similar disease processes have changed, which
felt to be an important variable in the presence or absence may have impacted our data. Lastly, our data were limited
of cervical injury, with injury correlating with advancing to short-term neurologic outcome; both functional and
age (21). Nonetheless, a high degree of suspicion for soft long-term outcome data were lacking. Future work should
tissue injury should be used when treating near-hanging examine the question of prognosis after near-hanging in-
patients, especially when a poor neurologic examination juries using a prospective design, multiple clinical sites,
makes examination more difficult (8,22,23). and formalized definitions for associated complications.
Lung injury was noted in our patient population, and
our findings are largely in agreement with previously CONCLUSIONS
published work. We and others have demonstrated that
pulmonary edema after near-hanging can be seen even In this single-center retrospective analysis of children af-
in patients who ultimately have a good neurologic ter a near-hanging event, we determined that patients who
outcome (20,24,25). We observed ARDS exclusively in either died or were discharged in a comatose state uni-
patients who either died or were persistently comatose, formly had an initial presentation with prehospital car-
and this is in contrast to previous adult studies that diac arrest and a GCS score of 3T. This combination of
observed that some near-hanging patients with ARDS findings was not seen in any patient who had a favorable
did make a meaningful recovery (15). Given the inci- neurologic outcome, defined as PCPC score of 1 to 4.
dence of pulmonary complications—almost 20% in our
study—clinicians must have a high suspicion for the po-
tential development of pulmonary edema in patients with REFERENCES
respiratory distress. In addition, clinicians may find that a
1. Kegler SR, Stone DM, Holland KM. Trends in suicide by level of
lung protective, open-lung (higher positive end- urbanization - United States, 1999-2015. MMWR Morb Mortal
expiratory pressure and a lower tidal volume) ventilator Wkly Rep 2017;66:270–3.
strategy within this cohort of patients may prove useful. 2. Yau RK, Paschall MJ. Epidemiology of asphyxiation suicides in the
United States, 2005-2014. Inj Epidemiol 2018;5:1.
In many disease processes within pediatric neurocriti- 3. Mathews EM, Woodward CJ, Musso MW, et al. Suicide attempts
cal care, the natural reporting of the disease process is presenting to trauma centers: trends across age groups using the Na-
impacted by withdrawal of life support. Often following tional Trauma Data Bank. Am J Emerg Med 2016;34:1620–4.
4. Bridge JA, Asti L, Horowitz LM, et al. Suicide trends among
counseling regarding an expected poor neurologic elementary school-aged children in the United States from 1993
outcome, families elect to withdraw life support as to 2012. JAMA Pediatr 2015;169:673–7.
opposed to proceeding with formal brain death testing. 5. Gandhi R, Taneja N, Mazumder P. Near hanging: early intervention
can save lives. Indian J Anaesth 2011;55:388–91.
To examine whether early withdrawal of life support 6. Atreya A, Kanchan T. Clinico-epidemiological study of near-
might have impacted our results, we examined patients hanging cases - an investigation from Nepal. J Forensic Leg Med
who ultimately died from their injuries and separated 2015;33:35–8.
7. Penney DJ, Stewart AH, Parr MJ. Prognostic outcome indicators
this group into patients who had life support withdrawn following hanging injuries. Resuscitation 2002;54:27–9.
from patients who expired despite medical intervention. 8. Salim A, Martin M, Sangthong B, et al. Near-hanging injuries: a 10-
There were no significant differences noted apart from year experience. Injury 2006;37:435–9.
9. Martin MJ, Weng J, Demetriades D, et al. Patterns of injury and
significantly longer hospital and PICU lengths of stay functional outcome after hanging: analysis of the National Trauma
in the patients where life support was withdrawn, sug- Data Bank. Am J Surg 2005;190:836–40.
Pediatric Near-Hanging and Outcome 27

10. Davies D, Lang M, Watts R. Paediatric hanging and strangulation 18. Hsu CH, Haac B, McQuillan KA, et al. Outcome of suicidal hanging
injuries: a 10-year retrospective description of clinical factors and patients and the role of targeted temperature management in
outcomes. Paediatr Child Health 2011;16:e78–81. hanging-induced cardiac arrest. J Trauma Acute Care Surg 2017;
11. Kim MJ, Yoon YS, Park JM, et al. Neurologic outcome of comatose 82:387–91.
survivors after hanging: a retrospective multicenter study. Am J 19. Wahlen BM, Thierbach AR. Near-hanging. Eur J Emerg Med 2002;
Emerg Med 2016;34:1467–72. 9:348–50.
12. Matsuyama T, Okuchi K, Seki T, et al. Prognostic factors in hanging 20. Berdai AM, Labib S, Harandou M. Postobstructive pulmonary
injuries. Am J Emerg Med 2004;22:207–10. edema following accidental near-hanging. Am J Case Rep 2013;
13. Fiser DH. Assessing the outcome of pediatric intensive care. J Pe- 14:350–3.
diatr 1992;121:68–74. 21. Feigin G. Frequency of neck organ fractures in hanging. Am J
14. Hackett AM, Kitsko DJ. Evaluation and management of pediatric Forensic Med Pathol 1999;20:128–30.
near-hanging injury. Int J Pediatr Otorhinolaryngol 2013;77: 22. Nikolic S, Micic J, Atanasijevic T, et al. Analysis of neck injuries in
1899–901. hanging. Am J Forensic Med Pathol 2003;24:179–82.
15. Mansoor S, Afshar M, Barrett M, et al. Acute respiratory distress 23. Borgquist O, Friberg H. Therapeutic hypothermia for comatose sur-
syndrome and outcomes after near hanging. Am J Emerg Med vivors after near-hanging-a retrospective analysis. Resuscitation
2015;33:359–62. 2009;80:210–2.
16. Gantois G, Parmentier-Decrucq E, Duburcq T, et al. Prognosis at 6 24. Sinha A, Sivanandan S, Ramesh P, et al. Post obstructive pulmonary
and 12 months after self-attempted hanging. Am J Emerg Med edema in a child who attempted suicidal hanging. Indian J Pediatr
2017;35:1672–6. 2008;75:1075–7.
17. Nichols SD, McCarthy MC, Ekeh AP, et al. Outcome of cervical 25. Viswanathan S, Muthu V, Remalayam B. Pulmonary edema in near
near-hanging injuries. J Trauma 2009;66:174–8. hanging. J Trauma Acute Care Surg 2012;72:297–301.
28 S. La Count et al.

ARTICLE SUMMARY
1. Why is this topic important?
Near-hanging is a common injury in the pediatric pop-
ulation. The majority of research examining this injury
type has been conducted in adults.
2. What does this study attempt to show?
This study attempts to show key clinical factors that
may portend poor outcome in this patient population.
3. What are the key findings?
Pediatric patients with near-hanging injury tended to
dichotomize, ultimately either doing well or quite poorly.
In patients who presented with an initial Glasgow Coma
Scale score of 3T and prehospital cardiac arrest, there
was uniform poor neurologic outcome.
4. How is patient care impacted?
This study may influence how caregivers advise fam-
ilies regarding prognosis in this patient population.