The Laryngoscope

C 2016 The American Laryngological,

V
Rhinological and Otological Society, Inc.

Characterizing Mortality in Pediatric Tracheostomy Patients

Jamie L. Funamura, MD; Sonia Yuen, BA; Kosuke Kawai, ScD; Ozgul Gergin, MD; Eelam Adil, MD, MBA;
Reza Rahbar, DMD, MD; Karen Watters, MB, BCh, MPH

Objectives/Hypothesis: To assess the longitudinal risk of death following tracheostomy in the pediatric age group.
Study Design: Retrospective cohort study.
Methods: Hospital records of 513 children (18 years) at a tertiary care children’s hospital who underwent tracheosto-
my between 1984 and 2015 were reviewed. The primary outcome measure was time from tracheostomy to death. Secondary
patient demographic and clinical characteristics were assessed, with likelihood of death using v2 tests and the Cox propor-
tional hazards model.
Results: Median age at time of tracheostomy was 0.8 years (interquartile range, 0.3–5.2 years).The highest mortality
rate (27.8%) was observed in patients in the 13- to 18-year-old age category; their mortality rate was significantly higher
when compared to the lowest mortality risk group patients (age 1–4 years, P 5 .031). Timing of death was evenly distribut-
ed: <90 days (37.6%), 90 days to 1 year (27.1%), and >1 year after tracheostomy (35.3%). Patients who underwent trache-
ostomy for cardiopulmonary disease had an increased risk of mortality compared with airway obstruction (adjusted hazard
ratio: 3.53, 95% confidence interval: 1.72-7.24, P < .001) and other indications. Adjusted hazard ratios for bronchopulmonary
dysplasia (BPD) and congenital heart disease (CHD) were 2.63 and a 2.61, respectively (P < .001).
Conclusions: Pediatric patients with tracheostomy have a high mortality rate, with an increased risk of death associated
with a cardiopulmonary indication for undergoing tracheostomy. The majority of deaths occur after the index hospitalization
during which the tracheostomy was performed. BPD and CHD are independent predictors of mortality in pediatric tracheosto-
my patients.
Key Words: Pediatric tracheostomy, tracheostomy indication, congenital heart disease, bronchopulmonary dysplasia.
Level of Evidence: 4
Laryngoscope, 127:1701–1706, 2017

INTRODUCTION The mortality rate in pediatric tracheostomy

Pediatric tracheostomy is a valuable and often nec- patients is higher than that which is typical for most
essary procedure in children with airway obstruction, pediatric otolaryngology procedures. Patient age, comor-
chronic ventilator dependence, and need for improved bidities, and type of hospital in which the tracheostomy
pulmonary toilet. Because of advances in medical tech- was performed may have an impact on outcomes. A
nology over the last decade, tracheostomy is now per- national inpatient quality-improvement database study
formed on younger patients with a greater incidence of demonstrated that tracheostomy placement in children
chronic and complex medical issues.1–6 Pediatric under 2 years of age has the highest rate of major com-
patients who undergo tracheostomy today have a high plications within 30 days of surgery.12 The risk of death
mortality rate, ranging from 13% to 19%.1,3,6–10 Less during the hospitalization when the tracheostomy was
than 3% of this mortality is directly attributable to placed is double for children cared for in a nonpediatric
tracheostomy-related adverse events.1,4,6–10 These num- hospital versus in a dedicated children’s hospital, with
bers are comparable to reported adult tracheostomy corresponding lower mortality rates in hospitals with
patient mortality rates (22%).11 higher volumes of pediatric tracheostomy procedures.2
Increased in-hospital mortality has also been noted in
From the Department of Otolaryngology and Communication children under the age of 1 year, who were born prema-
Enhancement (J.L.F., S.Y., K.K., E.A., R.R., K.W.), Boston Children’s Hospital, turely, and have a history of congenital heart disease.13
Boston, Massachusetts, U.S.A.; Department of Otolaryngology (O.G.),
Umraniye Education and Research Hospital, Istanbul, Turkey; and the
The safety and timing of tracheostomy in pediatric
Department of Otology and Laryngology (K.K., E.A., R.R., K.W.), Harvard patients with congenital heart disease is of particular
Medical School, Boston, Massachusetts, U.S.A. concern. Despite a low rate of tracheostomy procedures
Editor’s Note: This Manuscript was accepted for publication in children following congenital heart surgery (1.2%–
September 9, 2016.
Presented at the 2016 American Society of Pediatric Otolaryngolo- 2.7%), history of prior cardiac surgery has been identi-
gy Spring Meeting, Chicago, Illinois, U.S.A., May 20–22, 2016. fied as an independent risk factor for mortality in
The authors have no funding, financial relationships, or conflicts infants within 1 year after tracheostomy.14,15 Although
of interest to disclose.
Send correspondence to Karen F. Watters, MB, BCh, MPH, 300 children with congenital heart disease have a higher
Longwood Avenue, LO-367, Boston, MA 02115. E-mail: karen.watters@ mortality rate following tracheostomy than children
childrens.harvard.edu
without, they rarely die during the index hospitalization;
DOI: 10.1002/lary.26361 in-hospital mortality is equivalent to that of children

Laryngoscope 127: July 2017 Funamura et al.: Mortality in Pediatric Tracheostomy Patients
1701
 TABLE I. children’s hospital was performed. Patient data spanning a 30-
Characteristics of Pediatric Tracheostomy Patients (n 5 513). year period, from January 1, 1984 through December 31, 2015,
were included in the review and analysis. Approval for the
Characteristics No. or Median % or IQR study was obtained from the Boston Children’s Hospital’s Insti-
tutional Review Board.
Age at tracheostomy, yr 0.8 0.3–5.2
Patient charts were reviewed for the following key varia-
Gender, female 216 42.1% bles: age at tracheostomy, gender, primary indication for trache-
Comorbidities ostomy, comorbidities (preterm birth, bronchopulmonary
Bronchopulmonary dysplasia 111 21.7% dysplasia, congenital heart disease, neurologic or neuromuscu-
Congenital heart disease 103 20.1% lar disease, known syndrome, neoplastic disease, severe system-
ic infection), additional medical technology beyond
Infection 26 5.1%
tracheostomy (e.g., gastrostomy tube, ventilatory support), date
Neoplasia 34 6.6% of procedure, last clinical contact, and mortality. The medical
Neurologic/neuromuscular 238 46.5% records were also evaluated for cause of death when applicable.
Syndrome 124 24.2% Neurologic disorders included cerebral palsy, hypoxic ischemic
Other 179 35.0% encephalopathy, and seizure disorders. Genetic syndromes and
Multiple comorbidities associations such as trisomy 21, VACTERL association or 22q
deletion were categorized as syndromes. Only severe systemic
0 or 1 292 57.0%
infections such as sepsis or documented infection affecting mul-
 2 comorbidities 220 43.0% tiple organ systems were included under the infection category.
Primary indication for tracheostomy Preterm birth was defined as less than 37 weeks gestation.
Airway obstruction 105 20.5% Index hospitalization was defined as the hospitalization in
Cardiopulmonary 179 34.9% which the tracheostomy was performed.
Study data were collected and managed using REDCap
Craniofacial 47 9.2%
(Research Electronic Data Capture) technology.19 (Data were
Neurologic 162 31.6% collected and managed using REDCap electronic data capture
Trauma 20 3.9% tools hosted at Vanderbilt University. REDCap is a secure, Web-
Preterm (<37 weeks) 218 45.0% based application designed to support data capture for research
Ventilation support 354 69.4% studies, providing: 1) an intuitive interface for validated data
Gastrostomy tube 287 56.0% entry; 2) audit trails for tracking data manipulation and export
procedures; 3) automated export procedures for seamless data
Year at tracheostomy
downloads to common statistical packages; and 4) procedures
<2004 80 15.6% for importing data from external sources.) Inclusion criteria
2005–2009 112 21.8% were presence of a tracheostomy and a record of otolaryngology
2010–2015 321 62.6% consultation, which is the standard practice for all tracheostomy
Length of follow-up, yr 3.5 0.9–7.0 patients at our institution. Exclusion criteria were greater than
18 years of age at time of tracheostomy and insufficient data to
Mortality 85 16.6%
designate the key variables as described above.
Tracheostomy-related mortality 3 0.6%
Patient demographic and clinical characteristics were
Mortality during initial hospitalization 29 5.7% assessed with likelihood of death. Kaplan-Meier method was
used to construct survival curves. Hazard ratio (HR) was esti-
IQR 5 interquartile range.
mated using the Cox proportional hazards models. Univariate
and multivariate analyses were performed, (P < .05). We
without congenital heart disease.14–18 Further investiga-
tion of these associations is warranted because there is a
nationwide trend toward performing more tracheosto-
mies in children with congenital heart disease.17
The current literature suggests that there may be
independent risk factors for mortality following pediatric
tracheostomy.12–15 Further characterization of these
risks may be difficult to capture through national inpa-
tient databases that can be limited by coding error bias
and inability to follow individual patients longitudinally.
Conversely, longitudinal data analyzing the impact of
comorbidities on mortality in pediatric tracheostomy
patients have been limited to smaller case series. The
aim of this study is to compare the impact of specific
comorbidities on long-term mortality rates in a large
pediatric cohort that has undergone tracheostomy at a
tertiary pediatric hospital.

MATERIALS AND METHODS Fig. 1. Age distribution (years) of patients at time of tracheostomy.
A retrospective cohort study with chart review of all pedi- [Color figure can be viewed in the online issue, which is available
atric (0–18 years old) tracheostomy patients at a tertiary care at www.laryngoscope.com.]

Laryngoscope 127: July 2017 Funamura et al.: Mortality in Pediatric Tracheostomy Patients
1702
 TABLE II.
Predictors of Mortality in Pediatric Tracheostomy Patients.

Unadjusted HR Adjusted HR
No./Total % (95% CI) P (95% CI) P

Age category
<1 year 37/279 13.3% 0.96 (0.53-1.74) .89 0.83 (0.44-1.53) .54
1–4 years 15/100 15.0% Reference Reference
5–12 years 18/80 22.5% 1.54 (0.78-3.06) .22 1.89 (0.94-3.83) .076
13–18 years 15/54 27.8% 2.20 (1.08-4.51) .031 2.73 (1.26-5.91) .011
Comorbidities
Bronchopulmonary dysplasia 33/111 29.7% 2.93 (1.88-4.55) <.001 2.63 (1.60-4.32) <.001
Congenital heart disease 28/103 27.2% 2.63 (1.66-4.16) <.001 2.61 (1.51-4.51) <.001
Infection 15/26 57.7% 7.42 (4.20-13.11) <.001 4.21 (2.24-7.93) <.001
Neoplasia 10/34 29.4% 2.46 (1.27-4.77) .008 2.43 (1.19-4.97) .015
Neurologic 39/238 16.4% 0.86 (0.56-1.32) .49
Syndrome 23/124 18.5% 1.11 (0.69-1.80) .67
Other 37/179 20.7% 1.57 (1.02-2.42) .039
Multiple comorbidities
0 or 1 24/292 8.2% Reference Reference
2 61/220 27.7% 5.57 (2.23-13.92) <.001 2.03 (1.16-3.54) .013
Primary indication
Airway obstruction 9/105 8.5% Reference Reference
Cardiopulmonary 48/179 26.8% 3.79 (1.86-7.74) <.001 3.53 (1.72-7.24) <.001
Craniofacial 4/47 8.5% 0.87 (0.27-2.83) .82 0.76 (0.23-2.49) .65
Neurologic 22/162 13.6% 1.40 (0.65-3.05) .39 1.07 (0.48-2.38) .87
Trauma 2/20 10.5% 1.03 (0.22-4.78) .97 0.73 (0.15-3.43) .69
Ventilation support
Yes 71/354 20.1% 2.50 (1.41-4.44) .002
No 14/156 9.0% Reference
Gestational age
Preterm (<37 weeks) 30/218 13.8% 0.77 (0.48-1.21) .25
Term (37 weeks) 46/267 17.2% Reference
Gender
Female 35/216 16.2% 0.96 (0.62-1.48) .96
Male 50/297 16.8% Reference
Year category
2004 4/80 5.0% 0.04 (0.01-0.14) <.001
2005–2009 18/112 16.1% 0.38 (0.21-0.70) .002
2010–2015 63/321 19.6% Reference

Hazard ratio (HR) and 95% confidence interval (CI) were estimated from Cox proportional hazard model. HR corresponding to each comorbidity is com-
paring the risk of mortality in patients with comorbidity versus without comorbidity. Adjusted HRs were based on the multivariate model that includes age,
bronchopulmonary dysplasia, congenital heart disease, infection, neoplasia, and multiple comorbidities. We examined primary indication category in the sepa-
rate multivariate model that includes age.

initially considered any variables with P < .20 in the univariate 0.3–5.2 years) (Fig. 1). Cardiopulmonary disease was the
analysis as candidates for the multivariate model, and retained most common primary indication for tracheostomy
variables with P < .05 in the final multivariate model. For risk (34.9%), followed by neurologic and neuromuscular dis-
of death during the index hospitalization, we used multivariable orders (31.6%). Forty-three percent of patients had more
(logistic regression) analysis. All analyses were conducted using than one comorbidity. The majority of tracheostomies
SAS version 9.3 (SAS Institute, Cary, NC).
(62.6%) were performed in the last 5 years (2010–2015),
reflecting the increased numbers and complexity of our
RESULTS patient population from an expanding international
A total of 513 patients who had a tracheostomy referral base. The overall mortality rate was 16.6% (85
placed met study criteria. Patient characteristics are patients), with a median length of follow-up of 3.5 years
summarized in Table I. The median age at time of (IQR, 0.7–4.0 years). Three of the 85 (3.5%) deaths were
tracheostomy was 0.8 years (interquartile range [IQR], attributable to tracheostomy complications, including

Laryngoscope 127: July 2017 Funamura et al.: Mortality in Pediatric Tracheostomy Patients
1703
Fig. 2. Risk of mortality by primary indication for tracheostomy Fig. 4. Risk of mortality with and without congenital heart disease
(log rank test, P < .001). (CHD).

mucous plugging or accidental decannulation. Overall, group of 1 to 4 years old (adjusted HR: 2.73, 95% confi-
there was a 0.6% rate of tracheostomy-related adverse dence interval [CI]: 1.26-5.91, P 5 .011). Details of the
events resulting in death among all tracheostomy comorbid conditions of the oldest age group are depicted
patients. Twenty-nine of 85 (34.1%) deaths occurred dur- in Table III. Patients who underwent tracheostomy for
ing the index hospitalization. Overall, the timing of cardiopulmonary disease had an increased risk of mor-
death was evenly distributed between age groups: <90 tality compared with airway obstruction (adjusted HR:
days (37.6%), 90 days to 1 year (27.1%), and >1 year 3.53, 95% CI: 1.72–7.24, P < .001) and other indications.
after tracheostomy (35.3%). Gender and preterm birth were not associated with a
Older age and cardiopulmonary disease as a prima- higher risk of mortality. Higher mortality rates were
ry indication for tracheostomy were significantly associ- observed among patients undergoing tracheostomy in
ated with an increased risk of mortality in the the past 5 years compared with the preceding intervals
multivariate Cox model (Table II; Figs. 2–4). Similarly, studied (2004, 2005–2009; P < .001, P 5 .002).
comorbidities such as bronchopulmonary dysplasia, con- Subgroup analysis was performed for tracheostomy
genital heart disease, systemic infection, neoplastic dis- patients who died during the index hospitalization
ease, as well as combinations of more than one of these (Table IV). Older age was not significantly associated
comorbidities were also significantly associated with with higher mortality risk. A history of bronchopulmo-
increased mortality risk. The highest mortality rate nary dysplasia, neoplasia, severe systemic infection, or
(27.8%) was observed in patients 13 to 18 years old, multiple comorbidities was significantly associated with
which was significantly greater than the lowest risk a higher risk of death during the index hospitalization

TABLE III.
Details of Comorbidities in 13- to 18-Year-Old Patients.
Significant Comorbid Conditions No.

Cystic fibrosis with complications of lung 6 liver 5

transplant (acute and chronic rejection, post-
transplant lymphoproliferative disorder, respiratory
failure)
Pulmonary hypertension with complications of lung 2
transplantation (chronic rejection)
Leukemia/lymphoma with complications of bone marrow 2
transplant (multiorgan system failure, graft versus host
disease, bronchiolitis obliterans organizing pneumonia
requiring lung transplantation)
Sequelae of muscular dystrophy (dilated cardiomyopa- 2
thy, respiratory failure)
High cervical spine injury (gunshot wound, cord com- 2
pression from neurofibroma and progressive spinal
deformity)
Pulmonary metastases (epithelioid sarcoma) 1
Fig. 3. Risk of mortality with and without bronchopulmonary dys- Seizure disorder 1
plasia (BPD).

Laryngoscope 127: July 2017 Funamura et al.: Mortality in Pediatric Tracheostomy Patients
1704
 TABLE IV.
Predictors of Mortality During Index Hospitalization in Tracheostomy Patients.
No./Total % Crude OR (95% CI) P

Age category
<1 year 15/279 5.4% 2.78 (0.63-12.40) .18
1–4 years 2/100 2.0% Reference
5–12 years 7/80 8.8% 4.70 (0.95-23.28) .058
13–18 years 5/54 9.3% 5.00 (0.94-26.70) .059
Comorbidities
Bronchopulmonary dysplasia 17/111 15.3% 5.86 (2.71-12.69) <.001
Congenital heart disease 9/103 8.7% 1.86 (0.82-4.22) .14
Infection 5/26 19.2% 4.58 (1.59-13.20) .005
Neoplasia 7/34 20.6% 5.37 (2.11-13.69) <.001
Neurologic 9/238 3.8% 0.50 (0.22-1.12) .091
Syndrome 9/124 7.3% 1.44 (0.64-3.25) .38
Other 10/179 5.6% 0.98 (0.45-2.15) .96
Multiple comorbidities
0 or 1 6/292 2.1% Reference
2 23/220 10.5% 5.57 (2.23-13.92) <.001
Indication
Airway obstruction 4/105 3.8% Reference
Cardiopulmonary 21/179 11.7% 3.36 (1.11-10.06) .031
Craniofacial 1/47 2.1% 0.55 (0.06-5.05) .60
Neurologic 3/162 1.9% 0.48 (0.10-2.17) .34
Trauma 0/20 0% NA
Ventilation support
Yes 26/354 7.3% 4.04 (1.21-13.56) .024
No 3/156 1.9% Reference

Crude odds ratio (OR) and 95% confidence interval (CI) were estimated from the logistic regression model. OR corresponding to each comorbidity com-
pares the mortality in patients with comorbidity versus without comorbidity.
NA 5 not applicable.

(odds ratio [OR]: 5.86, 5.37, 4.58, 5.57, respectively); death. In doing so, we were able to collect data on
however, congenital heart disease was not (OR: 1.86; patients who died after a short defined period (such as
95% CI: 0.82-4.22). A primary indication of cardiopulmo- <30 days postoperatively or during the index hospitali-
nary disease for tracheostomy and ongoing need for ven- zation), information that is limited in previous stud-
tilator support were also associated with a higher ies.12,13 This is important because a large proportion of
mortality risk (P 5 .031, P 5 .024, respectively). deaths occurred remotely from the time of tracheostomy
(62.4% >90 days, 35.3% >1 year after tracheostomy)
and after discharge from the index hospitalization
DISCUSSION (65.9%).
This study complements the existing literature The highest risk of death over time in this study
regarding mortality in pediatric tracheostomy patients was found in patients who underwent tracheostomy for
by reporting longitudinal outcomes and comorbidities in the primary indication of cardiopulmonary disease. A
a large pediatric cohort. We report an overall mortality significantly decreased survival probability according to
rate of 16.6%, which is consistent with prior studies the primary indication for tracheostomy has not been
(13%–19%), and demonstrates the high mortality rate in previously demonstrated in the literature. It has been
this patient population.1,3,6–10 A tracheostomy-related noted that the primary indication for tracheostomy has
mortality rate of 0.6% is also comparable to rates (0.7%– been difficult to extract from national registries.13
3.6%) reported in the recent literature.1,4,6–10 The low Therefore, further large case series from similar tertiary
rate of tracheostomy-related mortality is favorable, but referral pediatric hospitals should address this topic to
effort should still be made to improve tracheostomy care support or refute our findings.
and education for caregivers. The relatively high overall Our study demonstrates that bronchopulmonary
mortality rate in children with tracheostomies highlights disease, congenital heart disease, severe systemic infec-
the need to determine specific risk factors for death. tion, and neoplasia are independent risk factors for mor-
In this study we were able to longitudinally follow tality. It is reasonable to assume that a severe acute
patients from the time of tracheostomy to the time of infection or a neoplastic disease process would increase

Laryngoscope 127: July 2017 Funamura et al.: Mortality in Pediatric Tracheostomy Patients
1705
risk of mortality. However, bronchopulmonary disease Acknowledgment
and congenital heart disease affect an increasing propor- We are grateful to David Roberson, MD, for his support
tion of children with tracheostomies, and therefore, fur- and role in the conception of this study.
ther studies are necessary to examine why patients with
these comorbidities have a higher mortality rate. In
CONCLUSION
addition, it would be worthwhile to compare mortality
Mortality in pediatric patients with tracheostomy is
rates between patients with these comorbidities with
high. There is an increased risk of death associated with
and without tracheostomy. The increased mortality rate
a cardiopulmonary disease as an indication for undergo-
in patients with bronchopulmonary disease or congenital
ing tracheostomy. Bronchopulmonary dysplasia and con-
heart disease in our study is similar to rates in a nation-
genital heart disease are independent predictors of
wide database study demonstrating increased mortality
mortality in pediatric tracheostomy patients. Further
in children who were born prematurely or with congeni-
studies may improve patient care outcomes in this high-
tal heart disease compared to those who did not have
these comorbid conditions.13 risk population.
Further study of our tracheostomy outcomes in chil-
dren with congenital heart disease may be warranted. A BIBLIOGRAPHY
review of the Kids’ Inpatient Database (1997–2009) dem- 1. Carron JD, Derkay CS, Strope GL, Nosonchuk JE, Darrow DH. Pediatric
onstrated that 9.6% of all tracheostomies were per- tracheotomies: changing indications and outcomes. Laryngoscope 2000;
110:1099–1104.
formed in children with congenital heart disease, with 2. Lewis CW, Carron JD, Perkins JA, Sie KC, Feudtner C. Tracheotomy in
an increasing proportion in each study year.17 Single- pediatric patients: a national perspective. Arch Otolaryngol Head Neck
Surg 2003;129:523–529.
institution studies support a poorer outcome for children 3. Zenk J, Fyrmpas G, Zimmermann T, Koch M, Constantinidis J, Iro H. Tra-
with congenital heart disease and tracheostomy com- cheostomy in young patients: indications and long-term outcome. Eur
Arch Otorhinolaryngol 2009;266:705–711.
pared with tracheostomy alone: mortality rates range 4. Shah RK, Lander L, Berry JG, et al. Tracheotomy outcomes and complica-
from 23.5 to 52.4%.14–18,20,21 Even more specifically, sin- tions: a national perspective. Laryngoscope 2012;122:25–29.
gle compared to biventricular physiology has been 5. Liu C, Heffernan C, Saluja C, et al. Indications, hospital course, and com-
plexity of patients undergoing tracheostomy at a tertiary care pediatric
observed in a case series to have a higher mortality rate hospital. Otolaryngol Head Neck Surg 2014;151:232–239.
(64% vs. 33.3%).16 The mortality rate for children with 6. Funamura JL, Durbin-Johnson B, Tollefson TT, Harrison J, Senders CW.
Pediatric tracheotomy: indications and decannulation outcomes. Laryn-
congenital heart disease and tracheostomy was similarly goscope 2014;124:1952–1958.
high (27.2%) in this study. These higher mortality rates 7. Butnaru CS, Colreavy MP, Ayari S, Froehlich P. Tracheotomy in children:
evolution in indications. Int J Pediatr Otorhinolaryngol 2006;70:115–
could be attributed to survival of a complex cardiac 119.
patient who would not have survived without recent 8. Carr MM, Poje CP, Kingston L, Kielma D, Heard C. Complications in pedi-
medical and surgical advances. If this is the case, it is atric tracheostomies. Laryngoscope 2001;111:1925–1928.
9. Mahadevan M, Barber C, Salkeld L, Douglas G, Mills N. Pediatric trache-
even more imperative to further characterize this otomy: 17 year review. Int J Pediatr Otorhinolaryngol 2007;71:1829–
patient population in future studies. 1835.
10. Ozmen S, Ozmen OA, Unal OF. Pediatric tracheotomies: a 37-year experi-
The results of this study differed from prior studies ence in 282 children. Int J Pediatr Otorhinolaryngol 2009;73:959–961.
with regard to the age group with the highest mortality 11. Halum SL, Ting JY, Plowman EK, et al. A multi-institutional analysis of
tracheotomy complications. Laryngoscope 2012;122:38–45.
rate. Previous work suggests that the youngest patients, 12. Mahida JB, Asti L, Boss EB, et al. Tracheostomy placement in children
<1 or 2 years of age, have the highest mortality younger than 2 years: 30-day outcomes using the National Surgical
rate.12,13 In this study, the highest mortality rate was in Quality Improvement Program Pediatric. JAMA Otolaryngol Head Neck
Surg 2016;142:241–246.
the 13- to 18-year-old age group. When these deaths 13. Berry JG, Graham RJ, Roberson DW, et al. Patient characteristics associ-
were examined closely (Table III), it became apparent ated with in-hospital mortality in children following tracheotomy. Arch
Dis Child 2010;95:703–710.
that the majority of these patients had terminal diagno- 14. Rosner E, Mastropietro CW. Prior cardiac surgery is independently associ-
ses. At our institution, there are large cancer, cystic ated with decreased survival following infant tracheostomy. Respir Care
2015;60:47–55.
fibrosis, and lung transplant programs, and our results 15. Costello JP, Emerson DA, Shu MK, et al. Outcomes of tracheostomy fol-
in this age category are likely reflective of a referral lowing congenital heart surgery: a contemporary experience. Congenit
bias. It also demonstrates a palliative tracheostomy Heart Dis 2015;10:e26–e29.
16. Cotts T, Hirsch J, Thorne M, Gajarski R. Tracheostomy after pediatric car-
practice that may not be a philosophy at other diac surgery: frequency, indications, and outcomes. J Thorac Cardiovasc
institutions. Surg 2011;141:413–418.
17. Maxwell BG, McMillan KN. Tracheostomy in children with congenital
We recognize that there are inherent limitations to heart disease: a national analysis of the Kids’ Inpatient Database. PeerJ
our study, in particular, this is a single-institution retro- 2014;2:e568.
18. Challapudi G, Natarajan G, Aggarwal S. Single-center experience of out-
spective study. We are also subject to a referral bias comes of tracheostomy in children with congenital heart disease. Conge-
within our institution, which may affect the generaliz- nit Heart Dis 2013;8:556–560.
19. Harris PA, Taylor R, Thielke R, Payne J, Gonzalez N, Conde JG. Research
ability of our results. However, the mortality rates and electronic data capture (REDCap)—a metadata-driven methodology and
independent risk factors for death in our study are con- workflow process for providing translational research informatics sup-
sistent with those reported in the literature. Looking for- port. J Biomed Inform 2009;42:377–381.
20. Alibrahim IJ, Kabbani MS, Abu-Sulaiman R, Al-Akhfash A, Mazrou KA.
ward, ongoing study of both national registries and large Outcome of tracheostomy after pediatric cardiac surgery. J Saudi Heart
case series will be needed to come to a complete under- Assoc 2012;24:163–168.
21. Hoskote A, Cohen G, Goldman A, Shekerdemian L. Tracheostomy in
standing of the pediatric tracheostomy patient popula- infants and children after cardiothoracic surgery: indications, associated
tion of today. risk factors, and timing. J Thorac Cardiovasc Surg 2005;130:1086–1093.

Laryngoscope 127: July 2017 Funamura et al.: Mortality in Pediatric Tracheostomy Patients
1706