- Email: [email protected]
Pediatric tracheostomy: Survival and long-term outcomes
International Journal of Pediatric Otorhinolaryngology 89 (2016) 81e85
Contents lists available at ScienceDirect
International Journal of Pediatric Otorhinolaryngology journal homepage: http://www.ijporlonline.com/
Pediatric tracheostomy: Survival and long-term outcomes Norihiko Tsuboi a, *, Kentaro Ide a, Nao Nishimura a, Satoshi Nakagawa a, Noriko Morimoto b a b
Critical Care Medicine, National Center for Child Health and Development, Tokyo, Japan Otorhinolaryngology, National Center for Child Health and Development, Tokyo, Japan
a r t i c l e i n f o
a b s t r a c t
Article history: Received 17 May 2016 Received in revised form 26 July 2016 Accepted 26 July 2016 Available online 28 July 2016
Objectives: The objective of this study was to investigate if there were any differences in survival and long-term outcomes between pediatric patients with and without neurological impairment who underwent tracheostomy. Methods: A retrospective chart review of pediatric patients (age 0e15 years) who underwent tracheostomy between March 2002 and December 2013 was conducted. Patients were categorized into two groups: those who were neurologically impaired (NI) (pediatric cerebral performance category, 3e6) and those who were not neurologically impaired (NN) (pediatric cerebral performance category, 1e2). Survival rates and cumulative incidence of weaning from mechanical ventilation or decannulation were calculated using the Kaplan-Meier method. Results: A total of 212 patients were included. Among them, 141 were categorized into NI group and 71 into NN group. Between the two groups, there were no significant differences in survival rates and cumulative incidence of weaning from mechanical ventilation. In total patients, one-year survival rate was 0.86 (95%CI 0.80e0.90) and five-year survival rate was 0.71 (0.62e0.78). One-year weaning rate was 0.58 (0.51e0.65) and five-year weaning rate was 0.66 (0.59e0.74). Decannulation rates were significantly lower in NI group than in NN group (p < 0.001). One-year and five-year decannulation rates were 0.04 (0.01e0.09) and 0.17 (0.10e0.29), respectively, in NI group, and 0.20 (0.12e0.33) and 0.54 (0.40e0.69), respectively, in NN group. Conclusions: In children who underwent tracheostomy, the decannulation rate was lower in those with neurological impairment compared with that in those without neurological impairment. There were no significant differences in survival or ventilator weaning between the two groups. © 2016 Elsevier Ireland Ltd. All rights reserved.
Keywords: Tracheostomy Pediatric intensive care Weaning Decannulation Survival rate Neurological impairment
1. Introduction The frequency and indications of pediatric tracheostomy have undergone notable changes over the last 30 years, and pediatric tracheostomy has recently become a relatively common procedure with approximately 5000 procedures performed in the United States each year [1]. Pediatric patients who require tracheostomy tend to be younger and with chronic diseases [1e7]. The current literature on pediatric tracheostomy contains limited objective data, especially those that describe Kaplan-Meier estimates of survival and long-term outcomes after pediatric tracheostomy. The aim of this study was to investigate if there are any
differences in survival and long-term outcomes between pediatric patients with and without neurological impairment who underwent tracheostomy. We hypothesized that the long-term outcomes would be worse in patients with neurological impairment regardless of their comorbidities. Thus, we categorized our pediatric patients into two groups using the pediatric cerebral performance category (PCPC) as those who were neurologically impaired (NI) (PCPC, 3e6) and those who were not neurologically impaired (NN) (PCPC, 1e2) at discharge from the pediatric intensive care unit (PICU). We evaluated survival and long-term outcomes by using the Kaplan-Meier method.
2. Materials and methods * Corresponding author. Critical Care Medicine, National Center for Child Health and Development, 2-10-1 Okura, Setagaya, Tokyo, 157-8535, Japan. E-mail address: [email protected] (N. Tsuboi). http://dx.doi.org/10.1016/j.ijporl.2016.07.033 0165-5876/© 2016 Elsevier Ireland Ltd. All rights reserved.
This is a retrospective observational study at the PICU of the National Center for Child Health and Development (NCCHD) in
82
N. Tsuboi et al. / International Journal of Pediatric Otorhinolaryngology 89 (2016) 81e85
Tokyo, Japan. This study was approved by the ethics committee of our institution (receipt number 1003). 2.1. Patients We enrolled all the patients aged under 16 years old who underwent tracheostomy between March 2002 and December 2013, and who were admitted to the PICU at the NCCHD. We extracted these patients from the operation register and the PICU register. Medical records were reviewed and the following data were recorded: intubation days before tracheostomy, disposition before tracheostomy, age at tracheostomy, primary diagnosis, follow-up duration after tracheostomy, PCPC at discharge from the PICU, occurrence of an event of weaning from mechanical ventilation, decannulation, death after tracheostomy, and the period from the tracheostomy until weaning from mechanical ventilation or decannulation. Outcomes were survival rates and cumulative incidence of weaning from mechanical ventilation or decannulation after tracheostomy. We estimated these outcomes at the end of June 2015. The unit is a multivalent PICU composed of 20 beds at a tertiary child hospital. 2.2. Tracheostomy procedure and postoperative management All tracheostomies were performed by otorhinolaryngologists at our institution. In general, systemic management was applied in the PICU during the postoperative acute phase (10e14 days after tracheostomy). Weaning from ventilation was achieved when patients could breathe spontaneously without hypercapnia. Decannulation was considered when all of the following preconditions were satisfied: restoration of the voice, ability to swallow, and no demand for oxygen. Before decannulation, the cannula was exchanged for one that is of smaller size to ensure adequate air circulation, and velopharyngeal training was given when necessary. 2.3. PCPC scale The PCPC scale was developed to describe the short-term outcome of pediatric intensive care by quantifying cognitive impairment8. It is a six-point graded scale of increasing disability from normal function (score ¼ 1) to death (score ¼ 6) (Table 1). A child with Down syndrome is usually classified as PCPC 3. Scale score has been found to be significantly associated with several measures of morbidity, including length of stay in the PICU, total hospital charges, discharge care needs, and the Pediatric Risk of Mortality Score [8]. 2.4. Statistical analyses Patients who underwent tracheostomy were categorized into two groups: NI (PCPC, 3e6) and NN (PCPC, 1e2). Patients' PCPC
scores were evaluated at discharge from the PICU by clinicians of the unit. Descriptive statistics were expressed as medians and interquartile ranges for continuous data, or absolute frequencies and percentage for categorical data. Bivariate comparisons between groups were performed using ManneWhitney U for continuous data, and Fisher exact for categorical data. Ninety-five percent confidence intervals (95% CI) were also calculated for relevant outcomes. Survival rates were calculated using the Kaplan-Meier method and cumulative incidence of weaning from mechanical ventilation or decannulation were demonstrated as one minus the Kaplan-Meier survival method. Comparisons between groups were performed using the log-rank test. The potential confounding factors were adjusted by Cox proportional hazards model. Two-sided p values less than 0.05 were considered statistically significant. All analyses were performed using R version 3.2.0.
3. Results A total of 212 patients out of 10,250 patients in the PICU during the study period were included. Among the included patients, 141 were categorized into NI group (PCPC, 3e6) and 71 into NN group (PCPC, 1e2). The characteristics of the two groups are shown in Table 2. The age at tracheostomy was significantly older in NI group than in NN group (p < 0.001). In total patients, 142 patients were followed-up for more than one year, and 55 patients for more than five years. Deaths were reported in 47 patients. Causes of death were progression of the underlying disease in 25 patients, sepsis in five, airway-related causes in four, pneumonia in four, and other causes in nine. Kaplan-Meier estimates of survival rates and cumulative incidence of weaning from mechanical ventilation or decannulation are summarized in Figs. 1e3. Between the two groups, there were no significant differences in survival rates and cumulative incidence of weaning from mechanical ventilation. In total patients, one-year survival rate was 0.86 (95%CI 0.80e0.90) and five-year survival rate was 0.71 (0.62e0.78). One-year weaning rate was 0.58 (0.51e0.65) and five-year weaning rate was 0.66 (0.59e0.74). Decannulation rates were significantly lower in NI group than in NN group (p < 0.001). In NI group, one-year and five-year decannulation rates were 0.04 (0.01e0.09) and 0.17 (0.10e0.29), respectively, and in NN group, 0.20 (0.12e0.33) and 0.54 (0.40e0.69), respectively. When adjusting for age in the Cox proportional hazards model, decannulation rates remained significantly lower in NI group than in NN group (p < 0.001). Thirty patients received additional airway surgical interventions under general anesthesia after tracheostomy to achieve or attempt decannulation. Among them, there were 21 cases of resection of tracheal granuloma, two cases of resection of upper airway tumor, and two cases of glossectomy. Bronchial foreign body removal, tracheoesophageal fistula repair, anterior tracheal wall suspension,
Table 1 Pediatric cerebral performance Category scale. Score Category 1 2 3 4 5 6
Normal Mild disability
Description
Normal; at age-appropriate level; school-age child attending regular school classroom Conscious, alert, and able to interact at age-appropriate level; school-age child attending regular school classroom, but grade perhaps not appropriate for age; possibility of mild neurologic deficit Moderate disability Conscious; sufficient cerebral function for age-appropriate independent activities of daily life; school-age child attending special education classroom and/or learning deficit present Severe disability Conscious; dependent on others for daily support because of impaired brain function Coma or vegetative Any degree of coma without the presence of all brain death criteria; unaware, even if awake in appearance, without interaction with state environment; cerebral unresponsiveness and no evidence of spontaneous eye-opening, and sleep-wake cycles Brain death Apnea, areflexia, and/or electroencephalographic silence
N. Tsuboi et al. / International Journal of Pediatric Otorhinolaryngology 89 (2016) 81e85
83
Table 2 Patient characteristics. NI (n ¼ 141) Age, month Disposition before tracheostomy Pediatric intensive care unit Neonatal intensive care unit General ward Home Intubation days before tracheostomy Follow-up duration, day Primary diagnosis Neurological disease Malformation syndrome Airway obstruction Craniofacial disease Trauma Pulmonary disease Metabolic disease Tumor Heart disease Skeletal disorder Liver failure Muscular disease Others
NN (n ¼ 71)
p
24
(8e81)
7
(5e17)
<0.001
119 13 4 5 21 888
84% 9% 3% 4% (15e58) (99e1915)
58 8 4 1 21 918
82% 11% 6% 1% (11e64) (215e1839)
0.696 0.633 0.446 0.666 0.479 0.567
74 18 11 7 7 5 5 4 4 2 1 1 2
52% 13% 8% 5% 5% 4% 4% 3% 3% 1% 1% 1% 1%
6 4 12 5 0 5 2 12 6 5 9 2 3
8% 6% 17% 7%
<0.001 0.152 0.060 0.541 0.098 0.308 1 <0.001 0.088 0.043 <0.001 0.260
7% 3% 17% 8% 7% 13% 3% 4%
NI: neurologically impaired group (PCPC, 3e6). NN: non-neurologically impaired group (PCPC, 1e2). Bivariate comparisons between groups were performed using ManneWhitney U for continuous data, and Fisher exact for categorical data. Medians and (interquartile ranges) are expressed for quantitative variables.
Fig. 1. Survival rates after tracheostomy. NI: neurologically impaired group (PCPC, 3e6). NN: non-neurologically impaired group (PCPC, 1e2).
Fig. 3. Cumulative incidence of decannulation after tracheostomy. NI: neurologically impaired group (PCPC, 3e6). NN: non-neurologically impaired group (PCPC, 1e2).
dissection of epiglottis adhesion, sclerotherapy of cervical lymphangioma, tracheal scar excision, and tracheal laser therapy were performed once respectively in this series. Neither tracheal balloon dilation nor laryngotracheal reconstruction with grafts were performed. 4. Discussion
Fig. 2. Cumulative incidence of weaning from mechanical ventilation after tracheostomy. NI: neurologically impaired group (PCPC, 3e6). NN: non-neurologically impaired group (PCPC, 1e2).
We reported the survival and long-term outcomes of patients after pediatric tracheostomy. Our findings were consistent with current reports and provide further information on long-term outcomes after pediatric tracheostomy [2e7,9,10]. Funamura et al. demonstrated a difference in the overall decannulation rates among pediatric patients undergoing tracheostomy for different indications [2]. In the study, pediatric tracheostomy indications were categorized into five groups according to the underlying causes: cardiopulmonary, craniofacial anomalies, neurological impairment, traumatic injury and upper airway obstruction. In other studies, similar categorizations of
84
N. Tsuboi et al. / International Journal of Pediatric Otorhinolaryngology 89 (2016) 81e85
pediatric tracheostomy indications have been adopted [1,3,7,9]. While these classifications are helpful, it is at times difficult to clearly classify all patients into these categories based on their comorbidities, such as those with both craniofacial anomalies and neurological impairment. Thus, we simply categorized our patients into two groups using the PCPC as those having neurological impairment and those without at discharge from the PICU. 4.1. Survival rates Recent reports on tracheostomy demonstrated a mortality rate of 8e20% [1e7,9,10], with the majority of deaths caused by a progression of underlying diseases. In our study, one-year mortality rate was 14% and five-year mortality rate was 29%. About half of the causes of death were progression of the underlying diseases, with the number increasing yearly. It is difficult to compare our mortality rate with those of previous reports because they did not use the Kaplan-Meier method to calculate the survival rates, nor did they state the individual follow-up duration after tracheostomy. In our study, four deaths due to airway-related causes are reported: two accidental tube removal and two tube obstruction. All four events occurred at home. To avoid accidental and fatal complication, it is necessary to provide sufficient instruction and to support the parents or guardians in terms of the management of tracheal cannula. To that end, we provide an education program prior to discharge for parents and guardians, including basic cardiopulmonary resuscitation, bag-valve mask ventilation, tracheal cannula exchange, and airway suctioning. 4.2. Weaning rates We considered that weaning from mechanical ventilation was one of the long-term outcomes after pediatric tracheostomy. Thus, we showed the incidence of weaning from mechanical ventilation calculated as one minus the Kaplan-Meier survival method. In 25 out of 123 cases that could wean from mechanical ventilation, the primary reason for tracheostomy was “difficulty in weaning from mechanical ventilation”. Among the 25 cases, 22 had pulmonary causes, two had cardiac causes and one had muscular cause. We speculate that these patients were able to wean off mechanical ventilation due to the various benefits that surgery yielded, including reduced effort in breathing, reduced dead space, easier elimination of pharyngeal secretions from the respiratory tract, reduced use of sedatives, and recovery of respiratory function owing to active rehabilitation or physical development. In contrast, previous pediatric reports did not show the weaning data in detail. 4.3. Decannulation rates Several recent reports which did not use the Kaplan-Meier method demonstrated 17e75% decannulation rates after tracheostomy, and showed lower deccanulation rates in patients with neurological impairment [3e7,9,10,12]. On the other hand, Leung and Berkowitz used the Kaplan-Meier method and reported that patients who had undergone tracheostomy for tracheobronchial toilet had a significantly shorter time to decannulation [11]. Funamura et al. also demonstrated higher decannulation rates and shorter time to decannulation in children undergoing tracheostomy for maxilla facial and laryngotracheal trauma compared with children undergoing tracheostomy for cardiopulmonary and neurological indications by using the Kaplan-Meier survival method [2]. We demonstrated lower deccanulation rates in patients with neurological impairment by using the Kaplan-Meier method. Higher decannulation rate of the patients without neurological impairment, which increased yearly, suggests that
those without neurological impairment have the potential to be decannulated as they become older regardless of their chronic respiratory problems. For neurologically impaired patients, the key factors of decannulation are the ability to wean from the mechanical ventilation and to maintain airway patency. We postulate that this is the reason why weaning rates were not significantly different between the two groups, as opposed to decannulation rates. In this study, there were 21 patients who were admitted to the neonatal intensive care unit (NICU) before receiving tracheostomy. Their primary diagnosis were five cases of neurological disease, four cases of craniofacial disease, three cases of malformation syndrome, tumor, or skeletal disease, two cases of airway obstruction, and a case of muscular disease. The 21 patients from the NICU had lower survival rates (p < 0.001), lower weaning rates (p ¼ 0.026), and lower decannulation rates (p ¼ 0.034), compared with the other patients. There were no significant differences in the proportion of patients from the NICU between NI and NN groups. This study has several limitations. First, there could be other factors that affected the survival and long-term outcomes. Second, the patients' PCPC scale was evaluated at discharge from the PICU. Some patients who were temporarily impaired upon discharge from the PICU may have made a neurological recovery later. Third, during the study period, there were a few patients who underwent tracheostomy but were not admitted to the PICU. Thus, the conclusions may not reflect “all” pediatric patients who underwent tracheostomy. Fourth, this was a single-center study. Therefore, there may be a risk of bias in patient selection. However, since our institution is a major pediatric tertiary hospital, the patient series presented in this study was likely to be a good representation of the distribution of cases in other pediatric tertiary hospitals. Wood et al. reported that the incidence of pediatric tracheostomies over five years among PICUs in 29 centers in Britain was 2.1% (1613 tracheostomy cases out of 78,504 PICU admissions), with infants under one year accounting for 46% [13]. Consistent with that study, the pediatric tracheostomy rate in our study was 2.1% (212/10,250), with 48% (101/212) of the patients being under one year old. 5. Conclusions By using the Kaplan-Meier method we demonstrated a lower decannulation rate in children with neurological impairment who underwent tracheostomy compared with that in children without neurological impairment who underwent tracheostomy. There were no significant differences in survival rates or the incidence of weaning from mechanical ventilation between the two groups. This information would be valuable to clinicians and families with children undergoing tracheostomy. Further investigations for clinical predictors of survival and long-term outcomes are warranted to elucidate the long-term clinical course after pediatric tracheostomy. Conflict of interest statement None. References [1] C.W. Lewis, J.D. Carron, J.A. Perkins, K.C. Sie, C. Feudtner, Tracheotomy in pediatric patients: a national perspective, Arch. Otolaryngol. Head. Neck Surg. 129 (2003) 523e529. [2] J.L. Funamura, B. Durbin-Johnson, T.T. Tollefson, J. Harrison, C.W. Senders, Pediatric tracheotomy: indications and decannulation outcomes, Laryngoscope 124 (2014) 1952e1958. [3] L.N. Ogilvie, J.K. Kozak, S. Chiu, R.J. Adderley, F.K. Kozak, Changes in pediatric tracheostomy 1982-2011: a Canadian tertiary children's hospital review, J. Pediatr. Surg. 49 (2014) 1549e1553.
N. Tsuboi et al. / International Journal of Pediatric Otorhinolaryngology 89 (2016) 81e85 [4] L. de Trey, E. Niedermann, D. Ghelfi, A. Gerber, C. Gysin, Pediatric tracheotomy: a 30-year experience, J. Pediatr. Surg. 48 (2013) 1470e1475. [5] S. Ozmen, O.A. Ozmen, O.F. Unal, Pediatric tracheotomies: a 37-year experience in 282 children, Int. J. Pediatr. Otorhinolaryngol. 73 (2009) 959e961. [6] M. Mahadevan, C. Barber, L. Salkeld, G. Douglas, N. Mills, Pediatric tracheotomy: 17 year review, Int. J. Pediatr. Otorhinolaryngol. 71 (2007) 1829e1835. [7] J.D. Carron, C.S. Derkay, G.L. Strope, J.E. Nosonchuk, D.H. Darrow, Pediatric tracheotomies: changing indications and outcomes, Laryngoscope 110 (2000) 1099e1104. [8] D.H. Fiser, Assessing the outcome of pediatric intensive care, J. Pediatr. 121 (1992) 68e74.
85
[9] W. Tantinikorn, C.M. Alper, C.D. Bluestone, M.L. Casselbrant, Outcome in pediatric tracheotomy, Am. J. Otolaryngol. 24 (2003) 131e137. [10] O. Dursun, D. Ozel, Early and long-term outcome after tracheostomy in children, Pediatr. Int. 53 (2011) 202e206. [11] R. Leung, R.G. Berkowitz, Decannulation and outcome following pediatric tracheostomy, Ann. Otol. Rhinol. Laryngol. 114 (2005) 743e748. [12] J.G. Berry, D.A. Graham, R.J. Graham, J. Zhou, H.L. Putney, J.E. O'Brien, et al., Predictors of clinical outcomes and hospital resource use of children after tracheotomy, Pediatrics 124 (2009) 563e572. [13] D. Wood, P. McShane, P. Davis, Tracheostomy in children admitted to paediatric intensive care, Arch. Dis. Child. 97 (2012) 866e869.
Contents lists available at ScienceDirect
International Journal of Pediatric Otorhinolaryngology journal homepage: http://www.ijporlonline.com/
Pediatric tracheostomy: Survival and long-term outcomes Norihiko Tsuboi a, *, Kentaro Ide a, Nao Nishimura a, Satoshi Nakagawa a, Noriko Morimoto b a b
Critical Care Medicine, National Center for Child Health and Development, Tokyo, Japan Otorhinolaryngology, National Center for Child Health and Development, Tokyo, Japan
a r t i c l e i n f o
a b s t r a c t
Article history: Received 17 May 2016 Received in revised form 26 July 2016 Accepted 26 July 2016 Available online 28 July 2016
Objectives: The objective of this study was to investigate if there were any differences in survival and long-term outcomes between pediatric patients with and without neurological impairment who underwent tracheostomy. Methods: A retrospective chart review of pediatric patients (age 0e15 years) who underwent tracheostomy between March 2002 and December 2013 was conducted. Patients were categorized into two groups: those who were neurologically impaired (NI) (pediatric cerebral performance category, 3e6) and those who were not neurologically impaired (NN) (pediatric cerebral performance category, 1e2). Survival rates and cumulative incidence of weaning from mechanical ventilation or decannulation were calculated using the Kaplan-Meier method. Results: A total of 212 patients were included. Among them, 141 were categorized into NI group and 71 into NN group. Between the two groups, there were no significant differences in survival rates and cumulative incidence of weaning from mechanical ventilation. In total patients, one-year survival rate was 0.86 (95%CI 0.80e0.90) and five-year survival rate was 0.71 (0.62e0.78). One-year weaning rate was 0.58 (0.51e0.65) and five-year weaning rate was 0.66 (0.59e0.74). Decannulation rates were significantly lower in NI group than in NN group (p < 0.001). One-year and five-year decannulation rates were 0.04 (0.01e0.09) and 0.17 (0.10e0.29), respectively, in NI group, and 0.20 (0.12e0.33) and 0.54 (0.40e0.69), respectively, in NN group. Conclusions: In children who underwent tracheostomy, the decannulation rate was lower in those with neurological impairment compared with that in those without neurological impairment. There were no significant differences in survival or ventilator weaning between the two groups. © 2016 Elsevier Ireland Ltd. All rights reserved.
Keywords: Tracheostomy Pediatric intensive care Weaning Decannulation Survival rate Neurological impairment
1. Introduction The frequency and indications of pediatric tracheostomy have undergone notable changes over the last 30 years, and pediatric tracheostomy has recently become a relatively common procedure with approximately 5000 procedures performed in the United States each year [1]. Pediatric patients who require tracheostomy tend to be younger and with chronic diseases [1e7]. The current literature on pediatric tracheostomy contains limited objective data, especially those that describe Kaplan-Meier estimates of survival and long-term outcomes after pediatric tracheostomy. The aim of this study was to investigate if there are any
differences in survival and long-term outcomes between pediatric patients with and without neurological impairment who underwent tracheostomy. We hypothesized that the long-term outcomes would be worse in patients with neurological impairment regardless of their comorbidities. Thus, we categorized our pediatric patients into two groups using the pediatric cerebral performance category (PCPC) as those who were neurologically impaired (NI) (PCPC, 3e6) and those who were not neurologically impaired (NN) (PCPC, 1e2) at discharge from the pediatric intensive care unit (PICU). We evaluated survival and long-term outcomes by using the Kaplan-Meier method.
2. Materials and methods * Corresponding author. Critical Care Medicine, National Center for Child Health and Development, 2-10-1 Okura, Setagaya, Tokyo, 157-8535, Japan. E-mail address: [email protected] (N. Tsuboi). http://dx.doi.org/10.1016/j.ijporl.2016.07.033 0165-5876/© 2016 Elsevier Ireland Ltd. All rights reserved.
This is a retrospective observational study at the PICU of the National Center for Child Health and Development (NCCHD) in
82
N. Tsuboi et al. / International Journal of Pediatric Otorhinolaryngology 89 (2016) 81e85
Tokyo, Japan. This study was approved by the ethics committee of our institution (receipt number 1003). 2.1. Patients We enrolled all the patients aged under 16 years old who underwent tracheostomy between March 2002 and December 2013, and who were admitted to the PICU at the NCCHD. We extracted these patients from the operation register and the PICU register. Medical records were reviewed and the following data were recorded: intubation days before tracheostomy, disposition before tracheostomy, age at tracheostomy, primary diagnosis, follow-up duration after tracheostomy, PCPC at discharge from the PICU, occurrence of an event of weaning from mechanical ventilation, decannulation, death after tracheostomy, and the period from the tracheostomy until weaning from mechanical ventilation or decannulation. Outcomes were survival rates and cumulative incidence of weaning from mechanical ventilation or decannulation after tracheostomy. We estimated these outcomes at the end of June 2015. The unit is a multivalent PICU composed of 20 beds at a tertiary child hospital. 2.2. Tracheostomy procedure and postoperative management All tracheostomies were performed by otorhinolaryngologists at our institution. In general, systemic management was applied in the PICU during the postoperative acute phase (10e14 days after tracheostomy). Weaning from ventilation was achieved when patients could breathe spontaneously without hypercapnia. Decannulation was considered when all of the following preconditions were satisfied: restoration of the voice, ability to swallow, and no demand for oxygen. Before decannulation, the cannula was exchanged for one that is of smaller size to ensure adequate air circulation, and velopharyngeal training was given when necessary. 2.3. PCPC scale The PCPC scale was developed to describe the short-term outcome of pediatric intensive care by quantifying cognitive impairment8. It is a six-point graded scale of increasing disability from normal function (score ¼ 1) to death (score ¼ 6) (Table 1). A child with Down syndrome is usually classified as PCPC 3. Scale score has been found to be significantly associated with several measures of morbidity, including length of stay in the PICU, total hospital charges, discharge care needs, and the Pediatric Risk of Mortality Score [8]. 2.4. Statistical analyses Patients who underwent tracheostomy were categorized into two groups: NI (PCPC, 3e6) and NN (PCPC, 1e2). Patients' PCPC
scores were evaluated at discharge from the PICU by clinicians of the unit. Descriptive statistics were expressed as medians and interquartile ranges for continuous data, or absolute frequencies and percentage for categorical data. Bivariate comparisons between groups were performed using ManneWhitney U for continuous data, and Fisher exact for categorical data. Ninety-five percent confidence intervals (95% CI) were also calculated for relevant outcomes. Survival rates were calculated using the Kaplan-Meier method and cumulative incidence of weaning from mechanical ventilation or decannulation were demonstrated as one minus the Kaplan-Meier survival method. Comparisons between groups were performed using the log-rank test. The potential confounding factors were adjusted by Cox proportional hazards model. Two-sided p values less than 0.05 were considered statistically significant. All analyses were performed using R version 3.2.0.
3. Results A total of 212 patients out of 10,250 patients in the PICU during the study period were included. Among the included patients, 141 were categorized into NI group (PCPC, 3e6) and 71 into NN group (PCPC, 1e2). The characteristics of the two groups are shown in Table 2. The age at tracheostomy was significantly older in NI group than in NN group (p < 0.001). In total patients, 142 patients were followed-up for more than one year, and 55 patients for more than five years. Deaths were reported in 47 patients. Causes of death were progression of the underlying disease in 25 patients, sepsis in five, airway-related causes in four, pneumonia in four, and other causes in nine. Kaplan-Meier estimates of survival rates and cumulative incidence of weaning from mechanical ventilation or decannulation are summarized in Figs. 1e3. Between the two groups, there were no significant differences in survival rates and cumulative incidence of weaning from mechanical ventilation. In total patients, one-year survival rate was 0.86 (95%CI 0.80e0.90) and five-year survival rate was 0.71 (0.62e0.78). One-year weaning rate was 0.58 (0.51e0.65) and five-year weaning rate was 0.66 (0.59e0.74). Decannulation rates were significantly lower in NI group than in NN group (p < 0.001). In NI group, one-year and five-year decannulation rates were 0.04 (0.01e0.09) and 0.17 (0.10e0.29), respectively, and in NN group, 0.20 (0.12e0.33) and 0.54 (0.40e0.69), respectively. When adjusting for age in the Cox proportional hazards model, decannulation rates remained significantly lower in NI group than in NN group (p < 0.001). Thirty patients received additional airway surgical interventions under general anesthesia after tracheostomy to achieve or attempt decannulation. Among them, there were 21 cases of resection of tracheal granuloma, two cases of resection of upper airway tumor, and two cases of glossectomy. Bronchial foreign body removal, tracheoesophageal fistula repair, anterior tracheal wall suspension,
Table 1 Pediatric cerebral performance Category scale. Score Category 1 2 3 4 5 6
Normal Mild disability
Description
Normal; at age-appropriate level; school-age child attending regular school classroom Conscious, alert, and able to interact at age-appropriate level; school-age child attending regular school classroom, but grade perhaps not appropriate for age; possibility of mild neurologic deficit Moderate disability Conscious; sufficient cerebral function for age-appropriate independent activities of daily life; school-age child attending special education classroom and/or learning deficit present Severe disability Conscious; dependent on others for daily support because of impaired brain function Coma or vegetative Any degree of coma without the presence of all brain death criteria; unaware, even if awake in appearance, without interaction with state environment; cerebral unresponsiveness and no evidence of spontaneous eye-opening, and sleep-wake cycles Brain death Apnea, areflexia, and/or electroencephalographic silence
N. Tsuboi et al. / International Journal of Pediatric Otorhinolaryngology 89 (2016) 81e85
83
Table 2 Patient characteristics. NI (n ¼ 141) Age, month Disposition before tracheostomy Pediatric intensive care unit Neonatal intensive care unit General ward Home Intubation days before tracheostomy Follow-up duration, day Primary diagnosis Neurological disease Malformation syndrome Airway obstruction Craniofacial disease Trauma Pulmonary disease Metabolic disease Tumor Heart disease Skeletal disorder Liver failure Muscular disease Others
NN (n ¼ 71)
p
24
(8e81)
7
(5e17)
<0.001
119 13 4 5 21 888
84% 9% 3% 4% (15e58) (99e1915)
58 8 4 1 21 918
82% 11% 6% 1% (11e64) (215e1839)
0.696 0.633 0.446 0.666 0.479 0.567
74 18 11 7 7 5 5 4 4 2 1 1 2
52% 13% 8% 5% 5% 4% 4% 3% 3% 1% 1% 1% 1%
6 4 12 5 0 5 2 12 6 5 9 2 3
8% 6% 17% 7%
<0.001 0.152 0.060 0.541 0.098 0.308 1 <0.001 0.088 0.043 <0.001 0.260
7% 3% 17% 8% 7% 13% 3% 4%
NI: neurologically impaired group (PCPC, 3e6). NN: non-neurologically impaired group (PCPC, 1e2). Bivariate comparisons between groups were performed using ManneWhitney U for continuous data, and Fisher exact for categorical data. Medians and (interquartile ranges) are expressed for quantitative variables.
Fig. 1. Survival rates after tracheostomy. NI: neurologically impaired group (PCPC, 3e6). NN: non-neurologically impaired group (PCPC, 1e2).
Fig. 3. Cumulative incidence of decannulation after tracheostomy. NI: neurologically impaired group (PCPC, 3e6). NN: non-neurologically impaired group (PCPC, 1e2).
dissection of epiglottis adhesion, sclerotherapy of cervical lymphangioma, tracheal scar excision, and tracheal laser therapy were performed once respectively in this series. Neither tracheal balloon dilation nor laryngotracheal reconstruction with grafts were performed. 4. Discussion
Fig. 2. Cumulative incidence of weaning from mechanical ventilation after tracheostomy. NI: neurologically impaired group (PCPC, 3e6). NN: non-neurologically impaired group (PCPC, 1e2).
We reported the survival and long-term outcomes of patients after pediatric tracheostomy. Our findings were consistent with current reports and provide further information on long-term outcomes after pediatric tracheostomy [2e7,9,10]. Funamura et al. demonstrated a difference in the overall decannulation rates among pediatric patients undergoing tracheostomy for different indications [2]. In the study, pediatric tracheostomy indications were categorized into five groups according to the underlying causes: cardiopulmonary, craniofacial anomalies, neurological impairment, traumatic injury and upper airway obstruction. In other studies, similar categorizations of
84
N. Tsuboi et al. / International Journal of Pediatric Otorhinolaryngology 89 (2016) 81e85
pediatric tracheostomy indications have been adopted [1,3,7,9]. While these classifications are helpful, it is at times difficult to clearly classify all patients into these categories based on their comorbidities, such as those with both craniofacial anomalies and neurological impairment. Thus, we simply categorized our patients into two groups using the PCPC as those having neurological impairment and those without at discharge from the PICU. 4.1. Survival rates Recent reports on tracheostomy demonstrated a mortality rate of 8e20% [1e7,9,10], with the majority of deaths caused by a progression of underlying diseases. In our study, one-year mortality rate was 14% and five-year mortality rate was 29%. About half of the causes of death were progression of the underlying diseases, with the number increasing yearly. It is difficult to compare our mortality rate with those of previous reports because they did not use the Kaplan-Meier method to calculate the survival rates, nor did they state the individual follow-up duration after tracheostomy. In our study, four deaths due to airway-related causes are reported: two accidental tube removal and two tube obstruction. All four events occurred at home. To avoid accidental and fatal complication, it is necessary to provide sufficient instruction and to support the parents or guardians in terms of the management of tracheal cannula. To that end, we provide an education program prior to discharge for parents and guardians, including basic cardiopulmonary resuscitation, bag-valve mask ventilation, tracheal cannula exchange, and airway suctioning. 4.2. Weaning rates We considered that weaning from mechanical ventilation was one of the long-term outcomes after pediatric tracheostomy. Thus, we showed the incidence of weaning from mechanical ventilation calculated as one minus the Kaplan-Meier survival method. In 25 out of 123 cases that could wean from mechanical ventilation, the primary reason for tracheostomy was “difficulty in weaning from mechanical ventilation”. Among the 25 cases, 22 had pulmonary causes, two had cardiac causes and one had muscular cause. We speculate that these patients were able to wean off mechanical ventilation due to the various benefits that surgery yielded, including reduced effort in breathing, reduced dead space, easier elimination of pharyngeal secretions from the respiratory tract, reduced use of sedatives, and recovery of respiratory function owing to active rehabilitation or physical development. In contrast, previous pediatric reports did not show the weaning data in detail. 4.3. Decannulation rates Several recent reports which did not use the Kaplan-Meier method demonstrated 17e75% decannulation rates after tracheostomy, and showed lower deccanulation rates in patients with neurological impairment [3e7,9,10,12]. On the other hand, Leung and Berkowitz used the Kaplan-Meier method and reported that patients who had undergone tracheostomy for tracheobronchial toilet had a significantly shorter time to decannulation [11]. Funamura et al. also demonstrated higher decannulation rates and shorter time to decannulation in children undergoing tracheostomy for maxilla facial and laryngotracheal trauma compared with children undergoing tracheostomy for cardiopulmonary and neurological indications by using the Kaplan-Meier survival method [2]. We demonstrated lower deccanulation rates in patients with neurological impairment by using the Kaplan-Meier method. Higher decannulation rate of the patients without neurological impairment, which increased yearly, suggests that
those without neurological impairment have the potential to be decannulated as they become older regardless of their chronic respiratory problems. For neurologically impaired patients, the key factors of decannulation are the ability to wean from the mechanical ventilation and to maintain airway patency. We postulate that this is the reason why weaning rates were not significantly different between the two groups, as opposed to decannulation rates. In this study, there were 21 patients who were admitted to the neonatal intensive care unit (NICU) before receiving tracheostomy. Their primary diagnosis were five cases of neurological disease, four cases of craniofacial disease, three cases of malformation syndrome, tumor, or skeletal disease, two cases of airway obstruction, and a case of muscular disease. The 21 patients from the NICU had lower survival rates (p < 0.001), lower weaning rates (p ¼ 0.026), and lower decannulation rates (p ¼ 0.034), compared with the other patients. There were no significant differences in the proportion of patients from the NICU between NI and NN groups. This study has several limitations. First, there could be other factors that affected the survival and long-term outcomes. Second, the patients' PCPC scale was evaluated at discharge from the PICU. Some patients who were temporarily impaired upon discharge from the PICU may have made a neurological recovery later. Third, during the study period, there were a few patients who underwent tracheostomy but were not admitted to the PICU. Thus, the conclusions may not reflect “all” pediatric patients who underwent tracheostomy. Fourth, this was a single-center study. Therefore, there may be a risk of bias in patient selection. However, since our institution is a major pediatric tertiary hospital, the patient series presented in this study was likely to be a good representation of the distribution of cases in other pediatric tertiary hospitals. Wood et al. reported that the incidence of pediatric tracheostomies over five years among PICUs in 29 centers in Britain was 2.1% (1613 tracheostomy cases out of 78,504 PICU admissions), with infants under one year accounting for 46% [13]. Consistent with that study, the pediatric tracheostomy rate in our study was 2.1% (212/10,250), with 48% (101/212) of the patients being under one year old. 5. Conclusions By using the Kaplan-Meier method we demonstrated a lower decannulation rate in children with neurological impairment who underwent tracheostomy compared with that in children without neurological impairment who underwent tracheostomy. There were no significant differences in survival rates or the incidence of weaning from mechanical ventilation between the two groups. This information would be valuable to clinicians and families with children undergoing tracheostomy. Further investigations for clinical predictors of survival and long-term outcomes are warranted to elucidate the long-term clinical course after pediatric tracheostomy. Conflict of interest statement None. References [1] C.W. Lewis, J.D. Carron, J.A. Perkins, K.C. Sie, C. Feudtner, Tracheotomy in pediatric patients: a national perspective, Arch. Otolaryngol. Head. Neck Surg. 129 (2003) 523e529. [2] J.L. Funamura, B. Durbin-Johnson, T.T. Tollefson, J. Harrison, C.W. Senders, Pediatric tracheotomy: indications and decannulation outcomes, Laryngoscope 124 (2014) 1952e1958. [3] L.N. Ogilvie, J.K. Kozak, S. Chiu, R.J. Adderley, F.K. Kozak, Changes in pediatric tracheostomy 1982-2011: a Canadian tertiary children's hospital review, J. Pediatr. Surg. 49 (2014) 1549e1553.
N. Tsuboi et al. / International Journal of Pediatric Otorhinolaryngology 89 (2016) 81e85 [4] L. de Trey, E. Niedermann, D. Ghelfi, A. Gerber, C. Gysin, Pediatric tracheotomy: a 30-year experience, J. Pediatr. Surg. 48 (2013) 1470e1475. [5] S. Ozmen, O.A. Ozmen, O.F. Unal, Pediatric tracheotomies: a 37-year experience in 282 children, Int. J. Pediatr. Otorhinolaryngol. 73 (2009) 959e961. [6] M. Mahadevan, C. Barber, L. Salkeld, G. Douglas, N. Mills, Pediatric tracheotomy: 17 year review, Int. J. Pediatr. Otorhinolaryngol. 71 (2007) 1829e1835. [7] J.D. Carron, C.S. Derkay, G.L. Strope, J.E. Nosonchuk, D.H. Darrow, Pediatric tracheotomies: changing indications and outcomes, Laryngoscope 110 (2000) 1099e1104. [8] D.H. Fiser, Assessing the outcome of pediatric intensive care, J. Pediatr. 121 (1992) 68e74.
85
[9] W. Tantinikorn, C.M. Alper, C.D. Bluestone, M.L. Casselbrant, Outcome in pediatric tracheotomy, Am. J. Otolaryngol. 24 (2003) 131e137. [10] O. Dursun, D. Ozel, Early and long-term outcome after tracheostomy in children, Pediatr. Int. 53 (2011) 202e206. [11] R. Leung, R.G. Berkowitz, Decannulation and outcome following pediatric tracheostomy, Ann. Otol. Rhinol. Laryngol. 114 (2005) 743e748. [12] J.G. Berry, D.A. Graham, R.J. Graham, J. Zhou, H.L. Putney, J.E. O'Brien, et al., Predictors of clinical outcomes and hospital resource use of children after tracheotomy, Pediatrics 124 (2009) 563e572. [13] D. Wood, P. McShane, P. Davis, Tracheostomy in children admitted to paediatric intensive care, Arch. Dis. Child. 97 (2012) 866e869.