Abstract
Purpose
The purpose of the study was to find the factors that were associated with tracheostomy procedures in ventilated pediatric trauma patients.
Methods
The Trauma Quality Improvement Program (TQIP) database of the calendar year 2017 through 2019 was accessed for the study. All patients <18 years old and who were on mechanical ventilation for more than 96 hours were included in the study. Multiple logistic regression analysis was performed to find the factors that were associated with a tracheostomy.
Results
Out of 2653 patients, 1907 (71.88%) patients underwent tracheostomy. The patients who underwent tracheostomy had a lower median [IQR] of Glasgow Coma Scale (GCS) (3 [3-8] vs 5 [3-10], P < .001) and had a higher proportion of severe spine injury (On Abbreviated Injury Scale [AIS]≥3) (11.6% vs 8.8%, P = .044) when compared with patients who did not have tracheostomy. Lower GCS scores and severe spine injury were associated with higher odds of tracheostomy, with all P values <.05. Higher proportion of tracheostomy procedures were performed at level I pediatric trauma centers as compared to non-designated pediatric centers (odds ratio [95% CI]: 1.848 [1.524-2.242], P < .001).
Conclusion
A lower GCS score, severe spine injury and highest level trauma centers were associated with a tracheostomy.
Key Takeaways
• Question. The purpose of the study was to find the factors that were associated with tracheostomy procedures in ventilated pediatric trauma patients. • Findings. The study showed patients who suffered from severe traumatic brain or spine injury are at higher risk for tracheostomy. • There is an association between highest level care pediatric trauma centers and the increased incidence of tracheostomy in this cohort of patients. Certain age specific factors found to be associated with tracheostomy in children. • Meanings. Early tracheostomy may be performed in high-risk patients.
Introduction
Tracheostomy is one of the common procedures performed in the intensive care unit (ICU) setting on ventilated trauma victims. 1 Prolonged mechanical ventilation is one of the reasons for tracheostomy because of the benefit of early liberation of the ventilator, fewer length of hospital stay (LOS), and Intensive care unit (ICU) days.2,3 Severe head injury patients occasionally undergo early tracheostomy due to potential benefit. 4 There are guidelines for the timing of tracheostomy in adult ventilated patients that say early tracheostomy should be performed when patients are expected to be on mechanical ventilation for 3 or more weeks 5 ; however, there are no such guidelines for the pediatric population.
Among many factors, severe traumatic brain injury (TBI) is one of the organ injuries that occasionally require tracheostomy for airway protection. 6 Much information regarding the trauma conditions that led to tracheostomy comes from adult studies. There is a paucity of information regarding the factors that result in tracheostomy in pediatric trauma patients. Therefore, we designed the study to find out what factors can lead to tracheostomy in the pediatric patients’ cohort so that the providers can engage the family and initiate early discussion of the procedure. The objective of our study was to identify factors that are associated with tracheostomy in pediatric trauma patients using the Trauma Quality Improvement Program (TQIP) database.
Methods
Ethical Approval
The study was performed from the de-identified data that the data is available to the researchers. As per the policy of the Hackensack Meridian Health IRB, the study was deemed exempt from the review.
All procedures followed were in accordance with the ethical standards of the Institutional Review Board of Hackensack Meridian Health and with the Helsinki Declaration of 1975, as revised in 2008. Since the data of TQIP are de-identified patient information available to the researchers, the study was exempted from Hackensack Meridian Health IRB review.
Data Acquisition and Inclusion and Exclusion Criteria
To perform the study, we accessed the Trauma Quality Improvement Program (TQIP) database of participant use files (PUF) of the calendar year from 2017 through 2019. The TQIP database is created and maintained by the American College of Surgeons (ACS). Trauma centers across the US share data with the ACS-TQIP voluntarily. Trauma registrars in each trauma center abstract the data into the trauma registry and deposit data into TQIP. Currently, more than 800 hospitals across the US participate in data-sharing programs with TQIP. All pediatric trauma patients <18 years old who were admitted to the hospital and placed on mechanical ventilation for longer than 96 hours were included in the study. Other variables included in the study are race, sex, injury type, injury severity score (ISS), Glasgow Coma Scale (GCS) score, abbreviated injury scale (AIS) score of body region injuries, massive blood transfusion [MBT] (≥1500cc Packed Red Blood Cells over initial 4 hours of hospital arrival), procedures to control the hemorrhage; laparotomy, thoracotomy, angioembolization, laparotomy and embolization, and certain comorbidities including congenital disease, prematurity, mental and personality disorder, substance abuse, and history of coagulopathy. Patients who died in the emergency department were excluded from the study. Any patient who stayed on the ventilator for less than 96 hours was excluded also. Data with missing information were excluded except where the pediatric trauma center verification level was missing. The missing information on the pediatric center designation was considered a category. The multiple imputations were not performed because the missing information did not look like missing at random. An abbreviated injury scale (AIS) score of ≥3 was considered severe injury.
Outcomes
The primary outcome of the study was to identify factors that are associated with tracheostomy.
Statistics
Patient demographic information and outcomes were summarized using summary statistics (median with interquartile range (IQR) [first quartile – third quartile] for continuous variables, and frequency and percentage for categorical variables). To compare the groups, who underwent tracheostomy and who did not go for tracheostomy, the Wilcoxon Rank Sum test was used for continuous variables, and the Chi-square test or Fisher exact test was used for the categorical variables, based on the sample size. A multiple logistic regression model was used to assess the chance of having a tracheostomy with those variables which had a P-value < .2 in the univariate analysis or reported as a factor for tracheostomy. The backward selection procedure was used to identify the factors which were significantly associated with the chance of having a tracheostomy. The parameter estimates from the fitted model were summarized using β coefficient estimates and 95% confidence intervals (CIs), and 95% CIs for the odds ratio (OR) as measures of precisions. The Hosmer-Lemeshow goodness-of-fit test was used to evaluate the model fitting. The receiver operating characteristic (ROC) curve was constructed and the corresponding area under the curve (AUC) with their confidence intervals was calculated by using DeLong’s method. 7 The 2-sided P-value was reported for each test. A P-value of <.05 was considered an indication of statistical significance. Statistical analysis was performed using the R language. 8
Results
Patient Characteristics and Univariate Analysis
Univariate Analysis Between the Patients Who had Tracheostomy vs No Tracheostomy.
Note. ISS; Injury severity score, GCS; Glasgow coma scale, AIS; Abbreviated injury scale, Q1-Q3; first quartile and Third quartile, *; statistically significant.
Multivariable Analysis
Multivariable Analysis.
Note. GCS; Glasgow Coma Scale, AIS; Abbreviated injury scale, * = Statistically significant.
The AUC [95%CI] was .720 [.701-.740]. The AUC shows that the chance of having a tracheostomy can be fairly estimated by using the proposed model The Hosmer-Lemeshow goodness of fit test showed a P value of .197. It means the model for finding the factors for tracheostomy was good Figure 1. Receiver operating characteristics (ROC) curve of factors associated with tracheostomy in ventilated pediatric trauma patients.
Multivariable Analysis of Factors Associated With Tracheostomy Based on Age Category.
Note. GCS; Glasgow Coma Scale, AIS: Abbreviated Injury Scale, Age is in years, Race other; age is not categorized as white, black, Asian, etc.,* ; statistically significant
The AUC of all 4 age categories of risk factor for tracheostomy ranged from .877 to .940 means that identified factors for the prediction of tracheostomy were in the range of good to excellent.
Discussion
Our study showed that of the pediatric patients who were placed on mechanical ventilation after sustaining a traumatic injury and stayed on the ventilator for more than 96 hours, approximately 72% underwent tracheostomy. The significant factors associated with the tracheostomy were lower GCS score and spine injury. The patients who were treated at level I pediatric trauma centers had almost double the odds of having the tracheostomy compared to patients who were treated at non-designated pediatric trauma centers. The history of prematurity was an independent risk factor for tracheostomy in children of all ages (up to 17 years old). Advanced age, history of congenital anomalies, and severe neck injury (AIS ≥3) were found to be risk factors for tracheostomy in the age range of 13-17 years old. The patient’s ISS was not associated with the tracheostomy procedure.
Studies have shown that early tracheostomy in TBI resulted in favorable outcomes during a 6 months follow-up 6 with the failure of successful liberation from mechanical ventilation in the adult population being one of the reasons for the need for tracheostomy in TBI patients. 9 Therefore, it is not surprising to find that the lower GCS was one of the factors associated with a tracheostomy. Our study showed that a decrease of one point of GCS (range 3-15) would lead to a 3.6% increased chance of tracheostomy. However, tracheostomy procedures occurred less frequently in pediatric patients compared to adult TBI patients. 10
The other factor our study identified that was associated with tracheostomy in children was spine injury. Higher cervical spine and associated spinal cord injury can result in acute respiratory failure and prolonged mechanical ventilation. The most common mechanism with spine injury was found to be motor vehicle crash in younger children and sport-related mechanism in older children and associated with significant spinal cord injury. 11 However, a study in adult cervical spinal cord injury patients from a multicenter trial found that 62.6% of patients can be successfully extubated at the time of discharge from mechanical ventilation, 12 raising a question regarding the need for tracheostomy in cervical spinal cord injury patients. The probable reasons for tracheostomy in spine injury in our study may be related to the providers’ understanding of the decreased function of respiratory mechanics and increased production of respiratory secretion following spinal cord injury can lead the patients on mechanical ventilation for a prolonged period.13-15
Our study found yet another unexpected factor: the pediatric trauma patients who were treated at level I trauma centers had almost double the odds of getting tracheostomy in our population compared to non-designated pediatric trauma centers. This risk factor was mostly seen in the age group of 13 years to 17 years old children. The reason for increased tracheostomy procedures at level I pediatric designated center is not completely understood. A prior report showed that severe pediatric trauma patients (ISS >15) who were treated at higher volume patient centers had a better survival probability and overall better outcomes. 16 Contrary to our results a paper presented at a Critical Care Medicine meeting a few years ago found a smaller number of tracheostomies were performed in pediatric trauma centers when compared to pediatric and adult (mixed center) and adult trauma centers (3.7% vs 6.9% vs 8.9%), respectively. The patient cohort in the study was all pediatric trauma patients who were intubated and on mechanical ventilation for ≥24 hours. 17 The same group of researchers has published another report that showed similar findings that adult trauma center physicians performed more tracheostomies than physicians working in only pediatric trauma centers. 18 The variation in results between our study with the published reports may be due to the selection of patients in the study and the design of the study where the study compared patients between pediatric trauma centers with pediatric and adult trauma centers. Our study compared the pediatric trauma centers level I to level II pediatric trauma centers and non-designated pediatric trauma centers.
Multivariable analysis based on different age categories showed that certain age specific factors were associated with tracheostomy. For example, prematurity was an independent risk factors for all age groups (up to 17 years of age), whereas in age category of up to 3 years old and 7 years to 12 years old, lower GCS was the only other factor that was associated with higher rate of tracheostomy. Similarly, age group of 13-17 years old, not only advanced age was independently associated with higher rate of tracheostomy but also history of congenital abnormality, prematurity, and severe neck injury (AIS ≥3) were associated with a higher rate of tracheostomy. Therefore, it is important for providers to consider age of a child when trying to identify the risk factor for tracheostomy.
Identifying factors associated with a tracheostomy is important information for providers to consider early tracheostomy in a high-risk patient. Two recent studies from the National Trauma Data Bank (NTDB) included ∼10 years of data. 18,19 They found that early tracheostomy was associated with lower LOS, ICU days, ventilator days, and incidence of pneumonia. The definition of the timing of early tracheostomy and the age ranges were different in the 2 studies; however, early tracheostomy was beneficial not only in severe TBI patients but also in non-TBI patients. 18
Limitation
The retrospective nature of the study carries an inherent bias, which was partly addressed by the use of a multivariable analysis to identify factors that can be associated with tracheostomy procedures in pediatric trauma patients. However, the lack of detailed information in the database regarding the data on respiratory mechanics and failure of successful extubation may have impacted the overall results. The exact reasons for a higher number of tracheostomies in level I pediatric trauma centers were not completely understood. However, there is a possibility that level I pediatric trauma centers had successfully extubated more patients earlier (within 4 days) compared to level II or non-designated pediatric trauma centers. Therefore, higher frequency of tracheostomy procedure seen at the level I trauma center could be related to the higher severity of the patients’ condition that required prolonged mechanical ventilation. Further prospective study may be needed to be performed to answer the question. Another limitation of the study was the lack of detailed information of level of spine injury.
In conclusion, most pediatric trauma patients who were on mechanical ventilation for more than 4 days underwent tracheostomy. Lower GCS score, severe spine injury, and history of prematurity were the main patient factors that were significantly associated with a tracheostomy. However, certain age specific factors were also found to be associated with higher rate of tracheostomy. A child’s age should be taken into consideration when determining the risk factors of tracheostomy in ventilated pediatric trauma patients.
Footnotes
Acknowledgments
Elli Gourna Paleoudis, MS, Ph.D. performed the critical reading and final editing of the manuscript.
Author Contributions
Nasim Ahmed (NA) conceived and designed the study. Nasim Ahmed was responsible for retrieving the study data, while Yen-Hong Kuo (YH K) performed the data analysis. NA and YHK both contributed to manuscript writing.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
Ethical Approval
All procedures followed were in accordance with the ethical standards of the Institutional Review Board of Meridian Health and with the Helsinki Declaration of 1975, as revised in 2008. Since the data of TQIP are de-identified patient information available to the researchers, the study was exempted from Hackensack Meridian Health IRB review.
Informed Consent
Informed consent Given this retrospective study was performed from the de-identified National database from the American College of Surgeons that is available to all researchers, this study was exempted from the Hackensack Meridian Health IRB review as per policy and no informed consent was required.
Data Availability
The data that support the findings of this study are available from [American College of Surgeons, contact information;