Tracheal intubation (TI) is one of the most common procedures in the PICU and can lead to severe hypoxemia and other life-threatening complications (1, 2). Adverse TI associated events (TIAEs) and hypoxemia below 80% during TI occur in approximately 15% and 13% of TIs, respectively (2, 3). Compared with healthy patients undergoing TI (e.g., during anesthesia for elective surgery), critically ill patients are at a much higher risk for adverse events during TI (4). Adverse TIAEs may lead to increased morbidity and mortality in critically ill children and have been used as quality improvement (QI) and patient safety indicators (5).
Previous studies have demonstrated that airway events might be associated with ICU outcomes such as length of stay in critically ill children. Extubation failure in infants after cardiac surgery is associated with a significantly longer ICU stay (6). Children experiencing an unplanned extubation during their PICU course have increased length of stay in both the PICU and the hospital (7). What remains unclear is whether factors around the time of intubation itself contribute to these clinical outcomes: length of PICU stay or mechanical ventilation (MV), and mortality in the critically ill children.
To date, no studies have shown an association between immediate events such as TIAEs or peri-intubation oxygen desaturation and important clinical outcomes in critically ill children. We hypothesized that the occurrence of TIAEs or peri-intubation oxygen desaturations is associated with ICU outcomes: PICU mortality, longer length of PICU stay, and longer length of MV. We also hypothesized that the occurrence of any TIAE alone is associated with these ICU outcomes. Finally, we hypothesized the occurrence of severe TIAE alone is associated with these ICU outcomes. We used data from the National Emergency Airway Registry for Children (NEAR4KIDS) Network, a prospective multicenter QI intubation safety registry, to investigate the impact of TIAEs and oxygen desaturations on clinical outcomes.
Study Design and Setting
We performed a retrospective analysis of prospectively collected observational data on TIs from 35 PICUs in the NEAR4KIDS Network. Participating sites were enrolled through the Pediatric Acute Lung Injury and Sepsis Investigators (PALISI) Network (8). Institutional Review Board approval was obtained at each participating site.
Patients under 18 years old admitted to a participating PICU from January 2013 to June 2015 were included in the analysis if they underwent primary TI in the PICU and had documented ICU outcomes. TIs to replace an existing tracheal tube (tracheal tube change) were excluded.
Data were extracted from the multicenter NEAR4KIDS database for all TIs occurring in participating PICUs. The NEAR4KIDS database was developed by members of the PALISI Network in collaboration with NEAR4KIDS and the National Emergency Airway Registry investigators to improve advanced-airway management for critically ill children (9, 10). The data collection form was developed and piloted in a single tertiary care PICU and adapted by NEAR4KIDS Network investigators for the multicenter investigation (2, 9).
Collected data included patient demographics (age, weight, sex), patient severity of illness (Pediatric Index of Mortality 2 [PIM2]), patient history (illness category, cyanotic heart disease), patient assessment (indication for TI, airway examination), airway management approach and medications, and adverse events. Each site leader developed a site-specific compliance plan to ensure greater than 95% capture rate for intubation encounters performed in the PICU. Two NEAR4KIDS compliance officers were responsible for review and approval for each site’s plan prior to the start of data collection. Site leaders subsequently reviewed their PICU intubation encounters and recorded PICU mortality, duration of PICU stay, and duration of MV for each patient.
Definitions and Outcome Measures
The NEAR4KIDS operational definition team defined adverse TIAEs a priori with two categories: nonsevere TIAEs and severe TIAEs (2, 9, 11). Severe TIAEs included cardiac arrest, esophageal intubation with delayed recognition, emesis with witnessed aspiration, hypotension requiring intervention, laryngospasm, malignant hyperthermia, pneumothorax, pneumomediastinum, or direct airway injury. Nonsevere TIAEs included mainstem bronchial intubation, esophageal intubation with immediate recognition, emesis without aspiration, hypertension requiring therapy, epistaxis, dental or lip trauma, medication error, arrhythmia, and pain or agitation requiring additional medication and causing a delay in intubation. Emesis was coded when gastric content was noticed in the oropharynx during airway management. Peri-intubation oxygen desaturation was defined as SpO2 less than 80% during an intubation attempt, when the highest documented SpO2 was greater than 90% after preoxygenation.
An airway management encounter, course, and attempt were defined a priori to improve reporting of intubation events (2). “Encounter” was defined as one completed advanced-airway management event, including intubation, for a patient. “Course” was defined as one method of TI and one set of medications. “Attempt” was defined as a single advanced airway maneuver and ending when the device is removed.
Primary outcomes were PICU mortality, duration of PICU stay, and duration of MV. Duration of PICU stay and MV were reported as number of days with 0–24 hours = 0, 25–48 hours = 1, 49–72 hours = 2, etc. If a patient failed extubation within 24 hours, it was not considered a successful extubation. Duration of MV was determined by time (in days) between TI and successful extubation.
Statistical analysis was performed using STATA 11.2 and 14.0 (StataCorp, College Station, TX). Data are presented as mean ± SD or median (interquartile range [IQR]) where appropriate. Univariate analysis for categorical variables was performed using chi-square test. Wilcoxon signed rank test was used for comparison of nonparametric variables. For multivariate analysis, logistic regression was used to identify independent associations between PICU outcomes and categorical variables. Linear regression was similarly used to identify independent associations between PICU outcomes and continuous variables. Natural log transformation was performed to ensure normality.
A multivariable regression model with the occurrence of any TIAEs or desaturation to SpO2 less than 80% as an exposure variable included patient-level covariates associated with occurrence of any TIAEs or desaturation to SpO2 less than 80% (age, PIM2, diagnosis, indication for respiratory failure, shock, procedural, upper airway obstruction, pulmonary toilet [suctioning and clearance of secretions], history of difficult airway, and symptom of upper airway obstruction). p value of less than or equal to 0.1 was used for inclusion criteria except for PIM2.
A multivariable regression model with any adverse TIAE included patient-level covariates associated with occurrence of any TIAE (diagnosis, indication for respiratory failure, shock, procedural, indication for therapeutic hyperventilation, history of difficult airway, and symptom of upper airway obstruction), age and PIM2 score. p value of less than or equal to 0.1 was used for inclusion criteria except age and PIM2 which were decided a priori.
A multivariable regression model with severe TIAE included covariates associated with occurrence of severe TIAE (age, diagnosis, indication for respiratory failure, shock, procedural, loss of airway protection, and symptom of upper airway obstruction) and PIM2 score; p value of less than or equal to 0.1 was used for inclusion criteria. Finally, sensitivity analyses were performed by repeating the above multivariable regression with a limited TI dataset from the sites that reported greater than 95% of PICU outcome data (i.e., missing data < 5%).
Site and Patient Characteristics
A total of 5,504 primary TI encounters were entered in to the NEAR4KIDS database by 35 participating PICUs during the study period (January 2013 to June 2015). A median of 108 (IQR, 58–229) TIs per site was reported during that time period. At least one TIAE was reported in 892 TIs (16.2%), whereas severe TIAEs were reported in 364 TIs (6.6%). TIAE or desaturation to SpO2 less than 80 % was reported in 1,617 (29%) of TIs (Table 1). The completeness of outcomes data (PICU mortality, PICU length of stay, duration of MV) ranged from 78% to 81% for all TI encounters.
The majority of TIs (46%) were performed in infants less than 1 year of age, followed by children 1–7 years old (33%) and older children (21%). Demographics and clinical factors associated with TIAEs are outlined in Table 1. There were statistically significant differences in age, diagnostic category, indication for TI, and upper airway obstruction between the group with TIAE or desaturation to SpO2 less than 80% and no event group (Table 1). Infants had a higher frequency of TIAE or desaturation to SpO2 less than 80%, as did children with a respiratory indication for TI. There were no significant differences in the PIM2 scores between event and nonevent groups.
The PICU mortality data were available in 4,471 (81%) of all 5,504 entered TI encounters. Among those reported, the PICU mortality was 9.6% (430/4,471 encounters). The overall median duration of PICU stay after TI was 12 days (IQR, 6–16 d) in 4,318 (78%) TI encounters. Data on duration of MV after TI encounter were available in 4,349 (79%) of TI encounters. The overall median MV duration was 4 days (IQR, 1–8 d).
The occurrence of TIAE or desaturation to SpO2 less than 80% was associated with longer duration of MV (5 vs 3 d; p < 0.001) and longer PICU stay (14 vs 11 d; p < 0.0001) but not with ICU mortality. The occurrence of TIAE was significantly associated with longer duration of MV (4 d [IQR, 2–9 d] vs 4 d [IQR, 1–8 d]; p < 0.0005) but not with duration of PICU stay or PICU mortality (Table 2). Only the occurrence of severe TIAEs was associated with longer duration of MV (5 vs 4 d; p = 0.003), longer PICU stay (15 vs 12 d; p = 0.035), and PICU mortality (17.9% vs 9.6%; p < 0.0001) (Table 2). The association between each severe TIAE and ICU outcomes is shown in Supplemental Table A (Supplemental Digital Content 1, http://links.lww.com/PCC/A380).
A multivariable analysis for individual ICU outcomes (PICU mortality, duration of PICU stay, and duration of MV) was performed, in which we accounted for patient-level factors associated with the occurrence of TIAEs as described in Table 1 and outlined in the legend of Table 3. In this regression model, only the occurrence of severe TIAEs was independently associated with increased PICU mortality (odds ratio = 1.80; 95% CI, 1.24–2.60; p = 0.002). The occurrence of any TIAE or desaturation to SpO2 less than 80% was not associated with an increase in PICU mortality.
The occurrence of TIAE or desaturation to SpO2 less than 80% was associated with an increase in the duration of MV (12% above baseline; p = 0.004). None of the immediate events (any TIAEs, severe TIAEs, or desaturation to SpO2 < 80%) were independently associated with the duration of PICU stay.
Sensitivity analyses with a limited TI dataset from sites that reported greater than 95% of PICU outcome data revealed similar results except that severe TIAEs were independently associated with duration of PICU stay (16% increase, p = 0.049), as shown in Supplemental Table B (Supplemental Digital Content 2, http://links.lww.com/PCC/A381).
Table 4 demonstrates the result of multivariable analyses for duration of MV. The occurrence of TIAE or desaturation to SpO2 less than 80% was independently associated with an increase in duration of MV. The PIM2 score, PICU admission diagnosis as respiratory disease, respiratory failure, or pulmonary toileting (suctioning and clearance of secretions) as an indication for TI were associated with longer duration of MV. Procedural indication or indication for upper airway obstruction was both associated with shorter duration of MV.
Table 5 demonstrates the result of multivariable analyses with severe TIAE as an exposure variable and PICU mortality as an outcome variable. The occurrence of severe TIAE was independently associated with higher PICU mortality. Older age, PIM2 score, cardiac disease, and hemodynamic instability as a TI indication were also associated with higher PICU mortality. Upper airway obstruction or procedural indication as a TI indication was independently associated with lower PICU mortality.
Using data from the NEAR4KIDS Registry database, we have shown that adverse events at the time of TI in the PICU, including oxygen desaturation to SpO2 less than 80%, are associated with important clinical outcomes. On univariate analysis, the combined event of TIAE or desaturation to SpO2 less than 80%, as well as any TIAE (severe or nonsevere), is associated with duration of MV, whereas severe TIAEs are associated with duration of MV and PICU stay as well as mortality. On multivariable analysis, TIAE or desaturation to SpO2 less than 80% is associated with duration of MV and severe TIAEs are associated with mortality even after we adjusted for patient-level confounders including severity of illness.
The association of TIAE or desaturation to SpO2 less than 80% with clinical outcomes raises several important questions. First, can future interventions reduce the occurrence of oxygen desaturations and TIAEs for critically ill children undergoing TI? Second, will these interventions reducing the occurrence of desaturation and TIAEs lead to better ICU outcomes?
In response to the first question, the NEAR4KIDS Network has developed and implemented a TI safety bundle checklist that is currently being evaluated. The bundle was developed by a multidisciplinary QI committee and includes risk factor assessment, TI plan generation, a preprocedure time-out to ensure appropriate preparation, and a postprocedure huddle to identify opportunities for QI (5). It has been recognized that difficult TIs are associated with a higher frequency of desaturations below 80% and adverse TIAEs; recognition prior to intubation may help the team be better prepared and potentially reduce TIAEs (12). The NEAR4KIDS Network is considering future QI interventions to decrease the frequency of TIAEs and oxygen desaturations in critically ill children. These interventions include apneic oxygenation and video laryngoscopy.
Apneic oxygenation, a method to provide oxygen during laryngoscopy, has been suggested as a possible intervention to decrease desaturation during TI. Adult study results are conflicting as to whether apneic oxygenation improves clinical outcomes (13–19). The adult studies that are supportive of apneic oxygenation to date have included high-risk patients but have been observational or had small numbers (13, 15, 17–19). Apneic oxygenation for TI in critically ill children has not yet been studied.
Video laryngoscopy is another technique which may be considered as a possible method to decrease TIAEs. Video laryngoscopy has been shown to be a reasonable alternative to direct laryngoscopy for pediatric patients (20, 21). However, a meta-analysis demonstrated that although video laryngoscopy improved glottic visualization in pediatric patients, the improved visualization was at the expense of a longer time to successful TI and an increase in the TI failure rate (22). The role of video laryngoscopy in TIs in the PICU requires further investigation.
The NEAR4KIDS collaborative has also implemented the checklist to ensure the skillset of the airway provider matches to the risk of TIs (5). Allowing a less-experienced trainee to attempt TI in an emergency situation places the child at increased risk of adverse TIAEs. Multiple attempts at intubation are also associated with increased risk of TIAEs (23). There is an increased frequency of adverse TIAEs when the initial person to attempt TI is a resident rather than a critical care fellow (24). Emergent TIs are a common occurrence in the PICU and are associated with TIAEs (25), in contrast to the operating room, where conditions are controlled, and the intubation is more often nonemergent. The safety of the child requires a more experienced physician to attempt TI initially, potentially leading to fewer opportunities for a trainee to obtain the necessary experience to perform the procedure skillfully. Alternative methods of training, including simulation and controlled settings for elective TI, may be necessary before allowing residents to perform TI in a critically ill child.
To answer the second question, well-thought QI intervention plans and diligent execution with apneic oxygenation, video laryngoscopy with an updated airway safety checklist, and a robust statistical analysis to evaluate the effect of these QI interventions on ICU outcomes are essential.
Our study has several limitations. Only about 80% of the TI reports included ICU mortality, length of PICU stay, and the duration of MV. The data were self-reported, which could introduce reporter bias. Although site-specific compliance plan was in place to ensure complete and accurate reporting, the individual data reported were not reviewed by persons outside each institution. Importantly, under-adjustment of risk factors as confounders may have biased our results. It is possible that the occurrence of desaturation or adverse TIAEs is simply an epiphenomenon in the association between a patient risk factor (e.g., severity of respiratory illness) and ICU outcomes. However, after we adjusted for patient-level confounders including severity of illness and indication for TIs (procedural vs urgent/emergent), the association between desaturation or TIAEs and the duration of MV remained significant. Because this was an observational study, unmeasured confounders may also have affected the results, and we were not able to control for them. Finally, we analyzed the data with composite TI events (i.e., any TIAE or desaturation, any TIAE, severe TIAE) as a priori decided. Each component of TIAEs, however, likely has a different weight in association with patient ICU outcomes, and we were not powered to delineate this difference.
Oxygen desaturations and TIAEs are common with TI in the PICU and are associated with longer duration of MV. Severe TIAEs are also associated with mortality. Although implementing QI interventions to reduce desaturation and TIAEs has its own face value, given the association between these immediate TI events and worse ICU outcomes could be noncausal, the clinical impact of future QI interventions should be rigorously evaluated whether these interventions would actually improve patient ICU outcomes.
1. Jaber S, Amraoui J, Lefrant JY, et al. Clinical practice and risk factors for immediate complications of endotracheal intubation in the intensive care unit: A prospective, multiple-center study. Crit Care Med 2006; 34:2355–2361.
2. Nishisaki A, Turner DA, Brown CA 3rd, et al; National Emergency Airway Registry for Children (NEAR4KIDS); Pediatric Acute Lung Injury and Sepsis Investigators (PALISI) Network: A National Emergency Airway Registry for children: Landscape of tracheal intubation in 15 PICUs. Crit Care Med 2013; 41:874–885.
3. Nett S, Emeriaud G, Jarvis JD, et al; NEAR4KIDS Investigators and Pediatric Acute Lung Injury and Sepsis Investigators (PALISI) Network: Site-level variance for adverse tracheal intubation-associated events across 15 North American PICUs: A report from the national emergency airway registry for children. Pediatr Crit Care Med 2014; 15:306–313.
4. Bowles TM, Freshwater-Turner DA, Janssen DJ, et al; RTIC Severn Group: Out-of-theatre tracheal intubation: Prospective multicentre study of clinical practice and adverse events. Br J Anaesth 2011; 107:687–692.
5. Li S, Rehder KJ, Giuliano JS Jr, et al; National Emergency Airway Registry for Children (NEAR4KIDS) Investigators; Pediatric Acute Lung Injury and Sepsis Investigator PALISI Network Investigators: Development of a quality improvement bundle to reduce tracheal intubation-associated events in pediatric ICUs. Am J Med Qual 2016; 31:47–55.
6. Baker-Smith CM, Wilhelm CM, Neish SR, et al. Predictors of prolonged length of intensive care unit stay after stage I palliation: A report from the national pediatric cardiology quality improvement collaborative. Pediatr Cardiol 2014; 35:431–440.
7. Roddy DJ, Spaeder MC, Pastor W, et al. Unplanned extubations in children: Impact on hospital cost and length of stay. Pediatr Crit Care Med 2015; 16:572–575.
8. Pediatric Acute Lung Injury and Sepsis Investigators (PALISI) Network, 2015. Available at: https://www.palisi.org/members
. Accessed February 8, 2017
9. Nishisaki A, Ferry S, Colborn S, et al; National Emergency Airway Registry (NEAR); National Emergency Airway Registry for kids (NEAR4KIDS) Investigators: Characterization of tracheal intubation process of care and safety outcomes in a tertiary pediatric intensive care unit. Pediatr Crit Care Med 2012; 13:e5–10.
10. National Emergency Airway Registry (NEAR). Available at: http://www.near.edu/
. Accessed November 11, 2015
11. Nishisaki A, Marwaha N, Kasinathan V, et al. Airway management in pediatric patients at referring hospitals compared to a receiving tertiary pediatric ICU. Resuscitation 2011; 82:386–390.
12. Graciano AL, Tamburro R, Thompson AE, et al. Incidence and associated factors of difficult tracheal intubations in pediatric ICUs: A report from National Emergency Airway Registry for Children: NEAR4KIDS. Intensive Care Med 2014; 40:1659–1669.
13. Wimalasena Y, Burns B, Reid C, et al. Apneic oxygenation was associated with decreased desaturation rates during rapid sequence intubation by an Australian helicopter emergency medicine service. Ann Emerg Med 2015; 65:371–376.
14. Vourc’h M, Asfar P, Volteau C, et al. High-flow nasal cannula oxygen during endotracheal intubation in hypoxemic patients: A randomized controlled clinical trial. Intensive Care Med 2015; 41:1538–1548.
15. Miguel-Montanes R, Hajage D, Messika J, et al. Use of high-flow nasal cannula oxygen therapy to prevent desaturation during tracheal intubation of intensive care patients with mild-to-moderate hypoxemia. Crit Care Med 2015; 43:574–583.
16. Semler MW, Janz DR, Lentz RJ, et al; FELLOW Investigators; Pragmatic Critical Care Research Group: Randomized trial of apneic oxygenation during endotracheal intubation of the critically ill. Am J Respir Crit Care Med 2016; 193:273–280.
17. Dyett JF, Moser MS, Tobin AE. Prospective observational study of emergency airway management in the critical care environment of a tertiary hospital in Melbourne. Anaesth Intensive Care 2015; 43:577–586.
18. Ramachandran SK, Cosnowski A, Shanks A, et al. Apneic oxygenation during prolonged laryngoscopy in obese patients: A randomized, controlled trial of nasal oxygen administration. J Clin Anesth 2010; 22:164–168.
19. Patel A, Nouraei SA. Transnasal humidified rapid-insufflation ventilatory exchange (THRIVE): A physiological method of increasing apnoea time in patients with difficult airways. Anaesthesia 2015; 70:323–329.
20. Ilies C, Fudickar A, Thee C, et al. Airway management in pediatric patients using the Glidescope Cobalt®: A feasibility study. Minerva Anestesiol 2012; 78:1019–1025.
21. Fiadjoe JE, Gurnaney H, Dalesio N, et al. A prospective randomized equivalence trial of the GlideScope Cobalt® video laryngoscope to traditional direct laryngoscopy in neonates and infants. Anesthesiology 2012; 116:622–628.
22. Sun Y, Lu Y, Huang Y, et al. Pediatric video laryngoscope versus direct laryngoscope: A meta-analysis of randomized controlled trials. Paediatr Anaesth 2014; 24:1056–1065.
23. Lee JH, Turner DA, Kamat P, et al; Pediatric Acute Lung Injury and Sepsis Investigators (PALISI); National Emergency Airway Registry for Children (NEAR4KIDS): The number of tracheal intubation attempts matters! A prospective multi-institutional pediatric observational study. BMC Pediatr 2016; 16:58
24. Sanders RC Jr, Giuliano JS Jr, Sullivan JE, et al; National Emergency Airway Registry for Children Investigators and Pediatric Acute Lung Injury and Sepsis Investigators Network: Level of trainee and tracheal intubation outcomes. Pediatrics 2013; 131:e821–e828.
25. Rehder KJ, Giuliano JS Jr, Napolitano N, et al; National Emergency Airway Registry for Children and Pediatric Acute Lung Injury and Sepsis Investigators: Increased occurrence of tracheal intubation-associated events during nights and weekends in the PICU. Crit Care Med 2015; 43:2668–2674.
adverse events; intensive care unit; intubation; outcomes; pediatric