Abstract
Neonatal tracheal intubation is a life-saving procedure that is associated with adverse events. The effect of video laryngoscopy on adverse events during neonatal intubation is unknown. In this single site retrospective cohort study, video laryngoscopy was independently associated with decreased risk for adverse events during neonatal intubation.
Background:
Neonatal tracheal intubation is a challenging yet life-saving procedure that is associated with adverse events (1–5). Video laryngoscopy (VL), which improves the view of the glottis during intubation, may lower the risk of such events (5, 6). Studies which investigated the impact of VL on adverse events in the pediatric and adult populations report conflicting results (6, 7). Evidence regarding the effect of VL on adverse events during neonatal intubation is insufficient (8–11). Randomized controlled trials demonstrated improved neonatal intubation success rates for trainees using VL compared with conventional laryngoscopy, but these trials only examined a limited number of adverse events (12, 13). We hypothesized that VL is associated with a decrease in adverse events during neonatal tracheal intubation.
Methods:
Design
We conducted a retrospective cohort study of neonatal intubations performed at our institution between July 1, 2013 and June 30, 2016. We retrospectively queried the National Emergency Airway Registry for Neonates (NEAR4NEOS), a prospectively developed database, for all intubation encounters in our neonatal intensive care unit (NICU). NEAR4NEOS is a multicenter neonatal airway registry that developed from the pediatric airway registry, National Emergency Airway Registry for Children (NEAR4KIDS) (14). Data on patient, provider, and practice characteristics and proximal outcomes are recorded at the time of intubation by the clinical team. The NEAR4NEOS registry is deemed an ongoing Quality Improvement Initiative by the Children’s Hospital of Philadelphia Institutional Review Board (IRB) and thus, this study of NEAR4NEOS data was not subject to IRB approval.
Setting
The Children’s Hospital of Philadelphia (CHOP) is an urban, academic, training hospital in the United States. The CHOP NICU is a level IV, 98 bed referral center with a small percentage of inborn patients with prenatally diagnosed congenital anomalies. Only intubations occurring in the NICU setting were included; intubations occurring in the specialized delivery center or other hospital units were excluded. Endotracheal tube exchanges (i.e. upsizing or replacing an existing endotracheal tube in an intubated patient) were excluded from this analysis. Patients remain intubated throughout most of the tube exchange procedure, and thus the process is distinct from a traditional intubation in a non-intubated patient. Intubations performed by neonatologists, neonatology fellows, pediatric residents and other NICU staff (nurse practitioners, physician’s assistants and hospitalists) were included. Our unit has a general guideline which limits the number of intubation attempts per provider (up to 2 attempts) and encourages the use of premedication for intubation. The types and dosages of premedication are at the discretion of the neonatologist. An attending neonatologist supervises the majority of NICU intubations.
Exposure
The exposure of interest was the first laryngoscopy device used for the intubation encounter: VL vs. conventional laryngoscopy. The most commonly utilized video laryngoscope in our unit was the C-MAC (Karl-Storz, Tuttlingen, Germany). The C-MAC was introduced into our unit in July 2014. The decision to use the C-MAC for intubation was based on the provider’s discretion. Intubations using devices other than the C-MAC video laryngoscope or conventional laryngoscope (i.e., other types of video laryngoscope or fiberoptic bronchoscope) were excluded, as these intubations were rare and typically occurred in patients with difficult airways.
Study Definitions
We utilized NEAR4NEOS operational definitions (1, 6). Briefly, an “encounter” is a single intubation procedure beginning with delivery of premedication and ending 20 minutes after securement of endotracheal tube. A “course” is defined as one method of intubation, including the initial device, approach, and medications used. An “attempt” begins with the insertion of a laryngoscope into the patient’s mouth and ends when the laryngoscope is withdrawn. There can be many attempts during a course and multiple courses within an encounter. Only the first course of each intubation encounter was included in this analysis as the study question related to the initial approach to intubation. If a patient underwent multiple intubation encounters in the NICU throughout the study period, the first course of each of these encounters were included in the analysis.
Adverse tracheal intubation associated events (TIAEs) are categorized by NEAR4NEOS as non-severe or severe. Examples of non-severe TIAEs include: mainstem bronchial intubation, esophageal intubation with immediate recognition, dysrhythmia (including heart rate < 60 beats per minute), lip trauma, and pain or agitation requiring additional medication with delay in intubation. Examples of severe TIAEs include: cardiac arrest with or without return of spontaneous circulation (ROSC), esophageal intubation with delayed recognition, pneumothorax, pneumomediastinum, and laryngospasm. Severe oxygen desaturations are defined as ≥ 20% decrease in SpO2 from the highest value prior to the procedure to the lowest value recorded at any point during the intubation. The highest SpO2 value is obtained just prior to the intubation, during bag mask ventilation and premedication administration (if administered). Severe oxygen desaturation events are collected separately from TIAEs.
The primary outcome of this study was any adverse TIAE occurring during the first intubation course. Secondary outcomes included severe TIAEs, severe oxygen desaturation events, first attempt success rate, number of attempts during the intubation course, and overall success rates in the first course.
Statistical Analysis
Statistical analysis was performed using STATA 14.0 (Stata Corp, College Station, TX). Baseline characteristics between groups who were intubated with VL and conventional laryngoscopy were analyzed using Chi square, Fisher’s exact, and Wilcoxon rank sum tests. Associations between the use of VL and adverse TIAEs, severe oxygen desaturation events, number of intubation attempts, and success rates were investigated using Chi square, Fisher’s exact, and Wilcoxon rank sum tests. Logistic regression was performed to determine the independent effect of VL on the outcomes of all adverse TIAEs, severe TIAEs, and severe oxygen desaturation events. In post-hoc analysis the regression models were adjusted for co-variates that significantly differed (p<0.05) between the VL and conventional laryngoscopy groups in univariable analysis.
Results:
Of 805 tracheal intubation encounters performed during the study period, 644 (80%) were performed with conventional laryngoscopy and 161 (20%) were performed with VL (Table 1). Compared with patients who underwent intubation with conventional laryngoscopy, patients intubated with VL were older (median 40 days, Interquartile Range [IQR] [10–82] vs. median 15 days, IQR [1–30], p<0.001) and larger (median 3.0 kg, IQR [1.9–3.7] vs. median 2.6 kg, IQR [1.6–3.3], p<0.001) at the time of intubation. Upper airway obstruction was a more common indication for intubation using VL. Compared with the conventional laryngoscopy group, the VL group more commonly received sedative/analgesic (89% vs. 78%, p=0.002) and paralytic premedication (80% vs. 44%, p<0.001).
Table 1:
Comparison of patient, practice, provider characteristics and outcomes
| Video Laryngoscopy (n=161) | Conventional Laryngoscopy (n=644) | p value | |
|---|---|---|---|
| Patient characteristics | |||
| Patient age, days (median, IQR) | 40 (10, 82) | 15 (1,30) | <0.001 |
| Birth gestational age, weeks (median, IQR) (n=513) | 33 (27, 38) | 33 (27, 37) | 0.728 |
| Current weight, kg (median, IQR) | 3.0 (1.9,3.7) | 2.6 (1.6,3.3) | <0.001 |
| Male gender | 94 (58) | 369 (57) | 0.803 |
| History of a difficult airway | 23 (14) | 67 (10) | 0.162 |
| Other Indication** | 40 (25) | 195 (30) | 0.175 |
| Practice characteristics | |||
| Sedative/analgesic administration | 143 (89) | 501 (78) | 0.002 |
| Paralytic administration | 129 (80) | 284 (44) | <0.001 |
| Other (hospitalist, physician’s assistant, nurse practitioner) | 91 (57) | 324 (50) | |
| Outcomes | |||
| Any TIAE | 10 (6) | 124 (19) | <0.001 |
| Severe TIAE | 3 (2) | 30 (5) | 0.110 |
| Severe oxygen desaturation | 70/151 (46) | 308/601 (51) | 0.283 |
| First course number of intubation attempts (median, IQR) | 1 (1,2) | 2 (1,3) | <0.001 |
| First intubation attempt success | 101 (63) | 284 (44) | <0.001 |
| First course success | 153 (95) | 625 (97) | 0.203 |
Unless otherwise indicated, values represent patient n (%).
Each intubation may have more than one indication.
Includes shock, procedural indication, hyperventilation, neurological weakness, drug administration (including surfactant), and no airway protective reflex.
Adverse TIAEs occurred in 134 intubations (17%) during the study period. The most common TIAEs were esophageal intubation with immediate recognition (10% of all encounters) and dysrhythmia including heart rate <60 bpm (3% of all encounters) (Table 3; online). Adverse TIAEs occurred less frequently in intubations performed with VL (6%) than conventional laryngoscopy (19%), p<0.001. There was no significant difference in the rate of severe TIAEs or severe oxygen desaturations between groups. Patients in the VL group underwent fewer intubation attempts (median 1, IQR [1–2] vs. median 2, IQR [1–3], p<0.001). Overall success rates of the first course between groups did not differ (95% vs. 97%, p=0.203), but first attempt success was higher in the VL group (63% vs. 44%, p<0.001).
Table 3.
Adverse tracheal intubation-associated events during the study period
| Video Laryngoscopy (n=161) | Conventional Laryngoscopy (n=644) | |
|---|---|---|
| Esophageal Intubation, Immediate Recognition | 2 (1) | 76 (12) |
| Dysrhythmia (includes Heart Rate <60 bpm) | 6 (4) | 18 (3) |
| Mainstem Intubation | 0 | 12 (2) |
| Esophageal Intubation - Delayed Recognition | 0 | 9 (1) |
| Chest Compressions (<1 minute duration) | 1 (<1) | 7 (1) |
| Cardiac Arrest - Survived | 1 (<1) | 6 (1) |
| Gum Trauma | 0 | 6 (1) |
| Emesis, No Aspiration | 2 (1) | 4 (<1) |
| Lip Trauma | 0 | 2 (<1) |
| Pain or agitation, requiring additional medications and delay | 0 | 1 (<1) |
| Emesis with Aspiration | 1 (<1) | 1 (<1) |
| Hypotension Requiring Intervention | 0 | 2 (<1) |
| Laryngospasm | 0 | 1 (<1) |
| Pneumothorax/ Pneumomediastinum | 0 | 1 (<1) |
Note: more than one TIAE could occur during a given intubation.
After adjusting for patient and practice characteristics that differed between groups in univariable analysis, (Table 2), VL remained significantly associated with a reduction in adverse TIAEs (Odds Ratio [OR] 0.43, 95% Confidence Interval [CI] 0.21, 0.87), but not with a reduction in severe TIAEs (OR 0.70, 95% CI 0.19, 2.53) or severe oxygen desaturation events (OR 1.06, 95% CI 0.73, 1.55).
Table 2:
Multivariable analysis of characteristics associated with tracheal-intubation adverse events
| Any Tracheal Intubation Associated Event Odds Ratio (95% Confidence Interval) | Severe Tracheal Intubation Associated Event Odds Ratio (95% Confidence Interval) | Severe Oxygen Desaturation Event Odds Ratio (95% Confidence Interval) | |
|---|---|---|---|
| Video laryngoscopy | 0.43 (0.21, 0.87) | 0.70 (0.19,2.53) | 1.06 (0.73, 1.55) |
| Sedative/analgesic administration | 0.81 (0.48, 1.37) | 0.68 (0.28, 1.65) | 0.84 (0.54, 1.30) |
| Paralytic medication | 0.35 (0.22, 0.56) | 0.24 (0.09, 0.66) | 0.83 (0.59, 1.16) |
| Current weight (kg) | 1.00 (0.86, 1.17) | 1.21 (0.92, 1.58) | 0.88 (0.79, 0.98) |
| Indication for intubation | |||
| Respiratory failure | 1.02 (0.67, 1.56) | 1.72 (0.79, 3.73) | 1.75 (1.26, 2.43) |
| Upper Airway Obstruction | 0.59 (0.21, 1.62) | 1.14 (0.25,5.28) | 2.32 (1.17, 4.60) |
| Unplanned extubation | 0.58 (0.32, 1.05) | 1.17 (0.43,3.17) | 1.99 (1.32, 3.01) |
| First Attempt Provider Role | |||
| Neonatology fellow (Reference) | -- | -- | -- |
| Neonatology attending | 1.60 (0.75,3.42) | 2.80 (0.90, 8.67) | 1.58 (0.86,2.91) |
| Pediatric resident | 2.55 (1.29, 5.05) | 1.12 (0.26,4.85) | 2.05 (1.18, 3.58) |
| Other NICU Staff (Nurse practitioner, Physician’s assistant, Hospitalist) | 1.69 (1.02, 2.79) | 1.58 (0.62,4.02) | 1.20 (0.86, 1.68) |
Numbers in bold represent statistically significant values, p<0.05.
Models include variables that were statistically significant in univariable analysis. Post-natal age was not included, as this is collinear with patient weight.
Discussion:
To our knowledge, this is the first study to report the effect of VL on comprehensive adverse events during neonatal intubation. In this cohort, VL was independently associated with a decreased risk for adverse TIAEs overall, but not with a decreased risk for severe TIAEs or severe oxygen desaturation events. VL was also associated with an improved first attempt success rate and a decreased number of intubation attempts.
To date, the available studies of VL in neonatal intubation have largely focused on success outcomes; there are limited data regarding the influence of VL on adverse TIAEs. O’Shea et al. performed a randomized trial of junior physicians performing neonatal intubation and found improved first attempt success rates when instructors could view the video laryngoscope screen during the procedure, compared with instructors without a visible screen (66% vs. 41%). Lowest oxygen saturation, lowest heart rate, and duration of intubation did not differ between groups (12). In a separate trial, Moussa et al. also demonstrated that residents performing neonatal intubation had higher success rates with VL than conventional laryngoscopy, (75% vs. 63%), but the time to successful intubation was longer with VL (57 vs. 47 seconds). Bradycardic episodes and minimum oxygen saturation did not vary significantly between groups. The rate of mucosal trauma was higher in the conventional laryngoscopy group (4% vs 0%) (13).
After controlling for important baseline patient characteristics that differed between groups, video laryngoscopy remained significantly associated with a decrease in any adverse TIAEs as a group but not severe TIAEs. Severe TIAEs were a rare event, occurring in only 4% of intubation encounters. The largest impact of VL on adverse TIAEs was observed in esophageal intubations with immediate recognition. While in some regards esophageal intubations represent “failed” intubation attempts, they add potential for additional harm from the invasive instrumentation of the upper airway and esophagus during endotracheal tube placement. Thus, we believe that esophageal intubations may be more detrimental to patient safety and should be considered separately from failed intubation attempts when the endotracheal tube is not inserted.
Consistent with our results, a large retrospective cohort study of 8,875 pediatric intubations found that VL was associated with a decreased risk of adverse TIAEs, but not with a decreased risk of severe TIAEs [7]. Similar to our findings, the authors reported a decrease in mainstem intubations and esophageal intubations with immediate recognition when using VL (Table 3; online).
Observational studies and small trials of adult patients have shown improved intubation success rates with VL. However, in a post-hoc analysis of a large randomized trial conducted by Lascarrou et al., VL was associated with higher rates of adverse events (7). It is possible that these contrasting results are due to anatomical differences in the neonatal airway, such as a more anterior glottis, which may lead to greater improvements in outcomes with VL in the neonatal population (15).
VL was not associated with a decrease in severe oxygen desaturations in our study. Previous studies have shown an increased or equivalent time to intubation when using VL compared to conventional laryngoscopy in pediatric and neonatal intubations (13, 16, 17). We did not collect data about the duration of intubation attempts in this study. However, increased duration of intubation attempts with VL is a possible explanation for the lack of effect on severe oxygen desaturation events in this study.
Provider level and paralytic administration were also associated with adverse TIAEs (Table 2), which is consistent with findings from previous studies (1, 18). Our results suggest that VL confers a further protective effect in addition to these factors.
One study limitation is this was a single site study in a level IV referral NICU, because our institution was one of the few NEAR4NEOS sites using VL and collecting data at the initiation of the study period. These results may not be generalizable to all NICUs. A future prospective interventional trial is underway in the NEAR4NEOS network to assess the impact of VL on TIAEs. Also, TIAEs are self-reported by the clinical team in the NEAR4NEOS registry and may underestimate the true TIAE rate.
Study strengths include multivariable analysis of a large number of intubations using prospectively collected data in the NEAR4NEOS registry. Additionally, this study evaluated a comprehensive list of adverse TIAEs using standardized operational definitions.
Conclusions:
We found that video laryngoscopy was independently associated with decreased adverse events during neonatal intubation. These findings suggest video laryngoscopy is a helpful tool to optimize the safety and success of neonatal intubation in the NICU setting.
Acknowledgements:
Special thanks to the National Emergency Airway Registry for Neonates (NEAR4NEOS) Investigators.
References
- 1.Foglia EE, Ades A, Napolitano N, Leffelman J, Nadkarni V, Nishisaki A. Factors Associated with Adverse Events during Tracheal Intubation in the NICU. Neonatology. 2015;108:23–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Hatch LD, Grubb PH, Lea AS, Walsh WF, Markham MH, Whitney GM, et al. Endotracheal Intubation in Neonates: A Prospective Study of Adverse Safety Events in 162 Infants. The Journal of pediatrics. 2016;168:62–6.e6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.O’Donnell CP, Kamlin CO, Davis PG, Morley CJ. Endotracheal intubation attempts during neonatal resuscitation: success rates, duration, and adverse effects. Pediatrics. 2006;117:e16–21. [DOI] [PubMed] [Google Scholar]
- 4.Sauer CW, Kong JY, Vaucher YE, Finer N, Proudfoot JA, Boutin MA, et al. Intubation Attempts Increase the Risk for Severe Intraventricular Hemorrhage in Preterm Infants-A Retrospective Cohort Study. The Journal of pediatrics. 2016;177:108–13. [DOI] [PubMed] [Google Scholar]
- 5.Sawyer T, Foglia E, Hatch LD, Moussa A, Ades A, Johnston L, et al. Improving neonatal intubation safety: A journey of a thousand miles. J Neonatal Perinatal Med. 2017;10:125–31. [DOI] [PubMed] [Google Scholar]
- 6.Grunwell JR, Kamat PP, Miksa M, Krishna A, Walson K, Simon D, et al. Trend and Outcomes of Video Laryngoscope Use Across PICUs. Pediatr Crit Care Med. 2017;18:741–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Lascarrou JB, Boisrame-Helms J, Bailly A, Le Thuaut A, Kamel T, Mercier E, et al. Video Laryngoscopy vs Direct Laryngoscopy on Successful First-Pass Orotracheal Intubation Among ICU Patients: A Randomized Clinical Trial. JAMA. 2017;317:483–93. [DOI] [PubMed] [Google Scholar]
- 8.Singh R, Singh P, Vajifdar H. A comparison of Truview infant EVO2 laryngoscope with the Miller blade in neonates and infants. Paediatr Anaesth. 2009;19:338–42. [DOI] [PubMed] [Google Scholar]
- 9.Mutlak H, Rolle U, Rosskopf W, Schalk R, Zacharowski K, Meininger D, et al. Comparison of the TruView infant EVO2 PCD and C-MAC video laryngoscopes with direct Macintosh laryngoscopy for routine tracheal intubation in infants with normal airways. Clinics (Sao Paulo). 2014;69:23–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Hirabayashi Y, Otsuka Y. Early clinical experience with GlideScope video laryngoscope in 20 infants. Paediatr Anaesth. 2009;19:802–4. [DOI] [PubMed] [Google Scholar]
- 11.Holm-Knudsen R The difficult pediatric airway--a review of new devices for indirect laryngoscopy in children younger than two years of age. Paediatr Anaesth. 2011;21:98–103. [DOI] [PubMed] [Google Scholar]
- 12.O’Shea JE, Thio M, Kamlin CO, McGrory L, Wong C, John J, et al. Videolaryngoscopy to Teach Neonatal Intubation: A Randomized Trial. Pediatrics. 2015;136:912–9. [DOI] [PubMed] [Google Scholar]
- 13.Moussa A, Luangxay Y, Tremblay S, Lavoie J, Aube G, Savoie E, et al. Videolaryngoscope for Teaching Neonatal Endotracheal Intubation: A Randomized Controlled Trial. Pediatrics. 2016;137:e20152156. [DOI] [PubMed] [Google Scholar]
- 14.Nishisaki A, Turner DA, Brown CA 3rd, Walls RM, Nadkarni VM, National Emergency Airway Registry for C, et al. A National Emergency Airway Registry for children: landscape of tracheal intubation in 15 PICUs. Crit Care Med. 2013;41:874–85. [DOI] [PubMed] [Google Scholar]
- 15.Disorders of the neonatal airway : fundamentals for practice. In: Lioy J, Sobol SE, editors. New York: :: Springer New York; 2015. [Google Scholar]
- 16.Sun Y, Lu Y, Huang Y, Jiang H. Pediatric video laryngoscope versus direct laryngoscope: a meta-analysis of randomized controlled trials. Paediatr Anaesth. 2014;24:1056–65. [DOI] [PubMed] [Google Scholar]
- 17.Vlatten A, Aucoin S, Litz S, Macmanus B, Soder C. A comparison of the STORZ video laryngoscope and standard direct laryngoscopy for intubation in the Pediatric airway--a randomized clinical trial. Paediatr Anaesth. 2009;19:1102–7. [DOI] [PubMed] [Google Scholar]
- 18.Haubner LY, Barry JS, Johnston LC, Soghier L, Tatum PM, Kessler D, et al. Neonatal intubation performance: room for improvement in tertiary neonatal intensive care units. Resuscitation. 2013;84:1359–64. [DOI] [PubMed] [Google Scholar]