BACKGROUND: Neonatal endotracheal intubation is a necessary skill. However, success rates among junior doctors have fallen to <50%, largely owing to declining opportunities to intubate. Videolaryngoscopy allows instructor and trainee to share the view of the pharynx. We compared intubations guided by an instructor watching a videolaryngoscope screen with the traditional method where the instructor does not have this view.
METHODS: A randomized, controlled trial at a tertiary neonatal center recruited newborns from February 2013 to May 2014. Eligible intubations were performed orally on infants without facial or airway anomalies, in the delivery room or neonatal intensive care, by doctors with <6 months’ tertiary neonatal experience. Intubations were randomized to having the videolaryngoscope screen visible to the instructor or covered (control). The primary outcome was first-attempt intubation success rate confirmed by colorimetric detection of expired carbon dioxide.
RESULTS: Two hundred six first-attempt intubations were analyzed. Median (interquartile range) infant gestation was 29 (27 to 32) weeks, and weight was 1142 (816 to 1750) g. The success rate when the instructor was able to view the videolaryngoscope screen was 66% (69/104) compared with 41% (42/102) when the screen was covered (P < .001, OR 2.81, 95% CI 1.54 to 5.17). When premedication was used, the success rate in the intervention group was 72% (56/78) compared with 44% (35/79) in the control group (P < .001, OR 3.2, 95% CI 1.6 to 6.6).
CONCLUSIONS: Intubation success rates of inexperienced neonatal trainees significantly improved when the instructor was able to share their view on a videolaryngoscope screen.
- CI —
- confidence interval
- OR —
- odds ratio
What’s Known on This Subject:
Endotracheal intubation is a mandatory skill for neonatal trainees. It is a difficult skill to acquire, and success rates of junior doctors are low and falling.
What This Study Adds:
Videolaryngoscopy allows the supervisor to share the intubator’s view of the airway and provide more informed guidance. Teaching intubation using a videolaryngoscope with the screen visible to the instructor results in significantly higher success rates for inexperienced doctors.
Endotracheal intubation is a common, potentially life-saving intervention for newborn infants with respiratory failure.1 Intubation is a necessary skill for pediatric and neonatal trainees; however, it is a difficult skill to learn and maintain, and initial attempts are often unsuccessful.2–9 Reported first-attempt success rates of intubators with variable experience are 44% to 73%, and residents have the lowest success rates of 20% to 63%.2–9 Three recent studies report success by residents in <25% of attempts.2,3,5 With increasing use of noninvasive respiratory support,10 reductions in trainees’ working hours,11 increasing numbers of trainees, and changes in clinical recommendations, such as discontinuing routine intubation of infants delivered through meconium-stained liquor,1 there are fewer opportunities for neonatal trainees to acquire and maintain proficiency, and their success rates are consequently falling.4
Strategies have been developed to compensate for the reduction in clinical experience. A meta-analysis of studies of technology-enhanced simulation to teach adult intubation showed that this method was superior to no intervention.12 However, studies using simulation to teach neonatal and pediatric intubation have not demonstrated improved clinical performance.13,14 Animal models and cadaveric specimens are useful to demonstrate the anatomy but are very expensive and have limited availability.15
Successful intubation relies on the intubator being able to perform laryngoscopy to obtain a view of the infant’s airway and then recognize the anatomy displayed. Many novice intubators initially find this very challenging. Intubation instruction has traditionally relied on an apprenticeship model, in which a more experienced colleague supervises the novice. However, the instructor’s ability to provide guidance is limited by restricted access to the trainee’s view of the airway (Fig 1). Videolaryngoscopy offers a potential solution to this problem. Videolaryngoscopes use camera technology to visualize airway structures and facilitate endotracheal intubation. A recent systematic review found that insufficient evidence exists to recommend or refute the use of videolaryngoscopy for endotracheal intubation in neonates and called for randomized controlled studies to address efficacy and safety.16 The aim of this study was to determine if supervision using a modified traditional Miller videolaryngoscope improves pediatric residents’ first-attempt neonatal intubation success rates.
Patients and Study Design
This single-center, unblinded, randomized controlled trial was conducted from February 26, 2013, to May 26, 2014 at the Royal Women’s Hospital, Melbourne, Australia, a tertiary perinatal center with ∼7500 births and 300 infants <1500 g admitted to the NICU per year. Infants were eligible if they needed intubation and the intervention was going to be performed orally by a pediatric resident in their first 6 months of tertiary neonatal training. At the start of their neonatal rotation, all residents received intubation training, including practice on neonatal manikins. Their participation in the study was voluntary, and prior verbal informed consent was obtained. The need for intubation was determined by the clinical team and occurred either during resuscitation after birth or in the NICU. Infants were excluded if they had a facial, oral, or airway anomaly or were intubated nasally. The study was approved by the Royal Women’s Hospital research and ethics committees.
Prospective written consent by the infants’ parents or guardians was obtained whenever possible. If delivery was imminent, or the mother was in active labor or was recovering from the birth, it was considered inappropriate to approach before the intubation. Therefore, for infants <48 hours of age, when prospective consent was not possible, a deferral of consent was used as per the Australian National Health and Medical Research Council guidelines for studies in emergency medicine.17 The intubation was randomized, and retrospective consent to use data was obtained as soon as possible after the event. Consent was also requested to randomize further intubations if required. The process of deferred consent was approved by the Royal Women’s Hospital Ethics Committee.
All intubations were performed by using a modified traditional Miller videolaryngoscope (LaryFlex, Acutronics, Hirzel, Switzerland). A flexible fiber-optic cable threaded through the laryngoscope transmitted images from the blade tip to a nearby monitor. Two trolleys containing the videolaryngoscope system were kept within the NICU and the delivery suite. Intubation was performed after direct laryngoscopy with an additional view displayed on a computer-sized monitor (Fig 1). Reusable Miller blades sizes 1, 0, and 00 were used for term infants, preterm infants >1 kg, and infants <1 kg, respectively. The blades and fiber-optic cables were sterilized before each use and kept in sterile, sealed trays.
Premedication with fentanyl, atropine, and suxamethonium was used for elective intubations. A Neopuff Infant Resuscitator (Fisher & Paykel, Auckland, New Zealand) T-Piece was used to provide ventilation. Intubations were performed using sterile, single-use, uniform internal diameter, plastic endotracheal tubes (Mallinckrodt Medical, Athlone, Ireland). A stylet (Satin Slip intubation style, Mallinckrodt Medical) was available to stiffen the endotracheal tube at the resident’s request. Endotracheal tube placement was confirmed by a colorimetric exhaled carbon dioxide detector (Pedicap, Nellcor Puritan Bennett, Pleasanton, CA). Chest radiograph was performed to define tube position.
A computer-generated, variable-size block-randomization sequence was used. Allocation was stratified by the use of premedication (yes or no). Sequentially numbered opaque envelopes containing the randomization cards were kept on the videolaryngoscope trolleys. If an intubation was anticipated by the clinical team, the research team was notified and the equipment was set up. If this subsequently led to an intubation attempt by an eligible doctor, a randomization envelope was opened just before the intubation attempt. The unit of randomization was the endotracheal intubation. Infants reintubated subsequently were eligible for randomization again. However, only the first intubation attempt on each date was eligible for inclusion.
Attempts were supervised by 1 of 6 study investigators (Dr O’Shea, Dr Thio, Dr Kamlin, Dr McGrory, Dr John, and Dr Roberts). All 6 were trained to use the equipment, were shown several intubation recordings, and observed ≥3 supervised videolaryngoscopic intubations before supervising an intubation attempt. In the intervention group, the instructor was able to see the videolaryngoscope screen and offer verbal assistance during the intubation attempt (Fig 2). In the control group, the instructor offered verbal assistance but did not have access to images on the screen. The videolaryngoscope kept in the NICU had an attached laptop with the capacity to record images from the videolaryngoscope screen. Both intervention and control intubations in the premedication stratum were recorded. A standardized proforma on how to guide the intubation attempts was agreed on by all study investigators before study commencement (Fig 3).
A senior clinician who was not a member of the research team attended the intubation and decided when to terminate an unsuccessful attempt. Criteria to stop an attempt included falling heart rate, hypoxia with oxygen saturations <70%, attempt of >60 seconds’ duration, or the attending clinician’s discretion. Standardized debriefing was offered as soon as possible after the attempt. Residents were shown the video of the attempt (if recorded) and were encouraged to reflect on the positive and negative aspects of the attempt. The instructor then advised on what was done well and what could be improved. Residents were then allowed to watch the video again if they wished.
The primary outcome was first-attempt intubation success rate. Secondary outcomes included the infant’s lowest heart rate and oxygen saturation and duration of the attempt (defined as the time interval from insertion of the laryngoscope blade into the infant’s mouth until its removal). An independent data safety monitoring committee reviewed study outcomes after 100 intubations.
On the assumption of an incidence of 50% for the primary outcome,7 we needed 103 infants in each group to have a statistical power of 80% to detect a 20% absolute reduction in the risk of failure of intubation. All analyses were performed on an intent-to-treat basis. Data were analyzed by using Stata software (Intercooled 13, Stata Corp, College Station, TX). The data are presented as mean (SD) for normally distributed variables and median (interquartile range) when the distribution was skewed. The clinical characteristics and outcome variables were analyzed by using χ2 test, t test, and Mann-Whitney U test as appropriate. The results were adjusted for clustering by operator. P values were 2-sided, and values <0.05 were considered statistically significant.
Two hundred thirteen intubations in 168 infants (median 1 intubation per infant, range 1 to 4) were randomized during the study period, and 206 were included for analysis (104 screen visible and 102 screen covered) (Fig 4).
Demographic details of the infants are provided in Table 1. At the time of inclusion, 43% of infants weighed <1 kg. Study intubations were performed by 36 residents. They performed a median of 7 randomized intubations each, range 2 to 11. Details of the residents’ previous intubation experience are provided in Table 2. Images from the videolaryngoscope screen were recorded for 125 intubations (79.6% of premedicated intubations, 60.7% of total study cohort).
The first-attempt intubation success rate when the instructor was able to watch the videolaryngoscope screen was 66% (69/104) compared with 41% (42/102) when the screen was covered: unadjusted odds ratio (OR) 2.81 (95% confidence interval [CI] 1.54–5.17), P < .001; adjusted OR 2.82 (95% CI 1.44–5.52) (adjusted for clustering by resident). When premedication was given, the success rate in the intervention group was 72% (56/78) compared with 44% (35/79) in the control group: OR 3.2 (95% CI 1.6–6.6), P < .001. When no premedication was given, success rates in the intervention and control groups were 50% (13/26) and 30% (7/23), respectively: OR 2.3 (95% CI 0.6–8.8), P = .164. Success rates stratified by level of experience of the resident are presented in Table 3.
Secondary outcomes are presented in Table 4. There were no significant differences in rates of hypoxia or bradycardia or in the duration of the attempt between the intervention and control groups.
The intubation success rates of pediatric residents using direct laryngoscopy improved significantly when an instructor was able to provide guidance based on the shared view of the upper airway. This result was achieved without evidence of harm, as this finding was not associated with increased hypoxia, bradycardia, or a longer duration of intubation attempt.
This is the first study to use a videolaryngoscope to assist junior doctors learning the skills of direct laryngoscopy and intubation in neonates. Infants of a wide range of gestational ages and weights were included. Extremely low birth weight infants were well represented. Both elective (premedicated) and emergency (nonpremedicated) intubations were included. Intubations were individually randomized, thereby reducing selection bias, and a high percentage of all eligible intubations were included (76%). This technique is relevant to other professionals involved in neonatal resuscitation and airway management (eg, respiratory therapists) and could also be used to facilitate training in pediatric and adult intubation.
Videolaryngoscopes have been available for >10 years18 and are now an established tool for acute airway management.19–21 The videolaryngoscope screen displays an improved, magnified, wider laryngeal view (Fig 2) compared with direct laryngoscopy.22 Previously, this technique has typically involved the intubator performing videolaryngoscopy looking at the screen during an intubation attempt rather than performing direct laryngoscopy looking in the patient’s mouth. Experienced intubators’ success rates using videolaryngoscopy in this manner compared with direct laryngoscopy are as high or slightly higher in patients with normal airways,23–25 and significantly higher in patients with anticipated difficult airways.20,26,27 Inexperienced intubators using videolaryngoscopes compared with laryngoscopes had greater success intubating healthy adults with normal airways.28
However, learning to intubate using videolaryngoscopy may not translate into the same ability using a traditional laryngoscope. Videolaryngoscopes take time to set up, need maintenance, and are expensive. Therefore, proficiency needs to be achieved at intubation using direct laryngoscopy. There is only 1 previous randomized study in which the intubator performed direct laryngoscopy and the instructor was either able to see the images on the screen or not, a crossover study performed by Howard-Quijano et al.29 Intubations were randomized in blocks of 6 to either 3 with the screen visible to the instructor followed by 3 with the screen covered or the order reversed. The 6 intubations occurred over several days. There were 37 intubators, medical students or nonanesthetic trainees, all with <6 previous intubation attempts. All intubations were performed on healthy adults with normal airways. The instructors were anesthesiologists trained to teach intubation and use the equipment. The success rate was significantly higher when the instructor was able to see the screen (69% compared with 55%, P = .04).29
Videolaryngoscopes vary in style from modified traditional Miller or Macintosh laryngoscopes to devices with a short angulated blade and guide channel. Our hope was that the resident’s experience performing laryngoscopy with the videolaryngoscope would be comparable to standard direct laryngoscopy so that the skill they learned could translate to standard practice. To achieve this, ideally the laryngoscope handle and blades would be comparable. Several Miller neonatal laryngoscope blades are available.30 They have subtle differences in size and shape but are straight and mostly either flat bottomed or with a slight midline trough.31 We chose the Laryflex videolaryngoscope for the study because it can be used for direct laryngoscopy as well as videolaryngoscopy and its blades most closely resemble commonly used neonatal laryngoscope blades. The blade is straight until a slight downward slope near the tip, and the midline trough is deeper. This necessitates the endotracheal tube curling around a relatively higher lip of the blade to approach the larynx, which is held in a slightly different position. It is possible that these subtle differences could limit translation of the results from our study to standard direct neonatal laryngoscopy. These findings may encourage manufacturers to minimize differences between blades. The device used in this study had a free-standing monitor that was placed alongside the infant’s incubator. This resulted in the instructor looking at the infant while the blade was introduced in the mouth, and then looking away from the infant to see the screen. Other videolaryngoscopes link to smaller screens that can be placed closer to the patient, allowing the instructor to watch the images and the trainee intubating simultaneously. Future devices may improve this design, for example linking wirelessly to a handheld tablet or smartphone.
This study did not assess whether the improved rate of successful intubation when using a videolaryngoscope resulted in retention of the skill when the operator was unassisted. However, recent work by Moussa et al showed that success rates of residents who learned intubation using videolaryngoscopy were maintained when they converted to classic laryngoscopy.32
We found higher success rates when the infant was given premedication beforehand. This is consistent with previous studies that have shown that premedication improves intubating conditions, reduces the number of attempts, and decreases the risk of airway trauma.33–38 It has been previously reported that inexperienced intubators have longer attempt durations than their more experienced colleagues.5–7 The intubation durations reported in this study are similar to those of other studies.6,7 We used a standardized approach to providing instruction and feedback both during and after an attempt. Intubation instruction using traditional methods is challenging in that the instructor’s ability to offer feedback during the attempt is limited. The intervention in this study allowed the instructor to provide accurate, precise, concurrent instruction and feedback during an attempt. This allowed for informed guidance but also quick correction of errors and positive reinforcement of what was being done correctly. As part of a standardized debrief after the intubation, the residents watched video recordings of most of their attempts (both control and intervention intubations). This may have reinforced what they did well and helped explain an unsuccessful attempt. This method was received favorably by the residents and is likely to have contributed to the high success rates found in this study.
There are limitations to this study. Only 1 of a number of available videolaryngoscopes was tested. Results using other devices may differ. The number of instructors was limited to a small core group who were trained to supervise according to specific guidelines. Instructors with less training may be less successful in supervising inexperienced residents.
Currently, pediatric residents and neonatal fellows learning intubation face the challenge of reduced opportunity to practice. A US study found that from 1994 to 2002, the number of intubating opportunities per resident decreased by more than two-thirds, and success rates almost halved.4 The anesthesia literature suggests that ≥40 intubations are necessary to become proficient (defined as success rates of ≥80%).39,40 It is becoming increasingly challenging for trainees to log high numbers of intubation attempts. However, the technique described in this study may enable trainees to become proficient faster. The intervention described in this study has produced the highest reported success rate for novice neonatal intubators. This method, which allows the instructor to share the view of the trainee, may offer a solution to the low and falling intubation success rates of neonatal trainees.
We thank the infants, their parents, and the residents for their participation in the study. We also thank the nursing, medical, and research staff at the Royal Women’s Hospital and the members of the Data Safety Monitoring Committee. Finally, we thank In Vitro Technologies and Acutronics AG for their generous lease of the videolaryngoscope for its use during the trial.
- Accepted August 26, 2015.
- Address correspondence to Dr Joyce E. O’Shea, Department of Neonatology, Maternity Bldg, Royal Hospital for Children, 1345 Govan Road, Glasgow G51 4TF, Scotland, UK. E-mail:
Drs O’Shea and John designed the study with the assistance of Drs Kamlin and Thio and Profs Kuschel and Davis; data collection was completed by Drs O’Shea, Thio, Kamlin, McGrory, John, and Roberts and Ms Wong; Drs O’Shea and Thio and Ms Wong analyzed the data; Dr O’Shea prepared the first draft of the manuscript; and all authors contributed to the editing process and approved the final draft.
This trial is registered with Australian New Zealand Clinical Trials Registry, no. 12613000159752.
FINANCIAL DISCLOSURE: In Vitro Technologies provided a videolaryngoscope for the duration of the study but had no part in study design or data analysis. The authors have indicated they have no financial relationships relevant to this article to disclose.
FUNDING: Funded by The Royal Women’s Hospital, Melbourne, Australia, and the Australian National Health and Medical Research Council Program (grant 606789).
POTENTIAL CONFLICT OF INTEREST: The authors have indicated they have no potential conflicts of interest to disclose.
- Kattwinkel J,
- Perlman JM,
- Aziz K,
- et al
- Falck AJ,
- Escobedo MB,
- Baillargeon JG,
- Villard LG,
- Gunkel JH
- Jagsi R,
- Weinstein DF,
- Shapiro J,
- Kitch BT,
- Dorer D,
- Weissman JS
- Kaplan SJ,
- Carroll JT,
- Nematollahi S,
- Chuu A,
- Adamas-Rappaport W,
- Ong E
- ↵National Health and Medical Research Council. HREC consideration of research proposals involving persons highly dependent on medical care. In: The Human Research Handbook. February 2002, p 55. Available at: http://firstclinical.com/regdocs/doc/?showpage=55&db=INT_Australia_Ethics_Handbook. Accessed September 4, 2015
- Apfelbaum JL,
- Hagberg CA,
- Caplan RA,
- et al.,
- American Society of Anesthesiologists Task Force on Management of the Difficult Airway
- ↵Aziz M, Dillman D, Fu R, et al. Comparing effectiveness of the C-MAC videolaryngoscope versus direct laryngoscopy in the setting of the predicted difficult airway. Anesthesiology. 2012;116:629:36
- Sun DA,
- Warriner CB,
- Parsons DG,
- Klein R,
- Umedaly HS,
- Moult M
- ↵Moussa A, Luangxay Y, Tremblay S, et al. Videolaryngoscope vs classic laryngoscope in teaching neonatal endotracheal intubation: a randomized controlled trial. Pediatr Acad Soc Asian Soc Pediatr Res. 2014; abstract 1445.8.
- Malik MA,
- Subramaniam R,
- Maharaj CH,
- Harte BH,
- Laffey JG
- Jungbauer A,
- Schumann M,
- Brunkhorst V,
- Börgers A,
- Groeben H
- Howard-Quijano KJ,
- Huang YM,
- Matevosian R,
- Kaplan MB,
- Steadman RH
- Kieran EA,
- O’ Donnell CPF
- ↵Moussa A, Luangxay Y, Tremblay S, et al. Videolaryngoscope vs classic neonatal endotracheal intubation skill learned with the videloaryngoscope is maintained with the classic laryngoscope: phase 2 of a crossover randomized trial Pediatr Acad Soc Ann Meeting. 2015; abstract 1595.710
- Lemyre B,
- Cheng R,
- Gaboury I
- Dempsey EM,
- Al Hazzani F,
- Faucher D,
- Barrington KJ
- Roberts KD,
- Leone TA,
- Edwards WH,
- Rich WD,
- Finer NN
- Pereira e Silva Y,
- Gomez RS,
- Marcatto JO,
- Maximo TA,
- Barbosa RF,
- Simões e Silva AC
- Ghanta S,
- Abdel-Latif ME,
- Lui K,
- Ravindranathan H,
- Awad J,
- Oei J
- Copyright © 2015 by the American Academy of Pediatrics