Neonatal endotracheal intubation
- J P Wyllie, Department of Neonatology, The James Cook University Hospital, Marton Road, Middlesbrough TS4 3BW, UK; jonathan.wyllie{at}stees.nhs.uk
- Accepted 24 January 2008
Despite the fact that neonatal endotracheal intubation was described more than 2000 years ago,1 it was only in the 18th and 19th centuries that it began to be accepted as a worthwhile technique for ventilating lungs at birth.2–4 Nevertheless, the practice fell out of favour and many other strange methods were used to resuscitate babies at birth.5 However, in the early 20th century Flagg recommended endotracheal intubation for positive pressure ventilation of newly born babies in the USA6 and Blaikely and Gibberd made similar recommendations in the UK in 1935.7 Conversely, the only trial in humans of intubation at birth compared with the standard technique of resuscitation in a pressure chamber in 1966 found no difference between the two practices.8 It was with the developments in artificial respiration in the 1960s and 1970s that endotracheal intubation became an established part of the developing speciality of neonatal intensive care. Although it is now well established as a requirement for a few babies at birth and for infants receiving positive pressure ventilation, the interest in non-invasive ventilation and continuous positive airways pressure (CPAP)9 as alternatives underlines the lack of randomised human evidence for its use. This article will review some of the evidence and describe the technique of neonatal intubation.
WHY AND WHEN TO INTUBATE?
Endotracheal intubation is either an emergency or an elective/semi-elective procedure and examples of some indications are listed in table 1. Some intensive care units in the UK have experienced a reduction in emergency intubation at birth following the introduction of structured resuscitation training,10 although prior to this intubation rates varied from 1.5% to 12% depending upon the hospital.10 11 As few as 1 in 500 babies may need intubation at birth.12 However, it is clear that with less time for experiential training, a reduction in the skills required for intubation has occurred,13 although even trained practitioners, who regard themselves as skilled, often take longer than expected to intubate.14 As a procedure with known complications (box 1), it is therefore essential that all endotracheal intubations are justifiable and carried out in as controlled an environment as possible.
Clearly babies who require ventilatory support may need endotracheal intubation, although the widespread use of antenatal steroids for threatened preterm delivery has reduced the need in more mature babies. This has also been related to an increased use of CPAP which is the preferred initial mode of support in many units and was recently tested in the COIN trial (CPAP Or INtubation at birth), the results of which have yet to be published.
ANATOMY
The airway anatomy of babies is different from that of older children and adults. The neonatal glottis can be difficult to visualise because of the relatively large tongue, high anterior larynx and narrow, u-shaped epiglottis. Straight laryngoscope blades have been developed to provide a direct view, keeping all intermediate structures aside. The narrowest part of the airway is at the cricoid ring where the tracheal tube will fit closest to the walls of the airway.
NEONATAL PHYSIOLOGY
It is important to realise that in comparison to older children and adults, neonates have proportionately greater oxygen consumption and a small reserve. Desaturation during apnoea occurs rapidly and is quickly followed by bradycardia. The presence of a patent foramen ovale and right sided heart pressures which will be equivalent to left at birth may contribute to rapid systemic desaturation during intubation.
EQUIPMENT
A non-exhaustive list of appropriate equipment for intubation is given in box 2. The use of appropriately sized endotracheal tubes minimises trauma, airway resistance and excessive leak around the tube. Approximate appropriate endotracheal tube sizes are given in table 2, but the endotracheal tube should be a snug fit as it is traditionally uncuffed due to concern that the tracheal mucosa may be damaged, giving rise to subglottic stenosis. If a stylet is used, it must not protrude beyond the end of the endotracheal tube. The two main types of endotracheal tube used in the UK are made by Portex and Vygon and each has its supporters. The latter is softer and thicker for a given internal diameter.
Masks for ventilation prior to intubation or after a failed attempt should be soft and mould to the face. Correct use of these is important and achieving a seal may need tuition even for experienced practitioners whose abilities may not be as expert as they suppose.15
Ideally, it should be possible to regulate the volume which is used to ventilate babies during intubation and resuscitation and limit it to 4–8 ml/kg, but most systems only permit the limitation of pressure by means of a spring-loaded or caged magnet “pop-off” valve. These systems can be over-ridden by vigorous squeezing of the bag which determines flow. The most reliable way to deliver a set flow of gas at a set pressure is via a T-piece system such as the Tom Thumb (Viamed, Keighley, UK) or Neopuff (Fisher & Paykel, Auckland, New Zealand).16 17
A straight bladed laryngoscope is used for intubating due to the anatomical differences in small babies. Until competence is acquired trainees require close supervision. It is clear that one 4–6-month period as a junior doctor on a neonatal unit does not equate with competence at neonatal intubation.13
ENDOTRACHEAL INTUBATION
Intubation is a practical skill which cannot be learnt from a text book. The best way to learn and maintain the skill is in a controlled environment with senior supervision. This may be in a neonatal intensive care unit or on a neonatal anaesthetic list.
All equipment should be within easy reach and assistance available. Consideration must be given to maintaining the baby’s temperature during the procedure. The baby should be positioned on a flat surface with the neck partially extended. A roll under the shoulders can help to maintain the correct position. Over-extension is a common mistake which stretches the trachea and moves the larynx anteriorly, making it difficult to visualise. Holding the laryngoscope in the left hand while opening the mouth, gently insert the blade, taking care not to damage the gums. Most practitioners use a variation of the paediatric technique, inserting the blade down the right side of the mouth and moving the tongue to the left, advancing the blade until the tip sits just beyond the base of the tongue. In many babies, especially those who are small or premature, the tongue is relatively fixed to the floor of the mouth and it may be easier to insert the blade centrally in the mouth.
Looking down the blade, clear any secretions which obscure the view, and place the tip of the laryngoscope in the valeculla at the base of the epiglottis where it meets the tongue. Lifting the entire blade forwards slightly brings the larynx into view behind and under the epiglottis (fig 1A). The blade should not be levered as this will apply pressure to the upper gum and make the intubation more difficult. External pressure applied to the larynx may help to bring the laryngeal opening into the centre of the field of view (fig 1B). This external pressure is best applied by an assistant who will find it easier to keep the pressure central and controlled.
Box 1 Complications of endotracheal intubation
-
Pharyngeal, oesophageal and tracheal trauma or perforation
-
Intubation of mainstem bronchus
-
Subglottic stenosis
-
Accidental extubation
-
Post-extubation atelectasis
-
Respiratory infection
-
Erosion of nares or septum (nasal intubation)
-
Palatal grooving (oral intubation)
Bring the tip of the endotracheal tube in from the right corner of the mouth so as to avoid obscuring the view. If the vocal cords are closed in an emergency intubation, wait for them to open and then insert the tube until the vocal cord guide is at the level of the cords (fig 1C).
Hold the tube firmly while removing the laryngoscope. If a stylet has been used remove the stylet, commence or recommence ventilation and fix the tube in place.
Box 2 Equipment for intubation
-
Laryngoscope with appropriate flat blades
-
Endotracheal tubes, sizes 2.5, 3.0, 3.5 and 4.0 mm internal diameter
-
Appropriate stylet or introducer
-
Suction apparatus
-
Suction catheters including a Yankauer
-
Soft fitting face masks
-
Magill neonatal forceps
-
Self inflating bag or T-piece gas delivery system
-
Source of compressed air/oxygen with blender
-
Appropriate fixing device
-
Exhaled carbon dioxide detector
-
Cardio-respiratory monitor/saturation monitor
Confirmation of endotracheal intubation
From the previous description it is clear that the practitioner should have observed the endotracheal tube in place. This is the first clinical determination of successful intubation. Secondary evidence comes from a rising heart rate if it was low or maintenance of an acceptable heart rate and other physiological parameters. The best secondary evidence, if there is any doubt, comes from the presence of exhaled carbon dioxide as determined by a carbon dioxide detector or capnography.18 The use of an exhaled carbon dioxide detector is the only technique which has been evaluated for confirmation of endotracheal tube placement in neonates.19–22 Other techniques such as evaluation of condensed humidified gas during exhalation, chest movement and auscultation have been used but have not been evaluated in rigorously designed studies in neonates and are poor indicators in adults.23
In newborn babies a positive test indicating exhaled carbon dioxide confirms correct placement of the endotracheal tube within the trachea, whereas a negative test is strongly suggestive of oesophageal intubation.19 21 Poor or absent pulmonary blood flow may give false negative results, but tracheal tube placement is correctly identified in nearly all patients who are not in cardiac arrest. Exhaled carbon dioxide detectors identify oesophageal intubations faster than using clinical assessments in a research setting.19 20
All intubations should therefore have an exhaled carbon dioxide detector available, however it is important to realise that there are weight and clinical restrictions for some detectors. If colourimetric indicators are used below the recommended weights, then false negative readings are possible. This can be overcome by compressing the chest to ensure sufficient gas flow, which may be helpful in an emergency situation. Colourimetric detectors may also give false positive results if exposed to adrenaline or Curosurf.24
Time to intubate
Several texts and guidelines stipulate that neonatal intubation should take no more than 20 s. Beyond that time the baby should be re-oxygenated before a further attempt is undertaken. This is clearly not possible if the intubation is an emergency following failure of airway management. There is now evidence that even experienced practitioners require more than 20 s in an emergency situation.14 Sometimes it is recommended that the practitioner holds their breath and curtails the attempt when they need to breathe. This is counter-intuitive as a hypoxic practitioner is surely less likely to succeed and as a practice it cannot be recommended. The 20–30 s period seems reasonable in the elective situation with physiological monitoring and clearly endotracheal intubation should be achieved in the shortest possible time in all situations. However, in the monitored situation it should be the aim to effect intubation without saturations or heart rate falling outside the target ranges.
Routes of intubation
Intubation can be performed orally or nasally. The choice of route depends upon the circumstances and the preference of the clinician or practice in the department. Oral intubation in neonates is faster, more likely to be successful at the first attempt and less traumatic to perform25 and should always be used first in an emergency intubation. However, there is no difference between nasal and oral intubation in terms of tracheal injury, need for refixation or re-intubation.26 In neonates <1500 g who are nasally intubated, there is a higher incidence of post-extubation atelectasis, which seems to be associated with reduction in nasal airflow through the nares.27 28 Proponents believe that nasal intubation permits better fixation with less tube movement and a lower incidence of subglottic problems, but evidence for this comes from only two studies in the 1970s and 1980s29 30 and there may be potentially more tube movement with nasal tubes.31 Oral intubation can cause palatal grooving, which tends to resolve spontaneously, although longer term problems with dentition have been reported.32 33
Differences between emergency and elective intubation
The major difference between these two situations is time. In an emergency there is usually pressure on the practitioner to reverse a bad situation which is often due to ineffective airway control. All other avenues should have been tried, but it is often worth making sure that they have in fact been carried out effectively, as it may be possible to move to a semi-elective intubation if the heart rate can be brought above 100 bpm with better airway management techniques using mask ventilation. When senior help is summoned in an emergency, airway intervention is the most common action taken.
In an emergency it will usually be inappropriate to use sedation and unnecessary if the baby is unconscious. However, it is essential to determine that the endotracheal tube is in place, and in the absence of clear clinical signs an exhaled carbon dioxide detector should be used. Confirmation of tube position in a patient failing to respond moves the resuscitation forward instead of wasting time with the old adage “if in doubt pull it out”. The limitations of a negative finding in the absence of pulmonary flow or in extremely low birth weight babies must be understood.18
In elective or semi-elective intubations there is more time and the whole process can be controlled. The baby can be sedated and the entire team should understand what is to happen. If it is a tube change, then it is possible to leave the airway unprotected for the minimum amount of time. This is done by sedating and/or paralysing (see below) the neonate appropriately and continuing to ventilate down the old endotracheal tube while positioning the new one. It is then possible to take out the old and immediately insert the new, leaving the airway unprotected for seconds at most and hardly interrupting ventilation. This is most straightforward when changing an oral to a nasal tube but can also be done when changing oral or nasal tubes.
In an elective intubation care must be taken to keep physiological parameters as close to the target range as possible. The ideal therefore is for the heart rate to stay above 100 bpm and the oxygen saturations to remain within the target range. Usually this is focused upon keeping the saturations above a baseline level, but in premature neonates it is as important to avoid very high saturations which may put the neonate at increased risk of retinopathy of prematurity and possibly bronchopulmonary dysplasia.34
As endotracheal intubation is such an established part of neonatal intensive care, there may be a tendency to underestimate the importance of elective or semi-elective intubation. Endotracheal tube changes may be seen as routine. It must be understood by all members of the team that such procedures are potentially hazardous and often represent the period of greatest predictable risk to babies receiving intensive care.
Sedation
In 1998 most neonatal units in the UK did not give any sedation for elective or semi-elective endotracheal intubation and only 14% had a written policy.35 This was despite evidence of physiological and practical benefits.36–38 A similar situation has been reported in the USA38 39 and France.40 The need for induction of anaesthesia at intubation is accepted in paediatric and adult practice in all but emergency situations and yet uncertainty persists in neonatal practice. Barrington and Byrne reported on the safety and practicability of premedicating 253 neonates for intubation, using a regimen of atropine, fentanyl and suxamethonium, in a unit with the variable skill mix and experience that is inevitable in daily practice.41 Many units who give sedation use morphine, probably because of familiarisation rather than proven efficacy. Roberts et al demonstrated the efficacy of mivacurium, a muscle relaxant, in a two centre randomised controlled trial recruiting 41 infants.42 The addition of mivacurium decreased the time and number of attempts needed to successfully intubate while significantly reducing the incidence of severe desaturation. More recently, a randomised study by Ghanta et al compared propofol with morphine, atropine and succinylcholine (suxamethonium) and found that it was more effective as an induction agent with less hypoxia, reducing the time to achieve successful intubation.43 While further studies are necessary to establish the optimum drug regimen and dosages, there is sufficient evidence to support the routine practice of neonatal premedication prior to elective intubation.
For non-emergency intubation, neonatal units should have a written policy outlining sedation and/or muscle relaxants to be administered. The evidence supports atropine (10–20 μg/kg), fentanyl (2–4 μg/kg) and mivacurium (200 μg/kg) or propofol (2.5 mg/kg) as acceptable regimes. However, midazolam (100 μg/kg), succinylcholine (2 mg/kg), atracurium (300–500 μg/kg), pancuronium (100 μg/kg) and rocuronium (600 μg/kg) have also been used. The latter has the potential advantage of a rapid onset (<60 s) and an intermediate duration, but extensive literature covering the neonate and premature baby is lacking.
Endotracheal tube length
In addition to direct visualisation of the tube as it passes through the glottis, there are a number of different methods for estimating the depth of endotracheal tube placement. The easiest to remember is the rule of 7-8-9 (7 cm at the lips for 1 kg, 8 cm for 2 kg and 9 cm for 3 kg babies).44 For nasotracheal tubes add another centimetre. However, this rule must be used with great caution in babies below 750 g.
The position of the patient will cause endotracheal tubes to move,31 so if chest radiograms are used to prove correct placement, babies’ heads should be in a standard position as flexion of the neck decreases and extension increases both nasal–carina and oral–carina distances. Lateral rotation produces no significant changes.45
Several studies have described inappropriate endotracheal tube position in up to 30% of patients. Techniques with a proven accuracy of greater than 80% are foot length for nasotracheal46 and nasal–tragus length for oral intubation.47 Weight or length based charts derived from local demographic data can be equally accurate. However, audit within our institution has demonstrated that estimation from previous radiographs is the most accurate technique.
Fixation
There is little evidence which can be used to recommend any one method of endotracheal tube fixation. There are advantages and disadvantages to both oral and nasal routes. There has only been one randomised controlled trial of endotracheal tube fixation demonstrating the superiority of one technique over another48 and there are therefore almost as many techniques as intensive care units. As accidental extubation represents a potentially avoidable risk to any ventilated neonate, this area warrants further structured study.
Training
Endotracheal intubation is a practical skill for which there are still no adequate models upon which to train. Some manikins are misleading and the skill is best learnt in the clinical situation with full support. Administration of an anaesthetic or sedation and paralysis of the patient makes such training more controlled and increases the success rate. Teaching on babies after death may raise more objections amongst professionals than parents.49 50 It is certainly difficult to approach parents at this time, but there is no reason why it should cause any offence and it has certainly been my experience that some parents have expressed satisfaction that their tragedy has improved the training of a professional who wished to look after babies.
CONCLUSION
Endotracheal intubation is a practical skill which has become an integral part of neonatal intensive care. All neonatologists should be able to perform this technique but there remains uncertainty as to the best methods to use and the most effective type of training. More research is required in almost all of the areas covered by this article.
