Sie sind auf Seite 1von 7

Pediatric Anesthesia 2008 18: 525–531 doi:10.1111/j.1460-9592.2008.02530.

Effects of head posture on the oral, pharyngeal and

laryngeal axis alignment in infants and young
children by magnetic resonance imaging
R E N A U D V I A LE T M D * , A N D R É N A U M D †,
K A T H I A C H A U M O Î T RE M D ‡ A N D C L A U D E M A R T I N MD†
*Réanimation Pédiatrique et Néonatale, Département d’Anesthésie-Réanimation CHU Nord Bd.
P. Dramard, Marseille, †Département d’Anesthésie-Réanimation, CHU Nord Bd. P. Dramard,
Marseille and ‡Département d’Imagerie Médicale, CHU Nord Bd. P. Dramard, Marseille, France

Background: Objective anatomical studies supporting the different
recommendations for laryngoscopy in infants and young children are
scarce. The objective of this study was to measure by magnetic
resonance imaging (MRI) the consequences of head extension on the
oral, pharyngeal and laryngeal axes in infants and young children.
Methods: Thirty patients (age: 33 ± 28 months; weight 14 ± 9 kg),
under general inhalated anesthesia delivered via a laryngeal mask
airway, were studied in two anatomic positions: head in the resting
position and in simple extension. The following measurements were
made on each scan: the face and the neck axes, the pharyngeal axis, the
laryngeal axis, and the line of vision of glottis. The various angles
between these axes were defined: d angle between line of vision and
laryngeal axis, and b angle between pharyngeal axis and laryngeal
axis. From an anatomical point of view, laryngoscopy and passage of a
naso-tracheal tube would be facilitated if these angles are narrow.
Results: Placing the patient from the resting position into extension
led to a narrowing of the angle d but a widening of the angle b.
Conclusions: In infants and young children, under general anesthesia
and with a laryngeal mask airway in place, just a slight head extension
improves alignment of the line of vision of the glottis and the
laryngeal axis (narrowing of angle d) but worsened the alignment of
the pharyngeal and laryngeal axes (widening of angle b).

Keywords: head ⁄ *anatomy & histology; neck ⁄ *anatomy & histology;

magnetic resonance imaging; newborns; infants; posture

the operating room) are major concerns for pediatric

anesthetists and intensivists. Optimal head position
Optimal glottic visualization and easy-to-perform is commonly supposed to permit the alignment of
naso-tracheal intubation (as commonly performed in the oral, pharyngeal and laryngeal axis (1). Optimal
head positioning for laryngoscopy and intubation is
Correspondence to: R. Vialet, Réanimation Pédiatrique et Néonatale,
Département d’Anesthésie-Réanimation CHU Nord Bd. P. controversial. The guidelines recommended by the
Dramard 13 915, Marseille, France (email: experts vis-à-vis head position for laryngoscopy in

 2008 The Authors

Journal compilation  2008 Blackwell Publishing Ltd 525
5 26 R . V I A LE T ET AL .

children state that the head must be slightly dysmorphia were further considered for inclusion.
extended (2–5) or extended with elevation of the Parents were informed that the study required
occiput (1,5–8). It is also frequently stated that for additional images (i.e. images of the neck) with a
newborns, infants or children with enlarged heads, short prolongation in the duration of the examina-
padding below the occiput may not be required to tion, and an extension of the head performed by the
obtain the so-called ‘sniffing position’ (5,8–10). For anesthetist between two sequences. If parental con-
other experts, the sniffing position, even in the sent was obtained, the patient was included in the
newborn, requires elevation of the head (1,6) or the study. Standard preoperative examination was per-
shoulders (11). Not all of the recommendations are formed as usual and the anesthetic protocol was not
widely accepted since some experts recommend modified. The Ethics Committee of our institution
positioning the head in the resting position (7). In approved this study.
addition, when naso-tracheal intubation is per- Sevoflurane sedation was induced in an anesthe-
formed, head flexion is usually but not universally sia room outside the magnetic field. Pulse oximetry,
admitted as facilitating laryngeal passage of the tube heart rate, respiratory rate and capnography were
(1,3,7,12). monitored from the outset and throughout the
Despite an abundance of articles in the literature procedure. An intravenous cannula was inserted.
on the anatomical particularities of infants and Insufflation started with 2–7% sevoflurane (vapor-
children, objective studies supporting the theoretical izer setting). Intravenous access and laryngeal mask
background and the clinical relevance of these airway (LMA) insertion were performed under deep
different recommendations for laryngoscopy in general anesthesia, and then sevoflurane adjusted
infants and little children are scarce. Magnetic appropriately following conscious level. Once set-
resonance imaging (MRI) has greatly contributed to tled, patients were transferred to the scanner and
our knowledge of larynx development in children pulse oximetry, capnography, heart and respiratory
(13) and larynx conformation under sedation (14) were continuously monitored with a monitor placed
and permitted objective study of head positioning in an observation room. Warm coverings and ear
for intubation in adults (15). Our hypothesis was defenders were systematically used. After scanning,
that anatomic angles measurements in different infants were transferred to a recovery area away
positions can afford anatomical background to sup- from the magnetic field.
port methods to facilitate glottic visualization and The child lying in supine position directly on the
intubation. The main objective of this study was to flat MRI table, head was placed by the anesthetist at
measure by MRI the consequences of head extension first close to the resting position, i.e. axis of the face
on the oral, pharyngeal and laryngeal axes in (passing by the brow and the chin) close to the
children. The secondary objectives were to assess if horizontal plane. Head extension was performed by
head extension solely places the children in the the anesthetist who placed the head of the child as if
‘sniffing position’ and how head extension modifies he wanted to perform the intubation of the patient
the alignment of the pharyngeal and laryngeal axes. (but without any laryngoscopy attempt). Head
position was maintained by padding over the top
of the head with adhesive tape. No head ring nor
pillow were used. Only two anesthetists participated
Over an 11-month period, all children who had a in this study. In order to improve adjustment of their
scheduled cerebral MRI under general anesthesia practice, they positioned together the first nine
with at least two different acquisition sequences in included patients.
our hospital were considered for inclusion. MRI scan All of the MRI studies were performed with a 1.5
and indication of general anesthesia (to enable the Tesla system (Maestro Class; Siemens, Erlanger,
child to stay motionless during the MRI scan) was Germany) with an ear–nose–throat coil. The acqui-
indicated by pediatricians not involved in the study. sition technique was a spin-echo sequence with a
Following the standard preoperative examination, repetition time of 703 ms and echo time of 13 ms.
only patients graded ASA 1 with no expected T1-weighted images were obtained in the sagittal
difficult airway management and no facial or head plane.

 2008 The Authors

Journal compilation  2008 Blackwell Publishing Ltd, Pediatric Anesthesia, 18, 525–531
H E A D P O S T U R E A N D A IR W A Y I N I N F A N T S A N D C H I L D R E N 527

As proposed by Adnet et al. in a previous study in passing through the centers of the inferior (cricoid
adults (15), the following measurements (Figure 1) cartilage) and superior (base of epiglottis) laryngeal
were made on each MRI scan in resting position then orifices, and line of vision of glottis (LV) passing by
in extension: axis of the face (passing by the brow the lower end of the upper incisors and by the
and the chin), axis of the neck (passing by the former corniculate cartilage. The following angles were
end of the cricoid cartilage and the sternal fork), measured: angle of the face and the neck with the
pharyngeal axis (PA) passing through the anterior horizontal axis (respectively Face angle and Neck
portion of the atlas and of C2, laryngeal axis (LA) angle) to quantify extension; the angle d between LV
and LA, (theoretically, laryngoscopy would be facil-
itated if d angle is narrow); the angle b between PA
(a) and LA (pharyngeal-laryngeal alignment) were
measured (theoretically, passage of a naso-tracheal
tube would be facilitated if b angle is narrow) (15).
The amplitude of head extension was measured
for each individual patient by calculating the differ-
ence in Face angle before and after extension. The
effects of head posture were assessed by plotting
respectively d and b on Face angle. Statistical
analysis was performed using the t-test for paired
comparisons, linear regression and analysis with
analysis of variance to test the null hypothesis that
the slope of the best fit lines was equal to zero. (SPSS
version 13.0.1; SPSS Inc., IL, USA). A P-value of <
0.05 was considered as statistically significant.

Thirty patients with an average age of
33 ± 28 months (range: 1.5–105 months) and weigh-
ing 14 ± 9 kg (range: 6–32 kg) were studied. Weight
and age distribution are displayed in Figure 2. No
LMA cuff was positioned in the pharynx, and the
upper end of the cuff had a location opposite to the
C1 or C2 vertebra, what indicated that the LMA
were correctly positioned. Positioning the patient in
the resting position actually put the face and the
neck in slight flexion (Table 1) but axis of the face
was close to the horizontal plane. Positioning the
head in extension led to a slight face extension and a
neck flexion (Table 1). The amplitude of head exten-
sion, despite the fact that it was significant, was not
very marked (Table 1). Placing the patient from the
resting position into extension led to a narrowing of
the angle d but a widening of the angle b (worsening
of the alignment of the pharyngeal–laryngeal axis)
(Table 1). Figure 3a and b shows that extensions of
Figure 1 up to 28 were studied and that variations of angle d
Evolution of d and b and angles from the resting position (a) to
head extension (b). MA, mouth axis, PA, pharyngeal axis, LA, (Figure 3a) and b (Figure 3b) followed an inverse
laryngeal axis, LV, line of vision. linear relationship depending on head extension.

 2008 The Authors

Journal compilation  2008 Blackwell Publishing Ltd, Pediatric Anesthesia, 18, 525–531
5 28 R . V I A LE T ET AL .

Table 1
8 Head and neck angles to horizontal axis (in degrees)

Resting Paired differences

position Extension (extension – resting position)

6 Face extension )9 ± 7 a 11 ± 8 20 ± 9 *
Neck flexion 17 ± 9 22 ± 8 5±9*
d 38 ± 7 18 ± 8 19 ± 7 *
b 8±4 17 ± 5 )9 ± 7 *
n a
4 a negative value of extension indicates flexion.
*P < 0.001 vs resting position.

0 10 20 30 40 50 60 70 80 90 100 110
Age (months)




0 5 10 15 20 25 30 35
Weight (kg)

Figure 2
Age and weight histograms.
Figures 3
(a) When head extension (Face angle with the horizontal axis)
Figure 3a also indicates that obtaining a narrow d increases, d decreases (alignment between line of vision and
angle (< 15) was associated with only slight head laryngeal axis is improved) (b) but b angle increased (alignment
extension (15–25) (filled circles, Figure 3a). And between pharyngeal axis and laryngeal axis worsened). Values are
in degrees. Filled circles indicates an angle of < 15.
Figure 3b indicates that obtaining a narrow b angle
(< 15) was associated with a head extension of 0 or and hardly increased for b angle (Figure 4b), but
less (i.e. a head flexion) (filled circles, Figure 3b). these trends were not statistically significant (slopes
According to the age of the patient, variation during of the best fit lines do not differ to zero, P = 0.15 and
extension hardly decreased for d angle (Figure 4a) P = 0.75, respectively).

 2008 The Authors

Journal compilation  2008 Blackwell Publishing Ltd, Pediatric Anesthesia, 18, 525–531
H E A D P O S T U R E A N D A IR W A Y I N I N F A N T S A N D C H I L D R E N 529

(a) and challenged much of the logic of the sniffing

30 position. MRI scans were also used in adults to
study the airway obstruction induced by anesthesia
20 (16). Shorten et al. used MRI to study pediatric
airway in sedated children and studied airway
patency with two different positioning techniques
0 (sniffing position and shoulder elevation) (14). Hud-
gins et al. used computed tomographic (CT) scans,
–10 but the purpose of their work was to study the
normal pediatric larynx with images obtained in the
axial plane (17). Arens et al. used MRI to study sleep
airway obstruction but only reported data on the
growth relationships of the bones and the tissues
0 20 40 60 80 100 surrounding the upper airway (13). An anatomic
Age (months) study with the same topic as ours was performed
(b) using conventional radiography of the airway (10).
30 But to our knowledge, it is the first time that MRI
scan has been used to study the pediatric airway
during head posture modification in the perspective
of evaluating intubation conditions.
In the present study, the resting position actually
0 resulted in slight flexion in terms of Face angle to the
horizontal plane. Actually, the resting position has
–10 never been quantitatively defined nor precisely
standardized. In conscious patient the head at rest
shows some degree of flexion (18,19), and such
–30 flexion in the resting position (or ‘neutral position’)
has been previously described in unconscious chil-
0 20 40 60 80 100
dren anesthetized for scanner procedures (17). Head
Age (months)
extension was not complete, but in a manner
Figures 4 previously described during laryngoscopy (20–22),
(a) Changes of d and (b) b values during extension with age. and cervical extension almost reached previously
described in the sniffing position (15) or extension
during laryngoscopy (20,21,23). At last, laryngeal
extension observed in our study are qualitatively
comparable to previously published data (24). So,
The main result of our study is that head extension despite the fact that the head positioning was made
in infants and young children improves alignment of without laryngoscopy, extension of the head ob-
the line of vision of the glottis and the laryngeal axis served in our study was similar to those obtained
(narrowing of angle d). The reduction of angle d by during laryngoscopy. The large variability of our
head extension was greater in our study (19 vs 13) results is certainly due to the lack of standardization
than those reported by Adnet in adults (15). This of head extension. In the present study, head
confirms that in infants and young children, just a extension did not make it possible to obtain the
slight head extension provides good alignment of admitted values of 15 of head extension and 35 of
the line of vision of the glottis and the laryngeal axis. neck flexion for an actual sniffing position (12,25).
In 2001, Adnet et al. (15) used MRI for the first However, we observed that simple head extension is
time to study the conditions of intubation in eight associated with neck flexion. Our results support the
adults, and this work brought major objective fact that compared with adults, a simple head
elements to the comprehension of the technique extension in infants and young children brings head

 2008 The Authors

Journal compilation  2008 Blackwell Publishing Ltd, Pediatric Anesthesia, 18, 525–531
5 30 R . V I A LE T ET AL .

and neck angles closer to this position, and confirms are not equivalent (27–30). The term ‘line of vision’, as
that occipital padding to improve elevation of the defined by Adnet et al. (15), should not mislay the
head is not necessary to fulfill requirements of the reader, this line is an anatomic line which corresponds
sniffing position. Since the variations of the d and b only in theory to visualization, and may not reflect
angles throughout extension in this study were close actual visualization during laryngoscopy. The line of
to those reported by Adnet in adults (15), the vision as determined by MRI and the real line of vision
explanation is certainly due to the relative during laryngoscopy are most likely different due to
macrocephaly. soft tissue displacement during laryngoscopy, and
During extension, the b angle increased (like the our results should not be considered as surrogate
PA axis tips posteriorly). Our study indicates that a measures for optimal glottic visualization. However,
better alignment of the pharyngeal and laryngeal MRI assessments primarily performed on a small
axes is obtained with a slight flexion of the head, and number of patients using the same theoretical
that nasotracheal intubation may be made more approach (15) later showed clinical relevance in a
difficult when head is extended. large prospective randomized study (31).
The present study has several limitations. As In conclusion, the main findings of the present
patients were under general anesthesia, possible study are that in infants and young children under
modifications of anatomical structures induced by general anesthesia and a LMA in place, head
anesthesia were taken into account, but anatomical extension improves alignment of the line of vision
modifications due to the LMA could not be ruled of the glottis and the laryngeal axis, and that the
out. Changes of the surrounding anatomic struc- better alignment of the pharyngeal and laryngeal
tures due to LMA include an increase in the axes is obtained with a slight flexion of the head.
anteroposterior diameter of the pharynx, an eleva- Despite this study brings some science to common
tion of the epiglottis, a small amount of cervical pediatric anesthesia lore, further studies on intuba-
spine motion (2 of flexion) and anterior movement tion conditions should be performed to determine
of the bony and cartilaginous landmarks of the how these anatomical findings can influence our
neck (26). These modifications can be regarded as clinical practice. The crucial clinical point remains
negligible compared with the variations produced larynx visualization, and not anatomical axis align-
by the extension of the head on the various studied ment.
axes. Moreover, one can estimate that the displace-
ment induced by the LMA is of same magnitude
in resting and flexion positions, with limited
1 Stone DJ, Gal TJ. Airway management. In: Miller RD, ed.
consequences on the calculation of the observed
Anesthesia, 4th edn. New York: Churchill Livingstone; 2000:
differences. 1414–1451.
Owing to the quality of MRI definition, axis 2 Benali H. Intubation. In: Voyer M, ed. Soins intensifs et réani-
determinations were easy to make in our patients, mation du nouveau-né. Paris: Masson, 2002: 348–352.
3 Cloup M, ed. Intubation trachéale. Pédiatrie d’urgence. Paris:
except for the determination of the Neck axis Flammarion, 1993: 784–790.
because it was difficult to precisely locate the sternal 4 Obladen M. Ventilation artificielle. In: Obladen M, ed. Soins
fork in the younger infants. The results concerning intensifs pour nouveau-nés. Paris: Springer, 1998: 97–144.
5 Tremblay P, Guay J, Lapierre G. Intubation endotrachéale et
neck flexion must therefore be cautiously inter- soins des voies aériennes. In: Beaufils F, ed. Urgences et soins
preted. intensifs pédiatriques. Paris: Doin, 1994: 3–20.
The lack of statistical signification of the age on 6 James LS. Emergencies in delivery room. In: Martin RJ, ed.
Neonatal-perinatal Medicine, 3rd edn, St. Louis: Mosby, 1983:
our results was certainly due to the relatively small
number of patients and to the small range of ages 7 Backofen JE, Rogers MC. Emergency management of the air-
with a preponderance of infants. A study stratified way. In: Rogers MC, ed. Textbook of Pediatric Intensive Care.
to age should be performed to better address this Baltimore: Williams and Wilkins, 1987: 57–79.
8 McCoy E. The McCoy laryngoscope in infants and children.
important issue. The quantification of visualization Can J Anaesth 2004; 51: 101–105.
conditions with angle measurements, as we 9 Riasi J. The difficult pediatric airway. In: Benumof JL, ed.
attempted to do, can be criticized. Particularly ana- Airway management, principles and practice. St. Louis: Mosby,
1996: 585–637.
tomical axis alignment and laryngeal visualization

 2008 The Authors

Journal compilation  2008 Blackwell Publishing Ltd, Pediatric Anesthesia, 18, 525–531
H E A D P O S T U R E A N D A IR W A Y I N I N F A N T S A N D C H I L D R E N 531

10 Westhorpe RN. The position of the larynx in children and its 21 MacIntyre PA, McLeod AD, Hurley R et al. Cervical spine
relationship to the ease of intubation. Anaesth Intensive Care movements during laryngoscopy. Comparison of the Macin-
1987; 15: 384–388. tosh and McCoy laryngoscope blades. Anaesthesia 1999; 54:
11 Lipsky CL. Critical care techniques of the neonates: endotra- 413–418.
cheal intubation. In: Spitzer AR, ed. Intensive Care of the Fetus 22 Hastings RH, Vigil AC, Hanna R et al. Cervical spine move-
and Neonate. St. Louis: Mosby, 1996: 1153–1157. ment during laryngoscopy with the Bullard, Macintosh, and
12 Benumof JL, ed. Conventional orotracheal and nasotracheal Miller laryngoscopes. Anesthesiology 1995; 82: 859–869.
intubation. Airway Management, Principles and Practice. St. 23 Wahlen BM, Gercek E. Three-dimensional cervical spine
Louis: Mosby, 1996: 261–276. movement during intubation using the Macintosh and Bullard
13 Arens R, McDonough JM, Corbin AM et al. Linear dimensions laryngoscopes, the bonfils fibrescope and the intubating
of the upper airway structure during development: assessment laryngeal mask airway. Eur J Anaesthesiol 2004; 21: 907–913.
by magnetic resonance imaging. Am J Respir Crit Care Med 24 Goudsouzian NG, Denman W, Cleveland R et al. Radiologic
2002; 165: 117–122. localization of the laryngeal mask airway in children. Anes-
14 Shorten GD, Armstrong DC, Roy WI et al. Assessment of the thesiology 1992; 77: 1085–1089.
effect of head and neck position on upper airway anatomy in 25 Horton WA, Fahy L, Charters P. Defining a standard
sedated paediatric patients using magnetic resonance imaging. intubating position using ‘‘angle finder’’. Br J Anaesth 1989; 62:
Paediatr Anaesth 1995; 5: 243–248. 6–12.
15 Adnet F, Borron SW, Dumas JL et al. Study of the ‘‘sniffing 26 Brimacombe J, ed. Anatomy. Laryngeal Mask Anesthesia, 2nd ed.
position’’ by magnetic resonance imaging. Anesthesiology 2001; Philadelphia: Saunders, 2005: 73–104.
94: 83–86. 27 Froelich MA. The ‘‘sniffing position’’ is not an anatomic myth.
16 Shorten GD, Opie NJ, Graziotti P et al. Assessment of upper Anesthesiology 2001; 95: 1044.
airway anatomy in awake, sedated and anaesthetised patients 28 Weinberg GL. Another view of the ‘‘sniffing position’’.
using magnetic resonance imaging. Anaesth Intensive Care 1994; Anesthesiology 2001; 95: 1044–1045.
22: 165–169. 29 Lee LW. You can smell the difference. Anesthesiology 2001; 95:
17 Hudgins PA, Siegel J, Jacobs I et al. The normal pediatric lar- 1045.
ynx on CT and MR. AJNR Am J Neuroradiol 1997; 18: 239–245. 30 Stasiuk RB. There’s more to intubation than laryngoscopy.
18 Graf W, de Waele C, Vidal PP. Functional anatomy of the Anesthesiology 2002; 97: 751, author reply 754.
head-neck movement system of quadrupedal and bipedal 31 Adnet F, Baillard C, Borron SW et al. Randomized study
mammals. J Anat 1995; 186 (Pt 1): 55–74. comparing the ‘‘sniffing position’’ with simple head extension
19 Ferrario VF, Sforza C, Poggio CE et al. A three-dimensional for laryngoscopic view in elective surgery patients. Anesthesi-
non-invasive study of head flexion and extension in young ology 2001; 95: 836–841.
non-patient subjects. J Oral Rehabil 1997; 24: 361–368.
20 Rudolph C, Schneider JP, Wallenborn J et al. Movement of the
upper cervical spine during laryngoscopy: a comparison of the
Accepted 1 February 2008
Bonfils intubation fibrescope and the Macintosh laryngoscope.
Anaesthesia 2005; 60: 668–672.

 2008 The Authors

Journal compilation  2008 Blackwell Publishing Ltd, Pediatric Anesthesia, 18, 525–531