Anaesth Intensive Care 2002; 30: 771-774

Appropriate Size of Laryngeal Mask Airway for Children

G. P. Y. LOKE*, S. M. TAN†, A. S. B. NG‡
Department of Paediatric Anaesthesia, KK Women’s and Children’s Hospital, Singapore

SUMMARY
The aim of this crossover study was to determine the optimal size of laryngeal mask airway in children weighing 10
to 20 kg. In each of 67 apnoeic anaesthetized children, the size 2 and size 2½ laryngeal mask airways were inserted
consecutively by a skilled user and the cuff inflated to 60 cmH2 O. Each LMA was assessed for the ease of insertion
(by the number of attempts), oropharyngeal leak pressure, anatomical position (assessed fibreoptically) and the
volume of air required to achieve intracuff pressure of 60 cmH2 O. During the measurement of oropharyngeal leak
pressure, the airway pressure was not allowed to exceed 30 cmH2 O. There was no failed attempt at insertion with any
size. The oropharyngeal leak pressure was significantly less for the size 2 LMA compared to the size 2½ LMA
(P<0.001). The oesophagus was visible on three occasions, all with the size 2 LMA. Gastric insufflation occurred in
three patients, all with the size 2 LMA. The incidence of low oropharyngeal leak pressure (<10 cmH2 O) was low
(9.0%) and all occurred with the size 2 LMA. The fibreoptic bronchoscope scores were not significantly different
between the two sizes of LMAs. The volume of air to achieve intracuff pressure of 60 cmH2 O was much lower than
the maximum recommended volume (5.1 ml for size 2 and 6.2 ml for size 2½). We conclude that the size 2½ LMA
provides a better fit than size 2 in children 10 to 20 kg.
Key Words: ANAESTHESIA: paediatric. EQUIPMENT: laryngeal mask airway

The laryngeal mask airway (LMA) is being used
increasingly in paediatric anaesthesia. It is a scaled-
down version of the adult form and no direct post-
mortem specimen work has been published. The
anatomy of the larynx of children is known to be
different from that of adults. A higher and more
anterior larynx with a relatively large floppy epiglottis
may make the proper placement of the LMA difficult.
The manufacturer (Intavent, Henley-on-Thames,
U.K.) currently recommends the size of the LMA for
children based on weight (size 2 for 10 to 20 kg and
size 2½ for 20 to 30 kg)1. This weight-based formula
had been evaluated by several previous studies in
adults which suggested that “up-sizing” of LMA and
a gender-based formula (size 3 to size 4 for females
and size 4 to 5 for males) provided better fit for the
* M.B., B.S. (Singapore), Medical Officer and Trainee in Anaesthesia,
Department of Paediatric Anaesthesia, KK Women’s and Children’s
Hospital, Singapore.
† M.B., B.S. (Singapore), M.Med. Anaesthesia (Singapore), Registrar in
Anaesthesia, Changi General Hospital, Singapore.
‡ M.B., B.S.(Singapore), M.Med. Anaesthesia (Singapore), Senior
Consultant in Anaesthesia, Department of Paediatric Anaesthesia, KK
Women’s and Children’s Hospital, Singapore.
Address for reprints: Dr G. P. Y. Loke, Department of Paediatric
Anaesthesia, KK Women’s and Children’s Hospital, 100 Bukit Timah Road,
Singapore 229899.
Accepted for publication on July 28, 2002.
Anaesthesia and Intensive Care, Vol. 30, No. 6, December 2002
patients2-4. It is common practice to inflate the LMA
with the maximum recommended volumes of air1. It is
now suggested that a cuff pressure of 60 cmH2O is
more appropriate5.
In this study, we have compared the size 2 and the
size 2½ LMA for children weighing 10 to 20 kg in
terms of ease of insertion, oropharyngeal leak
pressure, anatomical position (assessed fibre-
optically) and the volume of air required to achieve
intracuff pressure of 60 cmH2O.
METHODS
After obtaining approval from the local Ethics
Committee, 67 ASA 1/2 children between 10 and
20 kg scheduled for elective surgery were recruited
into the study. Patients were excluded if they had res-
piratory tract pathology, were at risk of aspiration or
otherwise considered unsuitable for laryngeal mask
use. Their ages, heights, weights and body mass
indices were recorded.
Anaesthesia was induced either intravenously with
thiopentone 5 mg/kg or inhalationally with sevo-
flurane 8% and nitrous oxide 50%. Intravenous
fentanyl 1 µg/kg was then given to induce apnoea.
Adequate depth of anaesthesia was established with
bag and mask using sevoflurane and nitrous oxide.
The laryngeal mask was inserted when the end-tidal
sevoflurane concentration reached 5%.
772 G. P. Y. LOKE, S. M. TAN, A. S. B. NG

A size 2 laryngeal mask, as recommended by the RESULTS

manufacturers for this weight child1, was inserted The demographic data of the patients are
first. The laryngeal mask was inserted by an ex- presented in Table 3. The number of attempts,
perienced anaesthetist using the standard Brain tech- oropharyngeal leak pressures, fibreoptic scores and
nique6. Successful placement was judged by chest wall volume of air to achieve an intracuff pressure of
excursion and capnography. The cuff of the laryngeal 60 cmH2O are presented in Table 4. These para-
mask was then inflated with air until the intracuff meters are presented for patients 10 to 15 kg in
pressure reached 60 cmH2O5. The oropharyngeal Table 5 and for patients 15.1 to 20 kg in Table 6.
leak pressure (the airway pressure at which the There was no failed attempt at insertion with any
manometer dial reached stability with a fresh gas flow size. The oropharyngeal leak pressure was signifi-
of 3 l/min and the expiratory value of the circle system cantly less for the size 2 LMA compared to the size
closed) was next measured. It was classified as 21¼2 LMA (P<0.001). This was true for patients
high (≥20 cmH2O), medium (11-19 cmH2O) or low
( 10 cmH2O). The airway pressure was not allowed to TABLE 3
exceed 30 cmH2O. The mouth of the patient was then Demographic data. Mean (range)
opened to check if the cuff was visible. The position Male/female; n 57/10
of the laryngeal mask airway was assessed using a Age (y) 4 (1.5-9)
Height (m) 0.97 (0.76-1.22)
fibreoptic scope which was passed through the laryn- Weight (kg) 14.8 (10.0-20.0)
geal mask airway to a position just proximal to the BMI (kg.m–2) 15.5 (12.4-19.3)
mask aperture bars and the view was scored accord-
ing to criteria listed in Table 1. Any visibility of the
TABLE 4
oesophagus was also noted. Comparison of insertion attempts, oropharyngeal leak pressure
(OLP), cuff volume, fibreoptic bronchoscope scores (FOB) for size
TABLE 1 2 and 2½ LMA
Fibreoptic bronchoscope score7
Description Size 2 Size 2½
4 Only vocal cords seen
Number of attempts
3 Vocal cords and posterior part of epiglottis seen
1/2/F 67/0/0 66/1/0
2 Vocal cords and anterior part of epiglottis seen
OLP; cmH2O
1 Vocal cords not seen
High ≥20 cm H2O 23 (34.3%) 35 (52.2%)
Medium 11-19 cmH2O 38 (56.7%) 32 (47.8%)
Low 10 cmH2O 6 (9%) 0 (0%)
The first laryngeal mask airway was removed and a FOB scores; n (%)
4 VC only 12 (17.9%) 15 (14.9%)
second laryngeal mask airway (size 2½) was inserted. 3 VC+PE 5 (7.5%) 2 (3%)
The patient was ventilated with a facemask between 2 VC+AE 41 (61.2%) 40 (59.7%)
insertions. The fit and position of the second laryn- 1 No VC 9 (13.4%) 10 (14.9%)
Cuff volume mean
geal mask airway were assessed in a similar manner to (range); ml 5.1 (3.7-6.5) 6.2 (4.2-8.2)
the first. SaO2 (%) 94-100 95-100
The following parameters were recorded: the ease VC=vocal cord; PE=posterior epiglottis; AE=anterior epiglottis.
of insertion (Table 2; a failed attempt is when the
LMA had to be removed from the mouth), the TABLE 5
oropharyngeal leak pressure, the presence of cuff in Comparison of insertion attempts, oropharyngeal leak pressure
(OLP), cuff volume, fibreoptic bronchoscope scores (FOB) for size
the mouth, the fibreoptic bronchoscope scores, the 2 and 2½ LMA in patients weighing 10 to 15 kg
lowest SaO2 during the insertion of each laryngeal
Description Size 2 Size 2½
mask airway and any other adverse events (e.g. gastric
Number of attempts
insufflation as indicated by epigastric distension, 1/2/F 35/0/0 34/1/0
dislodgement, etc) that may have occurred. OLP; cmH2O
Statistical analysis was with paired t-test, Wilcoxon High ≥20 cm H2O 9 (25.7%) 19 (54.3%)
Medium 11-19 cmH2O 21 (60.0%) 16 (45.7%)
signed rank test and McNemar’s test. Significance Low 10 cmH2O 5 (14.3%) 0 (0%)
was taken as P<0.05. FOB scores; n (%)
4 VC only 5 (14.3%) 6 (17.1%)
3 VC+PE 4 (11.4%) 2 (5.7%)
TABLE 2
2 VC+AE 23 (65.7%) 21 (60.0%)
Ease of insertion of laryngeal mask airway 1 No VC 3 (8.6%) 6 (17.1%)
1 LMA inserted with 1 attempt Cuff volume mean
2 LMA inserted with 2 attempts (range); ml 5.3 (3.8-6.9) 6.2 (4.1-8.3)
F Unable to insert LMA with 2 attempts VC=vocal cords; PE=posterior epiglottis; AE=anterior epiglottis.

Anaesthesia and Intensive Care, Vol. 30, No. 6, December 2002

 PAEDIATRIC LMA SIZING
TABLE 6
Comparison of insertion attempts, oropharyngeal leak pressure
(OLP), cuff volume, fibreoptic bronchoscope scores (FOB) for size
2 and 2½ LMA in patients weighing 15.1-20 kg
Description Size 2
Number of attempts
1/2/F 32/0/0
OLP; cmH2O
High ≥20 cm H2O 14 (43.8%)
Medium 11-19 cmH2O 17 (53.1%)
Low 10 cmH2O 1 (3.1%)
FOB scores; n (%)
4 VC only 7 (21.9%)
3 VC+PE 1 (3.1%)
2 VC+AE 18 (56.3%)
1 No VC 6 (18.8%)
Cuff volume mean
(range); ml 4.8 (3.7-6.0) 6.1 (4.2-8.0)
VC=vocal cords; PE=posterior epiglottis; AE=anterior epiglottis.
weighing 10 to 15 kg (P=0.001) as well as for patients
weighing 15.1 to 20 kg (P=0.039).
The fibreoptic bronchoscope scores were not sig-
nificantly different between the two sizes of LMA.
The oesophagus was visible via the bronchoscope on
three occasions, all with the size 2 LMA. The cuff was
visible in the mouth on three occasions and all
occurred with the size 2½ LMA. There were two inci-
dences of LMA dislodgement, one with the size 2
LMA and one with the size 2½ LMA. The lowest
SaO2 was not different between the two sizes of
LMAs. Gastric insufflation occurred in three patients
with the size 2 LMA (oropharyngeal leak pressure
was less than 20 cmH2O in all three cases).
DISCUSSION
In this study we have demonstrated that the size 2½
LMA provides a better seal in paediatric patients
weighing 10 to 20 kg compared to the recommended
size 2 LMA. Several studies had shown that upsizing
of the LMA in adults (from size 3 to 4 in females and
from size 4 to 5 in males) provided a better seal as
reflected by higher oropharyngeal leak pressures2-4.
However, to date there are no similar data available
for paediatric patients. Although some anaesthetists
may be concerned about the possibility of adverse
effects due to inserting a larger LMA in children,
especially those weighing less than 15 kg, our data
showed no difference in the number of insertion
attempts or the incidence of dislodgement in these
children. Instead, a higher oropharyngeal leak
pressure was achieved.
Many anaesthetists manually ventilate patients via
the LMA at induction to check airway patency or to
compensate for periods of apnoea. Our data showed
that there is a higher proportion of patients with
Anaesthesia and Intensive Care, Vol. 30, No. 6, December 2002
773
an oropharyngeal leak pressure of more than
20 cmH2O with the size 2½ LMA compared to the
size 2 (52.2% vs 34.3%, P=0.055). Although this does
not reach statistical significance, the ability to
Size 2½
improve the seal in 18% of patients is probably
32/0/0
clinically significant, as a better seal should reduce
the risk of gastric insufflation (and associated prob-
16 (50.0%) lems such as diaphragmatic splinting and regurgita-
16 (50.0%)
0 (0%)
tion of gastric contents). The observation of gastric
insufflation in our patients is probably an underesti-
9 (28.1%) mate of the real incidence since epigastric distension
0 (0%)
19 (59.4%) is a rather insensitive indicator. It has been demon-
4 (12.5%) strated in adults that the incidence of gastric insuffla-
tion increases significantly when the peak airway
pressure is more than 20 cmH2O during positive pres-
sure ventilation via the LMA8-10.
A higher oropharyngeal leak pressure implies a
better seal around the glottis not only to prevent
leakage of gases into the stomach but also to prevent
contamination of the larynx with oropharyngeal
secretions. It has been suggested that in patients
undergoing spontaneous ventilation, oropharyngeal
leak pressure should be more than 10 cmH2O11 (the
approximate pressure of fluid at the posterior
pharyngeal wall if the oral cavity is flooded). The inci-
dence of oropharyngeal leak pressure of 10 cmH2O
or less (with intracuff pressure of 60 cmH2O) is low
(9.0%) and all occurred with the size 2 LMA.
Although it has been suggested that the intracuff
pressure may be reduced to 22 mmHg (28.6 cmH2O)
without significantly affecting tidal ventilation in
spontaneously breathing anaesthetized patients12, we
are reluctant to reduce the intracuff pressure in our
patients further (especially with the size 2 LMA)
as this may lead to increased risk of laryngeal
contamination.
It is interesting to note that a small volume of air
is sufficient to achieve an intracuff pressure of
60 cmH2O. This low intracuff pressure is currently
recommended by the manufacturer with the aim of
reducing pharyngeal morbidity5,6. This is logical since
the LMA is soft at low cuff volume/low cuff pressure
and therefore sufficiently compliant to adapt to the
different pharyngeal shapes. Thus, the incidence of
pharyngeal morbidity will be lower although this has
been challenged13. At high volumes, the LMA is
poorly compliant. This may lead to local high pres-
sure points resulting in tissue or nerve damage14,15
without improving the seal16. A survey among our
nurses revealed that 80% of them would inflate
the LMA with the maximum recommended volume
(10 ml for size 2 and 14 ml for size 2½)1. We should
review the volume of air used to inflate the LMA. In
774 G. P. Y. LOKE, S. M. TAN, A. S. B. NG
fact, it is recommended that the cuff should be
periodically deflated during anaesthesia to prevent
excessive increase in cuff volume/pressure due to
nitrous oxide diffusion17,18.
Fibreoptic assessment of LMA position showed
a high incidence of the epiglottis impinging on the
grille of the LMA (60.3% for size 2 and 58.8% for size
2½). This is consistent with previous studies which
found a high incidence of epiglottic downfolding in
children compared to adults19,20. This may be due to
the relatively larger and more floppy epiglottis in
children21. Although this constitutes endoscopic
evidence of airway obstruction, there was no obstruc-
tion to airflow in any of these cases. Presumably, the
area between the downfolded epiglottis and the
laryngeal aperture is large enough to allow un-
obstructed ventilation. The LMA may sometimes
enclose the epiglottis even if it is correctly posi-
tioned22. The results of the fibreoptic assessment
should be interpreted with caution as the fibre-
optic bronchoscope offers only a two-dimensional
assessment of the LMA position.
In conclusion, the use of a size 2½ LMA in children
weighing 10 to 20 kg provides a better fit compared to
size 2 LMA. We recommend that the volume of air
used to inflate the LMA be reduced from that
currently recommended.
REFERENCES
1. The laryngeal mask. Int Anesthesiol Clin 1998; 36:48-49.
2. Asai T, Murao K, Yukawa H, Shingu K. Re-evaluation of
appropriate size of the laryngeal mask airway. Br J Anaesth
1999; 83:478-479.
3. Asai T, Howell TK, Koga K, Morris S. Appropriate size and
inflation of the laryngeal mask airway. Br J Anaesth 1998;
80:470-474.
4. Berry A, Brimacombe JR, McManus KF, Goldblatt M. An
evaluation of the factors influencing selection of the optimal
size of laryngeal mask airway in normal adults. Anaesthesia
1998; 53:565-570.
5. Brain AIJ. Pressure in laryngeal mask airway cuffs. Anaesthesia
1996; 51:603.
6. Brimacombe J, Brain AIJ, Berry A. The laryngeal mask airway
instruction manual. Henley-on-Thames: Intavent Research
Limited, 1996.
7. Brimacombe J, Berry A. A proposed fibreoptic scoring system
to standardize the assessment of laryngeal airway mask posi-
tion. Anesth Analg 1993; 76:457.
8. Weiler N, Lature F, Eberle B, Goedecke R, Heinrichs W.
Respiratory mechanics, gastric insufflation pressure and air
leakage of the laryngeal mask airway. Anesth Analg 1997;
84:1025-1028.
9. Brimacombe JR. Positive-pressure ventilation with the size 5
laryngeal mask airway. J Clin Anesth 1997; 9:113-117.
10. Devitt JH, Wenstone R, Noel AG, O’Donnell MP. The
laryngeal mask airway and positive pressure ventilation.
Anesthesiology 1994; 80:550-555.
11. Brimacombe J, Berry A, Brain AIJ. Optimal intracuff pressures
with the laryngeal mask airway. Br J Anaesth 1996; 77:295-296.
12. Morris GN, Marjot R. laryngeal mask airway performance:
effect of cuff deflation during anaesthesia. Br J Anaesth 1996;
76:456-458.
13. Rieger A, Brunne B, Striebel HW. Intracuff pressure do not
predict laryngopharyngeal discomfort after use of the laryngeal
mask airway. Anesthesiology 1997; 87:63-67.
14. Nagai K, Sakuramoto C, Goto F. Unilateral hypoglossal nerve
paralysis following use of the laryngeal mask airway.
Anaesthesia 1994; 49:603-604.
15. King C, Street MK. Twelfth cranial nerve paralysis following
use of a laryngeal mask airway. Anaesthesia 1994; 49:786-787.
16. Keller C, Puhringer F, Brimacombe JR. Influence of cuff
volume on oropharyngeal leak pressure and fibreoptic position
with the laryngeal mask airway. Br J Anaesth 1998; 81:186-187.
17. Slater P, Lavies NG. Optimal laryngeal mask airway cuff
pressure. Anaesthesia 1996; 51:1187.
18. Lumb AB, Wrigley MW. The effect of nitrous oxide on laryn-
geal mask airway cuff pressure. Anaesthesia 1992; 47:320-323.
19. Rowbottom SJ, Simpson DL, Grubb D. The laryngeal mask
airway in children. A fibreoptic assessment of positioning.
Anaesthesia 1991; 46:489-491.
20. Elwood T, Cox RG. Laryngeal mask airway insertion with
a laryngoscope in paediatric patients. Can J Anaesth 1996;
43:435-437.
21. Chow BF, Lewis M, Jones SE. Laryngeal mask airway in
children: insertion technique (letter). Anaesthesia 1991;
46: 590-591.
22. Asai T. Difficulty in assessing the correct position of the
laryngeal mask airway. Br J Anaesth 1994; 72:366.
Anaesthesia and Intensive Care, Vol. 30, No. 6, December 2002