See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/51729042

Factors associated with difficult intubation in prehospital emergency

medicine

Article  in  European Journal of Emergency Medicine · October 2011

DOI: 10.1097/MEJ.0b013e32834d3e4f · Source: PubMed

CITATIONS READS

23 99

6 authors, including:

Yonathan Freund Francois-Xavier Duchateau

Assistance Publique – Hôpitaux de Paris Allianz Partners
144 PUBLICATIONS   1,875 CITATIONS    112 PUBLICATIONS   908 CITATIONS   

SEE PROFILE SEE PROFILE

Philippe Juvin
Assistance Publique – Hôpitaux de Paris
70 PUBLICATIONS   1,206 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

disaster View project

repatriation View project

All content following this page was uploaded by Francois-Xavier Duchateau on 13 August 2020.

The user has requested enhancement of the downloaded file.

 Original article 1

Factors associated with difficult intubation in prehospital

emergency medicine
Yonathan Freund, Francois-Xavier Duchateau, Marie-Laure Devaud,
Agnès Ricard-Hibon, Philippe Juvin and Jean Mantz

Objectives When managing airways in a prehospital in the standard intubation group and 23 (31%) in the
setting, emergency physicians have to deal with DI group (P < 0.01).
difficult intubation (DI), which increases morbidity and
Conclusion For prehospital orotracheal intubation,
mortality. The primary goal of this study was to
independent risk factors of DI are a mental-thyroid
determine predictors of DI in the out-of-hospital field
distance less than three fingers, a patient on the floor, and
faced by the French physician-staffed Emergency
a superior airways obstruction. Anticipation of DI could
Medical Service.
result in fewer attempts, and fewer complications, as the
Methods The study was a prospective, observational rate of complication increases with the difficulty of
study, including all consecutive patients intubated during intubation. European Journal of Emergency Medicine
a 30-month period. Patients having experienced standard 00:000–000 c 2011 Wolters Kluwer Health | Lippincott
intubation (two attempts or less) or DI (more than two Williams & Wilkins.
attempts) were compared. European Journal of Emergency Medicine 2011, 00:000–000
Results Six hundred and ninety-four patients were Keywords: airways, difficult intubation, emergency, prehospital
included: 70 (11%) were classified as DI and 583 as
Department of Anaesthetics and Intensive Care, Beaujon University Hospital,
standard intubations. Logistic regression showed Clichy, France
that airways obstruction [odds ratio (OR), 4.1; 95%
Correspondence to Yonathan Freund, MD, Emergency Department,
confidence interval (CI), 1.71–14.4], intubation on Pitié-Salpêtrière Hospital, 47-83 Boulevard de l’hôpital, 75013 Paris, France
the floor (OR, 2.6; 95% CI, 1.04–6.6), and a hyoid-mental Tel: + 33 1 42177912; fax: + 33 1 42177242;
e-mail: yonathanfreund@gmail.com
distance less than three fingers (OR, 2.3; 95% CI,
1.2–4.7) were independent predictors of DI. Immediate Received 22 July 2011 Accepted 16 September 2011
complications occurred in 89 patients (16%): 66 (11%)

Introduction Methods
When managing airways in a prehospital setting, emer- Study design
gency physicians have to deal with difficult intubation The study was a prospective observational study, which
(DI), which increases morbidity and mortality [1–3]. was conducted in the EMS unit of a teaching hospital,
Substantial progress in research has been made over the covering an area of 290 172 inhabitants during a 30-month
years in this field. A safe and effective rapid sequence period. The study has been approved by the Institutional
intubation (using etomidate and succinylcholine) has Review Board of Paris – North Hospitals, Paris 7
been generalized, and algorithms for the management of University, Assistance-Publique – Hôpitaux de Paris.
DI are widespread [4–6]. However, little is known about
the factors associated with prehospital DI. Most previous As a part of the French EMS system, physician-staffed
studies on prehospital intubation were carried out before ambulances are deployed if a particularly severe or life-
the use of rapid sequence induction was generalized. In threatening condition is suspected. The teams include a
a prehospital setting, intubation is often associated physician, a nurse, and a paramedic. When intubation is
with unexpected difficulties such as difficult environ- indicated, patients are intubated by an emergency
ments, uncomfortable positions, or lack of experience of physician, or by an anesthestic nurse with substantial
emergency physicians in charge of the patients. For these experience in both the EMS and operating theaters.
reasons, early identification of patients at risk of DI is Physicians are specialized in emergency medicine or
of major interest because alternative techniques and/or anesthetics and have at least 2 years experience in
airway rescue tools can be anticipated [7]. Therefore, the prehospital EMS care. From April 2008 to October 2010,
primary goal of the present study was to determine 10 225 such ambulances were deployed, and 694 patients
predictors of DI in the out-of-hospital field by French were intubated. Metal, single-use or reusable, laryngo-
Emergency Medical Services (EMS). scope blades with similar first-pass success were used [8].
0969-9546
c 2011 Wolters Kluwer Health | Lippincott Williams & Wilkins DOI: 10.1097/MEJ.0b013e32834d3e4f

Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
 2 European Journal of Emergency Medicine 2011, Vol 00 No 00

All consecutive intubated patients were included, and the Results

data were collected under the supervision of a senior Six hundred and ninety-four patients were intubated
attending. All forms were completed by the physician during the study period. Of these patients, 41 were
who performed intubation. excluded because an alternative technique was preven-
tively used at the physician’s discretion during the first or
the second attempt and the patients were successfully
Definition of difficult intubation
intubated before the third attempt (Fig. 1).
According to the French Society of Anaesthetics and
Intensive Care (Société Française d’Anesthésie Réanima- Demographic data
tion), a DI is considered present after two failed The mean age was 60.5 years, and the sex ratio was 1.6
laryngoscopic attempts for tube placement, or when (62% men). Three hundred and sixty-four (56%) patients
alternative measures are needed to ensure successful presented with cardiac arrest at the time of the
intubation. On the basis of this definition, patients were intubation, and were intubated without sedation. Among
recorded as follows: the remaining 330 patients, 289 patients underwent rapid
sequence intubation. Seventy patients (11%) were
(1) If more than two attempts were needed to obtain a classified as DI and 583 as StI. Median number of
proper endotracheal tube placement, the intubation attempts was one (Fig. 2). In the DI group, median
was recorded as difficult. number of laryngoscopic attempts was three (3–3.75).
(2) If an alternative technique was preventively and
successfully used (such as a gum elastic bougie)
before the third attempt, then the patient was Fig. 1
excluded, for we could not assess the difficulty of
the process. 694 Patients
(3) In other cases, when the procedure was successful in consecutively
less than three attempts without using alternative intubated
techniques, the intubation was recorded as a standard
intubation (StI).

Patients were divided into two groups: StI (two attempts

41 Excluded
or less) or DI. Variables recorded included age, sex, (alternative technique
estimated weight and height, Cormack classification, before the third attempt)
position of the patient during the procedure (on a bed,
on a stretcher, on the floor), medical condition of the
patient (cardiac arrest or other cause), drugs used for
induction, hyoid-mental distance, the presence of an
583 Standard
airways obstruction [or ear–nose–throat (ENT) pathol- 70 Difficult
intubations
ogy], number of attempts at intubation, possible alter- intubations (DI)
(Sti)
native techniques, and immediate complications. These
were compared between the two groups. Flow chart.

Statistical analysis
Expecting a DI rate of about 10%, and requiring a
minimum of 50 patients for the multivariate analysis, we Fig. 2
estimated that the sample size should exceed 500
Attempts
patients. All collected data were included in univariate 500
analysis, and factors statistically associated with a DI were 450
then pooled in the multivariate analysis. Results are 400
reported as mean values ± SD for continuous variables 350
300
and medians with interquartile ranges for discontinuous
250
data. Qualitative data are expressed as the percentage of 200
patients. Statistical analysis was performed using a 150
one-way ANOVA for quantitative data and a w2 test for 100
qualitative data. A multivariate analysis was also per- 50
0
formed, with all variables significantly associated with DI. 1 2 3 4 5 6 7
Statistical significance was defined as P < 0.05. Stat-View
5 (Abacus Concept, Berkeley, California, USA) was used Number of attempts.
for analysis.

Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
 Factors associated with difficult intubation Freund et al. 3

Cormack classification Table 3 Logistic regression

One hundred and eighty-six (28%) of the intubated Variable OR 95% Confidence interval P
patients were classified as Cormack 1, 389 (60%) as
Patient on the floor 2.60 1.04–6.6 0.04
Cormack 2, 55 (8%) as Cormack 3, and 22 (3%) as Cardiac arrest 1.60 0.796–3.195 0.19
Cormack 4. The Cormack classifications were signifi- Hyoid-mental < 3 fingers 2.34 1.157–4.745 0.02
cantly different depending on whether the intubation was Airways obstruction 4.09 1.171–14.278 0.03

considered to be difficult or not as shown in Table 1. OR, odds ratio.

The Cormack score was not analyzed as a predictive

factor of DI, because this criteria was unknown before
the patient is intubated. airways lesions (three in the DI and three in the StI
groups, respectively), oxygen desaturation (six in the DI
Factors associated with DI and 19 in the StI groups, respectively), and cardiac arrest
It was found that a mental-hyoid distance of less than (six in the DI and thee in the StI groups, respectively).
three fingers was more likely to be present in patients
undergoing a DI (16 vs. 9%, P = 0.03). Furthermore, Discussion
patients handled on the floor (89 vs. 68%, P < 0.001), The definition of DI used in this study is a pragmatic
patients with cardiac arrest (74 vs. 54%, P = 0.001), and comment on the difficulty of the procedure, rather than a
patients with obstructive airways conditions (6 vs. 1%, characteristic of the patient being intubated. In the field
P = 0.02) were more often present in the DI group. of prehospital medicine, this ‘real-life’ evaluation of the
Obese patients (defined as those with a BMI > 30) were difficulty is of much more interest than what difficulty
not found significantly more likely to present a DI. the intubation could pose under optimal conditions.
Univariate comparison is shown in Table 2.
The findings of this study can be summarized as follows:
Logistic regression showed that an airways obstruction airway obstruction, a hyoid-mental distance of less
[odds ratio (OR), 4.1; 95% confidence interval (CI), than three fingers, and a patient intubated on the
1.71–14.4], a patient intubated on the floor (OR, 2.6; 95% floor, were all independent predictors of DI in the
CI, 1.04–6.6), and a hyoid-mental distance less than three prehospital setting.
fingers (OR, 2.3; 95% CI, 1.2–4.7) were independent
The rate of DI observed (11%) is consistent with pre-
predictors of DI, as shown in Table 3.
existing literature [9–11] in which the same EMS system
Immediate complications occurred in 89 patients (16%): experiences rates of 2 [10] to 20% [12]. In a recent large
66 (11%) in the StI group and 23 (31%) in the DI group monocenter study, including 1442 consecutive out-
(P < 0.01). In the two groups, complications were: of-hospital intubations, the rate of DI was of 7.4% [9].
esophageal intubation (12 in the DI group and 21 in For paramedic-staffed EMS systems, the incidence has
the StI group), aspiration of gastric content (five in the been reported as up to 36% [13–15]. These variations
DI and six in the StI groups, respectively), dental or may result from the absence of a strict definition of DI,
and from major differences in staff.
Table 1 Cormack classification in the two groups Many factors have been associated with a DI based on
Difficult intubation Standard intubation the anaethetics and emergency medicine literature, for
Cormack classification N = 70 (%) N = 583 (%) P example, hyoid-mental distance of less than three fingers,
1 16 (23) 170 (29) < 0.001 thyroid-hyoid distance of less than two fingers, obesity,
2 22 (31) 367 (63) — poor neck mobility, obstructed airways, large tongue, and
3 18 (25) 37 (6) —
4 14 (20) 8 (1) —
large incisors. Most of these factors remain controversial,
especially obesity [11,16–18]. A recent study conducted
in an emergency department [19], showed that thyroid-
hyoid distance of less than two fingers was the sole
Table 2 Univariate analysis for determining difficult intubation
independent variable predictive of DI. The presence of a
StI DI P short neck (mental-hyoid distance less than three fingers,
N (%) 553 (89%) 70 (11%) or hyoid-thyroid distance less than two fingers), has been
Age 60.2 ± 19.3 63.2 ± 16.3 0.16 repeatedly proven to be an independent predictor of
Sex (Male) 224 (62%) 45 (64%) 0.7
BMI > 30 69 (12%) 13 (19%) 0.11 DI [11,19,20], as has the presence of an airways
Patient on the floor 397 (68%) 62 (89%) < 0.01 obstruction (for example facial trauma, or ENT neopla-
Cardiac arrest 312 (54%) 52 (74%) 0.001
Hyoid-mental < 3 fingers 52 (9%) 11 (16%) 0.03
sia). The impossibility for the operator to stand up has
Airways obstruction 9 (1%) 4 (6%) 0.02 also been described as an independent predictor of
Cervical immobilization 47 (8%) 5 (7%) 0.79 DI [9]. We found a strong association between intubation
Mean age ± standard deviation. on the floor and DI. This new finding is of importance,
DI, difficult intubation; StI, standard intubation. because of its uniquness to the prehospital setting.

Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
 4 European Journal of Emergency Medicine 2011, Vol 00 No 00

Indeed, it could not have been observed in the operating should be added to the ‘DI’ situations [20]. We chose to
room or the emergency department for obvious reasons, exclude patients who underwent a successful intubation
so pre-existing literature is not available for comparison. with the use of an alternative (such as gum elastic
bougie) before the third attempt, for we could not state
Orliaguet et al. [10] performed a prospective study in the whether the intubation would have been difficult or not.
field of the French physician-staffed EMS system, which The greater specificity gained from this decision could
exhibits some similarities with the present one, although have been at the expense of the exclusion of some DI
DI was defined differently. Similarly to our study, a patients from the analysis. Also, although data were
history of ENT and maxillofacial trauma were factors collected from a standardized form, the process was not
independently associated with prehospital DI, whereas blinded, creating a possible selection bias for the two
patients in cardiac arrest were no more likely to present a groups. Finally, as a monocentric study, our results may
DI. Operator’s position during intubation (standing, not apply to all prehospital units, nor even other EMS
kneeling, decubitus) was also identified as an indepen- systems, but independent risk factors of DI may apply
dent factor, which could be correlated with the position of regardless of the grade or specialty of the operator.
the patient (on the floor, on a stretcher, on a bed) in this
study. However, Combes et al. [9] also found that operator Conclusion
status (resident) was associated with a higher DI rate. The identification of risk factors for DI in the prehospital
Status of the operator was recorded but not analyzed in emergency context is of major concern, as DI is associated
the present study because governing policy recommends with significant morbidity. Anticipation of DI could result
switching the operator to a senior physician or a in fewer attempts to succeed, and fewer complications, as
anesthetics nurse in case of failure at the first attempt. the rate of complications increases with the difficulty of
Finally, although higher BMI was identified as a risk factor intubation. The presence of one or more of the predictors
for DI in the Combes study (OR, 1.0; 95% CI, 1.0–1.1), of DI (mental-thyroid distance less than three fingers, a
and elsewhere in the literature [17,21], we could not find patient on the floor, and a superior airways obstruction)
any significant association between a BMI greater than 30 should lead to the anticipation of the need to use
and DI. As emphasized by Brodsky et al. [17], there are alternative techniques for out-of-hospital airway control.
many reasons for this discrepancy. First, the standard
sniffing position for tracheal intubation differs between
obese and nonobese patients. Second, the different Acknowledgements
studies do not endorse the same definition of DI; as The authors would like to thank Dr E.C. Baker (King’s
difficult laryngoscopy is not synonymous with DI, using college hospital, London, UK) for having reviewed and
the Cormack grade for the definition of DI could lead to improved our work.
judgement bias. Again, obesity remains a controversial
predictor of DI. Conflicts of interest
There are no conflicts of interest.
Another original finding was the association between DI
and immediate complications, suggesting that DI con-
tributes to significantly higher morbidity and mortality. References
1 Bellhouse CP, Dore C. Criteria for estimating likelihood of difficulty of
This result strengthens the view that DI is a major endotracheal intubation with the Macintosh laryngoscope. Anaesth Intensive
concern in the practice of the EMS, and that identifying Care 1988; 16:329–337.
2 Benumof JL, Scheller MS. The importance of transtracheal jet ventilation
risk factors for DI could contribute appreciably to in the management of the difficult airway. Anesthesiology 1989; 71:
ameliorating morbidity and mortality in this setting. An 769–778.
early identification of predictors of DI could lead to 3 Samsoon GL, Young JR. Difficult tracheal intubation: a retrospective study.
Anaesthesia 1987; 42:487–490.
preventive use of alternative techniques to avoid 4 Benomof JC. Algorithm for difficult intubation. Acta Anaesthesiol Scand
complications. The operator should expect a higher risk 2001; 45:1186.
of complications when facing a patient with airway 5 Mercer SJ, Guha A. Testing adherence to the DAS algorithm for difficult
tracheal intubation during rapid sequence induction of anaesthesia.
obstruction (such as ENT neoplasia for example), a Anaesthesia 2010 [Epub ahead of print].
hyoid-mental distance of less than 3 cm, or a patient on 6 Saxena S. The ASA difficult airway algorithm: is it time to include video
the floor, which are the predictors of DI identified here. laryngoscopy and discourage blind and multiple intubation attempts in the
nonemergency pathway? Anesth Analg 2009; 108:1052.
Rapid and reliable identification of these factors is 7 Harris T, Ellis DY, Foster L, Lockey D. Cricoid pressure and laryngeal
essential in the EMS field, where the rate of DI is much manipulation in 402 pre-hospital emergency anaesthetics: essential safety
higher than in the operating room. measure or a hindrance to rapid safe intubation? Resuscitation 2010;
81:810–816.
8 Jabre P, Galinski M, Ricard-Hibon A, Devaud ML, Ruscev M, Kulstad E, et al.
There are some limitations in this study. A strict Out-of-hospital tracheal intubation with single-use versus reusable metal
definition of DI was used, that is, the absence of a laryngoscope blades: a multicenter randomized controlled trial. Ann Emerg
successful intubation after two attempts. The French Med 2011; 57:225–231.
9 Combes X, Jabre P, Jbeili C, Leroux B, Bastuji-Garin S, Margenet A, et al.
Society of Anesthesia and Intensive Care suggests that Prehospital standardization of medical airway management: incidence and
instances requiring the use of an alternative technique risk factors of difficult airway. Acad Emerg Med 2006; 13:828–834.

Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
 Factors associated with difficult intubation Freund et al. 5

10 Orliaguet G, Tartière S, Lejay M, Carli P. Prospective in-field evaluation 15 Stewart RD, Paris PM, Winter PM, Pelton GH, Cannon GM. Field
of orotracheal intubation by emergency medical service physicians. endotracheal intubation by paramedical personnel. Success rates and
JEUR 1997; 1997:27–32. complications. Chest 1984; 85:341–345.
11 Adnet F, Borron SW, Racine SX, Clemessy JL, Fournier JL, Plaisance P, et al. 16 Adams JP, Murphy PG. Obesity in anaesthesia and intensive care.
The intubation difficulty scale (IDS): proposal and evaluation of a new score Br J Anaesth 2000; 85:91–108.
characterizing the complexity of endotracheal intubation. Anesthesiology 17 Brodsky JB, Lemmens HJ, Brock-Utne JG, Vierra M, Saidman LJ. Morbid
1997; 87:1290–1297. obesity and tracheal intubation. Anesth Analg 2002; 94:732–736 table of
12 Adnet F, Jouriles NJ, Le Toumelin P, Hennequin B, Taillandier C, Rayeh F, contents.
et al. Survey of out-of-hospital emergency intubations in the French 18 Juvin P, Lavaut E, Dupont H, Lefevre P, Demetriou M, Dumoulin JL,
prehospital medical system: a multicenter study. Ann Emerg Med 1998; Desmonts JM. Difficult tracheal intubation is more common in obese than in
32:454–460. lean patients. Anesth Analg 2003; 97:595–600 table of contents.
13 Krisanda TJ, Eitel DR, Hess D, Ormanoski R, Bernini R, Sabulsky N. An 19 Soyuncu S, Eken C, Cete Y, Bektas F, Akcimen M. Determination of difficult
analysis of invasive airway management in a suburban emergency medical intubation in the ED. Am J Emerg Med 2009; 27:905–910.
services system. Prehosp Disaster Med 1992; 7:121–126. 20 Intubation difficile. Conférence d’experts, SFAR, Texte court, 2006.
14 Pointer JE. Clinical characteristics of paramedics’ performance of 21 Karkouti K, Rose DK, Wigglesworth D, Cohen MM. Predicting difficult
endotracheal intubation. J Emerg Med 1988; 6:505–509. intubation: a multivariable analysis. Can J Anaesth 2000; 47:730–739.

Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
View publication stats