Unanticipated difficult or failed intubation following induction of general anaesthesia for caesarean delivery remains potential life-threatening complication.1,2 As exposure to obstetric general anaesthesia continues to decline, and with it experience, this is unlikely to change. Ability to predict problem patients would be an advantage and several tests have been proposed to predict difficult laryngoscopy in pregnant women.3 Our aim was to investigate five bedside predictors that might identify women with a potential intubation difficulty immediately prior to emergency caesarean section.
Patients and methods
After Local Research Ethics Committee approval, this prospective observational study was conducted at the Vakif Gureba Hospital, Istanbul between March 2002 and June 2005. Written informed consent was obtained from all. Consecutive women requiring general anaesthesia for emergency caesarean section during the periods of duty of one anaesthesiologist, excluding those unable to provide adequate personal data and those with a history of airway or intubation difficulty, were recruited.
Prior to tracheal intubation, the age, height, current weight and weight gain during pregnancy were noted. Parity and gestation were not recorded. The current weight was measured on foot scales in the delivery room, and mouth opening and Mallampati score4 were assessed with the women standing immediately prior to transfer to the operating room. Mallampati scores were as follows:
- soft palate, fauces, uvula and tonsillar pillars visible
- soft palate, fauces and uvula visible
- soft palate and base of uvula visible
- soft palate not visible
Mouth opening was assessed using the maximal interincisor gap.
In the operating room lying supine, with the head on a gel-filled ring in the intubation position, sternomental distance and thyromental distance were measured with a graduated tape, and atlantooccipital extension was measured with a bubble goniometer. Sternomental distance was from the sternal notch to the mental protuberance. Thyromental distance was from the thyroid notch to the mental protuberance. Atlantooccipital extension was the angle between the horizontal and a line linking the tragus of the ear to the angle of the mouth.
General anaesthesia was induced with thiopentone 5–6 mg kg−1 and succinylcholine 1 mg kg−1 as part of a rapid sequence without cricoid pressure. Laryngoscopy was performed with a size 3 Macintosh laryngoscope and the view was graded according to Cormack and Lehane.5
Appearance at laryngoscopy was graded as follows:
- good view of glottis
- only posterior part of glottis seen
- only epiglottis visible
- epiglottis not visible
Difficult laryngoscopy was predicted by interincisor gap less than 3 cm, thyromental distance less than 6 cm, sternomental distance less than 12.5 cm, atlantooccipital extension of 80° or less and Mallampati score of at least 3. Difficult intubation was defined at laryngoscopy as Cormack and Lehane grades 3 or 4.2,6,7 For each assessment, the same method of examination was used.
For analysis, women were divided into two groups (easy or difficult) according to appearance at laryngoscopy. The easy group consisted of Cormack and Lehane grades 1 and 2 and the difficult group grades 3 and 4. Exact logistic regression was used to identify significant independent predictors for difficult intubation with two-tailed P value less than 0.05 as significant. Data are presented as mean (SD) and median [range] and count as appropriate. Analyses were performed using LogXact 8.0, Cytel Inc., Cambridge, Massachusetts, USA and Number Cruncher Statistical Systems (NCSS) 2007, NCSS Inc., Kaysville, Utah, USA.
In 3 years, 239 women were recruited. None of the patients had a history of difficult intubation. Cormack and Lehane grade of 2 or less (easy) was found in 225 and grade of at least 3 (difficult) in 14 women. The proportion of women with difficult laryngoscopy was 0.059 [95% confidence interval (CI) 0.032–0.096], which represents an incidence of one of 17 (95%CI from 1/31 to 1/10).
There were no significant effects or differences in age, height, weight, BMI or weight gain with respect to laryngoscopy and intubation (Table 1). Sixty-five (28.9%) and six (42.9%) patients had BMI more than 30 kg m−2 in easy and difficult groups, respectively (P = 0.36).
There were 24 abnormal tests in 20 women. Four women had two abnormal tests. One woman was in the Cormack and Lehane 3 group, whereas others were in Cormack and Lehane 1–2 group. Abnormal tests were thyromental distance and sternomental distance in all women. Interincisor gap and atlantooccipital extention were normal in both groups. Thyromental distance was abnormal in five of 225 women in group 1 and one of 14 in group 2. Sternomental distance was abnormal in six of 225 in group 1 and one of 14 in group 2. Mallampati scores I/II/III were 176/40/9 and 10/2/2 in groups Cormack and Lehane 1-2 and Cormack and Lehane 3, respectively. None of these differences reached statistical significance. The results for the predictive tests for difficult intubation are shown in Table 2 and the summary of the combination of the five tests is shown in Table 3.
A positive result from any of the five predictors combined had a sensitivity of 0.21 (95%CI 0.05–0.51), a specificity of 0.92 (95%CI 0.88–0.96), a positive predictive value of 0.15 (95%CI 0.032–0.38) and a negative predictive value of 0.95 (95%CI 0.91–0.97) for a Cormack and Lehane grade of at least 3 at laryngoscopy.
Assessment of the airway before general anaesthesia for caesarean section is recommended8 despite the poor predictive value of individual tests.9 Using a combination of tests can improve predictability of a difficult intubation, but10,11 the routine use of Mallampati scoring, neck mobility testing and thyromental distance measurement in elective anaesthesia cases failed to show the desired improvement.12,13 In addition, a meta-analysis of the performance of bedside screening came to the conclusion that the prediction of a difficult intubation was of limited value.14 Our results confirm that even in the emergency situation in women in labour, bedside airway assessment was unhelpful in predicting those with a Cormack and Lehane grade 3 appearance at laryngoscopy.
Five tests were chosen in the hope that by increasing the number of tests, predictability of a difficult intubation might be improved. The choice of the five tests reflects what can practically be achieved in the minutes between decision to deliver and emergency caesarean delivery. A combination of clinical and radiological tests might improve on the positive predictive value seen here.10,11 Sophistication in terms of acoustic reflectometry, airway photography and volume measurement15 might represent the state-of-the-art, but are not likely to become available as bedside assessments in the emergency situation.
Airway photography was employed to assess Mallampati score in pregnant women at 12 weeks' gestation and 38 weeks. The Mallampati score increased in line with weight gain as pregnancy progressed.16 If this was a simple correlation, then the incidence of difficult intubation would increase with BMI. Our data suggest that BMI, as a single measure, is not necessarily predictive, and others have considered the role of fluid retention in explaining differences seen in pregnancy.16 Sophisticated airway assessment has also been compared at onset and the end of labour and delivery. Airway assessment significantly worsened during labour, but without any correlation with duration of labour or the quantity of fluids administered.15 The authors advise airway assessment if anaesthesia is required during labour, and the purpose of our study was to evaluate this and similar advice in the genuine clinical setting of emergency caesarean section.
A major limitation of this study is the fact that the anaesthesiologist who made the assessments also performed and scored the laryngoscopy. Bias might be expected in making a Cormack and Lehane 3 scoring more likely in those women who had one or more abnormal assessments. If such bias had operated here, we should expect that it would serve to improve the predictive value of abnormal tests. In that event, our results would overstate the case. Either way, it is evident that even if our case is overstated, the conclusion is unchanged; the positive predictive value is so poor that much more than bias is required to make the assessments worthwhile. It should also be accepted that some compromise is necessary if studies of this nature are to be conducted in unpredictable emergency situations occurring around the clock.
Further evidence that bias played a small role is in the incidence of Cormack and Lehane 3 appearance at 5.9% (1: 17). This compares well with the incidence of difficult intubation from the meta-analysis at 5.8% (95%CI 4.5–7.5%).14 In obstetric patients, a higher incidence (7.9%) has been quoted.2 Because in the UK most anaesthesia for out of hours caesarean sections is given by anaesthetists in training, caution should be observed when comparisons are made with other countries, where senior anaesthesiologists undertake this role. Another factor that might influence the incidence is the use of cricoid pressure, which was not used here. Inexpert application of cricoid pressure can distort laryngeal anatomy and may complicate tracheal intubation.17
In conclusion, the five bedside airway assessments have been of limited value in the prediction of a Cormack and Lehane 3 appearance at laryngoscopy. The low sensitivity means that 79% (95%CI 49–95) of difficult intubations will be missed using this combination of assessments. Although perceived wisdom is that such assessments should be part of preoperative preparation for emergency caesarean section, the effort required might be put to better use ensuring that facilities for dealing with a Cormack and Lehane 3 appearance at laryngoscopy are routinely available for all cases and that a difficult intubation drill is ingrained in the minds of all anaesthesiologists performing general anaesthesia for emergency caesarean section.
1 Munnur U, de Boisblanc B, Suresh MS. Airway problems in pregnancy. Crit Care Med 2005; 33:259–268.
2 Ezri T, Szmuk P, Evron S, et al
. Difficult airway in obstetric
anesthesia: a review. Obstet Gynecol Surv 2001; 56:631–641.
3 Honarmand A, Safavi MR. Prediction of difficult laryngoscopy in obstetric
patients scheduled for caesarean delivery. Eur J Anaesthesiol 2008; 25:714–720.
4 Mallampati SR, Gatt SP, Gugino LD, et al
. A clinical sign to predict difficult tracheal intubation: a prospective study. Can Anaesth Soc J 1985; 32:429–434.
5 Cormack RS, Lehane J. Difficult tracheal intubation in obstetrics. Anaesthesia
6 Kuczkowski KM, Reisner LS, Benumof JL. The difficult airway: risk, prophylaxis and management. In: Chestnut DH, editor. 3rd ed. Elsevier Mosby; 2004. pp. 535–561.
7 Rose KD, Cohen MM. The incidence of air-way problems depends on the definition used. Can J Anaesth 1996; 43:30–34.
8 Glosten B. Anesthesia for obstetrics. In: Miller RD, editor. Anesthesia
. New York: Churchill Livingstone; 2000. pp. 2024–2068.
9 Karkouti K, Rose KD, Wigglesworth D, Cohen MM. Predicting difficult intubation: a multivariable analysis. Can J Anesth 2000; 47:730–739.
10 Naguib M, Malabarey T, AlSatli RA, et al
. Predictive models for difficult laryngoscopy and intubation. A clinical, radiologic and three-dimensional computer imaging study. Can J Anaesth 1999; 46:748–759.
11 Iohom G, Ronayne M, Cunningham AJ. Prediction of difficult tracheal intubation. Eur J Anaesthesiol 2003; 20:31–36.
12 Ulrich B, Listyo R, Gerig HJ, et al
. The difficult intubation: the value of BURP and 3 predictive tests of difficult intubation. Anaesthesist 1998; 47:45–50.
13 Yentis SM. Predicting difficult intubation: worthwhile exercise or pointless ritual? Anaesthesia
14 Shiga T, Wajima Z, Inoue T, Sakamoto A. Predicting difficult intubation in apparently normal patients: a meta-analysis of bedside screening test performance. Anesthesiology 2005; 103:429–437.
15 Kodali BS, Chandrasekhar S, Bulich LN, et al
. Airway changes during labor and delivery. Anesthesiology 2008; 108:347–349.
16 Pilkington S, Carli F, Dakin MJ, et al
. Increase in Mallampati score during pregnancy. Br J Anaesth 1995; 74:638–642.
17 Vanner R. Cricoid pressure. Int J Obstet Anesth 2009; 18:103–105.