The Mallampati classification system estimates the size of the tongue relative to the oral cavity.9 A high Mallampati class signifies unfavourable oropharyngeal anatomy for direct laryngoscopy.10 Haemodynamic responses and airway complications are more severe when intubation is difficult or emergent.11,12 BID-guided NTI is independent of direct laryngoscopy and we, therefore, hypothesised that it would be effective in patients with Mallampati class 3. We also hypothesised that BID may cause less haemodynamic changes due to the avoidance of direct stimulation induced by the Macintosh blade.
The primary endpoint under investigation in the present study was the haemodynamic response to NTI. In addition, intubation duration, first-attempt success rate and associated airway complications were compared between groups.
Patients and methods
The present study was approved by the Scientific Research Projects Approval Determination of Independent Ethics Committee of Shanghai Ninth People's Hospital (201026) and was registered at Clinical Trials.gov (NCT 01170455).
Consecutive normotensive patients with American Society of Anesthesiologists (ASA) classification of I or II, between 18 and 60 years old and requiring NTI as part of their anaesthetic management for elective oral and maxillofacial surgery, were screened. During the preoperative visit, the anaesthesiologist noted the patient's modified Mallampati class13 (evaluated while the patient was in the sitting position, with mouth open as wide as possible and tongue protruded maximally without phonation). Those who were classified with a Mallampati class 3 were enrolled. Patients were excluded if an awake NTI was planned (i.e. thyromental distance < 6 cm, inter-incisor distance < 3.5 cm, BMI > 30 kg m−2, neck scars, restricted neck movements or retrognathia). All patients who had oesophageal disease and severe obstructive sleep apnea were excluded. Patients with a history of previous head and neck surgeries were also excluded. Written consent was obtained from all participants. During preoperative interviews, patients were asked which nostril was more unobstructed. If both nostrils were equal and the surgeon had no objection, the right side was chosen.14 Patients were randomly assigned to either a Macintosh laryngoscope group (ML group) or a BID group by computer-generated randomisation codes. All NTIs were practiced by an attending anaesthesiologist experienced with both intubation devices (>100 NTIs with both the Macintosh laryngoscope and the BID).
No preanaesthetic medication was administered. In the operating room, an intravenous infusion of lactated Ringer's solution was started to manage fluid loss from the overnight fast. Cannulation of the radial artery of the right arm with a G22 gauge catheter was performed under local anaesthesia for continuous haemodynamic readings. Mean arterial pressure (MAP), heart rate (HR) and lead II ECG were measured with a multifunction monitor (Spacelabs, Spacelabs Medical, Issaquah, Washington, USA). Other monitoring included SpO2 and end-expiratory carbon dioxide (PetCO2). In both groups, the patient was placed in the supine position with neutral head and flexed neck over a 7-cm thick pillow. Prior to anaesthesia induction, the nasal mucosa was well prepared with 1% tetracaine hydrochloride jelly for 2 min and five drops of ephedrine hydrochloride nitrofurazone (containing approximately 2-mg ephedrine) in all patients. After a stabilisation period of 10 min, baseline haemodynamic values were recorded by a digital camera. Induction agents included fentanyl (2 μg kg−1) and propofol (2 mg kg−1). Upon loss of consciousness and jaw relaxation, manual ventilation was initiated. If manual ventilation was successful, rocuronium (0.6 mg kg−1) was administrated and post-induction values were recorded 5 min after induction. Unsuccessful manual ventilation led to study exclusion. Preformed endotracheal tubes (ETTs) (RAENasal, Mallinckrodt Medical, Athlone, Ireland; ID 6.5 mm in female and ID 7.0 mm in male patients) were used in current study, as it seemed to be used more frequently for oral and maxillofacial surgery.
In the ML group, NTI was performed in a conventional manner.15 First, a preformed ETT well lubricated with 1% tetracaine hydrochloride jelly was inserted into the nostril and advanced to the nasopharynx. Second, a standard Macintosh No.3 or No.4 curved blade was inserted into the patient's mouth to expose the glottis. If necessary, the BURP manoeuvre (backward, upward, right-sided pressure) on the thyroid cartilage was attempted to obtain good glottis visibility.16 Third, the preformed ETT was inserted into the trachea with the aid of Magill's forceps or head flexion, if necessary.
The procedure for BID was as follows. First, after being well lubricated with 1% tetracaine hydrochloride jelly, an oesophagus airway was inserted into the nostril and advanced into the oesophagus until an adequate depth (18 cm for female and 20 cm for male patients). If resistance was felt, the patient's neck was slightly flexed to facilitate oesophagus insertion. Second, the lighted catheter (powered by a battery cell) was inserted through the oesophagus airway. A slight chin lift manoeuvre was applied by a second assistant. A central glow above the laryngeal prominence that could not be moved downward indicated the possibility that the lighted catheter may be impeded in the epiglottic vallecula. In this case, the oesophagus airway was moved slightly downward. If the glow was unclear or absent in the neck during the downward movement, the lighted catheter was inserted into the oesophagus. This indicated that the ventilation hole might be under the glottis and, thus, we pulled the oesophagus airway out slightly.17,18 A bright, well circumscribed circle of glow in the midline at the level of the laryngeal prominence and disappearance at the level of the suprasternal notch indicated that the lighted catheter had passed through the glottis.3,8,17,18 Third, the battery cell was detached, and the oesophagus airway was removed carefully. Fourth, a well lubricated preformed ETT was stretched and guided over the lighted catheter. The ETT was rotated 90° counter clockwise for effectively facilitating advancement into the trachea, if resistance was met.
We stopped any attempt which lasted more than 60 s or which was associated with peripheral oxygen saturation of less than 90% or oesophagus intubation. In the case of an unsuccessful attempt, 100% oxygenation was given through a facemask for 3 min before the next attempt. Failure to intubate was defined as the inability to place the preformed ETT into the trachea after two unsuccessful attempts. At this point, the patient was excluded from NTI characteristics analysis and the other device was attempted. If this failed, fiberoptic NTI was performed by a senior attending anaesthesiologist. In both groups, successful NTI was confirmed with auscultation and end-tidal capnography. Anaesthesia was maintained with sevoflurane and intravenous fentanyl and rocuronium. The presence of persistent bleeding and severe mucosal or dental trauma were considered an indication for discontinuation of the procedure and exclusion from further study.
An observer using a digital camera collected haemodynamic response data every 15 s for the first 60 s of intubation. Data were then recorded every 30 s for an additional 4 min. The maximum values of MAP during observation were also recorded. Additionally, intubation duration, defined as the time from insertion of the device (ETT or oesophagus airway) into the nostril to the time when the preformed ETT was placed, was recorded by a second observer using a stopwatch. Five minutes after intubation, a third observer, who was blinded to group assignment, evaluated the severity of epistaxis. Epistaxis was estimated according to the distance blood travelled up the suction catheter and tube: none (no blood aspirated); mild (blood aspirated by <50 cm); moderate (blood aspirated by 50–300 cm); and severe (blood aspirated by >300 cm).14,19 The number of attempts and episodes of hypoxia during intubation (oxygen saturation less than 90%) was recorded with knowledge of group assignment.
An anaesthetist blinded to group assignment administered a questionnaire to all patients 24 h after the procedure. The questionnaire assessed nasal pain, sore throat and hoarseness on a 10-point visual analogue pain scale (VAS: 0 = none, 10 = worst).18,20
Sample size estimations were based on differences in MAP and HR. A difference of 20 mmHg or 20 beats/min in MAP or HR was considered clinically important. Given an expected SD of ±20 for MAP and HR from previous studies18 and accepting a two-tailed α-error of 5% and β-error of 20%, we estimated that 21 patients would be required in each group. To compensate for patients who would drop out of the study, 25 patients were enrolled in each group.
Data were analysed with Prism 5.0 for Windows (GraphPad Software Inc., La Jolla, California, USA) to calculate mean, SD, median and the interquartile range (IQR). Parametric data were compared using an unpaired t-test. Non-parametric data were analysed with the Mann–Whitney U-test. The rate of success on the first attempt was compared with Fisher's exact tests. Haemodynamic changes within each group were analysed with one-way analysis of variance for repeated measures. Dunnett's multiple comparisons were used to test the difference between the baseline and post-induction values and measurement taken at the different time points. The χ2-test for trend was used to compare the severity of epistaxis between the ML and BID groups. A P value less than 0.05 was considered significant.
A total of 57 eligible patients were interviewed 1 day before operation over a 2-month period. However, seven patients were excluded either because they refused to participate (n = 4) or the operations were delayed (n = 3). Thus, 50 eligible patients were enrolled for randomisation. However, four patients in the ML group were excluded from further study because NTI failed after two attempts. Therefore, only 46 patients were analysed for haemodynamic values and NTI-associated adverse events (Fig. 3).
There was no significant difference between the ML and BID groups in age, sex distribution, weight, height or BMI (Table 1). The overall success rate of NTI was 100% in the BID group and 84% in the ML group (Table 2). There was a significantly higher first attempt success rate in the BID group compared with the ML group (100 vs. 76%, respectively, P = 0.022) (Table 2). The NTI duration was 36 s (IQR 32–40) in the BID group and 33 s (IQR 25.5–41.5) in the ML group (Table 1).
After anaesthesia induction, MAP decreased significantly compared with baseline values in both groups, whereas there was no significant change in HR. Following this initial decrease, there was a significant increase in MAP observed in both groups within 15–240 s after induction. Compared with baseline values, the increase in MAP persisted longer (within 30–90 s) in the ML group than in the BID group (60 s). In addition, the ML group showed a significantly higher MAP value compared with the BID group within the 30–60 s period (Fig. 4). The difference between the maximum MAP and the post-induction value was significantly greater in the ML group than in the BID group (64.4 ± 16.1 vs. 45.9 ± 16.1 mmHg, P = 0.0003, Table 1). A significant HR increase was observed within 45–180 s after NTI in the ML group and within 45–120 s after NTI in the BID group. There was no significant difference in HR at any time points between groups (Fig. 5).
Epistaxis during NTI was less frequent and less severe in the BID group than in the ML group (P = 0.031, Table 1). Nasal pain, hoarseness and sore throat were self-limited adverse events and occurred similarly in both groups (Table 1). There was no persistent bleeding, severe mucosal or dental trauma or hypoxemia observed with NTI.
The three main causes of the haemodynamic response to NTI are the stimulation of the nasopharyngeal structures, oropharyngeal structures and trachea induced by laryngoscopy or ETTs.21,22 The current study clearly showed that NTI using BID caused a smaller and shorter haemodynamic response compared with Macintosh laryngoscope in anaesthetised Mallampati class 3 adults. The results strongly demonstrate that the BID technique may provide clinical advantages in attenuating the haemodynamic responses to NTI.
Prior research has shown that a higher Mallampati class indicates a greater rate of unfavourable events associated with Macintosh laryngoscope.10 For enhanced glottic exposure, it was necessary to increase the upward lifting force of the Macintosh blade and to compress the laryngeal prominence that distorted the oropharynx structure. The haemodynamic response to NTI could be significantly attenuated by reducing the upward lifting force of the laryngoscope.23 On the contrary, BID-guided NTI avoided direct laryngoscopy which minimised stimuli applied to the oropharyngeal structures during intubation.18 In addition, Magill's forceps were often used to direct the tip of the ETT through the glottis in the ML group.15 It is also likely that the curve of the ETT was causing the tip to abut the anterior laryngeal wall due to the anterior elevation of the larynx during invasive laryngoscopy, necessitating a combination of manoeuvres such as withdrawing the tube slightly and flexing the neck to facilitate ETT advancement. In contrast, the BID-guided NTI resulted in far less distortion of the anterior airway anatomy than the Macintosh laryngoscope. The shape of the oesophagus airway is in alignment with the nasopharyngeal curve. The lighted catheter is soft and flexible and could slide into the vocal cords easily with minor adjustment of the oesophagus airway.18 Furthermore, ETT insertion over the lighted catheter potentially provided a more direct route from the nasopharynx to the trachea, consequently causing a lesser haemodynamic response. Moreover, the BID technique comprises a ‘deliberate oesophagus insertion then withdraw’ procedure to bypass the epiglottis and locate the ventilation hole below it. Through this method, BID avoids the jaw thrust or jaw lift manoeuvre necessary to elevate the epiglottis from the posterior wall of the pharynx in lightwand-guided intubation. These manoeuvres are sufficient to cause a haemodynamic responses to elevate the epiglottis from the posterior wall of pharynx which is necessary in lightwand-guided intubation.1,24,25
Although trachea irritation caused by ETT produced a marked hypertensive response in both groups, the BID group showed significantly attenuated haemodynamic changes during the period from 30 to 60 s after NTI. Moreover, the maximum fluctuation of MAP in the BID group was also significantly attenuated. These results indicated that in Mallampati class 3 patients, the overall stimulation as a result of BID-guided NTI was less than that of Macintosh laryngoscope. The present study also showed moderate fluctuations in HR. There was an initial decrease in HR which was most likely caused by the stimulus of the nasal cavity mucosa evoking a nasal cardiac reflex. Although these results confirmed our initial hypothesis, it should be emphasised that the dosage of fentanyl and propfol used in the study could not sufficiently inhibit the haemodynamic response to NTI in either device, and further study is needed.
BID-guided NTI had a higher first attempt success rate than procedures performed with the Macintosh laryngoscope. Moreover, four patients initially assigned to the direct laryngoscope group were eventually intubated successfully with BID on the first attempt. Three patients (No. 1, 6 and 12) exhibited a Cormark–Lehane grade 4 glottic view even after the BURP manoeuvre was applied and could not be intubated through the Macintosh laryngoscope approach. In one patient (No. 15), the ETT could not be adequately aligned with the glottis even with the help of Magill's forceps despite good laryngoscopy. The NTI failed because the tip of the ETT deviated from the glottis inlet and oesophagus intubation accidentally occurred twice when the patient's head was flexed. The patient was eventually easily intubated on the first attempt with the BID. These results clearly demonstrate that the BID could serve as a promising backup alternative for failed NTI. The median time required for NTI was about 3 s longer in the BID group than in the ML group which was most likely due to the additional oesophagus airway placement prior to ETT intubation. However, a 3-s delay does not represent a major drawback, although a high success rate of intubation and a low frequency of airway complications are more crucial.
The most common NTI complication is epistaxis induced by the conventional ETT sharp-edged Murphy eye.14,26 The present study indicated that the BID was effective in reducing epistaxis during NTI despite two nasal insertions. Previous studies have suggested that epistaxis associated with NTI can be prevented by use of a curve-tipped suction catheter for guidance,27 mechanical dilation of nasopharyngeal airways28 and an oesophageal stethoscope-obturated ETT.14 In the BID group, the first insertion of the oesophagus airway seldom caused epistaxis (most likely due to a round and soft tip). In addition, the oesophagus airway also served as a nasal cavity dilator. The second nasal insertion of the ETT was guided by the lighted catheter which may also reduced nasal damage.27 There were no significant differences in the occurrence of nasal pain, sore throat or hoarseness between the BID and ML groups. However, surgical procedures tend to cause similar symptoms which may confound the results. In addition, these variables were our secondary endpoints in the present study. Further specific investigation is required to more thoroughly investigate this issue.
A potential advantage of the BID is that the patient can be ventilated until the lighted catheter is inserted into the trachea successfully. The oesophagus airway is a nasopharyngeal ventilation device and has been used as a nasal airway rescue device in a critical event.29 Furthermore, the oesophagus airway can be connected to a standard anaesthetic circuit. Moreover, the 3.5-mm lighted catheter can easily be aligned with the suction hole of an extension tube (Breathing System, Tyco Healthcare, Mallinckrodt Dar S.r.l, Italy) (Fig. 6). Passage through the suction hole allows mechanical ventilation through the oesophagus airway and monitoring of PetCO2 while the patient's mouth is sealed by adhesive plaster. Nofal3 reported this type of approach to oxygenate the patient during an awake NTI. We have found that, with a simple modification (sealing the mouth and nose), these procedures are effective for mechanical ventilation. The patients in the current study were intubated successfully within 60 s with no hypoxia precluding the necessity of this method. However, the ability to provide continuous mechanical ventilation can be of great benefit in cardiorespiratory compromised patients or prolonged BID-guided NTI.
An important factor in determining the success of BID-guided NTI is accurate depth of the oesophagus airway. Previous clinical experience has shown very small inter-patient deviation in the depth of oesophagus airway (about 20 cm for male and 18 cm for female patients), suggesting that we can easily apply these two constants on the basis of sex. In addition, the depth of the oesophagus airway can be adjusted according to the location of the glow as mentioned in the methods section.
The major limitation of the current study is, first, the lack of a lightwand control group. The lightwand is a widely accepted technique and should be directly compared to any new transillumination device. Second, the use of patients with Mallampati class 3 alone has only moderate diagnostic value.10,12,30 Further study to assess the performance of the BID in patients with more difficult laryngoscopy problems is required. Third, patients with cardiovascular diseases and hypertension were excluded from the present study. These patients may need greater protection from the stress reaction induced by NTI and this issue should be addressed in future studies. Finally, there was only one operator performing the procedures in the current study and, thus, the results may be biased by the anaesthesiologist's distinctive techniques.
In summary, the use of the BID in NTI leads to attenuated haemodynamic responses, a higher first-attempt success rate and decreased epistaxis as compared with Macintosh laryngoscope use. We conclude that in experienced hands, the BID is an effective alternative to the Macintosh laryngoscope for NTI in anaesthetised adults with Mallampati class 3.
This work was entirely funded by department internal resources. Both H.J. and Y.-S.Z. are the inventors of the BID. They were authorised by the Chinese Ministry of Health for BID-guided NTI technique promotion programs. No person involved in this study has any financial relationship with Anhui Xiao Shan Hygienic Material Co. Ltd.
The authors would like to thank all participants in this study for their willing cooperation.
1. Favaro R, Tordiglione P, Dilascio F, et al. Effective nasotracheal
intubation using a modified transillumination
technique. Can J Anaesth
2. Yamamura H, Yamamoto T, Mamiyama K. Device for blind nasal intubation. Anesthesiology
3. Nofal O. Awake light-aided blind nasal intubation: prototype device. Br J Anaesth
4. Asai T. Use of the endotrol endotracheal tube and a light wand for blind nasotracheal
5. Cheng KI, Chang MC, Lai TW, et al. A modified lightwand-guided nasotracheal
intubation technique for oromaxillofacial surgical patients. J Clin Anesth
6. Xue FS, Yang QY, Xu YC, et al. A simple method to use the preformed nasotracheal
tube during the lightwand-guided awake nasal intubation in the patients with difficult airways
. Chin Med J (Engl)
7. Chen TH, Tsai SK, Lin CJ, et al. Does the suggested lightwand bent length fit every patient? The relation between bent length and patient's thyroid prominence-to-mandibular angle distance. Anesthesiology
8. Jiang H, Sun Y, Zhu Y. A useful device for difficult nasal tracheal intubation in China. Eur J Anaesthesiol
9. Mallampati SR, Gatt SP, Gugino LD, et al. A clinical sign to predict difficult tracheal intubation: a prospective study. Can Anaesth Soc J
10. Lee A, Fan LT, Gin T, et al. A systematic review (meta-analysis) of the accuracy of the Mallampati tests to predict the difficult airway. Anesth Analg
11. Mort T. Emergency tracheal intubation: complications associated with repeated laryngoscopic attempts. Anesth Analg
12. Rhee K, Lee J, Kim J, et al. A comparison of lighted stylet (Surch-Lite) and direct laryngoscopic intubation in patients with high mallampati scores. Anesth Analg
13. Samsoon GLT, Young J. Difficult tracheal intubation: a retrospective study. Anaesthesia
14. Seo KS, Kim JH, Yang SM, et al. A new technique to reduce epistaxis and enhance navigability during nasotracheal
intubation. Anesth Analg
15. Benumof J. Conventional (laryngoscopic) orotracheal and nasotracheal
intubation (single-lumen tube). In: Airway management: principles and practice. St Louis, Missouri: Mosby-Year Book; 1996. pp. 261–276.
16. Takahata O, Kubota M, Mamiya K, et al. The efficacy of the ‘BURP’ maneuver during a difficult laryngoscopy. Anesth Analg
17. Sun Y, Jiang H, Zhu Y, et al. Blind intubation
device for nasotracheal
intubation in 100 oral and maxillofacial surgery patients with anticipated difficult airways
: a prospective evaluation. Eur J Anaesthesiol
18. Sun Y, Liu JX, Jiang H, et al. Cardiovascular responses and airway complications following awake nasal intubation with blind intubation
device and fibreoptic bronchoscope: a randomized controlled study. Eur J Anaesthesiol
19. Lu PP, Liu HP, Shyr MH, et al. Softened endotracheal tube reduces the incidence and severity of epistaxis following nasotracheal
intubation. Acta Anaesthesiol Sin
20. Joo HS, Kapoor S, Rose DK, et al. The intubating laryngeal mask airway after induction of general anesthesia versus awake fiberoptic intubation in patients with difficult airways
. Anesth Analg
21. Singh S, Smith J. Cardiovascular changes after the three stages of nasotracheal
intubation. Br J Anaesth
22. Zhang GH, Xue FS, Sun HY, et al. Comparative study of hemodynamic responses to orotracheal intubation with laryngeal mask airway and direct laryngoscope. Chin Med J (Engl)
23. Xue FS, Li XY, Liu QJ, et al. Circulatory responses to nasotracheal
intubation: comparison of GlideScope videolaryngoscope and Macintosh direct laryngoscope. Chin Med J (Engl)
24. Hirabayashi Y, Hiruta M, Kawakami T, et al. Effects of lightwand (Trachlight) compared with direct laryngoscopy on circulatory responses to tracheal intubation. Br J Anaesth
25. Davis L, Cook-Sather SD, Schreiner M. Lighted stylet tracheal intubation: a review. Anesth Analg
26. Piepho T, Thierbach A, Werner C. Nasotracheal
intubation: look before you leap. Br J Anaesth
27. Morimoto Y, Sugimura M, Hirose Y, et al. Nasotracheal
intubation under curve-tipped suction catheter guidance reduces epistaxis. Can J Anaesth
28. Kay J, Bryan R, Hart HB, et al. Sequential dilation: a useful adjunct in reducing blood loss from nasotracheal
1985; 63 (3A):A259.
29. Sun Y, Jiang H, Zhu Y. Airway rescue with an esophageal airway in a child with a huge venous malformation in oral and maxillofacial region: a case report. Paediatr Anaesth
30. Toshiya S, Zen’ichiro W, Tetsuo I, et al. Predicting difficult intubation in apparently normal patients: a meta-analysis of bedside screening test performance. Anesthesiology
Keywords:© 2011 European Society of Anaesthesiology
airways; anaesthesia; blind intubation; nasotracheal; transillumination