Anaesth Intensive Care 2008; 36: 717-721

Clinical Technique
View of the larynx obtained using the Miller blade and paraglossal approach, compared to that with the Macintosh blade
B. ACHEN*, O. C. TERBLANCHE, B. T. FINUCANE
Department of Anesthesiology and Pain Medicine, University of Alberta Hospital, Edmonton, Alberta, Canada

SUMMARy The purpose of this study was to determine if laryngoscopy using a Miller blade with a paraglossal approach would yield an improved view of the larynx compared to that obtained with a Macintosh blade using the standard approach. One-hundred and sixty-one patients, scheduled for elective surgery requiring tracheal intubation, voluntarily participated in this study. Patients were randomly assigned to one of the two groups (Miller vs. Macintosh). A standard general anaesthetic was administered. Comparisons were made of the percentage of the vocal cords visible at laryngoscopy. The view of the airway was also graded using the Cormack and Lehane scale. Statistical analysis using Fishers exact test was performed. A P value <0.05 was considered statistically significant. The time required to complete intubation and complications if any, were also recorded. Laryngoscopy using the Miller blade allowed 100% of the vocal cords to be viewed in 78% of cases, whereas this was achievable in only 53% with the Macintosh blade (P=0.0014). The Miller blade enabled greater than 25% of the vocal cords to be seen in 95% of the cases, whereas this was achievable in only 80% with the Macintosh laryngoscope (P=0.003). A grade 1 Cormack and Lehane view of the larynx was obtained in 96.5% of cases in the Miller group compared with 85% in the Macintosh group (P=0.02). Direct laryngoscopy using the Miller blade and paraglossal approach, afforded a much-improved view of the larynx in the majority of cases. For this reason trainees should learn laryngoscopy using both blades.
Key Words: laryngoscopy, larynx, blades, paraglossal

Most anaesthesiologists traditionally learn how to perform laryngoscopy and intubation using the Macintosh laryngoscope. Very few trainees use any other method of direct laryngoscopy. The literature is replete with reports of laryngoscope designs and newer methods of performing endotracheal intubation. We examined our ability to perform tracheal intubation using the Miller blade and the paraglossal approach. Many anaesthesiologists perform laryngoscopy with the Miller blade in the same way they would using a curved blade. The

paraglossal approach was described by Magill in 19301, but is seldom used or taught today. The aim of this study was to perform laryngoscopy using the Miller blade and paraglossal approach and the Macintosh blade using the conventional approach and then compare the view of the larynx obtained using these two different techniques. MATERIALS AND METHODS This study was approved by the Ethics Committee of the University of Alberta. A total of 161 patients were randomised to either the Macintosh or Miller group. A random number generator was used to facilitate the individual selections. The patients were randomised based on the last digit being odd or even. The Miller blade was selected for even numbers and the Macintosh blade for odd numbers. There were 11 more patients in the Miller group at the end of the study, with this

*M.D., F.R.C.P.C., Assistant Clinical Professor. M.D., F.R.C.P.C., Staff Anesthesiologist. M.B., B.Ch., F.R.C.P.C., Professor. Address for reprints: Dr B. Achen, Department of Anesthesiology and Pain Medicine, University of Alberta, Clinical Sciences Building, Room 8-120, Edmonton, Alberta T6G 2G3, Canada. Accepted for publication on June 5, 2008. Anaesthesia and Intensive Care, Vol. 36, No. 5, September 2008

718

B. Achen, O. c. TerBlAnche eT Al
using the Miller blade and paraglossal approach. A power analysis was performed to select the number of patients required in each group and to reduce the likelihood of a type II error. The likelihood of obtaining an acceptable view of the vocal apparatus in the control group (Macintosh blade) was estimated to be 80% and an acceptable clinically important difference was judged to be an increase of 15%. Based on a 22 study design (two blades and two outcomes) a total of 150 subjects were required (75 per group). Differences between groups were estimated using Fishers exact test (two-tailed) and the differences were judged to be significant when P <0.05. The Cormack and Lehane3 grading system was also performed and comparisons were made between groups. Following an explanation of the study and informed consent, patients were escorted to the operating room where standard monitors were attached. A standard anaesthetic consisting of propofol 1 to 2 mg/kg, fentanyl 2 to 3 g/kg, rocuronium 0.6 mg/kg was administered. All patients were placed in the sniffing position. Laryngoscopy was performed after complete abolition of the train-of-four response using a peripheral nerve stimulator applied at the ulnar nerve. At laryngoscopy the observer determined if 100% of the vocal cords were visible (optimal view) and if not, a determination was then made if more or less than 25% of the vocal cords were visible. The duration of intubation was defined as follows: the time taken from placement of the laryngoscope in the mouth to the time taken to remove the laryngoscope from the mouth following intubation. All the data were tabulated and then analysed using Fishers exact test. Data are expressed as percentages. A P value <0.05 was considered statistically significant. RESULTS Demographics, including age, weight and Mallampati score were compared and were similar in both groups (Table 1). Laryngoscopy using the Miller blade allowed 100% of the vocal cords to be viewed in 78% of cases, whereas laryngoscopy using the Macintosh blade allowed 100% of the vocal cords to be seen in only 53% of cases (Figure 1). This was statistically significant with a P value of 0.0014. In 95% of cases 25% or more of the vocal cords were visualised using the Miller blade (Figure 2). Twentyfive percent or more of the vocal cords were visualised in 80% of the cases using the Macintosh blade
Anaesthesia and Intensive Care, Vol. 36, No. 5, September 2008

randomisation method. This resulted in 86 patients being enrolled in the Miller group and 75 patients in the Macintosh group. Inclusion criteria were as follows: ASA I to III, ages 18 to 75 and elective surgery. Patients were excluded if rapid sequence induction was required, if intubation was deemed to be difficult or if emergency surgery was required. Patients invited to participate in this study were those attending a tertiary care centre in a large city in Canada. A complete evaluation of the airway was performed preoperatively. This included the Mallampati assessment2, thyromental distance, mouth-opening and neck flexion and extension. If the airway was judged to be difficult using the first three of these criteria, the patient was excluded from the study. The null hypothesis states that the percentage of the vocal cords visible when performing laryngoscopy with the Miller blade and paraglossal approach is comparable to that observed when using the Macintosh blade. Number 3 and 4 blades in each category were used for all intubations. Proficiency using the paraglossal approach with the Miller blade was acquired by the investigators by performing 25 intubations each before the study commenced. Following is a description of the paraglossal approach: this method requires insertion of the laryngoscope blade in the extreme corner of the mouth. An assistant retracts laterally on the cheek and the blade is inserted so that it passes immediately beside the tongue and then next to the tonsillar pillar on the right side of the oropharynx. Once the epiglottis is visualised it is gently lifted to expose the vocal cords. The laryngoscope blade remains lateral to the tongue at all times and no attempt is made to bring the blade base toward the midline. In this way the laryngoscope is in a paraglossal position and the endotracheal tube (ETT) is inserted in the lateral aspect of the mouth. The ETT is directed underneath the laryngoscope blade and its natural curvature brings it back up towards the vocal cords as it is advanced. No attempt is made to insert the ETT beside the blade as there is insufficient room to manoeuvre it successfully to the trachea. This method, once mastered, yields a very high success rate of endotracheal intubation. Bearing in mind that it is difficult to obtain accurate inter-observer data on the ease or difficulty of laryngeal exposure, we assumed, based on a pilot study/current literature, that an acceptable view of the vocal cords (25 to 100% visibility) was observed in approximately 80% of cases using the Macintosh blade. We anticipated an improvement of 10 to 20%

ImprOved vIew Of lArynx wITh mIller lAryngscOpe

TABle 1 Demographics Mean age Miller Macintosh
s

719

% Male 48.9% 51.5%

% Female 51.1% 48.5%

Mean weight 84 kg 79 kg

*MP #1 54% 53%

*MP #2 31% 38%

*MP #3 15% 9%

55.7 y 55.4 y

*MP refers to Mallampati classification, =standard deviation of 18.3 kg, =standard deviation of 17.9 kg, s=standard deviation of 17.0 y, d=standard deviation of 17.7 y.

(Figure 2). This also was statistically significant with a P value of 0.003. A grade 1 Cormack and Lehane view was obtained in 96.5% of cases in the Miller group
100.0% 90.0% 80.0% 70.0% 60.0% 50.0% 40.0% 30.0% 20.0% 10.0% 0.0%

78.0%

compared with 85% of cases in the Macintosh group (Figure 3) (P=0.02). A comparison of the time taken to perform intubation was slightly longer in the Miller blade group (22.7 vs. 17.2 seconds), however this time difference had no clinical significance and therefore did not warrant further statistical analysis. There were no serious complications observed during this study. DISCUSSION Modern laryngoscopy in anaesthesia evolved from direct laryngoscopy performed by ENT surgeons. Magill wrote a report about An Improved Laryngoscope for Anaesthetists in the New Inventions section of the Lancet in 19264. That device retained many of the features of the direct laryngoscope used by ENT surgeons, however the lighting was much improved and the distal end of the spatula was flat, wide and tilted slightly upwards, which was an advantage in elevating and controlling the epiglottis when passing catheters into the larynx. The Miller laryngoscope was introduced in 1941

53.0%

47.0%

22.0%

100% <100% Miller vs. Macintosh Glottis View Observed Miller Mac
fIgure 1: Comparison of the number of cases in the Miller group with a 100% view of the vocal cords compared with the number in the Macintosh group with a 100% view of the vocal cords.

100.0% 90.0% 80.0% 70.0% 60.0% 50.0% 40.0% 30.0% 20.0% 10.0% 0.0%

95.0% 80.0%

100.0%

96.5% 85.0%

80.0%

60.0%

20.0% 5.0%

40.0% 15.0% 3.5% 0.0% Grade 1 Grade 2 to 4 Miller vs. Macintosh Laryngeal Grade Mac Miller
fIgure 3: Comparison of the Cormack and Lehane grades obtained in the two groups shown in percentages.

20.0%

>/= 25% <25% Miller vs. Macintosh Vocal cords visualised Miller Mac

fIgure 2: Comparison of the number of cases in the Miller group with 25% or more of the vocal cords visible compared with the number in the Macintosh group with 25% of the vocal cords visible.
Anaesthesia and Intensive Care, Vol. 36, No. 5, September 2008

720

B. Achen, O. c. TerBlAnche eT Al
placing the ETT. This difference may be explained on that basis that we used the paraglossal approach, which was not used in Arinos study. The paraglossal approach does require diligence to master but is certainly worth the effort and the introduction of improvements in the Miller blade in recent years by Henderson facilitate the paraglossal approach because of the width and overall design of the blade. However there are some disadvantages to straight blades for laryngoscopy. The view is more restricted in patients with limited mouth opening and passage of the tube is difficult in some cases even when one uses the paraglossal approach. The time required to place the ETT in the larynx was slightly longer with the Miller blade (22.7 vs. 17.2 seconds) and this should improve with continued use. One might legitimately ask why the 100% and 25% visualisation measures of the vocal cords were chosen in this study. The reason these measurements were chosen was that the Cormack and Lehane classification (CL) does not allow for a specific and precise delineation of the view of the vocal cords. A CL grade I is the equivalent to 100% exposure of the vocal cords. A CL grade II is the equivalent to less than 25% exposure of the vocal cords; grade III represents a view of the epiglottis only and in a grade IV CL view of the epiglottis is not visible at all. Therefore visualisation of less than 25% of the vocal cords reasonably represents a CL grade II. Ideally in this study, we would have preferred to perform laryngoscopy and intubation using both blades in each patient; however this would be difficult to defend ethically. Also it would have been preferable to include patients with difficult airways in this study seeing that the straight blade and paraglossal approach may offer some advantages in those circumstances. However the diagnosis of the difficult airway and standardising degrees of difficulty would have been a real challenge. Therefore we decided to test our hypothesis in predictably easy airways first. Operator bias was another shortcoming of this study, however blinding of this study was not an option. The Macintosh blade has been in use for more than 60 years and during that time some of the disadvantages of the curvature of that blade have been expressed by laryngoscope designers and anaesthesiologists. Racz11 introduced the Improved Vision Macintosh blade a number of years ago because the curvature of the original Macintosh blade interfered with his ability to see the larynx. In his design, Racz cut into the curvature of the
Anaesthesia and Intensive Care, Vol. 36, No. 5, September 2008

at a time when endotracheal intubation was being performed more frequently. Miller did a comparison with other laryngoscope blades available at the time and cited a number of advantages. It came in just one size and was not designed for infants5. Macintosh introduced his laryngoscope in 1943. Information about this device was published in the New Inventions section of the Lancet6. Macintosh expressed concerns about the use of the long straight blade... Occasionally this manoeuvre jeopardises the patients upper teeth or takes a minor divot out of the posterior pharyngeal wall. He stated the following: The new laryngoscope is designed so that when its short curved blade is in position the tip will fit into the angle made by the epiglottis with the base of the tongue. He goes on to say I can expose the larynx more easily and at a lighter plane of anesthesia than with any of the standard laryngoscopes. Did Macintosh have some purpose in mind in designing a curve for the Macintosh blade? Perhaps the curve was designed to conform with the curved shape of the tongue? Macintosh himself made the following statement about the curve: I find now that users place too much emphasis on the curve For two years I have used blades ranging from those with a well marked curve to the perfectly straight I had some difficulty in deciding which was best for all round use. This quotation is found in a remote correspondence section of the Lancet in 19447. Therefore it would appear that when designing the Macintosh blade, Macintosh had no particular purpose for using a curve at all. The main idea was to design a blade so that the tip slotted into the vallecula and with some pressure on the hyoid bone the epiglottis rotated upwards thereby exposing the vocal cords. Macintosh observed that the vallecula was innervated by the glossopharyngeal nerve whereas the under surface of the epiglottis was innervated by the superior laryngeal nerve which was a branch of the vagus. He indicated that patients reacted far more vigorously physiologically to vagal than glossopharyngeal nerve stimulation. This probably was very relevant when inhalation agents were solely used to anaesthetise patients for endotracheal intubation. The results of this study demonstrate that the Miller blade provides a better view of the larynx than the Macintosh blade and this observation concurs with those of other investigators Bellhouse8 and Henderson9. The results contradict the findings from Arino et al10 who observed better intubating conditions with the Miller blade but had difficulty

ImprOved vIew Of lArynx wITh mIller lAryngscOpe

Macintosh blade thereby improving the line of sight to the larynx. Horton et al12 described the peardrop phenomenon when using the Macintosh blade in patients known to have difficult airways. The peardrop phenomenon is a radiologic sign seen during X-ray laryngoscopy and occurs as a result of compression of the oversized tongue by the Macintosh laryngoscope blade, causing posterior displacement of the epiglottis, thereby worsening the view of an already difficult exposure of the larynx. Macintoshs influence on anaesthesia practices in the United Kingdom and colonial countries was quite significant and still is, and this is the most likely reason why anaesthesiologists in these countries first learn how to perform laryngoscopy using the Macintosh blade. The other reason is that the Macintosh blade is easier to learn in the hands of a beginner. There are certain circumstances when the straight blade may offer some advantages. In some patients the epiglottis is very long and floppy making visualisation of the larynx more difficult. In cases when the epiglottis is difficult to elevate, the straight blade may work. The straight blade has some advantages in patients with awkward dentition. It is easier to pass a bougie using the straight laryngoscope because a greater proportion of the larynx can usually be seen. Finally, we sometimes encounter difficulty exposing the larynx with the Macintosh blade when the tongue is large in relation to the mandibular space. In that situation the Macintosh blade compounds the problem by pushing the epiglottis downwards thereby further impeding ones view of the vocal apparatus. This does not occur when one uses the paraglossal approach with a straight blade. It is unlikely that we will see a major practice change in the use of laryngoscopes. Those who first learn laryngoscopy using the Macintosh blade will most likely continue doing so and vice versa. However, with 36 months available to gain clinical experience in airway management in most training programs, trainees should get some exposure to a variety of airway adjuncts including straight laryngoscope blades. Henderson13,14 has recently designed an improved version of the Miller blade which has a wider spatula and is generally easier to use. Professionals in our discipline first learned

721

to perform laryngoscopy using the straight blade. In the words of Santayana, those who do not remember the past are condemned to relive it. In conclusion, direct laryngoscopy using the Miller laryngoscope blade and paraglossal approach, allowed an improved view of the larynx compared to that obtained with the Macintosh blade. Trainees should be given an opportunity to learn both techniques. ACKNOWLEDGEMENTS We are grateful for the assistance provided by Ms Marilyn Blake who has helped us a great deal in preparation of the manuscript. We would also like to acknowledge the statistical advice provided by Professor A. Clanachan. REFERENCES
1. Magill IW. Technique in endotracheal anesthesia. Br Med J 1930; 2:817-819. 2. Mallampati SE. Clinical signs to predict difficult tracheal intubation (hypothesis). Can Anaesth Soc J 1983; 30:316-317. 3. Cormack RS, Lehane J. Difficult tracheal intubation in obstetrics. Anaesthesia 1984; 39:1105-1111. 4. Magill IW. New Inventions. An Improved Laryngoscope for Anaesthetists. The Lancet 1926; March 6th :500. 5. Miller RA. A new laryngoscope. Anesthesiology 1941; 2:317320. 6. Macintosh RR. New Inventions. A new laryngoscope. Lancet 1943; 1:205. 7. Macintosh RR. Laryngoscope blades. Lancet 1944; 1:485. 8. Bellhouse CP. An angulated laryngoscope for routine and difficult tracheal intubation. Anesthesiology 1988; 69:126-129. 9. Henderson JJ. Questions about the macintosh laryngoscope and technique of laryngoscopy. Eur J Anaesthesiol 2000; 17:25. 10. Arino JJ, Velasco JM, Gasco C, Lopez-Timoneda F. Straight blades improve visualization of the larynx while curved blades increase ease of intubation: a comparison of the Macintosh, Miller, McCoy, Belscope and Lee-Fiberview blades. Can J Anesth 2003; 50:501-506. 11. Racz GB. Improved vision modification of the Macintosh laryngoscope. Anaesthesia 1984; 39:1249-1250. 12. Horton WA, Fahy L, Charters P. Factor analysis in difficult tracheal intubation: laryngoscopy-induced airway obstruction. Br J Anaesth 1990; 65:801-805. 13. Henderson JJ. Questions about the Macintosh Laryngoscope and the technique of laryngoscopy. Eur J Anaesthesiol 2000; 17:2-5. 14. Henderson JJ. Laryngeal view and ease of endotracheal intubation achieved with a new straight laryngoscope (Henderson Laryngoscope). Anesthesiology 1999; 91:563.

Anaesthesia and Intensive Care, Vol. 36, No. 5, September 2008