Secondary Logo

Journal Logo

Patient Safety: Research Report

Laryngoscopy and Tracheal Intubation in the Head-Elevated Position in Obese Patients: A Randomized, Controlled, Equivalence Trial

Rao, Srikantha L., MBBS, MS*; Kunselman, Allen R., MA; Schuler, H Gregg, MS, CCRC; DesHarnais, Susan, PhD

Section Editor(s): Brull, Sorin J.

Author Information
doi: 10.1213/ane.0b013e31818556ed
  • Free
  • Chinese Language Editions

Excellent intubating conditions are imperative for direct laryngoscopy and the efficient placement of a tracheal tube. The proper positioning of a patient before induction is a key step. Classic teaching has been to position the patient in the “sniffing” position, or supine with moderate head elevation and atlanto-occipital extension.1 Although some studies have reported that tracheal intubation is more difficult in obese as compared to normal weight patients2,3 others, who have reported no difference between the two groups, attributed it to a ramped position with the patient's head clearly elevated above their shoulders to improve laryngeal exposure in obese patients.4,5

In a randomized study with a crossover design of 40 non-obese patients, significantly improved laryngeal view was demonstrated by performing laryngoscopy in the 25-degree head-up position compared to the supine position in the same patients.6 In a study of 60 obese patients undergoing bariatric surgery, the ramped position improved laryngeal view when compared to a standard sniff position.7

The standard sniffing position is advantageous not only for laryngoscopy but also for mask ventilating the patient's lungs before tracheal intubation. Due to the change in lung and chest wall compliance, obese patients tend to desaturate quickly and may be better served if they are placed in the head-elevated position. In obese patients, administration of oxygen in the 25-degree head-up position allows a greater safety margin for induction of anesthesia than the supine position by achieving 23% higher oxygen tensions.8

In our hospital, the routine practice during patient positioning before induction is to use folded blankets, stacked under the patient's upper body, neck, and head to elevate the head. However, the number of blankets used differs among patients, and obtaining the optimum position in each patient is tedious, as it requires adding or removing blankets while repositioning the patient each time (trial and error method). Use of other devices including a commercially available foam pillow (Troop Elevation Pillow, Mercury Medical, Clearwater, FL) to achieve this position has been described in recent literature.9,10 This head-up position also can be achieved by a simple maneuver of configuring the operating room (OR) table, similar to a reclining chair with the back or trunk portion of the table up. With the patient lying on the OR table, the electronic table controls can be used to flex the table at the trunk-thigh hinge and raise the “back” or “trunk” section of the table up as necessary to achieve the optimum position. This can be done with or without the head-piece at the head end of the table. This prospective, randomized study was designed to objectively assess the two methods of achieving the head-up position, and the time required to secure the airway.


The study was approved by the Human Subjects Protection Office of the Penn State Hershey Medical Center, and it was conducted in the OR of a university teaching hospital with an anesthesia residency program. Written informed consent was obtained from 85 patients between March 2007 and August 2007. All 85 patients were either American Society of Anesthesiologists (ASA) class I or II, age 21–65 yr, and undergoing elective surgery under general anesthesia. Of the 85 patients, 64 (31 blanket group, 33 table-ramp group) underwent bariatric surgery (open or laparoscopic gastric bypass, or gastric banding), whereas the remainder 21 (12 blanket group, 9 table-ramp group) underwent orthopedic, urologic, or gynecologic surgery. Data from our pilot study of 11 subjects undergoing bariatric surgery were used to obtain an estimate of the variability in the time required to secure the airway to properly power our trial. The time required for securing an airway was measured using the current method of patient positioning using blankets.

The patient's height and weight on the day of preoperative evaluation were used to calculate the Body Mass Index (BMI), and only patients with a BMI >30 kg/m2 were approached to be enrolled in the study. Patients with a history of difficult intubation were not enrolled. Each patient's medical history was reviewed. A specific diagnosis of obstructive sleep apnea was sought in each medical record in those patients who had undergone a formal sleep study.11,12 Age and gender of the patient were noted. The visibility of oropharyngeal structures was assessed using the Mallampati Classification Method as modified by Samsoon and Young. Neck circumference (cm) at the level of the thyroid cartilage, width of mouth opening (inter-incisor distance) (cm), and thyromental distance (cm) were measured in the preoperative area and recorded. These variables were chosen as predictors of difficult intubation based on previous studies.3,13,14

After obtaining written informed consent, 2 mg of midazolam was administered IV to all patients before transport to the OR. Patients were then randomized using permuted blocks randomization scheme of size 6 (known only to the biostatistician during the study) with subjects equally allocated to be positioned using either the 1) blanket method or 2) OR table-ramp method. The trial was designed to be stopped if there were more than eight intubation failures (defined as more than three attempts at laryngoscopy and intubation) in the 85 study patients (Appendix).

In the OR, patients in the blanket group lay on a ramp made by layering multiple folded blankets on a flat OR table. Blankets were then added or removed to ensure that the patient's head was above the shoulders and the external auditory meatus and sternal notch were in the same horizontal plane (Fig. 1). Patients in the table-ramp group were placed on the flat OR table with a hospital pillow under their heads to elevate their occiput, and the OR table was then reconfigured to the position of a reclining chair. With the patient lying on the OR table, the electronic table controls were used to flex the table at the trunk-thigh hinge and raise the “back” or “trunk” section of the table up as necessary to achieve the optimum position of aligning the external auditory meatus to the sternal notch. This was done with the head-piece at the head end of the table in tall patients and without the head-piece in other patients. The foot portion of the table was lowered by flexing the leg-thigh hinge so that the knees were slightly flexed to avoid stretching the sciatic nerve (Fig. 2). An Amsco 3085 surgical table (Steris Corp. Mentor, OH) was used to position the patients undergoing bariatric surgery. During normal use, the head piece of the table can be moved independently to elevate the patient's head if necessary to improve glottic visualization. During bariatric surgery, the patients were positioned on the OR table at induction in such a manner that they could be placed in low lithotomy position during surgery without moving them. This required removal of the head-piece of the OR table and its placement at the foot end. As the patient's head now rested on the back or trunk section of the OR table instead of the head-piece, the ability to move the head-piece up to flex the patient's head, if necessary, during laryngoscopy was lost. Consequently, if further flexion was required we would have had to actively elevate the patient's head manually instead of elevating the head-piece. This maneuver was not required in any of the study patients.

Figure 1.
Figure 1.:
Head-elevated position achieved by placing multiple blankets under the patient's upper body.
Figure 2.
Figure 2.:
Head-elevated position achieved by table-ramp created by flexing the table at the trunk-thigh hinge and raising the back or trunk section of the table up. Note the head-piece is missing in this case.

Using routine monitoring, anesthesia was induced using 100 μg of fentanyl IV and 2–2.5 mg/kg (ideal body weight) propofol IV to achieve hypnosis (loss of eyelash reflex). Muscle relaxation was achieved with 1.0–1.5 mg/kg of succinylcholine or 0.6–1.0 mg/kg of rocuronium (lean body mass). The choice of muscle relaxant was at the discretion of the attending anesthesiologist in the room. Manual bag-mask ventilation was attempted before laryngoscopy. An oral airway was inserted if the airway was obstructed. After loss of visible twitches using the single 1 Hz stimulation at the ulnar nerve, laryngoscopy was performed using a Macintosh blade size 3. The same anesthesiologist positioned all patients. The first laryngoscopy and attempt at tracheal intubation were performed by the anesthesia resident or certified registered nurse anesthetist (CRNA) caring for the patient that day. Second and subsequent laryngoscopies using a Macintosh 4 or Miller 2 blade were performed by the attending anesthesiologist taking care of the patient. Each patient's trachea was intubated under direct vision using an endotracheal tube (8.0 mm ID for men; 7.5 mm ID for women).

After obtaining the best possible view, the laryngoscopist was allowed to choose between an unstyletted or styletted endotracheal tube. The best view obtained during laryngoscopy was graded by the Lehane-Cormack classification as reported by the laryngoscopist performing the intubation.15 A visual chart on each data collection form was used to review the classification with the laryngoscopist before documentation. The number of attempts at laryngoscopy and tracheal intubation were noted separately. The time interval between the loss of consciousness (loss of eyelash reflex) and detection of CO2 on the end-tidal CO2 monitor after the successful placement of the tracheal tube was noted as the time to secure the airway. Although the anesthesia resident or CRNA and the attending anesthesiologist administering the anesthetic could not be blinded to the method used to position the patient, they were not aware of the data points with regard to the outcome variables. They were, however, informed that after three failed attempts at laryngoscopy and intubation, the patient would be managed at the discretion of the attending anesthesiologist using the anesthesia department's difficult airway practice guidelines. A “call for help” would be generated and the attending anesthesiologist would be free to choose the next modality or airway device at their discretion. The plan per the research study protocol was to insert an intubating laryngeal mask airway and attempt ventilation and tracheal intubation through the laryngeal mask airway (under fiberoptic guidance if necessary).

Statistical Analysis

From our pilot study in 11 patients who underwent bariatric surgery and were positioned using blankets, an estimate of variability in the time to secure the airway was obtained (mean time = 160 s, sd = 81 s) for use in power and sample size estimation for this study. We hypothesized that the time to tracheal intubation in obese patients would be equivalent between the blanket and table-ramp groups. If the mean time to tracheal intubation in the blanket group is μb and the mean time in the table-ramp group is μt, our hypothesis was that the difference in the mean time to intubation between the two groups is no less than and no more than the maximum allowable limit of clinical significance (ΔL ≤ [μb − μt] ≤ ΔU). The sample size of 42 per group (84 total) was chosen using the maximum allowable limit of no clinical significance, Δ, of 55 s (i.e., [ΔL, ΔU] = [−55,55]), with 85% statistical power to reject the null hypotheses in favor of the alternative hypothesis that the two treatment groups are equivalent in time to tracheal intubation, assuming the expected difference in means to be zero and a common standard deviation of 81 s using two one-sided tests with a significance level of 0.05.

All secondary continuous outcomes were compared using the two-sample t-test (normal distribution) or the Mann-Whitney test (nonnormal distribution). Categorical data were compared using Pearsons χ2, Fisher's exact tests or a χ2 test for trend as appropriate. All analyses were performed with SAS for Windows Version 9.1 (SAS Institute, Cary, NC). Data are reported as mean (±sd).


Eighty-five patients (15 men and 70 women) were randomized to either the blanket group (n = 43) or the table-ramp group (n = 42). Three protocol violations were noted. In one patient randomized to the table-ramp group, who was recruited and consented for the study, anesthesia was induced without the investigator being present and was excluded from analysis as no data were collected. Two patients in the table-ramp group had their first laryngoscopy and tracheal intubation attempted by emergency medicine (nonanesthesia) residents. The second laryngoscopy was performed by the anesthesia resident. Both of these patients had their third laryngoscopy performed by the attending anesthesiologist, during which their tracheas were intubated. Although the time taken to secure the airways for these patients was the longest, the data from these two patients were included in the analysis on an intention-to-treat basis.

There were no differences in demographics between the two groups (Table 1). The two groups also appeared to be well balanced with respect to the standard treatment variables (Table 2). The arterial oxygenation saturation as measured by Spo2 preoperatively, after administration of oxygen and during the process of laryngoscopy and tracheal intubation was comparable in the two groups. Two patients in the blanket group and four patients in the table ramp group had a Spo2 reading of 95% or less during the process of laryngoscopy and intubation.

Table 1
Table 1:
Patient Variables
Table 2
Table 2:
Treatment Variables

The mean intubation time was 175 (66) s in the blanket group and 163 (71) s in the table-ramp group. Forty of 43, (93.0%) and 39 of 42 (92.8%) patients underwent successful tracheal intubation on the first attempt in the blanket group and the table-ramp group, respectively. Using the data from all 85 patients, the 95% confidence interval (corresponding to an α = 0.05 two one-sided tests) for equivalence is −36, 13 s. The two methods are equivalent, provided the bounds for equivalence are −55,55 s.

The number of times laryngoscopy was performed and tracheal intubation attempted in both groups of patients is shown in Table 3. The difference between the groups was not significant: P = 0.21 for laryngoscopy and P = 0.76 for tracheal intubation, respectively. The time to tracheal intubation did not increase significantly as BMI increased (Fig. 3).

Table 3
Table 3:
Outcome Variables
Figure 3.
Figure 3.:
The time taken to secure the airway did not increase with Body Mass Index (BMI). The bold line represents Linear Regression of time taken to secure the airway against BMI. The dashed lines represent the 95% confidence interval about the regression line.


Failure to properly position the patient before induction is a common mistake, especially among trainees. Although other techniques can be adopted to compensate for lack of proper positioning, the margin for error is smaller in obese patients.

While the ASA difficult airway algorithm suggests that preplanned strategies lead to improved outcome and it recommends a preformulated strategy for intubation of the difficult airway, it is silent with respect to proper patient positioning before tracheal intubation.16 As anesthesiologists, our experience from regional anesthesia is that time taken for proper positioning is time well spent in improving the chance of achieving a successful block.

While the table-ramp head-elevated position for obese patients may not be new, what is important is the fact that the optimal position should be achieved before intubation, regardless of the means to achieve this position. It may indeed be easier, for both the patient and the anesthesiologist, to achieve the ideal position by maneuvering the OR table, rather than inserting wedges and blankets under the patient's torso.

The head-elevated laryngoscopic position, as part of a preplanned strategy, along with administration of oxygen (denitrogenation) could be useful in delaying arterial desaturation as it would allow better conditions for both spontaneous and mask ventilation in an obese patient. The same head-elevated position can be used after surgery while weaning the patient off the ventilator during the extubation process. This is done more easily using the electronic controls of the OR table to recreate a table ramp during emergence instead of reinserting blankets under the patient.

Surgeons in our hospital prefer not to have blankets under the anesthetized patient during surgery. As a larger portion of the patient's back is in contact with the operating table surface when blankets are not used to position patients, the likelihood of injury to the patient's skin is minimized when blankets are removed after tracheal intubation. However, injury to OR personnel may occur when an attempt is made to lift or move patients so that the blankets can be removed from under the patients after tracheal intubation. Use of the electronic controls of the OR table to position patients avoids these problems.

The pilot study was performed to obtain an estimate of the variability to properly power a full study and to demonstrate equivalence of the two methods. The zone of equivalence with regard to the time required to secure the airway was based on the following clinical criteria. In theory, patients' lungs could be ventilated manually for a long time without harm by effective mask ventilation with laryngoscopy attempted intermittently provided arterial oxygen and CO2 are maintained within normal limits. However, one must be cautious with this approach, as more than two attempts at laryngoscopy could lead to harm.17 Incorporating the recommendations of the Difficult Airway Society algorithm18 our anesthesia department recently adopted a local practice advisory of no more than three attempts at tracheal intubation before calling for help in daily clinical practice. In this context, we chose to define the equivalence bounds for time to intubation to be <60 s for our study and limit the number of attempts at laryngoscopy to three.

In the previously reported trials, comparing the supine position to the ramped-up position using blankets, all laryngoscopies and intubations were performed by one anesthesiologist, the author, who had several years of experience.5,7 In another study, where the 25-degree head-up position was compared to the supine position, all laryngoscopies were performed by one laryngoscopist.6 In general, the tendency to excessively narrow the conditions and hypotheses of a clinical trial to ensure the validity of its results has been criticized.19 Our intention was to conduct a study whose results would be broadly applicable.

We chose the time from loss of consciousness to the time CO2 is detected after placing a tracheal tube as our primary outcome variable. This is the period during which an unconscious patient with an unprotected airway undergoes mask ventilation and laryngoscopy and is at risk for hypoxemia, hypoventilation, gastric insufflation, and pulmonary aspiration. The period we chose is much longer than those used in other studies, where the time from beginning of laryngoscopy to insertion of the tracheal tube was measured. Also in our study, as laryngoscopy and tracheal intubation were performed by trainees with various levels of training and expertise (residents, CRNAs, and attending anesthesiologists), the times recorded for securing the airway show a wide range and large variance. We presume that is the main reason why our standard deviation is much larger than that reported in the other studies.

In our study, where the treatment differences between the groups are about the same size as patient-to-patient variability, a rigorous trial (via a randomized design) is the only reliable way to separate treatment effect from noise. We understand that trials are difficult to apply, when studying treatments that depend on the proficiency of the practitioner rather than drugs, as it is harder to isolate the therapeutic component of interest from the proficiency of the provider.20

As the proportion of obese patients has been steadily increasing,21 the method of positioning such patients becomes increasingly important. Traditional teaching has emphasized the supine sniffing position. Similarly in obese patients, the head-elevated position achieved by the simple maneuver of elevating the back section of the OR table should be used, providing a greater margin for safety, especially with trainees.

In conclusion, proper positioning before laryngoscopy and tracheal intubation is important. Our data suggest that the head-elevated laryngoscopic position achieved by using blankets under a patient's head and shoulders or by configuring the OR table into a back-up position are equivalent with regard to the quality of laryngeal exposure and time required to achieve tracheal intubation. On the basis of our results, we propose that positioning patients in the head-elevated position by elevating the back or trunk section of the OR table can be considered by clinicians as part of their preformulated strategy in their daily clinical practice in managing the airways of obese patients.


The rate of difficult intubation in the normal population is 6.2% (95% CI 4.6%–8.3%) and 15.8% (95% CI 14.3%–17.5%) in obese patients. Assuming the failure to intubate a patient is not related to the randomized treatment, if more than eight intubation failures occur in 85 patients, the exact binomial upper 95% confidence limit will be above 0.18.20 That is, in order to have intubation failures have maximum threshold of 20%, then eight or fewer failures must occur given 85 patients, or the trial would be stopped.


1. Bannister FB, Macbeth RG. Direct laryngoscopy and tracheal intubation. Lancet 1944;244, 6325:651–4
2. LinksJuvin P, Lavaut E, Dupont H, Lefevre P, Demetriou M, Dumoulin JL, Desmonts JM. Difficult tracheal intubation is more common in obese than in lean patients. Anesth Analg 2003;97:595–600
3. 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; 103:429–37
4. Levitan RM, Ochroch AE. Airway management and direct laryngoscopy: a review and update. Crit Care Clin North Am 2000;16:373–88
5. Brodsky JB, Lemmens HJ, Brock-Utne JG, Vierra M, Saidman LJ. Morbid obesity and tracheal intubation. Anesth Analg 2002; 94:732–6
6. Lee BJ, Kang JM, Kim DO. Laryngeal exposure during laryngoscopy is better in the 25 degrees back-up position than in the supine position. Br J Anaesth 2007;99:581–6
7. Collins JS, Lemmens HJ, Brodsky JB, Brock-Utne JG, Levitan RM. Laryngoscopy and morbid obesity: a comparison of the “sniff' and “ramped” positions. Obes Surg 2004;14:1–5
8. Dixon BJ, Dixon JB, Carden JR, Burn AJ, Schachter LM, Playfair JM, Laurie CP, O'Brien PE. Preoxygenation is more effective in the 25 degrees head-up position than in the supine position in severely obese patients: a randomized controlled study. Anesthesiology 2005;102:1110–5
9. Nissen MD, Gayes JM. An inflatable, multichambered upper body support for the placement of the obese patient in the head-elevated laryngoscopy position. Anesth Analg 2007;104: 1305–6
10. Wong AB, Moore MS. Positioning of obese patients in out-of-operating room locations Anesth Analg 2007;104:1306
11. Olson EJ, Moore WR, Morgenthaler TI, Gay PC, Staats BA. Obstructive sleep apnea-hypopnea syndrome. Mayo Clin Proc 2003;78:1545–52
12. Kushida CA, Littner MR, Morgenthaler T, Alessi CA, Bailey D, Coleman J Jr, Friedman L, Hirshkowitz M, Kapen S, Kramer M, Lee-Chiong T, Loube DL, Owens J, Pancer JP, Wise M. Practice parameters for the indications for polysomnography and related procedures: an update for 2005. Sleep 2005;28:499–521
13. Jacobsen J, Jensen E, Waldau T, Poulsen TD. Preoperative evaluation of intubation conditions in patients scheduled for elective surgery. Acta Anaesthesiol Scand 1996;40:421–4
14. Karkouti K, Rose DK, Wigglesworth D, Cohen MM. Predicting difficult intubation: a multivariable analysis. Can J Anaesth 2000;47:730–9
15. Wilson ME, Spiegelhalter D, Robertson JA, Lesser P. Predicting difficult intubation. Br J Anaesth 1988;61:211–16
16. Practice guidelines for management of the difficult airway: a report by the Am Society of Anesthesiologists Task Force on Management of the Difficult Airway. Anesthesiology 1993; 78:597–602
17. Kluger MT, Short TG. Aspiration during anesthesia: a review of 133 cases from the Australian Anesthetic Incident Monitoring Study (AIMS). Anesthesia 1999;5491:19–26
18. Difficult Airway Society, Guidelines: Available at: (date last accessed May 15, 2008)
19. Coleman CH, Menikoff JA, Goldner JA, Dubler NN. The ethics and regulation of research with human subjects. Matthew Bender and Company, LexisNexis Group, 2005:259
20. Piantadosi S. Clinical trials: a methodologic perspective. New York: John Wiley & Sons, 1997
21. CDC Behavioral Risk Factor Surveillance System available at: (date last accessed May 15, 2008)
© 2008 International Anesthesia Research Society