Apnoeic Oxygenation during Simulated Difficult Intubation in Obese Patients: Comparison of Buccal Ring, Adair and Elwyn Tube Versus Nasal Cannula: A Prospective Randomized Controlled Trial.

Background: Apnoeic oxygenation is an established method of increasing safe apnoea times during intubation and this is of more importance in obese patients. The usefulness of buccal Ring, Adair and Elwyn (RAE) oxygenation has been established in previous studies, however a head-to-head comparison with nasal cannula (NC) is lacking. Aim: The aim of this study was to compare apnoea time with buccal RAE (BR) versus NC in obese patients. Setting and Design: This was a prospective, nonblinded randomized controlled trial conducted in a tertiary hospital where fifty American Society of Anaesthesiologists Physical Status Class I and II, obese patients with body mass index ≥30, posted for elective surgery were included. Materials and
Methods: Following adequate preoxygenation and standard induction of anaesthesia, a prolonged simulated difficult laryngoscopy was performed during which oxygen was provided via either BR or NC. The primary outcome was time to desaturation to <95% or 10 min, which ever occurred first. Other outcomes recorded were lowest saturation, time to resaturation and highest end tidal carbon di oxide. Statistical Analysis: Mean with standard deviation (SD) or median with inter quartile range were used for continuous variables and absolute number with percentage were used for categorical variables. The primary outcome was analyzed using Kaplan-Meier survival curves, and log-rank tests were applied.
Results: Patient characteristics were similar in both arms. The mean apnoea time in seconds (SD) in the BR group, 375.3 (116.6) was higher than the NC group 316.1 (94.1), P = 0.054. From the Kapan Meier curves the probability of desaturating to <95% was earlier in the NC group than the BR group (P = 0.092). The other outcomes were similar in both groups.
Conclusion: This is the first study that demonstrates that oxygenation via a BR is better than NC in providing apnoeic oxygenation in obese patients and can safely be used when NC are contraindicated. Copyright:

INTRODUCTION

The global prevalence of obesity looms large. In 2016, about 13% of the world's adult population, 11% of men and 15% of women were found to be obese.[1] In India, the prevalence of obesity is estimated to triple by 2040-9.5% and 13.9% among men and women respectively, thus increasing the number of obese patients presenting to the emergency departments, operating rooms and critical care units.[2]

Apnoea time is the time following cessation of breathing, to allow for placement of a definitive airway. It is of significance as once the saturation drops to 90%, further desaturation is rapid. The safe apnoea time in nonobese patients is around 9 min and 2–4 min in the obese.[3] Additionally, obesity is an established risk factor for difficult ventilation and intubation.[4] This makes active oxygenation challenging.

Apnoeic oxygenation is a proven method of oxygen insufflation during the period of apnoea that helps increase the safe apnoea time.[567] The aim of this study was to compare two methods of apnoeic oxygenation, buccal Ring, Adair and Elwyn (RAE) (BR) against nasal cannula (NC), during simulated difficult intubation in obese patients.

MATERIALS AND METHODS

Trial design

This was an open label, parallel arm, prospective randomized controlled trial. The institutional local review board (IRB) and ethical committee approved the trial (IRB Min. No. 11137, Date-January 24, 2018).

Participants

Adult patients (ASA class I and II) body mass index (BMI) >30 scheduled for elective surgery under general anaesthesia were included. Exclusion criteria included those with chronic respiratory disease, uncontrolled hypertension, ischaemic heart disease, congestive heart failure, elevated intracranial pressure, gastro oesophageal reflux disease, saturation <98% after pre oxygenation, Grade 3 or 4 mask ventilation, current or past Grade 3 or 4 laryngoscopy, current or past allergy to the drugs being used.

Setting

The study was conducted in the operating rooms of a tertiary care hospital in South India.

Study period

The study was done over a duration of 4 months.

Intervention

Written informed consent was obtained for participation in the study and use of the patient data for research and educational purposes. The procedure followed the guidelines laid down in Declaration of Helsinki (2013). Fifty patients were randomized into the two arms: BR-apnoeic oxygenation with BR and NC-apnoeic oxygenation with NC. All patients were monitored with standard ASA monitors, bi spectral index (BIS) and a neuromuscular monitor. The RAE tube was prepared by cutting an oral RAE tube size 3.5 mm at the distal end above the Murphy's eye. The blunt end of the tube was placed in the left buccal space and the cut end of the tube was connected to the oxygen source via a universal connector.

Patients were premedicated with injection midazolam 0.02 mg.kg−1. Pre-oxygenation was done keeping patients in 30° reverse Trendelenburg position, with 10 Lmin−1 of oxygen, using a tight-fitting bag with the adjustable pressure limiting valve fully opened, until an end tidal oxygen of more than 90%. Patients were then induced with 2 mg.kg−1 of propofol, 2 μg.kg−1 of fentanyl and muscle relaxation with 1.2 mg kg-1 of rocuronium (after assessing the ability to ventilate). At this point the face mask was removed and apnoeic oxygenation commenced, with 15Lmin-1 through a flow meter via the buccal tube placed at the angle of the mouth for the BR group, and via NC for the NC group. This point was considered time zero (T0). Maintenance of anaesthesia from this point was with a propofol infusion at 150 μg.kg−1 min−1 to maintain a BIS of 40–60. Train of four was used to ensure adequate muscle relaxation. Video laryngoscopy was performed after 1 min of rocuronium administration, the ability to intubate assessed and a grade 3 Cormack and Lehane grade was simulated for 10 min or until the saturation dropped to <95%.

An arterial blood gas was taken at the start and end of the apnoea period to note the partial pressure of carbon dioxide (PaCO2) and partial pressure of oxygen (PaO2). Patients were intubated after 10 min of apnoea or earlier if the saturation dropped to <95% and positive pressure ventilation was initiated. Ventilation via the endotracheal tube was set with tidal volumes of 10 mL.kg−1 at 12 breaths.min−1 and 5 cm H2O of positive end-expiratory pressure. The end-tidal carbon dioxide (EtCO2) value at the start of ventilation was noted, as was the lowest SpO2. Time to desaturation to <95% and time to attain saturation of 100% after intubation and positive pressure ventilation was noted. Haemodynamics were monitored and maintained to 20% of baseline with ephedrine or phenylephrine boluses.

Outcomes

Primary outcome was apnoea time. Apnoea time was defined as the time to reach saturation <95% during the study period of 10 min. Secondary objectives were-(1) the lowest saturation (2) the time to return to 100% saturation after commencement of positive pressure ventilation once the endotracheal tube was secured (3) the highest EtCO2 (4) PaCO2 and PaO2 at the start and end of the apnoea period.

Sample size

Considering a non-inferiority margin of 90 s for BR tube oxygenation from nasal oxygenation with a standard deviation (SD) of 1.5 min, and to detect a difference with 80% power and 5% error, a sample size of 25 patients was calculated for each arm.[3]

Randomization

Computer generated block randomization was used. The random allocation sequence was generated by the statistician prior to commencement of the study. Allocation concealment was by using sequentially numbered opaque envelopes that were opened just prior to the start of the case. The primary investigator enrolled participants and assigned them to the appropriate intervention. Participants, investigators, and the statistician could not be blinded to the intervention.

Statistical methods

The descriptive statistics such as Mean with SD or median with inter quartile range (IQR) were used for continuous variables and absolute number with percentage were used for categorical variables. To compare the continuous variables across the group, independent t-test was used and for categorical data, Chi-square test was used. The primary outcome was analyzed using Kaplan-Meier survival curves, and log-rank tests were applied. Apnea times with Spo2≥95% were reported in each group for comparison. P < 0.05 were considered as the statistically significant variables. All analysis was performed using the software using Statistical Package for Social Sciences for Windows (SPSS Inc. Released 2017, version 23.0, Armonk, New York, USA).

RESULTS

During the study period, 25 patients were randomized to each arm. Participant flow is depicted in the Figure 1.

Figure 1

CONSORT flow diagram

Baseline characteristics with respect to mean age, BMI and sex distribution were similar [Table 1]. However due to the female preponderance the type of obesity was predominantly gynecoid (68%). There were no morbidly obese patients in the study.

Table 1

Demography and clinical characteristics (baseline parameters)

Variable	BR (n=25)	NC (n=25)	P
Age (years), mean (SD)	43.9 (12.3)	44.9 (12.2)	0.774
Females, n (%)	19 (76.0)	23 (92.0)	0.247
Weight (kg), mean (SD)	77.9 (12.4)	82.3 (12.6)	0.223
Height (cm), mean (SD)	156.5 (10.1)	157.2 (9.5)	0.807
BMI (kg/m2), mean (SD)	32.6 (2.4)	33.4 (2.6)	0.237
BMI 30-35, n (%)	20 (80)	20 (80)
BMI >35, n (%)	5 (20)	5 (20)
Obesity type-gynacoid, n (%)	17 (68.0)	17 (68.0)	1.000

BMI: Body mass index, BR: Buccal Ring, Adair and Elwyn, NC: Nasal cannula, SD: Standard deviation

The mean apnoea time (SD) in the BR group was higher than the NC group, 375.3 s (116.6) and 316.1 s (94.1) respectively with a p value of 0.054. The minimum saturation, resaturation time, maximum EtCO2, PaCO2 and PaO2 at the end of apnoea time was similar between groups [Table 2].

Table 2

Comparison of apnoea time and physiological parameters

Parameters	Group (n=25)		P
	BR	NC
Baseline SpO2 (%)	98.8	98.5	0.36
Start of apnoea time PaO2 (mmHg)	328 (97.8)	258 (87.2)	0.01
End of apnoea time PaO2 (mmHg)	80.5 (40.1)	69.2 (10.3)	0.19
Apnoea time (s)	375.3 (116.6)	316.1 (94.1)	0.05
Minimum SpO2 (%)	90.6 (3.1)	90.3 (2.6)	0.69
Resaturation time (s)	126.7 (53.7)	148.3 (54.7)	0.17
Maximum EtCO2	45.9 (6.2)	44.1 (5.9)	0.29
End PaCO2 (mmHg)	55.6 (6.4)	52.5 (6.8)	0.11
End PaO2 (mmHg)	80.5 (40.1)	69.2 (10.3)	0.19

Results are mean (SD). BR: Buccal Ring, Adair and Elwyn, NC: Nasal cannula, SD: Standard deviation, SpO2: Saturation, PaO2: Partial pressure of oxygen, PaCO2: Partial pressure of carbon di oxide, EtCO2: End tidal carbon di oxide

The comparisons of oxygen saturation profile during the study period for each patient is highlighted in Figure 2. This visually demonstrates early desaturation in the NC group. There were three patients in the BR group and one patient in the NC group who completed the entire 10 min of study duration (P = 0.609). Six patients in the BR group and seven patients in the NC group desaturated to <90% by the end of intubation (P = 0.675). No discomfort, harm or unintended effects were seen in any patients.

Figure 2

Arterial oxygen saturation profiles during study period including lowest recorded value after study termination for each patient in both groups

Figure 3 further illustrates the comparison of apnoea times between the two groups depicted in the Kaplan Meier graph. The graph demonstrates the probability of desaturating to <95% was earlier in the NC group than the BR group (P = 0.092).

Figure 3

Kaplan Meir graph-(Event– Saturation ≤95%). Censoring includes subjects who are nonevents

DISCUSSION

In our study, we found that the mean safe apnoea time in the BR group was 375.3 (116.6) and NC group was 316.1 (94.1) (P = 0.054). In this noninferiority trial, the results demonstrate moderately significant, longer apnea times in the BR group compared to the NC group (59 s).

The mean apnoea time attained in the NC group was comparable to Ramachandra et al., where apnoeic oxygenation through NC had a mean safe apnoea time of 317.4 s (61.2) verses 209.4 (79.8) in the control arm, in the obese population.[8]

Heard et al. demonstrated a median (IQR) safe apnoea time of 750 (389–750) with the BR tube and 296 (244–314) seconds (p < 0.001) in the control arm in an obese population.[3] A similar study in the non-obese by Toner et al. also obtained a median (IQR) apnoea time with SpO2 >94%; of 750 s (750–750) with buccal apnoeic oxygenation, verses 444 (402–522); (P < 0.001) in the sham oxygenation arm.[9] The mean apnoea time we obtained with buccal oxygenation is approximately half of what Heard and Toner demonstrated. This difference could be explained by a difference in conduct of anaesthesia; these studies used target controlled infusion (TCI) and time zero began with commencement of TCI, (at loss of verbal response) when respiratory efforts may have still been present. In our study, time zero commenced just after the muscle relaxant was given and patients were truly apnoeic. Accordingly, our time zero commenced later than theirs.

We noticed that the mean (SD) PaO2 in mm Hg at the start of apnoea time was statistically different in both arms, 328 (97.8) in the BR verses 258 (87.2) in the NC arm (P = 0.01). This can be explained by the increased leak in the NC group during preoxygenation compared to the BR group. Despite a tight-fitting mask, there is an inadvertent leak bilaterally by the sides of the NC and unilaterally with the BR. It is possible that could have contributed to the longer apnoea time in the BR group.

The minimum SpO2 mean (SD) in both the arms was around 90% (2.6), as opposed to 94% (4.4) demonstrated by Ramachandran et al. using nasal prongs which is likely to be due to a shorter study duration-6 min. Heard et al. obtained a median (IQR) minimum saturation of 97% (92–99). The mean resaturation time in our study was longer than the 49 s reported by Ramachandran et al., probably due to the shorter study duration of 6 min in their study.[8] The other secondary end points such as highest EtCO2 and end PaCO2 were comparable to other studies.[389]

Isono and Ishikawa stated succinctly that the final goal of airway management is the maintenance of oxygenation and rapid recovery from hypoxemia.[10] Various clinical studies have evaluated the efficacy of apnoeic oxygenation during the induction of anaesthesia using the NC, nasopharyngeal (NP) catheter, modified laryngoscopes, buccal oxygenation and high flow NC (HFNC). There is no consensus over the superiority of one over the other for routine use, considering cost effectiveness and availability. The NP catheter has been shown to yield excellent oxygenation by maintaining high supra laryngeal oxygen delivery but poses a considerable risk of barotrauma if the oxygen flows do not vent through the mouth or nose.[11] Modified laryngoscopes with a 14 Fr catheter attached to the side of the blade have been used for apnoeic oxygenation with similar benefits and risks compared to the NP catheter, but with an added disadvantage of commencement of oxygenation only after insertion of the laryngoscope.[1213] The NC was found to have moderate gains in apnoea time but its efficacy may be limited in obese patients by inadequate velo pharyngeal patency under anaesthesia.[14] In contrast HFNC generates CPAP by delivering flows up to 70 L. min−1 and thereby in the process overcomes velopharyngeal obstruction and augments FRC.[15] Disadvantages posed by HFNC are that the pressure it generates is highly dependent on mouth closing, often it is an intermittent technique as the equipment needs to be removed if bag and mask ventilation is necessitated. Pharyngeal pressures can rise when the oral and nasal escape routes are attenuated. In the critically ill patients, it has not proven to be efficacious, further it is expensive and not readily available.[16]

Heard et al. introduced a novel method of apnoeic oxygenation using a modified RAE tube to provide buccal oxygenation and found that 65% of patients in this arm maintained their SpO2 above 95% for the duration of apnoea period, which was 12.5 min, one of the longest study durations undertaken in the obese population.[3] Toner et al. furthered this study looking at tracheal oxygen concentrations, tracheal pressure and CO2 accumulation during apnoeic oxygenation and concluded that buccal oxygen delivery with a patent airway was safe without untoward CO2 retention or barotrauma and maintained tracheal oxygen concentrations of >90%.[9]

With no studies comparing nasal and buccal oxygenation we conducted this study. The following points are the strengths in our study. This study was an RCT with no dropouts. The primary investigator was present for each case. The limitations include non-blinded and a single centre study. This was a noninferiority study with a sample size of 50, and a higher sample size may be warranted to corroborate findings. The BR costs marginally more and may not be as freely available as the NC. Blinding was not practically possible as the interventions are visibly different. A good grade of laryngoscopy did not always ensure easy and quick intubation, leading to low minimum saturations. This may have been because ramping which is known to improve the grade of laryngoscopy was not part of the study.[4] Though available literature advocates the use of PEEP during preoxygenation for the obese, our study did not include this modality.[4]

CONCLUSION

Buccal oxygenation with an oral 3.5 mm RAE tube is an effective method of apnoeic oxygenation and gives safe apnoea times higher than the NC. It is a reliable alternative and can be used in situations where nasal oxygenation is contraindicated or not feasible.

Financial support and sponsorship

Fluid Research Grant, Christian Medical College, Vellore.

Conflicts of interest

There are no conflicts of interest.