Atomised lidocaine for airway topical anaesthesia in the morbidly obese: 1% compared with 2%

Summary

Airway anaesthesia using atomised lidocaine for awake oral fibreoptic intubation in morbidly obese patients was evaluated using two doses of local anaesthetic. In this randomised, blinded prospective study, 40 ml of atomised 1% (n = 11) or 2% (n = 10) lidocaine was administered with high oxygen flow as carrier. Outcomes included time for intubation, patient tolerance to airway manipulation, haemodynamic parameters, the bronchoscopist’s overall satisfaction, and serial serum lidocaine concentrations. Patients receiving lidocaine 1% had a longer mean (SD) time from the start of topicalisation to tracheal tube cuff inflation than those receiving lidocaine 2% (8.6 (0.9) min vs 6.9 (0.5) min, respectively; p < 0.05). Patients in the 1% cohort demonstrated increased responses to airway manipulation (p < 0.0001), reflecting lower bronchoscopist’s satisfaction scores (p < 0.03). Haemodynamic responses to topicalisation and airway manipulation were similar in both groups. Peak plasma concentration was lower in the 1% group (mean (SD) 1.4 (0.3) and 3.8 (0.5) μg.ml⁻¹, respectively; p < 0.001). Airway anaesthesia using atomised lidocaine for awake oral fibreoptic intubation in the morbidly obese is efficacious, rapid and safe. Compared with lidocaine 1%, the 2% dose provides superior intubating conditions.

Safe airway management in morbidly obese patients requiring general anaesthesia is of paramount importance and inadequate preparation may have dire consequences. At induction of anaesthesia, these patients can experience rapid arterial oxygen desaturation [1] and aspiration [2], and difficulties may be encountered at all levels of airway management including ventilation and intubation [3–6]. Securing the airway using awake fibreoptic intubation is considered to be a safe strategy [7], but this technique requires superior airway anaesthesia if excessive sedation is to be avoided. In the morbidly obese patient, obscured landmarks may limit the usefulness of invasive techniques (such as nerve blocks) to achieve airway anaesthesia [8]. Recently, we demonstrated that atomised lidocaine 2% or 4% is an efficacious, rapid and safe method to achieve topical airway anaesthesia in morbidly obese patients. However, the lowest optimum dose was not established [9]. This study was undertaken to compare atomised lidocaine 1% and 2% for airway anaesthesia during awake fibreoptic intubation in morbidly obese patients undergoing bariatric surgery. In a prospective, blinded and randomised investigation, main outcome measures included procedural time, patient tolerance to airway manipulation, haemodynamic parameters and serum lidocaine concentrations. The bronchoscopist’s satisfaction level with intubating conditions and patients’ tolerance to airway topicalisation were also assessed.

Methods

Following institutional ethics board approval, morbidly obese patients (BMI > 50 kg.m⁻²) undergoing gastric bypass surgery were approached for inclusion into our study if the pre-operative airway assessment indicated awake intubation. Exclusion criteria included known sensitivity to local anaesthetics and significant hepatic or renal insufficiency. After obtaining written informed consent, patients were randomly assigned (random numbers table) to receive 40 ml of either 1% (400 mg) or 2% (800 mg) lidocaine for airway topicalisation. These doses were based on our previous study [9] showing that superior airway anaesthesia permitting awake oral fibreoptic intubation was achieved with 40 ml of atomised lidocaine 2%. Both the bronchoscopist and an objective observer were blinded to the study medication. Patients were carefully positioned supine on the operating table using the Troop Elevation Pillow and Head Cradle^® (Mercury Medical, Clearwater, FL, USA) to achieve a standardised ideal ‘head elevated’ position. While oxygen was administered by facemask, standard monitors were applied including ECG, pulse oximeter, and non-invasive blood pressure monitor. An 18-G cannula was inserted into an upper limb vein for drug and fluid administration, and a 20-G cannula was inserted into a radial artery for arterial blood pressure monitoring and to obtain serial blood samples for determination of plasma lidocaine levels. Oral sodium citrate 30 ml and iv metoclopramide 10 mg and ondansetron 4 mg were given for aspiration prophylaxis, with glycopyrronium 0.3 mg iv for its anti-sialogogue effect. Midazolam (1–2 mg) and fentanyl (100–150 μg) were titrated iv to produce an awake, yet calm and co-operative patient, as previously described [9]. Using high-flow oxygen (10 l.min⁻¹), the study drug was administered via a glass atomiser with an adjustable tip (DV-15-RD; Sunrise Medical, Montreal, QC, Canada) to the oropharynx, with the patient breathing deeply through the mouth, and with the nares clamped. Atomisation was performed during inspiration and expiration, with the mist directed more posteriorly during inspiration and anterior to the soft and hard palate and tongue during expiration. Bronchoscopy and intubation then proceeded through an oral guide (Ovassapian airway). Following confirmation of correct tracheal tube placement, general anaesthesia was induced via inhalation of sevoflurane, followed by muscle paralysis with rocuronium.

The primary outcome measures were patients’ tolerance, heart rate and blood pressure responses to bronchoscopy and intubation, and serial plasma lidocaine levels. Patients’ tolerance to airway manipulation was assessed using a 3 point scale (0 = no response, 1 = mild gagging (gag response easily managed by gentle reassurance, not warranting additional sedation, and able to continue with the procedure), 2 = intolerable) by an independent observer during insertion of the oral guide, insertion of the bronchoscope into the oropharynx, advancement of the bronchoscope through the glottis, and advancement of the tracheal tube through the glottis as in our previous study [9]. Blood pressure and heart rate were recorded at these same points and during the process of airway topicalisation. Serial arterial blood samples were obtained for analysis of plasma lidocaine concentrations at baseline pre-atomisation and 5, 10, 15, 20, 25, 30, 45, 60 and 120 min after the conclusion of the atomisation. Blood samples were immediately spun and the plasma portion stored at −20 °C. A gas chromatograph mass spectrometry method was used for lidocaine concentration analysis (sensitivity 0.2 μg.ml⁻¹). The time for airway topicalisation and total time for airway manipulation (beginning of atomisation to tracheal tube cuff inflation) were recorded. A 1–5 point score was given for the bronchoscopist’s overall satisfaction with the bronchoscopy and intubating conditions (1 = worst, 5 = best). A 1–3 point score was given for the ability of the patient to tolerate airway topicalisation (1 = worst, 3 = best). The same individual (CW) performed airway topicalisation, bronchoscopy and intubation in all subjects while another individual (SBB) rated patient tolerance to airway manipulation.

Student’s t-test was used to compare the 1% and 2% group with respect to drug doses and the time required for airway manipulation. Spearman rank correlation and Mann–Whitney tests were used to assess bronchoscopists’ satisfaction with intubating conditions and patients’ tolerance to topicalisation scores. Patients’ responses to airway manipulation were assessed with the Fisher’s exact and Chi-squared tests. Repeated measures ANOVA and the Tukey–Kramer multiple comparisons tests were used to assess haemodynamic responses and lidocaine plasma levels. The results were deemed significant for p values < 0.05. Assuming a 0.85 proportion of ‘no response’ at the defined points during awake fibreoptic intubation using 2% lidocaine for airway topicalisation (based on previous clinical experience [9]), the minimum number of patients required for an alpha level of 0.05 (two-tailed) and a power of 0.8 to detect a 30% difference between the groups is 35 in each group. instat^® and graphpad prism^® (GraphPad Software Inc, San Diego, CA, USA) were used for data analysis and plots.

Results

There were 21 patients recruited (lidocaine 1% group n = 11, lidocaine 2% group n = 10) when the blinding was broken for interim analysis because of the clinical impression of decreased tolerance to airway manipulation in some patients. Patients’ characteristics were similar between the two groups (Table 1). The midazolam and fentanyl doses used in the two groups were indistinguishable (Table 1). The mean time for airway topicalisation (all patients) was 4.7 (0.7) min. The total time of airway manipulation (start of airway topicalisation until cuff inflation) was longer in the 1% group (8.6 (0.9) than in the 2% group 6.9 (0.5) min, respectively; p < 0.05). Patients in the 2% group had lower response scores to airway manipulation compared with those in the 1% group, with 80% of the responses graded as no response compared with only 39% of the 1% group (p = 0.0002). When the responses were analysed by rank (no response, mild gagging or intolerable) a difference between the groups was detected (p < 0.0001). No patient in the 2% group responded to placement of the oral guide (Fig. 1), and mild gagging was observed as the bronchoscope was manipulated through the oropharynx (n = 3) and glottis (n = 2). Insertion of the tracheal tube into the trachea produced mild gagging in three patients (Fig. 1). The median (IQR) bronchoscopists’ satisfaction scores with airway management in this group were 5 (4–5). In the 1% group there was mild gagging to oral guide insertion (n = 4), insertion of the bronchoscope through the oropharynx (n = 7) and glottis (n = 6), and insertion of the tracheal tube into the trachea (n = 6) (Fig. 1). In the 1% group, intolerance to insertion of the oral guide (n = 1) and bronchoscope advancement through the oropharynx (n = 2) and glottis (n = 1) was observed (Fig. 1), mandating additional sedation for fibreoptic intubation to succeed (no supplemental lidocaine was administered to any patients, and all patients’ tracheas were successfully intubated). The bronchoscopist’s satisfaction scores were lower (median (IQR) 4 (3–4) than in the 2% group; p < 0.03). The patients’ tolerance scores to airway topicalisation using 1% and 2% lidocaine were similar (2 (2–3) and 3 (3–3), respectively) and there was no correlation between the patients’ tolerance to topicalisation and the bronchoscopists’ satisfaction scores in either group.

Table 1.   Patients’ characteristics and doses of sedative drugs used during airway manipulation after topicalisation with 1% or 2% lidocaine. Data are number or mean (SD).

	Lidocaine 1% (n = 11)	Lidocaine 2% (n = 10)
Male/female	5/6	3/7
Age; years	41.4 (9.7)	45.9 (11.4)
Height; cm	168 (12.1)	165 (10.5)
Weight; kg	164 (10.1)	158 (11.7)
BMI; kg.m^−2	56.3 (6.7)	54.7 (8.5)
Midazolam; mg	1.8 (0.4)	1.9 (0.6)
Fentanyl; μg	145 (41)	140 (39)

Figure 1.  Patient tolerance to stages of airway manipulation after topicalisation with 1% or 2% lidocaine. TT = tracheal tube.

Haemodynamic parameters were not different between the two groups (Fig. 2), but increases in heart rate and blood pressures were recorded during airway management in each cohort. Peak plasma lidocaine concentrations were greater in the 2% group than the 1% group (mean (SD) 3.8 (0.5) μg.ml⁻¹ vs 1.4 (0.3) μg.ml⁻¹; p < 0.001) and occurred at 5 min following the end of the atomistion (Fig. 3). There were no signs of lidocaine toxicity before induction, during anaesthesia, or in the recovery period. All patients’ tracheas were extubated in the operating room at the completion of surgery and transferred to the post-anaesthesia care unit in a stable condition.

Figure 2.  Mean (SD) in heart rate (upper panel) and mean arterial pressure (lower panel) following atomisation with lidocaine 1% or 2%, and airway manipulation. Mean (SD); **p < 0.01, ***p < 0.001, compared with baseline within each group. TT = tracheal tube.

Figure 3.  Mean (SD) plasma lidocaine concentrations before (baseline, T = 0) and after airway topicalisation with 1% or 2% lidocaine. *p < 0.001, ^#p < 0.05, 1% compared with 2% group.

Discussion

This study confirms our previous report that topicalisation of the airway using atomised lidocaine 2% provides superior airway anaesthesia to 1% for awake fibreoptic intubation in the morbidly obese patient [9]. This technique has particular merit because it is effective, rapid and safe, and requires minimal patient sedation to achieve excellent intubating conditions. The present study extends our previous work [9] because it illustrates that airway anaesthesia with atomised lidocaine 1% provides measurably less satisfaction with intubating conditions from the bronchoscopist’s perspective, based on the increased patients’ responsiveness to airway manipulation during the intubation procedure. This confirms our impressions during the study that triggered an interim analysis. We speculate that the lighter level of airway anaesthesia achieved with lidocaine 1% may have contributed to the increased heart rate and blood pressure trends during airway manipulation compared with the 2% group. We emphasise that by study design, relatively light levels of sedation were produced with small titrated doses of midazolam and fentanyl such that patients were alert, co-operative and able to maintain spontaneous respiration. As there is an inverse relationship between the level of sedation and the required depth of airway insensibility achieved with application of local anaesthetic, topicalisation with lidocaine 1% may be acceptable particularly if an increased level of sedation is targeted. An improvement of the present report compared with our previous work includes the standardisation of head positioning, achieved with the Troop Elevation Pillow and Head Cradle, thus eliminating an important confounding variable that may interfere with overall airway management.

We have suggested that in some clinical situations atomisation is a superior method of achieving dense airway anaesthesia compared with other techniques such as nebulisation and nerve blocks [9]. Nebulisation may result in untimely administration of large doses of local anaesthetic and may produce unpredictable anaesthesia depending on the amount of drug lost to the atmosphere [8, 10]. Invasive techniques (such as nerve blocks or translaryngeal injection) may be difficult to perform in some patients – particularly the morbidly obese, because of obscured landmarks. With atomisation, a more predictable dose of local anaesthetic may be delivered to the patient because the high-powered mist is directed around the patient’s mouth. As drug administration is provided throughout the respiratory cycle, this may be particularly important in the early phase of inspiration where peak flow is greatest. Techniques that provide intermittent drug administration, such as metered-dose sprays, may not take advantage of this. In addition, this technique permits the operator to spray targeted areas of the oropharynx (for example, the tongue, hard and soft palate, pharynx) in a controlled manner that is coincident with the patient’s behaviour, and to ensure that high-flow drug administration continues particularly during deep inspiration (such as following an episode of forced expiration, or coughing).

A major advantage of atomisation of local anaesthetic is the rapidity at which the airway can be anaesthetised. In this study airway anaesthesia was achieved in less than 5 min. The comparatively lighter degree of airway anaesthesia achieved with lidocaine 1% did not lead to coughing or breath holding, and this is also reflected in the similar patient tolerance scores to topicalisation in both groups. Although comparative information is lacking in the morbidly obese, topicalisation via nebulisation in normally sized patients requires more than 10 min [8, 11]. In a recent paper using the ‘spray-as-you-go’ technique for wake intubation (2–4% lidocaine), topicalisation took more than 20 min [12]. Notwithstanding that adequate non-invasive airway anaesthesia can be achieved with a variety of methods, a prolonged time for topicalisation may be unacceptable in certain critical situations. The rapidity of the atomisation technique, coupled with the continuous administration of high-flow oxygen as carrier that facilitates oxygenation, helps to ensure safe and timely airway management in a high-risk population such as the morbidly obese.

In both the 1% and 2% groups, the total dose of local anaesthetic administered exceeded that normally recommended for lidocaine (5 mg.kg⁻¹) but there were no signs of clinical toxicity, as was the case with our previous study [9]. Importantly, with both groups, the plasma concentration did not reach the accepted toxic plasma level of 5–10 μg.ml⁻¹. Similarly, to achieve non-invasive airway topicalisation, larger than recommended doses of lidocaine have been administered via gel, nebulisation, or ‘spray-as-you’ go techniques without adverse consequences or excessive blood concentrations [12–18]. With regard to this study, the discrepancy between the relatively large doses of administered lidocaine and low plasma concentration may be explained by several factors. Drug may be lost to atmosphere (particularly during expiration) or swallowed. Furthermore, the plasma volume of morbidly obese patients may be increased [19] but this is difficult to predict. Using lean body mass instead of total body weight or ideal body weight is considered a reasonable starting point to calculate plasma volume. However, large inter-patient variability probably exists regarding the ratio of fat to lean tissue mass and plasma volume that constitutes the excess weight of the patient [20, 21]. Finally, in the morbidly obese there may be altered absorption of local anaesthetic from the airway mucosa.

In a previous study, we suggested that the ‘art’ of airway topicalisation using the atomiser technique may influence the amount of drug absorbed in an unpredictable manner [9]. This includes the success in achieving continuous deep inspiration for several minutes, maintaining clamping of the nares, and the ‘painting technique’ of oropharyngeal structures with lidocaine spray throughout the respiratory cycle. We acknowledge that the peak plasma lidocaine concentration may have occurred before the 5 min sampling mark; an issue we considered unlikely on the basis of previously published data [13–16]. With these considerations in mind we recognise that in certain situations (such as decreased lean body mass, hepatic or renal disease), airway topicalisation by whichever means has the potential to achieve toxic plasma concentrations of local anaesthetic drug. Hypercarbia can decrease the seizure threshold for local anaesthetic plasma concentration [22] and so particular care should be exercised when embarking on airway topicalisation in patients with respiratory distress. It is, of course, naïve to think that any one technique of airway anaesthesia is ideally suited to every clinical situation [23], and each must be evaluated on its strengths and weaknesses in the context of the patient and the procedure.

A limitation of this study is the relatively small numbers of patients studied in each group. Differences between groups, such as the haemodynamic responses to airway manipulation, may not have reached statistical significance (that is, a type II error). Another limitation is the subjective nature of the scoring methods we used (patients’ tolerance to topicalisation and airway manipulation, bronchoscopists’ satisfaction). We believe, however, that there is a clear difference in the 1% and 2% groups with regard to their responses to airway manipulation. This is not only derived from our results, but by the extensive experience we have with using atomised lidocaine 2% for airway topicalisation and the uncharacteristically light airway anaesthesia we achieved with a subset of patients in this study.

In summary, we have shown that airway topicalisation in the morbidly obese using atomised lidocaine 2% is a rapid, efficacious and safe method to achieve airway anaesthesia. Topicalisation with atomised lidocaine 1% provided measurably inferior airway anaesthesia.