Inhaled Albuterol, but Not Intravenous Lidocaine, Protects Against Intubation-induced Bronchoconstriction in Asthma

The study was approved by the hospital’s ethics and research committee. One hundred tenpatients scheduled for elective surgery requiring general anesthesia and tracheal intubation were recruited over an 8-yr period. Informed consent was obtained. All patients had a diagnosis of asthma by their primary care physician for at least 1 yr, and all had been treated for reactive airways disease with inhaler therapy in the month before surgery. Patients with significant cardiac disease or those requiring awake or fiberoptic intubation were excluded. Patients were instructed not to take any of their regular asthma medicines on the day of surgery. Patients were without asthma medicine for at least 10 h. Patients were given routine spirometry (unless they had had spirometry within 3 months of surgery).

Our study tested the ability of two drugs to protect against intubation-induced bronchoconstriction: part 1 tested the efficacy of intravenous lidocaine, and part 2 tested the efficacy of inhaled albuterol. Part 2 was performed after the completion of part 1. Each part was a randomized, double-blind, placebo-controlled trial. Study drug or placebo was dispensed by the pharmacy and administered to the patient at the predetermined time.

To assess the protective effect of drugs and placebos, we measured lower pulmonary resistance (R^L). Because of the relatively high and variable resistance of the upper airway, we could not make meaningful comparisons of resistance measured before and after intubation. We therefore had to infer the occurrence of intubation-induced bronchoconstriction from measurements made entirely after intubation, i.e. , after the event we were detecting. We reasoned that significant bronchoconstriction caused by intubation would result in a relatively high resistance immediately after intubation, and that such acutely increased resistance might diminish with time and with the inhalation of isoflurane, an anesthetic agent known to be a brochodilator. 10,14,15Therefore, patients with high pulmonary resistance (R^L≥ 5 cm H²O · l⁻¹· s⁻¹) before administration of isoflurane, whose resistance subsequently decreased by 50% or more while breathing isoflurane, were deemed to have “responded” to intubation with bronchoconstriction. Patients treated with a protective drug would be less likely than those treated with placebo to show a response.

Transpulmonary pressure was measured with a differential transducer (model LCVR; Celesco, Canoga Park, CA) connected between an esophageal balloon catheter and a tracheal catheter positioned at the tip of the endotracheal tube (Portex Jet Ventilator Adaptor; SIMS Portex Ltd., Hythe, UK). The esophageal balloon was 12 cm long, 3 cm in perimeter, and mounted on a polyethylene catheter (2.80-mm OD, 1.77-mm ID). Airflow was measured with a pneumotachometer (Fleisch No. 1; Rusch, Inc., Duluth, GA) and pressure transducer. The flow signal was integrated electrically to indicate volume and calibrated with a 3-l syringe. At prescribed times after intubation, pressure, flow, and volume signals were displayed as pressure–volume and flow–volume traces on a storage oscilloscope, photographed, and later analyzed by the method of Neergard and Wirtz. 3,10,20We calculated R^L(excluding the upper airway) as the pressure difference between inspiration and expiration at midtidal volume divided by the corresponding flow difference, as previously described. 3,10,20Traces were very consistent from breath to breath, and we analyzed one representative breath at each time point.

In part 1, intravenous lidocaine (1.5 mg/kg) or saline was given as a bolus dose 2.5–3 min before planned intubation. In part 2, albuterol or albuterol–placebo was administered in four puffs from a metered dose inhaler, 15–20 min before planned intubation. All patients were premedicated with midazolam (1–2 mg administered intravenously), and after standard noninvasive monitoring was applied, they were given 100% oxygen to breathe. Anesthesia was induced with propofol (2 mg/kg), fentanyl (3 μg/kg), and vecuronium (0.1 mg/kg) and was maintained with a propofol infusion (100–200 μg · kg⁻¹· min⁻¹) and 50% nitrous oxide in oxygen by mask. In part 1, intravenous drugs were administered in the following order: fentanyl, propofol, vecuronium, study drug. All patients underwent laryngoscopy and tracheal intubation within 3 min of the start of induction. After laryngoscopy and tracheal intubation (7.5-mm endotracheal tube for men and 7.0 mm for women), an esophageal balloon catheter was positioned with its tip 40 cm from the incisors, and the tracheal catheter was advanced to the distal endotracheal tube.

Patients underwent ventilation with equal parts oxygen and nitrous oxide (50/50), using tidal volumes of 10 ml/kg at 8 breaths/min using a square waveform inspiratory flow (Ohmeda Anesthesia Ventilator 7800 series; Datex-Ohmeda, Madison, WI). Respiratory measurements were made as soon as possible after intubation, within 1 min, and at approximately 5-min intervals thereafter for a total of five measurements. Surgical preparation and draping were allowed to begin after obtaining the first two measurements and the start of isoflurane. Heart rate and mean systemic blood pressure were also recorded. After the initial two measurements, isoflurane inhalation (2% inspired concentration) was begun, and the propofol infusion was discontinued. After three more measurements, the study protocol was finished, and respiratory measuring equipment was removed.

Data are presented as mean, SD, median, and range, or as number and percent of patients. For each phase of the study, treated and placebo groups were compared with respect to preoperative respiratory variables, intraoperative hemodynamic and pulmonary measurements, and R^Lusing a two-sample t  test (age, weight, mean blood pressure), Wilcoxon rank sum test (forced expiratory volume in 1 s [FEV¹], heart rate, R^L), or the Fisher exact test (sex, asthma medications). To compare changes in hemodynamic parameters, the number of patients whose postintubation heart rate and mean blood pressure measurements changed by more than 20% from preintubation were compared between groups using the Fisher exact test. The effects of isoflurane were assessed within each group by comparing the difference between maximum preisoflurane and minimum postisoflurane R^Lusing the Wilcoxon signed rank test.

Subjects were originally categorized as responders if the maximum R^L in the first two measurements after intubation, and before administration of isoflurane, was greater than or equal to 5 cm H²O · l⁻¹· s⁻¹and subsequently decreased by 50% or more while breathing isoflurane. In a post hoc  analysis, we explored secondary definitions of response, as described in Results. Response rates of treated and placebo patients were compared using the Fisher exact test.

Figure 1shows R^Lchanges after intubation in a representative patient who met the original definition of response to intubation and in a patient who did not. The initially high pulmonary resistance progressively diminished with time or the start of isoflurane administration.

Although several patients recorded high pulmonary resistances and evidence of expiratory obstruction, only one patient (intravenous lidocaine study group) demonstrated severe bronchoconstriction requiring albuterol treatment and early administration of isoflurane. In this patient we were able to obtain one measurement before rescue therapy. This patient had received intravenous lidocaine and was listed as a responder.

Sixty patients were randomized, 30 in each group. Three patients receiving placebo were excluded from analysis, two because of change in the anesthetic plan (no tracheal intubation necessary) and one because of inadequate oscilloscope records. There were no significant differences between lidocaine and placebo groups in preoperative patient characteristics, preoperative FEV¹, or percent change in FEV¹after bronchodilation. There were no differences in preintubation hemodynamic values or in the proportion of patients whose heart rate or mean blood pressure changed by more than 20% with intubation, suggesting that the groups were anesthetized to similar depth (table 1).

There was no significant difference in R^L measurements after intubation and before isoflurane (mean R^L, 8.2 [SD 9.1] cm H²O · l⁻¹· s⁻¹for lidocaine, 7.6 [SD 6.7] cm H²O · l⁻¹· s⁻¹for placebo;table 1and fig. 2). There was also no significant difference in R^Lmeasurements after the administration of isoflurane (4.3 [SD 3.6] cm H²O · l⁻¹· s⁻¹for lidocaine vs.  4.3 [SD 4.5] cm H²O · l⁻¹· s⁻¹for placebo). In addition, there was no significant difference in the rate of response to tracheal intubation (table 2and fig. 3).

Fifty patients were randomized, 25 in each group. Two patients receiving placebo were excluded from analysis, one because of change in the anesthetic plan (no tracheal intubation necessary) and one because of inadequate measurements. The median FEV¹during preoperative evaluation was significantly greater in the albuterol group (102%vs.  90% of predicted;P = 0.04). However, the change in FEV1 (%) after bronchodilation was not different between the two groups. There were no differences in preintubation hemodynamic values or in the number of patients whose heart rate or mean blood pressure changed by more than 20% with intubation (table 1).

The R^Lafter intubation and before isoflurane was lower in the treated group (mean R^L, 5.3 [SD 3.0] cm H²O · l⁻¹· s⁻¹for albuterol, 8.9 [SD 7.4] cm H²O · l⁻¹· s⁻¹for placebo;P = 0.04;table 1and fig. 2). Although the R^L after intubation was also lower for the albuterol group, it was not statistically significant (3.8 [SD 1.8] cm H²O · l⁻¹· s⁻¹for albuterol vs.  4.9 [SD 2.7] cm H²O · l⁻¹· s⁻¹for placebo). There were significantly fewer responders (original definition) in the albuterol group: 1 versus  8 (P < 0.01;table 2and fig. 3).

There were statistically significant reductions in R^Lafter the administration of isoflurane in all study groups (table 1and fig. 2). The reduction in R^Lwas not significantly greater in the intravenous lidocaine group and both placebo groups compared with the albuterol group (albuterol vs.  inhaled placebo, P = 0.05; albuterol vs.  lidocaine or intravenous placebo, P > 0.05).

In determining the protective effect of a drug, our choice of definition of response to intubation was critical because inappropriate criteria for response could obscure differences between groups. Therefore, we performed post hoc  examination of the ability of other definitions of response to reveal hidden effects. We varied the definition of “high” resistance in the initial measurements from 5 to 3 cm H²O · l⁻¹· s⁻¹and analyzed the data using 3 cm H²O · l⁻¹· s⁻¹. It also seemed possible that intubation-induced bronchoconstriction might not resolve with time and isoflurane inhalation; therefore, we combined criteria for response to include patients whose initial R^Land minimal R^Lafter isoflurane were both greater than 5 or 7 cm H²O · l⁻¹· s⁻¹. The results were not different with any of the secondary definitions of response (table 2and fig. 3).

In our study, 19 of 105 patients (18%) smoked tobacco. Smokers were not significantly different from nonsmokers in R^Lbefore isoflurane (mean R^Lof nonsmokers, 7.3 [SD 7.3] cm H²O · l⁻¹· s⁻¹vs.  8.2 [SD 5.8] cm H²O · l⁻¹· s⁻¹for smokers), R^Lafter isoflurane (mean R^Lof nonsmokers, 4.2 [SD 3.4] cm H²O · l⁻¹· s⁻¹vs.  4.8 [SD 2.8] cm H²O · l⁻¹· s⁻¹for smokers), or rate of response to intubation. The study groups were not different regarding tobacco use or R^Lmeasured in smokers and nonsmokers before or after administration of isoflurane.

Depth of anesthesia, as inferred from the stability of heart rate and mean blood pressure, was similar in drug and placebo groups in which similar numbers of patients demonstrated 20% change in heart rate or mean blood pressure. For each part of the study, no association was found between changes in heart rate or mean blood pressure and response to intubation. Similarly, there was no association between preoperative FEV¹or change in FEV¹after bronchodilator and airway response to intubation.

Our results show that intravenous lidocaine, 1.5 mg/kg, given within 3 min before intubation, was not effective in preventing postintubation bronchospasm in asthmatic patients undergoing general anesthesia with tracheal intubation after a propofol induction. However, inhaled albuterol was effective. The efficacy of albuterol was expected, as it is an effective bronchodilator in patients with asthma and in the setting of general anesthesia with tracheal intubation. 2,3,13,14,24By contrast, although intravenous lidocaine has been described as useful in preventing intubation-induced bronchospasm, 2,20,24its efficacy is less well established. 26Previous studies demonstrating bronchodilation after intravenous lidocaine did so either in nonintubated patients or during provocative tests after intubation and stabilization. 15,20,23–25,27,28Other studies demonstrating beneficial effects of lidocaine on tracheal muscle tone were performed in vitro . 17,19,21Intravenous lidocaine alone has been shown to have minimal effects on bronchial tone. 23,24,29,30In the setting of a histamine challenge, doses up to 10 mg/kg may even result in airway narrowing consistent with bronchoconstriction, 25although bronchodilation may result with similar doses after methacholine challenge. 31 

Recommendations for the use of intravenous lidocaine have been based principally on laboratory investigations in humans or animals or in vitro  experiments. 17–24,27,31Several of these studies have tested the efficacy of intravenous lidocaine in the setting of direct or indirect airway irritants such as histamine, 7,24,25,27,28acetylcholine or methylcholine, 29,31or mechanical irritants such as distilled water or inhaled lidocaine. 12,22Few, if any, studies have tested the benefits of prophylactic lidocaine in the setting of intubation-induced bronchoconstriction. Of the studies assessing the affects of lidocaine on airway tone in asthmatic patients, none involved general anesthesia with intubation, and only three examined intravenous lidocaine. 9,11,23,24,28Inhaled lidocaine administered to asthmatic patients is reported to cause increased bronchomotor tone or constriction, whereas intravenous lidocaine was reported to cause bronchodilation. 9,11,23,24,28 

Although our results show no benefit of lidocaine in these conditions, it must be recognized that our data may have been affected by the anesthetic regimen used. Opioids have been shown to inhibit airway responses to a variety of stimuli. 32–34Likewise, propofol has been shown to attenuate airway response to a variety of stimuli compared with thiobarbiturates, the latter of which is commonly used in both clinical practice and studies of bronchial response. 4,6,12,15,22,35Eames et al.  6reported a postintubation R^Lof 8.1 cm H²O · l⁻¹· s⁻¹after propofol compared with 11.3 cm H²O · l⁻¹· s⁻¹for thiopental. In the present study, in which propofol was used in all patients, immediate postintubation R^Ls were similar to that reported by Eames et al.  for all groups except the albuterol group, in which it was lower, suggesting that albuterol adds significantly to the previously demonstrated protective effect of propofol. 6This was also demonstrated in a recent investigation by Wu et al.  36In this study, fenoterol, and not ipratropium, was effective in decreasing respiratory resistance after intubation in patients anesthetized with propofol. The investigators noted that use of propofol did not prevent increased resistance immediately after intubation, and that the “protection was not absolute in asthmatic patients.”36Finally, the addition of isoflurane resulted in a small but significant reduction of R^Lin the albuterol group, a result that has been demonstrated elsewhere. 13,14 

Another variable among investigations is the dose and timing of intravenous lidocaine. Doses ranging from 1 to 10 mg/kg, with and without infusions, have been administered from 1 to 10 min before laryngoscopy and intubation with a wide range of effects on hemodynamic changes, the cough reflex, and pulmonary resistance. In the current study, a dose of 1.5 mg/kg intravenous lidocaine was administered within 3 min of intubation. Other studies have shown similar dosing to be effective in reducing the minimal alveolar–anesthetic concentration of inhalation agents, 37blunting hemodynamic changes with laryngoscopy and intubation, 8,38,39attenuating the cough reflex, 12,40,41and decreasing bronchomotor tone, i.e. , bronchodilation. 24On the other hand, two studies have not demonstrated a significant benefit from this dose of intravenous lidocaine. 29,42Furthermore, Hirota et al.  25demonstrated a reduction in bronchial area, i.e. , bronchoconstriction, with doses of 1 or 10 mg/kg intravenous lidocaine. Yokioka et al.  40studied doses ranging from 0.5 to 2.0 mg/kg given 1–15 min before intubation and concluded that 1.5 mg/kg or greater given 1–3 min before airway stimulation suppressed the cough reflex.

We did not demonstrate any significant effect of tobacco use on the airway response to tracheal intubation. Although other studies have shown that tobacco use may increase airway response to intubation, the relative airway effects between tobacco and preexisting asthma are unknown. 3,6We also did not demonstrate any association between preoperative FEV¹, or change in FEV¹after bronchodilation, and bronchial response to intubation. Our data suggest that preoperative FEV¹may not be useful in predicting which patients may exhibit bronchoconstriction with intubation. This is not surprising because asthma is an episodic disease. During preoperative testing, patients are in a stable condition. At this point a known bronchodilator may not have any significant change on airflow and may therefore not be predictive of response to a noxious stimuli such as an endotracheal tube. Provocative testing before intubation may have been a better indicator of airway response to a noxious stimuli.

Sample size may be a limitation for the first part of the study. It is not known what fraction of asthmatic patients show clinically significant bronchoconstriction in response to tracheal intubation. In part 1, the response rate (original definition) was approximately 20% in both groups, whereas the response rate in the placebo group in part 2 was 35%. From our data, we cannot exclude the possibility that lidocaine had a small but clinically significant effect. For example, if lidocaine reduced the incidence of response from 35% to 20%, we would have needed approximately 140 patients per group to have an 80% chance of demonstrating an effect. With 30 patients in each group, a study has an 80% chance of detecting a reduction in response from, for example, 35% to 6.5% or from 30% to 3.5%, which is similar to the response rates in the second phase of our study.

In conclusion, our study demonstrates that intravenous lidocaine was not effective in reducing the airway responsiveness to tracheal intubation. Despite a number of studies showing bronchodilation after administration of both intravenous and topical local anesthetic agents, no study has demonstrated a protective effect of these agents in preventing bronchospasm after intubation in patients undergoing general anesthesia. Our study also showed that inhaled albuterol is effective in reducing airway responsiveness to intubation, in agreement with previous studies. Given the lack of effect of intravenous lidocaine (1.5 mg/kg) as well as an absence of convincing data in the literature, we do not recommend the routine use of intravenous lidocaine to reduce bronchospasm after tracheal intubation, at least in patients given propofol. However, inhaled albuterol is effective for this purpose.