Randomized Double Blind Control Study Comparing the Efficacy of Intracuff Alkalinized Lidocaine to Low Dose Remifentanil Infusion in Attenuating the Endotracheal Tube Induced Emergence Phenomena

Emergence from general anesthesia is often complicated by the endotracheal tube (ETT) induced airway and circulatory reflexes which can lead to potentially dangerous complications. Considerable research has been focused on prevention of these emergence phenomena (EP). Nevertheless, the problem is still far from its final solution. Objective: To compare the efficacy of intracuff alkalinized lidocaine (ICL) vs low dose remifentanil infusion in attenuating the ETT-induced EP. Methods: 120 ASA I-III patients, aged 18-65 years, were randomly assigned to receive intracuff alkalinized lidocaine (2% lidocaine mixed 1:1 with 1.4% NaHCO3) or an intravenous (IV) remifentanil infusion (0.05-0.5 mcg/ kg/min) combined with intracuff saline during desflurane-based general anesthesia. At the end of surgery, after desflurane was turned off in the assigned group, low dose remifentanil, or its equivalent placebo was decreased to one-tenth of the mean dose but not less than 0.01 mcg/kg/min and it was continued until extubation. A blinded researcher observed each patient from the time desflurane was discontinued until at least five minutes after extubation. Coughing was evaluated as either present or not, and graded on a point scale based on severity. The patients were also observed for development of any adverse events along with the vital signs during this emergence phase. Results: The incidence (44% vs 67%, p=0.02) and severity of coughing, overall, was significantly less in the lidocaine group compared to remifentanil group). The lidocaine group also had a lower incidence of significant coughing (2-3 on point scale) (25% vs 49%, p=0.009). The mean arterial pressure (MAP) in the lidocaine group was lower than the remifentanil group at extubation and 5 minutes after extubation. Conclusions: Intracuff alkalinized lidocaine (ICL) is more effective in reducing the incidence and severity of coughing compared to a low dose remifentanil infusion during emergence from desflurane based anesthesia. Citation: Vandse R, Castellon-Larios K, Fujii J, Melibary S, Wei L, et al. (2014) Randomized Double Blind Control Study Comparing the Efficacy of Intracuff Alkalinized Lidocaine to Low Dose Remifentanil Infusion in Attenuating the Endotracheal Tube Induced Emergence Phenomena. J Anesth Clin Res 5: 435. doi:10.4172/2155-6148.1000435

The reported incidence of patient coughing during emergence is as high as 80-95% [1][2][3]. Unfortunately, this can result in a number of detrimental side effects including hypertension, tachycardia, increased bleeding from the surgical site and increased intracranial and intraocular pressures [6][7][8][9]. Various strategies have been employed to attenuate this response including extubation in a deep plane of anesthesia [10], administration of IV agents like short acting narcotics [11][12][13], IV lidocaine [14][15][16] or dexmedetomidine [17] and the topical or intracuff application of lidocaine [1][2][3][4][5]. Each of these methods has its own limitations and a reliable technique which would increase the ETT tolerance while facilitating rapid and complete emergence from general anesthesia has not been proven to date.
Even though, IV lidocaine has shown to suppress both mechanicallyand chemically-induced airway reflexes, [14][15][16] its narrow antitussive window accompanied with the potential for systemic toxicity makes it an unreliable agent as a cough suppressant [18,19]. One of the principal mechanisms for coughing during emergence is irritation of the respiratory mucosa by the ETT and its cuff [1,19,20]. Hence, anesthetizing the mucosa which is in direct contact with the ETT and its cuff should improve ETT tolerance. Lidocaine administered through the Laryngotracheal instilled topical anesthesia (LITA) tube is also effective in suppressing the cough [21,22]. However, the study by Crerar et al. demonstrated greater efficacy of intracuff alkalinized lidocaine compared to a LITA tube [22]. Estebe et al showed successful suppression of cough reflex and sore throat with as little as 40 mg of intra cuff alkalinized Lidocaine [23][24][25] Encouraged by these reports, we decided to use low dose of alkalinized lidocaine to inflate the ETT cuff.
Acting on the central and peripheral opioid receptors, intravenous opioids have been shown to be successful in suppressing cough during emergence [11][12][13]. They have also been shown to be effective in attenuating the cardiovascular response to extubation [12,26]. An advantage of remifentanil over other opioids is its rapid onset and offset of action. It has a short blood-brain equilibration time and rapid elimination which makes it easier to adjust the infusion rate to achieve the desired levels of analgesia [27,28]. Aouad et al. [13] demonstrated that a low-dose infusion of remifentanil during emergence suppresses coughing and provides a smooth emergence from isoflurane-based anesthesia.
Currently, there are no reports on head-to-head comparison of the remifentanil infusion with intracuff alkalinized lidocaine in suppressing the emergence coughing.
Our primary objective was to measure the incidence of coughing between the ICL vs low dose remifentanil infusion groups. Secondary objectives were to compare the severity of coughing, calculate the mean time to extubation, incidence and severity of sore (or dry) throat, and incidence of other side effects.

Methods
We conducted a single-center, prospective, randomized, doubleblinded and double-arm trial at The Ohio State University Wexner Medical Center. After obtaining approval from our Institutional Review Board and written informed consent from all patients, we included 120 ASA I-III patients aged 18-65 years who were scheduled to undergo elective surgery under general anesthesia requiring endotracheal intubation for longer than two hours. The types of surgeries are listed in Table 1. Pregnant or breastfeeding patients, those with a history of asthma, chronic obstructive pulmonary disease, chronic cough, recent respiratory infections, active gastrointestinal reflux disease, hiatal hernia, malignant hyperthermia, laryngeal or tracheal surgery or pathology, or patients with allergic to either of the study medications, as well as prisoners, were excluded from the study. Additionally, subjects were excluded if they had severe cardiovascular, pulmonary, renal, hepatic, metabolic, psychiatric, endocrine, or neuromuscular disorders and if a nasogastric tube was required after extubation. Lastly, patients with an anticipated difficult intubation as determined by the anesthesiologist or a failure to intubate on the first attempt served as additional exclusion criteria.
Patients were randomly allocated to either remifentanil infusion or ICL group according to a computer-generated random table. All patients received IV induction with lidocaine, propofol, and fentanyl, and the choice of muscle relaxant which was left to the discretion of the anesthesiologist. Endotracheal intubation with a high volume low pressure cuffed ETT was performed by direct laryngoscopy. Cuff inflation was performed with normal saline for the remifentanil group and with alkalinized 2% lidocaine (2% lidocaine with 1.4% NaHCO 3 in a 1:1 volume ratio prepared immediately before administration) for the lidocaine group. Initial minimal occlusive pressure was established by listening for an audible leak and the cuff pressure was measured and recorded at different times to insure pressures of 25 mmHg or less. Desflurane (0.8-1 MAC) was used for maintenance of general anesthesia. Fentanyl by intermittent boluses was used for analgesia in the lidocaine group, while the continuous infusion of remifentanil at a rate 0.05-0.5 mcg/kg/min was used in the remifentanil group. These were titrated to keep the MAP and HR within 10% of the preoperative baseline values. The normal saline was run as an infusion in the lidocaine group for blinding purposes.
Towards the end of surgery, at the start of closing, neuromuscular blockade was reversed with neostigmine and glycopyrrolate. Mean remifentanil dose was calculated by dividing the total maintenance dose by the patient's weight and duration of surgery. Upon skin closure, desflurane was turned off (T0) and remifentanil was reduced to one tenth of the mean dose but not less than 0.01 mcg/kg/min and was titrated to patient's spontaneous respiratory rate. Similarly fentanyl dosage was titrated to maintain spontaneous respiratory rate between 10-20/min. Tracheal and oral suctioning was done immediately after stopping the desflurane.
During the emergence phase (from T0 to extubation, T1) 100% O 2 was administered at 10 L/min and the patient was intermittently stimulated verbally or with gentile tactile stimulation. The patients were extubated when they met the standardized extubation criteria. The remifentanil or normal saline infusion was stopped at extubation and a longer acting narcotic was administered for post-operative pain control. Once desflurane was discontinued (T0), a blinded observer recorded the mean arterial blood pressure (MAP), heart rate (HR), SpO 2 , spontaneous respiratory rate (RR), end tidal CO 2 (ETCO 2 ) and desflurane concentrations at two minute intervals until 5 minutes after extubation. The number and grading of coughing episodes was also recorded ( Table 2).
A second blinded observer evaluated the patient for sore throat in the post anesthesia recovery unit with a visual analog scale (VAS, 0-10 cm) at one hour after extubation. All other side effects that the subjects experienced were recorded as well.
One hundred and twenty patients underwent 1:1 randomization into remifentanil infusion and intracuff alkalinized lidocaine groups. To test the primary hypothesis, a sample size of 60 per group was required to achieve a power of 80% to detect a difference of 24% between the two groups, assuming 40% cough incidence in the remifentanil group [13] and 16% in the ICL group [1] The sample size was calculated by a 2-sided Pearson Chi-square test at α=0.05. This sample size took into account a 10% dropout rate.

Statistical Analysis
First, the cough score was dichotomized to 0 and 1. A score of 1-3 was treated as an event occurrence (1=cough) while 0=no occurrence. The proportion of patients with coughing episodes was compared between the two groups by the Pearson's Chi-square test. Odds ratio and its 95% confidence interval was calculated as well. Results were summarized using descriptive statistics (means, standard deviations for continuous variables and frequencies for discrete data). The continuous variables were compared between treatment groups using the twotailed T-test, and the categorical variables were compared using the Chi-square test, or Fisher's exact test, as indicated. Bonferroni method has been used to adjust the multiple comparisons at different time point for both MAP and HR values. A P value<0.05 was considered as statistically significant.

Results
A total of 120 patients were enrolled: 60 subjects for each group. 10 subjects were excluded from analysis because of screening failures and one patient was excluded because of missing data. Thus, the total number of patients analyzed was 57 subjects for the remifentanil group and 52 for the ICL group ( Figure 1).
There was no statistically significant difference with regards to age, gender, BMI, smoking history, or history of hypertension between the lidocaine and remifentanil patient groups. Incidentally, the mean duration of surgery and intubation in the lidocaine group was significantly longer than in the remifentanil group (Table 3). However, there was no difference in mean extubation time between the groups: 6.6 ± 2.8 minutes vs 7.1 ± 2.8 minutes, p=0.39.The patients in lidocaine group had a significantly lower incidence of cough compared to remifentanil group: 44% vs 67%, p=0.02 (Table 4 and Figure 2). The lidocaine group also had significantly lower rates of cough before extubation: 35% vs. 65%, p=0.002. However there was no significant difference in post-extubation cough rates: 19% vs. 14%, p=0.47.The lidocaine group also had lower incidence of significant coughing (grade 2 and 3), when compared to the remifentanil group: 25% vs 49%, p=0.009.
A post-hoc subgroup analysis was performed in order to readjust the mean duration of surgery between both groups (Table 5). Subjects taken into consideration on this analysis were those who had surgeries lasting between 120 and no more than 300 minutes. From these, 35 subjects remained in the lidocaine group and 29 in the remifentanil group. Duration of surgery did not show any statistical significance (p=0.14). Nonetheless, significant coughing before extubation and overall coughing remained higher in the remifentanil group when compared to the lidocaine group: 55% vs. 20% (OR=4.9); p value=0.004 and 55% vs. 23% (OR=3.1); p value=0.008.
There was no significant difference between the groups with regard to incidence of sore throat, postoperative nausea, or desaturation after extubation. Additionally, there were no differences between the groups in the incidence or severity of sore throat in the PACU. The majority of patients experienced no significant adverse effects (Table 6). Baseline MAP and HR values were comparable between the two groups. Both groups had some sympathetic stimulation during emergence evidenced by raise in the HR and MAP compared to baseline. Between the two  groups, mean HR was similar during emergence however MAP was significantly higher in the remifentanil group at extubation and at 5 minutes after extubation (Figure 3).

Discussion
The effects of intra cuff lidocaine and remifentanil infusion in suppressing cough during emergence from general anesthesia have been thoroughly investigated [3,[11][12][13][23][24][25]. However, no study has directly compared the two agents to demonstrate a greater efficacy of one over the other. Comparison between different studies is limited due to variation in the protocol design and anesthetic agents (inhaled anesthetics, narcotics). Our results indicate that ICL is more effective in reducing emergence coughing compared to low dose remifentanil infusion in patients receiving desflurane anesthesia.
Repeated success has been demonstrated with the ETT cuffs filled with lidocaine in suppressing the cough reflex [1][2][3]. The quantity of lidocaine that diffuses through the cuff was found to be directly related to the concentration of lidocaine and time [29,30]. Hence, earlier studies [1] used large doses of concentrated solution of lidocaine (200-500 mg) despite the possible adverse reactions if the cuff would rupture. Later studies [23,24] found that alkalization of lidocaine increased the rate of diffusion as much as 65 to 100 times which allowed reduction in the lidocaine dose while achieving the same results. One of the drawbacks of using 8.4% NaHCO 3 for alkalinization of Lidocaine was the potential for tracheal irritation if the cuff ruptures. Estebe et al. [25] has shown 1.4% NaHCO 3 to be as effective as 8.4% NaHCO 3 . They showed successful attenuation of cough reflex and sore throat with as little as 40 mg of lidocaine alkalinized with 1.4% NaHCO 3 . The incidence of coughing in their study was only 5%, however, they used continuous infusion of sufentanil in all the patients. The anti-tussive effect of narcotics needs to be considered while reviewing their results. Fagan et al. [1] used 4% plain lidocaine with a mean dose of 244 ± 36 mg. The incidence of coughing in the lidocaine group in their study was 16%. Another study by Navarro et al. [31], also showed successful attenuation of emergence coughing and sore throat among smokers by using intra cuff alkalinized 2% lidocaine (28% in ICL vrs 80% in control). The mean dosage of lidocaine was 138 ± 52 mg. However, study by Wetzel et al. [32], didn't show any decrease in the emergence coughing by ICL when used for procedures lasting less than 1.5 hours.  The odds ratio and its 95% confidence intervals have been listed.    This was postulated to be due to inadequate time for the lidocaine to diffuse through the cuff wall.
The overall incidence of coughing in the lidocaine group in our study was 44%. Even though, lidocaine performed better than remifentanil, it is still not clear how some patients still continue to cough. There are several possible explanations: we used low doses of lidocaine as described by Estebe et al. [23][24][25]. The mean dose of lidocaine used in our study was only 57.6 mg. Since the amount of drug diffusing through the cuff wall is directly proportional to the concentration of the lidocaine, higher concentration might be more effective in suppressing the cough. The primary mechanism for coughing during emergence is due to irritation of the respiratory mucosa by the ETT and its cuff [1,19] and ICL mainly works by abolishing this reflex. This explains the success of ICL in most of the cases. However, since ICL only anesthetizes the small area in contact with the cuff, any stimulation along the tracheobronchial tree (e.g. irritation with oropharyngeal secretions) can still induce cough. Lastly, ICL might not have a uniform circumferential numbing effect. The diffusion might theoretically be higher on the posterior portion of the trachea (assuming the patient is supine) due to gravity, air collection in the cuff, and other factors. Uneven diffusion may be a probable explanation for uneven numbness and failure to prevent coughing in some patients.
Remifentanil is an ultra-short acting opioid analgesic which has been shown to facilitate the smooth emergence from general anesthesia [11][12][13]. The dose of remifentanil used in our study was based on previous investigation by Aouad et al. [12], who demonstrated that a low-dose infusion of remifentanil attenuates coughing and hemodynamic stimulation during emergence from isoflurane-based anesthesia. The authors of that publication maintained remifentanil at one tenth of the mean dose during emergence, which was calculated at the end of the surgery by dividing the total dose administered during maintenance by the patient's weight and the duration of surgery. The mean remifentanil infusion rate during emergence was 0.014 ± 0.011 mcg/kg/min. The incidence of coughing in the remifentanil group was significantly lower compared to control (40% vs 80%, P=0.002). We followed the same protocol as Aouad et al. in terms of remifentanil infusion but desflurane instead of isoflurane was used in our study. The incidence of coughing in the remifentanil group was much higher in our study-67%, and the mean remifentanil dose during extubation was 0.014 ± 0.005 mcg/kg/ min. The failure of remifentanil to suppress coughing in our study could be related to inadequate effect site concentration of remifentanil as shown in some of the recent studies [13]. Nevertheless, airway irritant property of different inhalational agents also needs to be considered.
Jun et al. [11] using target controlled infusion device(TCI), showed that maintaining the predicted effect-site concentration (Ce) of remifentanil at 1.5 ng/ml during emergence from sevoflurane anesthesia resulted in a 50% reduction in the incidence of coughing whereas patients with remifentanil Ce of 1.0 ng.ml did not show any significant cough suppression In a dose escalation study, Chen et al. [33], also showed successful cough suppression by remifentanil which was found to be dose dependent and more pronounced at higher concentrations of 2.0-2.5 ng/ml. Thus reiterates the importance of maintaining minimum plasma concentration of remifentanil, however, such a goal can be difficult to achieve by manually controlled infusion.
Yamasaki et al. [34], showed synergistic effect of topical lidocaine when combined with remifentanil. In that study, remifentanil infusion at 0.1 μg/kg/min was continued at the end of the surgery in both groups. One group also received 4% lidocaine through the LITA tube. According to the authors, the incidence of coughing in remifentanil only group was 61% whereas that in the combined group was 26%. Interestingly, incidence of coughing was still very high in remifentanil only group even though it was maintained at a very high dose of 0.1 mcg/kg/min indicating that remifentanil is not a reliable antitussive agent.
In our study, the two treatments did not differ in their recovery profiles. Mean time to extubation after discontinuation of desflurane was similar between the groups. This likely confers no advantage or disadvantage to either intervention from a timing and efficiency perspective. Neither intervention group proved to be more effective than the other in reducing the incidence of other side effects, including postoperative nausea, postintubation desaturation, and dry/sore throat.

Limitations
There are several limitations in our study. First, even though we followed a standardized protocol, we were unable to use the same anesthesiologist or the observer for all the patients, and this could have introduced a bias and inter-observer variability into our study. Second, we used a fixed dose of remifentanil and didn't measure the plasma concentration. It is important to test if with higher doses of remifentanil, as opposed to doses used in our study, it will be possible to reach results comparable with ICL. Third, the duration of surgery and intubation was greater in the lidocaine than the remifentanil group. Given that remifentanil is an ultra-short acting opioid, it could be assumed to be at steady state after two hours and prolonging the remifentanil infusion would likely entail a similar conclusion. Finally, we didn't have a control group to compare the effects of remifentanil.

Conclusion
The intra cuff alkalinized lidocaine is more effective than low dose remifentanil infusion in suppressing coughing during emergence from Desflurane based anesthesia in patients undergoing surgeries lasting for longer than 2 hours. ICL was also more effective than remifentanil for reducing severity when coughing did occur. Moreover, ICL is advantageous when compared with several other techniques including remifentanil infusion, because it is very easy to use, doesn't require any special infusion device, is inexpensive, and has an established excellent safety profile.