Introduction

Fentanyl is one of the common opioids used during the pre-induction period because of the rapid onset, short duration and reduced cardiovascular effects. However, cough is a frequently adverse effect after fentanyl injection [1]. Even though fentanyl-induced cough (FIC) occurs in a short period, immediate management should be provided [2,3,4,5,6,7] because FIC is related to the magnification of intraocular, intracranial and intra-abdominal pressure [2, 7] and also periorbital or conjunctival petechiae [2].

However, FIC mechanisms are not well elucidated, but several mechanisms have been described. For example, fentanyl stimulates μ receptors and elicits neuronal transmission to activate receptors on the brainstem or transmits through the C-fiber of the vagus nerve to stimulate mucosal receptors on the proximal bronchus causing bronchoconstriction and cough [8].

Nevertheless, prior studies showed various methods to attenuate or treat FIC including nonpharmacologic approaches to suppress the incidence of FIC such as preoperative incentive spirometry [9], the huffing maneuver [10] and swallowing instantly before receiving intravenous fentanyl [11]. Other methods included pharmacologic administration of preemptive low-dose fentanyl [12,13,14], pre-injection with clonidine (α2-adrenoceptor agonist) [15] or ketamine [16], durieux [17], propofol [18, 19], pheniramine maleate [20], dexamethasone [21], vecuronium [22], MaSo4 [23, 24], pre-inhalation of salbutamol or terbutaline [7, 25], beclomethasone or sodium chromoglycate [8], terbutaline inhalation [9] and lidocaine [26,27,28,29,30].

Interestingly, lidocaine is a local anesthetic medication which has rapidly expanded in common practice, is easy to use and can be applied in multiple routes such as topical, inhalation, perioperative intubation and extubation [26, 27] Moreover, regional anesthesia is a requisite high volume of local anesthetic [31,32,33,34] as a consequence of side effects or toxicity, especially involving many routes of administration in the same patient [35].

Unfortunately, the outcomes of using a very low dose (0.25 mg/kg of lidocaine) have not been investigated. Consequently, this constitutes the first prospective controlled study aimed to demonstrate the effectiveness of lidocaine 0.25 mg/kg to suppress FIC compared with lidocaine 0.5 to 1 mg/kg to minimize adverse effects of lidocaine from multiple sites of injection or high-volume injection in the same patient.

Methods

Study design

The study employed a prospective, single center, block-randomized, patient/assessor-blinded design and active-controlled trial. After the Institutional Review Board of the Royal Thai Army Medical Ethics Committee approved and registered the study in the Thai Clinical Trials Registry (TCTR20191207001), 343 patients provided informed consent and attended the operating theater, Phramongkutklao Hospital, from November 2019 to December 2020. The inclusion criteria included patients aged 18 to 60 years with scheduled general anesthesia and American Society of Anesthesiologists (ASA) physical status classification I or II. The exclusion criteria comprised patients indisposed to participate, patients reporting a record of substance use disorders, chronic opioid use, opioid tolerance or opioid-induced hyperalgesia, presenting a history of allergy to lidocaine or fentanyl, history of chronic obstructive pulmonary disease, bronchial asthma, recent respiratory tract infection less than two weeks. Additional criteria included impaired kidney or liver function, or having been treated with angiotensin converting enzyme inhibitors or beta blockers, symptomatic bradycardia, tachyarrhythmia, left bundle branch block and second- or third-degree atrioventricular block, pregnancy and language barrier.

Randomization and allocation

All patients completed preoperative evaluation using only one anesthesiologist, and any anesthetic premedication was prohibited. Solid food was allowed until 8 h before the scheduled operative time, and clear liquids were permitted until 3 h before this time. The patients were randomly assigned in four groups equivalently using a computer-generated table and sealed envelopes. The random records were preserved and released by an anesthesia resident uninvolved in this study.

In all, 0.9% NaCl, 0.25, 0.5 and 1 mg/kg of lidocaine were established in Groups I, II, III and IV, respectively. All participants and one anesthesiologist who injected the solution and assessed the outcomes were blinded to the allocation groups.

Procedures

The procedures were performed in the inpatient department under general anesthesia. A nurse anesthetist started an intravenous infusion of isotonic balance solution since the patients had been admitted. No pre-medication was given. All patients were monitored for noninvasive blood pressure, electrocardiography and pulse oximetry on arrival at the operating theater. Lidocaine or 0.9% NaCl was injected within 3 to 4 s before two min of 3 μg/kg intravenous fentanyl from 3 to 4 s. A stopwatch was monitored during the injection periods. Subsequently, anesthetic induction was performed after fentanyl injection for 3 min.

Outcome measurement

The primary result was incidence of cough among group comparisons. The secondary results involving severity and risk factor of cough and hemodynamic response (systolic blood pressure (SBP), diastolic blood pressure (DBP) and heart rate (HR) were documented at two time points: 3 min before fentanyl injection (baseline) and at 3 min after starting fentanyl injections. Severity of coughing was established as episodic number of coughs (mild 1 to 2, moderate 3 to 4 and severe 5 or more) [30]. All outcomes were assessed within 3 min after fentanyl injection and completely supervised by one blinded anesthesiologist.

Statistical analysis

The sample size was calculated from a related study showing a 20% reduction of FIC after intravenous lidocaine administration and assuming incidence of FIC was 30% [14]. The result showed 79 patients per group were obliged to attain a significance level of 0.05 with 80% power of test. Categorical data were presented as percentage, and continuous data were presented as mean and standard deviation (SD). One-way ANOVA with multiple comparison tests and Bonferroni correction and Chi-squared test were executed to compare the differences between groups in categorical and continuous variables. Univariable analysis was implemented with two-sample t-test for numerical and Chi-squared test or Fisher’s exact test (count less than 5) for hypothesized factors to correlate with cough. Multivariable analysis was performed using backward binary stepwise logistic regression and the last model used an enter method to define risk factors of FIC. All data were analyzed using SPSS, version 26.0 (IBM Corp. Released 2011, IBM SPSS for Windows, Armonk, NY, USA). Results were expressed as odds ratio (OR) with 95% confidence interval (CI). A P-value of less than 0.05 was considered statistically significant.

Results

Three hundred forty-two patients were screened for inclusion. Of these, 22 were excluded because the operation was canceled for 6 patients, and pre-medication was administrated for 16 patients leaving 320 eligible patients for randomization and allocated in equal groups of 80. After randomization, none of the patients dropped out of the study or discontinued the allocated intervention as shown in Fig. 1. No clinically significant difference was observed among groups (P > 0.05) regarding demographic data including age, sex, weight, height, body mass index, underlying disease, smoking, alcohol consumption, ASA classification and operation type as presented in Table 1

Fig. 1
figure 1

CONSORT diagram of the study

Full size image
Table 1 Demographic data and baseline characteristics
Full size table

Thirty-two patients (40%) of the control group (Group I) presented cough. The incidence of cough was noted among 15, 13 and 11 patients (18.8, 16.3 and 13.8%) in Groups II, III and IV, respectively. Significant differences were observed between Groups I and II (P = 0.003), Groups I and III (P = 0.001) and Groups I and IV (P < 0.001). However, no significant difference was observed in the incidence of cough between lidocaine groups. [Groups II and III (P = 0.677), Groups II and IV (P = 0.391) and Groups III and IV (P = 0.658)] as presented in Table 2 and Fig. 2. Severity of cough in Group II exhibited a weighty trend toward possible mild cough compared with Groups III and IV [Group II vs. III vs. IV: 11 (13.8%) vs. 8 (10%) vs. 8 (10%), respectively; P = 0.684], and none of the patients in Group IV presented severe cough. Unfortunately, no significant difference was found in severity of cough between lidocaine groups (Groups II vs. III vs. IV) (P > 0.05) as shown in Table 2.

Table 2 Incidence and severity of fentanyl-induced cough among the groups
Full size table
Fig. 2
figure 2

Comparison of incidence of cough between groups

Full size image

Overall, average SBP, DBP and HR were significantly reduced from baseline at 3 min after fentanyl injection in all groups (P < 0.05). However, no significant difference was observed among groups comparisons (P > 0.05) as shown in Tables 3, 4 and Fig. 3).

Table 3 Hemodynamic responses at pre- and post-fentanyl administration
Full size table
Table 4 Univariate logistic regression for fentanyl-induced cough
Full size table
Fig. 3
figure 3

Comparison of changes in hemodynamic response

Full size image

Univariate analysis showed significant risk factors were associated with FIC among patients aged younger than 40, body mass index less than 25, no underlying hypertension, diabetes mellitus and dyslipidemia, no history of smoking, ASA classification I and no pre-lidocaine administration (P < 0.05) as shown in Table 5.

Table 5 Multivariate logistic regression and the last model using a enter method to define risk factors of fentanyl-induced cough
Full size table

Multivariate analysis using backward binary stepwise logistic regression and the last model using an enter method to define risk factors of FIC revealed that patients younger than 40 had 2.96 times the risk of cough [P = 0.007; adjusted odds ratio (adjusted OR) 2.96, 95% CI 1.35 to 6.5], as well as patients reporting no smoking had 6.5 times the risk of cough [(P = 0.013; adjusted OR 6.5, 95%CI 1.49 to 28.41). In addition, no pre-lidocaine injection produced 4.52 times the incidence of cough [P = 0.001; adjusted OR 4.52, 95% CI 1.93 to 10.56].

Additionally, no serious adverse effects of lidocaine either neurotoxicity or cardiotoxity was found during the study.

Discussion

FIC occurs commonly, especially in pre-anesthetic induction. Related studies have revealed that the incidence of FIC varies between 18 and 65% [7, 8, 28, 29, 35,36,37,38]. The study showed the incidence of FIC occurred in up to 40% of patients, which was higher than that reported in related studies showing 34.22 to 35% after 3 μg/kg of fentanyl was administered using a peripheral intravenous route [28, 29]. The higher incidence of FIC was probably due to a rapid fentanyl injection from 3 to 4 s. Lui et al. observed 43% after 5 μg/kg of fentanyl was administered using a peripheral intravenous route over 5 s [25]. Lin et al. showed 65% following 2.5 μg/kg of fentanyl using a peripheral intravenous route within 2 s [36]. These results were coupled with findings of 28% cough incidence after peripheral intravascular fentanyl 1.5 μg/kg injection by Phua et al. [38]: The same incidence from 2 μg/kg of fentanyl injected by peripheral intravenous route more than 5 s was reported by Agarwal et al. [7]. Moreover, Lin JA et al. established a long duration of fentanyl injection diminished the incidence of FIC [37].

As mentioned above, the incidence of FIC depended on dose and velocity of fentanyl injection, which increased significantly with rapid injection time [37]. Yu H et al. revealed that diluted fentanyl 50 to 10 mcg/ml attenuated FIC [39]. Therefore, the study recommended a medium dose of fentanyl (3 μg/kg) and adopted the injection time of fentanyl from 3 to 4 s, for which dose and injection time is commonly used in daily practice.

Even though the mechanisms of FIC were not well demonstrated, numerous theories have been proposed to reduce central sympathetic system symptoms producing vagal dominance and provoking cough and reflex bronchoconstriction [7, 25, 40]. In addition, lidocaine has been proposed to block the peripheral cough receptors in the trachea [26, 27, 41] and provide suppression both through chemical and mechanical airway reflexes [42, 43]: thus, reducing bronchoconstriction.

With reference to multimodal analgesia, local anesthetic injection was increased following many procedures [31, 32]. Consequently, the accumulated dose of lidocaine in either different routes of injection or high volume of injection [33, 34] may lead to adverse side effects [44]. For this purpose, the study indicated very low dose of lidocaine (0.25 mg /kg) two min before fentanyl administration to alleviate the adverse outcomes from accumulated doses of lidocaine and revealed 0.25 mg/kg was effective as 0.5 to 1 mg/kg of lidocaine to suppress FIC compared with placebo. However, no difference in the incidence of cough was found within lidocaine groups according to Pandey et al. showing that 0.5 to 1.5 mg/kg of lidocaine significantly suppressed FIC [28] and Golmohammadi M et al. revealed that 1 mg/kg of lidocaine pre-injection significantly alleviated the incidence of FIC [13]. Furthermore, in a recent study, a meta-analysis showed no significant difference between low doses (0.5 to 1 mg/kg) and high doses (1.5 to 2 mg/kg) of lidocaine at reducing the incidence of FIC[45].

Interestingly, the study found lidocaine injection for 2 min before fentanyl injection could reduce the incidence of cough in which the duration was longer than related studies using fentanyl injection following lidocaine injection within 1 min [28, 29, 35, 38, 46]. More precisely, the onset of intravenous lidocaine showed in 45 to 30 s with maximum effect at 1 to 2 min [47] and Mikawa et al. found that intravenous lidocaine two min before extubation of an endotracheal tube reduced the cough reflex [48]. Consequently, the study reported 2 min prior fentanyl injection could provide the maximum peak effect of lidocaine.

In spite of significantly reduced FIC doses of lidocaine, the trend toward dose-dependent character and plasma level according to related studies indicated increased doses of lidocaine showed additional effects of reducing cough from endotracheal intubation [42]. However, no significant difference was found among lidocaine groups.

Nevertheless, the study showed statistical difference from baseline regarding the hemodynamic response (SBP, DBP and HR) in all groups, but no statistical difference was found between lidocaine and placebo groups after fentanyl injection. Similarly, related studies showed fentanyl and lidocaine injection alleviated hemodynamic response during tracheal intubation [49].

In addition, the study found risk factor of FIC increased in age below 40 and nonsmoking subjects the same as reported by related studies [36, 50, 51]. However, sex or underlying disease including diabetes mellitus, hypertension and dyslipidemia were unrelated to the risk factors of FIC, according to a recent study [52] in which incremental titration of fentanyl could be applied for patients with underlying who experienced acute and chronic intractable pain.

The present study encountered a number of limitations. Firstly, the study demonstrated lidocaine was administrated within 2 min before fentanyl injection, in which related studies disclosed the duration of cough reduction from lidocaine was 5 to 8 min [42, 53, 54]. However, the same studies found the peak effect of lidocaine at 1 to 2 min [47]. Secondly, a dose solely established by body weight might have produced different effects. Thirdly, the incidence and severity of cough were not assessed over 3 min after fentanyl administration for which FIC probably rose over 3 min after fentanyl injection. However, many related studies have summarized that FIC occurred within 15 s after injection and none of the patients presented cough after 15 s [20].

Conclusion

The study implied intravenous lidocaine 0.25 mg/kg for 2 min before fentanyl injection was the most effective dose to suppress FIC and could be applied in daily practice. Findings indicated patients aged less than 40 years and nonsmoking were risk factors of FIC, regardless of sex and underlying disease.