The Impact of Head Elevation on Prevalence and Severity of Emergence Cough in Male Patients during Endotracheal Extubation

Background

Extubation is the process of removing an endotracheal (ET) tube, which liberates a patient from the mechanical ventilators [1]. According to the American Thoracic Society and the American College of Chest Physicians Clinical Practice Guidelines, tracheal extubation is recommended to be managed with a ventilator liberation protocol, including weaning parameter assessment (eg, frequency-tidal volume ratio) or spontaneous breathing trial [2,3]. However, despite these preparations, adverse events associated with extubation can occur, such as patient distress, hemodynamic perturbations, compromised ventilation, and airway irritation or injury [4].

Emergence cough from general anesthesia occurs frequently, with an incidence of up to 76% [5]. Uncontrolled cough at perioperative extubation can increase thoracic and intracranial pressure, bleeding at the surgical site, and wound dehiscence, and can cause hemodynamic instability [6]. Thus, strategies for reducing cough have been proposed, such as smooth emergence techniques (deep extubation and no-touch extubation) or pharmacologic interventions (topical and intravenous [IV] lidocaine, IV opioid, and dexmedetomidine) [7–9]. However, there is limited evidence to support the efficacy of these techniques, and there are potential adverse effects (eg, local anesthetic toxicity, opioid abuse, vomiting, and delayed emergence), and they can be expensive [4]. Despite several studies on adjuvants for improving emergence, there are no adjuvant maneuvers to reduce the emergence cough to date [4].

The cough reflex is mediated by dual sensory neurons (nociceptor and mechanoreceptor) throughout the larynx, trachea, carina, and large intrapulmonary bronchi [10]. The curvature of the ET tube is deviated from the airway anatomy with a slight S-shaped curve, and thus can directly apply pressure to these structures [11,12]. This pressure activates mechanoreceptor, which is considered a main contributing factor provoking cough [10,13]. Mucosal pressure exerted by the ET tube contributes to grade 1 or 2 ulcers, laryngeal incompetence, and even airway structure deformation [14–16].

The 3-axis alignment theory remains a common airway concept for direct laryngoscopy despite a debate about the degree of elevation and extension of head alignment needed [17,18]. Neck flexion by head elevation causes the glottic aperture to move posteriorly, resulting in better alignment of the 3-axis [19]. In addition, head elevation, a commonly used maneuver to maintain airway patency in anesthesia, is possibly associated with better oxygen reserve [20]. Therefore, we hypothesized that head elevation lowers the pressure between the ET tube and the airway wall because the curved ET tube has a propensity to unbend. To the best of our knowledge, this is the first study to adjust the airway passage to the ET tube’s configuration by changing the head position to prevent emergence cough.

This study aimed to evaluate the effects of head position on the prevention of cough during extubation in 71 male patients requiring general anesthesia during cholecystectomy.

Material and Methods

ETHICS STATEMENT:

This study was approved by the Ethics committee of the Ajou’s Institutional Review Board (reference number AJIRB-MED-OBS-20-055, May 19, 2020) and registered at http://ClinicalTrials.gov (NCT04409652, May 25, 2020) before the first participant was enrolled. The protocol was in accordance with the ethical principles outlined in the Helsinki Declaration of 2013, which guides ethical research practices. Written informed consent was obtained from each patient after taking sufficient time to think about it on the day before surgery.

The participants consisted of male patients aged 19–85 years undergoing elective laparoscopic cholecystectomy from June 2020 to November 2022. The exclusion criteria were individuals with an American Society of Anesthesiologists classification of 4 or higher, a body mass index (BMI) of ≥30 g/m², a history of upper respiratory tract infection within the past 2 weeks, a diagnosis of asthma or chronic obstructive pulmonary disease, anticipated difficult intubation, uncontrolled hypertension or diabetes, and severe cardiovascular disease.

PATIENT ALLOCATION:

The participants were randomly assigned to 1 of the following 2 groups using R 3.4.0 (Vienna, Austria; http://www.R-project.org/): the control group and the head elevation group. The control group was intubated in a head-elevated position on a pillow and then maintained in a neutral position (by removal of pillow) after anesthesia induction until ET extubation. The head elevation group was maintained in a head-elevated position on a pillow from anesthesia induction until ET extubation. A protractor was used to adjust the position to a neck flexion angle of approximately 35° based on the relationship between the external auditory meatus and the sternal notch.

ANESTHESIA:

None of the patients were pre-medicated. In the operating room, patients received standard monitoring, including pulse oximetry for oxygen saturation, electrocardiography, non-invasive blood pressure measurements, and bispectral index. After a pillow (8–10 cm thick) was placed under the head of the patient, anesthesia induction was conducted with propofol of 2 mg/kg and remifentanil (Ultian, Hanlim Pharm. Co., Ltd., Korea) using target-controlled infusion based on Minto’s pharmacokinetics at a concentration of 1–5 ng/mL. Once loss of consciousness was achieved, IV rocuronium was administered at a dose of 1 mg/kg to facilitate adequate muscle relaxation. After ensuring sufficient muscle relaxation, intubation was performed using a 7.5 mm ET tube with the assistance of a video laryngoscope. Cuff pressure was adjusted to 20–25 mmHg using a hand pressure gauge. After anesthesia induction, the pillow was removed or kept according to the assigned group during surgery. Sevoflurane and remifentanil were adjusted to maintain bispectral index levels between 40 and 60. Mean arterial pressure (MAP) and heart rate (HR) were maintained within 20% of the baseline values by adjusting the anesthetic agents. Rocuronium was continuously infused to maintain a train-of-four count of 1–2.

All patients received IV acetaminophen of 20 mg/kg and ramosetron of 0.3 mg at the start of skin closure. At the completion of surgery, sevoflurane was stopped, and fresh gas flow was increased to 5 L/min with 100% oxygen. Remifentanil was maintained at a concentration of 1.0 ng/mL. Sugammadex 2 mg/kg was given to achieve a train-of-four ratio >0.9. Extubation was carried out by a single anesthesiologist after eye-opening upon verbal command and confirming adequate spontaneous ventilation. Remifentanil was immediately discontinued. After vital signs were monitored for 3 min, the patient was transferred to the Post-Anesthesia Care Unit (PACU).

DEFINITIONS AND MEASUREMENTS:

The primary outcome was the incidence and severity of cough around ET extubation. The secondary outcome included respiratory complications, time to extubation, and throat and overall pain score. Emergence cough was assessed at 3 time points: before extubation, at extubation, and after extubation, using a 4-point scale (cough grade: 0=no cough, 1 =a single cough, 2=more than 1 episode of non-sustained cough, and 3=sustained and repetitive cough). Cough was defined as having clinical meaning for grades 2 and 3. Severe cough was defined as only grade 3.

Preoperative data (age, weight, height, BMI, smoking history, and past medical history) were collected through electronic medical records and interviews. The number of intubation attempts, the end-tidal concentration of sevoflurane at the time of eye-opening upon verbal command without stimulation, operation time, and anesthesia time were also collected. Time to extubation was defined as the time from sevoflurane stop to extubation. HR and MAP were recorded at 5 time points: baseline (before anesthesia induction, T0), at the end of surgery (T1), immediately (T2), 3 min (T3) after extubation, and at the PACU (T4). Respiratory complications such as bradypnea (respiratory rate <8 times/min), laryngospasm, and desaturation (SpO₂ <95% despite oxygen supplementation) were assessed. Immediately on arrival in the PACU, sedation levels were assessed using the Ramsay Sedation Scale (6 levels: 1=anxious and agitated or restless or both; 6=no response to a light glabellar tap or auditory stimulus). Nausea, vomiting, and overall and throat pain using the numeric rating scale (0=no pain, 10=worst pain) were evaluated. A fentanyl dose of 0.5 mcg/kg was administered in the event of pain score ≥5.

STATISTICAL ANALYSIS:

The sample size was calculated based on the incidence of emergence cough as the primary outcome. By referring to the findings of a previous study [21], the incidence of cough was 63% when extubation was achieved at remifentanil infusion of concentration 1.0 ng/mL after general anesthesia using sevoflurane and remifentanil. We considered a difference in the incidence of cough >50% as clinically relevant. Based on a significance level of 5% and statistical power of 80%, each group required 36 subjects to compensate for dropouts. Finally, a total of 72 patients was enrolled.

All statistical analyses were performed using SPSS (version 25.0, IBM Corporation, Armonk, NY, USA). Statistical significance was defined as P value <0.05. Continuous variables were expressed as the mean±standard deviation or median (interquartile range). Normality was assessed using the Kolmogorov-Smirnov test. If data followed a normal distribution, independent t tests were used for analysis. If they did not meet the assumption of normality, the Mann-Whitney U test was employed. Categorical variables were compared using the chi-square test or Fisher’s exact test and were presented as numbers (percentages). Repeatedly measured variables such as HR and MAP were analyzed using linear mixed models and were expressed as mean±standard error.

Results

COUGH PROFILES:

Around the ET extubation, the incidences of cough were up to approximately 90% despite the exclusion of grade 1 cough, and the incidences of severe cough also were over 50% in both groups (Table 2). Consequently, the incidences of cough and severe cough were not different between the 2 groups. However, the severity of cough was significantly lower in the head elevation group compared to the control group before extubation (cough scale 0/5/8/23 in the control group vs 1/2/17/15 in the head elevation group, P=0.039), although it was similar at extubation and after extubation.

EMERGENCE AND RECOVERY PROFILES:

During anesthesia emergence, time to extubation did not differ between the 2 groups (Table 3). Of the complications, desaturation after extubation was observed in 1 patient in the control group but resolved with encouragement. In the recovery room, sedation status upon arrival was observed within the range of awake levels (sedation scores 1, 2, and 3) except for 2 patients in the head elevation group (sedation scores 4). Nausea and vomiting, overall pain, throat pain, and the number of patients receiving fentanyl as rescue were similar between groups.

HEMODYNAMICS:

HR and MAP were comparable between the 2 groups during the perioperative period (Pgroup*time=0.621 and Pgroup*time=0.901 for HR and MAP, respectively) (Figure 2).

Discussion

In this prospective, randomized controlled study, we found that head elevation using a pillow (neck flexion angle of 35°) significantly reduced the severity of cough just before ET extubation during emergence from general anesthesia, as compared with neutral position. However, it did not reduce the incidences of cough or severe cough.

Although the underlying mechanism of cough remains uncertain, it is widely recognized that various signals transmitted through accepting receptors (nociceptor and mechanoreceptor) around the airway mucosa are relayed by vagal afferent nerves to the brainstem of the central nervous system, where the cough reflex is generated and regulated [10,22]. Anesthetic agents and opioids depress the cough evocation at the brainstem, whereas peripherally acting agents (eg, local anesthetics) block nociceptor and mechanoreceptor activation to reduce the cough [10]. Importantly, mechanoreceptors positioned beneath the epithelium proximally in the large airways are relatively insensitive to most chemical mediators (except local anesthetics) but are exquisitely sensitive to stimuli delivered to the mucosal surface [10]. In the intubated condition, cough can be triggered by several factors, such as airway secretion, inhalation agents, and an ET tube. Regarding mechanoreceptors, the ET tube exerts direct pressure and mechanical stimulation on the airway but provokes cough when the cough-inhibitory effects of anesthesia disappear at the brainstem during emergence.

The ET tube is an essential device for safety in mechanically ventilated patients in acute care or undergoing general anesthesia. The standard ET tube has a curvature called “the Magill curve”, with an approximate 140±20 mm radius [13]. This standard tube does not correspond to normal anatomy, in which the passage from the laryngeal inlet to the upper trachea forms a slight S-shaped curve [23,24]. The natural curve of the ET tube intends to align vertically with respect to the patient. Therefore, ET tubes deviating from airway anatomy can exert an elastic load on the pharynx and posterior larynx due to the lack of compliance [11,12].

Medical devices including an ET tube impose pressure on the surrounding tissue, and the subsequent mucocutaneous injuries generally mimic the device shape and size [25]. An ET tube is the most common device causing medical device-related pressure injuries, accounting for an estimated 7.5% in all care settings and up to 28% in ICU settings [26]. Although the intensity of pressure injury is less than the ET tube fixation, the ET tube itself can press against mucosal tissue, with intensity reported as grade 1 and 2 for 4-point device-related pressure ulcer grading [14]. In addition, the presence of an ET tube can cause pressure at all sites with tissue-endotracheal interaction (eg, the oral cavity, oropharynx, larynx, and trachea). Of these, the most frequently reported sites associated with ET tube-related pressure injury were the lips and tongue [25]. Many studies have focused on ET tube-related pressure injury affecting the skin and external soft tissue because internal airway injury lacks pressure measurement for grading [27,28]. However, internal airway pressure is also important, considering that subglottic and trachea endoluminal pressure consequently leads to tracheal stenosis [15]. Therefore, changes in head and neck position for fitting the airway passage to the ET tube’s configuration as much as possible might be essential to minimize the airway wall pressure as an extrinsic factor triggering a cough.

Since Magill first described the “sniffing position” as an elevation of the occiput and extension of the head, Bannister et al reformed the laryngoscopy positioning by suggesting the need to align the oral, pharyngeal, and laryngeal axes, called the “3-axis alignment theory” [17]. Horton et al further progressed this concept by establishing the ideal angles as 15º for upper cervical flexion and 35° for lower cervical extension [29]. Based on this theory, head elevation improves the alignment of the 3 axes compared to a neutral position. This was confirmed in magnetic resonance imaging (MRI) studies of the head and neck region [12,30]. Greenland et al acquired MRI sagittal slices from 10 awake adult volunteers and found that the nasopharynx was below the glottis in the neutral position but above the glottis in the sniffing position [30]. Later, they employed a “2-curve theory” to explain airway configuration, in which tangent was assessed at the point where the primary and secondary curves meet in the S-shaped airway [12]. As a result, head elevation caused an anticlockwise rotation of the tangent, meaning that a curved ET tube was more likely to pass from the airway into the trachea with less impact on the airway wall in the head-elevated position than in the neutral position [12]. Thus, the head elevation position seemed to cause less pressure on the airway wall than a neutral position in the intubated state. It might be helpful that crucial radiological imaging of the patients in the 2 head positions was achieved in the surgical room.

We chose head elevation, not the sniffing position, to compare with the neutral position in our study. After the formation and acceptance of the 3-axis alignment theory, its correctness has been under debate. It has been argued that the sniffing position does not bring the 3 axes into alignment [18,31]. In 2015, Orbany et al evaluated the laryngeal view in head extension without head elevation, 6 cm head-elevated, and 10 cm head-elevated sniffing positions [32]. Sniffing positions improved the laryngoscopic view, but a more head-elevated sniffing position further improved the view. The mechanism is assumed to be that head elevation (neck flexion) moves the glottis posteriorly, thus compensating for anterior displacement of the glottis during head extension and possibly resulting in better alignment of the 3-axis [19]. Therefore, if the same pillow height is applied to each position in our study, the head-elevated position might be more in accord with the ET tube configuration than the sniffing position and consequently lower the direct pressure between the ET tube and the airway wall. Further studies assessing the effect of the sniffing position on emergence cough are needed.

Another possible explanation underlying the reduction in emergence cough associated with head elevation could be related to tracheal accordion. In our study, all patients in both groups were intubated in a head-elevated position on a pillow, and insertion of the ET tube was stopped as soon as the vocal cord indicator passed the laryngoscope vocal cords. While the head elevation group was maintained in this position, the control group was changed to the neutral position by removing the pillow. It is well documented that neck flexion or extension can either decrease or increase the tracheal length [33]; therefore, the change in head position in the control group might lead to neck extension and trachea lengthening, displacing the endotracheal tube cuff cephalad towards the vocal cords in our study. Given that the most sensitive regions of the airways for mechanically triggering cough are the larynx and the tracheal and bronchial bifurcation, stimuli in the mechanosensitive receptors located more proximally to the larynx in the control group seem to act in provoking cough to some degree [10,34].

Our findings are particularly important with respect to presenting the clinical implication that a simple manipulation of the head position to align the physiological structures of the larynx and pharynx with the ET tube’s configuration, not pharmacologic interventions (eg, opioids), could reduce emergence cough in intubated patients. Another strength of this study lies in the single-sex (male) enrollment, single surgery with short operation time, and the additional effect of lowering airway edema because pressure forces are associated with local edema [25].

There were a few limitations to the methods in our study. First, blinding of the extubator was not practically possible. Second, tracheal diameter was associated with height in men’s anatomic studies [35]. ET tube size was best determined by each patient’s height, instead of using a uniform size of 7.5 mm. Third, the validity of the sniffing position on 3-axis alignment was significant in obese but not in non-obese patients [18]. Thus, the effect of head elevation on emergence cough should be evaluated in obese patients. Finally, target-controlled infusion of remifentanil was maintained at a concentration of 1.0 ng/mL during anesthesia emergence. Because the remifentanil concentration used to suppress emergence cough in a prior study enrolling male patients was approximately 2.5 ng/mL, it might have led to the high incidences of coughing, at up to approximately 90%, and consequently failed to reduce the incidence [36].

Conclusions

Head elevation using a pillow significantly reduced the severity of cough before ET extubation during anesthesia emergence in male patients.