A randomized control study comparing the pharyngolaryngeal morbidity of laryngeal mask airway versus endotracheal tube

Abstract

Introduction:

Endotracheal tube (ETT) has been associated with various pharyngolaryngeal morbidities (PLMs) following general anesthesia (GA). Laryngeal mask airway (LMA), currently the most commonly used supraglottic airway device, has several advantageous over the ETT but has been associated with varying results of PLM. The aim of our study was to compare the PLM between them and to know whether LMA is a better alternative.

Materials and Methods:

One hundred and seventy American Society of Anesthesiologists Grades 1 and 2 women scheduled for elective mastectomy were included in the study, 85 each in either group, E Group (intubated with ETT) and L Group (using LMA) on a random basis. All patients received GA with controlled ventilation using a muscle relaxant. PLMs such as hoarseness, pain on phonation, sore throat, and difficulty in swallowing were documented by an interview done postoperatively. Peroperative parameters such as intubation attempts, trauma during airway device insertion, and intraoperative incidents were also analyzed. A sample size of 85 patients in each group was calculated in order to achieve a study power of 0.8 and alpha level was taken as 0.05. Data were analyzed using SPSS version 16 using Chi-square test, Mann–Whitney U-test and Fisher's exact test were used as nonparametric tests. A two-tailed P< 0.05 was considered significant.

Results:

Patients in E Group had statistically significant increased incidence of a sore throat and voice complaints whereas L Group showed a statistically significant increase of swallowing problems. There was also a significant correlation between traumatic insertion and sore throat, pain on swallowing in the L Group, which could be due to direct trauma.

Conclusions:

ETT was associated with an increased incidence of voice problems and sore throat whereas LMA had an increased incidence of dysphagia and odynophagia. Use of LMA changes the pharyngolaryngeal profile to a more acceptable one.

INTRODUCTION

Injuries to the airway are well-recognized complications of general anesthesia (GA). Claims to airway injuries are frequent in American Society of Anaesthesiologists (ASA) closed claims database in which injuries to the larynx represent about 33%, most of them relating to long lasting pain or voice complaints due to nerve apraxia and cartilage trauma.[1] Sore throat had been rated as the eighth most adverse effect in the postoperative period following GA which can contribute to postoperative morbidity and patient dissatisfaction.[2]

Several risk factors for postoperative sore throat due to an endotracheal tube (ETT) have been identified which include female gender, age, surgical site, use of nitrous oxide, and duration of surgery.[3] Laryngeal mask airway (LMA), the most popular supraglottic airway device, is said to have a lower incidence of postoperative sore throat and voice problems though wide variations in their incidence are reported.[4] In pediatric age group, the incidence of a sore throat between endotracheal intubation and LMA are not significant.[5] In this study, we compared the whole spectrum of pharyngolaryngeal morbidity (PLM) such as hoarseness, pain on phonation, dysphagia, odynophagia, and sore throat between the LMA and ETT after providing adequate anesthetic depth at the time of intubation, avoiding nitrous oxide, and duration of surgery restricting to 2 h which are significant predictors of postoperative sore throat.[3] In addition, we compared the incidence of various perioperative events and their significance on PLM.

MATERIALS AND METHODS

After obtaining approval from the Institutional Review Board and Ethics Committee a randomized cohort trial involving 170 patients undergoing elective mastectomy were allocated to either Group E (intubated with ETT) or Group L (LMA classic as the airway device) using a computer-generated random table.

After a written informed consent, 170 women of ASA status 1 and 2 in the age group 18–60 years scheduled for elective mastectomy lasting 1–2 h were included in our study. Exclusion criteria were difficult airway with Mallampatti (MP) Class 3 and 4, history of nausea, vomiting, or a sore throat within 24 h of surgery, obesity, and gastroesophageal reflux. Those with preoperative voice and swallowing problems and requiring more than three attempts for airway device placement were also excluded.

All patients were premedicated orally with alprazolam 0.5 mg on the night before and about 2 h before induction. In the premedication room baseline parameters such as heart rate, blood pressure, electrocardiogram, and oxygen saturation were recorded, intravenous (I/V) fluid started and midazolam 1 mg, fentanyl 2 µg/kg were given before induction. After preoxygenation, I/V lidocaine 1.5 mg/kg was given and anesthesia induced with propofol 2.0 mg/kg and muscle relaxation obtained with 0.1 mg/kg vecuronium. Patients were ventilated with 35% oxygen in air and isoflurane dial concentration adjusted to achieve a minimum alveolar concentration of 1 as per the available gas monitoring to ensure adequate depth throughout the procedure. Trachea intubated with appropriate sized cuffed oral ETT and confirmed with end-tidal carbon dioxide and auscultation for bilateral air entry. Cuff was inflated just to prevent air leak and tube was secured. LMA insertion involved preparation of the device (cuff deflated and well lubricated with KY jelly) and blind digital insertion performed and position confirmed as above. Cuff was inflated as per the instructions of the manufacturer and the device secured. A maximum of three attempts to position the device were allowed, before proceeding to alternate method of airway management and such cases were recorded and excluded from the study. Anesthesia was maintained with air-oxygen mixture with inspired oxygen of 35%. A perioperative data form containing intubation data and perioperative incidents were also filled which documents the number of attempts or any trauma to oropharynx during intubation/LMA insertion. More than one attempt was taken as a “perioperative incident.” Any intraoperative untoward incident such as abnormal rise in airway pressure, bucking, coming out of anesthesia during the procedure, and extubation time events were also noted.

The traumatic/atraumatic criteria was based on whether there was any blood on the ETT/LMA at the time of extubation and/or any visible damage to the oropharynx after extubation which includes trauma to the lips, teeth, or gum in case of endotracheal intubation. I/V diclofenac sodium/tramadol were given for postoperative pain relief. At the end of the surgery neuromuscular blockade was reversed, patient awakened, the airway device removed, and the patient shifted to the recovery room.

All patients underwent a structured interview at 12, 24, and 36 h postoperatively which was conducted by an anesthesia resident who did not know the nature of airway management device used. Patients were asked about sore throat (constant pain independent of swallowing), hoarseness, and or pain during phonation (as voice complaints) and about dysphagia/odynophagia (difficulty/pain on swallowing). Severity of pain due to sore throat and odynophagia were recorded in the visual analog pain scale (VAS).

Mann–Whitney U-test statistic was used when the variables were ordinal/interval, Fisher's exact test was used for small sample sizes and goodness of fit to the data was tested using Chi-square (χ²) test. A two-tailed P < 0.05 was considered significant. The statistical analysis was done using SPSS version 16 (SPSS, Inc., Chicago, IL).

RESULTS

Out of the 170 patients originally enrolled in the study, 12 patients could not be followed up (4 in the L Group and 8 in the ETT Group) as they were discharged before the interview could be completed. Hence, the total study population was only 158. Score on the VAS scale was categorized to mild (1–2); moderate (3–5) and severe (>6). Both the groups were comparable in terms of age, weight, duration of anesthesia, and anesthetic management.

Age distribution between the two airways management techniques were closely matched (P = 0.749). Regarding voice complaints, patients were sorted into three categories (no voice complaints; hoarseness only and pain during speech). At 12 h significantly more patients (P < 0.01) who were intubated with the ETT (21 out of 77) had a hoarse voice than those in the L (14 out of 81) group (27.3 vs. 17.3%). About 9.0% of the E Group (7 out of 77) had pain during phonation as compared to none for the L Group. After 24 h complains of hoarseness in the E Group decreased to 14.3% (11 out of 77) and pain on phonation decreased to 2.6% (2 out of 77) while 8.6% of the patients (7 out of 81) in the L Group had hoarseness, none had pain on phonation. However, this difference was not significant (P > 0.05). After 36 h 10.4% of the E Group (8 out of 77) had hoarseness whereas only 1.2% (1 out of 81) of the LMA Group had (P < 0.05). None had pain on phonation at 36 h [Table 1].

Table 1.

Comparison of hoarseness/pain on speaking based on group

Pain on swallowing is another distressing sequel of airway management. A greater number of patients who were in the L Group (27 out of 81) had pain during swallowing at 12 h (mild pain 18.5%, moderate pain 14.8%) compared to those (24 out of 77) in the E Group (mild pain 31.2%, none had moderate pain). However, this difference was not found to be significant (P > 0.05). At 24 h patients in the L Group (22 out of 81) had significantly (P < 0.01) more pain (mild 22.2%; moderate 4.9%) than in the E (6 out of 77) Group (mild 7.8%). This difference (P < 0.01) continued up to 36 h; L Group (mild pain 16%; E Group no pain reported). Though 31% of patients in the E Group (almost same as L Group) had pain on swallowing at 12 h, it reduced significantly in the next 24 h and no pain was reported at 36 h [Table 2].

Table 2.

Comparison of pain on swallowing based on group

At 12 h complaints of a sore throat were significantly more in the E Group (47 out of 77) than in the L (29 out of 81) Group (61.1% vs. 35.8%, P < 0.01). The severity was also more in the E Group (moderate pain 13% vs. 4.9%). However, at 24 and 36 h there were no significant differences between the two groups although a larger percentage of the E Group had a sore throat (35.15% vs. 23.5% and 22.1% vs. 14.8%, respectively) [Table 3].

Table 3.

Comparison of a sore throat based on group

“Perioperative events” is a broad-based term into which all the perioperative factors that could play a role in PLM have been placed. The main role of this variable was to filter out confounding factors. Comparison of perioperative events based on the group does not show any significant difference with P = 0.511 (16.9% vs. 21% and 83.1% vs. 79%) but trauma during intubation/LMA insertion showed a significant difference. Only 16.9% of patients in the E Group showed blood on cuff compared to 38.3% in the L Group [P = 0.003, Table 4]. The significance of perioperative events was checked only with the incidence of the complaints at 12 h. Fisher's exact test was used as there were cells in the 2 × 2 contingency table which had less number of observations precluding the use of χ² test. As per the study, there was no significant impact of various intraoperative incidents on the outcome variables except in the complaints of a sore throat in the E Group where the P = 0.0123 [Table 5]. There was also a significant correlation between traumatic insertion and sore throat (P < 0.001), pain on swallowing (P < 0.003) in the L Group, which could be due to direct trauma [Tables 6 and 7].

Table 4.

Comparison of trauma during intubation/insertion based on group

Table 5.

Comparison of perioperative incidents on the incidence of a sore throat in the endotracheal tube group at 12 h

Table 6.

Comparison of traumatic insertion on the incidence of pain on swallowing in the laryngeal mask airway group at 12 h

Table 7.

Comparison of traumatic insertion on the incidence of a sore throat in the laryngeal mask airway group at 12 h

DISCUSSION

A sore throat associated with intubation can be attributed to multiple factors such as tube size,[6] sex,[3,7] type of surgeries,[3] cuff design,[8] use of local anesthetic, and lubrication for the cuff.[9] Our incidence of a sore throat (61.1% at 12 h) for the ETT was higher than the report of 40% (n = 809) incidence by Biro et al.[10] but less than Radu et al.[11] where an incidence of 74%, but at 6 h. The greater and more significant incidence of a sore throat in the E Group in our study can be because of the placement of ETT to more sensitive tissues of the larynx and the trachea which causes sensitization to pain, thus contributing to a postoperative sore throat.[12] Christensen et al.[7] in their study reported a lower incidence of a sore throat (14.4%) whereas Ogata et al.[13] found an incidence of a sore throat to be 55% and 45%, 4 and 24 h after surgery.

Hoarseness is a common complication after endotracheal intubation. Yamanaka et al.[14] observed an incidence of 49% on the day of surgery, 29%, 11%, and 0.8% on 1, 3, and 7 postoperative days, respectively. The most common cause for postoperative hoarseness in patients undergoing procedures that do not involve the head and neck is swelling of the vocal cords.[15] Vocal cord paralysis and paresis have been reported after tracheal intubation despite the intubation being atraumatic and the site of the surgery remote from the head and neck.[16,17] The incidence of vocal cord granuloma the following intubation is reported to be between 1 in 800 and 1 in 20,000.[18] Arytenoid dislocation which is a rare complication following tracheal intubation that will result in prolonged hoarseness of voice has also been reported.[14]

A sore throat due to LMA can be multifactorial, and its incidence can be influenced by the depth of anesthesia at the moment of insertion, method of insertion, the number of attempts, and the presence of humidifier moisture exchanger in the circuit.[19] The incidence of a sore throat with the use of LMA varies widely and may be attributable to different methodologies.[8,20,21] Several studies have evaluated insertion techniques, lubricants and the effect of cuff pressure limitation in terms of ease of insertion of the LMA, pharyngeal trauma, and postoperative throat symptoms.[22,23] However, what was surprising in our study was a higher incidence of a sore throat in L Group (35.8%) as they underwent controlled ventilation, which might require higher sealing pressures for the cuff as compared to spontaneously breathing patients, thus resulting in more mucosal ischemic injury. Some authors[19] found no relationship between cuff pressure and PLM, whereas others[24,25] came to the conclusion that minimizing cuff pressures led to decreased sore throat.

In a meta-analysis of randomized control trials[26] comparing LMA with ETT, less incidence of voice problems with LMA was found which can be of advantage to professional voice users. LMA can cause transient changes in vocal cord function.[27] A total of 14 nerve injuries has been reported until 2004[28,29] following the use of LMA namely recurrent (7), hypoglossal (5) and lingual (2). In all these cases size, 3 and 4 LMA's were used. It should be pointed out that, when perfectly positioned, the LMA lies with the tip resting against the upper esophageal sphincter, with the sides facing the pyriform fossae.

This study was designed mainly for the presence or absence of a specific complaint rather than their severity. For having uniformity in the duration of the airway device placement, we confined our study population to mastectomies which can be considered as a limitation. A much-generalized study including male sex and pediatric population with varying duration of the procedure is required to further substantiate our findings. Another limitation of this study was it lacked enough power to estimate the more severe voice complaints, as it will require a much larger sample size. In our study, there was a significant incidence of trauma following LMA insertion even with adequate relaxation and anesthetic depth. It can, therefore, be logically inferred that trauma is likely to be greater when adequate relaxation and anesthetic depth are not provided. To reduce trauma, perhaps MP classification can also be considered while selecting the size of LMA.

CONCLUSIONS

Both ETT and LMA can cause significant PLM following GA which can affect patient compliance and quality of care. Though the incidence of dysphagia and odynophagia were found to be more with LMA, hoarseness, and sore throat which are major determinants for prolonged hospital stay and patient discomfort were found to be much reduced in our study when compared with ETT. Reassurance and symptomatic management for swallowing problems are all that is required to reduce patient's anxiety. Definitely LMA is a better alternative when used judiciously, gently, and under adequate anesthetic depth to reduce the incidence of PLM further. Cuff pressure monitoring should be recommended as a standard practice to reduce adverse outcomes of PLM.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest