Mehanna R, Hennessy A, Mannion S, O’Leary G, Sheahan P. Effect of Endotracheal Tube Size on Vocal Outcomes After ThyroidectomyA Randomized Clinical Trial. JAMA Otolaryngol Head Neck Surg. 2015;141(8):690-695. doi:10.1001/jamaoto.2015.1198
The optimum size of endotracheal tube (ETT) for general anesthesia remains unresolved. Choice of ETT size may be of particular relevance to thyroid surgery because of the increased risk of laryngeal trauma and concerns regarding postoperative vocal outcomes.
To test our hypothesis that intubation with a smaller ETT would lead to reduced postoperative vocal impairment and associated reduced laryngoscopic evidence of laryngeal trauma compared with intubation with a standard-size ETT.
Design, Setting, and Participants
This double-blind randomized clinical trial studied patients 18 years and older undergoing elective thyroidectomy at an academic teaching hospital from October 15, 2012, through June 13, 2013.
Patients were randomized to group 1 (standard-size ETT, 8.0 mm for men and 7.5 mm for women; n = 24) or group 2 (small ETT, 7.0 mm for men and 6.5 mm for women; n = 25). Patients were assessed preoperatively and at 24 hours and 3 weeks postoperatively.
Main Outcomes and Measures
Fiberoptic videolaryngoscopy with modified scoring system, voice assessment using the GRBAS (grade, roughness, breathiness, asthenia, strain) rating scale, vocal self-assessment using the 30-item Voice Handicap Index, and subjective pain score.
At 24 hours, no significant differences were found between patients in groups 1 and 2 in change in GRBAS scores, change in laryngoscopic score (1.71 vs 1.76, P = .90), or postoperative pain score (3.3 vs 3.2, P = .91). At 3 weeks, no significant differences were found in changes in the 30-item Voice Handicap Index score (−2.2 vs −1.3, P = .74), GRBAS scores, or laryngoscopic score (0.25 vs 0.16, P = .67).
Conclusions and Relevance
We did not find evidence that smaller ETT size for thyroidectomy has any significant effect on postoperative vocal outcomes, incidence of laryngeal trauma as assessed by laryngoscopy, or pain scores. However, because of the small sample size, our study may have been underpowered to detect small differences.
clinicaltrials.gov Identifier: NCT02136459
The optimum size of endotracheal tube (ETT) for general anesthesia remains a matter of debate.1 In 1930, Magill recommended the use of the “largest endotracheal tube which the larynx can comfortably accommodate.”2(p 819) More recently, there has been a gathering momentum toward selecting ETTs of decreasing diameter. This shift has been driven by a perceived impression of reduced laryngeal morbidity, better preservation of voice quality, and decreased incidence of sore throat with smaller-diameter ETTs, although strong objective evidence to support this impression is somewhat limited.1,3- 5
Despite a low incidence of major complications, thyroid surgery may be associated with postoperative laryngeal symptoms, including vocal changes in the absence of impaired vocal cord mobility, and sore throat.6 Patients undergoing thyroidectomy may be at particularly increased risk of postoperative laryngeal morbidity compared with other operations because of the strong attachments of the thyroid gland to the trachea, such that intraoperative mobilization of the thyroid leads to an inevitable degree of movement of the trachea and larynx against the fixed ETT, with consequent risk of trauma to tracheal mucosa, vocal cords, and arytenoids. Thus, it would seem reasonable to assume that thyroidectomy is likely to be at increased risk of ETT-related laryngeal trauma than operations outside the head and neck region where there is no movement of the larynx.
Vocal impairment after thyroidectomy is an area of particular concern because of the operative risks of vocal cord palsy secondary to recurrent laryngeal nerve (RLN) injury.6,7 However, postthyroidectomy vocal impairment can occur in cases in which the RLN is preserved.6,8- 10
Laryngeal trauma due to endotracheal intubation is probably more common than appreciated; however, given that symptoms of sore throat and hoarseness after most operations are generally self-limiting, it is not usually a major cause of concern. On the other hand, postoperative vocal symptoms may be a major cause of anxiety after thyroidectomy. Therefore, if smaller ETT size truly confers better preservation of vocal function after endotracheal intubation, then choice of ETT size may be of particular pertinence during thyroid surgery. However, to date, no previously published study has attempted to correlate postthyroidectomy objective laryngoscopic findings or vocal symptoms with ETT size.
The hypothesis of the present study was that intubation with the use of a smaller ETT would lead to less trauma to the larynx during thyroidectomy than intubation with the use of a standard-size tube, which would be reflected through less objective laryngeal alterations as assessed by fiberoptic laryngoscopic examination and vocal assessment. The rationale for this hypothesis was that a smaller tube would be less unyielding when the larynx is moved against it and thus would lead to less laryngeal injury. To address this question, we designed a randomized clinical trial that compared the effect of use of a smaller ETT with that of ETTs of the standard size used in our department.
The present study took place at the South Infirmary Victoria University Hospital, an academic teaching hospital, from October 15, 2012, through June 13, 2013. Ethical approval was obtained a priori from the Cork Clinical Research Ethics Committee. Written consent was provided by all patients.
Eligibility included patients 18 years and older scheduled to undergo thyroid surgery by 1 of 2 surgeons (P.S. or G.O.). A priori exclusion criteria were preexisting vocal palsy or major vocal cord disease, invasive thyroid cancers, and concomitant lateral neck dissection. Randomized patients who underwent intraoperative sacrifice of the RLN were also excluded. After commencement of the trial, it was decided to extend exclusion criteria to patients with significant pulmonary comorbidity after one such patient with significant pulmonary comorbidity, who was randomized to a smaller tube size, had high airway pressures and required a high fraction of inspired oxygen to maintain adequate oxygen saturation levels, which improved after changing to a larger ETT.
The study was an intent-to-treat, double-blind, randomized clinical trial. Participants were recruited from patients admitted electively for thyroidectomy on the evening before surgery. During this study period, the general policy was for patients scheduled for thyroidectomy to be admitted on the evening before surgery provided a bed was available. Participants were randomized into 1 of 2 groups. Group 1 underwent endotracheal intubation for thyroid surgery with a standard-size ETT (8 mm for men and 7.5 mm for women). Group 2 underwent endotracheal intubation with a smaller ETT (7 mm for men and 6.5 mm for women). Because of the larger trachea in men, male patients were intubated with a tube 0.5 mm larger than women. Randomization was performed using a system of random sequence of group assignment within a sealed envelope that was opened by the anesthetist just before induction of anesthesia. Preoperative evaluation and postoperative assessment were performed by a single investigator (R.M.) who was masked to the ETT size used and intraoperative data.
Preoperative evaluation was performed on the evening before surgery and comprised a detailed history along with completion of the 30-item Voice Handicap Index (VHI-30),11 voice assessment using the GRBAS (grade, roughness, breathiness, asthenia, strain) rating scale, and fiberoptic videolaryngoscopy, which was scored by assigning a score of 0 (absent), 1 (present), or 2 (severe) to each of the following findings: edema; erythema; polyp, nodule, or granuloma; and reduced movement (0 for normal movement, 1 for reduced movement, and 2 for absent movement).
Endotracheal intubation was performed by consultant anesthetists or senior registrars under consultant supervision using Mallinckrodt cuffed ETTs (Covidien). Intubation difficulty score,12 number of attempts at intubation, and use of a bougie were recorded. Once intubated, the ETT cuff was inflated to 25 cm H2O using a cuff manometer. Anesthesia was administered in a standardized fashion using an oxygen-sevoflurane mix. Nitrous oxide was not used because of the potential for diffusion to affect ETT cuff pressure. Patients received 1 to 2 µg/kg of fentanyl, 2 to 4 mg/kg of propofol, 0.1 mg/kg of vecuronium, 0.05 to 0.1 mg/kg of morphine, and 0.1 mg/kg of dexamethasone. Paracetamol, diclofenac, and ondansetron hydrochloride were given unless contraindicated.
Surgery was performed by 1 of 2 experienced surgeons (P.S. or G.O.) using a standard technique.13 For most cases, the sternohyoid muscle was retracted and the sternothyroid muscle divided. The external branch of the superior laryngeal nerve was not specifically sought, whereas the RLN was always sought. Neural monitoring was not used during the study period. At the conclusion of the operation, a data sheet was filled out indicating the type of surgery performed, intraoperative findings, and length of surgery.
The first postoperative assessment was performed on the first postoperative day with the following data recorded: subjective pain score (1 to 10 using a visual analog scale), GRBAS score, and videolaryngoscopic score. The second postoperative assessment was performed 3 weeks after surgery with the following data recorded: completion of VHI-30 questionnaire, GRBAS score, and videolaryngoscopic score.
Statistical analysis was performed using XLSTAT statistical software, version 2013.5.05 (Addinsoft). Normally, distributed data were analyzed using a paired t test for paired data or an unpaired t test for unpaired data. Nonnormally distributed data were analyzed using a Wilcoxon ranked pairs test for paired data or a Mann-Whitney test for unpaired data. A Fisher exact test was used on 2 × 2 contingency tables.
During the period of the study, 80 patients underwent thyroidectomy. Six were not eligible for inclusion because of invasive thyroid cancer and/or performance of concomitant neck dissection (5 patients) or age younger than 18 years (1 patient). Of the remaining 74 patients, 57 were recruited. Eight patients were subsequently excluded: 6 because of nonadherence with the randomization protocol by the anesthetist (randomization envelope not opened or wrong tube type used), 1 after randomization because of the presence of a high-oxygen requirement and the decision of the anesthesiologist to reintubate with a larger ETT, and 1 because of the intraoperative sacrifice of the RLN (Figure).
The final population thus consisted of 49 patients (43 women), with a mean age of 53 years (age range, 29-82 years). Twenty-six underwent total thyroidectomy, and 23 underwent thyroid lobectomy. Final histologic analysis revealed carcinoma (6 patients), microcarcinoma (5 patients), Graves disease (2 patients), and other benign condition (36 patients).
Twenty-four patients were randomized into group 1 (large ETT; 8 mm for men and 7.5 mm for women) and 25 into group 2 (small ETT; 7 mm for men and 6.5 mm for women). The characteristics of the patients in groups 1 and 2 are given in Table 1.
As rated by the intubation difficulty scale, in group 1, there were 21 easy intubations and 3 slightly difficult intubations. In group 2, there were 19 easy intubations, 5 slightly difficult intubations, and 1 moderately difficult intubation. Bougies were used in 2 patients in each group.
The mean (SD) preoperative GRBAS scores were as follows: grade, 0.37 (0.49); roughness, 0.41 (0.57); breathiness, 0.06 (0.24); asthenia, 0.06 (0.24); and strain, 0.18 (0.39). On the first postoperative day, scores were as follows: grade, 0.46 (0.58); roughness, 0.63 (0.63); breathiness, 0.02 (0.14); asthenia, 0.06 (0.24); and strain, 0.21 (0.41). None of the changes were significant (P = .33, P = .12, P = .42, P = .63, and P = .81, respectively). At 3 weeks after surgery, GRBAS scores were as follows: grade, 0.26 (0.44); roughness, 0.33 (0.47); breathiness, 0.02 (0.15); asthenia, 0.0 (0.0); and strain, 0.11 (0.32). No significant differences were found between GRBAS scores at baseline and at 3 weeks (P = .10, P = .24, P = .42, P = .15, and P = .23, respectively).
The mean preoperative VHI-30 score was 7.6 (10.5). At 3 weeks after surgery, this score decreased to 6.1 (8.4). The change was not significant (P = .17).
Before surgery, 6 patients had positive laryngoscopic findings (4 in group 1 and 2 in group 2). Five had mild erythema (4 involving arytenoids and 1 diffuse), and one had a nodule or granuloma. The mean preoperative laryngoscopic score was 0.1 (0.33).
On the first postoperative day, 41 patients had at least one positive laryngoscopic finding. Thirty-nine had erythema, 27 had edema, 2 had polyp or nodule, and 4 had reduced vocal movement. The mean postoperative day 1 laryngoscopic score was 1.9 (1.32), which was significantly greater than the preoperative score (P < .001).
At 3 weeks after surgery, positive laryngoscopic findings were present in 14 patients (erythema in 9, edema in 7, polyp or nodule in 1, and reduced vocal movement in 3). The mean (SD) 3-week postoperative laryngoscopic score was 0.5 (0.80), which was significantly less than the day 1 postoperative score (P < .001) but still significantly greater than the preoperative score (P = .04).
The 4 patients with reduced postoperative vocal mobility underwent further follow-up at 3 months. By this time point, all had regained fully normal vocal mobility. The incidence of transient RLN neurapraxia was thus 5.3% (4 of 75 nerves at risk).
The effect of tube size on the studied parameters is given in Table 2 (postoperative day 1) and Table 3 (3 weeks after surgery). No significant difference was found in mean (SD) pain scores on the first postoperative day between groups 1 and 2 (3.3 [2.1] vs 3.2 [1.9], P = .91) (Table 2). No significant differences were found between groups 1 and 2 in the changes in any of the components of GRBAS scores at the first postoperative day or 3 weeks after surgery (Table 2 and Table 3).
The mean (SD) VHI-30 score was 7.9 (9.1) in group 1 and 7.4 (11.9) in group 2. At 3 weeks after surgery, the mean VHI-30 scores were 5.9 (8.3) in group 1 and 6.3 (8.6) in group 2. The changes in VHI-30 scores were not significantly different between groups 1 and 2 (Table 3).
The mean (SD) total preoperative laryngoscopic score was 0.2 (0.4) in group 1 and 0.1 (0.3) in group 2 (P = .37). This score increased to 1.9 (1.4) in group 1 and 1.8 (1.2) in group 2 on the first postoperative day and then decreased to 0.5 (0.8) in group 1 and 0.3 (0.6) in group 2 at 3 weeks after surgery. No significant differences were found between groups 1 and 2 in the changes in laryngoscopic scores among any of the time points (Table 2 and Table 3).
The optimum size of ETT for general anesthesia remains unresolved. Larger ETTs have been associated with increased incidence of throat discomfort3,4 and hoarseness4 after nonthyroid surgery. In a randomized clinical trial, Jaensson et al3 reported that the 6.0-mm ETT caused less postoperative discomfort in women undergoing surgery than the 7.0-mm ETT, although there was no significant difference in postoperative hoarseness. The ETT size may be of particular relevance in thyroid surgery. de Pedro Netto et al9 reported a higher incidence of abnormal videolaryngoscopic findings, postoperative voice symptoms, and perturbations in acoustic analysis after thyroidectomy than after breast surgery. Hisham et al5 reported a higher incidence of postoperative sore throat after thyroidectomy with larger ETT size. However, to our knowledge, there are no previously published data regarding the effect of tube size on postoperative vocal outcomes after thyroidectomy.
The present study was a randomized clinical trial that compared objective and subjective laryngeal outcome measures between patients randomized to standard or small ETTs. The salient findings were as follows: (1) there was a high incidence of abnormal laryngoscopic findings immediately after surgery, but this incidence significantly decreased during 3 weeks; and (2) there were no significant differences between patients randomized to standard or small ETTs with respect to postoperative vocal outcomes, laryngoscopic signs of laryngeal trauma, or subjective symptoms. Thus, our findings do not support a significant effect of ETT size on postoperative vocal outcomes.
The high incidence of abnormal laryngoscopic findings immediately after surgery in both groups is worth mentioning. The incidence of abnormal findings at postoperative day 1 in the present study (84%) was higher than that reported by de Pedro Netto et al,9 who reported laryngoscopic alterations in 28% of patients 2 weeks after thyroidectomy, and Echternach et al,10 who reported laryngeal complications in 31% of patients 3 to 4 days after surgery. The reason for the higher incidence of abnormal laryngoscopic findings in our study was probably because the first postoperative assessment was performed much sooner (1 day) after surgery. On the other hand, the incidence of abnormal laryngoscopic findings had significantly reduced to 29% (14 of 49 patients) by the second postoperative assessment (3 weeks after surgery), which is similar to the incidence reported by de Pedro Netto et al9 and Echternach et al.10 By comparison, for nonthyroid operations, the incidence of abnormal laryngoscopic findings in patients undergoing laryngoscopy within 24 hours of anesthesia ranges from 8% to 42%14,15; however, other authors16,17 have reported an incidence of abnormal findings as high as 69% to 73%.
There was considerable variability in VHI-30 scores in patients in the present study. The reasons for this variability are not clear but may include the effect of goiter on subjective assessment of voice or concerns about the same in the case of some patients.
It is likely that RLN injury and ETT-related trauma are not the only causes of suboptimal vocal sequelae and other symptoms after thyroidectomy. Other possible mechanisms include laryngeal edema, injury to the superior laryngeal nerve, division of strap muscles,18 and postoperative fibrosis. Thus, despite best efforts to protect the RLN and minimize laryngeal trauma, avoidance of poor vocal outcomes after thyroidectomy cannot be guaranteed.
The major weaknesses of the present study were the relatively small study cohort and the necessary exclusion of 6 patients who were recruited to the study but excluded because of nonadherence to the randomization protocol. With regard to the study size, we were unable to perform a reliable a priori power study without pilot data. However, a post hoc power study calculated that the number of patients who would have been required in each group to find a significant difference would have been 1708 (for VHI-30 scores at 3 weeks), 12 312 (for laryngoscopic scores on day 1), or 1063 (for laryngoscopic scores at 3 weeks). On the other hand, it would appear that the differences between the groups in each of the parameters studied were genuinely so minimal that it is probably unlikely that a larger study would have found any significant difference. Nevertheless, small study size is always a concern with a negative study result. Further weaknesses were the performance of laryngoscopic and GRBAS assessment by a single assessor and lack of data regarding interrater or intrarater reliability, as well as lack of masking regarding preoperative vs postoperative status. However, given that this assessor was masked to how participants were randomized, we believe that any bias due to having only a single assessor is unlikely to have affected the outcome of the comparison between standard and small ETTs. Finally, our study does not exclude the possibility that use of an even smaller ETT in group 2 may have yielded different results and the possibility of a difference in main outcome measures between the groups.
On the other hand, strengths of the study include the prospective and randomized design; double-blinding, with the patients and investigators performing the postoperative assessments being masked to the size of ETT used; and inclusion of both patient- and physician-rated evaluation of vocal function and laryngeal morbidity on all patients.
Abnormal laryngoscopic findings are common in the immediate postoperative period in patients undergoing thyroidectomy. These abnormal findings mostly resolve within 3 weeks. We did not find any evidence that smaller ETT size has any effect on laryngoscopic findings, vocal outcomes, or postoperative pain scores.
Submitted for Publication: February 7, 2015; final revision received May 3, 2015; accepted May 7, 2015.
Corresponding Author: Patrick Sheahan, MB, MD, FRCSI (ORL-HNS). Department of Otolaryngology–Head & Neck Surgery, South Infirmary Victoria University Hospital, Cork, Ireland (Sheahan.firstname.lastname@example.org).
Published Online: July 9, 2015. doi:10.1001/jamaoto.2015.1198.
Author Contributions: Dr Sheahan had full access to all the data in the study and takes full responsibility for the integrity of the data and the accuracy of the data analysis.
Study concept and design: Mehanna, Hennessy, Mannion, Sheahan.
Acquisition, analysis, or interpretation of data: Mehanna, Hennessy, O’Leary, Sheahan.
Drafting of the manuscript: Mehanna, Mannion, Sheahan.
Critical revision of the manuscript for important intellectual content: Mannion, Hennessy, O’Leary, Sheahan.
Statistical analysis: Sheahan.
Administrative, technical, or material support: Mehanna, Hennessy, Mannion, Sheahan.
Study supervision: Mannion, Sheahan.
Conflict of Interest Disclosures: None reported.