2018 Section 6 - Laryngology, Voice Disorders, and Bronchoesophalogy

Lee et al

Surgery Volume 160, Number 6

The human voice is produced by a complex instrument comprising of intricate laryngeal struc- tures, which are in turn innervated by branches of the recurrent laryngeal nerve, aided by branches of the superior laryngeal nerve. The effect on the voice of a RLN palsy is recognized and docu- mented easily. However, the subtler changes to the voice in the absence of a nerve palsy are less well documented. With the overarching aim of better understand- ing the potential causes of voice change during thyroid operation, the aims of this study were to quantify the voice change after thyroid operation; to correlate the measured voice change to the extent of operation; and to correlate the degree of voice change to the amount of recurrent laryngeal nerve swelling. METHODS Ethics approval was granted by the institutional review board prior to the commencement of this study. The study patients were prospectively re- cruited during a 12-month period. All patients undergoing total and hemithyroidectomy in the authors’ institution were eligible. However, pa- tients having a concurrent lymph node dissection were excluded. All participating patients had pre- operative and postoperative flexible laryngoscopy to assess vocal cord function and underwent functional voice assessments using established tools. The voice was assessed both subjectively and objectively by speech pathologists using a professional speech and voice analysis system (WEVOSYS lingWAVES 2.5, WEVOSYS, Forch- heim, Germany). The Voice Disorder Index (VDI) was used for the subjective assessment. An abbreviated version of the Voice Handicap Index, it is a self-reported measure of suffering caused by dysphonia. 4 This in- strument gives insight into the voice-related quality of life by quantifying the voice problem as it relates to the common activities in daily living. The VDI score ranges from 0–40, from best to worst. It has been demonstrated that this abbreviated assess- ment performed equally well compared with the extended counterpart. 4 The Dysphonia Severity Index (DSI) was used for the objective assessment. It was designed to establish an objective and quantitative correlate of the perceived voice quality. The DSI is based on the weighted combination of highest fre- quency, lowest intensity, maximum phonation time, and jitter. A perceptually normal voice would score a DSI of equal to or greater than +5 and a severely dysphonic voice would have a

DSI of 5 or less, with +5 to 5 being the usual range. 5 The DSI has been used to objectively monitor the result of voice therapy and training programs and has been shown to have good interobserver consistency. 6,7 Although there is no consensus on how the voice is best assessed in order to determine the effect of thyroidec- tomy, these 2 assessment tools were chosen as they have been validated previously and are rela- tively simple to administer, without any invasive procedure. The preoperative assessments were performed up to 6 weeks preoperatively, and the postoperative assessments were performed day 1 postoperatively prior to discharge. Follow-up assessments were performed between 6–12 months postoperatively. Thyroidectomy was performed as previously described. 8 Dexamethasone (8 mg) was adminis- tered routinely at the beginning of each case. In- traoperatively, the RLN diameter was measured at 2 distinct and specific time points. It was first measured when the RLN was first identified and confirmed by neuromonitoring. The second mea- surement was taken immediately upon completion of the lobectomy. These measurements were made using Vernier calipers with a resolution of 0.1 mm. Intraoperative neuromonitoring (Medtronic NIM 3.0, Medtronic, Minneapolis, MN) is routinely used in all cases. Standard statistical methods were used to analyze the collected data, including t test, v 2 test, Wilcoxon signed-rank test, Pearson correla- tion, and linear regression. Stata version 12 was used for statistical analysis. RESULTS Of the 70 recruited patients, 62 (89%) completed all the necessary assessments and were included in the analysis. The mean age of the study cohort was 47.7 ± 16 years (range, 18–80 years). There were 53 females and 9 males, giving a ratio of 6:1. Thirty-seven (60%) patients had a total thyroidectomy, while 25 (40%) had a hemithyroi- dectomy. Temporary RLN palsy occurred in 4 patients (6.5%), and there was no permanent RLN palsy ( Table I ). In the overall cohort, the VDI score (subjective assessment) increased from 4.2 preoperatively to 9.4 postoperatively ( P < .01). This subjective dete- rioration in the voice also was seen in patients without postoperative RLN palsy, not just in the 4 patients who sustained a palsy ( Table II ). On average, the postoperative VDI score of the pa- tients with palsy was 19.9 points higher than those without a palsy ( P < .01); interestingly, patients

156