- Email: [email protected]
Subjective and Objective Voice Assessments After Recurrent Laryngeal Nerve-Preserved Total Thyroidectomy
ARTICLE IN PRESS Subjective and Objective Voice Assessments After Recurrent Laryngeal Nerve-Preserved Total Thyroidectomy Chariton E. Papadakis, Panagiota Asimakopoulou, Efklidis Proimos, George Perogamvrakis, Effrosyni Papoutsaki, and Theognosia Chimona, Chania, Greece Summary: Objectives. This study aims to investigate early voice changes after total thyroidectomy, to assess the improved parameters in intermediate postoperative intervals, to evaluate the effect of age on voice after thyroidectomy, and to determine the correlation between the objective and the subjective method outcomes. Study Design. This is a prospective, nonrandomized study. Materials and Methods. One hundred ninety-one participants, divided into two age groups, underwent three full voice assessments (preoperatively and 1 and 8 weeks after thyroidectomy) by means of videostroboscopy, perceptual evaluation, acoustic analysis, aerodynamic evaluation, and a self-evaluation questionnaire. Two control groups enrolled in the study: (1) patients with an indication of neck surgery not related to laryngeal nerve injury risk or strap muscle dissection and (2) patients with an indication of a non-neck surgery. Results. No statistically significant difference was found in any voice parameter, between preoperative and 1-week postoperative assessment regarding the control groups. A statistically significant difference was found between preoperative evaluation and 1 week after thyroidectomy for the total study population, as well as for the ≥40 years’ age subgroup for all parameters evaluated except for shimmer. The <40 years’ age subgroup showed a statistically significant difference in pitch, maximum phonation time, and grade, roughness, breathiness, asthenia, and strain (GRBAS) score between preoperative evaluation and 1 week after thyroidectomy. None of the parameters showed a statistical significant difference in the <40 years’ age subgroup at 8 weeks’ evaluation. The Voice Handicap Index (VHI) score correlated significantly with the GRBAS score preoperatively and postoperatively at 1 and 8 weeks’ evaluations. Furthermore, VHI correlated significantly with pitch a week postoperatively. GRBAS scores showed significant correlation not only with VHI but also with acoustic parameters including pitch, shimmer, and noise-to-harmonic ratio 1 and 8 weeks after thyroidectomy. Conclusions. Objective voice changes are common in the majority of the thyroidectomized patients in the early postoperative period. Our results revealed that these changes are related to thyroidectomy per se. Older patients (≥40 years of age) show acoustic and aerodynamic changes 8 weeks postoperatively, although they report no voice abnormalities and their perceptual evaluation is similar to the preoperative one.
INTRODUCTION Thyroid pathology shows a constant increase through the years and thyroidectomy has become the most common neck operation. Voice changes are often encountered after thyroidectomy, whereas recurrent laryngeal nerve (RLN) function does not solely predict functional voice outcome. Other mechanisms that have been reported to affect vocal function include injury to the external branch of the superior laryngeal nerve (SLN), postoperative inflammation, laryngeal edema due to vascular congestion, damage to the cricothyroid muscle, endotracheal intubationrelated trauma (eg, cricoarytenoid joint injury, granuloma formation), strap muscle trauma, and laryngotracheal fixation with surrounding tissues.1–3 Patients may present with a seemingly normal voice without any complaint of vocal dysfunction, or with easy fatigue, difficulty in singing or speaking in a loud voice, and hoarseness. Most of these voice changes resolve spontaneously, but there are cases where patients show maladaptive
Accepted for publication December 16, 2016. From the Otolaryngology Department, General Hospital of Chania, Chania, Greece. Address correspondence and reprint requests to Theognosia Chimona, Otolaryngology Department, General Hospital of Chania, Ag. Marinas 63, Chania, Crete 73300, Greece. E-mail: [email protected] Journal of Voice, Vol. ■■, No. ■■, pp. ■■-■■ 0892-1997 © 2017 The Voice Foundation. Published by Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.jvoice.2016.12.011
compensation, which may persist after the resolution of the underlying pathology. Multidimensional voice outcomes (videostroboscopic examination, patient self-evaluation, perceptual rating of voice quality, acoustic analysis, and aerodynamic variables) are essential to assess voice changes comprehensively. The high incidence of patients suffering from post-thyroidectomy voice changes, as well as the fact that almost half of the patients with unilateral RLN or SLN palsy before surgery may be asymptomatic, has led surgeons to incorporate laryngoscopy or videostroboscopy as a part of the preoperative evaluation. In terms of self-evaluation of dysphonia, numerous quality-of-life (QoL) instruments have been developed and used.4 Voice Handicap Index (VHI), translated and validated in many languages, remains the most popular and widely used tool for the evaluation of voice dysfunction.5 Perceptual evaluation is performed either by an experienced speech pathologist or a phoniatric specialist. The grade, roughness, breathiness, asthenia, and strain (GRBAS) scale has a proven inter-rater reliability and is recommended for clinical purposes by the European Laryngeal Research Group.6,7 Acoustic analysis of voice is often used as a supplementary tool in perceptual evaluation. Numerous acoustic parameters are available through a growing number of software programs. Most clinical studies analyze sustained habitual vowel production and use voice pitch, as well as perturbation measures such as jitter, shimmer, and
ARTICLE IN PRESS 2 noise-to-harmonic ratios (NHR). Jitter and shimmer are timebased perturbation parameters and quantify the degree of cycleto-cycle variability in the fundamental period and amplitude, respectively. NHR is also a time-based measure that integrates elements of both jitter and shimmer. NHR divides the mean amplitude of the noise components by the mean amplitude of the cyclic or periodic component of the voice signal.8 Maximum phonation time (MPT) does not represent a real aerodynamic variable, such as mean phonatory airflow, subglottal air pressure, laryngeal airway resistance, or phonation threshold pressure, but is thought to be a reasonable predictor of airflow and has been used by several studies as an indicator of vocal aerodynamic function. Voice changes after RLN-preserved thyroidectomy have been recently investigated. Previous studies have implicated different assessment tools at various time intervals after thyroidectomy. Complaints of voice dysfunction are reported by 37%–87% of patients after thyroidectomy, whereas acoustic analysis of speech shows lowered pitch and altered perturbation parameters in 18% of patients.9–12 In the present study, we used a voice assessment protocol including both objective and subjective tools. The aims of the study were (1) to investigate early voice changes after RLN-preserved total thyroidectomy, (2) to assess the improved parameters in intermediate postoperative interval, (3) to investigate the effect of age on voice after thyroidectomy, and (4) to find out the correlation between the objective and the subjective method outcomes. MATERIALS AND METHODS Patients Patients with indications for thyroid surgery for benign, suspicious, or malignant pathology were recruited by endocrinologists (institutional or private). The study population was divided into two age subgroups: (1) patients <40 years of age and (2) patients ≥40 years of age. Two control groups enrolled in the study: (1) patients with an indication for neck surgery not related to laryngeal nerve injury risk or strap muscle dissection (control group A) and (2) patients with an indication for a non-neck surgery (control group B). All patients signed a written consent for their participation in the study and in the surgical procedure. Exclusion criteria were (1) age under 18 years, (2) laryngostroboscopic findings indicating altered voice quality (vocal fold polyps or cysts, Reinke’s edema, posterior laryngitis due to laryngopharyngeal reflux, vocal fold immobility, etc), (3) history of a previous phonosurgery procedure and history of a previous thyroid surgery (subtotal thyroidectomy or lobectomy), and (4) history of a progressive neurological disease. Patients were also excluded postoperatively from the study in cases of (1) RLN or SLN injury, (2) pathologic mucosal findings attributed to endotracheal intubation, (3) a well-differentiated T3 or T4 TNM stage thyroid carcinoma, (4) lymph node or distal metastasis, or (5) loss to follow-up. Furthermore, control patients were excluded if intubation lasted less than 90 minutes. The study protocol was approved by the Institutional Review Board of the General Hospital of Chania (# 093/2012).
Journal of Voice, Vol. ■■, No. ■■, 2017
Surgery Total thyroidectomy was performed by extracapsular dissection with identification and preservation of the RLN by the same surgeon. In the vast majority of cases, intraoperative monitoring of the RLN was used (Medtronic NIM-Pulse 2.0). Accurate dissection and distal ligation of the thyroid’s superior pole vessels contributed critically to the preservation of the external branch of the SLN. Control group A consisted of 14 patients who underwent superficial parotidectomy (n = 4), selective neck dissection levels I–III (n = 1), excision of the submandibular gland (n = 3), branchial cleft cyst (n = 2), large neck lipoma (n = 2), desmoid neck tumor (n = 1), and large dermoid cyst of the mouth floor (n = 1). Control group B consisted of 24 patients who underwent ear surgery (n = 5), abdominal surgery (n = 9), and orthopedic surgery (n = 10). Study design Study participants underwent three full voice assessments (preoperatively and 1 and 8 weeks postoperatively). The preoperative measurements allowed baseline assessment of voice parameters for each patient and all subsequent assessments were compared to these measures. A persistent voice complication after the second postoperative assessment was followed up at a 1-month interval. Voice outcome measures included the following: (1) Videostroboscopy. Clinical examination was performed with videostroboscopy (70° rigid endoscope, type Xion endoSTROB, DIVAS EndoStrob; Xion Medical, Berlin, Germany). The examination was performed in all cases by the same expert. Phonatory characteristics of the larynx at rest and during speech were viewed and mucosal waves of the vocal folds were examined during phonation of the prolonged vowel /i/. (2) Perceptual examination. GRBAS scale, as has been suggested by Hirano, 13 was used by the same speech pathologist to grade voice quality. Grading was applied blindly to voice samples of the patients containing conversational speech, counting from 1 to 10 and pronouncing the vowels “α/e/i/o/ou.” The GRBAS scale allows a rating of voice over four parameters (roughness, breathiness, asthenicity, and strain) and an overall grade for voice quality. Each aspect is rated on a fourpoint scale ranging from 0 (normal) to 3 (severely abnormal). (3) Acoustic analysis. The voice signal was recorded digitally, via a head microphone (~30 cm head-to-mouth distance) placed at 45° from the mouth axis, in a quiet room (ambient noise <50 dBA) and saved in a computer, with a sampling frequency of 44,100 Hz and a resolution of 16 bits per sample. It consisted of a sustained /α/, voiced three times, at a comfortable pitch and loudness. Acoustic analysis was performed on voice samples by Praat system (University of Amsterdam, The Netherlands; http://www.fon.hum.uva.nl/praat), and the mean score for pitch, jitter, shimmer, and harmonic-tonoise ratio was calculated (mean duration of the analyzed sample: 5 seconds).
ARTICLE IN PRESS Chariton E. Papadakis, et al
3
Voice Assessments After RLN-Preserved Thyroidectomy
TABLE 1. Demographic and Pathologic Characteristics Age
Total (N = 191) Women (n = 137, 73%) Men (n = 54, 28%)
Pathology
Mean Age (SD)
Min-Max (y)
Papillary Carcinoma (T1 or T2, No, Mo), n (%)
Follicular Carcinoma (T1 or T2, No, Mo), n (%)
Benign Disorder,* n (%)
50.6 (14.3) 50.2 (14.4) 51.6 (13.9)
19–85 19–85 28–77
95 (49.7) 62 (45.2) 33 (61.2)
4 (2.1) 3 (2.2) 1 (1.8)
92 (48.2) 72 (52.6) 20 (37)
* Benign disorder: multinodular goiter, toxic or nontoxic adenoma, and Hashimoto thyroiditis. Abbreviation: max, maximum; min, minimum; SD, standard deviation.
(4) Aerodynamics. MPT was measured during a sustained /α/ after a deep inspiration at comfortable pitch and loudness (expressed in seconds). The maximum value of the three trials was considered as the valid one. (5) Self-evaluation. The Greek version of the VHI questionnaire was handed out to all patients.14 It is a 30-point scale questionnaire with three subscales (10 points each): functional (VHI-F), physical (VHI-P), and emotional (VHIE). Each item is scored from 0 to 4, according to the impact of the statement on the patient’s everyday life (0 = never, 1 = almost never, 2 = sometimes, 3 = almost always, and 4 = always). Subscale scores range from 0 to 40 and the overall score ranges from 0 to 120. A total score of 0–30 indicates mild dysphonia, 31–60 moderate dysphonia, 61–90 severe dysphonia, and 91–120 very severe dysphonia.5 Data analysis Voice outcome measures were computed by using the SPSS statistical software package (SPSS, v.22.0; SPSS, Inc., Chicago, IL). Changes in voice pitch, jitter, shimmer, and MPT values were assessed with a paired t test. The scores of NHR, VHI, and GRBAS were highly skewed and were analyzed with the nonparametric Wilcoxon test. Spearman correlations were used to evaluate associations of subjective evaluation with perceptual and other clinical parameters. Statistical analysis was performed for the total study population as well as for the two age subgroups and control groups.
RESULTS Of the total of 204 patients recruited, 4 were excluded due to unilateral RLN paresis after thyroidectomy; in 2 patients, laryngostroboscopy revealed focal folds’ mucosal abnormalities due to intubation; and 7 were lost to follow-up. After the implementation of the exclusion criteria, the study population included 191 patients with a mean age 50.6 years (range 19– 85 years, standard deviation [SD]: 14.3); 137 (73%) were women, and 54 (28%) were men. Forty-six patients were <40 years old, and 145 patients were ≥40 years old. Histology revealed a T1 or T2 well-differentiated thyroid carcinoma in 52% of the patients (Table 1). Summary descriptive statistics are presented as minimum and maximum values, as well as mean values with SD (Table 2). Control group A consisted of six men with a mean age of 54.8 years (SD: 18.3) and eight women with a mean age of 54.6 years (SD: 11.3). Control group B consisted of 11 men with a mean age 49.5 years (SD: 14.7) and 13 women with mean age 50.3 years (SD: 17.9). Table 3 shows the paired t test results for control groups A and B. No statistically significant difference was found in any voice parameter, between preoperative and 1-week postoperative assessments. Thus, control patients did not undergo the 8-week evaluation. Regarding the study population, a statistically significant difference was found between preoperative and 1-week postthyroidectomy evaluations for the total study population, as well as for the ≥40 years’ age subgroup for pitch (P < 0.001), jitter (P < 0.05), NHR (P < 0.001), MPT (P < 0.001), GRBAS score (P < 0.001), and VHI score (P < 0.001) (Table 4). The <40 years’
TABLE 2. Acoustic, Aerodynamic, Perceptual, and Subjective Parameters Preoperatively and 1 and and 8 Weeks Post Thyroidectomy Preoperatively
Postoperatively (1 wk)
Postoperatively (8 wk)
Analysis
Parameter
Min-Max
Mean (SD)
Min-Max
Mean (SD)
Min-Max
Mean (SD)
Acoustic
Pitch, men Pitch, women Jitter Shimmer NHR MPT GRBAS VHI
105.9–288 88.1–294.1 0–1.4 1.32–12.1 0–0.14 6.6–36.7 0–8 0–52
199.76 (49.9) 203.1 (47.1) 0.37 (0.19) 4.3 (2.4) 0.02 (0.02) 18.7 (5.7) 0.86 (1.6) 4.2 (7.1)
89.9–241.4 82.1–262.1 0.13–2.6 0–17.5 0–0.73 5–30.8 0–13 0–59
175 (43.02) 178.1 (43.1) 0.44 (0.32) 4.5 (2.6) 0.03 (0.06) 14.2 (4.49) 1.6 (1.9) 8.5 (11.7)
109.3–286.6 89.4–287 0.12–0.97 0–14.34 0–0.23 8.3–34.7 0–7 0–42
197.8 (48.04) 201.5 (45.8) 0.34 (0.17) 4.18 (2.25) 0.02 (0.03) 18.2 (5.1) 0.76 (1.2) 4.5 (7.3)
Aerodynamic Perceptual Subjective
Abbreviations: GRBAS, grade, roughness, breathiness, asthenia, and strain; max, maximum; min, minimum; MPT, maximum phonation time; NHR, noiseto-harmonic ratio; SD, standard deviation; VHI, Voice Handicap Index.
ARTICLE IN PRESS 4
Journal of Voice, Vol. ■■, No. ■■, 2017
TABLE 3. Paired t Test Results for Control Groups A and B, Between Voice Parameters Preop and 1 Week Postop Control group B: Men, n = 11; Women, n = 13 Non-neck Surgery
Control Group A: Men, n = 6; Women, n = 8 Neck Surgery with No RLN Risk and No Strap Muscle Dissection Parameter
Preop Mean (SD)
Postop 1 wk Mean (SD)
Pitch, men 127.8 (20.5) 130.4 (26.3) Pitch, women 221.08 (39.2) 220.4 (38.8) Jitter 0.33 (0.18) 0.32 (0.11) Shimmer 3.38 (1.3) 2.99 (1.58) NHR 0.007 (0.007) 0.0086 (0.006) MPT 18.02 (3.7) 17.3 (3.61) GRBAS 0.21 (0.4) 0.5 (0.8) VHI 1.21 (1.57) 3.43 (8.12)
95% Confidence Interval [−21.3 to 16.18] [−5.4 to 6.77] [−0.09 to 0.11] [−1.02 to 1.8] [−1.09 to 2.55]
P
Preop Mean (SD)
Postop 1 wk Mean (SD)
0.74 124.3 (17.6) 123.7 (16.4) 0.8 225.5 (11,7) 227.23 (13.9) 0.84 0.35 (0.12) 0.33 (0.10) 0.55 4.95 (1.75) 4.16 (1.73) 0.5 0.02 (0.02) 0.19 (0.017) 0.4 19.9 (5.9) 20.7 (5.8) 0.16 0.67 (1.09) 0.67 (0.81) 0.3 1.42 (1.74) 1.46 (1.8)
95% Confidence Interval [−2.7 to 3.9] [−6.9 to 3.45] [−0.009 to 0.6] [−0.17 to 1.7] [−1.9 to 0.45]
P 0.69 0.47 0.14 0.1 0.36 0.21 1.0 0.9
Note: NHR, GRBAS, and VHI were analyzed with nonparametric Wilcoxon test. Abbreviations: GRBAS, grade, roughness, breathiness, asthenia, and strain; MPT, maximum phonation time; NHR, noise-to-harmonic ratio; postop, postoperatively; preop, preoperatively; RLN, recurrent laryngeal nerve; SD, standard deviation; VHI, Voice Handicap Index.
age subgroup showed a statistically significant difference in pitch, MPT, and GRBAS score between preoperative and 1-week postthyroidectomy evaluations (Table 4). Eight weeks postoperatively, a statistically significant difference was found for the whole study population for NHR (P < 0.05) and MPT (P < 0.05), as well as for pitch in women (P = 0.024) in the ≥40 years’ age subgroup, compared with the initial evaluation. None of the parameters showed a statistically significant difference in the <40 years’ age subgroup at 8 weeks’ evaluation (P > 0.05) (Table 5). Table 6 shows Spearman correlation coefficients between subjective, perceptual, and objective parameters preoperatively and 1 and 8 weeks postoperatively for the total study population. The VHI scores correlated significantly (P < 0.01) with the GRBAS scores preoperatively (rho = 0.324) and postoperatively at 1-week (rho = 0.543) and 8-week (rho = 0.478) evaluations. Furthermore, the VHI scores correlated significantly (rho = 0.203, P < 0.01) with pitch 1 week postoperatively. The GRBAS scores showed significant correlation not only with VHI but also with the following acoustic parameters: pitch (rho = 0.264, P < 0.01; rho = 0.148, P < 0.05), shimmer (rho = 0.172, P < 0.05; rho = 0.168, P < 0.05), and NHR (rho = 0.148, P < 0.05; rho = 0.17, P < 0.05) at 1 and 8 weeks after thyroidectomy, respectively. DISCUSSION Thyroidectomy is one of the most frequent operations in our department where patients undergo surgery due to benign, suspicious, or malignant pathology. Traditionally, voice changes after thyroidectomy were attributed to iatrogenic RLN injury. However, patients even after RLN-preserved thyroidectomy frequently complain of voice changes (ie, hoarseness, fatigue, difficulty in singing or yelling), as well as swallowing problems. These alterations to vocal and throat functions may decrease patients’ QoL, the so-called post-thyroidectomy syndrome.15,16 Fibrosis, vascular changes, injury of the external branch of the SLN, and psychosocial reactions have been implicated in the etiology of this syndrome.10,17
In the present prospective study, a multidimensional voice evaluation was used, consisting of both objective and subjective measures, to assess voice changes after total thyroidectomy. The first post-thyroidectomy evaluation took place as early as the seventh postoperative day, under the condition of normal stroboscopic findings, that is, no indication of laryngeal mucosal or laryngeal nerve damage. Unilateral RLN injury leads to early hoarseness with a hypokinetic or fixed vocal fold at the paramedian position, loss of vibratory mucosal wave, and incomplete glottic closure. Stroboscopic characteristics of SLN injury include thinned, shortened, and bowed unilateral vocal fold; lack of tension and lower vertical plane of the paretic vocal fold; oblique glottis deviating to the paretic side posteriorly; and irregular vibratory motions during phonation.18 The uniform way of patients’ intervention (total thyroidectomy by the same surgeon) and evaluation (same phoniatric specialist and speech therapist) is one of the main advantages of the present study. The analysis of the results for the control groups showed no statistically significant difference in all voice parameters tested a week postoperatively (Table 3). Thus, we can conclude that voice changes in the study population are related to thyroidectomy and not to endotracheal intubation, or to postoperative neck discomfort or pain. Furthermore, we analyzed a quite large population subdivided into two age subgroups, a <40 years’ age group and a ≥40 years’ age group. Finally, we evaluated study parameters at 8 weeks, an intermediate time interval, even though previous studies report that voice disturbances are resolved 3–6 months after thyroidectomy.19,20 With this intermediate assessment, we were able to find out if there exist parameters that show earlier recovery and whether age affects voice outcome at this interval. Study limitations include the relatively small number of patients in the subgroup aged <40 years, the aerodynamic evaluation only by MPT, and the lack of a long-term evaluation which is currently contacted. We found a statistically significant difference in all evaluated parameters, with the exception of shimmer, between the
Parameter
Preop Total Mean (SD)
Postop Total 1 wk Mean (SD)
[95% CI] P
Postop Age <40 1 wk Mean (SD)
Preop Age <40 Mean (SD)
199.7 (49.9) 175.7 (43.02)
187.05 (54.76) 176.54 (40.73) 0.4 (0.3) 4.02 (2.07)
1.5 (1.7) 6.8 (9.7)
[8.05–20.36] 200.2 (50.44) 0.049* [16.7–30.25] 204.2 (48.3) <0.001* [−0.16 to 0.04] 0.38 (0.19) 0.25 [−0.79 to 0.61] 4.4 (2.1) 0.79 0.27 0.018 (0.02) [2.8–5.17] 19 (5.7) <0.001* 0.03* 0.81 (1.4) 0.06 3.9 (6.6)
Postop Age ≥40 1 wk Mean (SD) 173.7 (41.06) 178.7 (44.2) 0.46 (0.3) 4.7 (2.7) 0.03 (0.07) 14 (7.2) 1.6 (2.0) 9.06 (12.2)
[95% CI] P [18.1–34.75] <0.001* [21.0–30.06] <0.001* [−0.13 to −0.2] 0.007* [−0.85 to 0.33] 0.149 <0.001* [3.12–5.41] <0.001* <0.001* <0.001*
* Shows a statistically significant difference between the compared parameters. Abbreviations: CI, confidence interval; GRBAS, grade, roughness, breathiness, asthenia, and strain; MPT, maximum phonation time; NHR, noise-to-harmonic ratio; postop, postoperatively; preop, preoperatively; SD, standard deviation; VHI, Voice Handicap Index.
TABLE 5. Analysis of Acoustic, Aerodynamic, Perceptual, and Subjective Parameters Preop and 8 Weeks Postop for Total Study Population as Well as for Age Subgroups
Parameter
Preop Total Mean (SD)
Postop Total 8 wk Mean (SD)
Pitch, men
199.7 (49.9)
197.8 (48.0)
Pitch, women 203.1 (47.1)
201.5 (45.8)
Jitter
0.37 (0.19)
0.34 (0.7)
4.3 (2.4)
4.18 (2.25)
NHR MPT
0.02 (0.02) 18.7 (5.7)
0.02 (0.03) 18.2 (5.1)
GRBAS VHI
0.86 (1.6) 4.2 (7.1)
0.76 (1.2) 4.5 (7.3)
Shimmer
[95% CI] P
Preop Age <40 Mean (SD)
[−2.1 to 6.02] 197.26 (50.4) 0.34 [−0.58 to 3.8] 200.03 (44.2) 0.15 [−0.009 to 0.006] 0.34 (0.19) 0.14 [−0.25 to 0.48] 3.93 (1.9) 0.53 0.016* 0.01 (0.2) [0.03–0.9] 17.8 (5.9) 0.03* 0.78 1.02 (1.8) 0.82 5.07 (8.4)
Postop Age <40 8 wk Mean (SD) 195.65 (51.2) 200.08 (41.47) 0.34 (0.17) 3.97 (2.1) 0.02 (0.02) 17.8 (5.4) 0.95 (1.3) 3.9 (6.01)
[95% CI] P
Preop Age ≥40 Mean (SD)
[−3.9 to 7.12] 200.2 (48.3) 0.51 [−6.33 to 6.2] 204.26 (48.32) 0.98 [−0.073 to 0.07] 0.38 (0.19) 0.98 [−0.79 to 0.71] 4.4 (2.1) 0.9 0.56 0.018 (0.02) [−0.9 to 0.89] 19 (5.7) 0.97 0.72 0.81 (1.4) 0.56 3.9 (6.6)
Postop Age ≥40 8 Weeks Mean (SD) 198.18 (48.06) 202.01(47.58) 0.34 (0.16) 4.2 (2.2) 0.024 (0.03) 18.4 (4.9) 0.70 (1.2) 4.75 (7.7)
[95% CI] P [−2.7 to 6.7] 0.39 [0.30–4.1] 0.024* [−0.006 to 0.07] 0.09 [−0.26 to 0.59] 0.4 0.015* [0.1–1.19] 0.018* 0.57 0.53
5
* Shows a statistically significant difference between the compared parameters. Abbreviations: CI, confidence interval; GRBAS, grade, roughness, breathiness, asthenia, and strain; MPT, maximum phonation time; NHR, noise-to-harmonic ratio; postop, postoperatively; preop, preoperatively; SD, standard deviation; VHI, Voice Handicap Index.
ARTICLE IN PRESS
0.02 (0.02) 13.8 (4.5)
Preop Age ≥40 Mean (SD)
Voice Assessments After RLN-Preserved Thyroidectomy
197.26 (50.39) [16.7–31.35] <0.001* Pitch, women 203.1 (47.1) 178.1 (43.13) [21.2–28.7] 200.03 (44.2) <0.001* Jitter 0.37 (0.19) 0.44 (0.32) [−0.12 to −0.24] 0.34 (0.19) 0.003* Shimmer 4.3 (2.4) 4,5 (2.6) [−0.7 to 0.11] 3.93 (1.9) 0.154 NHR 0.02 (0.02) 0.03 (0.06) <0.001* 0.01 (0.2) MPT 18.7 (5.7) 14.2 (4.49) [4.03–5.11] 17.8 (5.9) <0.001* GRBAS 0.86 (1.6) 1.6 (1.9) <0.001* 1.02 (1.8) VHI 4.2 (7.1) 8.5 (11.7) <0.001* 5.07 (8.4) Pitch, men
[95% CI] P
Chariton E. Papadakis, et al
TABLE 4. Analysis of Acoustic, Aerodynamic, Perceptual, and Subjective Parameters Preop and 1 Week Postop for Total Study Population as Well as for Age Subgroups
ARTICLE IN PRESS 6
Journal of Voice, Vol. ■■, No. ■■, 2017
TABLE 6. Spearman Correlation Coefficients Between Subjective, Perceptual, and Objective Parameters Preop and at 1 and 8 Weeks Postop for the Total Study Population
Preop VHI GRBAS Postop 1 wk VHI GRBAS Postop 8 wk VHI GRBAS
Pitch
Jitter
Shimmer
NHR
MPT
GRBAS
VHI
−0.018 −0.154
0.65 0.123
0.10 0.22†
0.123 0.217†
−0.007 0.08
0.324† 1
1 0.324†
−0.203† −0.264†
0.033 0.020
0.47 0.172*
0.08 0.148*
−0.042 0.113
0.543† 1
1 0.543†
−0.042 −0.148*
0.004 0.028
0.083 0.168*
0.076 0.17*
0.048 −0.113
0.478† 1
1 0.478†
* Correlation is significant at the 0.05 level. † Correlation is significant at the 0.01 level. Abbreviations: GRBAS, grade, roughness, breathiness, asthenia, and strain; MPT, maximum phonation time; NHR, noise-to-harmonic ratio; postop, postoperatively; preop, preoperatively; VHI, Voice Handicap Index.
preoperative evaluation and the acoustic analysis 1 week after thyroidectomy for the whole study population (Table 4). At 8 weeks’ postoperative evaluation, the mean values of NHR and MPT showed a statistically significant difference compared to the preoperative evaluation. An interesting finding was that patients <40 years of age showed a statistically significant difference only in pitch, MPT, and GRBAS score, and these parameters’ mean values were restored to normal 8 weeks postoperatively (Tables 4 and 5). In the ≥40 years’ age subgroup at 1 week’s evaluation, statistical analysis showed the same results as those for the total study population, whereas at 8 weeks’ postthyroidectomy evaluation, there was still a statistically significant difference in pitch (women), NHR, and MPT. Patient’s age and extent of operation have been reported as independent factors for post-thyroidectomy lowered-pitch voice. Park et al found that female patients older than 32 years and those who underwent total thyroidectomy showed a statistically significant lower pitch compared to younger patients and less extended surgery.19 It has been suggested that calcified laryngeal cartilages, mucosal alterations, and laryngeal muscle atrophy, seen with the advance of age, are some of the causes of this functional deficit. Total thyroidectomy causes greater soft tissue injury, compared to lobectomy, and shows an early adverse effect on voice, which usually is resolved by 6 months.20 Extrinsic laryngeal muscle function is to lengthen or shorten the vocal folds and to regulate pitch by changing the relation of the thyroid gland to the cartilage laryngeal frame. Strap muscle function can be altered by cutting or more often by lateral retraction or by laryngeal fixation in the surgical bed. Thus, it is believed that with the advance of age, compensatory muscle mechanisms poorly counterweight the extended soft tissue damage after a total thyroidectomy. In our study we found that patients after thyroidectomy did not show altered cycle-to-cycle variability in amplitude. Thus, shimmer was the perturbation parameter evaluated with no significant change, although it has been reported to differ significantly with harmonic-to-noise ratio in previous studies and has been suggested as a predictor of overall voice quality.21,22 NHR and
MPT were the parameters that differed significantly for the total population throughout the study. In the younger age subgroup (<40 years), we found no significant change 8 weeks postoperatively. MPT is a predictor of airflow and has been used by several studies for the evaluation of vocal aerodynamic function. Solomon et al found no significant change in MPT in patients who underwent thyroidectomy or had no neck surgery 2 weeks and 3 months postoperatively.23 It has been suggested that MPT is rather an indicator of poor laryngeal valving, and this finding may explain the significant change in the values in the older subgroup. As far as subjective parameters are concerned, VHI is a useful QoL tool for self-evaluation of dysphonia and GRBAS is the main perceptual tool for the assessment of voice abnormalities.5,6 These two instruments have been found to have a significant correlation throughout the study. VHI scores did not correlate with other clinical objective parameters except pitch 1 week after surgery. For the subgroup of patients under 40 years of age, there was no statistically significant change in the VHI scores even 1 week after thyroidectomy. Older patients revealed a significant VHI score worsening. However, previous studies have reported discrepancies between subjective and objective voice changes in thyroidectomized patients.17,24,25 Although RLNpreserved post-thyroidectomy voice changes usually subside after 3–6 months and some patients do not report symptoms, they have to be informed about the possible voice function alteration.19,20 Furthermore, when patients are voice professionals, women, and over 40 years of age, voice rehabilitation might be useful soon after surgery to obtain quick recovery. CONCLUSIONS Objective voice changes are common in the majority of the thyroidectomized patients in the early postoperative period. These changes are related to the thyroidectomy per se and not to endotracheal intubation or to neck discomfort experienced in other neck operations. In younger patients (<40 years of age), a discordance between VHI and acoustic parameters exists. Furthermore, subjective tools do not correlate significantly with
ARTICLE IN PRESS Chariton E. Papadakis, et al
Voice Assessments After RLN-Preserved Thyroidectomy
objective acoustic parameters but show a significant correlation with perceptual assessment. Young patients return to their previous voice quality as early as 8 weeks’ post-thyroidectomy evaluation. Although older patients (≥40 years of age) report no abnormalities and the perceptual evaluation is similar to the preoperative one, they are found with acoustic and aerodynamic changes in parameters such as pitch (women), NHR, and MPT. REFERENCES 1. Finck C. Laryngeal dysfunction after thyroid surgery: diagnosis, evaluation and treatment. Acta Chir Belg. 2006;106:378–387. 2. Vilkman E, Sonninen A, Hurme P, et al. External laryngeal frame function in voice production revisited: a review. J Voice. 1996;10:78–92. 3. Hong KH, Kim YK. Phonatory characteristics of patients undergoing thyroidectomy without laryngeal nerve injury. Otolaryngol Head Neck Surg. 1997;117:399–404. 4. Zraick RI, Risner BY. Assessment of quality of life in persons with voice disorders. Curr Opin Otolaryngol Head Neck Surg. 2008;16:188–193. 5. Jacobson BH, Johnson A, Grywalsky C, et al. The voice handicap index (VHI): development and validation. Am J Speech Lang Pathol. 1997;6:66–70. 6. Dejonckere PH, Bradley P, Clemente P, et al. A basic protocol for functional assessment of voice pathology, especially for investigating the efficacy of (phonosurgical) treatments and evaluating new assessment techniques. Guideline elaborated by the Committee on Phoniatrics of the European Laryngological Society (ELS). Eur Arch Otorhinolaryngol. 2001;258:77–82. 7. De Bodt MS, Van de Heyning PH, Wuyts FL, et al. The perceptual evaluation of voice disorders. The perceptual evaluation of voice disorders. Acta Otorhinolaryngol Belg. 1996;50:283–291. 8. Kent RD, Read C. Introduction to the acoustic analysis of speech. In: The Acoustic Analysis of Speech. 2nd ed. Singular Thomson Learning; 2002:53– 103. 9. Soylu L, Ozbas S, Uslu HY, et al. The evaluation of the causes of subjective voice disturbances after thyroid surgery. Am J Surg. 2007;194:317–322. 10. Sinagra DL, Montesinos MR, Tacchi VA, et al. Voice changes after thyroidectomy without recurrent laryngeal nerve injury. J Am Coll Surg. 2004;199:556–560.
7
11. Van Lierde K, D’Haeseleer E, Wuyts FL, et al. Impact of thyroidectomy without laryngeal nerve injury on vocal quality characteristics: an objective multiparameter approach. Laryngoscope. 2010;120:338–345. 12. Nam IC, Bae JS, Chae BJ, et al. Therapeutic approach to patients with a lower-pitched voice after thyroidectomy. World J Surg. 2013;37:1940–1950. 13. Hirano M. Clinical Examination of Voice. New York, NY: Springer-Verlag Wien; 1981. 14. Helidoni ME, Murray T, Moschandreas J, et al. Cross-cultural adaptation and validation of the voice handicap index into Greek. J Voice. 2010;24:221– 227. 15. Lombardi CP, Raffaelli M, D’Alatri L, et al. Voice and swallowing changes after thyroidectomy in patients without inferior laryngeal nerve injuries. Surgery. 2006;140:1026–1032. 16. Ryu J, Ryu Y, Jung Y, et al. Extent of thyroidectomy affects vocal and throat functions: a prospective observational study of lobectomy versus total thyroidectomy. Surgery. 2013;154:611–620. 17. Stojadinovic A, Shaha AR, Orlikoff RF, et al. Prospective functional voice assessment in patients undergoing thyroid surgery. Ann Surg. 2002;236:823– 832. 18. Dursun G, Sataloff RT, Spiegel JR, et al. Superior laryngeal nerve paresis and paralysis. J Voice. 1996;10:206–211. 19. Park J-O, Bae J-S, Lee S-H, et al. The long-term prognosis of voice pitch change in female patients after thyroid surgery. World J Surg. 2016;40:2382– 2390. [Epub ahead of print]. 20. Vicente DA, Solomon NP, Avital I, et al. Voice outcomes after total thyroidectomy, partial thyroidectomy, or non-neck surgery using a prospective multifactorial assessment. J Am Coll Surg. 2014;219:152–163. 21. Maryn Y, Corthals P, Van Cauwenberge P, et al. Toward improved ecological validity in the acoustic measurement of overall voice quality: combining continuous speech and sustained vowels. J Voice. 2010;24:540–555. 22. Solomon N, Awan S, Helou L, et al. Acoustic analyses of thyroidectomyrelated changes in vowel phonation. J Voice. 2012;26:711–720. 23. Solomon N, Helou L, Makashay M, et al. Aerodynamic evaluation of the post-thyroidectomy voice. J Voice. 2012;26:454–461. 24. McIvor NP, Flint DJ, Gillibrand J, et al. Thyroid surgery and voice-related outcomes. Aust N Z J Surg. 2000;70:179–183. 25. Debruyne F, Ostyn F, Delaere P, et al. Acoustic analysis of the speaking voice after thyroidectomy. J Voice. 1997;11:479–482.
INTRODUCTION Thyroid pathology shows a constant increase through the years and thyroidectomy has become the most common neck operation. Voice changes are often encountered after thyroidectomy, whereas recurrent laryngeal nerve (RLN) function does not solely predict functional voice outcome. Other mechanisms that have been reported to affect vocal function include injury to the external branch of the superior laryngeal nerve (SLN), postoperative inflammation, laryngeal edema due to vascular congestion, damage to the cricothyroid muscle, endotracheal intubationrelated trauma (eg, cricoarytenoid joint injury, granuloma formation), strap muscle trauma, and laryngotracheal fixation with surrounding tissues.1–3 Patients may present with a seemingly normal voice without any complaint of vocal dysfunction, or with easy fatigue, difficulty in singing or speaking in a loud voice, and hoarseness. Most of these voice changes resolve spontaneously, but there are cases where patients show maladaptive
Accepted for publication December 16, 2016. From the Otolaryngology Department, General Hospital of Chania, Chania, Greece. Address correspondence and reprint requests to Theognosia Chimona, Otolaryngology Department, General Hospital of Chania, Ag. Marinas 63, Chania, Crete 73300, Greece. E-mail: [email protected] Journal of Voice, Vol. ■■, No. ■■, pp. ■■-■■ 0892-1997 © 2017 The Voice Foundation. Published by Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.jvoice.2016.12.011
compensation, which may persist after the resolution of the underlying pathology. Multidimensional voice outcomes (videostroboscopic examination, patient self-evaluation, perceptual rating of voice quality, acoustic analysis, and aerodynamic variables) are essential to assess voice changes comprehensively. The high incidence of patients suffering from post-thyroidectomy voice changes, as well as the fact that almost half of the patients with unilateral RLN or SLN palsy before surgery may be asymptomatic, has led surgeons to incorporate laryngoscopy or videostroboscopy as a part of the preoperative evaluation. In terms of self-evaluation of dysphonia, numerous quality-of-life (QoL) instruments have been developed and used.4 Voice Handicap Index (VHI), translated and validated in many languages, remains the most popular and widely used tool for the evaluation of voice dysfunction.5 Perceptual evaluation is performed either by an experienced speech pathologist or a phoniatric specialist. The grade, roughness, breathiness, asthenia, and strain (GRBAS) scale has a proven inter-rater reliability and is recommended for clinical purposes by the European Laryngeal Research Group.6,7 Acoustic analysis of voice is often used as a supplementary tool in perceptual evaluation. Numerous acoustic parameters are available through a growing number of software programs. Most clinical studies analyze sustained habitual vowel production and use voice pitch, as well as perturbation measures such as jitter, shimmer, and
ARTICLE IN PRESS 2 noise-to-harmonic ratios (NHR). Jitter and shimmer are timebased perturbation parameters and quantify the degree of cycleto-cycle variability in the fundamental period and amplitude, respectively. NHR is also a time-based measure that integrates elements of both jitter and shimmer. NHR divides the mean amplitude of the noise components by the mean amplitude of the cyclic or periodic component of the voice signal.8 Maximum phonation time (MPT) does not represent a real aerodynamic variable, such as mean phonatory airflow, subglottal air pressure, laryngeal airway resistance, or phonation threshold pressure, but is thought to be a reasonable predictor of airflow and has been used by several studies as an indicator of vocal aerodynamic function. Voice changes after RLN-preserved thyroidectomy have been recently investigated. Previous studies have implicated different assessment tools at various time intervals after thyroidectomy. Complaints of voice dysfunction are reported by 37%–87% of patients after thyroidectomy, whereas acoustic analysis of speech shows lowered pitch and altered perturbation parameters in 18% of patients.9–12 In the present study, we used a voice assessment protocol including both objective and subjective tools. The aims of the study were (1) to investigate early voice changes after RLN-preserved total thyroidectomy, (2) to assess the improved parameters in intermediate postoperative interval, (3) to investigate the effect of age on voice after thyroidectomy, and (4) to find out the correlation between the objective and the subjective method outcomes. MATERIALS AND METHODS Patients Patients with indications for thyroid surgery for benign, suspicious, or malignant pathology were recruited by endocrinologists (institutional or private). The study population was divided into two age subgroups: (1) patients <40 years of age and (2) patients ≥40 years of age. Two control groups enrolled in the study: (1) patients with an indication for neck surgery not related to laryngeal nerve injury risk or strap muscle dissection (control group A) and (2) patients with an indication for a non-neck surgery (control group B). All patients signed a written consent for their participation in the study and in the surgical procedure. Exclusion criteria were (1) age under 18 years, (2) laryngostroboscopic findings indicating altered voice quality (vocal fold polyps or cysts, Reinke’s edema, posterior laryngitis due to laryngopharyngeal reflux, vocal fold immobility, etc), (3) history of a previous phonosurgery procedure and history of a previous thyroid surgery (subtotal thyroidectomy or lobectomy), and (4) history of a progressive neurological disease. Patients were also excluded postoperatively from the study in cases of (1) RLN or SLN injury, (2) pathologic mucosal findings attributed to endotracheal intubation, (3) a well-differentiated T3 or T4 TNM stage thyroid carcinoma, (4) lymph node or distal metastasis, or (5) loss to follow-up. Furthermore, control patients were excluded if intubation lasted less than 90 minutes. The study protocol was approved by the Institutional Review Board of the General Hospital of Chania (# 093/2012).
Journal of Voice, Vol. ■■, No. ■■, 2017
Surgery Total thyroidectomy was performed by extracapsular dissection with identification and preservation of the RLN by the same surgeon. In the vast majority of cases, intraoperative monitoring of the RLN was used (Medtronic NIM-Pulse 2.0). Accurate dissection and distal ligation of the thyroid’s superior pole vessels contributed critically to the preservation of the external branch of the SLN. Control group A consisted of 14 patients who underwent superficial parotidectomy (n = 4), selective neck dissection levels I–III (n = 1), excision of the submandibular gland (n = 3), branchial cleft cyst (n = 2), large neck lipoma (n = 2), desmoid neck tumor (n = 1), and large dermoid cyst of the mouth floor (n = 1). Control group B consisted of 24 patients who underwent ear surgery (n = 5), abdominal surgery (n = 9), and orthopedic surgery (n = 10). Study design Study participants underwent three full voice assessments (preoperatively and 1 and 8 weeks postoperatively). The preoperative measurements allowed baseline assessment of voice parameters for each patient and all subsequent assessments were compared to these measures. A persistent voice complication after the second postoperative assessment was followed up at a 1-month interval. Voice outcome measures included the following: (1) Videostroboscopy. Clinical examination was performed with videostroboscopy (70° rigid endoscope, type Xion endoSTROB, DIVAS EndoStrob; Xion Medical, Berlin, Germany). The examination was performed in all cases by the same expert. Phonatory characteristics of the larynx at rest and during speech were viewed and mucosal waves of the vocal folds were examined during phonation of the prolonged vowel /i/. (2) Perceptual examination. GRBAS scale, as has been suggested by Hirano, 13 was used by the same speech pathologist to grade voice quality. Grading was applied blindly to voice samples of the patients containing conversational speech, counting from 1 to 10 and pronouncing the vowels “α/e/i/o/ou.” The GRBAS scale allows a rating of voice over four parameters (roughness, breathiness, asthenicity, and strain) and an overall grade for voice quality. Each aspect is rated on a fourpoint scale ranging from 0 (normal) to 3 (severely abnormal). (3) Acoustic analysis. The voice signal was recorded digitally, via a head microphone (~30 cm head-to-mouth distance) placed at 45° from the mouth axis, in a quiet room (ambient noise <50 dBA) and saved in a computer, with a sampling frequency of 44,100 Hz and a resolution of 16 bits per sample. It consisted of a sustained /α/, voiced three times, at a comfortable pitch and loudness. Acoustic analysis was performed on voice samples by Praat system (University of Amsterdam, The Netherlands; http://www.fon.hum.uva.nl/praat), and the mean score for pitch, jitter, shimmer, and harmonic-tonoise ratio was calculated (mean duration of the analyzed sample: 5 seconds).
ARTICLE IN PRESS Chariton E. Papadakis, et al
3
Voice Assessments After RLN-Preserved Thyroidectomy
TABLE 1. Demographic and Pathologic Characteristics Age
Total (N = 191) Women (n = 137, 73%) Men (n = 54, 28%)
Pathology
Mean Age (SD)
Min-Max (y)
Papillary Carcinoma (T1 or T2, No, Mo), n (%)
Follicular Carcinoma (T1 or T2, No, Mo), n (%)
Benign Disorder,* n (%)
50.6 (14.3) 50.2 (14.4) 51.6 (13.9)
19–85 19–85 28–77
95 (49.7) 62 (45.2) 33 (61.2)
4 (2.1) 3 (2.2) 1 (1.8)
92 (48.2) 72 (52.6) 20 (37)
* Benign disorder: multinodular goiter, toxic or nontoxic adenoma, and Hashimoto thyroiditis. Abbreviation: max, maximum; min, minimum; SD, standard deviation.
(4) Aerodynamics. MPT was measured during a sustained /α/ after a deep inspiration at comfortable pitch and loudness (expressed in seconds). The maximum value of the three trials was considered as the valid one. (5) Self-evaluation. The Greek version of the VHI questionnaire was handed out to all patients.14 It is a 30-point scale questionnaire with three subscales (10 points each): functional (VHI-F), physical (VHI-P), and emotional (VHIE). Each item is scored from 0 to 4, according to the impact of the statement on the patient’s everyday life (0 = never, 1 = almost never, 2 = sometimes, 3 = almost always, and 4 = always). Subscale scores range from 0 to 40 and the overall score ranges from 0 to 120. A total score of 0–30 indicates mild dysphonia, 31–60 moderate dysphonia, 61–90 severe dysphonia, and 91–120 very severe dysphonia.5 Data analysis Voice outcome measures were computed by using the SPSS statistical software package (SPSS, v.22.0; SPSS, Inc., Chicago, IL). Changes in voice pitch, jitter, shimmer, and MPT values were assessed with a paired t test. The scores of NHR, VHI, and GRBAS were highly skewed and were analyzed with the nonparametric Wilcoxon test. Spearman correlations were used to evaluate associations of subjective evaluation with perceptual and other clinical parameters. Statistical analysis was performed for the total study population as well as for the two age subgroups and control groups.
RESULTS Of the total of 204 patients recruited, 4 were excluded due to unilateral RLN paresis after thyroidectomy; in 2 patients, laryngostroboscopy revealed focal folds’ mucosal abnormalities due to intubation; and 7 were lost to follow-up. After the implementation of the exclusion criteria, the study population included 191 patients with a mean age 50.6 years (range 19– 85 years, standard deviation [SD]: 14.3); 137 (73%) were women, and 54 (28%) were men. Forty-six patients were <40 years old, and 145 patients were ≥40 years old. Histology revealed a T1 or T2 well-differentiated thyroid carcinoma in 52% of the patients (Table 1). Summary descriptive statistics are presented as minimum and maximum values, as well as mean values with SD (Table 2). Control group A consisted of six men with a mean age of 54.8 years (SD: 18.3) and eight women with a mean age of 54.6 years (SD: 11.3). Control group B consisted of 11 men with a mean age 49.5 years (SD: 14.7) and 13 women with mean age 50.3 years (SD: 17.9). Table 3 shows the paired t test results for control groups A and B. No statistically significant difference was found in any voice parameter, between preoperative and 1-week postoperative assessments. Thus, control patients did not undergo the 8-week evaluation. Regarding the study population, a statistically significant difference was found between preoperative and 1-week postthyroidectomy evaluations for the total study population, as well as for the ≥40 years’ age subgroup for pitch (P < 0.001), jitter (P < 0.05), NHR (P < 0.001), MPT (P < 0.001), GRBAS score (P < 0.001), and VHI score (P < 0.001) (Table 4). The <40 years’
TABLE 2. Acoustic, Aerodynamic, Perceptual, and Subjective Parameters Preoperatively and 1 and and 8 Weeks Post Thyroidectomy Preoperatively
Postoperatively (1 wk)
Postoperatively (8 wk)
Analysis
Parameter
Min-Max
Mean (SD)
Min-Max
Mean (SD)
Min-Max
Mean (SD)
Acoustic
Pitch, men Pitch, women Jitter Shimmer NHR MPT GRBAS VHI
105.9–288 88.1–294.1 0–1.4 1.32–12.1 0–0.14 6.6–36.7 0–8 0–52
199.76 (49.9) 203.1 (47.1) 0.37 (0.19) 4.3 (2.4) 0.02 (0.02) 18.7 (5.7) 0.86 (1.6) 4.2 (7.1)
89.9–241.4 82.1–262.1 0.13–2.6 0–17.5 0–0.73 5–30.8 0–13 0–59
175 (43.02) 178.1 (43.1) 0.44 (0.32) 4.5 (2.6) 0.03 (0.06) 14.2 (4.49) 1.6 (1.9) 8.5 (11.7)
109.3–286.6 89.4–287 0.12–0.97 0–14.34 0–0.23 8.3–34.7 0–7 0–42
197.8 (48.04) 201.5 (45.8) 0.34 (0.17) 4.18 (2.25) 0.02 (0.03) 18.2 (5.1) 0.76 (1.2) 4.5 (7.3)
Aerodynamic Perceptual Subjective
Abbreviations: GRBAS, grade, roughness, breathiness, asthenia, and strain; max, maximum; min, minimum; MPT, maximum phonation time; NHR, noiseto-harmonic ratio; SD, standard deviation; VHI, Voice Handicap Index.
ARTICLE IN PRESS 4
Journal of Voice, Vol. ■■, No. ■■, 2017
TABLE 3. Paired t Test Results for Control Groups A and B, Between Voice Parameters Preop and 1 Week Postop Control group B: Men, n = 11; Women, n = 13 Non-neck Surgery
Control Group A: Men, n = 6; Women, n = 8 Neck Surgery with No RLN Risk and No Strap Muscle Dissection Parameter
Preop Mean (SD)
Postop 1 wk Mean (SD)
Pitch, men 127.8 (20.5) 130.4 (26.3) Pitch, women 221.08 (39.2) 220.4 (38.8) Jitter 0.33 (0.18) 0.32 (0.11) Shimmer 3.38 (1.3) 2.99 (1.58) NHR 0.007 (0.007) 0.0086 (0.006) MPT 18.02 (3.7) 17.3 (3.61) GRBAS 0.21 (0.4) 0.5 (0.8) VHI 1.21 (1.57) 3.43 (8.12)
95% Confidence Interval [−21.3 to 16.18] [−5.4 to 6.77] [−0.09 to 0.11] [−1.02 to 1.8] [−1.09 to 2.55]
P
Preop Mean (SD)
Postop 1 wk Mean (SD)
0.74 124.3 (17.6) 123.7 (16.4) 0.8 225.5 (11,7) 227.23 (13.9) 0.84 0.35 (0.12) 0.33 (0.10) 0.55 4.95 (1.75) 4.16 (1.73) 0.5 0.02 (0.02) 0.19 (0.017) 0.4 19.9 (5.9) 20.7 (5.8) 0.16 0.67 (1.09) 0.67 (0.81) 0.3 1.42 (1.74) 1.46 (1.8)
95% Confidence Interval [−2.7 to 3.9] [−6.9 to 3.45] [−0.009 to 0.6] [−0.17 to 1.7] [−1.9 to 0.45]
P 0.69 0.47 0.14 0.1 0.36 0.21 1.0 0.9
Note: NHR, GRBAS, and VHI were analyzed with nonparametric Wilcoxon test. Abbreviations: GRBAS, grade, roughness, breathiness, asthenia, and strain; MPT, maximum phonation time; NHR, noise-to-harmonic ratio; postop, postoperatively; preop, preoperatively; RLN, recurrent laryngeal nerve; SD, standard deviation; VHI, Voice Handicap Index.
age subgroup showed a statistically significant difference in pitch, MPT, and GRBAS score between preoperative and 1-week postthyroidectomy evaluations (Table 4). Eight weeks postoperatively, a statistically significant difference was found for the whole study population for NHR (P < 0.05) and MPT (P < 0.05), as well as for pitch in women (P = 0.024) in the ≥40 years’ age subgroup, compared with the initial evaluation. None of the parameters showed a statistically significant difference in the <40 years’ age subgroup at 8 weeks’ evaluation (P > 0.05) (Table 5). Table 6 shows Spearman correlation coefficients between subjective, perceptual, and objective parameters preoperatively and 1 and 8 weeks postoperatively for the total study population. The VHI scores correlated significantly (P < 0.01) with the GRBAS scores preoperatively (rho = 0.324) and postoperatively at 1-week (rho = 0.543) and 8-week (rho = 0.478) evaluations. Furthermore, the VHI scores correlated significantly (rho = 0.203, P < 0.01) with pitch 1 week postoperatively. The GRBAS scores showed significant correlation not only with VHI but also with the following acoustic parameters: pitch (rho = 0.264, P < 0.01; rho = 0.148, P < 0.05), shimmer (rho = 0.172, P < 0.05; rho = 0.168, P < 0.05), and NHR (rho = 0.148, P < 0.05; rho = 0.17, P < 0.05) at 1 and 8 weeks after thyroidectomy, respectively. DISCUSSION Thyroidectomy is one of the most frequent operations in our department where patients undergo surgery due to benign, suspicious, or malignant pathology. Traditionally, voice changes after thyroidectomy were attributed to iatrogenic RLN injury. However, patients even after RLN-preserved thyroidectomy frequently complain of voice changes (ie, hoarseness, fatigue, difficulty in singing or yelling), as well as swallowing problems. These alterations to vocal and throat functions may decrease patients’ QoL, the so-called post-thyroidectomy syndrome.15,16 Fibrosis, vascular changes, injury of the external branch of the SLN, and psychosocial reactions have been implicated in the etiology of this syndrome.10,17
In the present prospective study, a multidimensional voice evaluation was used, consisting of both objective and subjective measures, to assess voice changes after total thyroidectomy. The first post-thyroidectomy evaluation took place as early as the seventh postoperative day, under the condition of normal stroboscopic findings, that is, no indication of laryngeal mucosal or laryngeal nerve damage. Unilateral RLN injury leads to early hoarseness with a hypokinetic or fixed vocal fold at the paramedian position, loss of vibratory mucosal wave, and incomplete glottic closure. Stroboscopic characteristics of SLN injury include thinned, shortened, and bowed unilateral vocal fold; lack of tension and lower vertical plane of the paretic vocal fold; oblique glottis deviating to the paretic side posteriorly; and irregular vibratory motions during phonation.18 The uniform way of patients’ intervention (total thyroidectomy by the same surgeon) and evaluation (same phoniatric specialist and speech therapist) is one of the main advantages of the present study. The analysis of the results for the control groups showed no statistically significant difference in all voice parameters tested a week postoperatively (Table 3). Thus, we can conclude that voice changes in the study population are related to thyroidectomy and not to endotracheal intubation, or to postoperative neck discomfort or pain. Furthermore, we analyzed a quite large population subdivided into two age subgroups, a <40 years’ age group and a ≥40 years’ age group. Finally, we evaluated study parameters at 8 weeks, an intermediate time interval, even though previous studies report that voice disturbances are resolved 3–6 months after thyroidectomy.19,20 With this intermediate assessment, we were able to find out if there exist parameters that show earlier recovery and whether age affects voice outcome at this interval. Study limitations include the relatively small number of patients in the subgroup aged <40 years, the aerodynamic evaluation only by MPT, and the lack of a long-term evaluation which is currently contacted. We found a statistically significant difference in all evaluated parameters, with the exception of shimmer, between the
Parameter
Preop Total Mean (SD)
Postop Total 1 wk Mean (SD)
[95% CI] P
Postop Age <40 1 wk Mean (SD)
Preop Age <40 Mean (SD)
199.7 (49.9) 175.7 (43.02)
187.05 (54.76) 176.54 (40.73) 0.4 (0.3) 4.02 (2.07)
1.5 (1.7) 6.8 (9.7)
[8.05–20.36] 200.2 (50.44) 0.049* [16.7–30.25] 204.2 (48.3) <0.001* [−0.16 to 0.04] 0.38 (0.19) 0.25 [−0.79 to 0.61] 4.4 (2.1) 0.79 0.27 0.018 (0.02) [2.8–5.17] 19 (5.7) <0.001* 0.03* 0.81 (1.4) 0.06 3.9 (6.6)
Postop Age ≥40 1 wk Mean (SD) 173.7 (41.06) 178.7 (44.2) 0.46 (0.3) 4.7 (2.7) 0.03 (0.07) 14 (7.2) 1.6 (2.0) 9.06 (12.2)
[95% CI] P [18.1–34.75] <0.001* [21.0–30.06] <0.001* [−0.13 to −0.2] 0.007* [−0.85 to 0.33] 0.149 <0.001* [3.12–5.41] <0.001* <0.001* <0.001*
* Shows a statistically significant difference between the compared parameters. Abbreviations: CI, confidence interval; GRBAS, grade, roughness, breathiness, asthenia, and strain; MPT, maximum phonation time; NHR, noise-to-harmonic ratio; postop, postoperatively; preop, preoperatively; SD, standard deviation; VHI, Voice Handicap Index.
TABLE 5. Analysis of Acoustic, Aerodynamic, Perceptual, and Subjective Parameters Preop and 8 Weeks Postop for Total Study Population as Well as for Age Subgroups
Parameter
Preop Total Mean (SD)
Postop Total 8 wk Mean (SD)
Pitch, men
199.7 (49.9)
197.8 (48.0)
Pitch, women 203.1 (47.1)
201.5 (45.8)
Jitter
0.37 (0.19)
0.34 (0.7)
4.3 (2.4)
4.18 (2.25)
NHR MPT
0.02 (0.02) 18.7 (5.7)
0.02 (0.03) 18.2 (5.1)
GRBAS VHI
0.86 (1.6) 4.2 (7.1)
0.76 (1.2) 4.5 (7.3)
Shimmer
[95% CI] P
Preop Age <40 Mean (SD)
[−2.1 to 6.02] 197.26 (50.4) 0.34 [−0.58 to 3.8] 200.03 (44.2) 0.15 [−0.009 to 0.006] 0.34 (0.19) 0.14 [−0.25 to 0.48] 3.93 (1.9) 0.53 0.016* 0.01 (0.2) [0.03–0.9] 17.8 (5.9) 0.03* 0.78 1.02 (1.8) 0.82 5.07 (8.4)
Postop Age <40 8 wk Mean (SD) 195.65 (51.2) 200.08 (41.47) 0.34 (0.17) 3.97 (2.1) 0.02 (0.02) 17.8 (5.4) 0.95 (1.3) 3.9 (6.01)
[95% CI] P
Preop Age ≥40 Mean (SD)
[−3.9 to 7.12] 200.2 (48.3) 0.51 [−6.33 to 6.2] 204.26 (48.32) 0.98 [−0.073 to 0.07] 0.38 (0.19) 0.98 [−0.79 to 0.71] 4.4 (2.1) 0.9 0.56 0.018 (0.02) [−0.9 to 0.89] 19 (5.7) 0.97 0.72 0.81 (1.4) 0.56 3.9 (6.6)
Postop Age ≥40 8 Weeks Mean (SD) 198.18 (48.06) 202.01(47.58) 0.34 (0.16) 4.2 (2.2) 0.024 (0.03) 18.4 (4.9) 0.70 (1.2) 4.75 (7.7)
[95% CI] P [−2.7 to 6.7] 0.39 [0.30–4.1] 0.024* [−0.006 to 0.07] 0.09 [−0.26 to 0.59] 0.4 0.015* [0.1–1.19] 0.018* 0.57 0.53
5
* Shows a statistically significant difference between the compared parameters. Abbreviations: CI, confidence interval; GRBAS, grade, roughness, breathiness, asthenia, and strain; MPT, maximum phonation time; NHR, noise-to-harmonic ratio; postop, postoperatively; preop, preoperatively; SD, standard deviation; VHI, Voice Handicap Index.
ARTICLE IN PRESS
0.02 (0.02) 13.8 (4.5)
Preop Age ≥40 Mean (SD)
Voice Assessments After RLN-Preserved Thyroidectomy
197.26 (50.39) [16.7–31.35] <0.001* Pitch, women 203.1 (47.1) 178.1 (43.13) [21.2–28.7] 200.03 (44.2) <0.001* Jitter 0.37 (0.19) 0.44 (0.32) [−0.12 to −0.24] 0.34 (0.19) 0.003* Shimmer 4.3 (2.4) 4,5 (2.6) [−0.7 to 0.11] 3.93 (1.9) 0.154 NHR 0.02 (0.02) 0.03 (0.06) <0.001* 0.01 (0.2) MPT 18.7 (5.7) 14.2 (4.49) [4.03–5.11] 17.8 (5.9) <0.001* GRBAS 0.86 (1.6) 1.6 (1.9) <0.001* 1.02 (1.8) VHI 4.2 (7.1) 8.5 (11.7) <0.001* 5.07 (8.4) Pitch, men
[95% CI] P
Chariton E. Papadakis, et al
TABLE 4. Analysis of Acoustic, Aerodynamic, Perceptual, and Subjective Parameters Preop and 1 Week Postop for Total Study Population as Well as for Age Subgroups
ARTICLE IN PRESS 6
Journal of Voice, Vol. ■■, No. ■■, 2017
TABLE 6. Spearman Correlation Coefficients Between Subjective, Perceptual, and Objective Parameters Preop and at 1 and 8 Weeks Postop for the Total Study Population
Preop VHI GRBAS Postop 1 wk VHI GRBAS Postop 8 wk VHI GRBAS
Pitch
Jitter
Shimmer
NHR
MPT
GRBAS
VHI
−0.018 −0.154
0.65 0.123
0.10 0.22†
0.123 0.217†
−0.007 0.08
0.324† 1
1 0.324†
−0.203† −0.264†
0.033 0.020
0.47 0.172*
0.08 0.148*
−0.042 0.113
0.543† 1
1 0.543†
−0.042 −0.148*
0.004 0.028
0.083 0.168*
0.076 0.17*
0.048 −0.113
0.478† 1
1 0.478†
* Correlation is significant at the 0.05 level. † Correlation is significant at the 0.01 level. Abbreviations: GRBAS, grade, roughness, breathiness, asthenia, and strain; MPT, maximum phonation time; NHR, noise-to-harmonic ratio; postop, postoperatively; preop, preoperatively; VHI, Voice Handicap Index.
preoperative evaluation and the acoustic analysis 1 week after thyroidectomy for the whole study population (Table 4). At 8 weeks’ postoperative evaluation, the mean values of NHR and MPT showed a statistically significant difference compared to the preoperative evaluation. An interesting finding was that patients <40 years of age showed a statistically significant difference only in pitch, MPT, and GRBAS score, and these parameters’ mean values were restored to normal 8 weeks postoperatively (Tables 4 and 5). In the ≥40 years’ age subgroup at 1 week’s evaluation, statistical analysis showed the same results as those for the total study population, whereas at 8 weeks’ postthyroidectomy evaluation, there was still a statistically significant difference in pitch (women), NHR, and MPT. Patient’s age and extent of operation have been reported as independent factors for post-thyroidectomy lowered-pitch voice. Park et al found that female patients older than 32 years and those who underwent total thyroidectomy showed a statistically significant lower pitch compared to younger patients and less extended surgery.19 It has been suggested that calcified laryngeal cartilages, mucosal alterations, and laryngeal muscle atrophy, seen with the advance of age, are some of the causes of this functional deficit. Total thyroidectomy causes greater soft tissue injury, compared to lobectomy, and shows an early adverse effect on voice, which usually is resolved by 6 months.20 Extrinsic laryngeal muscle function is to lengthen or shorten the vocal folds and to regulate pitch by changing the relation of the thyroid gland to the cartilage laryngeal frame. Strap muscle function can be altered by cutting or more often by lateral retraction or by laryngeal fixation in the surgical bed. Thus, it is believed that with the advance of age, compensatory muscle mechanisms poorly counterweight the extended soft tissue damage after a total thyroidectomy. In our study we found that patients after thyroidectomy did not show altered cycle-to-cycle variability in amplitude. Thus, shimmer was the perturbation parameter evaluated with no significant change, although it has been reported to differ significantly with harmonic-to-noise ratio in previous studies and has been suggested as a predictor of overall voice quality.21,22 NHR and
MPT were the parameters that differed significantly for the total population throughout the study. In the younger age subgroup (<40 years), we found no significant change 8 weeks postoperatively. MPT is a predictor of airflow and has been used by several studies for the evaluation of vocal aerodynamic function. Solomon et al found no significant change in MPT in patients who underwent thyroidectomy or had no neck surgery 2 weeks and 3 months postoperatively.23 It has been suggested that MPT is rather an indicator of poor laryngeal valving, and this finding may explain the significant change in the values in the older subgroup. As far as subjective parameters are concerned, VHI is a useful QoL tool for self-evaluation of dysphonia and GRBAS is the main perceptual tool for the assessment of voice abnormalities.5,6 These two instruments have been found to have a significant correlation throughout the study. VHI scores did not correlate with other clinical objective parameters except pitch 1 week after surgery. For the subgroup of patients under 40 years of age, there was no statistically significant change in the VHI scores even 1 week after thyroidectomy. Older patients revealed a significant VHI score worsening. However, previous studies have reported discrepancies between subjective and objective voice changes in thyroidectomized patients.17,24,25 Although RLNpreserved post-thyroidectomy voice changes usually subside after 3–6 months and some patients do not report symptoms, they have to be informed about the possible voice function alteration.19,20 Furthermore, when patients are voice professionals, women, and over 40 years of age, voice rehabilitation might be useful soon after surgery to obtain quick recovery. CONCLUSIONS Objective voice changes are common in the majority of the thyroidectomized patients in the early postoperative period. These changes are related to the thyroidectomy per se and not to endotracheal intubation or to neck discomfort experienced in other neck operations. In younger patients (<40 years of age), a discordance between VHI and acoustic parameters exists. Furthermore, subjective tools do not correlate significantly with
ARTICLE IN PRESS Chariton E. Papadakis, et al
Voice Assessments After RLN-Preserved Thyroidectomy
objective acoustic parameters but show a significant correlation with perceptual assessment. Young patients return to their previous voice quality as early as 8 weeks’ post-thyroidectomy evaluation. Although older patients (≥40 years of age) report no abnormalities and the perceptual evaluation is similar to the preoperative one, they are found with acoustic and aerodynamic changes in parameters such as pitch (women), NHR, and MPT. REFERENCES 1. Finck C. Laryngeal dysfunction after thyroid surgery: diagnosis, evaluation and treatment. Acta Chir Belg. 2006;106:378–387. 2. Vilkman E, Sonninen A, Hurme P, et al. External laryngeal frame function in voice production revisited: a review. J Voice. 1996;10:78–92. 3. Hong KH, Kim YK. Phonatory characteristics of patients undergoing thyroidectomy without laryngeal nerve injury. Otolaryngol Head Neck Surg. 1997;117:399–404. 4. Zraick RI, Risner BY. Assessment of quality of life in persons with voice disorders. Curr Opin Otolaryngol Head Neck Surg. 2008;16:188–193. 5. Jacobson BH, Johnson A, Grywalsky C, et al. The voice handicap index (VHI): development and validation. Am J Speech Lang Pathol. 1997;6:66–70. 6. Dejonckere PH, Bradley P, Clemente P, et al. A basic protocol for functional assessment of voice pathology, especially for investigating the efficacy of (phonosurgical) treatments and evaluating new assessment techniques. Guideline elaborated by the Committee on Phoniatrics of the European Laryngological Society (ELS). Eur Arch Otorhinolaryngol. 2001;258:77–82. 7. De Bodt MS, Van de Heyning PH, Wuyts FL, et al. The perceptual evaluation of voice disorders. The perceptual evaluation of voice disorders. Acta Otorhinolaryngol Belg. 1996;50:283–291. 8. Kent RD, Read C. Introduction to the acoustic analysis of speech. In: The Acoustic Analysis of Speech. 2nd ed. Singular Thomson Learning; 2002:53– 103. 9. Soylu L, Ozbas S, Uslu HY, et al. The evaluation of the causes of subjective voice disturbances after thyroid surgery. Am J Surg. 2007;194:317–322. 10. Sinagra DL, Montesinos MR, Tacchi VA, et al. Voice changes after thyroidectomy without recurrent laryngeal nerve injury. J Am Coll Surg. 2004;199:556–560.
7
11. Van Lierde K, D’Haeseleer E, Wuyts FL, et al. Impact of thyroidectomy without laryngeal nerve injury on vocal quality characteristics: an objective multiparameter approach. Laryngoscope. 2010;120:338–345. 12. Nam IC, Bae JS, Chae BJ, et al. Therapeutic approach to patients with a lower-pitched voice after thyroidectomy. World J Surg. 2013;37:1940–1950. 13. Hirano M. Clinical Examination of Voice. New York, NY: Springer-Verlag Wien; 1981. 14. Helidoni ME, Murray T, Moschandreas J, et al. Cross-cultural adaptation and validation of the voice handicap index into Greek. J Voice. 2010;24:221– 227. 15. Lombardi CP, Raffaelli M, D’Alatri L, et al. Voice and swallowing changes after thyroidectomy in patients without inferior laryngeal nerve injuries. Surgery. 2006;140:1026–1032. 16. Ryu J, Ryu Y, Jung Y, et al. Extent of thyroidectomy affects vocal and throat functions: a prospective observational study of lobectomy versus total thyroidectomy. Surgery. 2013;154:611–620. 17. Stojadinovic A, Shaha AR, Orlikoff RF, et al. Prospective functional voice assessment in patients undergoing thyroid surgery. Ann Surg. 2002;236:823– 832. 18. Dursun G, Sataloff RT, Spiegel JR, et al. Superior laryngeal nerve paresis and paralysis. J Voice. 1996;10:206–211. 19. Park J-O, Bae J-S, Lee S-H, et al. The long-term prognosis of voice pitch change in female patients after thyroid surgery. World J Surg. 2016;40:2382– 2390. [Epub ahead of print]. 20. Vicente DA, Solomon NP, Avital I, et al. Voice outcomes after total thyroidectomy, partial thyroidectomy, or non-neck surgery using a prospective multifactorial assessment. J Am Coll Surg. 2014;219:152–163. 21. Maryn Y, Corthals P, Van Cauwenberge P, et al. Toward improved ecological validity in the acoustic measurement of overall voice quality: combining continuous speech and sustained vowels. J Voice. 2010;24:540–555. 22. Solomon N, Awan S, Helou L, et al. Acoustic analyses of thyroidectomyrelated changes in vowel phonation. J Voice. 2012;26:711–720. 23. Solomon N, Helou L, Makashay M, et al. Aerodynamic evaluation of the post-thyroidectomy voice. J Voice. 2012;26:454–461. 24. McIvor NP, Flint DJ, Gillibrand J, et al. Thyroid surgery and voice-related outcomes. Aust N Z J Surg. 2000;70:179–183. 25. Debruyne F, Ostyn F, Delaere P, et al. Acoustic analysis of the speaking voice after thyroidectomy. J Voice. 1997;11:479–482.