Voice and swallowing changes afterthyroidectomy in patients withoutinferior laryngeal nerve injuriesCelestino Pio Lombardi, MD,a Marco Raffaelli, MD,a Lucia D’Alatri, MD,b

Maria Raffaella Marchese, MD,b Mario Rigante, MD,b Gaetano Paludetti, MD,b andRocco Bellantone, MD,a Rome, Italy

Background. We performed a prospective analysis on voice and swallowing alterations following totalthyroidectomy (TT), in the absence of recurrent nerve injury.Methods. Patients aged 21 to 65 years undergoing TT, in the absence of laryngeal/pulmonary disease,previous neck surgery, or malignant diseases, were subjected to videostrobolaryngoscopy (VSL), acousticvoice analysis (AVA), and maximum phonation time (MPT) tests preoperatively and 3 monthspostoperatively. Voice impairment scores (VIS) and swallowing impairment scores (SIS) were obtainedpreoperatively, and at 1 week, 1 month, and 3 months postoperatively.Results. Among the 127 selected patients, 39 completed the postoperative evaluation. No recurrentnerve injury was observed during the postoperative VSL in any of the patients. Preoperative andpostoperative AVA and MPT scores did not differ significantly. The mean postoperative VIS wassignificantly higher than the preoperative VIS at 1 week and 1 month after TT (13.7 and 9.6 vs 4.4,respectively; P � .05) but not 3 months after TT (6.7). The mean SIS was higher than thepreoperative SIS at 1 week, 1 month, and 3 months after TT (10.3, 6.0, and 2.8 vs 0.5, respectively;P � .05).Conclusions. Physicians should inform patients that transient voice and swallowing symptoms mayoccur following total thyroidectomy, and our data suggest mild symptoms may occur in the majority ofoperated patients. (Surgery 2006;140:1026-34.)

From the Division of Endocrine Surgery, Department of Surgery,a and the Institute of Otolaryngology,b Università

Cattolica del Sacro Cuore, Rome, Italy

phonatory and swallowing alterations fol-lowing thyroidectomy are well-known outcomes oflaryngeal nerve injury.1-4 A lesion of the recurrentinferior laryngeal nerve (ILN) is considered to bethe main cause of voice changes.1-6 Injury to theexternal branch of the superior laryngeal nerve(EBSLN) is another complication that has signifi-cant implications, especially for patients in voice-related professions.3,7,8 Prevention of voiceimpairment is the primary goal of thyroid surgery.Despite the existence of surgical adjuncts, such as

Presented at the 27th Annual Meeting of the American Associ-ation of Endocrine Surgeons, New York, New York, May, 2006.

Accepted for publication August 8, 2006.

Reprint requests: Dr Marco Raffaelli, Divisione di Endocri-nochirurgia, Istituto di Clinica Chirurgica, Università Cattolicadel Sacro Cuore, L.go A. Gemelli 8, 00168 – Rome – Italy.E-mail: marcoraffaelli@rm.unicatt.it.

0039-6060/$ - see front matter

© 2006 Mosby, Inc. All rights reserved.

doi:10.1016/j.surg.2006.08.008

1026 SURGERY

nerve monitoring,6,9 there is no effective substitutefor a thorough knowledge of neck anatomy andmeticulous surgical techniques to identify and pro-tect laryngeal nerves during thyroidectomy.5,9

However, most of the voice alterations thatoccur following thyroidectomy are self-limitedand not related to impaired nerve func-tion.3,4,10-12 These alterations have been attrib-uted to arytenoid trauma following endotrachealintubation, surgical trauma, laryngotracheal fix-ation of the strap muscles, or lesion of the peri-thyroidal neural plexus.2-4,11,12

Swallowing problems and aspiration have beenreported in about 30% of patients with unilateralILN injury.2,9,13 These symptoms may arise fromcordal or cricopharyngeal dysfunctions (the formerof which relates to the inability to close the glottiscompletely) related to alterations of the motor andsensory branches of the ILN to the inferior con-strictor muscle and laryngeal mucosa.9 Moreover,some patients report nonspecific swallowing

changes and discomfort, and occasional dysphagia

Surgery Lombardi et al 1027Volume 140, Number 6

after thyroid resection, even long after the surgicalprocedure.2

Subjective voice changes have been investigatedin recent years3,4,10-12; however, less attention hasbeen paid to swallowing alterations after uncompli-cated thyroidectomy. These symptoms are usuallydismissed by the clinician or attributed to orotra-cheal intubation, because they do not have a dra-matic impact on the patients’ quality of life, andcannot be translated into objective data by appro-priate complementary tests. Nevertheless, patientsare usually concerned if these symptoms occur.

In this prospective study, we analyzed patientswho were undergoing total thyroidectomy (TT) toidentify voice and swallowing alterations in the ab-sence of clinical ILN injury. We also looked atpossible causes of these symptoms and their impli-cations.

MATERIAL AND METHODSPatients who underwent TT between May and

September 2005 were considered eligible for thisstudy. Exclusion criteria included being less than21 or greater than 65 years of age; prior vocal foldparalysis; history of voice or laryngeal diseases re-quiring therapy, speech disorders or pulmonarydiseases; previous neck surgery; and malignant dis-eases. Patients showing post-thyroidectomy evi-dence of ILN or EBSLN palsy also were excluded.Among the 188 patients who underwent TT duringthis period, 127 met the inclusion criteria andsigned an informed consent to participate in thestudy.

Surgical technique. Patients undergoing bothconventional and video-assisted thyroidectomywere included in the study. All surgical procedureswere performed by an experienced endocrine sur-geon or by a resident operating under supervision.TT was defined as total bilateral extracapsular thy-roidectomy. Surgical techniques for conventionaland video-assisted thyroidectomy have been de-scribed previously.14,15 Typically, after skin inci-sion, the strap muscles are separated along themidline and then reapproximated at the end of theoperation. In every thyroid surgery, we recommenda systematic and prompt identification of theILN.14,15 Following its identification, the surgeonshould track the ILN upward to the point of entryinto the larynx. Every attempt should be made topreserve all of its branches during dissection. If thenerve is not found at its usual place, the surgeonshould be aware of the possibility of a nonrecurrentILN. Because the EBSLN is not exposed routinely,the superior thyroid artery and vein are ligated

individually, close to the thyroid capsule to avoid

injury to the EBSLN. When the EBSLN is notreadily identifiable, no further dissection is pur-sued to avoid inadvertent nerve injury. In addition,it is important to avoid producing lesions of thecricothyroid muscle from electrocoagulation ormanual retraction.

Patient assessment. Vocal and swallowing func-tions were assessed with videostrobolaryngoscopy(VSL), acoustic analysis, and patients’ subjectivevoice and swallowing evaluations. The assessmentprotocol was planned as follows: the VSL was con-ducted the day before operation, the second post-operative day, and 3 months after the operation; anacoustic analysis was performed before the opera-tion, and again 3 months after operation; patients’subjective voice and swallowing evaluations wereconducted preoperatively, 1 week, 1 month, and 3months following operation.

Videostrobolaryngoscopy. The VSLs were per-formed with a 70-degree rigid laryngoscope and aPentax stroboscope (RLS 9100B model; Kay Elem-etrics, Lincoln Park, NJ) to assess vocal fold motionimpairment (indicative of ILN injury), as well asbowing and inferior displacement of the vocal fold;regularity and symmetry of the mucosal travelingwave; and degree of glottic closure (all indicative ofEBSLN injury).

Acoustic voice Analysis (AVA). Objective voiceanalysis was performed with a multidimensionalvoice program (MDVP) (model 5105; Kay Elemet-rics) and a voice range profile program (VRP)(model 4326; Kay Elemetrics), using a computer-ized speech lab (model 4300B; Kay Elemetrics),recording with a unidirectional dynamic micro-phone (Shure, Evanston, Ill), which was positionedat a 45-degree angle and at distance of 20 cm fromthe patient’s mouth. The microphone saturationinput was fixed at 6/9 of CH1, and the environ-mental noise was �30 dB SPL. For the MDVP anal-ysis, after 3 training emissions, a sustained vowel/a/, produced at a comfortable pitch and loudnesslevel for at least 5 seconds, was recorded with asampling rate of 50 kHz. We selected an interval of3 seconds from the mid-portion of each vowel. Thefollowing parameters were analyzed: mean fre-quency (F0, Hz), noise-to-harmonic ratio (NHR),jitter (%), and shimmer (%). For the VRP analysis,the patients were instructed to phonate using asustained vowel /a/ as loud and as soft as possiblefrom the lowest to the highest frequencies. Anautomated procedure was used to obtain a VRP.Measurements derived from the VRP included theassessment of the lowest (Flow) and the highest(Fhigh) frequencies (Hz); the number of semitones;

and the mean intensity ranges for all semitones

thus th

1028 Lombardi et al SurgeryDecember 2006

(mean dB range, dB, SPL). Finally, the maximumphonation time (MPT) was obtained by having thepatient sustain the vowel /a/ for as long as possibleon a single breath. The longest of 3 attempts wasrecorded as the MPT.

Subjective voice and swallowing evaluation. Aseries of questions related to the frequencies ofvoice and swallowing abnormalities were asked ofeach patient, using 2 questionnaires specifically de-fined for this study (Tables I and II). The scoresresulting from these questionnaires were defined asthe voice impairment score (VIS) and swallowingimpairment score (SIS). The VIS varied from aminimum score of 0 (no voice alterations) to amaximum of 40 (highest voice impairment),whereas the SIS ranged from a minimum of 0 (noswallowing alterations) to a maximum of 20 (high-est swallowing impairment).

Statistical analysis. Statistical analyses were per-formed with the software package SPSS (version10.0 for Windows; SPSS Inc., Chicago, Ill). Thechi-square test was used for categorical variables,and the analysis of variance (ANOVA) was used forcontinuous variables. A P value less than .05 wasconsidered to be significant.

RESULTSAmong the 127 patient cohort, 4 patients (3.1%)

experienced transient laryngeal nerve palsy, docu-mented by postoperative VSL, and were excludedfrom this study. All patients recovered within 2months of operation. Among the remaining 123patients (68.3%), 84 refused to undergo a fol-low-up consultation because of the absence of anysubjective symptoms 1 month after the operationand/or they resided too far from the hospital. Ofthe initial 127 patients, 39 (31.7%) completed the

Table I. Subjective voice evaluation questionnaire

Never 0

1. My voice is hoarse.2. My voice is breathy and weak.3. I have difficulty in singing and yelling.4. The pitch of my voice is lowered.5. My voice is monotonous.6. I run out of air when I talk.7. My voice changes during the day.8. I use a great effort to speak.9. The clarity of my voice is unpredictable.10. I feel vocal fatigue.VIS score � Total

�VIS, voice impairment score.Each individual’s VIS was the sum of all responses in the questionnaire,

3-month follow-up procedure and were included in

the study. All of these patients were female, with amean age of 42.6 � 11.2 years (range, 24 to 65). Afinal histological examination of the resected tissueconfirmed benign diseases in all the patient cases.The mean weight (� standard deviation [SD]) ofthe surgical specimen was 25.8 � 9.2 grams (range,17 to 54).

According to the VSL data, none of the patientsdemonstrated any vocal fold movement limitationand/or signs of palsy of the EBSLN before surgery,on the second postoperative day, or 3 months afterthe operation. The mean MPT was 19.1 secondsbefore surgery and 19.7 seconds 3 months later (P� .05; Table III). The objective parameters ob-tained from the MDVP and VRP analyses are sum-marized in Table III. No significant differenceswere found among the preoperative and postoper-ative mean values of the MDVP and VRP parame-ters before and after TT (P � .05; Table III).

Subjective voice and swallowing evaluation. Thenumber of patients reporting each vocal and swal-lowing symptom at each time point (preopera-tively, 1 week, 1 month, and 3 monthspostoperatively) is shown in Tables IV and V. Pre-operatively, 15 of 39 patients presented with voicesymptoms and 2 presented with swallowing symp-toms. Most of the patients complained of voice andswallowing symptoms 1 week after surgery (31 of 39patients; 79.5%).

The prevalence of preoperative and postopera-tive voice symptoms is shown in Fig 1. At 1 weekafter surgery, 31 of 39 patients (79.5%) demon-strated hoarseness; 18 of 39 (46.1%) had weak andbreathy voice; 25 of 39 patients (64.1%) had diffi-culty singing and yelling; 16 of 39 (41.0%) hadlower-pitched voices; 14 of 39 (35.9%) possessedmonotonous voices; 18 of 39 (46.1%) exhibited

early never 1 Sometimes 2 Often 3 Always 4

e possible score range was 0 to 40.

N

reduced air intake capacity during speech; 21 of 39

naire, th

Surgery Lombardi et al 1029Volume 140, Number 6

(53.8%) experienced voice changes throughoutthe day; 14 of 39 (35.9%) had great difficulty speak-ing; 10 of 39 (25.6%) had an unpredictable changeof voice clarity; and 16 of 39 patients (41.0%) re-ported vocal fatigue (Table IV).

Four patients reporting preoperative voice symp-toms showed early (1-week) postoperative improve-ment. In general, the voice changes graduallyimproved over time (Fig 1). The mean VIS wassignificantly increased at 1 week and 1 month fol-lowing surgery (P � .05; Table VI). At 3 monthsafter surgery, the VIS remained higher than that atpreoperative levels, but the difference was not sig-nificant (P � .05). After an initial impairment, theVIS showed a trend toward the preoperative levels.At 3 months after surgery, the patients’ VIS valueswere significantly improved with respect to thoseobtained at 1 week and 1 month after TT (P � .05;Table VI).

The prevalence of preoperative and postopera-tive swallowing symptoms is summarized in Fig 2. At1 week after surgery, 21 of 39 patients (53.8%)

Table II. Subjective swallowing evaluation question

N

1. I use a great effort to swallow.2. I feel a throat obstacle during swallowing.3. I feel pharyngeal annoyance during bolus transit.4. I cough during bolus transit.5. I feel sensation of foreign body in pharynx.SIS score � Total

�SIS, Swallowing impairment score.Each individual’s SIS score was the sum of all responses in the question

Table III. Comparison of preoperative and 3-month postoperative results of MDVP andVRP analyses

PreOPmean (� SD)

3-month POmean (� SD) P value

F0 (Hz) 190.4 (� 37.4) 191.5 (� 36.2) NSJitter (%) 0.5 (� 0.2) 0.5 (� 0.3) NSShimmer (%) 3.6 (� 1.2) 3.3 (� 1.0) NSNHR (dB) 0.1 (� 0.01) 0.1 (� 0.0) NSMPT (s) 19.1 (� 7.9) 19.7 (� 7.3) NSFlow (Hz) 162.0 (� 31.8) 158.1 (� 29.8) NSFhigh (Hz) 402.4 (� 74.9) 402.2 (� 90.2) NSNumber of

semitones16.8 (� 4.0) 17.0 (� 3.8) NS

Mean dB range 16.0 (� 3.3) 15.9 (� 3.4) NS

�MDVP, Multidimensional voice program; VRP, voice range profile;PreOP, preoperative; PO, postoperative; SD, standard deviation; NS, notsignificant; NHR, noise-to-harmonic ratio; MPT, maximum phonationtime.

reported difficulty in swallowing; 12 of 39 (30.8%)

had the sensation of a throat obstruction; 25 of 39(64.1%) reported an annoyance during bolus tran-sit; 16 of 39 (41.0%) coughed during bolus transit;and 16 of 39 (41.0%) experienced a sensation of aforeign body in the pharynx (Table V). The swal-lowing impairments improved gradually over time(Fig 2). The mean postoperative SIS was signifi-cantly increased above the preoperative SIS 1 week,1 month, and 3 months after surgery (P � .05;Table VI).

DISCUSSIONThe results of this study confirm that thyroidec-

tomized patients complain frequently of voice andswallowing changes, even after uncomplicated sur-geries. Even if these symptoms do not translate intoobjective findings, their presence elicits anxietyamong patients if they have not been informed ofthe risks. These symptoms should be regarded as atrue complication of thyroid surgery, even if theyare often neglected by the consulted physicians.For this reason, we investigated post-thyroidectomyoutcomes. Previously, studies investigating theprevalence of voice changes in patients undergoingthyroidectomy3,4,10-12 have demonstrated that mostof these effects may occur in the absence of VSLevidence of laryngeal nerve injuries (ILN, EBSLN).In our study, voice alterations after uncomplicatedthyroidectomy include a broad spectrum of symp-toms that are usually temporary and resolve rela-tively quickly. Several causes have been postulated,including orotracheal intubation3; modification ofthe vascular supply and venous drainage of thelarynx11; cricothyroid dysfunction3,4,11; laryngotra-cheal fixation with impairment of vertical move-ment3,10-12; strap muscle malfunction due todenervation or direct injury (section)10-12; localpain in the neck; and/or a psychologic reaction tothe postoperative situation.11 Because some objec-tive voice changes have been demonstrated insymptomatic patients after TT, a routine compre-hensive functional voice analysis has been sug-

Nearly never 1 Sometimes 2 Often 3 Always 4

us the possible score range was 0 to 20.

naire

ever 0

gested before and after thyroid surgery to provide

tionnair

g impai

1030 Lombardi et al SurgeryDecember 2006

an early diagnosis of postoperative vocal distur-bances and timely voice rehabilitation.3,11,12 How-ever, previous studies reported discrepanciesbetween subjective and objective voice changes inthyroidectomized patients,3,11,12 and noteworthydiscordance between objective deterioration invoice and subjective symptoms exists in about onethird of symptomatic patients.3,12 The results of our

Table IV. Subjective evaluation*

PreO

1. My voice is hoarse. 142. My voice is breathy and weak. 83. I have difficulty in singing and yelling. 64. The pitch of my voice is lowered. 25. My voice is monotonous. 66. I run out of air when I talk. 127. My voice changes during the day. 88. I use a great effort to speak. 129. The clarity of my voice is unpredictable. 1010. I feel vocal fatigue. 4

PreOP, preoperative; PO, postoperative.*The number of patients reporting subjective symptoms in the VIS ques

Table V. Subjective evaluation*

1. I use a great effort to swallow.2. I feel a throat obstacle during swallowing.3. I feel pharyngeal annoyance during bolus transit.4. I cough during bolus transit.5. I feel sensation of foreign body in pharynx.

�PreOP, preoperative; PO, postoperative.*The number of patients reporting subjective symptoms in the swallowin

Fig 1. The number of patients that exhibited hoarseness(1), breathy and weak voice (2), difficulty in singing andyelling (3), lower-pitched voice (4), monotonous voice(5), reduced inhalation capacity during speech (6), voicechanges during the day (7), great effort in speaking (8),unpredictable change of voice clarity (9), or vocal fatigue(10) before, 1 week, 1 month and 3 months after thy-roidectomy.

study demonstrate that the majority of patients

complained of vocal symptoms in the absence ofany objective alteration of AVA. Furthermore, AVAparameters may remain unaltered even in symp-tomatic patients.

The discrepancies between subjective and objec-tive voice and swallowing analyses could be ex-plained, in part, by the fact that, in our series,postoperative AVA was performed 3 months aftersurgery, when most of the subjective symptoms al-ready had resolved. This could be considered oneof the major limitations of this study. Another lim-itation could be that we elected not to use laryngealelectromyographic analysis to confirm the integrityof the laryngeal nerves. We discussed using electro-myography with some of our patients before initi-ating the study, and we noticed a generalreluctance on their part to participate in this inva-sive procedure. However, it recently has been dem-onstrated that the objective voice measures we usedfor AVA correlate well with electromyographic find-ings.16

Our investigation of swallowing impairment af-ter thyroidectomy in the absence of laryngeal nerveinjuries revealed that it is common for patients toexperience discomfort during swallowing or occa-sional dysphagia after thyroidectomy.2 These symp-toms usually resolve after a short period, but may

1-week PO 1-month PO 3-month PO

31 21 1618 10 1225 31 2516 12 1014 10 818 25 2121 18 1214 18 1610 14 1216 12 12

e at the different time points during the study.

P 1-week PO 1-month PO 3-month PO

21 10 612 10 425 16 216 10 1216 16 14

rment score questionnaire at the different time points during the study.

P

PreO

00022

persist long after surgery in approximately 30% of

Surgery Lombardi et al 1031Volume 140, Number 6

patients.2 To our knowledge, only 1 paper has beenpublished regarding the prevalence of long-termswallowing disturbances in patients who underwentuncomplicated thyroidectomy.2 In addition, noprospective study addressing the prevalence ofearly post-thyroidectomy swallowing changes hasbeen published thus far.

Because we did not establish a well-validatedmethod to measure subjective voice and swallowingproblems, we prepared 2 specifically designedquestionnaires that, with simple questions, providethe spectrum of symptoms typically reported bypatients. The severity of the symptoms was ex-pressed by a score of 0 to 4; the sum of all theresults gave a total score for vocal (VIS) and swal-lowing (SIS). These scores were not validated inother studies or in the general population, whichcould be regarded as a major limitation of ourstudy. Nonetheless, the scores provided a tool toevaluate the patients selected for this study.

Moreover, the results were quite surprising be-cause post-TT swallowing symptoms were reportedmore frequently than voice changes, even if theyrecurred in the same patients at the same time.Furthermore, the swallowing symptoms showed a

Table VI. Comparison of preoperative and postop

PreOP 1-week PO

VIS 4.4 (� 7.0) 13.7 (� 10.7)mean (� SD)SIS 0.5 (� 1.2) 10.3 (� 4.9)mean (� SD)

�VIS, Voice impairment score; SIS, swallowing impairment score; PreOP*These values are reported as mean (�SD).†VIS 3-month PO was significantly lower than 1-week PO and 1-month PO(refer to text for details).

Fig 2. The number of patients that presented with agreat effort to swallow (1), sensation of throat obstruc-tion during swallowing (2), pharyngeal annoyance dur-ing bolus transit (3), cough during bolus transit (4), orsensation of foreign body in pharynx (5) before and 3months after thyroidectomy.

tendency to persist longer than the voice changes.

Three months following operation, the mean SISwas still significantly impaired compared with thepreoperative SIS. Because of the frequent associa-tion between subjective voice and swallowingchanges, these parameters should have the sameanatomic cause. Orotracheal intubation could playsome role in the development of these symptoms,as already supposed for the voice changes.3,11 How-ever, Pereira et al2 compared thyroidectomizedand cholecystectomized patients retrospectively,and demonstrated that long-term symptoms arenot related to orotracheal intubation.

Because most of the patients included in thisstudy did not complete the follow-up evaluationdue to a lack of symptoms, the incidence of voiceand swallowing impairment after thyroidectomymay be overestimated. This should be consideredwhen evaluating the results. Another shortcomingof this study is the lack of long-term evaluation.Nonetheless, our results strongly confirm that voiceand swallowing symptoms are common even afteruncomplicated thyroidectomy, regardless of theirexact prevalence.

We believe that, in the absence of any VSL de-monstrable laryngeal nerve injury, the factors thatcould determine voice and swallowing disturbancesare related primarily to the normal healing process,or, possibly in a small percentage of cases, to somesubclinical postoperative hematoma. As proposedpreviously by others,2,3,11,12 after thyroid gland re-moval, the strap muscles become the sole supportof the laryngotracheal unit, which joins the musclesthrough scar tissue formation.2 This results in la-ryngotracheal fixation with impairment of verticalmovement. This problem seems more evident inthe case of strap muscle division.2,11,12 However,strap muscle functional deficits and scar formationmay result from their elongation and retractionwithout division.2 It is important to highlight thefact that we retract strap muscles routinely, withoutsectioning them.

We observed that most of the symptoms resolved

e VIS and SIS*

1-month PO 3-months PO P value

9.6 (� 8.9) 6.7 (� 7.1)† �.05

6.0 (� 2.9) 2.8 (� 2.8) �.05

rative; PO, postoperative; SD, standard deviation.

.05); VIS 3-month PO was not significantly higher than PreOP (P � .05)

erativ

, preope

(P �

within 3 months of TT. This strongly supports the

1032 Lombardi et al SurgeryDecember 2006

possibility that these functional problems were re-lated to the normal healing process. Moreover, afunctional component related to local neck painand an emotional and psychologic reaction to thepostoperative stress cannot be excluded. Nonethe-less, as demonstrated previously,2 upper aerodiges-tive symptoms can persist for years in some patientsafter an “uncomplicated” thyroidectomy. This sug-gests that some symptoms, besides normal scar for-mation, may be due to an intraoperative injury ofthe fine anastomotic branches connecting the ILNand the EBSLN and/or 1 of their anastomosis withthe sympathetic cervical chain.2,17 Some smallbranches of the laryngeal nerves, together withother cervical branches, participate in the auto-nomic, sensory, and motor innervation of pharyn-geal and laryngeal structures.2,17 This perivisceralnerve plexus could be involved in the motor andsensor innervation of the pharynx and the larynx.Injury to this plexus during thyroidectomy couldunderlie at least some of the postoperative discom-fort, which is variable in relation to the wide arrayof the nerve anatomy and disposition possibili-ties.2,17

Regardless of the cause, these post-thyroidec-tomy vocal and swallowing changes, which are re-ported frequently, can dramatically reduce apatient’s quality of life. Additional studies based ona larger cohort series and including data gatheredfrom a longer follow-up period are necessary todetermine the exact prevalence, clinical impact,pathogenesis, and outcomes of vocal and swallow-ing changes for prevention and treatment of thesesymptoms. We also believe that caregivers shouldbe aware of the specific subjective discomforts thatcommonly follow thyroid surgery. In addition, phy-sicians should inform patients that transient voiceand swallowing symptoms may frequently occur fol-lowing total thyroidectomy, even in absence of la-ryngeal nerve injuries.

REFERENCES1. Rosato L, Carlevato MT, De Toma G, Avenia N. Recurrent

laryngeal nerve damage and phonetic modifications aftertotal thyroidectomy: surgical malpractice only or predict-able sequence? World J Surg 2005;29:780-4.

2. Pereira JA, Girvent M, Sancho JJ, Parada C, Sitges-Serra A.Prevalence of long-term upper aero-digestive symptoms af-ter uncomplicated bilateral thyroidectomy. Surgery 2003;133:318-22.

3. Stojadinovic A, Shaha AR, Orlikoff RF, Nissan A, KornakMF, Singh B, et al. Prospective functional voice assessmentin patients undergoing thyroid surgery. Ann Surg 2002;236:823-32.

4. Sinagra DL, Montesinos MR, Tacchi VA, Moreno JC, Falco

JE, Mezzadri NA, et al. Voice changes after thyroidectomy

recurrent laryngeal nerve injury. J Am Coll Surg 2004;199:556-60.

5. Hermann M, Alk G, Roka R, Glaser K, Freismuth M. Laryn-geal recurrent nerve injury in surgery for benign thyroiddiseases. Effect of nerve dissection and impact of individualsurgeon in more than 27,000 nerves at risk. Ann Surg2002;235:261-8.

6. Dralle H, Sekulla C, Haerting J, Timmerman W, NeumannHJ, Kruse E, et al. Risk factors of paralysis and functionaloutcome after recurrent laryngeal nerve monitoring in thy-roid surgery. Surgery 2004;136:1310-22.

7. Bellantone R, Boscherini M, Lombardi CP, Bossola M, Ru-bino F, De Crea C, et al. Is the identification of the externalbranch of the superior laryngeal nerve mandatory in thyroidoperation? Results of a prospective randomized study. Sur-gery 2001;130:1055-9.

8. Friedman M, LoSavio P, Ibrahim H. Superior laryngealnerve identification and preservation in thyroidectomy.Arch Otolaryngol Head Neck Surg 2002;128:296-303.

9. Randolph GW. Surgical anatomy of the recurrent laryngealnerve. In: Randolph GW, editor. Surgery of the thyroid andparathyroid glands Philadelphia, PA: Saunders; 2003. p300-42.

10. Hong KH, Kim YK. Phonatory characteristics of patientsundergoing thyroidectomy without laryngeal nerve injury.Otolaryngol Head Neck Surg 1997;117:399-404.

11. Debruyne F, Ostyn F, Delaere P, Wellens W. Acoustic anal-ysis of the speaking voice after thyroidectomy. J Voice1997;11:479-82.

12. McIvor NP, Flint DJ, Gillibrand J, Morton RP. Thyroidsurgery and voice-related outcomes. Aust N Z J Surg2000;70:179-83.

13. Bou-Malhab F, Hans S, Perie S, Laccourreye O, Brasnu D.Swallowing disorders in unilateral recurrent laryngeal nerveparalysis. Ann Otolaryngol Chir Cervicofac 2000;117:26-33.

14. Bellantone R, Lombardi CP, Bossola M, Boscherini M, DeCrea C, Alesina PF, et al. Total thyroidectomy for manage-ment of benign thyroid disease: a review of 526 cases. WorldJ Surg 2002;26:1468-71.

15. Bellantone R, Lombardi CP, Raffaelli M, Boscherini M, DeCrea C, Traini E. Video-assisted thyroidectomy: report of aninitial experience on 53 cases. J Am Coll Surg 2002;194:610-4.

16. Robinson JL, Mandel S, Sataloff RT. Objective voice mea-sures in nonsinging patients with unilateral superior laryn-geal nerve paresis. J Voice 2005;19:665-7.

17. Raffaelli M, Iacobone M, Henry JF. The “false” nonrecur-rent inferior laryngeal nerve. Surgery 2000;128:1082-7.

DISCUSSIONDr Ashok R. Shaha (New York, NY): Excellent

presentation, very good study. I think these studiesare difficult, so I congratulate you for the same. Ihave been very interested in these problems, as youknow from our previous publication on voice inthyroid surgery. But the swallowing problem lastsmuch longer than your study indicates.

Three months is not the right time to evaluatelong-term swallowing problems. Almost everybodyhas some difficulty in swallowing one week after,and improve up to 3, 4 months. But if you followthese patients for 6, 9, 12 months and a few years,

you will find that 5% of the patients complaining of

Surgery Lombardi et al 1033Volume 140, Number 6

vague discomfort in the throat. They can’t swallowbig tablets, they will say “I have to drink a littlemore water (to swallow the pills),” and they willcomplain that “My swallowing has not been thesame.” I think this really needs to be extendedmore than three months to critically analyze theswallowing function after thyroidectomy.

I like to call this condition “Post-ThyroidectomyCentral Compartment Syndrome.” I won’t say “Sha-ha’s Syndrome” because you don’t want your nameon an unpleasant complication. But these patientsdo get these symptoms, and I have a feeling thereare 2 reasons. You gave a battery of reasons, but themost important reason is scarring. And the secondis the damage we do to branches of the nervecomplex, the anastomosing nerve branches. Manytimes you will find a tiny branch of the recurrentlaryngeal nerve going to the cricopharyngeal mus-cle. These tiny nerve branches are injured, which isthe main reason these people have continuingproblems.

Dr Celestino P. Lombardi: I was waiting for yourquestion since I know your studies very well. I com-pletely agree with you. This paper was also writtento point out that we usually don’t care about swal-lowing problems of thyroidectomized patients, butthey have this problem. And of course 3 months isa too short time to assess definitive recovery ofthese patients. The problem exists, even if it is aminor problem, and so we have to inform thepatient about it.

Dr Thomas J. Musholt (Mainz, Germany): I con-gratulate you on the study because I know that it isvery difficult to perform this kind of phoniatricstudy and to convince patients to repeatedly un-dergo the unpleasant laryngoscopies involved.

You used a multidimensional voice program(MDVP) protocol to assess the patients’ voice. Butat Mainz University, we had difficulties getting re-producible results because this kind of phoneticassessment is very dependent on the patient’s co-operation. So if you perform voice assessment byMDVP in the same patient repeatedly, you will getslightly different results and voice changes that arenot related to voice impairments in this patient atall. So I wonder if you documented how wide therange of results or variability that you gatheredfrom one examination to another examination inthe same proband, unrelated to surgery?

And the second question is, you asked the pa-tients about subjective symptoms of voice impair-ments and difficulties in swallowing, and youcannot always detect objectively measurable alter-

ations in these patients. Can you tell us about the

correlation between subjective symptoms and ob-jective findings in these patients?

Dr Lombardi: For the first question, I think thatevery kind of analysis may be useful because wehave to compare the homogeneous group, and sothis analysis may be useful for our group. We havemany, many methods to compare this differencefollowing different analyses, and there are severalkinds of analyses most of which are computerized.I think that each analysis can be useful if the studyis well planned.

About the correlation between objective andsubjective, you saw in the slides that there is not acorrelation because you have a patient with somevoice problems, as well as swallowing problems,without any objective alterations. This confirmsother studies that found a discrepancy betweensubjective and objective symptoms.

We wanted to propose this kind of score becauseit can be very useful in the evaluation of voice andswallowing impairment. The patient has to answervery simple questions, which allow us to preciselyunderstand what the patient feels. Of course, that issubjective, but all of our patients are subject. It isimportant to pay attention to the subjective sensa-tions of the patient.

Dr Cord Sturgeon (Chicago, Ill): I enjoyed yourtalk, and I think it is very important to have hard,objective data like these for speech and swallowingdifficulties after operations. My question is, haveyou looked at the control groups in order to try toboth validate the magnitude of your findings andtry to link causation to the operation?

For example, have you performed all these stud-ies and assessments in patients who have under-gone, for example, laparoscopic cholecystectomy,or perhaps even a local anesthesia-type operation,so that you can correlate the findings after thyroidor parathyroid surgery, and see how different thatis from just intubation, for example?

Dr Lombardi: I am seeing Professor Sitges Serrabehind you and I know his wonderful study. So Iask him the question. I think that in his study about10% to 12% of cholecystectomized patients hadswallowing problems, if I remember correctly.

We know, and we assessed, that some of thepatients had voice problems before the surgery. Sothe same patients could have problems also forlocal anesthesia in whatever kind of operation. Andat the end of my speech, I said that maybe weoverestimated the swallowing and voice changesafter thyroidectomy. I think that this happenedbecause most of the patients were lost at follow-up,since they had no symptoms and said to be totally

healthy.

1034 Lombardi et al SurgeryDecember 2006

I think that the problem exists also before theoperation. The problem is present also in otherkind of operations because the intubation may be aproblem, like in cholecystectomy. But of course thethyroidectomy or any cervical operation might in-crease the risk of this kind of problem.

Dr Antonio Stiges-Serra (Barcelona, Spain): If Icould do the study again, I would choose anothercontrol, maybe carotid artery surgery. Because thatis neck surgery, that is intubation far away from thepharyngeal nerve, the plexus nerve where probablythe stroma is. So I also criticize my own paper. If Iwould do the study again, I would choose necksurgery but a little bit farther away from the thy-mus. That would be a nice control group.

Dr Greg Randolph (Boston, Mass): I applaudyour work, I think it is important work, and I ap-plaud the Program Committee for including it.And I think that your voice analysis is state-of-the-art acoustic voice analysis that we would offer atMassachusetts Eye and Ear Infirmary in the voicelab, and it is a reproducible test.

You found that vocal changes were reported bynearly 75% of patients and swallowing changes in84% of patients. My experience is, if you looked at

a subgroup of singers, those changes would be even

more common and last longer all in the setting ofpatients who have normal, gross, glottic function,no vocal cord paralysis.

So I think this really demonstrates that voicesubjective instruments are not in any way predictiveof postoperative laryngeal function, and that if weare ever to know how we are doing with thyroidsurgery, we need to look at the larynx, not ask thepatient whether they have vocal cord paralysis.Thus far, estimates of vocal cord paralysis afterthyroidectomy have been a blend of what the pa-tient reports and what we want to hear. So we haveto look at the larynx.

I just wanted to make that as a comment andthen ask you if you have looked at any patients whoare kind of vocal athletes, singers, or actresses, anddetermined their rate of these subjective voice com-plaints.

Dr Lombardi: Of course that was not included inthis kind of study, but when we have some profes-sional singers, we try to do the best and the evalu-ation is always done.

I think that also in these cases, as you suggest, wecan learn by your experience. Also, using somenerve monitoring may be always difficult to predict

the final outcome.