Synchronous Electrical Stimulation of Laryngeal

Muscles: An Alternative for Enhancing Recovery

of Unilateral Recurrent Laryngeal Nerve Paralysis

*,†Alejandro Garcia Perez, ‡Xochiquetzal Hern�andez L�opez, ‡V�ıctor Manuel Valadez Jim�enez,*Arturo Minor Mart�ınez, and §Pablo Antonio Ysunza, *yzMexico City, and Veracruz, Mexico xRoyal Oak, Michigan

Summary: Background. Although electrical stimulation of the larynx has been widely studied for treating voice

AccepFrom t

cional, Mracruz, MCity, MeCraniofaAddre

Speech a3535 WebeaumonJourna0892-1� 201http://d

disorders, its effectiveness has not been assessed under safety and comfortable conditions. This article describes design,theoretical issues, and preliminary evaluation of an innovative system for transdermal electrical stimulation of the lar-ynx. The proposed design includes synchronization of electrical stimuli with laryngeal neuromuscular activity.Objective. To study whether synchronous electrical stimulation of the larynx could be helpful for improving voicequality in patients with dysphonia due to unilateral recurrent laryngeal nerve paralysis (URLNP).Materials and Methods. A 3-year prospective study was carried out at the Instituto Nacional de Rehabilitacion inthe Mexico City. Ten patients were subjected to transdermal current electrical stimulation synchronized with the funda-mental frequency of the vibration of the vocal folds during phonation. The stimulation was triggered during the phase ofmaximum glottal occlusion. A complete acoustic voice analysis was performed before and after the period of electricalstimulation.Results. Acoustic analysis revealed significant improvements in all parameters after the stimulation period.Conclusion. Transdermal synchronous electrical stimulation of vocal folds seems to be a safe and reliable procedurefor enhancing voice quality in patients with (URLNP).Key Words: Larynx–Electrical Stimulation–Voice–Therapy–Paralysis.

INTRODUCTION

Dysphonia, which includes hoarseness and breathiness, is definedas a decrease in voice quality and an increase in vocal effort. It isthe most commonly found symptom in patients with unilateralrecurrent laryngeal nerve paralysis (URLNP).1 It has been re-ported that dysphonia significantly disrupts the quality of life ofaffected individuals.2–4 URLNP can be the result of an injury ordamage to the 10th cranial nerve or its peripheral branch (ie, therecurrent larynx nerve). The etiology of this disorder includesiatrogenic situations such as inadvertent injuries during surgery,complications from endotracheal intubation. Also, noniatrogeniccauses, such as trauma involving the neck or chest, masscompression, and viral infections have been reported. Finally,drugs with potential neurotoxicity effect, such as Vincristine,have been reported as a rare etiology of vocal fold paralysis.5

The diagnosis of URLNP is a complex task and should becarried out by experienced clinicians. To reach an accuratediagnosis and rule out that vocal fold immobility is not derivedfrom mechanical causes (eg, from neoplasm, arytenoid carti-lage dislocation, or cricoarytenoid arthritis), it is usually neces-sary to combine several evaluation procedures, includingexternal palpation of the laryngeal region, videolaryngoscopy,

ted for publication January 8, 2014.he *Centro de Investigaci�on y Estudios Avanzados del Instituto Polit�ecnico Na-exico City,Mexico; yInstituto Tecnologico Superior de Poza Rica, Poza Rica, Ve-exico; zServicio de Foniatr�ıa del Instituto Nacional de Rehabilitaci�on, Mexicoxico; and the xDepartment of Speech and Language Pathology, Ian Jacksoncial Clinic, Beaumont Health System, Royal Oak, Michigan.ss correspondence and reprint requests to Pablo Antonio Ysunza, Department ofnd Language Pathology, Beaumont Health System, Medical Office Building,st 13 Mile Road, Suite 101, Royal Oak, MI 48073. E-mail: antonio.ysunza@t.edul of Voice, Vol. -, No. -, pp. 1-7997/$36.004 The Voice Foundationx.doi.org/10.1016/j.jvoice.2014.01.004

stroboscopic endoscopy, laryngeal electromyography (EMG),imaging, and laboratory studies.6,7

Once the diagnosis has been established, the managementof URLNP primarily depends on its associated symptomsand recovery prognosis, which in turn depend on the severityof the paralysis, the position of the paralyzed vocal fold, andits etiology.8–10 At present time, the two main options fortreatment are voice therapy and surgical procedures.Unfortunately, few studies provide evidence supporting theefficacy of voice therapy. Moreover, the studies addressingthe efficacy of voice therapy have yielded inconclusiveresults.11–14 Nonetheless, it is generally accepted that,under certain conditions, voice therapy may contribute toenhance voice quality by improving glottal occlusion as aresult of training the intrinsic laryngeal muscles to developcompensatory functions. However, the paralysis persists inmost of the cases, and the long-term effectiveness of voicetherapy has yet to be demonstrated.15

Surgical methods are also available for treating URLNP. Inparticular, phonosurgical techniques aimed to correct vibratorymovements of vocal folds during phonation by correcting thevocal fold position or tension or by increasing the vocal fold vol-ume.16,17 Surgical options are diverse and they have been inconstant development. Several studies support the effectivenessof these methods.18–20 However, given its invasive nature,surgery also presents disadvantages such as the risk of damaginghealthy structures and the need, in some cases, to be performedrepeatedly.21–23 In addition, although some surgical procedurescan be reversible by removing the implants, other proceduresare actually irreversible.24

Functional electrical stimulation (FES) is a technique thatdates back to the middle of the 19th century. It uses electriccurrents to induce contractions in paralyzed muscles and has

FIGURE 1. Microphone and electrodes placement.

Journal of Voice, Vol. -, No. -, 20142

been widely studied as a therapeutic option for treating para-lyzed muscles, especially distal muscles.25,26 In many cases,FES has led to the recovery of voluntary movement of theparalyzed muscles. However, the physiological mechanismby which this occurs is still unclear.

The use of FES in the treatment of voice disorders has beenstudied almost since its origins. A review of the literature re-veals that, despite the significant scientific interest in FES,the modern technological tools have become available, andin our improved understanding of larynx physiology, FEShas not become an effective therapeutic intervention to treatdysphonia in patients with URLNP.27–29 The main obstaclesthat, until now, have precluded the use of FES as aneffective therapeutic technique for treating voice disordersare the great complexity of the laryngeal physiology, and thedifficulty of producing sufficient neuromuscular stimulationusing superficial electrodes without causing pain or skinaffectations in the stimulated area.30,31

The purpose of this article was to study the use of an innova-tive system for transcutaneous electrical stimulation of thevocal folds. The device considers two theoretical issues:

The first concerns the fact that during sustained phonation,the minimum impedance through the laryngeal tissue coincideswith the phase of maximum glottal occlusion in each vibratorycycle,32 so that if the electrical stimuli are matched with suchphase, the intensity of the stimuli required to produce thedesired effects could be reduced.

The second issue is derived from the theory of Rushton33

about the physiological mechanism through which electricalstimulation promotes restorative synaptic modifications thatin some cases leads to the recovery of muscle functionality.This theory, based respectively on the widely studied Hebbiantheory,34 suggests that the occurrence of such neuromuscularrestorative process is highly dependent on the coincidence ofthe electrical stimuli with the conscious, voluntary effort ofthe patient, and the consequent coincidence between theoccurrence of dromic and antidromic impulses on Hebb-typecells. The results of the studies carried out by Ptok andStrack35 and Lagorio et al36 showed an increment on the effi-cacy of FES therapies when combined with voice exercises,which may be explained in accordance with the theory ofRushton.

The proposed stimulation system features, in particular, thesynchronization of electrical stimuli with the vocal fold vibra-tion that occurs during sustained phonation by the patient, insuch a way that one current pulse is triggered and deliveredto the laryngeal tissue in each cycle, making the stimulationfrequency being the same as the fundamental frequency (F0)of voice. The system uses the electrical signal of voice as syn-chronization source for triggering the electrical pulses, and itprovides the possibility of selecting the phase of vibration ofthe vocal folds in which the stimuli are delivered (maximal oc-clusion phase). These characteristics allow the therapy withFES to take advantage of the previously described approachesby making the electrical pulses coincide with the maximumglottal occlusion and with conscious voluntary effort of thepatient.

MATERIALS AND METHODS

A synchronous electrical stimulator and a purpose-built strobo-scopic light source were developed at the Centro de EstudiosAvanzados (CINVESTAV) del Instituto Politecnico Nacionalin Mexico City. A 3-year prospective study, approved by theInstitutional Review Board of the Instituto Nacional de Reha-bilitacion in Mexico City, was conducted from June 1, 2007,to June 1, 2010.

Synchronous electrical stimulator

During sustained phonation, the electrical stimulator generated,at each cycle of vocal fold movement, a square current pulsewith specific parameters. To this end, the voice signal of the pa-tient was captured through an electret-type microphone thatwas placed over the thyroid cartilage. The microphone washeld in place by an elastic band (Figure 1). The acquired signalwas then electronically filtered and conditioned to obtain aquasi-sine waveform of constant amplitude and whose fre-quency was the F0 of voice of each patient (ie, the frequencycorresponds to the frequency of the vibration of the vocalfold). The signal was harnessed to obtain a synchronizationpulse at the zero crossing of each rising edge which was usedas synchronization source for triggering the electrical stimulithrough the processing of a microcontroller. The amplitudeand duration of the synchronization pulses were 5 V and 140microseconds, respectively.The equipment features a digital user interface consisting of a

screen and keyboard that allows the physician to program andvisualize the stimulation parameters to be used. The samemicro-controller (PIC16F877A from Microchip Technology, Inc.,Chandler, AZ) controls the user interface and uses the synchroni-zation pulses for generating stimulation pulses through an analogcircuit. Carbon rubbed electrodes were used (REF 79966 - Inte-lect Advanced/IntelectMobile; DJO, UK, LTD, Guilford Surrey,UK) for performing the percutaneous electrical stimulation.The intensity of the electrical stimulus and the duration of the

pulses can be adjusted from 0 to 10 mA and from 0.1 to 0.5 mil-liseconds. A value of 500 Hz was arbitrarily considered as themaximum F0 of the patients for being candidates for using

FIGURE 2. Output from digital oscilloscope. Four channels are dis-

played (CH1–CH4 top to bottom). The horizontal axis is 5 millisec-

onds per major division. CH1 shows the voice signal captured by the

microphone during sustained phonation of the phoneme /i/. In CH2,

the conditioned voice signal is displayed. CH3 shows the synchroniza-

tion pulses. In CH4, stimulation pulses delayed by Dt with respect to

the synchronization pulses are displayed. The panels on the right-

hand side provide some measurements.

FIGURE 3. (A) Typical electroglottographic waveform. (B) Micro-

phone signal during phonation of sustained vowel /i/. (C) Frontal sec-

tion through the mid-portion of the glottis at different phases of vocal

fold oscillation. Point 3 corresponds to minimum impedance, which is

achieved at maximum closure of the glottis. Dt represents the dephas-

ing time between the stimulation and synchronization pulses, provided

that synchronization pulses were produced in the point 1. By varying

the value of Dt, the stimuli can be applied during different phases of

vocal fold vibration.

Alejandro Garcia Perez, et al Enhancing Recovery of URLNP 3

this equipment, so that in any case, the duration of the pulseswould never exceed 25% of the period of the signal of voiceto prevent muscular fatigue (ie, duty-cycle was always lessthan 25%).

The equipment allows to modify dephasing time Dt betweenthe stimulation and synchronization pulses via software, suchcapability allows to select with high accuracy the phase of thevibratory cycle of the vocal folds inwhich the electrical stimulusis triggered. The dephasing time was implemented as a fractionof the period of the voice signal, which was done so that thedephasing time thus determined corresponded with the samephase of movement of the vocal folds, even at different fre-quencies. Also, the electrical stimulator featured a port throughwhich the voice signal captured by the microphone could beaccessed, along with the conditioned signal, the synchronizedpulses, and a sample of the stimulation pulses. Figure 2 showsthese signals displayed in descending order on a digital four-channel oscilloscope (TDS640 from Tektronix, Inc., Beaverton,OR) for sustained phonation of the phoneme /i/.

Purpose-built stroboscopic light source

A videolaryngostroboscopic device was developed to identifythe phase of the vocal fold vibration in which each stimuluswas provided (ie, point 1, 2, 3 4, 5, or 6 on Figure 3). To achievethis aim, a stroboscopic light source (SLS) was designed andbuilt. The SLS used a sample of the electrical stimulation pulsesas a synchronization source to generate light pulses. The lightpulses generated by the SLS were 0.3-millisecond long andwere triggered 4 microseconds after the onset of each electricalstimulus (CH4 in Figure 2). For a periodic signal of voice, eachlight pulse illuminated the vocal folds in the same position (ie,at the same point in each vibration cycle), thereby producing astatic image of the vocal folds, which made possible to observe

their position essentially at the moment in which the stimuluswas applied.

To determine the dephasing time Dt that makes the electricalstimulus coincide with maximum occlusion of the vocal folds(which corresponds to minimum impedance), experienced cli-nicians examined five healthy volunteers by videolaryngostro-boscopy using the SLS synchronized with the stimulationpulses generated by the stimulator but without applying electri-cal stimuli. Each volunteer was asked to utter the phoneme /i/(of the Spanish Language) in a sustained manner and, duringphonation, the image of the vocal folds was observed in themonitor. The freezing effect caused by the stroboscopic light al-lowed the clinician to observe the vibration of the vocal foldsapparently fixed in a given position, which was varied bychanging Dt until they were observed at the maximum levelof occlusion (Figure 4).

FIGURE 4. Laryngoscopy of vocal folds during electrical stimuli.

The vocal folds appear to be static at the position of maximum occlu-

sion during the sustained production of the phoneme /i/. The vocal fold

may be viewed in various positions by varying Dt.

Journal of Voice, Vol. -, No. -, 20144

It was found that, in all cases, the same dephasing time Dtwas required to observe the vocal folds in the position ofmaximum occlusion. Based on these results, Dt was fixed at0.10 T for all patients that participated in the study, where Tis the inverse of the F0 of voice.

All images were captured with a high resolution 1/300 charge-coupled device Camera (ATMOS Cam 31; ATMOS Medizin-Technik GmbH & Co., Hampshire, UK) through an ultra slimnasopharyngoscope with 3.4 mm diameter, 120� field of viewand 30 cm working length (FNL-10P2; Pentax Ricoh ImagingCo., Ltd., Tokyo, Japan). In all cases, the nasopharyngoscopewas introduced through the most permeable nostril and posi-tioned so that the region of the arytenoid cartilages and thevocal folds were fully visible.

Study population

Patients from the Division of Voice Disorders of the InstitutoNacional de Rehabilitacion in Mexico City were studied. Pa-tients with a confirmed diagnosis of URLNP were selected. Itshould be pointed out that for this study, patient participationwas voluntary and those who did not wish to participatereceived the standard voice therapy with no disadvantages atall. Patients who met the inclusion criteria and accepted toenroll in the study signed a written informed consent, in accor-dance with the Declaration of Helsinki. The study protocol wasapproved by the Internal Review Board of the Institute.

The following inclusion criteria were considered. All partic-ipants should present with chronic dysphonia (between 10 and24 months after onset), caused by URLNP of postsurgical etiol-ogy. The diagnosis had to be confirmed by one experiencedclinician and it was based on history, clinical, electromyo-graphic, and videolaryngostroboscopic findings. The integrity

of the superior laryngeal nerve had to be corroborated by anormal EMG of the cricothyroid muscle in all cases. The partic-ipants should be capable of producing sustained phonation forat least 3 seconds, and the F0 of voice did should not exceed500 Hz. Patients who had received voice therapy during thelast 5 months or who had undergone phonosurgical procedureswere excluded from the study. Patients with heart-rhythm disor-ders, pacemakers, metallic implants, or cochlear implants werealso excluded.

Study protocol

The study consisted of objectively and quantitatively evaluatingthe effect of synchronous transcutaneous electrical stimulationon the voice quality of patients with URLNP. Using Dr. SpeechVocal Assessment Version 3 (Tiger Electronics, Inc., Shanghai,China), the voice acoustics of each patient was analyzedfollowing the same methodology before and after they hadundergone 10 sessions of stimulation therapy. The acoustic pa-rameters evaluated were as follows: F0, jitter (%), shimmer (%),harmonics-to-noise ratio (HNR), and normalized noise energy(NNE). Themaximum phonation time (MPT) was also assessedbefore and after the stimulation period.The patients’ voices were recorded in an anechoic chamber

with a one-way microphone positioned approximately 20 cmin front of the mouth. The microphone was pointed towardthe mouth and positioned slightly beneath the chin to reducethe effects of airflow. The microphone, met the Guidelines forselecting microphones for human voice,37 sensitivity was�69.5 dB with a 600 U impedance and frequency responsefrom 50 to15 000 Hz. All recordings were carried out beforenoon to avoid any possible abnormalities from vocal fatigueor overuse. During the recording, the patients were seated ina comfortable position. For both the pre- and post-therapy eval-uations, each patient was asked to utter the phonemes /a/, /e/, /i/,/o/, and /u/ (vowel sounds of the Spanish Language) in a sus-tained manner and in habitual tone and loudness for as longas was possible. Each phoneme was recorded in three consecu-tive trials. From the three recordings of each phoneme, a singlesample between 0.5 and 1.0 seconds long was used to calculatethe acoustic parameters; reasons for this were to deal with thenormal irregularities found in voice recordings, and the implicitmathematical limitations of the algorithms applied by the soft-ware, which are specially emphasized when analyzing patho-logic voices.38 For this aim, the principal member of thetechnical team analyzed the voice recorders and chose the sam-ples by visually searching for the most stable portion of the re-cordings in terms of waveform, frequency, and intensity. Theexaminer was blinded to all information about the recorders,including the name, sex, and age of the patients, the correspond-ing vowel, and the type of evaluation (pre- or poststimulation).Moreover, MPTwas measured manually by the physicians us-

inga stopwatch. For this purpose, each subjectwas asked to inhaledeeply and utter an /a/ in habitual tone and loudness for as long aspossible. Three exercises were practiced in each case, and fromthe three measures obtained, the largest value was taken.Findings from routine clinical and videostrobolaryngoscopic

examinations carried out during the diagnostic process and at

TABLE 1.

Results of Acoustic Analysis Pre- and Post-Therapy

Phonemes /a/ /e/ /i/ /o/ /u/

F0Pre 187.6 (57.23) 188.5 (55.78) 189.8 (59.71) 187.9 (57.63) 189.4 (56.96)Post 197.8 (63.33) 197.3 (64.62) 202.8 (64.52) 198.0 (65.15) 197.2 (65.40)P 0.3750 0.2754 0.1641 0.2754 0.2324

Jitter (%)Pre 4.76 (2.48) 5.16 (2.26) 4.91 (2.56) 5.12 (2.76) 5.77 (2.54)Post 1.63 (1.25) 1.71 (1.45) 1.54 (1.39) 1.72 (1.25) 1.70 (1.37)P 0.0137 0.0039 0.0195 0.0098 0.0039

Shimmer (%)Pre 11.71 (3.42) 12.28 (3.40) 11.63 (4.22) 13.06 (3.70) 12.23 (4.44)Post 6.56 (2.36) 6.29 (2.14) 6.47 (2.38) 6.76 (2.13) 6.36 (2.49)P 0.0039 0.0020 0.0078 0.0020 0.0039

HNR (dB)Pre 10.69 (4.91) 11.44 (4.46) 12.12 (4.23) 10.66 (4.40) 11.46 (5.28)Post 19.49 (4.93) 19.58 (6.43) 20.84 (4.87) 19.73 (5.74) 20.57 (5.96)P 0.0020 0.0020 0.0039 0.0039 0.0020

NNE (dB)Pre �5.89 (3.15) �6.16 (3.19) �6.04 (3.20) �6.04 (3.26) �5.38 (3.42)Post �13.76 (5.04) �13.18 (4.63) �12.61 (3.48) �13.96 (5.47) �13.65 (5.70)P 0.0273 0.0020 0.0039 0.0273 0.0098

Notes: Values in the form of mean (standard deviation). Significant differences indicated by P < 0.05.

Alejandro Garcia Perez, et al Enhancing Recovery of URLNP 5

the end of the therapies were also registered. Videostrobolar-yngoscopic examinations were performed by using the work-station ATMOS C31 from ATMOS MedizinTechnik GmbH &Co. KG.

After concluding the initial evaluation, each patient underwent10 weekly 30-minute-long stimulation sessions, which consistedsolely of applying synchronous transcutaneous electrical stimula-tion using the synchronous electrical stimulator. During the ses-sions, the patients were seated comfortably. Male patients wereasked to attend clean shaven. The skin under the chin and onthe larynxwas thoroughly cleanedwith alcohol, and a conductivegel was applied to minimize the electrode skin impedance. Theelectrodes were placed in the laryngeal area, as shown inFigure 1, with the active electrode on the side of the paralyticvocal fold.Theelastic bandused tofix themicrophonewas placedon top of the electrodes to help maintain good electrical contact.

Except for frequency, the same stimulation parameters wereused for all patients and in all stimulation sessions. The patientswere exposed to 150 microseconds, 7 mA pulses. The frequencyof the electrical stimuli depended on the F0 of voice of each sub-ject. Both the pulse width and intensity were chosen in order forthem to be comfortable for the patient but high enough toproducea response through the stimulation of nondenervated or partiallydenervated muscles, whose chronaxy values lie between 0.001and 1 milliseconds. At the very beginning of each therapy, thepulse width and amplitude were set to 150 microseconds and0mA, respectively, then, during patient phonation, the amplitudewas gradually increased to reach 7 mA. None of the patientsexperienced discomfort or painful sensation during the therapy.

For 30 minutes, a physician motivated the given patient toinhale deeply and utter the phoneme /i/ and to try to maximize

the phonation timeduring each exercise.Onaverage, each patientcompleted 127 (s ¼ 42) exercises during each session. Duringphonation, the stimulator delivered current pulses synchronizedwith the vibration of the vocal fold. In the first therapy session,each patient underwent videolaryngostroboscopic analysiswhereby the physician used the SLS to verify that the electricalstimuli coincided with maximum occlusion of the vocal folds.This goal was achieved using the same Dt for all the patients.

Statistical analysis

We compared the acoustic parameters F0, jitter, shimmer, HNR,NNE, and the MPT obtained before and after the therapies byusing the nonparametric Wilcoxon signed rank test for paireddata within GraphPad Prism version 5.04 for Windows (Graph-Pad Software, La Jolla, CA) with 95% set as the level of statis-tical significance and two-tailed P values.

RESULTS

A total of 20 patients were selected for the study group. Thegroup included seven men and 13 women. Mean age [standarddeviation (s), range] was 32 years (9.89, 19–49).

From the initial number of patients, five of them wereexcluded at the beginning of the protocol because they failedto meet the inclusion criteria described herein. Five more pa-tients left before completing the study protocol because theywere could not travel to the hospital. Finally, 10 patientscompleted the study, three men and seven women with amean age of 33.87 (9.86, 21–49).

For the five phonemes analyzed, significant differences werefound for jitter, shimmer, HNR, and NNE. The F0 of the voicedid not change significantly for any of the phonemes (Table 1).

Journal of Voice, Vol. -, No. -, 20146

The MPT increased from 3.9 seconds (s ¼ 1.1) to 9.6 seconds(s ¼ 3.02) (P ¼ 0.0056).

Although the paralysis persisted in all the patients, the resultsof the acoustic analysis revealed a marked improvement in thepatients’ voice quality. After the stimulation period, videolar-yngostroboscopic examinations of the patients revealed an in-crease in muscular tone on the paralyzed vocal fold in themajority of cases, better compensation of the mobile—unin-jured vocal fold by contracting longitudinally widening thefold and enhancing better closure of the glottal chink.

DISCUSSION

The results of this study suggest that synchronous electricalstimulation has a positive effect on the voice quality of patientswith chronic dysphonia caused by UVFP. With the parametersused for the stimulation, none of the patients experienceddiscomfort. The majority experienced a light, painless, tinglingsensation. However, because the stimulation frequency dependson the F0 of each patient’s voice, patients with excessively deepor sharp voice tonalities may not be candidates for this therapy.From the literature, the range of frequencies for which this ther-apy is most likely to be effective is from 100 to 300 Hz.39

The results indicate that, for the five phonemes analyzed, theacoustic analysis of voice provides consistent informationregarding the voice quality (Table 1). Because all patientsincluded in this study had been developing vocal fold paralysisfor at least 10 months, the influence of possible spontaneous re-covery was considered to be minimal. However, this possibilitycannot be completely disregarded.

The acoustic analysis of the voice is a method that enables theobjective measurement of voice quality, and its use in the eval-uation of treatments has been widely documented.30–42 The F0

determines the tone of speech and provides information on thephysiological structure of the larynx. Jitter and shimmerindicate small variations in F0 and cycle-to-cycle amplitudevariations, respectively. These variations result from small dif-ferences in weight, tension, and biochemical characteristics ofthe vocal fold, as well as from small differences in neural con-trol. The disturbances from jitter and shimmer correlate withperceived roughness and hoarseness. The HNR and NNE pa-rameters are obtained by calculating the noise energy that isproduced as a result of the air turbulence that is generated prin-cipally during the glottal occlusion phase. The level of acousticnoise can be greater in pathologic voices, which correlates withperceived hoarseness.43,44

MPT is a good indicator of the degree of glottal closure dur-ing phonation, and the consequent phonation control. However,MPT is also sensitive to changes in lung function (eg, lung aircapacity).45 Even when synchronous electrical stimulation iscombined with sustained phonation, lung function is not ex-pected to be significantly affected. As such, it was assumedthat an increase in MPT was the result of an improvement inglottal occlusion. It should be pointed out that this assumptionmay not be valid in studies that combine electrical stimulationwith voice therapy because one of the goals of voice therapy isusually to improve lung function.

Patients with severe dysphonia and a largely aperiodic signalmay present a significant challenge for using the system for syn-chronous electrical stimulation described herein. Due to themathematical restrictions of the implemented algorithms in allacoustic analysis software, an acceptable level of periodicity isnecessary for yielding useful results. All patients included inthe study group for this article, although they showed dysphonicvoices, showed reasonably acceptable levels of periodicity.More-over, patient with extremely severe dysphonia were excludedbecause the software was unable to measure the acoustic param-eters due to the severity of the noise and acoustic perturbations.Nonetheless, it should be considered that synchronous stim-

ulation is not entirely dependent of determining F0. Each cycleof the stimulation signal corresponds to a new cycle of thesignal, which is being synchronized, that is, to the cycle ofthe voice signal. Each pulse is triggered by the cross aroundzero of the rising edge of the voice signal (plus a delay inDt). Thus, in cases of intermittently aperiodic signals, the resul-tant stimulation signal has similar aperiodicities than the voicesignal. Moreover, the voice signal is directly related with theelectroglottographic curve that measures glottis impedance,which is dependent on glottal occlusion. Theoretically, eachstimulation pulse is generated in the same occlusion phase ofthe vocal folds in each period, independently of the F0. Incontrast, Dt delay is a function of the F0, which in this caseis obtained from the duration of the last period. Therefore,even in severely dysphonic voices with extremely abnormalshimmer and jitter values, this stimulation device could stillbe useful.It can be assumed that the maximal occlusion lasts at least a

sixth of the period, which represents a tolerance value of at least16.7%. Finally, we can assume that the occurrence of someisolated cycles in which the stimulation would not reach exactlythe phase of maximum occlusion would be statistically non-significant.There are some reports concerning the use of the VitalStim

(Vista, CA) system in cases of dysphagia and vocal fold paresis.However, it is necessary to point out the differences between thestimulation used by the VitalStim system and the synchronousstimulation system described in this article. The stimulation ofthe VitalStim system is not synchronized. The stimulation fre-quency is selected by the user and it is independent frequency ofthe voice signal.

CONCLUSION

The results of this study show that the use of synchronous stim-ulation can provide an effective option for addressing one of themost important problems related with electrical treatments ofvoice disorders, which is, as exposed herein, the limited phys-iological effect that can be achieved by using stimulation pa-rameters within safety and comfortable ranges for the patients.The results also suggest that synchronous electrical stimu-

lation contributes to improve voice quality in patients withULRNP. However, it will be necessary to study the potentialadvantages provided by this innovative procedure in largergroups of patients. Furthermore, the outcome after synchronous

Alejandro Garcia Perez, et al Enhancing Recovery of URLNP 7

stimulation should be compared with conventional methods fortreating ULRNP. The limitations of this procedure will alsohave to be established, especially regarding the limited rangeof allowable frequencies, which in turn will have to be furtherstudied. Finally, from the results of this study, it is evidentthat the electrical voice signal can be used not only as a syn-chronization source, as in the case of SLSs, but also as a refer-ence for triggering the electrical stimuli during the selectedphase of vocal folds vibratory cycle.

Acknowledgments

Theauthorswant to thankAliciaVilleda,MD,NormaL.Garrido,MD, andAraceli P. S�anchez,MD, for their valuable contributionsto this project.

REFERENCES1. Koufman JA, Postma GN, Cummins MM, Blalock PD. Vocal fold paresis.

Otolaryngol Head Neck Surg. 2000;122:537–541.

2. Murry T, Rosen CA. Outcome measurements and quality of life in voice

disorders. Otolaryngol Clin North Am. 2000;33:905–916.

3. Spector BC, Netterville JL, Billante C, Clary J, Reinisch L, Smith TL.

Quality-of-life assessment in patients with unilateral vocal cord paralysis.

Otolaryngol Head Neck Surg. 2001;125:176–182.

4. Borel S, Crevier-Buchman L, Tessier C, Hans S, Laccourreye O, Brasnu D.

Quality of life before and after thyroplasty for vocal fold paralysis. Rev Lar-

yngol Otol Rhinol (Bord). 2004;125:287–290.

5. Laccourreye O, Papon JF, Kania R, M�enard M, Brasnu D, Hans S. Unilat-

eral laryngeal paralyses: epidemiological data and therapeutic progress.

Presse Med. 2003;32:781–786.

6. Rubin AD, Sataloff RT. Vocal fold paresis and paralysis. Otolaryngol Clin

North Am. 2007;40:1109–1131.

7. Ysunza A, Landeros L, Pamplona M, Silva A, Prado H, Fajardo G. The role

of electromyography in vocal cord paralysis. Gac Med Mex. 2008;144:

303–308.

8. Havas T, Lowinger D, Priestley J. Unilateral vocal fold paralysis: causes,

options and outcomes. Aust N Z J Surg. 1999;69:509–513.

9. Kelchner LN, Stemple JC, Gerdeman E, Le Borgne W, Adam S. Etiology,

pathophysiology, treatment choices, and voice results for unilateral adductor

vocal fold paralysis: a 3-year retrospective. J Voice. 1999;13:592–601.

10. Ptok M, Strack D. Voice exercise therapy versus electrostimulation therapy

in patients with unilateral vocal fold paralysis. HNO. 2005;53:1092–1097.

11. Speyer R, Wieneke GH, Dejonckere PH. Documentation of progress in

voice therapy: perceptual, acoustic, and laryngostroboscopic findings pre-

therapy and posttherapy. J Voice. 2004;18:325–340.

12. Schindler A, Bottero A, Capaccio P, Ginocchio D, Adorni F, Ottaviani F.

Vocal improvement after voice therapy in unilateral vocal fold paralysis.

J Voice. 2008;22:113–118.

13. Speyer R. Effects of voice therapy: a systematic review. J Voice. 2008;22:

565–580.

14. Cantarella G, Viglione S, Forti S, Pignataro L. Voice therapy for laryngeal

hemiplegia: the role of timing of initiation of therapy. J Rehabil Med. 2010;

42:442–446.

15. Miller S. Voice therapy for vocal fold paralysis. Otolaryngol Clin North

Am. 2004;37:105–119.

16. Bihari A,M�esz�arosK, Rem�enyiA, Lichtenberger G.Voice quality improve-

ment after management of unilateral vocal cord paralysis with different

techniques. Eur Arch Otorhinolaryngol. 2006;263:1115–1120.

17. Sittel C, Bosch N, Plinkert PK. Surgical voice rehabilitation in unilateral

vocal fold paralysis. Chirurg. 2008;79:1055–1064.

18. Laccourreye O, Papon JF, M�enard M, Crevier-Buchman L, Brasnu D,

Hans S. Treatment of recurrent unilateral paralysis with thyroplasty with

Montgomery implant. Ann Chir. 2001;126:768–771.

19. Paniello RC. Laryngeal reinnervation. Otolaryngol Clin North Am. 2004;

37:161–181.

20. Mallur PS, Rosen CA. Vocal fold injection: review of indications, techniques,

andmaterials for augmentation.ClinExpOtorhinolaryngol. 2010;3:177–182.

21. Tucker HM, Wanamaker J, Trott M, Hicks D. Complications of laryngeal

framework surgery (phonosurgery). Laryngoscope. 1993;103:525–528.

22. Rosen CA. Complications of phonosurgery: results of a national survey.

Laryngoscope. 1998;108:1697–1703.

23. Young VN, Zullo TG, Rosen CA. Analysis of laryngeal framework surgery:

10-year follow-up to a national survey.Laryngoscope. 2010;120:1602–1608.

24. Hartl DM, Travagli JP, Leboulleux S, Baudin E, Brasnu DF,

Schlumberger M. Clinical review: current concepts in the management of

unilateral recurrent laryngeal nerve paralysis after thyroid surgery. J Clin

Endocrinol Metab. 2005;90:3084–3088.

25. Ahlborn P, Schachner M, Irintchev A. One hour electrical stimulation ac-

celerates functional recovery after femoral nerve repair. Exp Neurol.

2007;208:137–144.

26. Bajd T, Munih M. Basic functional electrical stimulation (FES) of extrem-

ities: an engineer’s view. Technol Health Care. 2010;18:361–369.

27. Katada A, Nonaka S, Adachi M, et al. Functional electrical stimulation of

laryngeal adductor muscle restores mobility of vocal fold and improves

voice sounds in cats with unilateral laryngeal paralysis. Neurosci Res.

2004;50:153–159.

28. GilmanM, Gilman SL. Electrotherapy and the human voice: a literature re-

view of the historical origins and contemporary applications. J Voice. 2008;

22:219–231.

29. Ptok M, Strack D. Therapeutic effects of electrical stimulation therapy on

vocal fold vibration irregularity. HNO. 2009;57:1157–1162.

30. Humbert IA, Poletto CJ, Saxon KG, Kearney PR, Ludlow CL. The effect of

surface electrical stimulation on vocal fold position. Laryngoscope. 2008;

118:14–19.

31. Clark H, Lazarus C, Arvedson J, Schooling T, Frymark T. Evidence-based

systematic review: effects of neuromuscular electrical stimulation on swal-

lowing and neural activation. Am J Speech Lang Pathol. 2009;18:361–375.

32. Unger J, Meyer T, Herbst CT, Fitch WT, Dollinger M, Lohscheller J. Pho-

novibrographic wavegrams: visualizing vocal fold kinematics. JAcoust Soc

Am. 2013;133:1055–1064.

33. Rushton DN. Functional electrical stimulation and rehabilitation—an hy-

pothesis. Med Eng Phys. 2003;25:75–78.

34. HebbDO.TheOrganisationofBehavior.NewYork,NY:Wiley&Sons; 1949.

35. Ptok M, Strack D. Electrical stimulation-supported voice exercises are su-

perior to voice exercise therapy alone in patients with unilateral recurrent

laryngeal nerve paresis: results from a prospective, randomized clinical

trial. Muscle Nerve. 2008;38:1005–1011.

36. Lagorio LA, Carnaby-Mann GD, Crary MA. Treatment of vocal fold

bowing using neuromuscular electrical stimulation. Arch Otolaryngol

Head Neck Surg. 2010;136:398–403.

37. Svec JG, Granqvist S. Guidelines for selecting microphones for human

voice production research. Am J Speech Lang Pathol. 2010;19:356–368.

38. Little MA, Costello DA, Harries ML. Objective dysphonia quantification in

vocal fold paralysis: comparing nonlinear with classical measures. J Voice.

2011;25:21–31.

39. Stein RB. Functional electrical stimulation. In: Larry RS, ed. Encyclopedia

of Neuroscience. Oxford, UK: Academic Press; 2009:399–407. %! Func-

tional Electrical Stimulation %@ 978-390-308-045046-045049.

40. 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 Lar-

yngological Society (ELS). Eur Arch Otorhinolaryngol. 2001;258:77–82.

41. Wang CC, Huang HT. Voice acoustic analysis of normal Taiwanese adults.

J Chin Med Assoc. 2004;67:179–184.

42. Chernobel’skii SI. Application of acoustic analysis of the voice to diagnosis

and treatment of functional dysphonia.VestnOtorinolaringol. 2009;5:40–42.

43. Bhuta T, Patrick L, Garnett JD. Perceptual evaluation of voice quality and

its correlation with acoustic measurements. J Voice. 2004;18:299–304.

44. Toran KC, Lal BK. Objective analysis of voice in normal young adults.

Kathmandu Univ Med J (KUMJ). 2009;7:374–377.

45. Speyer R, Bogaardt HC, Passos VL, et al. Maximum phonation time: vari-

ability and reliability. J Voice. 2010;24:281–284.