Voice Disorders in the Elderly

*Naomi D. Gregory, *Swapna Chandran, Deborah Lurie, and *Robert T. Sataloff, *yPhiladelphia, Pennsylvania Summary: Objectives. Despite the prevalence of voice disorders, as well as the physiological and functional changes of the aging larynx, there is a lack of data analyzing dysphonia in the geriatric population. The goal of this study was to investigate dysphonia in this cohort. Study Design. Retrospective chart review. Methods. This study analyzes the histories, demographics, Voice Handicap Index (VHI) questionnaires, and objective voice measures (OVMs) for 175 patients with voice complaints, age ranging from 65 to 89 years. Diagnoses of any vocal fold pathology were made via strobovideolaryngoscopy and laryngeal electromyography (LEMG) at the time of presentation. Results. Strobovideolaryngoscopy revealed that laryngeal laryngopharyngeal reux in 91% (N 159) was the most common diagnosis associated with the voice complaints, followed by muscle tension dysphonia in 73% (N 127) and paresis in 72% (N 126). Of the 175 patients in this study, 27% (N 48) of patients had a history of antecedent event, which might have contributed to their current dysphonia, most commonly upper respiratory tract infection in 27% (N 13) and endotracheal intubation in 21% (N 10). Ninety-three percent (N 153) of patients who underwent LEMG had weakness in the distribution of at least one nerve. VHI scores varied greatly, ranging from 4 to 104, with an average score of 43.9. When VHI scores were correlated with OVMs, correlations were found with mean jitter (%), jitter (abs.), maximum phonation time (s), and shimmer (%). When OVM scores were compared with KayPENTAX normative thresholds, 69.7% of subjects were found to be above the threshold for soft phonation index. Conclusion. Our studies identied at least one pathologic factor contributing to dysphonia in all elderly patients presenting with voice complaints. The high average VHI score indicated that these geriatric patients experienced signicant dissatisfaction because of their dysphonia. The problem was of sufcient magnitude to result in a high percentage of patients proceeding with treatment. Additional research is needed to determine normative values for OVMs and other assessments in the elderly population and establish whether normative values in common use are appropriate for this population. Key Words: Geriatric laryngologyHoarsenessDysphoniaVoice disorders in the elderlyAge-associated voice disordersGeriatric voice pathology.

INTRODUCTION The larynx undergoes many age-related physiological and structural changes. The laryngeal muscles atrophy, the elastic bers of the vocal ligament become thinner, mucus glands degenerate, laryngeal cartilages ossify, and the epithelium thickens.1,2 These changes can lead to decreased vocal fold mass, inadequate approximation of the vocal folds, and other vocal fold alterations that can affect vocal fold function.2,3 The quality of voice resulting from air loss, laryngeal tension, tremor, and altered fundamental frequency may allow listeners to easily differentiate some elderly voices from younger voices.3 Ultimately, these variations may lead to an undesired vocal quality. This article assesses the demographics, most common complaints, diagnoses, objective voice measures (OVMs), Voice Handicap Index (VHI) scores, and interventions of patients older than 65 years with voice complaints in a tertiary care laryngology practice.

METHODS This study is a retrospective institutional review board-approved chart review of 175 consecutive geriatric patients seen in a tertiary care laryngology practice in Philadelphia, Pennsylvania. All patients aged 65 years and older who presented to the ofce with a primary voice complaint between 1992 and 2008 and for whom data were available were included in the study. Patients were excluded only if strobovideolaryngoscopic data, OVMs, and other information evaluated in this study had not been obtained or were unavailable. At the time of study, voice complaints were dened as any change in voice quality or ability perceived by the patient as negative. Demographic information collected included sex, age, and profession. Patient complaints were recorded, as were events
TABLE 1. Profession of Each Patient Profession (N 175) n (%) 25 (14) 24 (14) 18 (10) 17 (10) 14 (8) 36 (21) 41 (23)

Accepted for publication October 28, 2010. From the *Department of Otolaryngology - Head and Neck Surgery, Drexel University College of Medicine, Philadelphia, Pennsylvania; and the yDepartment of Mathematics and Computer Science, St. Josephs University, Philadelphia, Pennsylvania. Address correspondence and reprint requests to Robert T. Sataloff, Department of Otolaryngology - Head and Neck Surgery, Drexel University College of Medicine, 1721 Pine Street, Philadelphia, PA 19103. E-mail: rtsataloff@phillyent.com Journal of Voice, Vol. 26, No. 2, pp. 254258 0892-1997/$36.00 2012 The Voice Foundation doi:10.1016/j.jvoice.2010.10.024

Professional singer/speaker/religious leader Teacher/professor Sales/customer service Physician/nurse Business executive/manager Other: attorney, social worker, realtor, construction engineer None/unknown

Naomi D. Gregory, et al

Voice Disorders in the Elderly

255

TABLE 2. Chief Complaints of Each Patient Chief Complaint Hoarseness Decreased volume Throat clearing Raspiness Fatigue Cough Breathiness Breaks Decreased range Globus sensation Tremor Dysphagia Total, n (%) 124 (71) 79 (45) 76 (43) 25 (14) 65 (37) 40 (23) 38 (22) 24 (14) 29 (17) 17 (10) 7 (4) 11 (6) Men, n (%) 53 (30) 32 (18) 33 (19) 11 (6) 26 (15) 17 (10) 13 (7) 8 (5) 13 (7) 7 (4) 1 (1) 4 (2) Women, n (%) 71 (41) 47 (27) 43 (25) 14 (8) 39 (22) 23 (13) 25 (14) 16 (9) 16 (9) 10 (6) 6 (3) 7 (4)

The above-mentioned complaints were seen in ve patients or less and were not included in the table: deeper voice, belching, sore throat, burning taste, strain, increased effort, and increased pitch.

antecedent to the onset of symptoms. Diagnostic observations were recorded as made during examination including strobovideolaryngoscopy. All laryngeal diagnoses were made by an experienced laryngologist who performed and/or reviewed the recorded examination. Laryngeal electromyography (LEMG) data were recorded, as well. LEMG was performed by an experienced neurologist who is also board certied in electrophysiology. OVMs were obtained using the Kay PENTAX (Lincoln Park, NJ) voice analysis system and compared with standard normative values. These data were available for 167 (95%) of the 175 patients enrolled in this study; however, not all of these patients had data on all 10 measures assessed. As these data were collected retrospectively from 1992 to 2008, additional measures were added during this time period with subsequent software updates to the KayPENTAX operating system. VHI scores were obtained at the time of each patients rst ofce evaluation. Any medical or surgical interventions were documented. Finally, descriptive statistics were computed. Spearman rank order correlations were used to measure the correlation between VHI and OVMs. All data were compiled and stored in an excel spreadsheet in a password-protected computer. RESULTS One hundred seventy-ve charts of patients who were aged 65 years or older were reviewed. At the time of data collection, age range was 6589 years. Average age was 74.5 years (standard deviation [SD], 6.38 years). Of the 175 patients, 46% (N 81) were men and 54% (N 94) were women. A variety of professions requiring substantial voice use were represented among the patients (Table 1). Hoarseness was the most common complaint as reported by 124 (71%) patients; 71 (57%) of whom were women, and 53 (43%) of whom were men. Decreased volume and frequent throat clearing were the second and third most common complaints, respectively (Table 2). Forty-eight (27%) of the 175 patients reported some event preceding the onset of their voice

complaints or symptoms (Table 3). The three most common events were upper respiratory tract infection (N 13; 27%), intubation (N 10; 21%), and trauma (N 7; 15%). Trauma in this study included fall, motor vehicle accident, bicycle accident, and gas leak exposure. Diagnoses were obtained with strobovideolaryngoscopy for all the patients, and 159 (91%) patients were found to have signs of laryngeal pharyngeal reux (LPR). Eighty-ve of these were women, and 74 were men. Other common ndings included muscle tension dysphonia (MTD) in 73% (N 127) and paresis in 72% (N 126). Diagnoses are summarized in Table 4. Multiple diagnoses were given to patients if there were multiple pathologic ndings detected on strobovideolaryngoscopy. LEMG was obtained for 164 subjects and conrmed that 153 (93%) patients had weakness in the distribution of at least one nerve. One hundred twenty-one (74%) patients had only 70 90% recruitment of their left and/or right superior laryngeal nerve, whereas 32 (20%) patients had less than 70% recruitment. Twenty-one (13%) patients were found to have tremor, ve (3%) had LEMG ndings suggestive of spasmodic dysphonia, and three (2%) were diagnosed with myasthenia gravis. Table 5

TABLE 3. Antecedent Events that Occurred Immediately Before Voice Changes Antecedent Event (N 48) Upper respiratory infection Intubation Trauma Carotid endarterectomy Thyroid surgery Vocal fold surgery Cerebrovascular accident Other surgery Other: hormone replacement therapy, thyroiditis, and anterior cervical fusion n (%) 13 (27) 10 (21) 7 (15) 4 (8.33) 4 (8.33) 4 (8.33) 2 (4) 2 (4) 2 (4)

256
TABLE 4. Diagnoses as Made by an Experienced Laryngologist Using Strobovideolaryngology Diagnosis LPR MTD Paresis (by clinical assessment only, not LEMG) Vocal fold mass Glottic insufciency Varicosity/ectasia Reinkes edema Tremor Vocal fold stiffness Vocal fold scar/brosis Leukoplakia Cancer Other Total, n (%) 159 (91) 127 (73) 126 (72) 54 (31) 33 (19) 33 (19) 24 (14) 22 (13) 21 (12) 20 (11) 5 (3) 3 (2) 50 (29)

Journal of Voice, Vol. 26, No. 2, 2012

Men, n (%) 74 (42) 58 (33) 57 (33) 25 (14) 20 (11) 13 (7) 9 (5) 7 (4) 13 (7) 9 (5) 5 (3) 3 (2) 31 (18)

Women, n (%) 85 (49) 69 (39) 69 (39) 29 (17) 13 (7) 20 (11) 15 (9) 15 (9) 8 (5) 11 (6) 0 (0) 0 (0) 19 (11)

Diagnosis of other includes spasmodic dysphonia, pseudosulcus vocalis, hemorrhage, uctuating neuroasymmetry, neurogenic dysphonia, anterior glottic web, sulcus vergeture, sulcus vocalis, and candida.

summarizes additional LEMG data and neurological ndings (which were counted separately). OVMs were available for 95% (N 167) of the subjects in this study. These data were interpreted using the normative data provided by KayPENTAX, as can be reviewed in Table 6. The percentage of subjects who were above the KayPENTAX normative thresholds may be seen in Table 6. Notably, 69.7% of subjects were above the threshold for soft phonation index (SPI). Fifty VHI questionnaires were completed; 26 were lled out by women and 24 by men. The range of calculated VHIs was 4 104 (average 43.9; SD 28.2). The VHI results may be seen in Table 7. Signicant (P < 0.05) positive correlations were found between total VHI and percent jitter, absolute jitter, percent shimmer, and absolute shimmer; signicant negative correlation was found between total VHI and mean phonation time (seconds) (Spearman rank order correlations, Table 8). Various treatments were used for these patients. Eighty-three patients underwent some type of interventional procedure, in the ofce or operating room. Sixteen patients underwent a type I thyroplasty, 14 had some type of mass excised, and 34 had vocal fold injections with Botulinum toxin (N 14), fat (N 12), or dexamethosone (N 8). In addition, voice therapy was offered to all patients. Eighty-three (47%) patients completed six or more sessions of voice therapy. Forty-eight (58%) of these individuals were women, and 35 were men (42%). Finally, most patients were started on a medication,

particularly for LPR. One hundred forty-eight (84%) patients were started on a proton pump inhibitor, and one hundred nine of these also were started on Ranitidine 300 mg once daily. Three patients were treated with pyridostigmine for treatment of myasthenia gravis conrmed by electromyography.

DISCUSSION The prevalence of voice disorders or dysphonia in elderly individuals has been reported to be between 12% and 47%.46 Despite this high prevalence, studies are rare. Indeed, there is no agreement even on what parameters should be studied. This shortage of data may be because the effect of age alone on the voice is difcult to determine. Older patients may suffer from an array of comorbid conditions and take multiple medications. In a study by Woo et al, 81 of 151 elderly patients had systemic illnesses involving one or more systems. The two most common comorbid conditions were pulmonary disease and hypertensive cardiac disease.7 Furthermore, there are many potential adverse effects from common medications prescribed in the aging population that may also affect vocal quality.8 In the study by Woo et al, more than one-third of patients were on one or more medications.7 Of the 175 elderly patients, the most common voice complaints included hoarseness, inability to project/decreased volume, and excessive throat clearing/phlegm. Diagnoses

TABLE 5. LEMG Results Obtained by a Neurologist, Board Certied in Electrophysiology Laryngeal EMG (N 164) Superior laryngeal nerve (7090% recruitment) Superior laryngeal nerve (<70% recruitment) Tremor Spasmodic dysphonia Total, n (%) 121 (74) 32 (20) 21 (13) 5 (3) Men, n (%) 55 (34) 16 (10) 6 (4) 2 (1) Women, n (%) 66 (40) 16 (10) 15 (9) 3 (2)

Values reect weakness in the left or right vocal fold, or both. Each nding was counted individually.

Naomi D. Gregory, et al

Voice Disorders in the Elderly

257

TABLE 6. Percentage of Subjects Who are Above the KayPENTAX Normative Thresholds Measure NHR VTI SPI F0 (Hz) Jitter (%) Jitter (abs.) Shimmer (%) Shimmer (abs) MFR (mL/s) MPT (s) Description Noise to harmonic ratio Voice turbulence index Soft phonation index Average fundamental frequency Variability of frequency Variability of amplitude Shimmer in dB Mean ow rate Maximum phonation time Normative Threshold From MDVP (KayPENTAX) 0.19 0.061 14.12 n/a 1.04 83.2 3.81 0.35 n/a n/a Percent of Subjects Above the Threshold 26.4 (43/163) 13.7 (21/153) 69.7 (106/152) 57.8 (67/116) 50.7 (75/148) 49.1 (57/116) 48.3 (72/149)

Abbreviations: MDVP, Multidimensional Voice Program.

associated with these complaints most commonly included laryngopharyngeal reux, paresis, MTD, and vocal fold mass. Laryngopharyngeal reux and MTD may be considered secondary diagnoses in many patients, although these conditions are important contributors to voice impairment. Woo et al also noted a high incidence of secondary diagnoses contributing to dysphonia, especially in patients with benign vocal lesions and inammatory processes.7 Previous studies have demonstrated vocal fold bowing, paralysis, benign vocal fold lesions, voice tremor, and spasmodic dysphonia as the most common diagnoses in the elderly.3 Other studies have identied common strobovideoscopic ndings in the elderly to include prominent vocal processes and membranous spindle-shaped glottic chink during phonation.9 The high incidence of specic strobovideolaryngoscopic ndings in our study and other studies should aid in eliminating the vague diagnosis of presbylaryngis in most elderly patients. This term should be used strictly when no pathologic ndings are identied, except for those of age relatedanatomic (atrophy) and physiological changes.7,10 Our review identied at least one pathologic factor contributing to dysphonia, other than aging changes, in all the patients. Most patients in this study received a multidisciplinary approach to treatment including medication, voice therapy, and/or surgical intervention. Medical therapy was aimed largely at treating the inammatory changes related to LPR. Forty-seven percent of patients had six or more sessions of voice therapy. The remaining patients either did not complete six full sessions or refused voice therapy, although all patients were offered this therapeutic approach during their initial visit. A recent study by Berg et al11 found a statistically signicant improvement in the voice-related quality of life (VRQOL) measure in subjects who received voice therapy, with 74% experiencing an overall improvement in VRQOL (voice turbulence improves the quality of life). One limitation of our study is that VHI scores were only calculated at the time of the initial visit, not after therapeutic interventions. Finally, almost onethird of patients were able to receive surgical intervention either in the ofce or in the operating room. All patients who had surgical intervention had received voice therapy.

These ndings are signicant because more than 50% of patients reported signicant quality of life impairment resulting from their dysphonia.3 Indeed, the high average score of the VHI in this study also suggests that geriatric voice patients who seek medical attention experience signicant dissatisfaction because of their dysphonia. Consequently, geriatric dysphonia may have serious psychosocial implications; communication disorders may be associated with social withdrawal, depression, or anxiety. When the VHI results were correlated with the OVMs, signicant correlations with mean jitter (%), jitter (abs.), shimmer (%), and maximum phonation time (MPT) were found, suggesting that these particular parameters correlate with high levels of voice dissatisfaction among patients older than 65 years. Specically, as VHI scores increased, so did jitter (%), jitter (abs.), and shimmer (%). MPT was lower in patients with VHI scores more than 60 than in those with lower VHI scores. Voice measures obtained for this population were also compared with the normative thresholds as determined by KayPENTAX. Interestingly, 69.7% of subjects were found to be above the threshold for SPI. Conversely, only 13.7% of subjects were above the normative threshold for VT index. Additional studies will be required to determine signicance of these observations. In future studies, OVMs in patients aged 65 years and older with dysphonia should be compared with a group of young patients with similar voice abnormalities and also should be obtained for a cohort of elderly patients without voice complaints to determine whether standard normative values are appropriate in this population. Multidisciplinary treatment should be considered for every elderly voice patient, even when comorbid states, inability to
TABLE 7. VHI Obtained from a Questionnaire at the Initial Visit VHI (N 50, 29%) Total P (physical) F (functional) E (emotional) Value Range 4104 238 037 036 Value Average (SD) 43.9 (28.2) 17.06 (8.9) 13.4 (11.1) 13.5 (9.8)

258
TABLE 8. Correlation Between VHI and OVMs OVM NHR VTI SPI F0 (Hz) Jitter (%) Jitter (abs.) Shimmer (%) Shimmer (dB) MRF (mL/s) MPT (s)
*P < 0.05.

Journal of Voice, Vol. 26, No. 2, 2012

Spearman Rank Order Correlation Coefcient 0.126 0.060 0.139 0.269 0.465* 0.325* 0.425* 0.418* 0.070 0.576*

accurate and comprehensive diagnosis possible and important. Furthermore, with increasing medical technology, more elderly patients are continuing to have active vocal lives, either recreationally or professionally, emphasizing the importance of geriatric vocal health. Additional research is necessary to establish parameters of normal and abnormal function in elderly patients and determine whether parameters used for younger patients can be applied appropriately in a geriatric population. Where needed, geriatric normative values should be established and used routinely. REFERENCES
1. Baken RJ. The aged voice: a new hypothesis. J Voice. 2005;19:317325. 2. Sataloff RT, Linville SE. The effects of age on the voice In: Sataloff RT, ed. Professional Voice: The Science and Art of Clinical Care. 3rd ed. San Diego, CA: Plural Publishing, Inc.; 2005:497512. 3. Lundy DS, Silva C, Casiano RR, et al. Cause of hoarseness in elderly patients. Otolaryngol Head Neck Surg. 1998;118:481485. 4. Golub JS, Chen P, Otto K, et al. Prevalence of perceived dysphonia in a geriatric population. J Am Geriatr Soc. 2006;54:17361739. 5. Roy N, Stemple J, Merrill RM, et al. Epidemiology of voice disorders in the elderly: preliminary ndings. Laryngoscope. 2007;117:628633. 6. Turley R, Cohen S. Impact of voice and swallowing problems in the elderly. Otolaryngol Head Neck Surg. 2009;140:3336. 7. Woo P, Casper J, Colton R, et al. Dysphonia in the aging: physiology versus disease. Laryngoscope. 1992;102:139144. 8. Abaza MM, Levy S, Hawkshaw M, et al. Effect of medications on the voice. Otolaryngol Clin North Am. 2007;40:10811090. 9. Hagen P, Lyons G, Nuss D, et al. Dysphonia in the elderly: diagnosis and management of age-related voice changes. South Med J. 1996;89:204207. 10. Pontes P, Brasolotto A, Behlau M. Glottic characteristics and voice complaint in the elderly. J Voice. 2005;19:8493. 11. Berg EE, Hapner E, Klein A, et al. Voice therapy improves quality of life in age-related dysphonia: a case-control study. J Voice. 2008;22:7074.

travel, or altered living conditions make such treatment logistically difcult. The nearly universal participation in some type of treatment regimen suggests that many patients aged 65 years and older are indeed willing to undergo treatment to improve their voice dysfunction. CONCLUSION Evaluation of vocal complaints in a growing elderly population will become increasingly more important. The diagnosis of presbylaryngis should be used precisely and sparingly, with increasing technology for diagnosing subtle vocal abnormalities and a multidisciplinary approach for intervention, making