Journal of Voice

Vol. 14, No. 3, pp. 378-386

© 2000 The Voice Foundation

Voice-Related Quality of Life (V-RQOL) Following

Type I Thyroplasty for Unilateral Vocal Fold Paralysis

*?Norman D. Hogikyan, ~Walter R Wodchis, *§Jeffrey E. Terrell,

*Carol R. Bradford, and IfRamon M. Esclamado
*Department of Otolaryngology--Head and Neck Surgery, University of Michigan Health System; 1"VocalHealth Center,
University of Michigan Health System; .~School of Public Health, University of Michigan, Ann Arbor, Michigan;
§Health Services and Development Center for Excellence, VeteransAffairs Hospital, Ann Arbor, Michigan;
fFormerly affiliated with Department of Otolaryngology--Head and Neck Surgery, University of Michigan
Health System, Ann Arbor, Michigan: currently affiliated with Department of Otolaryngology--
Head and Neck Surgery, The Cleveland Clinic Foundation, Cleveland, Ohio

Summary:Unilateral vocal fold paralysis is a common clinical problem which

frequently causes severe dysphonia. Various treatment options exist for this con-
dition, with the type I thyroplasty being one of the more commonly performed
surgical procedures for vocal rehabilitation. The Voice-Related Quality of Life (V-
RQOL) Measure is a validated outcomes instrument for voice disorders. This
study measured the V-RQOL of patients with unilateral vocal fold paralysis who
had undergone a type I thyroplasty and compared these scores to those of patients
with untreated and uncompensated unilateral vocal fold paralysis and to normals.
Treated patients had significantly higher domain and overall V-RQOL scores than
untreated patients, but also scored lower than normals. These differences were true
across gender and age. Patients who were more distant from surgery had lower V-
RQOL scores than those who had more recently been treated. It is concluded that
type I thyroplasty leads to a significantly higher V-RQOL for patients with unilat-
eral vocal fold paralysis. This study also demonstrates further the utility of patient-
oriented measures of treatment outcome. Key Words: Unilateral vocal fold paral-
ysis-Voice-related quality of life (V-RQOL)--Type I thyroplasty.

Unilateral vocal fold paralysis (UVFP) usually phagia. Treatment options for UVFP are varied, and
causes severe dysphonia and can also lead to dys- were recently reviewed in an article by the principal
authorJ Laryngeal framework surgery consisting of
the type I thyroplasty (TPY I) is one of the more
Accepted for publication July 30, 1999. commonly employed surgical treatment options for
This paper was presented at The Voice Foundation's 28th patients with persistent dysphonia due to UVFP. This
Annual Symposium: Care of the Professional Voice, June 6, procedure medializes the paralyzed vocal fold to en-
1999, Philadelphia, Pennsylvania,. hance glottic closure and vibratory dynamics. The
Address correspondence and reprint requests to Norman D.
surgical literature contains several papers describing
Hogikyan, Department of Otolaryngology,University of Michi-
gan Medical Center, 1904 Taubman Center, 1500 East Medical procedures considered to be precursors of the con-
Center Drive, Ann Arbor, Michigan 48109-0312, USA. temporary operation,2, 3 but Isshiki et al is credited
e-mail: nhogikya@umich.edu with description of the fundamental procedure used

378
 VOICE-RELATED QUALITY OF LIFE (V-RQOL) FOLLOWING TYPE I THYROPLASTY 379

today4,5 and popularized in the United States by MATERIALS AND METHODS

Koufman and others. 6
Numerous authors have examined treatment out- Patient groups
comes following TPY I using perceptual, acoustic, All patients treated surgically for UVFP through
aerodynamic, and videostroboscopic assessments of the University of Michigan Health System during the
voice and of laryngeal function. 7-14 It is uniformly period from July 1, 1993 through August 31, 1998
felt that this procedure can improve and in some cas- were identified through a search of the institutional
es eliminate the dysphonia associated with UVFP as surgical database. Patients who had undergone only
measured by such methods. Noting the importance of an injection laryngoplasty using Gelfoam (Pharma-
patient-oriented measures of outcome, Harries and cia and Upjohn, Kalamazoo, Michigan) paste, a sub-
Morrison wrote that "the degree of patient satisfac- stance intended to provide temporary voice improve-
tion is one of the most important parameters of the ment, were then excluded from the search results.
success of surgical treatment, perhaps more than any Seventy-three patients who had undergone 77 proce-
objective changes.''11 Two recent papers 15,16 ex- dures were identified. These surgical procedures
plored issues of quality of life (QOL) as related were: 57 TPY I, 4 revision TPY I, 10 laryngeal rein-
specifically to vocal fold paralysis, and a third pa- nervation operations, and 6 injection laryngoplasties.
per 17 introduced an instrument to measure QOL is- The 61 total TPY I had been performed in 58 pa-
sues related to voice disorders in general, or voice-re- tients, and by 4 different surgeons. Of these 58 pa-
lated quality of life (V-RQOL). Baba and colleagues tients: 9 were dead, 2 had previously undergone im-
used a nonvalidated survey to demonstrate a negative plant removal for complications without replacement,
impact of persistent recurrent laryngeal nerve injury 1 has a revision pending, 1 had developed cognitive
upon the QOL of patients undergoing esophagecto- deficits precluding questionnaire completion, and 1
my with regional lymphadenectomy for treatment of had recovered mobility in the treated vocal fold.
esophageal carcinoma. 15 Gliklich and colleagues re-
These 14 patients were ineligible for the study, leav-
cently validated a voice outcome survey (VOS)
ing 44 eligible patients in this group. The V-RQOL
specifically for the purpose of determining treatment
Measure was mailed to these patients via United
outcome of patients with UVFP. 16 This 5-question
States mail with a cover letter explaining the purpose
instrument contains items related to both voice and
swallowing, and was shown to be valid, reliable, and of the study, and asking them to return the question-
responsive. The authors felt that the brevity of the in- naire in a provided self-addressed and stamped enve-
strument made it easy to administer, but also made it lope. Each patient was also asked to give a rating of
less comprehensive or diagnostically discriminating his or her current voice using a 5-point Likert-style
than a lengthier questionnaire. scale of global voice quality.
The Voice-Related Quality of Life (V-RQOL) Twenty consecutive new patients with UVFP who
Measure is a two-domain, 10-item instrument that were seen by the senior author from July 1, 1998
has been shown in a diverse population of dysphonic through February 28, 1999, but had not yet under-
patients to be valid, reliable, and responsive to change gone surgical treatment by August 31, 1998 were al-
(Appendix A). It is now used clinically to determine so reviewed. To enhance the comparability of this
the degree of impact that a voice disorder is having group with the preoperative status of the treated
upon a patient's V-RQOL, and to measure treatment group, patients with well compensated UVFP who
outcomes in voice patients. The purpose of the cur- were not candidates for a future TPY I at our institu-
rent study was to measure the V-RQOL of patients tion were then eliminated from the list. We define a
treated for UVFP with a TPY I during a recent 5-year paralysis as well compensated when complete glottic
period, and to compare these values with those of closure is seen with phonation during stroboscopy,
subsequent new patients with untreated and uncom- the speaking voice is perceptually normal to mildly
pensated UVFP, and nonvoice patients with no voice- dysphonic, and the patient experiences little to no vo-
related complaints. This study was undertaken with cal difficulties with their habitual voice use. Three
the approval of the University of Michigan Institu- patients met these criteria, leaving 17 dysphonic pa-
tional Review Board for Human Studies. tients with UVFP constituting the untreated group.

Journal of Voice, Vol. 14, No. 3, 2000

380 NORMAN D. HOGIKYAN ET AL

These patients had presented for care during the time other patients in the same study population. Sensitiv-
period when the V-RQOL Measure had already been ity analyses showed this to have an equivalent effect
introduced into routine clinical use, and thus the V- to removing the item from the individual's scoring
RQOL and global voice rating data were obtained formula (data not shown). The only item that re-
from patient charts. quired imputation asked about employment related
Archival V-RQOL data for 22 nonvoice patients V-RQOL. Because some patients were retired, they
were used for the third group in the present study. left this item blank.
These 22 patients had no voice complaints, and their Mean V-RQOL scores, standard deviations, and
scores are used for purposes of comparisons with a the range of scores were calculated for patients in
normal or nonvoice-disorder population. They are each of the three populations. Standard t tests were
heretofore referred to as the normal group or normals. used to test whether scores from the three popula-
tions were similar. Most comparisons were conduct-
Statistical analyses ed for responses of treated and untreated patients.
Survey responses from the returned questionnaires Reported P values are based on two-tailed signifi-
were recorded. V-RQOL responses from normals, cance tests. An F test determined that the sample
treated, and untreated patients were merged into a variance for normals were significantly different
single database. An indicator variable was used to from treated patients (F = 1.00, P = 0.00), while sam-
distinguish between patient groups. SAS v6.12 and ple variance from treated and untreated patient popu-
Microsoft Excel were used to manage the data and lations were not significantly different (F = 0.85, P =
calculate results, including statistical tests. 0.64). As a result, t tests for treated and untreated pa-
The etiologies for UVFP were grouped into four tients assume equal variances and tests for treated
categories: idiopathic, postsurgical (nonmalignant), versus normals assume unequal variances. P values
malignancy, and other. Frequencies were calculated from tests are reported in the accompanying tables.
for demographic characteristics and by etiology. The
V-RQOL overall scores were calculated along with RESULTS
Social-Emotional and Physical Functioning domain
scores. These scores are based on 10, 4, and 6 items, Thirty of the 44 eligible treated patients returned
respectively. The formula used to calculate all three questionnaires for a response rate of 68%. Demo-
scores (Appendix B) provides an index of V-RQOL graphic data for the three study populations, and eti-
on a scale of 0-100, where 0 is extremely poor V- ology of paralysis for the treated and untreated
RQOL and 100 is the highest possible V-RQOL that groups are shown in Table 1.
one could expect to attain. Table 2 shows the means and ranges of domain and
Missing data were imputed using the mean re- overall V-RQOL scores by study group. Scores for
sponse for the relevant item based on responses of the treated group were significantly higher than for

T A B L E 1. Demographic Data for All Groups, and Etiology of Paralysis

for Treated and Untreated Groups
Normals Untreated Treated

n 22 17 30
Mean age 47.64 52.56 61.26
Female 13 (59%) 8 (47%) 14 (47%)
Diagnosis category
Idiopathic 5 (29%) 3 (10%)
Postsurgical (nonmalignant) 6 (35%) 19 (63%)
Malignancy 4 (24%) 7 (23%)
Other 2 02%) 1 (3 %)

Journal of Voice, Vol. 14, No. 3, 2000

 VOICE-RELATED QUALITY OF LIFE (V-RQOL) FOLLOWING TYPE I THYROPLASTY 381

T A B L E 2. Means and Ranges for Domain and Overall V-RQOL Scores by Study Group

Normals Untreated Treated

Mean (n = 22) (SD) Mean (n = 17) (SD) Mean* (n = 30) (SD)

Social-emotional 98.86 (4.16) 39.71 (28.64) 75.00 (24.18)

(range) (81.25, 100) (0.00, 87.50) (25.00, 100)
Physical functioning 97.35 (5.38) 27.85 (20.52) 72.70 (18.18)
(range) (85.00, 100) (0.00, 66.67) (37.50, 100)
Overall V-RQOL 97.95 (3.71 ) 32.59 (21.38) 73.62 (19.50)
(range) (75.00, 100) (0.00, 75.00) (35.00, 100)

*All P-values < 0.01 (treated vs untreated, and treated vs normals).

Abbreviations: SD, standard deviation.

the untreated group, and were also significantly low- DISCUSSION

er than the scores for normals. Categorical self-rat-
ings of global voice quality and mean overall V- Results of this study clearly demonstrate that the
RQOL scores by voice rating category are shown in average V-RQOL was much higher in treated versus
Table 3 for the treated and untreated groups. Most of untreated patients. This was true for the social-emo-
the untreated patients rated their voices as poor, tional and physical functioning domains, as well as
while very good was the most common rating in the for the overall scores. Magnitude of this difference
treated group. Relative magnitude of V-RQOL scores was convincing, with treated scores approximately
correlated well with voice rating category. double the untreated scores, and the level of signifi-
In Table 4, mean overall V-RQOL scores by gender cance was also impressive at <0.01. These unequivo-
cal findings accomplish the primary goal of this
and age category are shown for each study group.
study, which was to determine whether such a differ-
Treated scores were significantly different from un-
ence existed or not. They also contribute an impor-
treated scores within each category. Mean V-RQOL
tant new dimension of support for use o f T P Y I in the
scores versus time since surgery are shown in Table treatment of UVFP.
5 for the treated group. Patients who were over 2 Because the V-RQOL Measure is a relatively new
years from surgery had significantly lower scores instrument, preoperative data were not available for
than patients within 1 year of their TPY I. the treated group. There were however data for a set
of dysphonic patients with untreated and uncompen-
sated UVFP readily available from the pool of new
T A B L E 3. Self-ratings of Global Voice Quality and patients presenting for care to the same institution in
Overall V-RQOL Scores by Voice Rating for Treated more recent months. We feel that the vocal capabili-
and Untreated Groups ties and V-RQOL of the untreated group can be ap-
propriately considered representative of the presumed
Untreated Treated
preoperative status of the treated group.
Voice Rating n V-RQOL n V-RQOL The self-ratings of global voice quality also showed
a different pattern in the treated versus the untreated
Poor 13 28.01 groups (Table 3), with poor being the most common
Fair 4 47.50 8 57.80 rating in untreated patients and very good the most
Good 10 63.87 common in the treated group. Overall V-RQOL
Very good 11 92.50 scores correlated well with global voice ratings.
To try to better understand the range in V-RQOL
Excellent 1 90.00
scores demonstrated by the untreated and treated

Journal of Voice, Vol. 14, No. 3, 2000

382 N O R MAN D. HOGIKYAN E T AL

T A B L E 4. Overall V-RQOL Scores by Gender and Age for Each Study Group

Normals Untreated Treated

Mean (SD) Mean (SD) Mean* (SD)

Female 96.67 (5.45) 27.66 (24.66) 67.14 (20.28)

Male 98.85 (2.19) 36.98 (18.33) 79.29 (17.46)
Age (years)
Under 45 99.06 (1.29) 29.58 (27.54) 80.74 (18.62)
45-64 95.63 (5.79) 30.30 (16.69) 77.95 (17.77)
65+ 99.58 (1.02) 38.96 (21.44) **67.67 (20.60)

*All P-values < 0.01 (treated vs untreated), except ** < 0.05.

T A B L E 5. Domain and Overall V-RQOL Scores by Time Since Surgery for Treated Group

< 1 Year 1-2 Years Over 2 Years

Mean (n = 8) (SD) Mean (n = 6) (SD) Mean* (n = 16) (SD)

Social-emotional 85.94 (15.86) 80.21 (28.05) 67.58 (22.67)

Physical functioning 84.90 (13.17) 75.69 ( 17.91) 65.48 (16.19)
Overall V-RQOL 85.31 ( 13.66) 77.50 (21.75) 66.32 ( 17.06)

* Significantly different from < 1 year (P < 0.05).

groups, the groups were broken down by gender and period. Longitudinal data over a longer time period
age categories (Table 4). A statistically significant are not available from the current literature. It is pos-
difference remained between the two groups though, sible that there truly is a decline in postoperative vo-
looking across the gender and age categories. There cal capabilities over a period of years; continued den-
was a tendency within the treated group for male pa- ervation atrophy or fibrosis in vocal fold musculature
tients to score higher than females, and for older pa- could lead to further decline in glottic closure or vi-
tients to score lower than younger ones; however, the bratory dynamics. An alternative explanation may be
standard deviations were fairly large and differences that vocal capabilities remain constant, but patients'
were not significant. Differences among the various expectations become nigher with time and they are
etiologies for UVFP were also looked at (data not more bothered by vocal problems. We have found that
shown); however, the relatively small numbers in patients who have fairly recently been operated upon
some groups precluded meaningful comparisons. will think of their voices in comparison with the un-
We found the data for the treated group regarding treated paralysis voice, while those who are much fur-
V-RQOL score and time from surgery particularly in- ther from the time of surgery tend to think of their
teresting, although in many ways it probably raises vocal capabilities in comparison with normals. The is-
more questions than it answers. The patients who sue of change over time will also be answered by the
were >2 years postsurgery had a significantly lower ongoing prospective study at our institution exploring
V-RQOL than patients who were <1 year from a wide variety of vocal outcome parameters longitu-
surgery, but all remained higher than untreated scores. dinally in patients treated surgically for UVFP.
Lu et al looked at longitudinal changes in voice fol- Further comment is warranted by the fact that
lowing TPY I, 14 and found that perceptual, videostro- mean V-RQOL scores in the treated group were sig-
boscopic, acoustic, and aerodynamic measures re- nificantly less than those of normals. The relative dif-
mained fairly stable over a 6-month postoperative ference in scores was not as great as for treated ver-

Journal of Voice, Vol. 14, No. 3, 2000

 VOICE-RELATED Q U A L I T Y O F L I F E (V-RQOL) F O L L O W I N G T Y P E I T H Y R O P L A S T Y 383

sus untreated patients, but the difference did exist. ment, information at the core of patient satisfaction
This concurs with the clinical observation that treated can be reliably measured.
patients generally feel that their voices perform well in
most circumstances, but will not be optimal in times CONCLUSIONS
when high vocal intensity or endurance is required.
This should not really be surprising when considering In this study, the average V-RQOL scores of pa-
the fact that the TPY I is a static, geometric solution tients with UVFP who had been treated with a TPY I
for a dynamic neurological problem of laryngeal func- were compared to those of untreated patients and to
tion. These data speak to the fact that although this is normals. Treated patients scored significantly higher
an extremely good treatment option, there is still re- than untreated patients, but were also lower than nor-
search to be done in treatment of UVFP. The ideal so- mals. The V-RQOL scores of patients who were more
lution, ability to restore motion-specific innervation to distant from surgery were lower than those who had
the paralyzed vocal fold, is not yet at hand. been operated upon more recently. Further research is
This paper also contributes to the growing interest necessary to better define longitudinal V-RQOL and
in measuring patient-oriented outcomes for patients voice quality changes in patients undergoing TPY I.
Patient-oriented variables such as V-RQOL are valu-
with voice disorders, and demonstrates further the
able ways of measuring treatment outcomes.
utility of such data. As noted earlier by Harries and
Morrison, "the degree of patient satisfaction is one of Acknowledgment: The authors wish to acknowledge Dr
the most important parameters of the success of sur- Jack Wheeler and the University of Michigan Health Ser-
gical treatment. ''1~ Using the V-RQOL Measure or vices Research Initiative for technical assistance and fi-
other validated patient-oriented outcomes instru- nancial support related to this project.

Journal of Voice, Vol. 14, No. 3, 2000

384 NORMAN D. HOGIKYAN ET AL

APPENDIX A

Voice-Related Quality of Life (V-RQOL) Measure

University of Michigan

NAME: DATE:
We are trying to learn more about how a voice problem can interfere with your day-to-day activities. On this
paper, you will find a list of possible voice-related problems. Please answer all questions based upon what your
voice has been like over the past two weeks. There are no "right" or "wrong" answers.
Considering both how severe the problem is when you get it, and how frequently it happens, please rate each
item below on how "bad" it is (that is, the amount of each problem that you have). Use the following scale for
rating the amount of the problem:
1 = None, not a problem
2 = A small amount
3 = A moderate (medium) amount
4 = A lot
5 = Problem is as " b a d as it can be"
Because of my voice, H o w much of a problem is this?
1. I have trouble speaking loudly or being heard in noisy situations. 1 2 3 4 5
2. I run out of air and need to take frequent breaths when talking. 1 2 3 4 5
3. I sometimes do not know what will come out when I begin speaking. 1 2 3 4 5
4. I am sometimes anxious or frustrated (because of my voice). 1 2 3 4 5
5. I sometimes get depressed (because of my voice). 1 2 3 4 5
6. I have trouble using the telephone (because of my voice). 1 2 3 4 5
7. I have trouble doing my job or practicing my profession (because of my voice). 1 2 3 4 5
8. I avoid going out socially (because of my voice). 1 2 3 4 5
9. I have to repeat myself to be understood. 1 2 3 4 5
10. I have become less outgoing (because of my voice). 1 2 3 4 5

The overall quality of my voice during the last 2 weeks has been (please circle):
Poor Fair Good Very Good Excellent

Journal of Voice, Vol. 14, No. 3, 2000

 VOICE-RELATED QUALITY OF LIFE (V-RQOL) FOLLOWING TYPE I THYROPLASTY 385

APPENDIX B

Scoring Algorithm for V-RQOL Measure

V-RQOL General Scoring Algorithm

100 - (Raw Score - # items in domain or total I x 100
\ Highest Possible Raw Score - # items }

Social-Emotional Domain (Items 4, 5, 8, 10)

100 - (Raw Sc°re

× 1 0 0 1- 64)

Physical Functioning Domain (Items 1, 2, 3, 6, 7, 9)

100 - (Raw Sc°re

× 1 0 0 2- 46)

Total Score (Items 1-10)

1 0 0 - ( Raw Sc°re
40 - 10) X100

Example for Total Score

If raw score is 30 (such as if a "medium problem" exists with all items), then:

100- ~-~ x 100

= 100 - (0.5 X 100) = 100 - 50 = 50 standard score

Journal of Voice, Vol. 14, No. 3, 2000

386 NORMAN D. HOGIKYAN ET AL

REFERENCES 10. D'Antonio LL, Wigley TL, Zimmerman GJ. Quantitative

measures of laryngeal function following teflon injection or
1. Hoff PT, Hogikyan ND. Unilateral vocal fold paralysis. Curr thyroplasty type I. Laryngoscope. 1995;105:256-262.
Opin Otolaryngol Head Neck Surg. 1996;4:176-181. 11. Harries ML, Morrison M. Short-term results of laryngeal
2. Payr. Plastik am Schildknorpel zur Behebung der Folgen ein- framework surgery-thyroplasty type I: a pilot study. J Oto-
seitiger Stimmbandlahmung. Dtsch Med Wochenschr. 1915; laryngol. 1995;24:281-287.
43:1265. 12. Omori K, Kacker A, Slavit D, Blaugrand SM. Quantitative
3. Meurman Y. Operative mediofixation of the vocal cord in videostroboscopic measurement of glottal gap and vocal
complete unilateral paralysis. Arch Otolaryngol. 1952;55: function: an analysis of thyroplasty type I. Ann Otol Rhinol
544-553. Laryngol. 1996;105:280-285.
4. Isshiki N, Morita H, Okamura H, Hiramoto M. Thyroplasty as 13. Adams SG, Irish JC, Durkin LC, Wong DLH, Brown DH.
a new phonosurgical technique. Acta Otolaryngol (Stockh). Evaluation of vocal function in unilateral vocal fold paraly-
1974;78:451-457. sis following thyroplastic surgery. J OtolaryngoL 1996;25:
165-170.
5. Isshiki N, Okamura H, Ishikawa T. Thyroplasty type I (later-
14. Lu FL, Casiano RR, Lundy DS, Xue JW. Longitudinal eval-
al compression) for dysphonia due to vocal cord paralysis or
uation of vocal function after thyroplasty type I in the treat-
atrophy. Acta Otolaryngol (Stockh). 1975;80:465-473.
ment of unilateral vocal paralysis. Laryngoscope. 1996; 106:
6. Koufman JA. Laryngoplasty for vocal cord medialization: an
573-577.
alternative to Teflon. Laryngoscope. t986;96:726-730,
I5. Baba M, Natsugoe S, Shimado M, Na~ano S, Noguchi Y,
7. Gray SD, Barkmeier J, Jones D, Titze I, Dmker D. Vocal eval- Kawachi K, Kusano C, Aikou T. Does hoarseness of voice
uation of thyroplastic surgery in the treatment of unilateral vo- from recurrent nerve paralysis after esophagectomy for car-
cal fold paralysis. Laryngoscope. 1992;102:415-421. cinoma influence patient quality of life? J Am Coll Surg.
8. Netterville JL, Stone RE, Lukens ES, Civantos FJ, Ossoff 1999;188:231-236.
RH. Silastic medialization and arytenoid adduction: the Van- 16. Gliklich RE, Glovsky RM, Montgomery WW. Validation of
derbilt experience. Ann Otol Rhinol Laryngol. 1993;102:413- a voice outcome survey for unilateral vocal cord paralysis.
424. Otolaryngol Head Neck Surg. 1999;120!153-158.
9. Leder SB, Sasaki CT. Long-term changes in vocal quality fol- 17. Hogikyan ND, Sethuraman G. Validation of an instrument to
lowing Isskiki thyroplasty type I. Laryngoscope. 1994;104: measure voice-related quality of life (V-RQOL). J Voice.
275-277. 1999;13:557-569.

Journal of Voice, Vol. 14, No. 3, 2000