Download Voice Changes after Androgen Therapy for

The Laryngoscope
Lippincott Williams & Wilkins, Inc.
© 2004 The American Laryngological,
Rhinological and Otological Society, Inc.
Voice Changes after Androgen Therapy for
Hypogonadotrophic Hypogonadism
Timur Akcam, MD; Erol Bolu, MD; Albert L. Merati, MD; Coskun Durmus, MD; Mustafa Gerek, MD;
Yalcin Ozkaptan, MD
Objectives/Hypothesis: Males with isolated hypogonadotropic hypogonadism (IHH) fail to undergo
normal sexual development, including the lack of
masculinization of the larynx. The objective of this
study was to measure the mean vocal fundamental
frequency (MF0) in IHH patients and determine the
impact of androgen treatment. An additional aim was
to compare the MF0 between IHH patients and controls. Study Design: Prospective observational study.
Methods: Twenty-four patients with IHH were identified along with 30 normal males and females. Voice
recordings were obtained on all subjects. Androgen
therapy was administered to the IHH patients. The
MF0 and serum sex hormone levels were measured
before treatment and at intervals during therapy.
These results were compared with the pretreatment
data within the IHH group. Voice parameters were
also compared between the pre- and posttreatment
IHH patients and the normal males and females. Results: The MF0 in untreated IHH patients was 229 ⴞ 41
Hz. This was intermediate between the normal male
(150 ⴞ 22 Hz, P < .001) and normal female patients
(256 ⴞ 29 Hz, P < .01). After treatment, the MF0 in the
IHH group decreased to 173 ⴞ 30 Hz (P < .0001); indeed, their posttreatment MF0 approached that of
normal males (P < .08). Serum hormone levels responded to the injected testosterone, but these levels
did not directly correlate with MF0. Conclusions: MF0
in IHH patients is intermediate between normal male
and female levels. After treatment with testosterone,
these values approach the range of normal males.
This prospective study details the impact of androgens on the larynx and vocal function in patients with
IHH. Key Words: Hypogonadism, voice, puberty, androgen, larynx.
Laryngoscope, 114:1587–1591, 2004
From the Department of Otolaryngology–Head and Neck Surgery
(T.A., C.D., M.G., Y.O.), Gulhane Military Medical Academy, Ankara, Turkey;
the Department of Endocrinology and Metabolism (E.B.), Gulhane Military
Medical Academy, Ankara, Turkey; and the Division of Laryngology
(A.L.M.), Department of Otolaryngology and Communication Sciences, Medical College of Wisconsin, Milwaukee, WI, U.S.A.
Editor’s Note: This Manuscript was accepted for publication February 10, 2004.
Send Correspondence to Dr. Timur Akcam, Assistant Professor, Department of Otolaryngology–Head and Neck Surgery, Gulhane Military
Medical Academy, Etlik-Ankara, 06018 Turkey. E-mail: takcam@gata.
edu.tr
Laryngoscope 114: September 2004
INTRODUCTION
The differing characteristics of male and female
voices are easily recognized, particularly those changes
that occur around the time of puberty. Normal sexual
development at this time is characterized by maturation
of the genitals, the appearance of secondary sexual characteristics, as well as changes in mood and behavior.1 In
sexually mature males, the processes of spermatogenesis
and sex steroid production are regulated by the pituitary
gonadotropins luteinizing hormone (LH) and folliclestimulating hormone (FSH). In male idiopathic hypogonadotropic hypogonadism (IHH), gonadotropin secretion is
impaired, resulting in a failure of sexual maturation.
The impact of sexual development on the voice is
illustrated by the retention of feminine voice characteristics in boys who undergo orchiectomy before puberty.2
Similarly, when androgenic agents are used in females,
significant voice changes may occur.3–5 These effects may
also be seen in trained male singers with hypogonadism
who are given testosterone replacement. In the latter case,
significant vocal register and voice quality changes have
been reported.6
The primary objective of this study was to determine
the mean fundamental voice frequency (MF0) in a population of males with IHH and compare this with normal
males and females. The secondary objective was to determine the impact of testosterone treatment by measuring
the MF0 over the course of androgen treatment.
MATERIALS AND METHODS
Subjects
In this prospectively acquired series, 24 patients with IHH
were identified from the endocrinology clinic at our institution.
Patients who had previously received androgenic agents or gonadotropin were excluded. The diagnosis of IHH followed the
initial recognition of inappropriate sexual development, retardation of spontaneous puberty, or the lack of pubertal changes until
18 years old. The diagnosis was confirmed by determination of a
serum testosterone concentration below normal expected levels
as well as serum FSH and LH concentrations at or below normal
levels. Computed tomography and magnetic resonance imaging of
the brain were performed to evaluate for the presence of either
pituitary or hypothalamic masses. The final confirmation required a normal smell test, an XY karyotype determination, and
Akcam et al.: Voice Changes after Androgen Therapy
1587
the presence of gonadotropin response to repetitive doses of gonadotropin releasing hormone. The 24 subjects were identified
once these criteria were met. None had a positive family history
of IHH. The patients were classified as Tanner stages I and II
according to secondary sexual characteristics. The Tanner classification system is listed in Table I.
Thirty male and female adults who had completed normal
pubertal maturation (Tanner stage V) were also studied as control groups for comparison with the male IHH patients. Every
subject in the study was examined by an otolaryngologist to
exclude any obvious pathology that could affect the voice. This
included an otolaryngologic history and physical. Laryngoscopy
was performed in all patients. Once identified for the study, the
patients with IHH and the control groups underwent pretreatment voice recording and blood tests. Data were then gathered
during and after treatment as outline below. All institutional
guidelines for human subject data collection and study were
followed.
Hormone Studies
Serum levels of total testosterone (TT) and estradiol (E2)
were measured by immunochemoluminescence using a commercial kit (Bayer Corporation, Tarrytown, NY). Free testosterone
(FT) was measured by radioimmunoassay (Diagnostic Product
Corporation, Texas). Dehydroepiandrosterone sulfate (DHEAS)
and sex hormone binding globulin (SHBG) were measured by
immunochemoluminescence using a commercial kit, IMMULITE
1 (Diagnostic Product Corporation, Los Angeles, CA).
Hormone levels were first determined at baseline before the
initiation of intramuscular injections of Sustanon 250 mg (Organon), a commercially available preparation of testosterone propionate 30 mg, testosterone phenpropionate 60 mg, testosterone
isocaproate 60 mg, and testosterone decanoate 100 mg. After the
initial injection, Sustanon was administered to IHH patients
every 21 days for 3 months. The serum assays were repeated at 5,
10, and 90 days after the initiation of therapy. Hormone assessments at month 3 occurred 7 days after the last injection. All
blood samples were collected from fasting patients and controls
between 8:00 AM and 8:30 AM.
Voice Analysis
Acoustic parameters were determined by the Multi Dimensional Voice Program (model 5105, Ver. 2.3, Kay Elemetrics Corp,
Lincoln Park, NJ). The voice samples were recorded by using a
microphone (Shure, Dynamic cardioid microphone C606N, Niles,
IL) placed 10 cm away and 45° below from the mouth. A midvowel
segment on a sustained /a/, with a duration of 3 seconds at
habitual loudness and pitch, was used. MF0 for the voice samples
were determined. The voice analyses of the male IHH patients
were performed before the treatment and at the 5th and 10th
days and 3rd months of the therapy, concurring with the days
that the blood samples were taken.
Statistical Analysis
All the statistical analyses were performed by the SPSS 10.0
(Statistical Package for Social Sciences, SPSS Inc., Chicago, IL)
program. Descriptive statistics were shown as mean ⫾ SD. To
investigate the differences between measurements, the repeated
measurement procedure of the SPSS program was used; for the
data that were not normally distributed the Friedman test was
used. When considering the differences between two groups of
data, the Wilcoxon signed rank test with Bonferroni correction
was used. The male, female, and male IHH groups were compared by using the analysis of variance (ANOVA) module, with
Bonferroni used as a post hoc test. When parametric assumptions
were not appropriate, the Mann-Whitney U test with Bonferroni
correction was used. For the correlations among the consecutive
differences, the Pearson coefficient of correlation was used. A
priori, a P value less than .05 was determined to be statistically
significant.
The lower limits of detection for TT, estradiol (E2), FT,
DHEAS, and SHBG were 10 ng/dL, 10 pg/mL, 0,18 pg/mL, 2
␮g/dL, 0.22 nmol/L, 13 nmol/L, respectively. The intra-assay and
interassay coefficients of variation were 2.3% to 6.2% and 1.4% to
4.7% for TT, 2.3% to 3.0% and 1.5% to 2.9% for LH, 2.0% to 2.9%
and 0.3% to 2.7% for FSH, 4.0% to 12.1% and 4.5% to 8.1% for E2,
3.7% to 6.2% and 7.3% to 9.7% for FT, 7.6% to 9.5% and 8.1% to
15.0% for DHEAS, 4.1% to 7.7% and 5.8% to 13.0% for SH.
RESULTS
There were 24 patients in the IHH group (mean age
20.6 years, range 19 –22). The normal male group had 30
subjects (mean age 20.9 years, range 20 –23) as did the
normal female group (mean 21.1 years, range 20 –23). The
MF0 of groups is presented in Table II. The differences
between the groups are statistically significant. (F ⫽
91.402; P ⬍ .001). The mean fundamental frequency of the
IHH patients before treatment was intermediate between
the male and female groups, being significantly lower
than in the female subjects (P ⬍ .01) and higher than that
of the male subjects (P ⬍ .001).
The hormone profile of the male IHH patients
throughout the course of the study is summarized in Table
III. FT, TT, SHBG, and E2 levels all changed significantly
during the study (respectively, ␹2 ⫽ 57.301, P ⬍ .001; ␹2 ⫽
53.850, P ⬍ .001; ␹2 ⫽ 45.650, P ⬍ .001; ␹2 ⫽ 13.434, P ⬍
TABLE I.
Pubertal Development in Boys, (from Tanner JM. Growth at adolescence. Oxford: Blackwell Scientific, 1962).
Genital Development
Stage
I
II
III
IV
V
Pubic Hair Development
Characteristics
Prepubertal; testicular length ⬍2.5 cm
The testes are over 2.5 cm in the longest diameter,
and the scrotum is thinning and reddening
Growth of the penis occurs in width, and length and
further growth of the testes is noted
Penis is further enlarged, and testes are larger with
darker scrotal skin color
Genitalia are adult in size and shape
Laryngoscope 114: September 2004
1588
Stage
Characteristics
I
II
Prepubertal; no pubic hair
Sparse growth of slightly pigmented, slightly curved
pubic hair mainly at the base of the penis
Thicker, curlier hair spread to the mons pubis
III
IV
V
Adult-type hair that does not yet spread to the
medial thighs
Adult-type hair spread to the medial thighs
Akcam et al.: Voice Changes after Androgen Therapy
TABLE II.
Fundamental Frequencies of IHH Subjects Compared with Male
and Female Control Groups before Androgen Therapy.
Mean Fo ⫾
Standard Deviation
(in Hz)
Group
IHH males
Males
229.33 ⫾ 41.29
150.40 ⫾ 22.01
Females
256.17 ⫾ 29.01
P ⬍ .001 compared with
IHH males
P ⬍ .01 compared with
IHH males
Male versus female (P ⬍ .001) using two-tailed t-test).
IHH ⫽ isolated hypogonadotropic hypogonadism.
.01). The change seen in DHEAS levels did not reach
statistically significance (␹2 ⫽ 6.891, P ⫽ .075). FT and TT
showed a peak level at day 5, reaching a level that was
significantly higher than the baseline values (respectively,
z ⫽ 4.286, P ⬍ .001; z ⫽ 4.257, P ⬍ .001). At day 10, FT
levels decreased significantly (respectively, z ⫽ 4.257, P ⬍
.001; z ⫽ 4.086, P ⬍ .001), although they remained significantly higher than the baseline values (respectively, z ⫽
4.286, P ⬍ .001; z ⫽ 4.229, P ⬍ .001). The FT and TT levels
did not change significantly from day 10 posttreatment
levels through the third month (z ⫽ 2.457, P ⫽ .14; z ⫽
1.429, P ⫽ .153). They did, however, again, remain higher
than the baseline levels (z ⫽ 3.331, P ⫽ .01; z ⫽ 4.114,
P ⬍ .001). SHBG levels showed a significant decrease at
day 5 (z ⫽ 4.286, P ⬍ .001), day 10 (z ⫽ 4.286, P ⬍ .001),
and at the third month (z ⫽ 2.972, P ⫽ .003) when compared with baseline. E2 levels significantly increased at
day 5 (z ⫽ 2.776, P ⫽ .006) and 10 (z ⫽ 2.978, P ⫽ .003),
but the change at 3 months compared with baseline was
not statistically significant (z ⫽ 1.860, P ⫽ .063).
Voice Analysis Results after Treatment
The mean F0 of the male IHH subjects before the
treatment was 229.33 ⫾ 41.29 Hz. This decreased to
218.13 ⫾ 30.81 at day 5 (F ⫽ 6.713; P ⫽ .016). The mean
F0 returned to the previously high values at day 10, at
226.17 ⫾ 38.29 Hz, showing no significant difference with
baseline values (F ⫽ 0.579, P ⫽ .455). At the 3-month
analysis, however, mean F0 had decreased to 173.13 ⫾
29.76 Hz, a statistically significant change from baseline
values (F ⫽ 82.664, P ⬍ .001). The progression of MF0
results in the IHH group over time is displayed in Figure
1. When compared with the normal female group, the
mean F0 of male IHH patients after 3 months of treatment
was found to be significantly different (z ⫽ 6.093, P ⬍
.001). Although the mean F0 of the treated IHH patients
was still higher than those of the control males, the difference between mean F0 of normal male and IHH male
groups was no longer statistically significant (z ⫽ 1.723,
P ⫽ .085). These results are presented in bar graph format
in Figure 2. The mean differences of the F0 between baseline measurements and the values at the day 5, 10, and
the third month posttreatment are presented in Table IV.
When the correlation between consecutive differences of
MF0 and consecutive differences of hormonal parameters
was investigated, no significant correlation was detected.
DISCUSSION
Although habitual pitch and fundamental frequency
of may be an indicator of vocal and sexual maturity, the
genotype may not correspond to the perceived sex of the
voice. In this study, it was determined that untreated
male IHH patients have a mean F0 intermediate between
normal males and females. The difference between the
MF0 of normal male and IHH subjects objectively demonstrates the impact of endocrine abnormalities on voice
parameters. This is further supported by the subsequent
trend toward normalization the F0 of IHH after androgen
treatment as reported in this article.
Laryngeal maturation caused by androgen stimulation after puberty is a fundamental basis for the difference
between the male and female voice, although the direct
mechanisms for this are complex. Beckford et al.7 investigated the effect of androgen stimulation to the laryngeal
structures in a sheep model and reported that the thyroid
cartilage demonstrated a greater response to male sex
hormones than did the cricoid or arytenoid cartilages. In
addition, it was found that larynges exposed to higher
androgen levels tended to have a more acute thyroid angle. Specific anatomic measurements, such as distances
between key landmarks of the larynx, are affected by
androgen exposure. One example of this is the androgensensitive changes of the arytenoid cartilage. In Beckford’s
animal study, the distances between the vocal process and
the base of the arytenoid increases with androgen stimulation, as does the distance from vocal process to the
muscular process. The relative descent of the larynx, and
subsequent enlargement of the vocal tract and resonance
system with age and maturation, are additional factors
affecting voice parameters.8
TABLE III.
Serum Concentrations of Hormones in Male IHH Subjects before Treatment, and at the 5th day, 10th day, and 3rd month after the
Commencement of Therapy.
Baseline
5th Day
10th Day
3rd Month
50.26 ⫾ 32.33
3.02 ⫾ 3.01
188.20 ⫾ 89.89
21.78 ⫾ 9.42
45.42 ⫾ 21.43
1006.92 ⫾ 356.63
51.28 ⫾ 19.93
165.90 ⫾ 56.36
43.96 ⫾ 26.63
23.12 ⫾ 10.50
334.12 ⫾ 172.36
18.68 ⫾ 8.79
135.26 ⫾ 64.39
35.17 ⫾ 14.75
22.32 ⫾ 10.04
285.27 ⫾ 278.39
12.34 ⫾ 13.50
145.49 ⫾ 65.37
31.92 ⫾ 23.89
32.35 ⫾ 17.18
Hormone
Total testosterone (ng/dL)
Free testosterone (pg/mL)
DHEA-S (␮g/dL)
Estradiol (pg/mL)
SHBG (nmol/L)
DHEA-S ⫽ dehydroepiandrosterone; SHBG ⫽ sex hormone binding globulin; IHH ⫽ isolated hypogonadotropic hypogonadism.
Laryngoscope 114: September 2004
Akcam et al.: Voice Changes after Androgen Therapy
1589
Fig. 1. Course of mean vocal fundamental frequency (MF0) determinations over time for the isolated hypogonadotropic hypogonadism (IHH) group.
The presence of a difference between the F0 of male
IHH patients who have deficiency of testosterone and the
F0 of normal postpubertal female subjects may be explained by the development of some male genotypic characteristics despite the hormonal deficiency. Castrated individuals may have the external morphologic structure of
a male with a lung capacity of 3.5 to 5 L and the muscular
abdomen and pelvic girdle necessary to sustain a male
voice and a male-type bony architecture of resonating
chambers.2
Changes in fundamental frequency are associated
with androgen development and puberty characteristics.
MF0 is related to height, pubic hair stage (PH), testis
Fig. 2. Comparison of the mean F0 between control groups and the
pre- and posttreatment isolated hypogonadotropic hypogonadism(IHH) group.
Laryngoscope 114: September 2004
1590
volume, TT, and serum hormone binding globulin in boys
during puberty.9 Harries et al.8 reported that although
there was a correlation between testis volume and voice
parameters, there was no relation between salivary testosterone concentrations and voice parameters of males at
puberty. These investigators indicated that maximum vocal change occurred between stages G3 and G4 of the
Tanner Classification. Vuorenkoski et al.10 stated that the
decrease in speaking fundamental frequency (SFF) occurred gradually as correlated with PH, and, although the
mean lower frequency (LF) began to decrease earlier than
mean SFF, the greatest change occurred later in mean LF
(between PH 3 and 4) than in the mean SFF (between PH
2 and 3).10 Although assessment of a single measurement
of F0 to determine the developmental stage of a patient
has limited value, analysis of voice parameters over time,
however, may contribute to evaluation of progress of a
person’s development. When testosterone was administered to male IHH patients in this study, a significant
decrease in mean F0 was noted at the fifth day following
treatment, the same time that the highest levels of TT,
FT, and E2 were present. As the hormone levels decreased
at day 10, however, mean F0 nearly returned to pretreatment levels. We postulate that changes in MF0 occurring
shortly after androgen administration may have been
caused by the edema-forming effect of testosterone. The
interfibrillar space of vocal folds is filled with a colloidal
mass, consisting of polysaccharides and hyaluronic acid,
and is capable of retaining a great amount of water.5 Any
factor changing the water binding capacity will have an
effect on colloidal mass, potentially affecting F0. The transient decline at day 5 and subsequent increase at day 10 of
mean F0 is likely the result of from water shifts in the
interfibrillary space.
At the end of the third month of treatment, the MF0
of male IHH patients tended toward the MF0 of the normal males whose secondary sex characteristics were well
developed. It is known that testosterone replacement increases muscle mass in castrated animals.11 Bhasin et
al.12 also showed that supraphysiologic doses of testosterone in normal men caused significant increases in the
cross-sectional areas of the triceps and the quadriceps
within 10 weeks. Androgen treatment induces myogenesis
in the developing larynx of juvenile Xenopus laevis.13 In
an animal study with Xenopus laevis, it is shown that
dehydrotestosterone (DHT) treatment converts the laryngeal muscle to a fast-twitch fiber type and results in hypertrophy of laryngeal muscle fibers, and androgen is
more powerful determinant of muscle fiber size than innervation,14 with all of these investigations illustrating
the impact of androgens on the physiology of the vocal
apparatus.
Talaat et al.15 demonstrated in a murine model that
laryngeal muscle changes were noticeable after 1 month of
testosterone administration and that they became more
evident after the second and third months of treatment.
The changes were basically in the form of hypertrophy and
hyperplasia, affecting mainly the medial part of thyroarytenoid muscle. The hypertrophy of the muscle fibers was
irreversible at 2 months after discontinuation of the drug.
On the basis if these previous reports and the findings of
Akcam et al.: Voice Changes after Androgen Therapy
TABLE IV.
Mean Changes of MF0 in Consecutive Analysis in the IHH Group.
Mean Difference ⫾ SD
(numbers are in Hz)
Baseline–5th day
Baseline–10th day
Baseline–3rd month
5th day–10th day
10th day–3rd month
11.19 ⫾ 21.17
3.59 ⫾ 20.34
56.19 ⫾ 30.34
⫺8.04 ⫾ 19.63
53.03 ⫾ 30.12
BIBLIOGRAPHY
F ⫽ 6.713; P ⫽ .016
F ⫽ 0.579; P ⫽ .455
F ⫽ 82.664; P ⬍ .001
F ⫽ 0.497; P ⫽ .488
F ⫽ 103.290; P ⬍ .001
MF0 ⫽ mean vocal fundamental frequency; IHH ⫽ isolated hypogonadotropic hypogonadism.
this article, it is postulated that the mechanism of change
in mean F0 of the IHH patients after testosterone administration is caused by an increase in the mass of vocal
muscle.
CONCLUSION
To our knowledge, this is the first report to investigate voice parameters in a prospectively acquired series of
IHH patients. We have shown that untreated males with
IHH have a mean F0 that is intermediate between normal
males and females. After treatment with testosterone, the
MF0 changes to approximately that of normal males. Although the hormonal profile of this patient population
reflects their underlying condition, there was no persistent correlation detected between mean F0 and hormone
levels over the course of treatment. The utility of voice
parameters, such as the MF0, in the characterization of
normal development may be the subject of future investigations. All patients receiving androgenic agents should
be counseled regarding the likelihood for short- and longterm changes in vocal pitch.
Acknowledgments
The authors would like to thank Yavuz Sanisoglu,
PhD. from the Department of Biostatistics at Gulhane
Laryngoscope 114: September 2004
Military Medical Academy, Ankara, Turkey, for his assistance.
1. Tanner JM. Growth at Adolescence. Oxford: Blackwell Scientific, 1962.
2. Abitbol J, Abitbol P, Abitbol B. Sex hormones and the female
voice. J Voice 1999;13:424 – 446.
3. Baker J. A report on alterations to the speaking and singing
voices of four women following hormonal therapy with
virilizing agents. J Voice 1999;13:496 –507.
4. Boothroyd CV, Lepre F. Permanent voice change resulting
from Danazol Therapy. Aust NZ J Obstet Gynaecol 1990;
30:275–276.
5. Damste PH. Voice change in adult women caused by virilizing
agents. J Speech Hear Disord 1967;32:126 –132.
6. King A, Ashby J, Nelson C. Effects of testosterone replacement on a male professional singer. J Voice 2001;15:
553–557.
7. Beckford NS, Rood SR, Schaid D, Schanbacher B. Androgen
stimulation and laryngeal development. Ann Otol Rhinol
Laryngol 1985;94(6 Pt 1):634 – 640.
8. Harries ML, Walker JM, Williams DM, et al. Changes in the
male voice at puberty. Arch Dis Child 1997;77:445– 447.
9. Pedersen MF, Moller S, Krabbe S, Bennett P. Fundamental
voice frequency measured by electroglottography during
continuous speech. A new exact secondary sex characteristic in boys in puberty. Int J Pediatr Otorhinolaryngol
1986;11:21–10.
10. Vuorenkoski V, Lenko HL, Tjernlund P, et al. Fundamental
voice frequency during normal and abnormal growth and
after androgen treatment. Arch Dis Child 1978;53:
201–209.
11. Griggs RC, Kingston W, Jozefowicz RF, et al. Effect of testosterone on muscle mass and muscle protein synthesis.
J Appl Physiol 1989;66:498 –503.
12. Bhasin S, Storer TW, Berman N, et al. The effects of supraphysiologic doses of testosterone on muscle size and
strength in normal men. N Engl J Med 1996;335:1–7.
13. Sassoon D, Segil N, Kelley D. Androgen-induced myogenesis
and chondrogenesis in the larynx of xenopus laevis. Dev
Biol 1986;113:135–140.
14. Tobias M, Marin M, Kelley DB. The roles of sex, innervation
and androgen in laryngeal muscle of Xenopus laevis.
J Neurosci 1993;13:324 –333.
15. Talaat M, Angelo A, Talaat AM, et al. Histological and histochemical study of effects of anabolic steroids on the female larynx. Ann Otol Rhinol Laryngol 1987;96:468 – 447.
Akcam et al.: Voice Changes after Androgen Therapy
1591