[Downloaded free from http://www.cancerjournal.net on Wednesday, April 12, 2017, IP: 14.139.190.14]
Original Article
Acoustic analysis of voice in nonlaryngeal
head and neck cancer patients
post chemoradiotherapy
ABSTRACT
Background: Concurrent chemoradiotherapy (CCRT) used for definitive management of locally advanced head and neck squamous
cell carcinoma (HNSCC) allows organ preservation at the cost of preservation of function. Vocal cords, being within the field of
irradiation, undergo acute and chronic changes which adversely impacts the patients’ voice.
Aims: To assess the acute changes in the acoustic characteristics of voice post‑CCRT in patients with nonlaryngeal HNSCC.
Materials and Methods: Thirty patients with HNSCC treated with CCRT, a total dose of 66–70 Gy/33–35 fractions at five
fractions/week, with weekly cisplatin. Acoustic analysis (AA) and laryngoscopic examination performed at baseline, 6 weeks, and
3 months post‑CCRT. Statistical analysis of the parameters using ANOVA and Student’s t‑test was performed.
Results: Of the thirty patients, 26 patients completed CCRT. At 6 weeks post‑CCRT, among 14/26 patients, most (11/14 [78.57%])
developed Grade III toxicity. On AA, both increase and decrease in mean F0 from baseline was observed. An increase (P < 0.05)
in each, i.e., jitter, shimmer, and noise to harmonics ratio (NHR) were recorded. At 3 months post‑CCRT, among 8/14 available,
most (6/8 [75%]) showed Grade II toxicity. The mean F0 reduced for both genders; jitter and shimmer, and NHR values maintained
an increase (P > 0.05).
Conclusions: Periodic AA allows quantification of voice changes and mapping of vocal toxicity induced by CCRT.
KEY WORDS: Acoustic analysis, chemoradiation, head and neck cancer
INTRODUCTION
Concurrent chemoradiotherapy (CCRT), the current
standard of care for the definitive management
of locally advanced head and neck squamous cell
carcinoma (HNSCC), is worthly called an organ
preservation technique.[1,2] However, the normal
tissue toxicity compromises the functional outcome
of this treatment. The larynx is one such organ,
whose function is affected due to its integral
location within the field of radiation. Wide field
head and neck irradiation adversely affects the
voice, even in the absence of malignant laryngeal
pathology.
Acute laryngeal toxicity induced by radiotherapy
manifests as hoarseness of voice, pain in the throat
and whispered speech with marked edema of the
vocal cords and arytenoids, along with fibrinous
exudates.[3-6] Apart from the dose received by the
larynx, the xerostomia and its associated reduction
in pharyngeal lubrication also contribute to the vocal
dysfunction.[7] Thus, radiation‑induced dysphonia in
the form of hoarseness, breathiness, voice breaks,
repeated cough during speech, and whispered
speech, impairs the patient’s ability to communicate
effectively, in turn, affecting the posttreatment
quality of life (QoL).[8] The objective method of
analyzing the voice changes has been through
acoustic analysis (AA) and electoglottography.[9,10]
The current prospective study was undertaken to
assess the effect of CCRT on voice characteristics
through AA. It allows quantification of the
CCRT‑induced changes in the acoustic characteristics
of voice, which enable comparison of the laryngeal
outcome of various chemo-radiation fractionation
schedules.
This is an open access article distributed under the terms of the Creative Commons
Attribution‑NonCommercial‑ShareAlike 3.0 License, which allows others to remix,
tweak, and build upon the work non‑commercially, as long as the author is credited
and the new creations are licensed under the identical terms.
Nikhila
Radhakrishna,
B. K. Yamini1,
Amrut Sadashiv
Kadam2,
N. Shivashankar1,
Chendil
Vishwanathan2,
Rajesh Javarappa2
Department of
Radiotherapy,
Jawaharlal Institute
of Postgraduate
Medical Education and
Research, Puducherry,
1
Department of
Speech Pathology
and Audiology,
National Institute
of Mental Health
and Neurosciences,
2
Department of
Radiotherapy,
Bangalore Medical
College and Research
Institute, Bengaluru,
Karnataka, India
For correspondence:
Dr. B. K. Yamini,
Department of
Speech Pathology
and Audiology,
National Institute
of Mental Health
and Neurosciences,
Hosur Road,
Bengaluru ‑ 560 029,
Karnataka, India.
E‑mail: yaminihk@
gmail.com
Access this article online
Website: www.cancerjournal.net
DOI: ***
PMID: ***
Quick Response Code:
For reprints contact: reprints@medknow.com
Cite this article as: ???
© 2017 Journal of Cancer Research and Therapeutics | Published by Wolters Kluwer - Medknow
1
[Downloaded free from http://www.cancerjournal.net on Wednesday, April 12, 2017, IP: 14.139.190.14]
Radhakrishna, et al.: Acoustic analysis in post CRT HNC
Objective
To assess the acute changes in the acoustic characteristics of
voice post‑CCRT in patients with nonlaryngeal HNSCC.
MATERIALS AND METHODS
Thirty patients with histologically proven squamous
cell carcinoma of the head and neck with informed
consent were enrolled. After ethical clearance, the study
was conducted at the Department of Radiotherapy of
a Government Medical College. They included patients
aged between 18 and 70 years of both genders, with a
Karnofsky Performance Score >70, who were planned to
receive radical intent of treatment with a conventional
fractionation schedule. Patients with primary laryngeal
invasion, laryngeal cancers, comorbidities (diabetes mellitus,
neurological ailments, etc.), postoperative cases, professional
users of voice, and patients with preexisting vocal
pathology were excluded from the study. Staging workup
including confirmation of noninvolvement of laryngeal
apparatus was performed by indirect laryngoscopy (IDL),
nasopharyngolaryngoscopy (NPL), and radiological imaging.
Treatment protocol
Radical treatment to a total dose of 66–70 Gy in 33–35 fractions
at 2 Gy/fraction/day was delivered along with weekly cisplatin.
Posterior electron boost added where indicated.
Acoustic analysis
AA was performed at baseline, 6 weeks posttreatment, and
3 months posttreatment. The voice signal was recorded digitally
using Computerized Speech Labs model 4500 (Kay Labs) on the
Multidimensional Voice Protocol advanced version through
a microphone (~12 cm mouth – microphone distance) with
a sampling frequency of 50,000 Hz and resolution of 16 bits
per sample. The patient was seated comfortably in a quiet
room. Sustained phonation of vowel “|a|,” voiced thrice, at
comfortable pitch and loudness was recorded with a pause
of 60 s after each trial. Good samples of verified signals were
saved for further analysis. The parameters of voice analyzed
were fundamental frequency (F0), jitter, shimmer, and noise
to harmonics ratio (NHR).
AA evaluates the signature characteristics of voice, namely
the fundamental frequency (F0), jitter, shimmer, and NHR.[9]
Laryngeal toxicities of CCRT in patients with nonlaryngeal
HNSCC is evaluated by AA.[6,7,11] These studies have reported
a significant worsening of acoustic parameters such as jitter,
relative amplitude perturbation, amplitude perturbation
quotient, pitch amplitude, spectral flatness ratio, and
phonation frequency range. Similarly, aerodynamic measures
such as mean phonation time, mean airflow, and vocal fold
diadochokinetic rate were reported to have decreased among
the nonlaryngeal group. Perceptual analysis of voice using
the grade, roughness, breathiness, asthenia, and strain
has also demonstrated a significant worsening of voice
2
quality in the early postirradiation period in patients.[6,11]
Maximum phonation time and words per minute on AA and
jitter on electroglottography have all shown a significant
difference.[10] Such alterations in the voice have been found to
have a profound impact on the voice‑related QoL.[9,11]
Statistical analysis
Pre‑ and post‑treatment observations were analyzed
using Student’s t‑test and ANOVA. Results on continuous
measurements were studied by mean ± standard deviation.
Significance was assessed at 5%.
RESULTS
Flowchart of study patients is tabulated in Figure 1. The mean
age of the cohort was 55 years. Twenty‑three percent of the
patients presented with T2 tumors, 40% with T3, and 37%
with T4 tumors; primaries of the oral cavity and oropharynx
being the most common.
Although all 30 patients had baseline AA, only 14 were
available at 6 weeks post‑CCRT and 8 patients at 3 months
post‑CCRT. Given this and to keep the homogeneity of patient
numbers for comparisons, results were analyzed for different
time periods separately, that is, for 14 patients from baseline
to 6 weeks [Table 1] and for 8 patients from baseline to
3 months [Table 2].
Baseline to 6 weeks post‑CCRT: IDL examination of 14 patients
revealed Grade II laryngeal toxicity in 3 patients and
Grade III laryngeal toxicity in 11 patients [Figure 2]. AA
demonstrated [Table 1] a decrease in the F0 from baseline in
ten patients (five males and five females) and an increase in the
F0 in four patients (two males and two females). There was a
statistically significant increase in the jitter % (P = 0.03) and
shimmer (in dB) (P = 0.009) and NHR (P = 0.02) in comparison
with the baseline [Figure 3].
Baseline to 3 months post‑CCRT: Ten patients were available
out of which two patients underwent emergency tracheostomy
because of Grade IV laryngeal toxicity. Endoscopic evaluation
of eight patients revealed residual Grade I toxicity in two
patients and Grade II toxicity in six patients [Figure 2]. AA
showed a sustained increase in jitter, shimmer, and NHR
compared to baseline (P > 0.05) [Figure 4]. The F0 values
were lower than the mean baseline F0 for both genders, but
not significant.
DISCUSSION
Our study aimed at understanding the effect of CCRT on the
various acoustic parameters of voice. Kazi et al.[10] described
the tumor‑induced vocal cord edema and obstruction of
airflow through the glottis that causes distortion of voice.
Meleca et al.[12] have studied advanced laryngeal squamous
cell carcinoma treated nonsurgically and described the
Journal of Cancer Research and Therapeutics - Month 2017 - Volume XX - Issue XX
[Downloaded free from http://www.cancerjournal.net on Wednesday, April 12, 2017, IP: 14.139.190.14]
Radhakrishna, et al.: Acoustic analysis in post CRT HNC
tumor‑induced neuromuscular weakness of the cords, which
differs from the radiation induced mucositis and fibrosis of
the laryngeal tissues. Thus, it was considered essential to
rule out primary tumor affecting the laryngeal structures
by performing IDL, NPL, and radiographic examinations at
baseline.
At 6 weeks post‑CCRT, IDL and NPL evaluation revealed marked
edema of vocal cords, arytenoids, and false cords. Of the
14 patients available for analysis patients who had received
a total dose of more >66 Gy had Grade III laryngeal toxicity.
These findings support the study by Dornfeld et al.[13] who had
shown a steep drop in the laryngeal function above a dose of
66 Gy delivered by intensity modulated radiotherapy (IMRT).
Sanguineti et al.[14] recommend restricting the mean dose to
larynx to <43.5 Gy at 2 Gy/fraction to minimize the risk of
edema of Grade II and above. On univariate analysis, their
study suggests the association of laryngeal edema with neck
stage, nodal diameter, mean laryngeal dose, and laryngeal
V30–V70 Gy. The patient reported voice outcomes used by
Rinkel et al.[15] had also demonstrated deviant speech handicap
index scores in 55% patients treated by IMRT/3‑dimensional
conformal radiation therapy. Current practices of larynx
sparing techniques of IMRT by whole neck field IMRT or
junctioned IMRT have allowed reduction of mean laryngeal
doses between 29.1 and 38.8 Gy based on proximity to the
Initial patient recruitment
(n = 30)
No. of patients completed CCRT
(n = 26)
No. of patients lost during CCRT
(n = 4)
Patients defaulted (n = 2)
No. of patients evaluated at
6 weeks( n = 14)
No. of
patients at 3
months
(n = 10)
Patients died (n = 2)
No. of patients lost (n = 12)
6 patients died within 4
weeks of completing CCRT
No. of patients lost 3
months post CCRT (n = 4)
1- Died (poor nutrition)
1- Salvage surgery
1- Progressive disease
1- Lost to follow up
2 patients developed
progressive disease
4 patients lost to follow up
Figure 1: Flowchart of the study patients
12
Baseline
grade 0
grade I
grade II
grade III
grade IV
10
8
2.5
2
1.5
6
1
4
0.5
2
0
6 weeks post CCRT
0
6 weeks post CRT
Mean Jitter (%)
3 months post CRT
Figure 2: Comparison of laryngeal toxicity grades post CCRT
Mean Shimmer (dB)
Mean NHR
Figure 3: Acoustic analysis from baseline to 6 weeks
Journal of Cancer Research and Therapeutics - Month 2017 - Volume XX - Issue XX
3
[Downloaded free from http://www.cancerjournal.net on Wednesday, April 12, 2017, IP: 14.139.190.14]
Radhakrishna, et al.: Acoustic analysis in post CRT HNC
Table 1: Acoustic analysis from baseline to 6 weeks
Timeline
Mean F0 males (Hz)
n=7
Mean F0 females (Hz)
n=7
Mean Jitter(%) n=14
122.7 (SD=+15)
128.4 (SD=+27)
210.05 (SD=+ 23)
219.64 (SD=+23)
0.71 (SD=+0.70)
2.2.12 (SD=+2.20)
0.31 (SD=+0.10)
0.73 (SD=+0.55)
0.63
0.45
0.030
0.009
Baseline
6 weeks post CCRT
t test
P value
Mean Shimmer (dB) n=14 Mean NHR n=14
0.13 (SD=+0.02)
0.20 (SD=+0.11)
0.027
Table 2: Acoustic analysis from baseline to 3months
Timeline
Baseline
6 weeks post CCRT
3 months post CCRT
P value
1.8
Baseline
Mean F0 males (Hz)
n=4
Mean F0 females (Hz)
n=4
Mean Jitter(%) n=8
Mean Shimmer (dB)
n=8
Mean NHR n=8
115.8 (SD=+14)
134.4 (SD=+33)
108.4 (SD=+15)
P>0.05
212.3 (SD=+23)
213.4 (SD=+18)
197.4 (SD=+9)
P>0.05
0.7343 (SD=+0.4)
1.668 (SD=+1.1)
1.13 (SD=+0.68)
P>0.05
0.29 (SD=+0.06)
0.51 (SD=+0.26)
0.44 (SD=+0.25 )
P>0.05
0.13 (SD=+0.01)
0.16 (SD=+0.03)
0.15 (SD=+0.04)
P>0.05
6 weeks post CCRT
amplitude of its vibration. Thus, CCRT‑induced edema alters
the vibrational characteristics of the cords.
3 months post CCRT
1.6
1.4
1.2
1
0.8
0.6
0.4
0.2
0
Mean Jitter (%)
Mean Shimmer (dB)
Mean NHR
Figure 4: Acoustic analysis from baseline to 3 months
larynx.[16] Laryngeal sparing was not feasible in patients with
locally advanced primaries with nodal disease by conventional
two‑dimensional plans.
Evaluation of the vocal cords at the end of 3 months post‑CCRT
showed varying degrees of resolution of the vocal cord
edema and erythema. Secretions were much lesser with the
progression of time from treatment. These are attributed to
the xerostomia, fibrosis, dehydration, and dryness of laryngeal
mucosa.[4,17]
AA has been used for objective assessment of voice by Hamdan
et al.,[6] Paleri et al.,[11] and Fung et al.[7] While most studies have
used the vowel |i| for analysis,[11] we have used sustained
phonation of the vowel |a| for AA. AA performed at 6 weeks
posttreatment revealed an insignificant decrease in the F0
in ten patients due to the increase in the effective mass of
the cords due to edema in agreement with the findings of
other studies[6,7] which reported no significant change in F0.
We also recorded an increase in the F0 in four patients (two
males and two females) which may be attributable to the
altered vibrational length of the cord due to the presence of
mucositis‑induced pseudomembrane. The jitter and shimmer
values showed a significant increase from the baseline values.
These reflect the alterations in the tissue of the vocal cords,
which cause an increase in the perturbations of frequency and
4
At 3 months post‑CCRT, two out of the four patients who
had shown an increase in the F0 values maintained the rise
in comparison to the baseline. The remaining evaluated
patients sustained a decrease in the F0 values as compared
to the baseline. The jitter, shimmer, and NHR values showed
an insignificant reduction in comparison to the values at
6 weeks although they did not return to baseline. However,
Paleri et al.[11] reported a significant reduction in the F0 and a
significant increase in both jitter and shimmer at 3 months
post‑CCRT; the same parameters evaluated at 1‑year post‑CCRT
showed that the changes had reversed in direction although
not to the baseline values. The present study reflects that
the acute laryngeal toxicity induced by CCRT begin to settle
around 3 months posttreatment. Long‑term follow‑up with a
larger number of participants will be required to validate the
same, as well as to identify the right time to initiate voice
conservation in these patients.
CONCLUSIONS
Patients with nonlaryngeal HNSCC treated with CCRT
experience significant alterations in the voice which can be
objectively mapped using AA. However, larger sample size and
long‑term follow‑up would be required to assess the long‑term
vocal toxicity, its impact on posttreatment voice‑related QoL
and measures necessary to reduce this toxicity.
Acknowledgments
We would like to acknowledge Dr. Iqbal Ahmed, Professor and
Head, Department of Radiotherapy, Bangalore Medical College
and Research Institute, Bengaluru, Karnataka, India.
Financial support and sponsorship
Nil.
Conflicts of interest
There are no conflicts of interest.
Journal of Cancer Research and Therapeutics - Month 2017 - Volume XX - Issue XX
[Downloaded free from http://www.cancerjournal.net on Wednesday, April 12, 2017, IP: 14.139.190.14]
Radhakrishna, et al.: Acoustic analysis in post CRT HNC
REFERENCES
1.
Kulkarni MR. Head and neck cancer burden in India. Int J Head Neck
Surg 2013;4:29‑35.
2. Pignon JP, le Maître A, Maillard E, Bourhis J; MACH‑NC Collaborative
Group. Meta‑analysis of chemotherapy in head and neck cancer
(MACH‑NC): An update on 93 randomised trials and 17,346 patients.
Radiother Oncol 2009;92:4‑14.
3. Sato K, Nakashima T. Effect of irradiation on the human laryngeal
glands. Ann Otol Rhinol Laryngol 2008;117:734‑9.
4. Berg EE, Kolachala V, Branski RC, Muller S, Johns MM. Pathologic
effects of external‑beam irradiation on human vocal folds. Ann Otol
Rhinol Laryngol 2011;120:748‑54.
5. Rancati T, Schwarz M, Allen AM, Feng F, Popovtzer A, Mittal B, et al.
Radiation dose‑volume effects in the larynx and pharynx. Int J Radiat
Oncol Biol Phys 2010;76 3 Suppl: S64‑9.
6. Hamdan AL, Geara F, Rameh C, Husseini ST, Eid T, Fuleihan N.
Vocal changes following radiotherapy to the head and neck for
non‑laryngeal tumors. Eur Arch Otorhinolaryngol 2009;266:1435‑9.
7. Fung K, Yoo J, Leeper HA, Hawkins S, Heeneman H, Doyle PC, et al.
Vocal function following radiation for non‑laryngeal versus laryngeal
tumors of the head and neck. Laryngoscope 2001;111 (11 Pt 1):1920‑4.
8. Jones SM, Carding PN, Drinnan MJ. Exploring the relationship
between severity of dysphonia and voice‑related quality of life. Clin
Otolaryngol 2006;31:411‑7.
9. Awan SN. The Voice Diagnostic Protocol: A Practical Guide to the
Diagnosis of Voice Disorders. Maryland: Aspen Publication; 2001.
10. Kazi R, Venkitaraman R, Johnson C, Prasad V, Clarke P, Rhys‑Evans P,
11.
12.
13.
14.
15.
16.
17.
Journal of Cancer Research and Therapeutics - Month 2017 - Volume XX - Issue XX
et al. Electroglottographic comparison of voice outcomes in
patients with advanced laryngopharyngeal cancer treated by
chemoradiotherapy or total laryngectomy. Int J Radiat Oncol Biol
Phys 2008;70:344‑52.
Paleri V, Carding P, Chatterjee S, Kelly C, Wilson JA, Welch A, et al.
Voice outcomes after concurrent chemoradiotherapy for advanced
nonlaryngeal head and neck cancer: A prospective study. Head Neck
2012;34:1747‑52.
Meleca RJ, Dworkin JP, Kewson DT, Stachler RJ, Hill SL. Functional
outcomes following nonsurgical treatment for advanced‑stage
laryngeal carcinoma. Laryngoscope 2003;113:720‑8.
Dornfeld K, Simmons JR, Karnell L, Karnell M, Funk G, Yao M, et al.
Radiation doses to structures within and adjacent to the larynx are
correlated with long‑term diet‑ and speech‑related quality of life. Int
J Radiat Oncol Biol Phys 2007;68:750‑7.
Sanguineti G, Adapala P, Endres EJ, Brack C, Fiorino C, Sormani MP,
et al. Dosimetric predictors of laryngeal edema. Int J Radiat Oncol
Biol Phys 2007;68:741‑9.
Rinkel RN, Verdonck‑de Leeuw IM, Doornaert P, Buter J, de Bree R,
Langendijk JA, et al. Prevalence of swallowing and speech problems
in daily life after chemoradiation for head and neck cancer based on
cut‑off scores of the patient‑reported outcome measures SWAL‑QOL
and SHI. Eur Arch Otorhinolaryngol 2016;273:1849‑55.
Webster GJ, Rowbottom CG, Ho KF, Slevin NJ, Mackay RI. Evaluation
of larynx‑sparing techniques with IMRT when treating the head and
neck. Int J Radiat Oncol Biol Phys 2008;72:617‑22.
Lazarus CL. Effects of chemoradiotherapy on voice and swallowing.
Curr Opin Otolaryngol Head Neck Surg 2009;17:172‑8.
5