Acoustic Analysis of Voice in Laryngopharyngeal Cancers Pre and Post Radiotherapy

Abstract

Laryngopharyngeal cancers are one of the most commonly diagnosed head and neck malignancies frequently presenting primarily with change in voice. Radiotherapy being the main modality of treatment for early cancers continues to affect voice. Hence, acoustic analysis of the voice offers quantifiable values of several parameters delineating the obvious effect of the therapy. A total number of 60 patients, diagnosed with laryngopharyngeal cancers undergoing radiotherapy underwent acoustic voice assessment using Dr. speech software pre-treatment and at 1 and 3 months post radiotherapy. Data analysis was done using Mann–Whitney test and Wilcoxon signed rank test and a significant p value was obtained. The results of the study showed fundamental frequency (F0) and noise to harmonic ratio (NHR) to be the most affected in comparison to Jitter and Shimmer. The F0 and NHR values across baseline evaluation, first month and third month follow up showed a steady deterioration which was significant. The deterioration noted from the first to third month was not statistically significant. Across genders both F0 and NHR deterioration is more in males than in females. Radiotherapy causes definitive alterations in some acoustic measures of voice, which make the voice disharmonic and hoarse with contribution of harshness and breathiness. The effect is more pronounced on vocal parameters that are structure and projection based as evidenced by deterioration in values noted in F0 and NHR. Persistent deteriorated acoustic parameters for a longer duration of time are more likely which emphasizes the need for early voice rehabilitation.

Introduction

Laryngopharyngeal cancers are one of the most commonly diagnosed head and neck malignancies (30% of all CA in India) [1]. The most common symptoms being change in voice—hoarseness, dysphonia. Radiotherapy is the main modality of treatment for early cancers and it is used as an adjuvant therapy or the primary treatment by patient’s choice in advanced cancers and it continues to affect voice [2–4]. This definitely calls for an intensive analysis on the most intriguing subject of change in voice pre and post treatment by this modality.

Fatigue of the voice with much usage, inability to sing, reduced loudness, hoarse voice quality, and inability to shout in irradiated patients for laryngopharyngeal cancers or cancers where larynx was included in the field of radiation is commonly reported. The aim of this study is to analyse the change in voice by evaluating acoustic parameters in diagnosed cases of laryngopharyngeal cancers pre and post radiotherapy.

Materials and Methods

All patients with laryngeal and hypopharyngeal cancers diagnosed by examination and confirmed by histopathological report, visiting or admitted at Father Muller Medical College from September 2015 to July 2017, willing to undergo voice analysis and cooperate for follow ups were included in the study. 60 patients, undergoing radiotherapy or chemoradiotherapy for laryngopharyngeal cancers were selected by means of purposive sampling technique. With an appropriate informed consent taken to participate in the study, the patients were subjected to a detailed clinical examination, Direct Laryngoscopy and Biopsy for diagnostic purposes. Tracheostomized patients, recurrent disease, any surgical intervention other than a diagnostic biopsy and patients not willing for follow up were excluded.

A detailed history and a detailed throat examination was carried out. One day prior to radiotherapy the subject underwent acoustic analysis of voice. The following parameters were measured:

• Fundamental frequency (F0) Lowest frequency of a periodic waveform of the vocal cord

• Jitter (Jitt) Varying pitch in the voice which causes a rough sound

• Shimmer A frequent back and forth change in amplitude (from soft to louder) in the voice

• Noise to harmonics ratio (NHR)

  N = Mean energy of the differences between individual periods and average waveform

  H = Energy of the averaged waveform.

Collected data was analyzed using SPSS software version 23. Analysis was done using Mann–Whitney test and Wilcoxon signed rank test. A significant p value was obtained.

Results

The mean fundamental frequency (F0) obtained in the subjects during the baseline evaluation was noted to be higher. A drop in mean fundamental frequency was noted during both the first month and the third month post radiotherapy follow up respectively [Male = 100.81 (Pre), 82.3 (1 m), 80.52 (3 m); p = .000] ]Female = 198.98 (Pre), 165.48 (1 m), p = .007; 159.55 (3 m), p = .002]. The drop in fundamental frequency from the first month to the third month follow up was more evident in female subjects but was negligible in the male subjects who participated in the study (p = .000). (Table 1)

Table 1.

F0 variation across evaluation pre and post radiotherapy

Parameter	Sex	N	Mean	SD	Median (IQR)	Wilcoxon signed rank test p value
						At 1 MTH–base line	At 3 MTH–base line	At 3 MTH–At 1 MTH
F0	Male
	Base line	30	98.90	9.77	100.81 (92.91–107.06)	.000 HS	0.000 HS	.828 NS
	At 1 MTH	30	81.77	7.40	82.3 (76.5–87.95)
	At 3 MTH	30	82.19	11.45	80.52 (72.2–90.72)
	Female
	Base line	30	191.41	59.56	198.96 (159.91–239.09)	.007 HS	.002 HS	.026 Sig
	At 1 MTH	30	166.78	17.81	165.48 (150.42–179.24)
	At 3 MTH	30	158.03	17.57	159.55 (140.91–171.66)

HS highly significant, NS not significant, sig significant

In the case of noise to harmonic ratio (NHR), the mean NHR obtained in the subjects during the baseline evaluation was noted to be high but not close to norms. Fall in mean NHR was noted during both the first month and third month post radiotherapy [Male = 30 (Pre), 17 (1 m), 17 (3 m); p = .000] [Female = 30 (Pre), 16 (1 m), 13 (3 m), p = .000]. The drop in NHR from the first month to the third month follow up was more evident in female subjects but was negligible in the male subjects included in the study (p =.027 (1 m), .000 (3 m) respectively) (Fig. 1).

Fig. 1.

NHR variation across evaluation pre and post radiotherapy

The mean Shimmer value obtained in the subjects during the baseline evaluation was noted to be higher in case of females when compared to the norms, while in the case of male subjects the increase though present was not very deviant from the norm [Male = 2.76 (Pre), Female = 3 (Pre)]. A rise in mean Shimmer values was noted during both the first month and third month post radiotherapy [Male = 4.27 (1 m), 4.51 (3 m)], but the increased shimmer value noted in male subjects during the first month follow up post radiotherapy indicated an increase that was not significant (p = .072) (Fig. 2). The rise in Shimmer value from the first month to the third month follow up, though present was not evident enough to create significance. Shimmer variation across gender was found to be not significant [p =.579 (Pre), .387 (1 m), .767 (3 m)].

Fig. 2.

Shimmer variation across pre and post radiotherapy

The mean Jitter value obtained in the subjects during the baseline evaluation was noted to be lower in the case of male subjects than the female subjects [Male = 0.83 (Pre), 1.03 (1 m), 2.96 (3 m); Female = 1.18 (Pre), 2.95 (1 m), 2.96 (3 m)]. Female subjects showed an increased mean Jitter value that was of significance when compared to the norms (p =.000). A rise in mean Jitter values was seen during both the first month post radiotherapy follow up and the third month post radiotherapy follow up which was found to be significant [Males: p =.027 (1 m), .000 (3 m); Females: p = .000 (1 m), .000 (3 m)]. The rise in jitter value from the first month to the third month follow up, though present was not found to be significant (p = .975). Jitter variation across gender was found to be not significant [p = .448 (1 m), .181 (3 m)] (Fig. 3).

Fig. 3.

Jitter variation across evaluation pre and post radiotherapy

Discussion

Voice preservation is an equally important aspect of the management of laryngopharyngeal cancers, in special regard to an early stage cancer. Radiotherapy is the treatment modality of choice for the early cancers and has resulted in both excellent control rates and voice preservation [5]. Organ preservation protocols [6–8] may be preferred on a personal choice basis by the patient placing radiotherapy as the primary modality of treatment in advanced cancers though it originally plays an adjunctive role to surgery in treatment of such cancers. The importance of voice preservation goes beyond simply retaining the ability to vocalize.

Importance of voice assessment rests on the fact that voice is the mode of expression as needed in daily communication which is severely affected in these cancers. This in turn drastically affects the quality of life in such patients [9]. Quality of voice after radiotherapy has been poorly described in literature. Limited data exists on voice quality assessment pre and post radiotherapy in these cancers. The acoustic analysis of the voice offers quantifiable values of several parameters such as fundamental frequency, jitter, harmonics-to-noise ratio, and shimmer delineating the obvious effect of the therapy [10]. Our study revealed that certain voice parameters get altered after radiotherapy which falls in line with other similar studies [11, 12].

In our study subjects between the age of 40–70 years diagnosed with laryngopharyngeal carcinoma were enrolled as the population under study. Out of the 60 patients studied, 30 were female and 30 were male. Objective analysis was done for four parameters, fundamental frequency, jitter, shimmer and noise to harmonics ratio for all patients enrolled as a participant of this study. A baseline evaluation was done pre radiotherapy and two follow up evaluations at the first month and third month post radiotherapy was carried respectively. In our study both fundamental frequency (F0) values and noise to harmonic ratio values (NHR) noted across baseline evaluation, first month follow up and third month follow up showed a steady deterioration. The deterioration noted from the first to third month follow up though present was not statistically significant. Across genders both fundamental frequency and noise to harmonic ratio (vocal structure and projection based parameters) were the most affected parameters in comparison to Jitter and Shimmer.

It was also revealed that the deterioration in F0 and NHR were more notable in males than in females. F0 and NHR was also noted to be deteriorated more in cases with predominant laryngopharyngeal cancers and higher TNM scores than those with early laryngopharyngeal cancers and lower TNM scores.

Jitter and Shimmer (vibratory pattern based parameters) were noted to be more affected in individuals with predominant laryngopharyngeal cancers and higher TNM scores than the individuals with early laryngopharyngeal cancers and lower TNM scores. Increased Jitter and Shimmer values though present across genders were not evident enough to be considered of significance. This could be because the radiotherapy follow up duration being short it is not sufficient enough to cause major vibratory pattern alterations.

In tandem with several other studies, the fundamental frequency and noise to harmonic ratio in our study did show significant change by deteriorating post radiotherapy [13]. The preoperative fundamental frequencies and NHR values was affected when compared to the norm, but the significance of it was not evident to be considered as a notable finding. These results indicate that radiotherapy is a procedure that affect the larynx, especially in terms of musculature tone and voice projection. But it does not cause an alteration so severe that it hampers the vocal fold vibratory rate. This could be because of having subjects who were diagnosed with laryngopharyngeal carcinomas. Had the category been only laryngeal carcinomas then vibratory parameters such as jitter and Shimmer would have shown a significant increase post radiotherapy.

Recent studies have proved to have good results with organ preservation protocols that use radiochemotherapy [11, 14]. Although survival time is not much affected, laryngeal function is not always preserved, as stated by Hirsch et al. [15].

According to Leeper et al. [16] those authors concluded that, although subjective measurements of voice quality are important, objective measurements are necessary to assess subtle voice changes with time.

According to Behrman et al. [17], 80% of prospectively studied patients had changes in their voice quality 1 year after radiotherapy. 40% of their patients still complained of hoarseness, and 56% of throat clearing, even 3 years after treatment. Carrara-de Angelis et al. [18] concluded that of 15 patients who had received radiochemotherapy for laryngeal cancer, 33% had adequate voice quality or mild dysphonia, 40% moderate dysphonia, and 27% severe dysphonia.

Caminero et al. [19] reported that 11% had normal voices, 44% had mild dysphonia, 28% had moderate dysphonia, and 17% had severe dysphonia. Those authors also reported VHI results that were close to normal. Our results were slightly better, with the following characteristic voices: hoarse (76%), hoarse-breathing (12%), rough (20%), and crepitating (16%). On the GRBAS scale, 24% had normal voice, 40% mild dysphonia, and 36% moderate dysphonia. The patients recruited in our study with chronic voice change had elevated noise to harmonics ratios post radiotherapy, thus statistical analysis revealed a significance due to the variation noted.

The possible reason for no significant changes noted in Jitter and Shimmer could be because of the fact that the study population taken have been diagnosed with laryngopharyngeal cancers which do not mainly affect the vocal cords but are spread across the vocal tract. Hence radiotherapy will not be focused on the glottis region alone, therefore no evident findings have been noted in jitter and shimmer which are vibratory pattern based acoustic parameters. Hence in order to track vibratory pattern changes alone, laryngeal carcinomas have to be segregated and studied. To limit the factors under study TNM staging and its correlation to acoustic parameters was not extensively analyzed and also the subjective analysis of voice though recorded on GRBAS scale was not evaluated with respect to acoustic voice parameters in this study which could pave way for further research and insight in this field. Also advantage of early vocal rehabilitation [20, 21] can be dealt with in another study for real benefit assessment.

Conclusion

Radiotherapy causes definitive alterations in some acoustic measures of voice making it disharmonic and hoarse with an adequate contribution of harshness and breathiness perceived equally. The effect of RT is more pronounced on vocal parameters that are structure and projection based rather than on those that are vibratory pattern based as evidenced by deterioration in values noted in F0 and NHR from baseline. Likely persistent deteriorated acoustic parameters for longer time are attributable to post RT edema and fibrosis which emphasizes recognizable need for early voice rehabilitation.

Author Contributions

All contributors contributed significantly to the project.

Funding

This research was supported and funded by parent institution (Father Muller Charitable Institutions).

Availability of Data Material

All patients undergoing treatment in the Radiotherapy department of Father Muller Medical College within the time frame of the study and fitting the inclusion criteria were analysed.

Compliance with Ethical Standards

Code Availability

Not applicable.

Conflict of interest

There are no conflicts of interest.

Ethical Approval

This project was approved by the dedicated ethical committee of the parent institution.

Consent for Participate

Appropriate informed consent was taken from the participants in their native language (Kannada/Malayalam/Hindi) full disclosure of details of study.

Consent for Publication

All contributors consent to publish.

Informed Consent

Appropriate informed consent was taken in the native language of the patient to participate in the study and no animal experiments were performed.