INTRODUCTIONSupracricoid Partial Laryngectomy (SCPL) was first presented in Vienna, in 1959, by Meyer and Rieder (1). This surgery consists of the removing of the entire supraglottis, the false and true vocal folds and the thyroid cartilage including the paraglottic and pre-epiglottic spaces. The cricoid cartilage, hyoid bone, and at least one aritenoid are preseved. Phonatory and swallowing functions are maintained by the movement of the spared aritenoid to the tongue base. There are two types of reconstruction techniques: cricohyoidopexy (SCPL-CHP) and cricohyoidoepiglottopexy (SCPL-CHEP). The Near-Total Laryngectomy (NTL) was first used by Pearson in 1981 (2). The resection of the laryngeal structures in NTL, corresponds to those in Total Laryngectomy (TL), but preserves a small part of cricoid cartilage and whole (2/ 3) of the vocal folds (with respective innervation), preserving the intelligibility of the voice in patients. The advantages of both techniques is the high oncologic control (higher than 80%) with the maintenance of the laryngeal functions in 95 % (3). Other studies indicate acoustic, temporal and perceptive results away from the standard of normality. However, both surgeries showed good results in this study, specially for the speech intelligibility (4, 5, 6, 3, 7). It couldnt be found in literature a temporal acoustic and perceptive comparison between NTL and SCPL. It was only found between NTL and TL (Total Laryngectomy) or a comparison with hemilanryngectomy (8). In the presence of this data the aim of this study was to compare acoustic analysis of voice production in patients after NTL and SCPL.
PATIENTS AND METHODS Twenty-eight patients were prospectively reviewed to assess voice parameters. There were 25 men of about average of 60 years old (57.8 ± 10.8 years). Of all, 16 underwent a SCPL with CHEP and 9 underwent a NTL. All of them were operated in the Department Phonoaudiology of the Instituto do Cancer Arnaldo Vieira de Carvalho, Sao Paulo - Brazil (Table 1). All the patients were undergone a speech treatment in a minimum period of one year. The other treatments like radiotherapy and chemiotherapy couldnt be considered as criteria for inclusion or exclusion, as well as the frequency during the speech treatment. In this project, patients with previously neurologic alterations or fluency alterations were excluded.
The Institutional Ethics Committee gave scientific and ethical approval without restrictions (073/04) and the subjects written informed that consent was obtained, after explanation of the general nature of this study.
The patients were submitted to a questionary that included: patients identification, type of surgery (Near-Total, Supracricoid with CHP or CHEP), clinical staging of laryngeal cancer and other complementary treatments (radiotherapy and / or chemiotherapy post-surgery and speech therapy). All patientes were submitted to a perceptive temporal and acoustic assessment.
Voice Perception Analysis The patients voices were recorded in a special silent room with the MDVP (model CSL 4300B from Kay Elemetrics). Each patient stood 15 cm distant from the microphone (Shure SM 48). It was used 10-s of spontaneous conversation based on Boston´s test figure (9) for vocal quality and speech intelligibility judgement.
The samples from all patients recordings were assessed by three speech-language pathologists experts in the care of patients with head and neck cancer. The voices recordings were assessed according to G (global); R (roughness); B (breath voice); A (asthenia); and S (strain), GRBAS scale criteria (10). For each criterium and each sample, valuers should judge with a score ranging from 0 (normal) to 3 (severe).
For the analysis of speech intelligibility it was used the Visual Analogic Scale, that presents a scale of punctuation (ranging from 0 to 10), where high values indicates great speech intelligibility, and lower values indicate a deficient intelligibility. The voice samples from each patient were presented at random in a way that the listeners were unaware of the type of surgery that the patients were undergone.
The voice material consisted of pieces of approximately three seconds of the most stable portion of the sustained vowel /a/.
Acoustic and Speech Aerodynamic Testing Acoustic analysis was carried out using the vowel /a/. The computerized program Multi Dimension Voice (MDVP) of Kay Elemetrics (model 4300B) was used for acoustic analysis. The following measures were obtained from each patient:
1) Fundamental Frequency;
2) Jitter: Jitter Percentage (JITT), Frequency Period Perturbation Quocient (PPQ);
3) Shimmer: Shimmer Percentage (Shim), Amplitude Perturbation Quocient (APQ), and
4) Noise-to-Harmonic Ratio (NHR).
The Maximum Phonation Time (MPT) was extrated from the sustained vowel /a/, in just one expiration, during a three time measure, using the higher value.
Student t and Mann-Whitneys tests were applied to compare averages when appropriate. The agreement among observers was evaluated by means of Kappa coeficient. In order to verify the association among categorical variables were applied the chi-square and likelihood ratio tests. P-values < 0.05 were considered meaningful.
RESULTS In this report, both groups showed almost same average age, NTL = 59.2 ± 10.4 years old and SCPL = 57.0 ± 11.2 years old (p= 0.631 for the student test).
Concordance Values In the analysis concordance among the three speech language pathologists for the inqueries: global, roughness, breath voice, asthenia, strain, instability, pitch, loudness and intelligibility showed meaninful concordance between valuers 2 and 3 for the inqueries: pitch, loudness and intelligibility. Between valuers 1 and 3 it was observed agreement for strain and global. It couldnt have any concordance for asthenia, instability, roughness, and breath voice among the valuers (Table 2).
Perceptual Analysis Voice Perceptive Analysis X Surgery
The NTL group of patients presented worse general degree of speech quality than SCPL group of patients. This comparison was not statistically meaningful (p=0.754). according to the general degree of vocal quality the LNT group presented worse results compared to the LCPL group.
According to the intelligibility, valuers 2 and 3 agreed the most in NTL group. The average was 5, and in SCPL it was 6, in such case, the values were statistically meaningful (p=0.021). Due to this results the SCPL group presented greater intelligibility than the NTL group (Picture 1).
Picture 1. Squematic representation (Box-plot) for intelligibility results in both surgeries.
There were predominance of low pitch for SCPL (56.3%) and high pitch for NTL (66.7%) (p= 0.207). For loudness query, the NTL (77.8%) presented adequated loudness values higher than SCPL (62.5 %) (p=0.266).
The NTL presented degrees 0, 1, 2 and 3 for strain compared to SCPL that presented degrees 0, 1 and 2 for strain. Both surgeries presented higher values for degree 1 (p=0.012) (Picture 2).
Picture 2. Strain X Surgery.
In acoustic analysis the average fundamental frequency found for vowel /a/ was 120 Hz for NTL and 141.6 Hz for SCPL (p= 0.343). Referring to the results of shimmer, jitter, APQ, PPQ and NHR, in NTL group. It was obtained higher values for LNT than for LHSC. This comparison was not statisticaly meaningful (Table 3).
The MPT (maximum phonation time) values were 9.6 sec for NTL and 11.8 sec for SCPL (p= 0.339). As for fundamental frequency or MPT it was observed bigger values for SCPL, however these results were not statistically meaningful either.
DISCUSSION Normally, the vocal quality suffers great impact after the resections of the larynx. In this research, the SCPL presented predominant hoarse, breath voice and instable vocal quality while NTL has predominance of strain vocal quality.
Although this difference was not statiscally meaningful, probably due to the size of the sample, it could have occured because most of the patients submitted to the Near-Total Laringectomy had undergone radiotherapy and neck dissection, being that, such facts had occured with minor frequency in the other group. Literature describes (4) that radiotherapy associated with neck dissection can reduce the mobility and the vibration of the remaining structures, and consequently generate a more strain vocal quality.
MAKIEFF and ZACHAREK (3, 11) evidenced from the moderate to severe breath-hoarse vocal quality ratings rendered for all of the patients submitted to the SPCL. According to literature (7), NTL presents predominantely hoarse vocal quality, but some patients presented vocal variations.
Comparing both groups, it could be observed greater strain and worse global degree in NTL. Patients undergone to SPCL presented predominance for low pitch and for NTL, a high pitch. Both surgeries resulted in adequate loudness, so it can be concluded that both surgery technics with speech therapy warrant a good shut of the remaining structures.
This research demonstrated that both surgery technics promoted adequate speech intelligibility. SCPL group presented better results than NTL group.
According to the literature (4, 5, 6) the listeners tend to fulfill the assessment of speech intelligibility influenced many times by speech quality, speech rhythm and loudness. This way, it can infer that listeners of this report have tendency to classify speech intelligibility for NTL group with reduced scores due to the strain vocal quality presented by these patients. In contrast, hoarse vocal quality associated to a high pitch, that was detected in this work, in patients submitted to the SPCL, the literature has pointed it as a voice of great social acceptance.(12).
During this research, higher values were found for the average fundamental frequency of sustained vowel /a/ in SPCL (141.6 Hz) when compared to NTL group (120.6 hz), without statiscal significance. A previous study found in SCPL fundamental frequency with values like 150 Hz (4). Premalatha (8) reported values for NTL as 96.04 Hz for fundamental frequency. Infante (13), reported high values for fundamental frequency in NTL, when compared with normality.
The Jitter, APQ and NHR sustained vowel /a/ presented higher values for NTL, without statiscal significance. Laccourreye (14) reported values for NTL: Jitter - 6.82 %, Shimmer - 19.84% and NHR - 0.66 %. Literature has demonstrated that high values for fundamental frequency (fO) and for acoustic parameters are related to the strain and vibrant tissue reduction (4). One more time to emphasize consequences of radiotherapy is that the hardness of the tissues can result of a lesser quantity of the vibrant tissue. Besides that, the neck dissection is more agressive in NTL than in SCPL turning the region into a more rigid one.
In this research, the acoustic values for Jitter, Shimmer, PPQ, APQ and NHR showed alterations compared to the normality for both groups. On the other hand, the fundamental frequency acoustic parameter was found within the normality standard for both surgery techniques.
In relation to MPT, SCPL presented higher values compared to NTL. According to the literature, SCPL shows 17 sec while MPT and NTL presents 6.7 sec (5.8). This difference might be related to the inefficient shutting of the stoma for NTL patients, as both used pulmonary air flow. Maximum phonation time was, on average, half of the time of normal laryngeal speakers, due to excessive transglottal airflow.
It was observed through this study a low level of concordance among the 3 valuers. Valuers 2 and 3 agreed for pitch, loudness and intelligibility and the valuers 1 and 3 agreed for strain and global.
KREIMAN et al (15) observed that listeners with bigger experience in voice have a great variability in the perception of pathological voices, while lay listeners who are not exposed to pathologic voices, consider them with the same strategy used for normal voices. So lay listeners present less variability in pathologic voice analysis. Authors atributted this variability to the previous experience of the valuers. The authors concluded that perfect agreement and reliability are not achieved, not even from a theoretical viewpoint. The valuers that contributed for this report, demonstrated low concordance, valuers 2 and 3 showed concordance with pitch, loudness and intelligibility, and valuers 1 and 3 with strain and global score.
The main limitations of our study had been in relation to the reduced number of patients for each type of surgery; heterogeneity among the groups (surgery time, radiotherapy treatment). This way, we stand out the need of more studies with bigger samples for truly detect statistically if there is difference between the surgeries or if both promote the same vocal impact.
CONCLUSION In this present research it is not presentend any statistical difference in the vocal parameters studied in spite the vocal behavior between the surgeries. The SCPL presented better results for speech intelligibility and for vocal quality and pertubation measure when compared to the NTL.
REFERENCES 1. Mayer EH, Rieder W. Technique de layngectomie permettant de conserver la perméabilité respiratoire (La cricohiodopexie). Ann Otolaryngol Chir Cervicofac. 1959; 76:677-81.
2. Pearson BW. Subtotal Laryngectomy, Laryngoscope. 1981; 91:1904-12.
3. Makeieff M, Barbotte E, Giovanni A, Guerrier B. Acoustic and Aerodynamic Measurement of Speech Production after Supracricoid Partial Laryngectomy. Laryngoscope. 2005; 115(3):546-551.
4. Vale LP, Amaral TCRDM, Góis JF, Fukuyama EE, Valentim PJ, Valle-Neto MJR, Barbosa LHF, Ramos HVL, Neves LR, Settanni FAP, Feitosa AA. In: Barros, APB, Arakawa L, Tonini MD, Carvalho VA. Laringectomia Parcial Supracricoídea: Avaliação Perceptiva Auditiva e Acústica da voz. In: Fonoaudiologia em Cancerologia. Imprensa Oficial; 2000. p246-52.
5. Weinstein GS, Laccourreye O, Ruiz C, Dooley P, Chalian A, Mirza N. Larinx Preservation with Supracricoid Partial Laryngectomy with Cricohyodoepiglottopexy: Correlation of Videostroboscopic Findings and Voice Parameters. Ann Otol Rhinol Laryngol. 2002; 111:1-6.
6. Bron L, Pasche P, Brossard E, Monnier P, Schweizer V. Functional Análisis Alter Supracrioid Partial Laryngectomy With cricohyoidoepiglottopexy. Laryngoscope. 2002; 112(7):1289-1293.
7. Figueiredo ES, Wannmacher L, Forte AP, Soneghet RM, Campos CA, Rocha GB et al. Aspectos Fonaoudiológicos no pós-operatório da laringectomia Near Total. In: Fonoaudiologia em Cancerologia. Imprensa Oficial; 2000. p. 207-13.
8. Premalatha M, Shenoy AM, Anantha N. Speech Evaluation after Near Total Laryngectomy and Total Laryngectomy - A Compartive Acoustic Analysis. Indian J. Cancer. 1994; 31(4):244-249.
9. Goodglass H, Kaplan E. The assessment of aphasia and related disorders. Philadelphia: Lea & Fegiber, 1972.
10. Hirano M. Clinical Examination of Voice. New York: Springer Verlag, 1981.
11. Zacharek MA, Pasha R, Meleca RJ, Dworkin JP, Stachler RJ, Jacobs JR et al. Functional Outcomes after Supracricoid Laryngectomy. Laryngoscope. 2001; 111:1558-64.
12. Behlau M. Voz; O livro do Especialista. 1ª ed. Rio de Janeiro:Editora Revinter;2005, Vol 2, pp. 233-237.
13. Infante SFC, Preciado LJA, Peres CAF, Infante VF. Laryngectomía Casi total. Comentários Clínicos y estúdio de la Voz. ORL-DIPS, 2003; 30(1):27-33.
14. Laccourreye O, Crevier-Buchman L, Muscatello L, Hans S, Ménard M, Brasnu D. Speech And Voice Characterisitcs After Near-Total Laryngectomy - A Preliminary Prospective Study. Ann Otol Rhinol Laryngol. 1998; 107:1061-65.
15. Kreiman J, Gerrat B, Kempster G, Erman A, Berke GS. Perceptual evaluation of voice quality review, tutorial, and a framework for future research. J Speech Hear Res. 1993; 36:21-40.
ANNEX1. Especialista em Voz. Fonoaudióloga.
2. Mestranda. Fonoaudióloga do Instituto do Câncer Dr. Arnaldo Vieira de Carvalho.
3. Doutora. Cirurgiã de Cabeça e Pescoço do Instituto do Câncer Dr. Arnaldo Vieira de Carvalho.
Instituição: Instituto do Câncer Dr. Arnaldo Vieira de Carvalho, São Paulo, Brasil.
Endereço para correspondência: Janaina Bueno da Silva - Rua José Pavoni, 575 - Centro - Rincão / SP - CEP- 14830-000.
Este artigo foi submetido no SGP (Sistema de Gestão de Publicações) da R@IO em 13 de março de 2007. Cod. 229. Artigo aceito em 11 de junho de 2007.
PDF em Português
Texto em Inglês
