Skip to main content
PMC home page
Journal of Applied Oral Science logoLink to Journal of Applied Oral Science
. 2006 Dec;14(6):421–426. doi: 10.1590/S1678-77572006000600007

SPEECH EVALUATION WITH AND WITHOUT PALATAL OBTURATOR IN PATIENTS SUBMITTED TO MAXILLECTOMY

AVALIAÇÃO DE FALA COM E SEM OBTURADOR DE PALATO EM PACIENTES SUBMETIDOS À MAXILECTOMIA

Viviane de Carvalho-Teles 1,✉, Maria Inês Pegoraro-Krook 2, José Roberto Pereira Lauris 3
PMCID: PMC4327294  PMID: 19089242

Abstract

Most patients who have undergone resection of the maxillae due to benign or malignant tumors in the palatomaxillary region present with speech and swallowing disorders. Coupling of the oral and nasal cavities increases nasal resonance, resulting in hypernasality and unintelligible speech. Prosthodontic rehabilitation of maxillary resections with effective separation of the oral and nasal cavities can improve speech and esthetics, and assist the psychosocial adjustment of the patient as well. The objective of this study was to evaluate the efficacy of the palatal obturator prosthesis on speech intelligibility and resonance of 23 patients with age ranging from 18 to 83 years (Mean = 49.5 years), who had undergone inframedial-structural maxillectomy. The patients were requested to count from 1 to 20, to repeat 21 words and to spontaneously speak for 15 seconds, once with and again without the prosthesis, for tape recording purposes. The resonance and speech intelligibility were judged by 5 speech language pathologists from the tape recordings samples. The results have shown that the majority of patients (82.6%) significantly improved their speech intelligibility, and 16 patients (69.9%) exhibited a significant hypernasality reduction with the obturator in place. The results of this study indicated that maxillary obturator prosthesis was efficient to improve the speech intelligibility and resonance in patients who had undergone maxillectomy.

Keywords: Maxillary neoplasm, Palatal obturator, Speech intelligibility, Hypernasality

INTRODUCTION

One of the main impacts of patients submitted to maxillectomy is the impairment of speech intelligibility. The undesirable coupling between the oral and nasal cavities reduces intraoral air pressure during speech production causing articulatory imprecision, hypernasal speech, nasal air emission, and reduced vocal loudness8. One of the main problems faced by them is the impairment of speech intelligibility, which interferes with the quality of life.

The defects created by maxillectomy can be repaired by prosthetic obturation or even reconstruction, using free and microsurgical transplants, grafts, and distant or regional flaps. However, an obturator is still considered to be one of the best rehabilitation tools in maxillary resections due to its rapid accomplishment, low cost, and the possibility of modification according to the patient needs.

The aim of obturator prosthesis is to obliterate the undesired communication between the oral and nasal cavities created by the tumor resection surgery, and to improve speech intelligibility and swallowing.

However, studies in the literature correlating maxillary obturator prosthesis with functional aspects and perceptual bases of speech are scarce1,10,11,13,14,16,17,18. Many of these studies are case reports and/or shows heterogeneous groups with individuals who have involvement of soft palate, that may not be as successful as in those with resections limited only to the hard palate. Yoshida, et al.17 (1990) observed improvement in speech intelligibility after prosthetic treatment in only four of eight maxillectomized patients studied. Three patients of this group with soft defects achieved slight improvement in syllable intelligibility with obturator.

Sullivan, et al.13 (2002) reported speech intelligibility, speaking rate and communication effectiveness results for 32 patients who had undergone partial surgical resection of the maxilla with a wide range of defects involving the hard and soft palate. Their results revealed that prosthodontic intervention resulted in a 33% increase in sentence intelligibility, a 26-word-per-minute increase in speaking rate, and a 4.2 scale point improvement (0-7 scale) in hypernasality across a wide range of maxillary and soft palate defects.

The objective of the present study was to determine the efficacy of palatal obturator prosthesis in speech intelligibility and resonance in patients submitted to inframedial-structural maxillectomy.

MATERIAL AND METHODS

The records of patients submitted to treatment at the São Paulo Oncology Center Foundation between 1995 and 2001 were reviewed in order to select those who had undergone maxillectomy. One-hundred patients were identified and invited by telephone and/or telegram contact. Thirty-three (33%) of them could not be located and/or did not respond to the invitation. Of the remaining 67 (67%), 19 (19%) had died, 10 (10%) had recurrences and resections in other regions of the oral cavity and/or oropharynx, 4 (4%) had undergone maxillectomy restricted only to the alveolar margin, 3 (3%) had undergone chemotherapy and/or radiotherapy, 3 (3%) were living in long distance areas, 3 (3%) were not Brazilians, and 2 (2%) presented difficulties in motor locomotion, therefore were not included in the sample. The remaining 23 patients (23%) who had undergone inframedial-structural maxillectomy with an oronasal fistula, and were wearing a stable maxillary obturator prosthesis for at least 3 months, were selected for the sample.

Thus, 23 patients participated in the study, 17 females and 6 males, with ages ranging from 18 to 83 years (mean = 49.5 years) who had been submitted to inframedial-structural resection along the maxilla. Out of these, 10 had been submitted to radiotherapy and 6 to speech therapy.

All patients were asked to answer a questionnaire with the following data: patient identification, history of the disease, type of surgery (extent and characteristics of palatal resection), complementary treatments (pre- or postoperative radiotherapy and/or chemotherapy), type of prosthesis, time of prosthesis wearing and speech therapy (Table 1).

TABLE 1. Biographical data of patients who participated in the study.

Patients # Age (y) Gender Extent Palatal Resection Radiotherapy Speech Therapy Type of Prosthesis Type of prosthesis wearing (years and months)
1. 83 F Left L NO NO CMD 1y
2. 77 F Right L + AR NO YES CMD 3m
3. 76 F Left L + AR NO NO CMD 2y 5m
4. 73 F Anterior C + AR YES NO CMD 4y 5m
5. 55 F C YES NO RPD 5y 4m
6. 54 F Bilateral + left AR YES NO CMD 2y5m
7. 52 F Left L + AR NO NO CMD 1y4m
8. 47 F Left L + AR NO NO RPD 2y2m
9. 46 F C NO NO RPD 3m
10. 44 F Right L + AR NO NO RPD 1y2m
11. 40 F C NO NO CMD 3m
12. 35 F Left L + AR YES NO RPD 10y
13. 29 F Left L + AR YES YES RPD 2y4m
14. 27 F Right L NO YES RPD 2y5m
15. 25 F Left L + AR NO YES RPD 2y5m
16. 18 F Right L + AR YES NO RPD 4y6m
17. 18 F C NO NO RPD 1y4m
18. 77 M Right L + AR NO NO CMD 9y5m
19. 74 M Right L NO NO CMD 5y4m
20. 65 M Right L YES NO RPD 9y
21. 62 M Right L YES YES CMD 2y
22. 42 M C + Anterior AR YES NO RPD 5y4m
23. 19 M Right L + AR YES NO RPD 5y
Mean 49.4 3y4m
Open in a new tab

Legend: L = lateral; C= central; AR = alveolar ridge; CMD = complete maxillary denture; RPD= removable partial maxillary denture.

The patients who met all inclusion criteria had their speech tape recorded in a silent room with a digital tape recorder (MD-RS37). During recordings, each patient remained standing with the microphone (Lesson SM 58P4) positioned at 12 cm from his/her mouth. The patients were asked to count from 1 to 20, to repeat 21 words which included all phonemes of the Brazilian Portuguese language at initial, medial, and final positions, as well as to perform 15 sec of spontaneous conversation, with and without the obturator prosthesis. All speech recording samples were randomly edited on a CD.

Speech resonance and intelligibility from all patient recordings were assessed by five speech-language pathologists experienced in the care of patients with head and neck cancer. All speech recordings and perceptual judgments were carried out in a silent room.

The speech samples from each patient, with and without the obturator, were presented in random order so that the listeners were unaware of whether the patient was wearing or not the obturator. Speech intelligibility was assessed from a 15s sequence of spontaneous conversation, and from the 1 to 20 counting recordings for all patients with and without the obturator.

Speech intelligibility from spontaneous conversation and counting recordings were assessed using a 6-point scale according to Pegoraro-Krook9 (1995), where 1 represented normal speech intelligibility, and 6 represented severely impaired speech intelligibility. Speech intelligibility was also judged as the percentage of the correct words that the listeners were able to correctly identify, immediately after hearing the speech stimulus. Speech resonance was judged upon spontaneous speech and counting stimuli, using a 5-point scale, in which 1 represented normal resonance, and 5 represented very severe hyper or hyponasality.

The values established for each patient were extracted by median of five speech-language pathologists. Comparisons of the speech intelligibility and resonance results of all speech stimulus recordings, with and without the obturator were analyzed by the Wilcoxon test. Correlations between the spontaneous speech intelligibility and intelligibility upon word repetition, with and without the obturator, spontaneous speech intelligibility and resonance with and without the obturator, and intelligibility upon word repetition and resonance with and without the obturator were established using Spearman's Correlation Coefficient Test (r). The concordance between the five speech-language pathologists was extracted using the Kendall Coefficient of Concordance Test (W).

RESULTS

The concordance under the conditions without and with obturator among the five speech-language pathologists obtained using the Kendall Coefficient of Concordance Test (W) were 0.70 and 0.85 (p<0.001) for spontaneous speech intelligibility, respectively, 0.79 and 0.85 (p<0.001) for word intelligibility, respectively, and 0.80 and 0.62 (p<0.001) for speech resonance respectively.

Assessment of speech intelligibility by the listeners on the basis of the spontaneous speech recordings revealed a significant improvement in intelligibility with the obturator in place for 19 (82.6%) of the 23 patients, while no difference was found for the remaining 4 (17.4%). Mean levels of spontaneous conversation with and without a prosthesis were 0.78 and 1.30 respectively. Speech intelligibility judgments results based on repetition of the 21 words demonstrated that the listeners were able to identify 66.7% of correct words without the obturator and 90.4 % with the obturators for 23 patients. The speech resonance results demonstrated a significant reduction in hypernasality with the obturators in place for 16 (69.6%) of the 23 patients. Of the 7 remaining patients, 5 (21.7%) did not show significant difference in hypernasality when wearing the obturator, and 2 (8.7%) presented a slight worsening. Mean levels of speech resonance judgments with and without the obturator were 1.9 and 3.2 respectively (Table 2). The results of speech intelligibility judgments upon spontaneous speech and counting recordings, and upon the difference in correct words were statistically significant (p<0,001) between the conditions with and without obturator. Statistically significant results were found in speech resonance (p<0.001) between the two conditions.

TABLE 2. Means and standard deviation (SD) scores for spontaneous speech, word intelligibility, and speech resonance for 23 patients with and without palatal obturator prosthesis.

Variable Without Prosthesis With Prosthesis Wilcoxon test p
Mean SD Mean SD
Spontaneous (level) 1.30 0.34 0.78 0.37 < 0.0001
Word Intelligibility (%) 66.7 3.20 90.4 1.20 < 0.0001
Speech Resonance (level) 3.23 1.45 1.94 0.61 < 0.0001
Open in a new tab

Spearman's Correlation Coefficient Test (r) showed significant correlation between spontaneous speech intelligibility and resonance in the absence of the obturator (r = -0.78, p<0.001); between word intelligibility and speech resonance in the absence of the obturator (r = -0.83, p<0.001); between word intelligibility and speech resonance in the presence of the obturator (r = -0.87, p<0.001); and between word and spontaneous speech intelligibility in the absence of the obturator (r = -0.80, p<0.001). No significant correlation was observed between spontaneous speech intelligibility in the presence of the obturator and any of the variables analyzed.

DISCUSSION

The standard measurement of communicative function is speech intelligibility since speech is a social instrument, with most significant measurements starting with the extent to which speech can be understood2.

In the present study, all assessments of spontaneous speech intelligibility demonstrated improvement in intelligibility with the use of a prosthesis in most patients, in agreement with results reported in the literature1,10,11,13,17.

According to Rieger, et al.11 (2003) there are several important background patient characteristics that cannot be ignored when looking at either clinical speech measurements or patient perceptions regarding obturator function. Comparison of the two assessments of speech intelligibility has shown that under the condition without prosthesis 15 patients (65.2%) presented poor levels of speech intelligibility, while in the remaining 8 patients (34.8%) intelligibility ranged from mild to normal. We therefore tried to identify the factors that differed in each group.

The first factor was the extent of surgery. Out of 15 patients with lower speech intelligibility scores, 10 (66.7%) had undergone wider resections of the maxilla comprising the length of the alveolar border. In contrast, only 4 (50%) of the 8 patients with better speech intelligibility scores showed wider maxillectomies. Kornblith, et al.3 (1996) reported that obturators were more functional during communication and swallowing in patients with smaller resections of the hard palate. Moreover, Bohle, et al.1 (2005) demonstrated for 55 patients who underwent ablative head and neck cancer therapy that as the percentage of resection of palate or tongue increased, the intelligibility of speech decreased. Sullivan, et al.13 (2002) observed individuals with complete unilateral hard and complete unilateral soft palatal defects and found a mean intelligibility score of 46.5% with the prosthesis removed and 79.7% with it inserted. In addition, maxillectomies, particularly the larger ones, restrict the contact between the tongue and palate, impairing speech intelligibility5.

Another important factor was radiotherapy. Out of 15 patients with low speech intelligibility scores, 8 (53.4%) had been submitted to this treatment. One of the main consequences of radiotherapy is xerostomia. According to Novaes7 (1999), doses ranging from 65 to 70 Gy may lead to persistence of this alteration. Treatment of xerostomia is palliative and artificial saliva can be used to relieve this discomfort but the outcome is poor. Reduced saliva production and/or the presence of thicker saliva (an increased mucous component compared to the serous component) impair stability and retention of the dental prosthesis, mastication and swallowing, as well as causing difficulties in sound articulation15. Rieger, et al.11 (2003) observed that a drier mouth was associated with poorer intelligibility for patients who had resection of maxilla with history of radiotherapy.

A last factor was speech therapy. Most patients (approximately 75%) have not undergone this therapy probably because most of them lived in towns where this treatment was not available. Lack of speech therapy did not interfere with the general results of speech intelligibility evaluations which, according to Pegoraro-Krook9 (1995), was probably due to the fact that the patients were normal speakers before cancer surgery, and anatomical repositioning alone in combination with palatal obturation was sufficient to recover speech without the need for speech therapy. However, we may infer that patients whose speech intelligibility continued to be poor with the prosthesis in place could have benefited from speech therapy since many of them showed imprecise articulation and increased speaking velocity, among others.

Hypernasality is a perceptive phenomenon which affects the quality of life due to its impact on speech intelligibility6. In the present study, the speech resonance results have demonstrated a reduction in hypernasality with the use of the prosthesis in most patients (69.6%), in agreement with some studies in the literature1,9,13

Patients #10, #14, #15, #5 and #19 did not have improved speech resonance when wearing the prosthesis. In the first three patients, speech resonance had already been considered to be normal or close to normal without prosthesis and therefore its use did not change this result. In addition to that, their prostheses have shown better stability and adaptation, which allowed a better articulatory pattern.

In contrast, patients #5 and #19, who continued to have moderate to severe hypernasality and patients #6 and #11, who have shown worsening of hypernasality with the use of the prosthesis, presented stuck or imprecise articulation, reduced loudness and increased speech velocity despite good stability and retention of the prosthesis. None of these patients had undergone speech therapy. According to Russo and Behlau12 (1993), disorders of the articulatory pattern and speech velocity and rhythm frequently compromise speech. We therefore believe that the assessment of hypernasality by the listeners might be influenced by other parameters related to articulation, such as vocal quality, pitch, rhythm and inflection, with a consequent difficulty in differentiating speech resonance parameters from other communication variables1,8.

A significant correlation (r = -0.78) was observed between the judgment of spontaneous speech intelligibility and the judgment of speech resonance without the prosthesis, suggesting that the greater the impairment in patient's hypernasality, the higher will be the degree of impairment of speech intelligibility. Thus, patients with higher levels of hypernasality presented significant difficulties in making themselves understood. According to Yoshida, et al.18 (2000), mild hypernasality can generate only a discrete distortion in speech, while severe hypernasality leads to a drastic reduction in speech intelligibility, compromising the individual's oral communication and social contact. Bohle, et al.1 (2005) observed that perceptual ratings of resonance were associated with percent intelligibility of words and sentences, i.e. the data indicate that perceptual ratings of resonance indicative of increased hypernasality were associated with poorer intelligibility.

Another significant correlation was observed between the variables speech intelligibility upon word repetition and speech resonance both with and without the prosthesis (r = -0.87 and r = -0.83, respectively). These results demonstrate that, in both conditions, the larger the number of correctly repeated words, the lower the level of hypernasality. Studies reported a significant increase in the percentage of correctly identified words, associated with a reduction in hypernasality, after prosthetic treatment of maxillectomized patients1,13,17.

The correlation observed between the assessment of spontaneous speech intelligibility and intelligibility upon word repetition in the absence of a prosthesis was also statistically significant (r = -0.80), i.e. the larger the number of correctly repeated words, the more intelligible will be the spontaneous speech of these patients. Similar results were observed in the literature10.

No significant correlation was observed between spontaneous speech intelligibility in the presence of a prosthesis and any of the variables analyzed, probably due to the heterogeneity of the sample and the small number of patients participating in the study9,11.

In addition, significant interclass agreement was observed for the assessment of spontaneous speech intelligibility (r = 0.70 and r = 0.85) and word intelligibility (r = 0.79 and r = 0.85) both with and without the use of the prosthesis, respectively. However, a low, although statistically significant, level of agreement was observed between the listeners of this study in speech resonance assessment without prosthesis (r = 0.62). Yoshida, et al.18 (2000) confirmed that perceptive evaluation shows a poor intra and inter-judgment agreement and suggested the use of objective instruments for the study of speech resonance. In contrast, Kreiman, et al.4 (1993) reported that total agreement among listeners is not expected, even in the case of experienced examiners, due to the difficulty in assessing minimal changes occurring after prosthetic treatment in a perceptive manner. The authors concluded that perfect agreement and reliability are not achieved, not even from a theoretical viewpoint.

CONCLUSION

The present study has concluded that obturator prosthesis contributed to improved speech intelligibility in maxillectomized patients. However, the success of prosthetic treatment might be limited by factors such as radiotherapy, the extent of maxillary surgery and speech therapy. We therefore emphasize the integration of an interdisciplinary team in order to increase the efficacy of rehabilitation and the quality of life of these patients.

REFERENCES

  • 1.Bohle G, 3rd, Rieger J, Huryn J, Verbel D, Hwang F, Zlotolow I. Efficacy of speech aid prostheses for acquired defects of the soft palate and velopharyngeal inadequacy-clinical assessments and cephalometric analysis: a Memorial Sloan-Kettering Study. Head Neck. 2005;27(3):195–207. doi: 10.1002/hed.10360. [DOI] [PubMed] [Google Scholar]
  • 2.Fletcher SG, Jacob RF, Kelly DH, Krieger BM, Martin DE, Myers EN, et al. Speech production in rehabilitation of head and neck cancer patients. Head Neck. 1991;13(8):1–15. [Google Scholar]
  • 3.Kornblith AB, Zoltov IM, Gooen J, Huryn JM, Lerner T, Strpmg EW, et al. Quality of life of maxillectomy patients using an obturator prosthesis. Head Neck. 1996;18(4):323–334. doi: 10.1002/(SICI)1097-0347(199607/08)18:4<323::AID-HED3>3.0.CO;2-#. [DOI] [PubMed] [Google Scholar]
  • 4.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. doi: 10.1044/jshr.3601.21. [DOI] [PubMed] [Google Scholar]
  • 5.Matsuy Y, Ohno K, Shirota T, Imai S, Yamashita Y, Michi K. Speech function following maxillectomy reconstructed by rectus abdominis myocutaneous flap. J Craniomaxillofac Surg. 1995;23:160–164. doi: 10.1016/s1010-5182(05)80005-8. [DOI] [PubMed] [Google Scholar]
  • 6.Mckinnon SL, Hess CV, Landry RG. Reactions of college students to speech disorders. J Commun Disord. 1986;19:75–82. doi: 10.1016/0021-9924(86)90005-5. [DOI] [PubMed] [Google Scholar]
  • 7.Novaes PERS. A atuação da fonoaudiologia no câncer de cabeça e pescoço. In: Carrara-Angelis E, Fúria LB, Mourão LF, Kowalski LP, editors. Radioterapia em neoplasias de cabeça e pescoço. São Paulo: Lovise; 1999. pp. 33–41. [Google Scholar]
  • 8.Plank DM, Weinberg B, Chalian VA. Evaluation of speech following prosthetic obturation of surgically acquired maxillary defects. J Prosthet Dent. 1981;45(6):626–638. doi: 10.1016/0022-3913(81)90423-6. [DOI] [PubMed] [Google Scholar]
  • 9.Pegoraro-Krook M I. Universidade Federal de São Paulo, Escola Paulista de Medicina. São Paulo (SP): 1995. Avaliação dos resultados de fala de pacientes que apresentam inadequação velofaríngea e que utilizam prótese de palato [tese] [Google Scholar]
  • 10.Rieger J, Wolfaardt J, Seikaly H, Jha N. Speech outcomes in patients rehabilitated with maxillary obturator prostheses after maxillectomy: a prospective study. Int J Prosthodont. 2002;15(2):139–144. [PubMed] [Google Scholar]
  • 11.Rieger JM, Wolfaardt JF, Jha N, Seikaly H. Maxillary obturators: the relationship between patient satisfaction and speech outcome. Head Neck. 2003;25(11):895–903. doi: 10.1002/hed.10299. [DOI] [PubMed] [Google Scholar]
  • 12.Russo I, Behlau M. Percepção da fala: análise acústica do português brasileiro. São Paulo: Lovise; 1993. [Google Scholar]
  • 13.Sullivan M, Gaebler C, Beukelman D, Mahanna G, Marshall J, Lydiatt D, et al. Impact of palatal prosthodontic intervention on communication performance of patients maxillectomy defects: a multilevel outcome study. Head Neck. 2002;24(6):530–538. doi: 10.1002/hed.10095. [DOI] [PubMed] [Google Scholar]
  • 14.Tobey EA, Lincks J. Acoustic analyses of speech changes after maxillectomy and prosthodontic management. J Prosthet Dent. 1989;62:449–455. doi: 10.1016/0022-3913(89)90180-7. [DOI] [PubMed] [Google Scholar]
  • 15.Toljanic JA, Zucuskie TG. Use of a palatal reservoir in denture patients with xerostomia. J Prosthet Dent. 1984;52(4):540–544. doi: 10.1016/0022-3913(84)90342-1. [DOI] [PubMed] [Google Scholar]
  • 16.Umino S, Masuda G, Ono S, Fujita AK. Speech intelligibility following maxillectomy with and without prostheses: an analysis of 54 cases. J Oral Rehabil. 1998;25(2):153–158. doi: 10.1046/j.1365-2842.1998.00238.x. [DOI] [PubMed] [Google Scholar]
  • 17.Yoshida H, Michi K, Ohsawa T. Prosthetic treatment for speech disorder due to surgically acquired maxillary defects. J Oral Rehabil. 1990;17(b):565–571. doi: 10.1111/j.1365-2842.1990.tb01427.x. [DOI] [PubMed] [Google Scholar]
  • 18.Yoshida H, Furuya Y, Shimodaira K, Kanazawa T, Kataoka R, Takahashi K. Spectral characteristics of hypernasality in maxillectomy patients. J Oral Rehabil. 2000;27(8):723–730. doi: 10.1046/j.1365-2842.2000.00537.x. [DOI] [PubMed] [Google Scholar]

Articles from Journal of Applied Oral Science are provided here courtesy of Bauru School of Dentistry

RESOURCES