Evaluation of the closest speaking space in different dental and skeletal occlusions

Olcay akar, Prof Dr Med Dent,a Canan Bural, Dr Med Dent,b Tongu Sln, Dr Med Dent,c Evren zta, Dr Med Dent,d and Glnaz Maran, Dr Med Dente Istanbul University, Faculty of Dentistry, Istanbul, Turkey
Statement of problem. The closest speaking space (CSS) together with the vertical overlap of anterior teeth during the production of the /s/ sound have not been previously investigated with respect to differences in dental and skeletal orthodontic classifications. Purpose. The purpose of this study was to investigate the CSS in dental and skeletal occlusions and to analyze the cause and effect relationship of the CSS and the amount of the vertical overlap of anterior teeth. Material and methods. Poly vinylsiloxane interocclusal registration material was placed bilaterally onto the occlusal surfaces of premolar and molar teeth of 155 native Turkish speaking adolescent and young adult dentate participants, who were then asked to pronounce the word seyis. The thinnest point between the maxillary and mandibular teeth was recorded in millimeters as the CSS. The occlusion of each participant was classified according to the Angle dental and Steiner skeletal classifications. The differences in CSS values within each classification were statistically analyzed with the Kruskal-Wallis test, and the correlation between the CSS and the vertical overlap was statistically analyzed with the Spearman Rho Correlation tests (P<.05). Results. The differences in the CSS were only significant between Angle Class II division 2 and Class III groups (P=.034), while the differences in the CSS between skeletal classes were not significant. The correlation between the amount of CSS and the amount of vertical overlap was not significant. Conclusions. The results showed that regardless of dental and skeletal occlusions, average CSS values could be used to determine the occlusal vertical dimension of prosthetic restorations. (J Prosthet Dent 2013;109:222-226)

Clinical Implications

Within the limitations of this clinical study, average closest speaking space values of 1 to 3 mm can be used in the clinical determination of the occusal vertical dimension of prosthetic restorations, regardless of dental and skeletal classified occlusions; no correlation was found between the amount of vertical overlap and closest speaking space. The use of the Turkish word seyis was appropriate for registering the closest speaking space in native Turkish speakers.
Mandibular movements are complex and vary in speaking, masticating, and swallowing. Speaking is helpful in the clinical determination of parameters such as occlusal vertia b

cal dimension (OVD), tooth position, and anterior guidance.1-8 Pound9 indicated that vertical and horizontal overlap, occlusal classification, optimum OVD, incisal guidance, cusp

height, and the anterior display of mandibular teeth are the key factors in developing clear speech. In speech, the production of sibilant sounds requires a small space

Professor, Department of Removable Prosthodontics. Assistant Professor, Department of Prosthodontics. c Associate Professor, Department of Prosthodontics. d Associate Professor, Department of Orthodontics. e Associate Professor, Department of Orthodontics.

The Journal of Prosthetic Dentistry

akar et al

April 2013
between the teeth. When pronouncing these sounds, the occlusal surfaces and incisal edges of the mandibular teeth and the maxillary teeth are in their closest relation to each other; this is the closest speaking space (CSS).2,3,5,9-11 The CSS is the minimum distance between the teeth that occurs during the pronunciation of words containing sibilant sounds, particularly those in combination with the vowels /e/ and /i/.12 The CSS identifies and establishes the acceptable OVD so that the teeth do not contact during speech and the anterior teeth or occlusal rims are separated by about 2 mm.1,13 The amount of CSS produced by pronouncing words containing sibilant sounds varies with the participants10,14 depending on anatomic and morphologic factors such as skeletal variations,8 vertical overlap,10,15-17 increased OVD,18 and the absence of teeth.19 The CSS has been extensively studied and mean values have varied from 0.7 to 3.1 mm in participants with normal occlusion and dentition.3,5,12,13,17 Pound2 described the amount of posterior speaking spaces for occlusal classifications with respect to the amount of vertical and horizontal overlap of anterior teeth during the production of the sibilants. The amount of posterior speaking spaces should be in a range of 1.5 to 3 mm for Class I and less than 2 mm for Class III occlusions, although this space might controversially be increased up to 10 mm in patients with Class II occlusion. This trend was later scientifically compared and confirmed in only 2 clinical studies within a broad range of varied findings on CSS.5,20 In both studies, the maxillomandibular relationships were classified according to the incisor relationship, which was based on the contact of the mandibular incisors with the palatal surface of the maxillary incisors21,22; no orthodontic methods were used to classify the participants. The CSS has been investigated by using different test materials for speech such as the phonation of a phoneme,15,17,23,24 a single word,10,12,17,25-29 a sentence,25,30,31 a passage,5,13,16,20,25,32 or counting.33,34 It can be registered by extraoral (indirect) and intraoral (direct) methods by means of manual17 or radiographic measurements,15,17 or electronic kinesiographs. 5,10,12,13,16-18,20,23-27,29-34 The limitations of electronic registration methods are the impairment of the orofacial region or difficulty in adjusting the probes, which is time consuming.17 A new technique for the clinical registration of CSS with vinyl polysiloxane interocclusal registration material has been suggested, and only minimal differences compared with kinesiography were noted.12 The purpose of this study was to investigate the CSS in the Angle dental and Steiner skeletal Class I, II, and III occlusions and to analyze the cause and effect relationship of CSS with respect to the amount of vertical overlap. The null hypothesis was that the CSS would not differ in different dental and skeletal occlusions. Recording of CSS The participants were seated in a dental chair and the recording procedures were fully explained. Before recordings, participants were asked to repeat the Turkish word seyis (horse rider in English). Participants were then given a few minutes to become familiar with the word by repeating it aloud while an examiner checked the accuracy of the pronunciation. The participants were asked to repeat the word at a normal conversational rate and volume. Participants were instructed to sit in an upright position with their heads unsupported. The examiner placed approximately 1.5 cm of vinyl polysiloxane interocclusal registration material (Futar-D; Kettenbach GmbH & Co KG, Eschenburg, Germany) bilaterally on the occlusal surfaces of mandibular premolar and molar teeth. Then the participants were instructed to close their lips, swallow, and repeat the word seyis 10 times and then not close or move their mandible for 30 seconds while the material polymerized. The record material was not placed on the anterior teeth, so as not to interfere with the /s/ sound and to determine that the participants pronunciation was accurate. After they had polymerized, the interocclusal records were removed from the mouth and identified as the left or right side. The recorded CSSs were measured as suggested by Rizzatti et al.12 The minimum thickness on the occlusal records at the first premolar region was measured with a digital gauge (Guilin; Guanglu Measuring Instrument Co Ltd, Shanghai, China) with a 0.01 mm resolution. The recording material used in the study has been shown to have a vertical discrepancy of about 23 m 1 hour after setting.35 To minimize dimensional changes, measurements were completed 1 hour after intraoral recordings, and the resulting value was recorded as CSS in mm. The operator who made the measurements was not aware of the participants dental or skeletal classification.

223

MATERIAL AND METHODS

Participants The selection criteria for the participants were the absence of systemic disease, signs or symptoms of temporomandibular dysfunction, speech defects, an anterior open occlusal relationship, tongue thrusting, lisping, and the presence of intact first and second premolars in occlusion. A statistical power analysis indicated that a minimum of 165 participants who were accepted for orthodontic treatment at Istanbul University, Faculty of Dentistry, Department of Orthodontics, (between January 2009 and December 2009) was required. A total of 210 native Turkish speaking, adolescent and young adult dentate individuals were included in the study. An informed consent was obtained from each participant or parent before the orthodontic treatment and recording of the CSS.

akar et al

224
Dental and skeletal classifications Each participants occlusion was evaluated by means of casts according to the Angle classification method.36 The skeletal classification was performed by means of cephalometric radiographs made with the same machine (Orthophos XG Plus; Sirona Dental Systems, Bensheim, Germany) according to ANB angle (Steiner cephalometric analysis), which is the angle between point A (Subspinale, the most posterior midline point in the concavity between the anterior nasal spine and the prosthion), nasion (the most anterior point on the frontonasal suture in the midsagittal plane), and point B (Supramentale, the most posterior midline point in the concavity of the mandible between the pogonion and infradentale) in themaxilla. It is the most commonly used measurement for evaluating anteroposterior disharmony of the jaws.37 Participants were classified as Class I: 0ANB5 degrees, Class II: ANB>5 degrees, Class III: ANB<0 degrees. The vertical overlap was measured as the vertical overlap from the labio-incisal edge of the right maxillary central incisor with the greatest overlap to the labio-incisal edge of the corresponding mandibular right central incisor and was expressed as a percentage (%). Descriptive analysis was performed to calculate the mean and standard deviation (SD) values for the CSS and the vertical overlap for dental and skeletal occlusions. The Wilcoxon test was used to determine whether a difference between the 148 right and 144 left interocclusal records existed. To test interobserver and intraobserver reliability, a total of 40 occlusal records were measured twice, with a 1-week interval by 2 different observers and analyzed with the Spearman Rho test. Gender differences in the CSS were analyzed with the Mann Whitney U test, and the differences in the CSS within dental or skeletal classifications were analyzed with the Kruskal-Wallis test. Comparisons of the CSS or vertical overlap among groups within the classifications were analyzed with the Mann Whitney U test, and the correlation between the CSS and the amount of vertical overlap was analyzed with the Spearman Rho Correlation test. The significance level was set at =.05. All statistical analyses were performed with SPSS software (SPSS for Windows, v11.5; SPSS Inc, Chicago, Ill).

Volume 109 Issue 4

statistically reliable. Interobserver (r =0.89, P<.001) and intraobserver (r=0.91, P<.001) reliability rates were high. The mean and SD of the CSS were 2.27 1.17 mm for the 148 right interocclusal records and 2.15 1.06 mm for the 144 left interocclusal records. The difference between the 2 sides was not statistically significant (P=.737), and so data belonging to the 148 right interocclusal records of the 155 (100 females, 55 males) participants were used in the further statistical analysis. The mean age of the 55 males was 14.96 years (range 11 to 22 years), and the mean age of the 100 females was 15.50 years (range 10 to 28 years). The overall mean and SD of the CSS (range 0.11 to 5.85 mm) was 2.27 1.17 mm. The difference in the CSS between males and females was not significant (P=.747). The mean and SD of the CSS of Angle Class I (55 participants) was 2.28 0.89 mm, of Class II division 1 (44 participants) 2.16 1.49 mm, of Class II division 2 (29 participants) 2.66 1.16 mm, and of Class III (20 participants) 1.92 0.95 mm. The mean and SD of the amount of vertical overlap (%) was 20.13 18.02 for Angle Class I, 42.37 26.90, for Class II Division 1, 48.75 35.90 for Class II

RESULTS
Of 210 participants, 35 had some difficulties (inaccurate pronunciation while recording with the registration material in the mouth, gagging) while making the records, and 30 had interocclusal records with no exact thickness or measurable area around the premolar region. Therefore, the records of 65 participants were discarded and the data obtained by measuring the records of 155 participants (100 females, 55 males) were used. Also, 7 right and 11 left records of participants in the final group could not be measured because of the lack of necessary thickness. As a result 148 right and 144 left interocclusal measurements were performed and evaluated. The Spearman Rho test found that the measurement of the CSS was

Table I. Means and SDs of CSS (mm) and vertical overlap (%) in dental and skeletal Class I, II, and III classification
Angle Dental Classification Class I Class II Div 1 Class II Div 2 (n=55) (n=44) (n=29) Mean SD Mean SD Mean SD
CSS 2.28 0.89 18.02 2.16 42.37 1.49 26.90 2.66* 48.75 1.16 35.90

Class III (n=20) Mean SD

1.92* 16.25 0.95 24.03

Vertical overlap (%) 20.13

Steiner Skeletal Classification Class I Class II Class III (n=77) (n=46) (n=25) Mean SD Mean SD Mean SD
CSS Vertical overlap % 2.24 26.60 1.07 23.18 2.46 47.48 1.44 31.63 2.00 13.75 0.84 21.02

* Indicates statistically significant differences among groups (P<.05).

The Journal of Prosthetic Dentistry

akar et al

April 2013
division 2, and 16.25 24.03 for Class III occlusions. The differences in CSS were only significant between Angle Class II division 2 and Class III occlusions (P =.034) (Table I). The mean and SD of the CSS was 2.24 1.07 mm for skeletal Class I (77 participants), 2.46 1.44 mm for Class II (46 participants), and 2.00 0.84 mm for Class III (25 participants). The mean and SD of the amount of vertical overlap (%) was 26.60 23.18 for skeletal Class I, 47.48 31.63 for skeletal Class II, and 13.75 21.02 for skeletal Class III occlusions. The differences in CSS within skeletal Class I, II, and III occlusions were not statistically significant (P =.452) (Table I). The correlation between the CSS and the amount of vertical overlap was not statistically significant (Spearman r=.110, P=.249). thickness or area around the premolar region was identified, so these records were discarded. These observations can be regarded as the limitations of this method. The present finding that the CSS did not differ between genders supports the finding20 for participants with mean age of 20 years. This result might indicate that CSS is similar in adolescents and young adult participants without gender differences. Common qualitative methods of recording occlusal traits include the British Standard Institute (BSI) classification for incisors and the Angle classification for molar relationships.21,22 The present findings indicate that the amount of CSS did not differ between the Angle dental and cephalometric skeletal classifications. The previous studies of Burnett and Clifford5 and Howell,20 however, used the incisor classifications. Burnett and Clifford5 reported that the difference in the CSS between occlusions was only significant for Class III, which showed the lowest CSS value, and emphasized that the CSS should be interpreted with the low number (n=5) of Class III participants in mind. However, with a higher number of participants, Howell20 reported no significant differences among classes. The CSS findings of these studies were close for only Class III participants, while the other findings were quite different.5,20 The same trend for Class III participants was also observed in the present study, but for the Angle Class II participants, the CSS findings for Class II division 1 and division 2 participants conflicted. This result might be due to the difference in the Angle and incisor classification systems. The results of this study indicate that the CSS does not significantly differ when a skeletal classification is considered. However, when the Angle classification was analyzed, a significant difference in the CSS was observed between the Class II division 2 and Class III, in which the mean CSS finding was lower than in Class II division 2. The increased vertical overlap in Class II division 2 and the decreased overlap in Class III occlusions might have resulted in this apparent significant difference. This finding suggests the vertical overlap of the anterior teeth affected the CSS. However, a correlation suggesting a cause and effect relationship between the CSS and vertical overlap was not found in the present study. Thus the null hypothesis of this study was confirmed. Most clinical research on the CSS has been performed by using English words containing sibilants to register the /s/ sound, for example, Mississippi, sixty-six, and yes25 and nonEnglish words such as sessantasei12 and seis,23,26,27 or complete sentences.30 However, as no available study exists investigating the CSS in native Turkish speakers, the findings of this study cannot be compared. The average CSS of 2.27 mm in the present study was close to the previous findings pertaining to dentate participants.5,16,17,20,25,26,32,33 The word seyis was chosen as a test for speech because translated in English the word horse rider is easily understood and that, its pronunciation in Turkish is similar to its original pronunciation in Spanish, starting and ending with an /s/ sound.26 Depending on the average value and the general trend in CSS values within the classes of malocclusion, the use of the word seyis in the evaluation of the CSS in native Turkish speakers seems appropriate. Although it is held that the CSS is constant throughout life,3 further investigations of the CSS in native Turkish speakers of different ages and dentitions are necessary. The present study was conducted on young adult dentate participants with dental and skeletal malocclusions needing orthodontic treatment. However, a direct relationship between the severity of malocclusion and misarticulation has not yet been scientifically proved. Further evaluation and comparison of the CSS of the same participants are suggested after correcting the vertical overlap, incisors, premolar and molar teeth

225

DISCUSSION
The null hypothesis that the CSS would not differ in different dental and skeletal occlusions was not rejected. CSS is one of the methods used to determine OVD and the esthetics of prosthetic restorations. This clinical study aimed to test the amount of CSS with respect to the differences in dental and skeletal orthodontic classifications. Rizzatti et al12 recommended the use of vinyl polysiloxane interocclusal registration material over the kinesiographic method to determine the CSS because it is simple and less expensive. In the present study, it was also observed that the majority of the participants easily pronounced the word seyis with the interocclusal registration material between the posterior teeth. It should be noted that 16% of the participants (35 participants out of 210) were not successful in recording the CSS perhaps because of atypical situations,20 personal skills, or reaction to a foreign material in the oral cavity (vomiting, gagging, or difficulty in speech). On the interocclusal records, of 14% of the participants (30 participants out of 210), no exact

akar et al

226
positions, and the occlusion with orthodontic treatment; this is especially true of Class II divisions 1 and 2 and Class III malocclusions.
8. Mehringer EJ. The use of speech patterns as an aid in prosthodontic reconstruction. J Prosthet Dent 1963;13:825-36. 9. Pound E. The mandibular movements of speech and their seven related values. J Prosthet Dent 1966;16:835-43. 10.Burnett CA. Mandibular incisor position for English consonant sounds. Int J Prosthodont 1999;12:263-71. 11.The Glossary of Prosthodontic Terms. J Prosthet Dent 2005;94:24. 12.Rizzatti A, Ceruti P, Mussano F, Erovigni F, Preti G. A new clinical method for evaluating the closest speaking space in dentulous and edentulous subjects: a pilot study. Int J Prosthodont 2007;20:259-62. 13.Howell PGT. Incisal relationships during speech. J Prosthet Dent 1986;56:93-9. 14.Miralles R, Dodds C, , Palazzi C, Jaramillo C, Quezada V, Ormeo G, et al. Vertical dimension. Part 1: comparison of clinical freeway space. Cranio 2001;19:230-6. 15.Benediktsson E. Variation in tongue and jaw position in s sound production in relation to front teeth occlusion. Acta Odontol Scand 1958;15:275-304. 16.Burnett CA, Clifford TJ. The mandibular speech envelope in subjects with and without incisal tooth wear. Int J Prosthodont 1999;12:514-8. 17.Meier B, Luck O, Harzer W. Interocclusal clearance during speech and in mandibular rest position. A comparison between different measuring methods. J Orofac Orthop 2003;64:121-34. 18.Burnett CA, Clifford TJ. A preliminary investigation into the effect of increased occlusal vertical dimension on mandibular movement during speech. J Dent 1992;20:221-4. 19.Agnello JG, Wictorin L. A study of phonetic changes in edentulous patients following complete denture treatment. J Prosthet Dent 1972;27:133-9. 20.Howell PG. The variation in the size and shape of the human speech pattern with incisor-tooth relation. Arch Oral Biol 1987;32:587-92. 21.Angle EH. Classification of malocclusion. Dental Cosmos 1899;41:248-64. 22.British Standard Institute. British Standard Incisor Classication. Glossary of Dental Terms BS 4492. London: British Standard Institute; 1983 23.de Souza RF, Compagnoni MA. Relation between speaking space of the /s/ sound and freeway space in dentate and edentate subjects. Braz Oral Res 2004;18:333-7. 24.Rodrigues Garcia RC, Oliveira VM, Del Bel Cury AA. Effect of new dentures on interocclusal distance during speech. Int J Prosthodont 2003;16:533-7. 25.Burnett CA, Clifford TJ. Closest speaking space during the production of sibilant sounds and its value in establishing the vertical dimension of occlusion. J Dent Res 1993;72:964-67.

Volume 109 Issue 4

26.de Souza RF, Compagnoni MA, Leles CR, Sadalla KB. Association between the speaking space of /s/ sound and incisal overlaps in dentate and edentate subjects. J Appl Oral Sci 2005;13:413-7. 27.de Souza RF, Marra J, Pero AC, Compagnoni MA. Effect of denture fabrication and wear on closest speaking space and interocclusal distance during deglutition. J Prosthet Dent 2007;97:381-8. 28.Schierano G, Mozzati M, Bassi F, Preti G. Influence of the thickness of the resin palatal vault on the closest speaking space with complete dentures. J Oral Rehabil 2001;28:903-8. 29.Burnett CA. Mandibular incisor position for English consonant sounds. Int J Prosthodont 1999;12:263-71. 30.Lu GH, Chow TW, So LK, Clark RK. A computer-aided study of speaking spaces. J Dent 1993;21:289-96. 31.George JP. Using the Kinesiograph to measure mandibular movements during speech: a pilot study. J Prosthet Dent 1983;49:263-70. 32.Burnett CA. Reproducibility of the speech envelope and interocclusal dimensions in dentate subjects. Int J Prosthodont 1994;7:543-8. 33.Gillings BR. Jaw movements in young adult men during speech. J Prosthet Dent 1973;29:567-76. 34.Rivera-Morales WC, Mohl ND. Variability of closest speaking space compared with interocclusal distance in dentulous subjects. J Prosthet Dent 1991;65:228-32. 35.Ghazal M, Ludwig K, Habil RN, Kern M. Evaluation of vertical accuracy of interocclusal recording materials. Quintessence Int 2008;39:727-32. 36.Ackerman JL, Nguyen T, Proffit WR. The decision-making process in orthodontics. In: Orthodontics current principles and techniques. Graber LW, Vanarsdall RL Jr, Vig KWL. 5th ed. St. Louis: Elsevier, Mosby, p. 21-2. 37.Caufield PW. Tracing technique and identification of landmarks. In: Radiographic cephalometry. From basics to 3-D imaging. Jacobson A, Jacobson RL. eds: 2nd ed. Chicago Ill: Quintessence Books, p. 49, 99. Corresponding author: Dr Canan Bural Istanbul University, Faculty of Dentistry Department of Removable Prosthodontics Capa, 34390, Istanbul TURKEY Fax: +90212-525-35-85 E-mail: cbural@istanbul.edu.tr Copyright 2013 by the Editorial Council for The Journal of Prosthetic Dentistry.

CONCLUSION
Within the limitations of this clinical study, the following conclusions can be drawn: 1. The CSS does not differ between males and females. 2. The CSS does not differ according to the skeletal classification. 3. The difference in the CSS was only significant between Angle Class II division 2 and Class III occlusions. 4. No correlation was found between the amount of vertical overlap and the CSS. 5. The use of the Turkish word seyis is appropriate for registering the CSS in native Turkish speakers. 6. Irrespective of dental and skeletal classified occlusions, average CSS values can be used to determine the OVD of prosthetic restorations.

REFERENCES
1. Spear FM. Fundamental occlusal therapy considerations. In: Science and practice of occlusion. McNeill C, ed; Hong Kong: Quintessence Publishing Co; 1997. p. 421-34. 2. Pound E. Let /S/ be your guide. J Prosthet Dent 1977;38:482-9. 3. Silverman MM. The speaking method in measuring vertical dimension. 1952. J Prosthet Dent 2001;85:427-31. 4. Curtis TA, Langer Y, Curtis DA, Carpenter R. Occlusal considerations for partially or completely edentulous skeletal Class II patients. Part II: Treatment concepts. J Prosthet Dent 1988;60:334-42. 5. Burnett CA. Clinical rest and closest speech positions in the determination of occlusal vertical dimension. J Oral Rehabil 2000;27:714-9. 6. Murrell GA. Phonetics, function, and anterior occlusion. J Prosthet Dent 1974;32:23-31. 7. McCord JF, Firestone HJ, Grant AA. Phonetic determinants of tooth placement in complete dentures. Quintessence Int 1994;25:341-5.

The Journal of Prosthetic Dentistry

akar et al