Journal of
Oral Rehabilitation
Journal of Oral Rehabilitation 2015 42; 670–677
Anatomy of the larynx and pharynx: effects of age, gender and height revealed by multidetector computed tomography Y. INAMOTO*†, E. SAITOH*, S. OKADA†, H. KAGAYA*, S. SHIBATA*, M. BABA‡, K. ONOGI*, S. HASHIMOTO§, K. KATADA¶, P. WATTANAPAN** & J. B. PALMER†† *Department of Rehabilitation Medicine, School of Medicine, Fujita Health University, Aichi, †Faculty of Rehabilitation, School of Health Sciences, Fujita Health University, Aichi, ‡Japanese Red Cross Ashikaga Hospital, Tochigi, §Department of Hygiene, Fujita Health University, Aichi, ¶Department of Radiology, Fujita Health University, Aichi, Japan, **Institute of Medicine, Suranaree University of Technology, Muang, Nakornratchasima, Thailand and
††
Department of Physical Medicine and Rehabilitation, Department of Otolaryngology-Head and Neck Surgery, and
Center for Functional Anatomy and Evolution, Johns Hopkins University, Baltimore, Maryland, USA
SUMMARY Although oropharyngeal and laryngeal structures are essential for swallowing, the threedimensional (3D) anatomy is not well understood, due in part to limitations of available measuring techniques. This study uses 3D images acquired by 320-row area detector computed tomography (‘320ADCT’), to measure the pharynx and larynx and to investigate the effects of age, gender and height. Fifty-four healthy volunteers (30 male, 24 female, 23–77 years) underwent one single-phase volume scan (035 s) with 320-ADCT during resting tidal breathing. Six measurements of the pharynx and two of larynx were performed. Bivariate statistical methods were used to analyse the effects of gender, age and height on these measurements. Length and volume were significantly larger for men than for women for every measurement (P < 005) and increased with height (P < 005). Multiple regression analysis was performed to understand the interactions of gender, height and
Background Gender and age are likely to influence oropharyngeal and laryngeal structure (1, 2). However, the normative reference values of oropharyngeal and laryngeal structures are not yet well established, due to the limitations of the available measuring techniques. © 2015 John Wiley & Sons Ltd
age. Gender, height and age each had significant effects on certain values. The volume of the larynx and hypopharynx was significantly affected by height and age. The length of pharynx was associated with gender and age. Length of the vocal folds and distance from the valleculae to the vocal folds were significantly affected by gender (P < 005). These results suggest that age, gender and height have independent and interacting effects on the morphology of the pharynx and larynx. Three-dimensional imaging and morphometrics using 320-ADCT are powerful tools for efficiently and reliably observing and measuring the pharynx and larynx. KEYWORDS: anatomy, larynx, pharynx, multidetector computed tomography, deglutition, deglutition disorders Accepted for publication 21 March 2015
Until now, cadaveric, stereoendoscopic and radiographic measurements methods have been used. Although, cadaveric measurement has been employed extensively, it cannot be of reference to a dynamic movement in living subjects directly because of lack of natural muscle tone and fixation of tissues. Stereoendoscopic methods have procedural problems; there doi: 10.1111/joor.12298
LARYNX AND PHARYNX REVEALED BY MUTLIDETECTOR COMPUTED TOMOGRAPHY is insufficient field of view (FOV) and lack of quantitative data due to variations in lens-to-field distance. Moreover, those two methods have mainly focused on the measurement of laryngeal structures and rarely applied to the measurement of pharyngeal structures. Measurements using radiographic images are relatively new compared with the other two. It started from lateral radiography, and later, computed tomographic images (CT) and magnetic resonance imaging (MRI) were used. With technological advances, threedimensional images became available for volumetric analysis and accurate visualisation of structures. To date, insufficient data have been gathered to compare the data acquired by each method. Measurement varies based on measurement method. From cadaveric studies, the length of the vocal folds was between 230 and 282 mm in men and between 163 and 193 mm in women and the membranous portion of vocal folds measured between 145 and 166 mm in men and between 94 and 123 mm in women (3–7). Stereoendoscopic measurement showed that the membranous portion of vocal folds was between 110 and 180 mm in men and between 70 and 140 mm in women (8–12). Lateral X-ray images showed that the total vocal fold length was between 160 and 266 mm in men and from 130 to 220 mm in women (13, 14). CT images showed that the membranous portion of vocal fold was 148 mm and total vocal fold length was 258 mm in men (15). Measurements obtained with CT and MRI showed that the pharyngeal volume was between 130861 and 415570 mm3 in men and between 72766 and 321430 mm3 in women; pharyngeal length was between 526 and 761 mm in men and between 432 and 681 mm in women (1, 2, 16–20). These studies reported large differences among individuals, although the sources of this variability remain unclear. Several studies suggest that gender and age have an effect on vocal fold length and pharyngeal length (1, 2, 4, 8, 19). However, few have examined the interactions of height, age and gender. A 320-row area detector computed tomography (hereinafter called ‘320-ADCT’) is equipped with 320 rows of 05 mm detectors along the body axis. Three-dimensional images of pharynx and larynx can be acquired in 035 s with a non-helical scan. As the images are isotropic, volume data in which coronal © 2015 John Wiley & Sons Ltd
(X), sagittal (Y) and axial (Z) sections have almost same high-resolution structures (yielding voxels of 047 9 047 9 05 mm, volume of 011 mm3) can be depicted on any arbitrary cross section in any orientation (21). This feature provides accurate quantitative measurements and has great potential for describing the anatomical characteristics of pharynx and larynx. This research aimed to (i) measure the pharynx and larynx using the three-dimensional images acquired by 320-ADCT, thus providing the normative reference standard; and (ii) investigate the effects of age, gender and height on morphology of the larynx and pharynx.
Methods Fifty-four healthy volunteers (30 male and 24 female), 41 15 years old (range 23–77 years) participated in the study. Their average heights were 170 5 cm (male) and 156 5 cm (female). All subjects provided informed consent for participation after thorough explanation of the purpose and procedure and the risk of radiation exposure. This study protocol was approved by the Institutional Review Board at our university. Procedure A 320-ADCT* was used. A chair specially designed for examination of swallowing with CT† was placed on the opposite side of the CT table. Subjects were seated comfortably on the chair, reclining at an angle of 45 degrees with the head and neck in neutral position (22). Then each subject underwent one single-phase volume scan with instructions to relax and breath quietly (tidal breathing). We did not instruct the subjects to exhale or inhale, close or open their mouths, nor clench their teeth. CT data acquisition and processing The scanning parameters were set at a FOV of 240 mm, tube voltage/current = 120 kV/30 mA and slice thickness of 05 mm. The scan time was 035 s,
*Aquilion ONE; Toshiba Medical Systems Corporation, Otawara, Tochigi, Japan. † CT reclining chair; TomeiBrace Co., Ltd., Seto and Aska Corp., Kariya, Japan.
671
672
Y . I N A M O T O et al. which provided reduction in the radiation dose while limiting motion artefacts. Single-phase volume scanning was performed with 035 s duration for one rotation of the tube. The CT dose index (CTDI) and dose length product (DLP) with these scanning parameters were 19 mGy and 308 mGy cm, respectively. The scanning range was 160 mm from the skull base to the upper oesophagus. CT images were restructured using a full reconstruction method. Multiplanar reconstruction (hereinafter, ‘MPR’) and 3D-CT images were created using a volume-rendering method. We chose 300 HU or less to depict air in the pharyngeal and laryngeal cavities. This value was determined by the raters, who were experienced with diagnostic 3DCT images. First, MPR images under soft tissues conditions and 3D-CT images were compared. The raters selected the most appropriate cut-off value in which images of the on 3D-CT MPR images were the closest possible match. Measurements Length and volume were measured three dimensionally through software installed on a CT scanner. Wherever possible, the measurements were based on standard anatomical nomenclature. The operational definition of each measurement is shown in Table 1 and Fig. 1. Volume of oropharynx (V-OP, Fig. 1-1b) – It is bounded superiorly by the palatal plane passing through the anterior and posterior nasal spines (ANS and PNS), and parallel to the infraorbital line. The anterior boundary was a plane passing through the PNS (the junction of the hard and soft palate) and perpendicular to the palatal plane. The inferior boundary was a plane passing through the lowermost point of the valleculae and parallel to the palatal plane. The lateral margins were defined by the lateral walls of the pharynx. Volume of the laryngeal and hypopharyngeal cavities (V-LH, Fig. 1-1b) – It is not feasible to reliably mark the soft tissue border between the hypopharynx and the supraglottic larynx with CT. Thus, we measured a combined volume of the two regions. The VLP volume was bounded superiorly by the inferior boundary of V-OP and inferiorly by the superior surface of the true vocal folds (for the laryngeal part) and the inferior edge of the pyriform sinus (for the pharyngeal part).
Length of pharynx (L-P, Fig. 1-1a) – It is the distance from the PNS to the superior surface of the true vocal folds along a line perpendicular to the palatal plane. Width of pharynx (W-P, Fig. 1-1c–e) – it is the maximum side-to-side breadth of the pharynx. This location was identified on 3D-images and measured on axial MPR images. Distance between the bottom of the valleculae and the vocal folds (D-VV, Fig. 1-2a) – It is the distance from the inferior limit of the valleculae to the superior surface of the true vocal folds, measured in midsagittal slices. Anteroposterior diameter of the true vocal folds (L-TVF, Fig. 1-2b–c) – It is the distance between the anterior commissure and the most posterior boundary of the folds, measured in axial sections. To capture the vocal folds, the angle was adjusted first in midsagittal sections and then measured on axial sections using as landmarks the vocal processes. Volume of pyriform sinus (V-PS, Fig. 1-3a–d) – It is defined as a space surrounded by the pharyngeal wall (posteriorly and laterally) and the aryepiglottic folds medially. The superior boundary was a plane passing through the superior end of the aryepiglottic folds and parallel to the upper surface of the true vocal folds. Length of the pyriform sinus (L-PS, Fig. 1-3c) – It is defined as the maximum distance between the most inferior point of the pyriform sinus and its superior boundary. The V-PS and L-PS were averaged between left and right sides. Statistical analysis The simple comparison between men and women for each measurement was performed using t-test. Simple regression analysis was performed to see the effect of height and age. To explore the interaction of gender, height and age, multiple regression analysis was used with the independent values of gender (1 = male, 0 = female) and height. All the results are presented as mean standard deviation, with P < 005 being the threshold for statistical significance. The statistical analysis was performed using SPSS‡.
‡
IBM SPSS statistics 21; IBM, Chicago, IL, USA. © 2015 John Wiley & Sons Ltd
LARYNX AND PHARYNX REVEALED BY MUTLIDETECTOR COMPUTED TOMOGRAPHY Table 1. Definition of measurements Definition V-OP
Volume of oropharyngeal cavity
V-LH
Volume of laryngeal and hypopharyngeal cavities
L-P
Length of pharynx
W-P D-VV L-TVF V-PS
Width of pharynx Distance between the bottom of valleculae and vocal folds Length of true vocal folds Volume of pyriform sinus
L-PS
Length of pyriform sinus
Bounded superiorly by the palatal plane passing through the anterior nasal spine (ANS) and posterior nasal spine (PNS) and parallel to the infraorbital line. The anterior boundary was a passing through the PNS and perpendicular to the palatal plane. The inferiorly boundary was a plane passing through the lowermost point of the valleculae and parallel to the palatal plane Bounded superiorly by the inferior boundary of V-OP and inferiorly by the superior surface of the true vocal folds (for the laryngeal part) and the inferior edge of the pyriform sinus (for the pharyngeal part) Distance from the PNS to the superior surface of true vocal folds along a line perpendicular to the palatal plane Maximum side-to-side breadth of the pharynx Distance from the inferior limit of the valleculae to the superior surface of the true vocal folds Distance between the anterior commissure and the most posterior boundary of the folds Surrounded by the pharyngeal wall (posteriorly and laterally) and the aryepiglottic folds medially. The superior boundary was a plane passing through the superior end of the aryepiglottic folds and parallel to the upper surface of the true vocal folds Maximum distance between the most inferior point of the pyriform sinus and its superior boundary
Results
Discussion
Multiplanar reconstruction images and 3D-CT images of targeted structures were created for measurement. The average values for men and women in all the measurements are shown in Table 2. Strong genderrelated differences were identified. Values were significantly larger for men than women for every measurement (P < 005). Even after normalising for height, the pharyngeal and laryngeal values were significantly larger in men than in women, with the exception of the two pyriform sinus measures (V-PS and L-PS). Simple regression analysis and the scatter plots by height identified strong height-related differences (Fig. 2). Values increased significantly with height for all measurements (P < 005). On the other hand, age had relatively little effect (Fig. 2). Only V-LH became significantly smaller with increasing age. To understand the total effect of gender, height and age, multiple regression analysis was performed (Table 3). Gender effect was identified in L-P, D-VV and L-TVF (P < 005), height effect was identified in V-LH and W-P (P < 005), and age effect was identified in V-LH and L-P (P < 005).
Morphological measurement of the pharynx and larynx has been widely performed using cadaveric, stereoendoscopic, radiographic, CT and MRI methods. However, there are variability of those reported values due to the methodological limitations, and therefore, the referential values of for each part of the pharynx and larynx has not been well established. Although CT and MRI methods provide the volumetric analysis with three-dimensional images, limitations of scanning range and image acquisition time have made it difficult to acquire consistent results between studies based on the same measurement definitions. The 320-ADCT, used in this study, has excellent space and time resolution. The wide scanning range enabled image acquisition of pharynx and larynx at the same time without the image gaps which are typical using helical CT. In this study, volume and linear measures of the pharynx and larynx were performed with excellent resolution in three dimensions. Use of 320-ADCT for structural measurements has several advantages. First, high spatial resolution (05 mm slice thickness) enabled precise depiction of the targeted structures. The relative error of distance
© 2015 John Wiley & Sons Ltd
673
674
Y . I N A M O T O et al.
1a
1b
1c
2a
2b
2c
3a
3b
3c
1d
1e
3d
Fig. 1. Definition of structural measurements using 3D-CT and MPR images. (1) V-OP: volume of oropharyngeal cavity (1-b), V-LH: volume of laryngeal and hypopharyngeal cavity (1-b), L-P: length of pharynx (1-a), W-P: width of pharynx (1-c–e). W-P was identified from 3D images (1-c) of V-LH and depicted on axial MPR images (1-e). (2) D-VV: distance between the bottom of the valleculae and the vocal folds (2-a), L-TVF: length of true vocal folds (2-b, c). TVF was identified from midsagittal sections (2-b) and depicted on axial sections with the landmark of vocal process (2-c). (3) V-PS: volume of pyriform sinus, L-PS: length of pyriform sinus. Area coloured yellow was pyriform sinus. 3-a: anterior, 3-b, d: lateral, 3-c: posterior view of 3D-CT images.
measurement was within 034% the actual measured values, indicating high precision (23). In addition, the short scanning time minimises the radiation dose as well as the motion artefact during the scan (such as the one accompanied by breathing). The radiation dose, CTDIvol 08 mGy and DLPmGy.cm121 mGy, was notably low compared with the routine CT dose of CTDIvol 60 mG and DLPmGy.cm1050 mGy (23). Cone-beam computed tomography would provide lower radiation exposure; however, the longer scanning time and low resolution of soft tissue would preclude the high quality of imaging required for morphometric analysis. Moreover, construction of 3D images and measurements were readily performed in the present study. These factors support the application of 320-ADCT for morphometric analysis and formulation of reference values for the pharynx and larynx.
All measurements differed significantly with gender and height. As men were significantly taller than women, the larger values in men could be explained by height effects. However, the multiple regression analysis confirmed that there were indeed gender differences in some values after adjusting for height. This gender effect was noted especially in laryngeal measurement values. Gender differences in laryngeal anatomy are well known. The differences in length of the true vocal cords are believed to account for the differences in vocal pitch range between genders. We confirmed significantly longer D-VV and L-TVF in men compared with women as reported previously. Regarding the effect of age, there were significant differences in volume of larynx and hypopharynx (VLH) and in length of pharynx (L-P). The V-LH decreased significantly and the L-P increased significantly with ageing. In a previous study, © 2015 John Wiley & Sons Ltd
LARYNX AND PHARYNX REVEALED BY MUTLIDETECTOR COMPUTED TOMOGRAPHY Table 2. Comparison of each parameter in men and women (independent t-test) Women n = 24 Mean (s.d.)
P value Difference
15
15
10
10 5
135 55 760 355 259 167 10 166
(43) (15) (67) (39) (36) (20) (05) (40)
Oral Rehabilitation
Journal of Oral Rehabilitation 2015 42; 670–677
Anatomy of the larynx and pharynx: effects of age, gender and height revealed by multidetector computed tomography Y. INAMOTO*†, E. SAITOH*, S. OKADA†, H. KAGAYA*, S. SHIBATA*, M. BABA‡, K. ONOGI*, S. HASHIMOTO§, K. KATADA¶, P. WATTANAPAN** & J. B. PALMER†† *Department of Rehabilitation Medicine, School of Medicine, Fujita Health University, Aichi, †Faculty of Rehabilitation, School of Health Sciences, Fujita Health University, Aichi, ‡Japanese Red Cross Ashikaga Hospital, Tochigi, §Department of Hygiene, Fujita Health University, Aichi, ¶Department of Radiology, Fujita Health University, Aichi, Japan, **Institute of Medicine, Suranaree University of Technology, Muang, Nakornratchasima, Thailand and
††
Department of Physical Medicine and Rehabilitation, Department of Otolaryngology-Head and Neck Surgery, and
Center for Functional Anatomy and Evolution, Johns Hopkins University, Baltimore, Maryland, USA
SUMMARY Although oropharyngeal and laryngeal structures are essential for swallowing, the threedimensional (3D) anatomy is not well understood, due in part to limitations of available measuring techniques. This study uses 3D images acquired by 320-row area detector computed tomography (‘320ADCT’), to measure the pharynx and larynx and to investigate the effects of age, gender and height. Fifty-four healthy volunteers (30 male, 24 female, 23–77 years) underwent one single-phase volume scan (035 s) with 320-ADCT during resting tidal breathing. Six measurements of the pharynx and two of larynx were performed. Bivariate statistical methods were used to analyse the effects of gender, age and height on these measurements. Length and volume were significantly larger for men than for women for every measurement (P < 005) and increased with height (P < 005). Multiple regression analysis was performed to understand the interactions of gender, height and
Background Gender and age are likely to influence oropharyngeal and laryngeal structure (1, 2). However, the normative reference values of oropharyngeal and laryngeal structures are not yet well established, due to the limitations of the available measuring techniques. © 2015 John Wiley & Sons Ltd
age. Gender, height and age each had significant effects on certain values. The volume of the larynx and hypopharynx was significantly affected by height and age. The length of pharynx was associated with gender and age. Length of the vocal folds and distance from the valleculae to the vocal folds were significantly affected by gender (P < 005). These results suggest that age, gender and height have independent and interacting effects on the morphology of the pharynx and larynx. Three-dimensional imaging and morphometrics using 320-ADCT are powerful tools for efficiently and reliably observing and measuring the pharynx and larynx. KEYWORDS: anatomy, larynx, pharynx, multidetector computed tomography, deglutition, deglutition disorders Accepted for publication 21 March 2015
Until now, cadaveric, stereoendoscopic and radiographic measurements methods have been used. Although, cadaveric measurement has been employed extensively, it cannot be of reference to a dynamic movement in living subjects directly because of lack of natural muscle tone and fixation of tissues. Stereoendoscopic methods have procedural problems; there doi: 10.1111/joor.12298
LARYNX AND PHARYNX REVEALED BY MUTLIDETECTOR COMPUTED TOMOGRAPHY is insufficient field of view (FOV) and lack of quantitative data due to variations in lens-to-field distance. Moreover, those two methods have mainly focused on the measurement of laryngeal structures and rarely applied to the measurement of pharyngeal structures. Measurements using radiographic images are relatively new compared with the other two. It started from lateral radiography, and later, computed tomographic images (CT) and magnetic resonance imaging (MRI) were used. With technological advances, threedimensional images became available for volumetric analysis and accurate visualisation of structures. To date, insufficient data have been gathered to compare the data acquired by each method. Measurement varies based on measurement method. From cadaveric studies, the length of the vocal folds was between 230 and 282 mm in men and between 163 and 193 mm in women and the membranous portion of vocal folds measured between 145 and 166 mm in men and between 94 and 123 mm in women (3–7). Stereoendoscopic measurement showed that the membranous portion of vocal folds was between 110 and 180 mm in men and between 70 and 140 mm in women (8–12). Lateral X-ray images showed that the total vocal fold length was between 160 and 266 mm in men and from 130 to 220 mm in women (13, 14). CT images showed that the membranous portion of vocal fold was 148 mm and total vocal fold length was 258 mm in men (15). Measurements obtained with CT and MRI showed that the pharyngeal volume was between 130861 and 415570 mm3 in men and between 72766 and 321430 mm3 in women; pharyngeal length was between 526 and 761 mm in men and between 432 and 681 mm in women (1, 2, 16–20). These studies reported large differences among individuals, although the sources of this variability remain unclear. Several studies suggest that gender and age have an effect on vocal fold length and pharyngeal length (1, 2, 4, 8, 19). However, few have examined the interactions of height, age and gender. A 320-row area detector computed tomography (hereinafter called ‘320-ADCT’) is equipped with 320 rows of 05 mm detectors along the body axis. Three-dimensional images of pharynx and larynx can be acquired in 035 s with a non-helical scan. As the images are isotropic, volume data in which coronal © 2015 John Wiley & Sons Ltd
(X), sagittal (Y) and axial (Z) sections have almost same high-resolution structures (yielding voxels of 047 9 047 9 05 mm, volume of 011 mm3) can be depicted on any arbitrary cross section in any orientation (21). This feature provides accurate quantitative measurements and has great potential for describing the anatomical characteristics of pharynx and larynx. This research aimed to (i) measure the pharynx and larynx using the three-dimensional images acquired by 320-ADCT, thus providing the normative reference standard; and (ii) investigate the effects of age, gender and height on morphology of the larynx and pharynx.
Methods Fifty-four healthy volunteers (30 male and 24 female), 41 15 years old (range 23–77 years) participated in the study. Their average heights were 170 5 cm (male) and 156 5 cm (female). All subjects provided informed consent for participation after thorough explanation of the purpose and procedure and the risk of radiation exposure. This study protocol was approved by the Institutional Review Board at our university. Procedure A 320-ADCT* was used. A chair specially designed for examination of swallowing with CT† was placed on the opposite side of the CT table. Subjects were seated comfortably on the chair, reclining at an angle of 45 degrees with the head and neck in neutral position (22). Then each subject underwent one single-phase volume scan with instructions to relax and breath quietly (tidal breathing). We did not instruct the subjects to exhale or inhale, close or open their mouths, nor clench their teeth. CT data acquisition and processing The scanning parameters were set at a FOV of 240 mm, tube voltage/current = 120 kV/30 mA and slice thickness of 05 mm. The scan time was 035 s,
*Aquilion ONE; Toshiba Medical Systems Corporation, Otawara, Tochigi, Japan. † CT reclining chair; TomeiBrace Co., Ltd., Seto and Aska Corp., Kariya, Japan.
671
672
Y . I N A M O T O et al. which provided reduction in the radiation dose while limiting motion artefacts. Single-phase volume scanning was performed with 035 s duration for one rotation of the tube. The CT dose index (CTDI) and dose length product (DLP) with these scanning parameters were 19 mGy and 308 mGy cm, respectively. The scanning range was 160 mm from the skull base to the upper oesophagus. CT images were restructured using a full reconstruction method. Multiplanar reconstruction (hereinafter, ‘MPR’) and 3D-CT images were created using a volume-rendering method. We chose 300 HU or less to depict air in the pharyngeal and laryngeal cavities. This value was determined by the raters, who were experienced with diagnostic 3DCT images. First, MPR images under soft tissues conditions and 3D-CT images were compared. The raters selected the most appropriate cut-off value in which images of the on 3D-CT MPR images were the closest possible match. Measurements Length and volume were measured three dimensionally through software installed on a CT scanner. Wherever possible, the measurements were based on standard anatomical nomenclature. The operational definition of each measurement is shown in Table 1 and Fig. 1. Volume of oropharynx (V-OP, Fig. 1-1b) – It is bounded superiorly by the palatal plane passing through the anterior and posterior nasal spines (ANS and PNS), and parallel to the infraorbital line. The anterior boundary was a plane passing through the PNS (the junction of the hard and soft palate) and perpendicular to the palatal plane. The inferior boundary was a plane passing through the lowermost point of the valleculae and parallel to the palatal plane. The lateral margins were defined by the lateral walls of the pharynx. Volume of the laryngeal and hypopharyngeal cavities (V-LH, Fig. 1-1b) – It is not feasible to reliably mark the soft tissue border between the hypopharynx and the supraglottic larynx with CT. Thus, we measured a combined volume of the two regions. The VLP volume was bounded superiorly by the inferior boundary of V-OP and inferiorly by the superior surface of the true vocal folds (for the laryngeal part) and the inferior edge of the pyriform sinus (for the pharyngeal part).
Length of pharynx (L-P, Fig. 1-1a) – It is the distance from the PNS to the superior surface of the true vocal folds along a line perpendicular to the palatal plane. Width of pharynx (W-P, Fig. 1-1c–e) – it is the maximum side-to-side breadth of the pharynx. This location was identified on 3D-images and measured on axial MPR images. Distance between the bottom of the valleculae and the vocal folds (D-VV, Fig. 1-2a) – It is the distance from the inferior limit of the valleculae to the superior surface of the true vocal folds, measured in midsagittal slices. Anteroposterior diameter of the true vocal folds (L-TVF, Fig. 1-2b–c) – It is the distance between the anterior commissure and the most posterior boundary of the folds, measured in axial sections. To capture the vocal folds, the angle was adjusted first in midsagittal sections and then measured on axial sections using as landmarks the vocal processes. Volume of pyriform sinus (V-PS, Fig. 1-3a–d) – It is defined as a space surrounded by the pharyngeal wall (posteriorly and laterally) and the aryepiglottic folds medially. The superior boundary was a plane passing through the superior end of the aryepiglottic folds and parallel to the upper surface of the true vocal folds. Length of the pyriform sinus (L-PS, Fig. 1-3c) – It is defined as the maximum distance between the most inferior point of the pyriform sinus and its superior boundary. The V-PS and L-PS were averaged between left and right sides. Statistical analysis The simple comparison between men and women for each measurement was performed using t-test. Simple regression analysis was performed to see the effect of height and age. To explore the interaction of gender, height and age, multiple regression analysis was used with the independent values of gender (1 = male, 0 = female) and height. All the results are presented as mean standard deviation, with P < 005 being the threshold for statistical significance. The statistical analysis was performed using SPSS‡.
‡
IBM SPSS statistics 21; IBM, Chicago, IL, USA. © 2015 John Wiley & Sons Ltd
LARYNX AND PHARYNX REVEALED BY MUTLIDETECTOR COMPUTED TOMOGRAPHY Table 1. Definition of measurements Definition V-OP
Volume of oropharyngeal cavity
V-LH
Volume of laryngeal and hypopharyngeal cavities
L-P
Length of pharynx
W-P D-VV L-TVF V-PS
Width of pharynx Distance between the bottom of valleculae and vocal folds Length of true vocal folds Volume of pyriform sinus
L-PS
Length of pyriform sinus
Bounded superiorly by the palatal plane passing through the anterior nasal spine (ANS) and posterior nasal spine (PNS) and parallel to the infraorbital line. The anterior boundary was a passing through the PNS and perpendicular to the palatal plane. The inferiorly boundary was a plane passing through the lowermost point of the valleculae and parallel to the palatal plane Bounded superiorly by the inferior boundary of V-OP and inferiorly by the superior surface of the true vocal folds (for the laryngeal part) and the inferior edge of the pyriform sinus (for the pharyngeal part) Distance from the PNS to the superior surface of true vocal folds along a line perpendicular to the palatal plane Maximum side-to-side breadth of the pharynx Distance from the inferior limit of the valleculae to the superior surface of the true vocal folds Distance between the anterior commissure and the most posterior boundary of the folds Surrounded by the pharyngeal wall (posteriorly and laterally) and the aryepiglottic folds medially. The superior boundary was a plane passing through the superior end of the aryepiglottic folds and parallel to the upper surface of the true vocal folds Maximum distance between the most inferior point of the pyriform sinus and its superior boundary
Results
Discussion
Multiplanar reconstruction images and 3D-CT images of targeted structures were created for measurement. The average values for men and women in all the measurements are shown in Table 2. Strong genderrelated differences were identified. Values were significantly larger for men than women for every measurement (P < 005). Even after normalising for height, the pharyngeal and laryngeal values were significantly larger in men than in women, with the exception of the two pyriform sinus measures (V-PS and L-PS). Simple regression analysis and the scatter plots by height identified strong height-related differences (Fig. 2). Values increased significantly with height for all measurements (P < 005). On the other hand, age had relatively little effect (Fig. 2). Only V-LH became significantly smaller with increasing age. To understand the total effect of gender, height and age, multiple regression analysis was performed (Table 3). Gender effect was identified in L-P, D-VV and L-TVF (P < 005), height effect was identified in V-LH and W-P (P < 005), and age effect was identified in V-LH and L-P (P < 005).
Morphological measurement of the pharynx and larynx has been widely performed using cadaveric, stereoendoscopic, radiographic, CT and MRI methods. However, there are variability of those reported values due to the methodological limitations, and therefore, the referential values of for each part of the pharynx and larynx has not been well established. Although CT and MRI methods provide the volumetric analysis with three-dimensional images, limitations of scanning range and image acquisition time have made it difficult to acquire consistent results between studies based on the same measurement definitions. The 320-ADCT, used in this study, has excellent space and time resolution. The wide scanning range enabled image acquisition of pharynx and larynx at the same time without the image gaps which are typical using helical CT. In this study, volume and linear measures of the pharynx and larynx were performed with excellent resolution in three dimensions. Use of 320-ADCT for structural measurements has several advantages. First, high spatial resolution (05 mm slice thickness) enabled precise depiction of the targeted structures. The relative error of distance
© 2015 John Wiley & Sons Ltd
673
674
Y . I N A M O T O et al.
1a
1b
1c
2a
2b
2c
3a
3b
3c
1d
1e
3d
Fig. 1. Definition of structural measurements using 3D-CT and MPR images. (1) V-OP: volume of oropharyngeal cavity (1-b), V-LH: volume of laryngeal and hypopharyngeal cavity (1-b), L-P: length of pharynx (1-a), W-P: width of pharynx (1-c–e). W-P was identified from 3D images (1-c) of V-LH and depicted on axial MPR images (1-e). (2) D-VV: distance between the bottom of the valleculae and the vocal folds (2-a), L-TVF: length of true vocal folds (2-b, c). TVF was identified from midsagittal sections (2-b) and depicted on axial sections with the landmark of vocal process (2-c). (3) V-PS: volume of pyriform sinus, L-PS: length of pyriform sinus. Area coloured yellow was pyriform sinus. 3-a: anterior, 3-b, d: lateral, 3-c: posterior view of 3D-CT images.
measurement was within 034% the actual measured values, indicating high precision (23). In addition, the short scanning time minimises the radiation dose as well as the motion artefact during the scan (such as the one accompanied by breathing). The radiation dose, CTDIvol 08 mGy and DLPmGy.cm121 mGy, was notably low compared with the routine CT dose of CTDIvol 60 mG and DLPmGy.cm1050 mGy (23). Cone-beam computed tomography would provide lower radiation exposure; however, the longer scanning time and low resolution of soft tissue would preclude the high quality of imaging required for morphometric analysis. Moreover, construction of 3D images and measurements were readily performed in the present study. These factors support the application of 320-ADCT for morphometric analysis and formulation of reference values for the pharynx and larynx.
All measurements differed significantly with gender and height. As men were significantly taller than women, the larger values in men could be explained by height effects. However, the multiple regression analysis confirmed that there were indeed gender differences in some values after adjusting for height. This gender effect was noted especially in laryngeal measurement values. Gender differences in laryngeal anatomy are well known. The differences in length of the true vocal cords are believed to account for the differences in vocal pitch range between genders. We confirmed significantly longer D-VV and L-TVF in men compared with women as reported previously. Regarding the effect of age, there were significant differences in volume of larynx and hypopharynx (VLH) and in length of pharynx (L-P). The V-LH decreased significantly and the L-P increased significantly with ageing. In a previous study, © 2015 John Wiley & Sons Ltd
LARYNX AND PHARYNX REVEALED BY MUTLIDETECTOR COMPUTED TOMOGRAPHY Table 2. Comparison of each parameter in men and women (independent t-test) Women n = 24 Mean (s.d.)
P value Difference
15
15
10
10 5
135 55 760 355 259 167 10 166
(43) (15) (67) (39) (36) (20) (05) (40)
