The Effect of Posture on Upper Airway Dimensions in Normal Subjects and in Patients with the Sleep Apnea/Hypopnea Syndrome1 , 2
N. YILDIRIM, M. F. FITZPATRICK, K. F. WHYTE, R. JALLEH, A. J. A. WIGHTMAN, and N. J. DOUGLAS
Introduction
The obstructive sleep apnea/hypopnea syndrome (SAHS) is caused by repetitive narrowing and occlusion of the upper airway during sleep (1). Patients with the syndrome have narrow (2) and floppy (3) upper airways during wakefulness in comparison with normal subjects. The efficacy of upper airway surgery depends upon the location of airways obstruction, the response to uvulopalatopharyngoplasty being influenced by upper airway dimensions (4) and the site and extent of maximal airway narrowing (5). Some techniques currently used to determine the site of maximal airway narrowing in awake patients are carried out in the erect posture, and others are carried out with the patient supine. Lateral radiography cephalometry (6, 7)- and acoustic reflection (2) are carried out with the patient erect, whereas both computerized tomographic (8, 9) and magnetic resonance (10)imaging are carried out in supine patients. Clarification of the effect of posture on upper airway dimensions is thus important in allowing comparison between the results obtained by these different techniques. In addition, knowledge of the effect of posture on the upper airway is of interest as many patients with untreated SAHS report that they sleep better sitting upright than when lying in bed. We have, therefore, assessed the effect of posture on upper airway size in both normal nonsnoring subjects and in patients with the obstructive SAHS. The only technique that can assess the dimensions of the retropalatal and more distal upper airway in both erect and supine posture is cephalometry, which was thus utilized in this study. Methods Subjects Thirty-three nonsnoring normal subjects (15 men; mean age, 45 ± SD 16 yr; body mass index, 26 ± 5 kg/m) and 29 patients (26 men; mean age, 54 ± 11 yr; BMI, 32 ± 7 kg/m),
SUMMARY The effect of posture on upper airway dimensions was assessed for two reasons. First, some patients with untreated sleep apnea/hypopnea syndrome (SAHS) report they sleep better sitting upright. Second, to allow comparison of the differing techniques used to determine the site of maxlmel airway narrowing In ewake patients with SAHS, as some are carried out In the erect and others In the supine posture. Lateral cephalometry was therefore carried out In 33 nonsnorlng normal subjects and In 29 patients with obstructive SAHS (mean apneas plus hypopneas, 46 per hour; range, 17to 103). In both normal subjects and patients, uvular width was Increased (p < 0.05) In the supine posture, and this was associated with significant narrowing of the retropalatalalrway In the patients with SAHS (erect, 5.0 ± SO 2.6 mm; supine, 3.6 ± 2.8 mm; p < 0.01).In both normal subjects and patients, the retroglossal hypopharynx widened (p < 0.05) In the supine posture (e.g., in patients with SAHS, posterior airway space was: erect, 11.5 ± 4.5 mm; supine, 13A ±4.8 mm; p = 0.003). In the supine posture there was anterior movement of the hyoid and neck flexion In both groups. However, a study of the effect of neck flexion In the erect posture showed that neck flexion produced no changes In alrwey caliber. Thus, posture Is an Important detarmlnant of upper AM REV RESPIR DIS 1991; 144:845-S47 airway dimensions.
The normal subjects were recruited from relatives of patients with SAHS and from patients referred with nonrespiratory sleep complaints. All the patients with SAHS had been referred for evaluation because of clinical features of the SAHS. Each underwent polysomnography where electroencephalogram, electrooculogram, and electromyogram were recorded using our standard electrode placement (11), respiration was monitored by an inductance plethysmograph (Respitraces; Ambulatory Monitoring, Ardsley, NJ), airflow was measured by oronasal thermocouples, and oxygen saturation was measured by an Ohmeda 3700 oximeter (Ohmeda, Boulder, CO). Both sleep (12) and respiratory patterns (13, 14)were scored by standard criteria. The mean apnea plus hypopnea index of the normal subjects was 5 ± 3 (range, 0 to 11)per hour and of the patients with SAHS, 46 ± 27 (range, 17 to 103) per hour.
Radiology Cephalometric roentgenograms were carried out with the subjects both seated and supine with the patient facing at 90° to the X-ray beam. The patients were directed to gaze forward for the erect films and upwards for the supine films, holding their heads in a natural position. They were asked to close their mouths with their normal resting occlusion and their lips together, to allow the tongue to relax onto the floor of the mouth and not to swallow during the exposures. Radiographs were taken with the patient exhaling slowly
from a deep breath. The X-ray cone was positioned exactly 5 feet from the radiographic film and the film was placed against the left side of the face. Twenty-three cephalometric measurements were obtained from each subject in both erect and supine postures (figure I). These included distances between each of the following: anterior nasal spine (ANS), posterior nasal spine (PNS), gonion (Go), and gnathion (Gn). In addition, the distances from the midpoint of the sella (S) to the gonion and the hyoid (H) were measured. The distances from H to the mandibular plane (MP) and H to the intersection of a line along the posterior border of the mandibular process and the temporal bone (articulare; AR), H-Go, and Go-AR were measured. The greatest posterior convexity of the soft palate (uvular protrusion; UP) and the tip of the uvula (UT) were identified, and uvular length (UL; distance from PNS to UT) and uvular width (diameter of the soft palate at the widest point; UW) were measured as were distances from the posteri(Received in original form March 1, 1990 and in revised form March 4, 1991) 1 From the Respiratory Medicine Unit, Department of Medicine and Department of Radiology, City Hospital, Edinburgh, United Kingdom. , Correspondence should be addressed to N. J. Douglas, Respiratory Medicine Unit, Department of Medicine and Department of Radiology, City Hospital, Edinburgh, ENlO, 5SB, UK.
845
846
YILDIRIM, FITZPATRICK, WHYTE, JALLEH, WIGHTMAN, AND DOUGLAS
TABLE 1 CEPHALOMETRIC MEASUREMENTS IN NORMAL SUBJECTS AND IN PATIENTS WITH SAHS
l
Normal SUbjects Variables
ANS
Ar
H
::J Fig. 1. Diagrammatic representation of cephalometric measurementsmade.Fordefinition of abbreviations, see table 1 and texl.
or pharyngeal wall on a line from B (supramentale, the deepest point of the contour of the mandibular alveolus between the pogonion and the central incisor) through Go, posterior pharyngeal wall (PhW) to UP, UT, H, and PNS. The posterior airway space (PAS) the linear measurement between the base of the tongue (TB) and PHW on a line from point B through Go. PhW-TB (parallel to the mandibular plane) and PNS-TB were also measured. Two angles were measured: Go-Gn-H (the angle measured from Go to Gn to H) and the angle of the neck (the acute angle created by the intersection of a line drawn through the anterior border of the second cervical vertebra [C2l with a line drawn through the anterior border ofthe fourth cervical vertebra [C4]). Cephalometry films were also repeated in the erect posture with slight neck flexion in 19 of the normal subjects and in six patients with SAHS. This was to examine whether a change in neck angle could effect airway dimensions during supine cephalometry.
Statistical Analysis Comparisons weremade using Student's t test, with the Bonferroni correction for multiple comparisons when appropriate.
Results
The means and SD of each of the linear and angular measurements in the normal subjects and in the patients with SAHS are presented in table 1. In the normal subjects, .uvular width increased in the supine posture, but there was no significant change in uvula to pharyngeal wall dimensions. There was wideningof the retroglossal airwayas evidenced by a significant increase in the tongue base to pharyngeal wall dimension (p < 0.03). There was also an increase
UL, mm UW, mm UP-PhW, mm UT-PhW, mm TB-PhW. mm H-PhW, mm PAS, mm H-MP, mm Go-Go-H, degree Ang Nck, degree
Erect
46.8 10.2 6.0 8.7 10.0 34.8 11.9 23.2 24.6 16.5
± ± ± ±
± ± ±
± ±
±
6.2 2.1 3.4 3.1 4.1 4.0 4.6 7.6 8.3 9.7
Patients with SAHS
Supine
48.8 11.2 5.4 8.4 11.9 37.6 12.7 22.4 22.8 13.0
± ± ± ± ± ± ± ± ± ±
5.5 2.4t 3.1 3.2 4.5t 4.6t 4.7 6.7 7.5 9.6t
Erect
Supine
51.6 ± 5.8 11.7±2.6 5.0 ± 2.6 9.5 ± 3.4 10.2 ± 4.0 39.5 ± 6.5 11.5 ± 4.5 31.2 ± 6.7 33.4 ± 6.7 19.0 ± 10.0
52.4 13.0 3.6 7.7 11.4 41.3 13.4 31.2 34.1 14.0
± ± ± ± ± ± ± ± ± ±
7.0§ 2.6:j:§ 2.8* 5.1t 4.7 6.3*§ 4.8* 6.6§ 8.3§ 8.3*
Definition of abbreviations: Ul ; uvular length; UW ; uvular width; UP-PhW ; uvular protrusion to pharyngeal wall; UT-PhW ; uvular tip to pharyngeal wall; TB-PhW = base of tongue to PhW; H-PhW ; hyoid to PhW; PAS = posterior airway space; H-MP = hyoid to mandibular plane; Go-Gn-H ; gonion to gnathion to hyoid; Ang Nck ; angle of the neck. • Values are mean SO. t p < 0.05, erect versus supine. :1= p < 0.01, erect versus supine. § p < O.D1, supine normal SUbjects versus supine patients with SAHS.
in the hyoid to pharyngeal wall distance in the supine posture (p < 0.02). There was no other change in airway dimensions in the normal subjects, but the angle of the neck was significantly more flexed in the supine posture (p < 0.05). In the patients with SAHS, uvula width was also increased in the supine posture, this time associated with narrowing of the retropalatal airway as shown by reductions in the dimensions from both the uvular protuberance and the uvular tip to the pharyngeal wall (p < 0.05). There was also widening of the retroglossal airway in the supine posture as shown by an increase in the posterior airway space (p = 0.003) with an increase in the hyoid to pharyngeal wall dimension. The angle of the neck was also more flexed in the supine position (p = 0.0001). There were no other postural changes in cephalometric dimensions. There was no difference in the effect of posture on the percentage change in any cephalometric variable between normal subjects and patients with SAHS. There was no significant change in any airway dimension with neck flexion in the erect posture either in the 19 normal subjects (neck flexion, 11 ± 6) or in the six patients with SAHS (flexion, 10 ± 6). Comparison of supine airway dimensions in normal subjects and patients with SAHS showed that the patients had significantly longer and wider uvulas (table 1), with a tendency (p = 0.1) to have a narrower distance between the uvular protrusion and pharyngeal wall. There was no difference in the retroglossal airway between patients and control subjects, but the patients' hyoids were held significantly lower and more anteriorly than those of the normal subjects.
Correlation coefficientsbetween body mass index, apnea/hypopnea index, lowest oxygen saturation, and cephalometric measurements in both the erect and the supine postures are given in table 2 for the patients with SAHS. Although no cephalometric variable in the erect posture was significantly correlated with either apnea/hypopnea index or lowest oxygensaturation during sleep, in the supine posture the hyoid to mandibular plane distance was significantly associated with both the apnea/hypopnea index and the lowest oxygen saturation during sleep. Discussion
This study has shown that posture has a significant effect on upper airway caliTABLE 2 CORRELATION COEFFICIENTS BETWEEN BODY MASS INDEX (BMI), APNEAlHYPOPNEA INDEX (AHI), LOWEST OXYGEN SATURATION (LoSAT), AND CEPHALOMETRIC MEASUREMENTS IN BOTH THE ERECT AND THE SUPINE POSTURES
AHI LoSAT H-MPe PASe TB-PhWe HP-PhWe UT-PhWe H-PhWe H-MPs PASs TB-PhWs UP-PhWs UT-PhWs H-PhWs
BMI
AHI
LoSAT
0.45 t 0.53* 0.07 0.57* 0.52 -0.06 0.004 0.39t 0.33 0.67* 0.54* -0.07 0.24 0.23
0.52* 0.26 0.35 0.31 0.19 0.29 0.29 0.52* 0.20 0.18 0.02 0.20 0.32
-0.19 -0.20 -0.17 0.20 0.11 0.02 -0.43t -0.25 -0.14 0.27 0.06 0.05
Definition of abbreviations: e __ erect position; 5 = supine position. For other definitions, see table 1 and text. • p < 0.05.
t
p < O.D1 .
EFFECT OF POSTURE ON UPPER AIRWAYS IN SLEEP APNEAlHYPOPNEA SYNDROME
ber. Assumption of the supine posture causes widening of the uvula, anterior displacement of the tongue, and anterior displacement of the hyoid. Cephalometry carried out in the erect posture has been widely used in the assessment of patients with obstructive sleep apnea/hypopnea (4, 7, 15). This technique has been used in the preoperative assessment of upper airway dimension prior to uvulopalatopharyngoplasty (16). This operation carries a significant failure rate (17, 18), ranging from 35070 (15) to 85% (19, 20). Although uvulopalatopharyngoplasty failure has been associated with narrowing of the posterior airway space in some studies (4), this has not been confirmed in others (15). The results of the current study may help explain some of the differences in relationships between upper airway dimensions assessed in different postures and the response to upper airway surgery (4, 5, 13). The postural changes in airway dimensions may explain why some untreated patients with SAHS sleep better when sitting up. Upper airway obstruction in many patients with SAHS occurs behind the palate (21, 22) when sleeping supine. Assumption of the erect posture willwiden the retropalatal airway - at least in the awake state - and this probably explains the beneficial effect of sleeping upright for some patients. The causes of the postural changes in upper airway dimension require further investigation. Possible contributing factors include neck flexion, gravity, and alteration in upper airway reflexes. The neck was held on average at 50 greater flexion in the erect than in the supine posture in the patients with SAHS and 3.50 greater flexion in the normal subjects. However, our demonstration of the lack of effect of neck flexion on airway caliber in the erect posture does not suggest that neck flexion explains our observations of differing airway dimensions in the different postures. Furthermore, cervical extension is associated with pharyngeal widening in both conscious (23) and anesthetized (24)subjects. Gravity would cause the soft palate and tongue to fall back in the supine posture, thus narrowing the upper airway at both levels, whereas the airway widened behind the tongue. It is thus possible that assumption of the supine posture causes widening of the uvula, perhaps because of decreased venous return, and there may be backward displacement of the uvula, which was more obvious in the patients with sleep apnea although there was a trend to narrowing of the retropalatal air-
way (p = 0.14) also in the normal subjects. Were it possible to image this level of the upper airway in three dimensions in both the erect and supine posture, this question could be clearly answered, but that is not possible by current techniques. Such narrowing of the airway at the level of the uvula produced by gravity would increase the resistance to airflow, which might reflexly increase genioglossal and geniohyoid tone, thus widening the hypopharyngeal airway. Recent evidence (25) suggests that genioglossal tone progressively increases in response to progressively more negative intraluminal pressure. Thus, we suspect that the observed postural changes in upper airway dimensions result from posterior movement of the soft palate and uvula in the supine posture because of gravity, with a subsequent reflex increase in genioglossal and geniohyoid tone. Correlations have been made between upper airway dimensions in the erect posture and severity of the SAHS (7). The preliminary results in the current study indicate that such correlations may be strengthened when measurements are made in the supine posture, and particularly the hyoid to mandibular plane distance becomes highly significantly correlated with both the apnea plus hypopnea index and the lowest oxygen saturation achieved. However, this was not the primary purpose of the current study, and these results were obtained on 29 patients only and will require confirmation in larger numbers. Cephalostats were not used in this study as patients do not sleep in cephalostats. Our intention was to look at the effect of nocturnal postures on airway dimensions, and this could only be achieved without the use of cephalostats. The study, therefore, demonstrates that posture is an important determinant of upper airway caliber in awake subjects. The ideal assessment of the upper airway in patients with SAHS is in the lying posture during sleep, but this is not as yet available on a routine basis. It may be that measurements made in the lying posture during wakefulness are the best currently available substitute. References 1. Remmers JE, De Groot WJ, Sauerland EK, Anch AM. Pathogenesis of upper airway occlusion during sleep. J Appl Physiol1978; 44:931-8. 2. Rivlin J, Hoffstein V, Kalbfleisch J, McNicholas W, Zamel N, Bryan AC. Upper airway morphology in patients with idiopathic obstructive sleep apnea. Am Rev Respir Dis 1984; 129:353-60. 3. Brown IG, Bradley TD, Phillipson EA, Zamel N, Hoffstein V. Pharyngeal compliance in snoring subjects with and without obstructive sleep apnea. Am Rev Respir Dis 1985; 132:211-5.
847 4. Riley R, Guilleminault C, Powell N, Blair Simmons E Palatopharyngoplasty failure, cephalometry roentgenograms and obstructive sleep apnea. Otolaryngol Head Neck Surg 1985; 93:240-3. 5. Shepard JW, Thawley SE. Evaluation of the upper airway by computerized tomography in patients undergoing uvulopalatopharyngoplasty for obstructive sleep apnea. Am Rev Respir Dis 1989; 140:711-6. 6. Riley R, Guilleminault C, Herran J, Powell N. Cephalometric analyses in flow volume loops in obstructive sleep apnea patients. Sleep 1983;6:30311. 7. Partinen M, Guilleminault C, Quera-salva M-A, Jamieson A. Obstructive sleep apnea and cephalometric roentgenograms. Chest 1988;93:1199-1205. 8. Haponik EF, Smith PL, Bohlman ME, Allen RP, Goldman SM, Bleeker ER. Computerised tomography in obstructive sleep apnea. Am Rev Respir Dis 1983; 127:221-6. 9. Shepard JW, Stanson AW, Sheedy PF, Westbrook PRo Fast type CT evaluation of the upper airway during wakefulness in patients with obstructive sleep apnea (abstract). Am Rev Respir Dis 1989; 139:A372. 10. Abbey NC, Block AJ, Green D, Mancuso A, Hellard TW. Measurement of pharyngeal volume by digitized magnetic resonance imaging. Am Rev Respir Dis 1989; 140:717-23. 11. Douglas NJ, Calverley PMA, Leggett RJE, et al, Transient hypoxaemia during sleep in chronic bronchitis and emphysema. Lancet 1979; 1:1-4. 12. Rechtschaffen A, Kales A, eds, A manual of standardized terminology, techniques and scoring system for sleep stages of human sleep. Bethesda, MD. National Institute of Neurological Disease and Blindness, 1968. (NIH publication no. 204). 13. Guilleminault C, Hoed J, Mitzler MM. Clinical overview of the sleep apnea syndromes. In: Guilleminault C, Dement WC, eds. Sleep apnea syndromes. New York: Alan R Liss, 1978; 1-12. 14. Gould GA, Whyte KF, Rhind GB, et al. The sleep hypopnea syndrome. Am Rev Respir Dis 1988; 137:895-8. 15. Gislason T, Lindholm C, Almqvist M, et af. Uvulopalatopharyngoplasty in the sleep apnea syndrome. Arch Otolaryngol Head Neck Surg 1988; 114:45-51. 16. Fujita S, Conway WA, Zorick'F, et al. Evaluation of the effectiveness of uvulopalatopharyngoplasty. Laryngoscope 1985; 95:70-4. 17. Conway W, Fujita S, Zorick F, et af. Uvulopalatopharyngoplasty, Chest 1985; 88:385-7. 18. Wetmore SJ, Scrima L, Snyderman NL, Hiller FC. Postoperative evaluation of sleep apnea after uvulopalatopharyngoplasty. Laryngoscope 1986; 96:738-41. 19. Schoen LS, Annand VK, Weisenberger S. Upper-airway surgery for treating obstructive sleep apnea. Arch Otolaryngol Head Neck Surg 1987; 113: 850-3. 20. Walker EB, Frith RW, Harding DA, Cant BA. Uvulopalatopharyngoplasty in severeidiopathic obstructive sleep apnea syndrome. Thorax 1989; 44: 205-8. 21. Hudgel DW. Variable site of airway narrowing among obstructive sleep apnea patients. J Appl Physiol 1986; 61:1403-9. 22. Chaban R, Cole P, Hoffstein V. Site of upper airway obstruction in patients with idiopathic obstructive sleep apnea. Laryngoscope 1988;98:641-7. 23. Shelton RL, Bosma JE Maintenance of the pharyngeal airway. J Appl Physioll%2; 17:209-14. 24. Safar P, Ecscarraga LA, Chang E Upper airway obstruction in the unconscious patient. J Appl Physiol 1959; 14:760-4. 25. Innes JA, Horner RL, Guz A. Negative pharyngeal pressure causes reflex upper airway dilator muscle activation in man. Clin Sci 1990; 78:16.
N. YILDIRIM, M. F. FITZPATRICK, K. F. WHYTE, R. JALLEH, A. J. A. WIGHTMAN, and N. J. DOUGLAS
Introduction
The obstructive sleep apnea/hypopnea syndrome (SAHS) is caused by repetitive narrowing and occlusion of the upper airway during sleep (1). Patients with the syndrome have narrow (2) and floppy (3) upper airways during wakefulness in comparison with normal subjects. The efficacy of upper airway surgery depends upon the location of airways obstruction, the response to uvulopalatopharyngoplasty being influenced by upper airway dimensions (4) and the site and extent of maximal airway narrowing (5). Some techniques currently used to determine the site of maximal airway narrowing in awake patients are carried out in the erect posture, and others are carried out with the patient supine. Lateral radiography cephalometry (6, 7)- and acoustic reflection (2) are carried out with the patient erect, whereas both computerized tomographic (8, 9) and magnetic resonance (10)imaging are carried out in supine patients. Clarification of the effect of posture on upper airway dimensions is thus important in allowing comparison between the results obtained by these different techniques. In addition, knowledge of the effect of posture on the upper airway is of interest as many patients with untreated SAHS report that they sleep better sitting upright than when lying in bed. We have, therefore, assessed the effect of posture on upper airway size in both normal nonsnoring subjects and in patients with the obstructive SAHS. The only technique that can assess the dimensions of the retropalatal and more distal upper airway in both erect and supine posture is cephalometry, which was thus utilized in this study. Methods Subjects Thirty-three nonsnoring normal subjects (15 men; mean age, 45 ± SD 16 yr; body mass index, 26 ± 5 kg/m) and 29 patients (26 men; mean age, 54 ± 11 yr; BMI, 32 ± 7 kg/m),
SUMMARY The effect of posture on upper airway dimensions was assessed for two reasons. First, some patients with untreated sleep apnea/hypopnea syndrome (SAHS) report they sleep better sitting upright. Second, to allow comparison of the differing techniques used to determine the site of maxlmel airway narrowing In ewake patients with SAHS, as some are carried out In the erect and others In the supine posture. Lateral cephalometry was therefore carried out In 33 nonsnorlng normal subjects and In 29 patients with obstructive SAHS (mean apneas plus hypopneas, 46 per hour; range, 17to 103). In both normal subjects and patients, uvular width was Increased (p < 0.05) In the supine posture, and this was associated with significant narrowing of the retropalatalalrway In the patients with SAHS (erect, 5.0 ± SO 2.6 mm; supine, 3.6 ± 2.8 mm; p < 0.01).In both normal subjects and patients, the retroglossal hypopharynx widened (p < 0.05) In the supine posture (e.g., in patients with SAHS, posterior airway space was: erect, 11.5 ± 4.5 mm; supine, 13A ±4.8 mm; p = 0.003). In the supine posture there was anterior movement of the hyoid and neck flexion In both groups. However, a study of the effect of neck flexion In the erect posture showed that neck flexion produced no changes In alrwey caliber. Thus, posture Is an Important detarmlnant of upper AM REV RESPIR DIS 1991; 144:845-S47 airway dimensions.
The normal subjects were recruited from relatives of patients with SAHS and from patients referred with nonrespiratory sleep complaints. All the patients with SAHS had been referred for evaluation because of clinical features of the SAHS. Each underwent polysomnography where electroencephalogram, electrooculogram, and electromyogram were recorded using our standard electrode placement (11), respiration was monitored by an inductance plethysmograph (Respitraces; Ambulatory Monitoring, Ardsley, NJ), airflow was measured by oronasal thermocouples, and oxygen saturation was measured by an Ohmeda 3700 oximeter (Ohmeda, Boulder, CO). Both sleep (12) and respiratory patterns (13, 14)were scored by standard criteria. The mean apnea plus hypopnea index of the normal subjects was 5 ± 3 (range, 0 to 11)per hour and of the patients with SAHS, 46 ± 27 (range, 17 to 103) per hour.
Radiology Cephalometric roentgenograms were carried out with the subjects both seated and supine with the patient facing at 90° to the X-ray beam. The patients were directed to gaze forward for the erect films and upwards for the supine films, holding their heads in a natural position. They were asked to close their mouths with their normal resting occlusion and their lips together, to allow the tongue to relax onto the floor of the mouth and not to swallow during the exposures. Radiographs were taken with the patient exhaling slowly
from a deep breath. The X-ray cone was positioned exactly 5 feet from the radiographic film and the film was placed against the left side of the face. Twenty-three cephalometric measurements were obtained from each subject in both erect and supine postures (figure I). These included distances between each of the following: anterior nasal spine (ANS), posterior nasal spine (PNS), gonion (Go), and gnathion (Gn). In addition, the distances from the midpoint of the sella (S) to the gonion and the hyoid (H) were measured. The distances from H to the mandibular plane (MP) and H to the intersection of a line along the posterior border of the mandibular process and the temporal bone (articulare; AR), H-Go, and Go-AR were measured. The greatest posterior convexity of the soft palate (uvular protrusion; UP) and the tip of the uvula (UT) were identified, and uvular length (UL; distance from PNS to UT) and uvular width (diameter of the soft palate at the widest point; UW) were measured as were distances from the posteri(Received in original form March 1, 1990 and in revised form March 4, 1991) 1 From the Respiratory Medicine Unit, Department of Medicine and Department of Radiology, City Hospital, Edinburgh, United Kingdom. , Correspondence should be addressed to N. J. Douglas, Respiratory Medicine Unit, Department of Medicine and Department of Radiology, City Hospital, Edinburgh, ENlO, 5SB, UK.
845
846
YILDIRIM, FITZPATRICK, WHYTE, JALLEH, WIGHTMAN, AND DOUGLAS
TABLE 1 CEPHALOMETRIC MEASUREMENTS IN NORMAL SUBJECTS AND IN PATIENTS WITH SAHS
l
Normal SUbjects Variables
ANS
Ar
H
::J Fig. 1. Diagrammatic representation of cephalometric measurementsmade.Fordefinition of abbreviations, see table 1 and texl.
or pharyngeal wall on a line from B (supramentale, the deepest point of the contour of the mandibular alveolus between the pogonion and the central incisor) through Go, posterior pharyngeal wall (PhW) to UP, UT, H, and PNS. The posterior airway space (PAS) the linear measurement between the base of the tongue (TB) and PHW on a line from point B through Go. PhW-TB (parallel to the mandibular plane) and PNS-TB were also measured. Two angles were measured: Go-Gn-H (the angle measured from Go to Gn to H) and the angle of the neck (the acute angle created by the intersection of a line drawn through the anterior border of the second cervical vertebra [C2l with a line drawn through the anterior border ofthe fourth cervical vertebra [C4]). Cephalometry films were also repeated in the erect posture with slight neck flexion in 19 of the normal subjects and in six patients with SAHS. This was to examine whether a change in neck angle could effect airway dimensions during supine cephalometry.
Statistical Analysis Comparisons weremade using Student's t test, with the Bonferroni correction for multiple comparisons when appropriate.
Results
The means and SD of each of the linear and angular measurements in the normal subjects and in the patients with SAHS are presented in table 1. In the normal subjects, .uvular width increased in the supine posture, but there was no significant change in uvula to pharyngeal wall dimensions. There was wideningof the retroglossal airwayas evidenced by a significant increase in the tongue base to pharyngeal wall dimension (p < 0.03). There was also an increase
UL, mm UW, mm UP-PhW, mm UT-PhW, mm TB-PhW. mm H-PhW, mm PAS, mm H-MP, mm Go-Go-H, degree Ang Nck, degree
Erect
46.8 10.2 6.0 8.7 10.0 34.8 11.9 23.2 24.6 16.5
± ± ± ±
± ± ±
± ±
±
6.2 2.1 3.4 3.1 4.1 4.0 4.6 7.6 8.3 9.7
Patients with SAHS
Supine
48.8 11.2 5.4 8.4 11.9 37.6 12.7 22.4 22.8 13.0
± ± ± ± ± ± ± ± ± ±
5.5 2.4t 3.1 3.2 4.5t 4.6t 4.7 6.7 7.5 9.6t
Erect
Supine
51.6 ± 5.8 11.7±2.6 5.0 ± 2.6 9.5 ± 3.4 10.2 ± 4.0 39.5 ± 6.5 11.5 ± 4.5 31.2 ± 6.7 33.4 ± 6.7 19.0 ± 10.0
52.4 13.0 3.6 7.7 11.4 41.3 13.4 31.2 34.1 14.0
± ± ± ± ± ± ± ± ± ±
7.0§ 2.6:j:§ 2.8* 5.1t 4.7 6.3*§ 4.8* 6.6§ 8.3§ 8.3*
Definition of abbreviations: Ul ; uvular length; UW ; uvular width; UP-PhW ; uvular protrusion to pharyngeal wall; UT-PhW ; uvular tip to pharyngeal wall; TB-PhW = base of tongue to PhW; H-PhW ; hyoid to PhW; PAS = posterior airway space; H-MP = hyoid to mandibular plane; Go-Gn-H ; gonion to gnathion to hyoid; Ang Nck ; angle of the neck. • Values are mean SO. t p < 0.05, erect versus supine. :1= p < 0.01, erect versus supine. § p < O.D1, supine normal SUbjects versus supine patients with SAHS.
in the hyoid to pharyngeal wall distance in the supine posture (p < 0.02). There was no other change in airway dimensions in the normal subjects, but the angle of the neck was significantly more flexed in the supine posture (p < 0.05). In the patients with SAHS, uvula width was also increased in the supine posture, this time associated with narrowing of the retropalatal airway as shown by reductions in the dimensions from both the uvular protuberance and the uvular tip to the pharyngeal wall (p < 0.05). There was also widening of the retroglossal airway in the supine posture as shown by an increase in the posterior airway space (p = 0.003) with an increase in the hyoid to pharyngeal wall dimension. The angle of the neck was also more flexed in the supine position (p = 0.0001). There were no other postural changes in cephalometric dimensions. There was no difference in the effect of posture on the percentage change in any cephalometric variable between normal subjects and patients with SAHS. There was no significant change in any airway dimension with neck flexion in the erect posture either in the 19 normal subjects (neck flexion, 11 ± 6) or in the six patients with SAHS (flexion, 10 ± 6). Comparison of supine airway dimensions in normal subjects and patients with SAHS showed that the patients had significantly longer and wider uvulas (table 1), with a tendency (p = 0.1) to have a narrower distance between the uvular protrusion and pharyngeal wall. There was no difference in the retroglossal airway between patients and control subjects, but the patients' hyoids were held significantly lower and more anteriorly than those of the normal subjects.
Correlation coefficientsbetween body mass index, apnea/hypopnea index, lowest oxygen saturation, and cephalometric measurements in both the erect and the supine postures are given in table 2 for the patients with SAHS. Although no cephalometric variable in the erect posture was significantly correlated with either apnea/hypopnea index or lowest oxygensaturation during sleep, in the supine posture the hyoid to mandibular plane distance was significantly associated with both the apnea/hypopnea index and the lowest oxygen saturation during sleep. Discussion
This study has shown that posture has a significant effect on upper airway caliTABLE 2 CORRELATION COEFFICIENTS BETWEEN BODY MASS INDEX (BMI), APNEAlHYPOPNEA INDEX (AHI), LOWEST OXYGEN SATURATION (LoSAT), AND CEPHALOMETRIC MEASUREMENTS IN BOTH THE ERECT AND THE SUPINE POSTURES
AHI LoSAT H-MPe PASe TB-PhWe HP-PhWe UT-PhWe H-PhWe H-MPs PASs TB-PhWs UP-PhWs UT-PhWs H-PhWs
BMI
AHI
LoSAT
0.45 t 0.53* 0.07 0.57* 0.52 -0.06 0.004 0.39t 0.33 0.67* 0.54* -0.07 0.24 0.23
0.52* 0.26 0.35 0.31 0.19 0.29 0.29 0.52* 0.20 0.18 0.02 0.20 0.32
-0.19 -0.20 -0.17 0.20 0.11 0.02 -0.43t -0.25 -0.14 0.27 0.06 0.05
Definition of abbreviations: e __ erect position; 5 = supine position. For other definitions, see table 1 and text. • p < 0.05.
t
p < O.D1 .
EFFECT OF POSTURE ON UPPER AIRWAYS IN SLEEP APNEAlHYPOPNEA SYNDROME
ber. Assumption of the supine posture causes widening of the uvula, anterior displacement of the tongue, and anterior displacement of the hyoid. Cephalometry carried out in the erect posture has been widely used in the assessment of patients with obstructive sleep apnea/hypopnea (4, 7, 15). This technique has been used in the preoperative assessment of upper airway dimension prior to uvulopalatopharyngoplasty (16). This operation carries a significant failure rate (17, 18), ranging from 35070 (15) to 85% (19, 20). Although uvulopalatopharyngoplasty failure has been associated with narrowing of the posterior airway space in some studies (4), this has not been confirmed in others (15). The results of the current study may help explain some of the differences in relationships between upper airway dimensions assessed in different postures and the response to upper airway surgery (4, 5, 13). The postural changes in airway dimensions may explain why some untreated patients with SAHS sleep better when sitting up. Upper airway obstruction in many patients with SAHS occurs behind the palate (21, 22) when sleeping supine. Assumption of the erect posture willwiden the retropalatal airway - at least in the awake state - and this probably explains the beneficial effect of sleeping upright for some patients. The causes of the postural changes in upper airway dimension require further investigation. Possible contributing factors include neck flexion, gravity, and alteration in upper airway reflexes. The neck was held on average at 50 greater flexion in the erect than in the supine posture in the patients with SAHS and 3.50 greater flexion in the normal subjects. However, our demonstration of the lack of effect of neck flexion on airway caliber in the erect posture does not suggest that neck flexion explains our observations of differing airway dimensions in the different postures. Furthermore, cervical extension is associated with pharyngeal widening in both conscious (23) and anesthetized (24)subjects. Gravity would cause the soft palate and tongue to fall back in the supine posture, thus narrowing the upper airway at both levels, whereas the airway widened behind the tongue. It is thus possible that assumption of the supine posture causes widening of the uvula, perhaps because of decreased venous return, and there may be backward displacement of the uvula, which was more obvious in the patients with sleep apnea although there was a trend to narrowing of the retropalatal air-
way (p = 0.14) also in the normal subjects. Were it possible to image this level of the upper airway in three dimensions in both the erect and supine posture, this question could be clearly answered, but that is not possible by current techniques. Such narrowing of the airway at the level of the uvula produced by gravity would increase the resistance to airflow, which might reflexly increase genioglossal and geniohyoid tone, thus widening the hypopharyngeal airway. Recent evidence (25) suggests that genioglossal tone progressively increases in response to progressively more negative intraluminal pressure. Thus, we suspect that the observed postural changes in upper airway dimensions result from posterior movement of the soft palate and uvula in the supine posture because of gravity, with a subsequent reflex increase in genioglossal and geniohyoid tone. Correlations have been made between upper airway dimensions in the erect posture and severity of the SAHS (7). The preliminary results in the current study indicate that such correlations may be strengthened when measurements are made in the supine posture, and particularly the hyoid to mandibular plane distance becomes highly significantly correlated with both the apnea plus hypopnea index and the lowest oxygen saturation achieved. However, this was not the primary purpose of the current study, and these results were obtained on 29 patients only and will require confirmation in larger numbers. Cephalostats were not used in this study as patients do not sleep in cephalostats. Our intention was to look at the effect of nocturnal postures on airway dimensions, and this could only be achieved without the use of cephalostats. The study, therefore, demonstrates that posture is an important determinant of upper airway caliber in awake subjects. The ideal assessment of the upper airway in patients with SAHS is in the lying posture during sleep, but this is not as yet available on a routine basis. It may be that measurements made in the lying posture during wakefulness are the best currently available substitute. References 1. Remmers JE, De Groot WJ, Sauerland EK, Anch AM. Pathogenesis of upper airway occlusion during sleep. J Appl Physiol1978; 44:931-8. 2. Rivlin J, Hoffstein V, Kalbfleisch J, McNicholas W, Zamel N, Bryan AC. Upper airway morphology in patients with idiopathic obstructive sleep apnea. Am Rev Respir Dis 1984; 129:353-60. 3. Brown IG, Bradley TD, Phillipson EA, Zamel N, Hoffstein V. Pharyngeal compliance in snoring subjects with and without obstructive sleep apnea. Am Rev Respir Dis 1985; 132:211-5.
847 4. Riley R, Guilleminault C, Powell N, Blair Simmons E Palatopharyngoplasty failure, cephalometry roentgenograms and obstructive sleep apnea. Otolaryngol Head Neck Surg 1985; 93:240-3. 5. Shepard JW, Thawley SE. Evaluation of the upper airway by computerized tomography in patients undergoing uvulopalatopharyngoplasty for obstructive sleep apnea. Am Rev Respir Dis 1989; 140:711-6. 6. Riley R, Guilleminault C, Herran J, Powell N. Cephalometric analyses in flow volume loops in obstructive sleep apnea patients. Sleep 1983;6:30311. 7. Partinen M, Guilleminault C, Quera-salva M-A, Jamieson A. Obstructive sleep apnea and cephalometric roentgenograms. Chest 1988;93:1199-1205. 8. Haponik EF, Smith PL, Bohlman ME, Allen RP, Goldman SM, Bleeker ER. Computerised tomography in obstructive sleep apnea. Am Rev Respir Dis 1983; 127:221-6. 9. Shepard JW, Stanson AW, Sheedy PF, Westbrook PRo Fast type CT evaluation of the upper airway during wakefulness in patients with obstructive sleep apnea (abstract). Am Rev Respir Dis 1989; 139:A372. 10. Abbey NC, Block AJ, Green D, Mancuso A, Hellard TW. Measurement of pharyngeal volume by digitized magnetic resonance imaging. Am Rev Respir Dis 1989; 140:717-23. 11. Douglas NJ, Calverley PMA, Leggett RJE, et al, Transient hypoxaemia during sleep in chronic bronchitis and emphysema. Lancet 1979; 1:1-4. 12. Rechtschaffen A, Kales A, eds, A manual of standardized terminology, techniques and scoring system for sleep stages of human sleep. Bethesda, MD. National Institute of Neurological Disease and Blindness, 1968. (NIH publication no. 204). 13. Guilleminault C, Hoed J, Mitzler MM. Clinical overview of the sleep apnea syndromes. In: Guilleminault C, Dement WC, eds. Sleep apnea syndromes. New York: Alan R Liss, 1978; 1-12. 14. Gould GA, Whyte KF, Rhind GB, et al. The sleep hypopnea syndrome. Am Rev Respir Dis 1988; 137:895-8. 15. Gislason T, Lindholm C, Almqvist M, et af. Uvulopalatopharyngoplasty in the sleep apnea syndrome. Arch Otolaryngol Head Neck Surg 1988; 114:45-51. 16. Fujita S, Conway WA, Zorick'F, et al. Evaluation of the effectiveness of uvulopalatopharyngoplasty. Laryngoscope 1985; 95:70-4. 17. Conway W, Fujita S, Zorick F, et af. Uvulopalatopharyngoplasty, Chest 1985; 88:385-7. 18. Wetmore SJ, Scrima L, Snyderman NL, Hiller FC. Postoperative evaluation of sleep apnea after uvulopalatopharyngoplasty. Laryngoscope 1986; 96:738-41. 19. Schoen LS, Annand VK, Weisenberger S. Upper-airway surgery for treating obstructive sleep apnea. Arch Otolaryngol Head Neck Surg 1987; 113: 850-3. 20. Walker EB, Frith RW, Harding DA, Cant BA. Uvulopalatopharyngoplasty in severeidiopathic obstructive sleep apnea syndrome. Thorax 1989; 44: 205-8. 21. Hudgel DW. Variable site of airway narrowing among obstructive sleep apnea patients. J Appl Physiol 1986; 61:1403-9. 22. Chaban R, Cole P, Hoffstein V. Site of upper airway obstruction in patients with idiopathic obstructive sleep apnea. Laryngoscope 1988;98:641-7. 23. Shelton RL, Bosma JE Maintenance of the pharyngeal airway. J Appl Physioll%2; 17:209-14. 24. Safar P, Ecscarraga LA, Chang E Upper airway obstruction in the unconscious patient. J Appl Physiol 1959; 14:760-4. 25. Innes JA, Horner RL, Guz A. Negative pharyngeal pressure causes reflex upper airway dilator muscle activation in man. Clin Sci 1990; 78:16.
