| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Guest Access | Sign In via User Name/Password
|
|
|
|||||||
Guest Access | Sign In via User Name/Password |
|||||||||
* From the West Australian Sleep Disorders Research Institute (Ms. Shepherd, Ms. Jensen, Ms. Maddison, and Dr. Hillman), Department of Pulmonary Physiology, Sir Charles Gairdner Hospital, Nedlands, Western Australia; and School of Anatomy and Human Biology (Dr. Eastwood), University of Western Australia, Perth, Australia.
Correspondence to: Peter R. Eastwood, PhD, West Australian Sleep Disorders Research Institute, Internal Mailbox 201, Queen Elizabeth II Medical Centre, Nedlands, Western Australia, 6009; e-mail: Peter.Eastwood{at}health.wa.gov.au
Abstract
Background: Upper airway (UA) patency during inspiration is determined by the balance between dilating forces generated by UA dilator muscle activity and collapsing forces related to the decreased intraluminal pressure that accompanies flow generated by inspiratory muscle activity. It is possible that the relative strengths of UA dilator and inspiratory pump muscles could be an important determinant of the susceptibility to UA collapse during sleep (ie, obstructive sleep apnea [OSA]).
Methods: Measurements of tongue protrusion (TP) force and maximum inspiratory pressure (PImax) were obtained in 94 patients admitted for overnight polysomnography for suspected OSA, quantified by apnea-hypopnea index (AHI).
Results: There was a direct linear relationship between TP force and PImax (r2 = 0.37, p < 0.001). A high ratio of TP force to PImax (greater than group 90th percentile, 0.027 kg/cm H2O) appeared to protect against OSA, as moderate-to-severe OSA (AHI > 20/h) was not observed in any individual with a ratio above this threshold. AHI was not linearly related to TP force, PImax, or the ratio of TP force to PImax.
Conclusions: UA muscle strength is linearly related to inspiratory pump muscle strength. The ratio of UA muscle strength (TP force) and inspiratory pump muscle strength (PImax) was not different between individuals with and without OSA; however, a high wakeful ratio of TP force to PImax appears to be associated with a reduced propensity to moderate-to-severe OSA.
Key Words: collapsibility • diaphragm • genioglossus • obstructive sleep apnea • respiratory muscle strength • tongue protrusion • upper airway
Obstructive sleep apnea (OSA) is characterized by repetitive collapse of the upper airway (UA) during sleep. The tendency for the UA to collapse during sleep is related to the balance between dilating forces generated by UA dilator muscle activity and collapsing forces associated with the intraluminal pressure reductions that accompany flow generated by inspiratory muscle activity.1 Hence, it is possible that the relative strength of UA dilator and inspiratory pump muscles could be an important determinant of susceptibility to OSA. While substantial weakness of pharyngeal dilator muscles predisposes to OSA, as evident in neuromuscular disease,2 it is unknown whether a predisposition also exists in individuals in whom pharyngeal muscle dilator force-generating capacity is more subtly reduced relative to force-generating capacity of the inspiratory muscles. If so, such a predisposition may be evident from measures of the relative strength of these muscles.
Furthermore, it is unclear whether OSA itself influences UA dilator muscle strength. Abnormal structural changes including an increase in type IIa muscle fibers in UA muscles have been reported in individuals with OSA.345 These changes are consistent with an increased capacity for the UA muscles to produce force.3 Alternatively, other structural changes, possibly caused by inflammatory processes resulting from snoring-related vibration of pharyngeal walls,6 could decrease the capacity of the UA muscles to produce force. Functional studies provide conflicting findings with measures of in vitro twitch tension of UA muscle obtained from patients with OSA being either greater than nonapneic snorers,3 or not different from normal control subjects.5 The force developed during maximal voluntary tongue protrusion (TP) provides an index of functional strength of the genioglossus muscle, the major dilator muscle in the UA, as well as other intrinsic tongue muscles. The only study7 to measure TP force in OSA reported a weak negative relationship between TP force and severity of OSA, a finding attributed to an age-related decrease in TP force and increase in severity of OSA.
It is also possible that OSA affects the force-generating capacity of inspiratory muscles either directly or because of coexisting obesity that is commonly present in individuals with OSA. Previous studies have reported impaired ventilatory capacity in obese subjects compared to normal individuals,89 most likely as a consequence of reduced chest wall compliance.10 Conversely, inspiratory muscle strength could be preserved or even increased in OSA, as a consequence of the training effect of exaggerated inspiratory efforts made repeatedly throughout sleep against a narrowed or obstructed UA.
The ratio of maximum TP force to maximum inspiratory pressure (PImax) provides a simple index of the relative strength of UA dilator and inspiratory muscles. This study sought to determine whether there was any relationship between the ratio of TP force to PImax and the propensity for OSA. We hypothesized that waking measures of the ratio of UA muscle force to inspiratory pump muscle force would better predict severity of OSA than waking measures of UA muscle force or respiratory pump muscle force alone, such that individuals with greater UA muscle force relative to pump muscle force would be protected from UA collapse during sleep.
Materials and Methods
Subjects
Ninety-four consecutive patients (52 men and 42 women) attending the Respiratory Sleep Disorders Clinic at Sir Charles Gairdner Hospital for overnight polysomnography were recruited into the study (Table 1
). Subjects were excluded from participation if they had been or were being treated with continuous positive airway pressure, had undergone previous surgical treatment for sleep-disordered breathing, or if they had neuromuscular or pulmonary disease. This study was approved by the Sir Charles Gairdner Hospital Human Research Ethics Committee. Informed consent was obtained from each subject prior to participation in the study
|
Measurements
TP Force:
A lingual force transducer was constructed based on the device described by Mortimore et al.11 The mouthpiece consisted of a 1.0-cm diameter nylon plate connected to a load cell. Movement of the plate on the load cell was negligible. Behind the nylon plate was a groove 2.0 mm deep and 2.0 mm wide. Subjects were asked to rest their upper and lower incisors in the groove while supporting the transducer with one hand, and then to push their tongue as hard as possible against the nylon plate for approximately 2 s. Subjects were encouraged to produce a maximal effort during each attempt. The change in force with time was displayed on an oscilloscope positioned in front of the subject.
Maximum TP force was defined as the peak force generated in three attempts in which the peak force differed by < 5%. In the majority of cases, this required no more than six attempts. Subjects rested for approximately 30 s between trials. Measurements were obtained in both seated and supine postures with the head in a neutral position (when sitting, they were instructed to look straight ahead; when supine, the head was maintained in a neutral position using a head rest [Shea head rest; Gyrus ENT; Memphis, TN]). The force transducer was connected to an amplifier (Analog Devices; Norwood, MA), and data were recorded at 100 Hz on a data acquisition and analysis system (Powerlab model 16s; ADInstruments; Sydney, Australia).
PImax:
PImax was measured using a noncollapsible mouthpiece. A leak of 1.0 mm in diameter was created in the mouthpiece to minimize generation of pressure by the facial muscles. Pressure within the mouthpiece was monitored with a pressure transducer (Monitoring Kit; Abbott; Sligow, Ireland) and custom-made amplifier (Medical Technology and Physics, Sir Charles Gairdner Hospital; Nedlands, Western Australia) and recorded at 100 Hz on a data acquisition and analysis system (Powerlab model 16s).
Subjects were encouraged to perform a maximum inspiratory effort from normal FRC or RV against an occluded mouthpiece. Care was taken to ensure a stable pattern of quiet breathing prior to the FRC measurement and for complete expiration prior to the RV measurements. Each effort was maintained for approximately 2 s. Subjects rested for at least 30 s between efforts. PImax was defined as the peak pressure generated in three efforts in which the peak pressure differed by < 5%. In most cases, no more than six efforts were required. All measurements were made while the subject was seated with the head in a neutral position.
Polysomnography:
Overnight polysomnography was performed using a computerized data acquisition system (E-Series; Compumedics; Melbourne, Australia). This included monitoring of EEG (C4-A1 and C3-A2) with gold cup electrodes attached to the scalp, left and right electrooculograms, submental electromyogram, and ECG recorded with surface electrodes, abdominal and thoracic effort (inductance plethysmography), nasal and oral airflow, nasal pressure, leg movements, oxygen saturation, body position (mercury switch position sensor positioned on patients’ chest), and sound intensity measured by microphone suspended above the subject. Sleep stage analysis was performed according to standard criteria.12 Respiratory events during sleep were also analyzed according to recognized criteria.13 Severity of sleep-disordered breathing was reflected in the apnea-hypopnea index (AHI), which is defined as the number of apneas and hypopneas per hour of sleep. An AHI of 20/h was defined as the cutoff for clinically significant OSA, as an AHI > 20/h has been shown to be associated with significantly greater mortality than an AHI < 20/h.14
Data Analysis
Comparisons of variables between male and female subjects and between subjects with and without OSA were performed using unpaired t tests. Linear regression analyses were used to examine the relationships between TP force, PImax, and AHI. For these analyses, AHI was logarithmically transformed to normalize its distribution.7 Predicted values for PImax-FRC and PImax-RV were obtained from those described by Hamilton et al15 and Black and Hyatt.16 The 
2 test was used to determine whether the 90th percentile of the ratio of TP force to PImax could be used to identify individuals with and without OSA. When appropriate, TP force and PImax were corrected for age and BMI by dividing through by these variables. Data are presented as mean ± SD, and p < 0.05 was considered significant for all comparisons.
Results
Subjects
A total of 94 subjects were studied (Table 1). As a group, they were obese (BMI, 32 ± 8 kg/m2) and had moderate OSA (AHI, 25 ± 30/h). Compared to female subjects, male subjects were taller and had greater neck circumference. BMI, AHI, neck circumference, and cricomental space were greater in subjects with OSA vs subjects without OSA (Table 2
). On average, subjects slept for 5.8 ± 1.7 h, had an average sleep efficiency of 76.1 ± 12.6%, and spent 2.3 ± 3.4% of the night in stage 1 sleep, 66.8 ± 12.6% in stage 2 sleep, 9.2 ± 6.4% in stage 3 sleep, 14.6 ± 11.4% in stage 4 sleep, and 16.9 ± 7.3% in rapid eye movement sleep.
|
TP force measured supine was greater in individuals with OSA than those without OSA (Table 2), but was not different when TP force was corrected for age and BMI. No difference was observed between the groups when TP force was measured seated (Table 2).
No significant relationships were observed between log AHI and TP force whether measured in the seated or supine postures either for all subjects combined or when separated into those with and without OSA (Fig 1 ). TP force decreased with increasing age in subjects with OSA whether measured seated (r2 = 0.25, p < 0.01) or supine (r2 = 0.17, p < 0.01). No such relationship was observed in subjects without OSA whether measured seated (r2 = 0.01, p = 0.473) or supine (r2 = 0.0002, p = 0.917). No significant relationships were observed between TP force (seated or supine) and BMI either for all subjects combined or when separated into those with and without OSA.
|
No significant relationships were observed between log AHI and PImax-FRC (Fig 2 ) or PImax-RV in subjects with or without OSA or when all subjects were combined. PImax-FRC and PImax-RV both decreased with increasing age in all subjects whether considered as a single group (r2 = 0.17, p < 0.001, or r2 = 0.23, p < 0.001, respectively), or when separated into those with OSA (r2 = 0.29, p < 0.001, or r2 = 0.36, p < 0.001) or without OSA (r2 = 0.14, p = 0.007, or r2 = 0.18, p = 0.002). No significant relationships were observed between PImax and BMI whether measured from FRC or RV.
|
|
2]. This was also the case for the ratio of TP force to PImax when measured supine and from RV, respectively (results not shown). When the ratio of TP force to PImax was controlled for age and BMI, a significant threshold effect remained between TP force (seated or supine) and PImax (FRC and RV), such that moderate-to-severe OSA was not observed in any individuals with a ratio greater than that corresponding to the 90th percentile for the group. No relationships were observed between the ratio of TP force to PImax and duration of apneas or hypopneas, mean oxygen saturation overnight, or arousal index.
|
This study measured the strength of UA dilator and inspiratory pump muscles in individuals with and without OSA, and examined the potential for measurements of the relative force-generating capacity of these muscles to identify individuals who may be susceptible to UA collapse during sleep (ie, OSA). The major findings of the study were as follows: (1) in both male and female subjects and when all subjects were considered as a single group, TP force was greater when measured supine than seated; (2) supine TP force was greater in individuals with OSA than without OSA, but not when TP force was corrected for age and BMI; (3) PImax was not different between individuals with and without OSA; (4) UA muscle strength (TP force) was linearly related to inspiratory pump muscle strength; and (5) a very high ratio of TP force to PImax was associated with a reduced propensity to moderate-to-severe OSA.
UA Muscle Strength and OSA
There are several reasons why waking measurements of TP force may reflect the capacity of the dilator muscles to maintain airway patency during sleep. Firstly, the genioglossus is the major dilator muscle of the UA,117 being primarily responsible for maintenance of pharyngeal patency. Secondly, it has previously been shown that genioglossus electromyogram activity is related to TP strength; thus, a measure of TP force should reflect its maximal capacity to dilate the UA.18 Thirdly, the genioglossus muscle has been shown to have an increased proportion of type II muscle fibers and an increase in anaerobic enzyme markers in patients with OSA, compared to normal individuals and nonapneic snorers.35 Such changes are consistent with an increased capacity of the UA muscles to produce force,3 which could be reflected in wakeful, voluntary maximal efforts. Fourthly, studies have shown that training the genioglossus muscle with electrical stimulation during wakefulness can decrease snoring,1920 improve symptoms of daytime sleepiness, and decrease AHI.20
The present study found that supine TP force was significantly greater in individuals with moderate-to-severe OSA; however, such a difference was not apparent when measurements of TP force were controlled for age and BMI. This most likely reflects the observation that TP force decreases with increasing age and BMI.7 The lack of relationship between log AHI and TP force in our group of subjects and the weak correlation between these measures reported by Mortimore et al7 argue against an effect of OSA on the capacity of the UA muscles to generate force.5 While such a conclusion differs to studies345 showing an increase in type II muscle fibers and an increase in in vitro twitch tension in UA muscle fibers from OSA patients,3 the weak and variable nature of these relationships reported in the present and other studies3457 argue against a major training effect on the UA muscles as a result of repetitive overnight muscle activation in response to UA narrowing in OSA. Further, they suggest that any abnormal structural changes in UA muscles reported in individuals with OSA345 do not have a major influence on the capacity of the tongue to produce force. The clinical significance of these structural changes in relation to OSA remains unknown.
Respiratory Pump Muscle Strength and OSA
To date, no study has specifically examined the relationship between respiratory pump muscle strength and the severity of OSA. It is possible that the strength of the diaphragm and other inspiratory muscles could be increased in patients with OSA by virtue of the "training effect" of repetitive overnight inspiratory efforts against an occluded or narrowed airway. Alternatively, it is possible that in obese subjects in whom the risk of OSA is increased,21 a decrease in chest wall compliance10 could impair the force-generating capacity of the inspiratory muscles. However, against the latter possibility we found no relationship between absolute or percentage of predicted PImax and BMI in our subjects, consistent with findings of an earlier study.22
The present study found no difference in PImax-FRC or PImax-RV between individuals with and without OSA, and no relationship between PImax and log AHI. These results suggest that OSA has no overnight training effect on respiratory pump muscles but contrast with those of Mezzanotte et al,23 who measured PImax in 14 OSA subjects and 11 normal control subjects and reported a significantly higher PImax in the control group. The reasons for the disparate findings are unclear but may relate to between-study differences in sample size, characteristics of the control group, and methodology.
Ratio of UA Muscle Strength and Pump Muscle Strength
The notion of the importance to UA patency of the balance between genioglossal force tending to open the airway and inspiratory muscle force tending to close it through generation of negative intrapharyngeal pressure has been extant since the seminal work of Remmers et al.1 Our findings suggest that relative force-generating capacity of these muscles is important, with relatively strong UA muscles protecting against moderate-to-severe OSA. It is unclear whether this is the result of cause or consequence. However, our findings of a lack of association between TP force and AHI, PImax and AHI, and the finding of a linear association between TP force and PImax suggests that any change in force-generating capacity of the UA dilator and inspiratory muscles is an inherent characteristic rather than acquired as a consequence of OSA.
Other factors may have played a role in the lack of association between AHI and TP force, PImax, or the ratio of TP force to PImax. Because measures of TP force and PImax are reliant on subject effort, it is possible that they were underestimated in some individuals. However, we believe this unlikely, as subjects were highly motivated, strongly encouraged, and the results reproducible, differing by no more than 5% over a minimum of three efforts. Further, in the case of PImax, group means were > 100% of the predicted value. We chose to measure PImax in the seated position because the effect of supine position on PImax is known,24 and we were keen to make the measurements in a condition that best simulated how it would be used clinically (ie, seated). Because the effect of posture on TP force was unknown, we measured it both seated and supine. While it could be argued that it would be more appropriate to compare measurements of TP force and PImax supine, as it is in the posture that we sleep, it is likely that the strong association between TP force and PImax observed when seated would be replicated when supine. While it could also be argued that the lack of association between AHI and TP force, PImax, or the ratio of TP force to PImax reflects the fact that AHI is measured during sleep while the latter measures are made during wakefulness, this is by necessity as wakefulness is a necessary condition to assess maximum voluntary strength.
In conclusion, while there was no direct relationship between the ratio of UA muscle strength (TP force) to inspiratory pump muscle strength (PImax) and AHI, it was notable that moderate-to-severe OSA was not observed in any individual with a ratio exceeding the 90th percentile for the group (0.027 kg/cm H2O), suggesting that a very high waking ratio may indeed be protective of UA collapse during sleep.
Acknowledgements
We wish to thank the technical staff of the West Australian Sleep Disorders Research Institute for their support.
Footnotes
Abbreviations: AHI = apnea-hypopnea index; BMI = body mass index; FRC = functional residual capacity; OSA = obstructive sleep apnea; PImax = maximum inspiratory pressure; PImax-FRC = maximum inspiratory pressure measured at functional residual capacity; PImax-RV = maximum inspiratory pressure measured at residual volume; RV = residual volume; TP = tongue protrusion; UA = upper airway
This article was submitted through the West Australian Sleep Disorders Research Institute, Department of Pulmonary Physiology, Sir Charles Gairdner Hospital, Nedlands, Western Australia.
Ms. Shepherd, Dr. Eastwood, and Dr. Hillman contributed to the study conception, design, data interpretation, and preparation and revision of the manuscript. Ms. Shepherd, Ms. Jensen, and Ms. Maddison were primarily responsible for data collection.
This study was funded, in part, by National Health and Medical Research Council (Australia) Grant No. 303218. Dr. Eastwood was supported by a National Health and Medical Research Council (Australia) R. Douglas Wright Fellowship (No. 294404).
The authors have no conflict of interests with any part of this study to disclose.
Received for publication December 9, 2005. Accepted for publication June 13, 2006.
References
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
20/h, n = 52) [top right] and with OSA (AHI > 20/h, n = 42) [bottom right]. The solid lines represent regression lines.
