- Email: [email protected]
Comparison of supine-only and REM-only obstructive sleep apnoea
Sleep Medicine 13 (2012) 875–878
Contents lists available at SciVerse ScienceDirect
Sleep Medicine journal homepage: www.elsevier.com/locate/sleep
Original Article
Comparison of supine-only and REM-only obstructive sleep apnoea Andrew Gillman a, Teanau Roebuck a, Sally Ho a, Esther van Braak a, Matthew T. Naughton a,b,⇑ a b
Department of Allergy, Immunology and Respiratory Medicine, Alfred Hospital, Melbourne, Victoria, Australia Monash University, Melbourne, Victoria, Australia
a r t i c l e
i n f o
Article history: Received 8 August 2011 Received in revised form 11 December 2011 Accepted 22 January 2012 Available online 24 May 2012 Keywords: Sleep apnoea Supine position REM sleep Digital monitoring Clinic population Therapy
a b s t r a c t Objectives: The effect of body position and sleep state on sleep apnoea have major clinical implications in the management of patients, yet are infrequently reported in the scientific literature. The aim of this study was to compare and contrast the prevalence and severity of supine-only and rapid eye movement (REM)-only obstructive sleep apnoea (OSA) in a population. Methods: Prospective cohort analysis of the influence of supine body position and REM sleep on the severity of apnoea in 100 consecutive patients with OSA (apnoea–hypopnoea index [AHI] > 5) using attended polysomnography with continuous digital monitoring in an accredited sleep laboratory. Supine-only OSA was defined as a supine:non-supine AHI ratio of >2:1 and non-supine AHI <5 events/ h. REM-only OSA was defined as an REM:non-REM ratio of >2:1 and non-REM AHI <5 events/h. Results: Supine sleep time represented a greater proportion of total sleep time than REM sleep time (40% vs 13%). The prevalence of supine-only OSA was more than twofold greater than that of REM-only OSA (23% and 10%, respectively). The supine-only group had greater overall AHI (mean 12.6 ± 6.1 vs 7.2 ± 2.2 events/h; P < 0.01) than the REM-only group. No significant differences in gender, age, or sleepiness were found between the two groups. Conclusions: Supine-only OSA is more common and is associated with a greater AHI than REM-only OSA. Ó 2012 Elsevier B.V. All rights reserved.
1. Introduction Considerable variance in the severity of obstructive sleep apnoea (OSA), as measured by the apnoea–hypopnoea index (AHI), can be explained by several factors, which include the effects of either sleep state or body position. The effects of sleep state upon upper airway and respiratory pump muscle activity and cortical arousability are well known. Of the various sleep states, rapid eye movement (REM) sleep has long been known to have a significant effect on OSA, with greater length of apnoeas, greater hypoxaemia and greater hypercapnia than apnoeas during stages 1 + 2 non-REM sleep [1]. In contrast, slow wave sleep is characterized by long periods of snoring with trivial hypoxaemia and an absence of arousals or discreet (obstructive or central) apnoeic events. In stages 1 and 2 non-REM sleep, however, sleep arousals are commonplace, leading to instability of respiratory control and both obstructive and central apnoeas. Thus, the severity of sleep apnoea in non-REM sleep is often considerably less than that observed in REM sleep. One study found that patients with OSA restricted to REM sleep were more likely to be female, to have less severe OSA, and to be less obese [2]. ⇑ Corresponding author at: Department of Allergy, Immunology and Respiratory Medicine, Alfred Hospital, P.O. Box 315, Prahran 3181, Victoria, Australia. Tel.: +61 3 9076 3770; fax: +61 3 9076 3601. E-mail address: [email protected] (M.T. Naughton). 1389-9457/$ - see front matter Ó 2012 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.sleep.2012.01.016
Similarly, body position has a strong influence on OSA. A supine position and its associated gravitational effects on the tongue and mandible make the upper airway more collapsible, which is thought to be responsible for the worsening of OSA compared with a non-supine position [3]. Similar body positional effects in OSA have been observed in central sleep apnoea associated with heart failure, where the effects appear to be pulmonary rather than upper airway in origin [4]. Compared with non-supine OSA patients, patients with supine-only OSA tend to be younger, male, and to have a lower body mass index (BMI) [5]. The importance of identifying supine-only OSA is underscored by: (a) its association with milder disease compared with non-supine OSA (i.e., lower overall AHI) [5]; (b) a greater therapeutic response to oral appliances [6] and positional therapy [7–9]; and (c) the impact on the optimal fixed and autotitrating continuous positive airway pressure (CPAP) settings [10]. A further reason to make a distinction between REM-only and supine-only OSA is borne by the therapeutic option of positional therapy. In patients with supine-related OSA, two studies have suggested that positional therapy (i.e., instructing patients to avoid the supine position) is associated with a reduction in systemic blood pressure [11]. However, the definitions of REM OSA [12] and supine OSA [3,5– 7,9,10] in previous published papers have been inconsistent. In general, the definitions have been based upon a 2:1 ratio of either REM:non-REM AHI or supine:non-supine AHI. This variation in def-
876
A. Gillman et al. / Sleep Medicine 13 (2012) 875–878
inition significantly influences the prevalence of both REM-only and supine-only OSA. Moreover, as the inflection point of the total AHI is approximately 5 events/h for the development of cardiovascular and other complications and mortality [14], it is important to identify patients in whom positional therapy may be a safe option. Although similarities exist in the clinical descriptions of REMonly and supine-only OSA patients (i.e., younger, lower BMI) compared with non-REM and non-supine OSA patients, the difference in severity of OSA between REM-only and supine-only patients using similar definitions is unknown. The aims of this study were twofold. The first aim was to compare the clinical features of supine-only OSA with REM-only OSA using strict functional definitions to assess gender bias, BMI, and overall OSA severity based upon AHI and minimum oxygen saturation (SpO2). The second aim was to assess the prevalence of supineonly OSA using a strict functional definition (supine:non-supine event ratio >2 plus a non-supine AHI <5 events/h) and a lenient, more traditional definition (supine:non-supine AHI ratio >2) in a population of patients with suspected OSA attending a university-based sleep clinic.
for sleep stage using standard criteria, as described previously [4]. Apnoeas were defined as an absence of air flow for at least 10 s. Hypopnoeas were defined as a discernable fall in respiratory effort or flow associated with either a P3% SpO2 desaturation or an arousal. Body position was assessed with a Compumedics positional sensor device with digital video correction. For inclusion in the study, subjects had to sleep for at least 4 h with at least 15 min of REM sleep and to spend at least 30 min in a supine position and 30 min in a non-supine position during sleep. Supine-only and REM-only OSA were defined as AHI <5 (i.e., strict definition). Supine-only OSA could involve any sleep stage, and REM-only OSA could involve any body position. Supine-only OSA was subdivided using a lenient definition (2:1 AHI ratio) and a strict definition (AHI < 5 events/h). Data are presented as mean ± standard deviation. Unpaired ttests were used for comparison between groups. Multivariate analysis was used to assess the relative and significant contributions of positional- and REM-related AHI towards the overall AHI using multiple variant analysis. P < 0.05 was taken to indicate significance.
2. Methods
3. Results
Consecutive subjects of either gender aged >18 years, diagnosed with OSA (AHI > 5 events/h) using laboratory-attended polysomnography at an Australian-Sleep-Association-accredited sleep laboratory within a large university based teaching hospital, were enrolled in the study. Patients with known cardiac, pulmonary or neurological disease were excluded. In total, 100 subjects were identified. Approval was obtained from the Alfred Ethics Committee and patients provided informed consent. Polysomnography (Compumedics, Abbottsford, Australia) included electro-encephalography (C3A2, C4A1, O3A2, O4A1), submental electromyography (EMG), electro-oculography, electrocardiography, chest and abdominal movement using inductance plethysmography, oronasal thermistor, nasal pressure, digital video observation of body position, SpO2 (3-s averaging time), and anterior tibialis EMG. All sleep studies were scored manually
The cohort was aged 53.6 ± 12.9 years with a BMI of 31.5 ± 8.6 kg/m2 and Epworth Sleepiness Scale (ESS) score of 8.9 ± 5.3. Sixty-nine percent of participants were male. Patients slept for 351 ± 6 min, with 40% (138 min) of total sleep time in a supine position and 13% (46 min) of total sleep time in REM sleep (Table 1). The patients slept with 1.7 ± 0.5 pillows. Typical polysomnograph examples of REM-only and supine-only OSA are shown in Fig. 1. The AHI values were 29.6 ± 23.2 events/h (overall), 43.2 ± 27 events/h (supine), 20.9 ± 23 events/h (non-supine), 33.5 ± 21 events/h (REM) and 28.4 ± 24.7 events/h (non-REM) (Table 1). The corresponding minimum SpO2 values were 88.9%, 88.2%, 90.4%, 88.3%, and 88.6%. The proportion of total sleep time spent in a supine position was significantly greater than the proportion of total sleep time spent in REM sleep (40% vs 13%). Body
Fig. 1. Polysomnographic summary graphs of two male patients of similar age (45 years) and body mass index (BMI): one with rapid eye movement (REM)-only obstructive sleep apnoea (OSA) and the other with supine-only OSA.
877
A. Gillman et al. / Sleep Medicine 13 (2012) 875–878 Table 1 Baseline demographics of entire group (n = 100).
Table 3 Comparison of lenient and strict definitions of supine-only obstructive sleep apnoea.
Age (years) Gender (M:F) Body mass index (kg/m2) Epworth Sleepiness Scale score Total sleep time (hours:minutes)
53.6 ± 12.9 69:31 31.5 ± 8.6 8.9 ± 5.3 5:51 ± 6
AHI (events/h) Total Supine Non-supine REM Non-REM
29.6 ± 23.2 43.2 ± 27.3 20.9 ± 22.9 33.5 ± 21.2 28.4 ± 24.7
Supine sleep time (% total sleep time) Number of pillows
39.6% 1.7 ± 0.5
AHI, apnoea–hypopnoea index; REM, rapid eye movement. Data are mean ± standard deviation.
Table 2 Comparison of rapid eye movement (REM)-only and supine-only obstructive sleep apnoea (OSA) using the strict definition (apnoea–hypopnoea index [AHI] < 5 events/ h).
Prevalence (%) Age (years) Males (%) Body mass index (kg/m2) Epworth Sleepiness Scale score Total AHI (events/h) Supine AHI (events/h) Non-supine AHI (events/h) Minimum oxygen satuation (%)
REM-only OSA
Supine-only OSA
P
10 44.3 ± 9.0 80 ± 4 28.1 ± 5.8 7.4 ± 4.7 7.2 ± 2.2 8.4 ± 4.5 5.5 ± 4.3 89 ± 1.6
23 50.2 ± 12.3 87 ± 3 30.2 ± 11.9 9.2 ± 6.1 12.6 ± 6.1 28 ± 16.4 3.1 ± 2.4 88 ± 1.3
<0.01 0.22 0.62 0.65 0.4 0.01 0.001 0.048 ns
ns, not significant. Data are mean ± standard deviation.
position and REM sleep explained the variance in overall AHI, whereas minimum SpO2 was explained by position but not REM sleep. The prevalence of supine-only OSA was approximately twofold greater than the prevalence of REM-only OSA (23% vs 10%) (Table 2). The overall AHI in the supine-only OSA group was significantly greater than that of the REM-only group (12.6 ± 6.1 vs 7.2 ± 2.2 events/h; P = 0.01). The supine-only and REM-only OSA patients were of similar age (50 ± 12 vs 44 ± 9 years; P = 0.22), similar weight (BMI 30 ± 12 vs 28 ± 6 kg/m2; P = 0.65), had similar symptoms of sleepiness (ESS score 9.2 ± 6.1 vs 7.4 ± 4.7; P = 0.4), and a similar proportion were male (71% vs 69%). There was a significant variance in the prevalence of supine-only OSA dependent upon the definition used for non-supine, despite all patients having a supine:non-supine ratio of >2. The prevalence varied from 23% (using a strict definition of non-supine AHI of <5 events/h) to 63% (using a more lenient definition where non-supine AHI was unstated) (Table 3). Within the supine-only OSA group, the lenient definition group had a significantly lower overall AHI and a trend to be younger with a lower BMI compared with the strictly defined non-supine OSA group, although there were no differences in the supine sleep time or the mean supine AHI compared with the non-supine OSA group. 4. Discussion This study examined the prevalence and clinical features of supine-only and REM-only OSA. There were several novel and clinically relevant findings. First, the supine sleep time was longer than the REM sleep time (40% vs 13%). Second, the prevalence of
Prevalence (%) Age (years) Males (%) Body mass index (kg/m2) Epworth Sleepiness Scale score Total AHI (events/h) Supine AHI (events/h) Non-supine AHI (events/h) Minimum oxygen saturation (%)
Lenient definition
Strict definition
63 53 71 31 9.8 23 43 11 83
23 50 87 30 9.1 12 28 3 88
P
ns ns ns ns <0.01 <0.01 <0.01 <0.01
AHI, apnoea–hypopnoea index; ns, not significant. Lenient definition: overall apnoea–hypopnoea index (AHI) > 5 events/h, and supine:non-supine AHI ratio > 2. Strict definition: overall AHI > 5 events/h and supine:non-supine AHI ratio > 2 plus non-supine AHI < 5 events/h.
supine-only OSA was greater than the prevalence of REM-only OSA. Third, the overall severity of OSA was greater in supine-only OSA than in REM-only OSA. Finally, when the various published definitions of supine-only OSA were used, the prevalence varied from 23% (strict definition) to 63% (lenient definition); these values are similar to those published in the literature, namely 27% [7] and 50–60% [5]. The prevalence of supine-only OSA in this study varied from 23% to 63% depending upon the definition used. Mador et al. [15] previously introduced a stricter definition of supine-only OSA with a non-supine AHI of <5 events/h. Based upon 326 patients studied with polysomnography in two centres (17% had split night, balance full night diagnostic) and body position by direct observation, they demonstrated that 27% of their cohort of OSA patients met this criterion for supine-only OSA, similar to the present value of 23%. The authors felt that split night studies did not provide sufficient time for a sufficient mixture of supine and non-supine sleep. The finding that body position is a more important determinant than sleep stage on overall AHI (and thus OSA severity) is novel. Multivariate analysis showed that body position had a more important effect than sleep stage on OSA severity as measured by total AHI and minimum SpO2. Supine-only OSA was more common than REM-only OSA and was associated with more severe markers of OSA (AHI) in an older and more obese population using multivariate analysis, although a trend of increased mean body weight was only observed in the supine-only OSA group (Table 2). Thus, greater attention should be paid to identifying supine-only OSA. The prevalence of REM-related OSA in this study (10%) was consistent with the results of Boujaoude et al. [16] who reported a prevalence of 13% using similar definitions. This study extends the work of Boujaoude et al. by comparing the REM-only and supine-only OSA groups. Importantly, those subjects with REM-only OSA in the present study had mild severity of OSA overall, as indicated by a mean AHI <10 events/h. These findings have important implications in clinical sleep medicine, namely monitoring with accurate diagnosis and management. Particular attention was undertaken in this study to use simultaneous digital video recordings of patients to verify body position during the sleep studies. However, since the development of the modern diagnostic sleep laboratory, most ‘‘quality assurance’’ attention in terms of polysomnography has been based around sleep stage monitoring, mainly for the exclusion of wakefulness and the identification of REM sleep, arousals and other sleep stages rather than monitoring body position. To the authors’ knowledge, there are no calibration details or guidelines from peer groups regarding body position sensors that are capable of accurate mea-
878
A. Gillman et al. / Sleep Medicine 13 (2012) 875–878
surement throughout the night. Importantly, the assessment of body position without video monitoring remains to be determined. This study found that supine-only OSA patients tended to be younger, less obese, and to have less severe OSA compared with patients with non-supine OSA [2]. Using the stricter definition of supine-only OSA may be considered important for the following reasons. First: a strict supine-only OSA definition identifies patients with no disease or mild disease in the non-supine position who would be suitable candidates for positional therapy [3,8,9]. Cartwright et al. [3] trained 10 patients with severe positional OSA (mean supine AHI 72 events/h, mean non-supine AHI 19 events/h) to sleep a further single night in the non-supine position using an alarmed motion sensor system. Polysomnography was used on both nights. The alarm resulted in a decrease in average supine sleep time from 51% to 2% of total sleep time, and a decrease in total AHI from 55 to 24 events/h. However, moderate persistant disease occurred, presumably due to the high non-supine AHI at baseline. In another short-term study comparing positional therapy with CPAP with 13 patients with mild-to-moderate supine-only OSA, Jokic et al. [8] demonstrated a significant improvement in total AHI from 17 to 10 events/h with positional therapy, and to three events/h with CPAP. Finally, a more recent randomized controlled four-week crossover study by Skinner et al. [9] compared positional therapy (using a thoracic antisupine band) with CPAP in 20 functional supine OSA patients (mean age 56 years, BMI 31 kg/m2, non-supine AHI < 10 events/h). Similar to Jokic et al. [8] and Cartwright et al. [7], they observed a decrease in overall AHI (based upon unattended cardiopulmonary and positional monitoring) from 23 to 12 events/h with positional therapy, and a decrease to five events/h with CPAP. Positional therapy has been shown to decrease systemic blood pressure (from 133/78 to 127/75 mm Hg) in patients with supine-related OSA [11], and similar levels have been seen with CPAP [13]. A second reason for accurate identification of supine-only OSA is related to the prescription of oral appliances to treat OSA. Marklund et al. [6] demonstrated that successful apnoea reduction with a mandibular advancement splint is related to a non-supine AHI < 10 events/h on diagnostic testing. A third and final reason relates to CPAP pressure requirements. Oksenberg et al. [10] examined the effect of body position and sleep stage on optimal nasal CPAP pressure in OSA patients. The optimal CPAP was higher in the supine position in both REM and non-REM sleep, with a cascading pressure sequence of supine REM > supine non-REM > non-supine REM and non-supine non-REM sleep. Patients with positional OSA are possibly more likely to respond to autotitrating CPAP devices; this hypothesis needs further clarification. In summary, this study demonstrated that supine-only OSA is common, occurring in 23–63% of a standard OSA population (depending upon the definition used); REM-only OSA is much less
common (10%). Supine-only OSA has a greater effect on the overall severity of OSA than REM-only OSA. Finally, accurate identification of supine-only OSA with monitoring is required if therapies such as positional therapy, mandibular advancement splints, or autotitrating CPAP devices are to be considered. 5. Conflict of Interest The ICMJE Uniform Disclosure Form for Potential Conflicts of Interest associated with this article can be viewed by clicking on the following link: http://dx.doi.org/10.1016/j.sleep.2012.01.016.
References [1] Findley L, Wilhoit S, Suratt P. Apnea duration and hypoxemia during REM sleep in patients with OSA. Chest 1985;8:432–6. [2] O’Connor C, Thornley K, Hanly P. Gender differences in the PSG features of OSA. Am J Respir Crit Care Med 2000;161:1465–72. [3] Cartwright RD. Effect of sleep position on sleep apnoea severity. Sleep 1984;7:110–4. [4] Szollosi I, Roebuck T, Thompson B, Naughton MT. Lateral sleeping position promotes ventilatory stability in central sleep apnea/Cheyne-Stokes respiration. Sleep 2006;29:1045–51. [5] Oksenberg A, Silverberg DS, Arons E, Radwan H. Positional vs non positional obstructive sleep apnoea patients: anthropomorphic, nocturnal polysomnographic and multiple sleep latency test data. Chest 1997;112:629–39. [6] Marklund M, Persson M, Franklin K. Treatment success with mandibular advancement device is related to supine-dependent sleep apnea. Chest 1998;114:1630–5. [7] Cartwright RD, Lloyd S, Lilie J, Kravitz H. Sleep position training for sleep apnea syndrome: a preliminary study. Sleep 1985;8:87–94. [8] Jokic R, Klimaszewski A, Crossley M, Sridharg G, Fitzpatrick MF. Positional treatment vs CPAP in patients with obstructive sleep apnoea syndrome. Chest 1999;115:771–81. [9] Skinner M, Kingshott R, Filsell S, et al. Efficacy of the tennis ball technique versus nCPAP in the management of position-dependent OSAS. Respirology 2008;13:708–15. [10] Oksenberg A, Silverberg D, Arons E, Radwan H. The sleep supine position has a major effect on optimal nasal continuous positive airway pressure. Chest 1999;116:1000–6. [11] Berger M, Oksenberg A, Silverberg DS, Arons E, Radwan H, Iaina A. Avoiding the supine position during sleep lowers 24 h blood pressure in obstructive sleep apnea (OSA) patients. J Hum Hypertens 1997;11:657–64. [12] Haba-Rubio J, Janssens J, Rochat T, Sforza E. Rapid eye movement-related disordered breathing. Chest 2005;128:3350–7. [13] Pepperell JC, Ramdassingh-Dow S, Crosthwaite N, Mullins R, Jenkinson C, Stradling JR, et al. Ambulatory blood pressure after therapeutic and subtherapeutic nasal continuous positive airway pressure for obstructive sleep apnoea: a randomised parallel trial. Lancet 2002;359:204–10. [14] Young T, Finn L, Peppard PE, Szklo-Coxe M, Austin D, Nieto FJ, et al. Sleepdisordered breathing and mortality: 18 year follow-up of the Wisconsin Sleep Cohort. Sleep 2008;31:1071–8. [15] Mador M, Kufel T, Magalang U, Rajesh SK, Watwe V, Grant BJB. Prevalence of positional sleep apnoea in patients undergoing polysomnography. Chest 2005;128:2130–7. [16] Boujaoude ZC, Abouzgheib WB, Aboujaoude RN, Dagher HN, Pratter MR, Akers SM, et al. The frequency of REM specific OSA. Chest 2003;124(4):74S.
Contents lists available at SciVerse ScienceDirect
Sleep Medicine journal homepage: www.elsevier.com/locate/sleep
Original Article
Comparison of supine-only and REM-only obstructive sleep apnoea Andrew Gillman a, Teanau Roebuck a, Sally Ho a, Esther van Braak a, Matthew T. Naughton a,b,⇑ a b
Department of Allergy, Immunology and Respiratory Medicine, Alfred Hospital, Melbourne, Victoria, Australia Monash University, Melbourne, Victoria, Australia
a r t i c l e
i n f o
Article history: Received 8 August 2011 Received in revised form 11 December 2011 Accepted 22 January 2012 Available online 24 May 2012 Keywords: Sleep apnoea Supine position REM sleep Digital monitoring Clinic population Therapy
a b s t r a c t Objectives: The effect of body position and sleep state on sleep apnoea have major clinical implications in the management of patients, yet are infrequently reported in the scientific literature. The aim of this study was to compare and contrast the prevalence and severity of supine-only and rapid eye movement (REM)-only obstructive sleep apnoea (OSA) in a population. Methods: Prospective cohort analysis of the influence of supine body position and REM sleep on the severity of apnoea in 100 consecutive patients with OSA (apnoea–hypopnoea index [AHI] > 5) using attended polysomnography with continuous digital monitoring in an accredited sleep laboratory. Supine-only OSA was defined as a supine:non-supine AHI ratio of >2:1 and non-supine AHI <5 events/ h. REM-only OSA was defined as an REM:non-REM ratio of >2:1 and non-REM AHI <5 events/h. Results: Supine sleep time represented a greater proportion of total sleep time than REM sleep time (40% vs 13%). The prevalence of supine-only OSA was more than twofold greater than that of REM-only OSA (23% and 10%, respectively). The supine-only group had greater overall AHI (mean 12.6 ± 6.1 vs 7.2 ± 2.2 events/h; P < 0.01) than the REM-only group. No significant differences in gender, age, or sleepiness were found between the two groups. Conclusions: Supine-only OSA is more common and is associated with a greater AHI than REM-only OSA. Ó 2012 Elsevier B.V. All rights reserved.
1. Introduction Considerable variance in the severity of obstructive sleep apnoea (OSA), as measured by the apnoea–hypopnoea index (AHI), can be explained by several factors, which include the effects of either sleep state or body position. The effects of sleep state upon upper airway and respiratory pump muscle activity and cortical arousability are well known. Of the various sleep states, rapid eye movement (REM) sleep has long been known to have a significant effect on OSA, with greater length of apnoeas, greater hypoxaemia and greater hypercapnia than apnoeas during stages 1 + 2 non-REM sleep [1]. In contrast, slow wave sleep is characterized by long periods of snoring with trivial hypoxaemia and an absence of arousals or discreet (obstructive or central) apnoeic events. In stages 1 and 2 non-REM sleep, however, sleep arousals are commonplace, leading to instability of respiratory control and both obstructive and central apnoeas. Thus, the severity of sleep apnoea in non-REM sleep is often considerably less than that observed in REM sleep. One study found that patients with OSA restricted to REM sleep were more likely to be female, to have less severe OSA, and to be less obese [2]. ⇑ Corresponding author at: Department of Allergy, Immunology and Respiratory Medicine, Alfred Hospital, P.O. Box 315, Prahran 3181, Victoria, Australia. Tel.: +61 3 9076 3770; fax: +61 3 9076 3601. E-mail address: [email protected] (M.T. Naughton). 1389-9457/$ - see front matter Ó 2012 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.sleep.2012.01.016
Similarly, body position has a strong influence on OSA. A supine position and its associated gravitational effects on the tongue and mandible make the upper airway more collapsible, which is thought to be responsible for the worsening of OSA compared with a non-supine position [3]. Similar body positional effects in OSA have been observed in central sleep apnoea associated with heart failure, where the effects appear to be pulmonary rather than upper airway in origin [4]. Compared with non-supine OSA patients, patients with supine-only OSA tend to be younger, male, and to have a lower body mass index (BMI) [5]. The importance of identifying supine-only OSA is underscored by: (a) its association with milder disease compared with non-supine OSA (i.e., lower overall AHI) [5]; (b) a greater therapeutic response to oral appliances [6] and positional therapy [7–9]; and (c) the impact on the optimal fixed and autotitrating continuous positive airway pressure (CPAP) settings [10]. A further reason to make a distinction between REM-only and supine-only OSA is borne by the therapeutic option of positional therapy. In patients with supine-related OSA, two studies have suggested that positional therapy (i.e., instructing patients to avoid the supine position) is associated with a reduction in systemic blood pressure [11]. However, the definitions of REM OSA [12] and supine OSA [3,5– 7,9,10] in previous published papers have been inconsistent. In general, the definitions have been based upon a 2:1 ratio of either REM:non-REM AHI or supine:non-supine AHI. This variation in def-
876
A. Gillman et al. / Sleep Medicine 13 (2012) 875–878
inition significantly influences the prevalence of both REM-only and supine-only OSA. Moreover, as the inflection point of the total AHI is approximately 5 events/h for the development of cardiovascular and other complications and mortality [14], it is important to identify patients in whom positional therapy may be a safe option. Although similarities exist in the clinical descriptions of REMonly and supine-only OSA patients (i.e., younger, lower BMI) compared with non-REM and non-supine OSA patients, the difference in severity of OSA between REM-only and supine-only patients using similar definitions is unknown. The aims of this study were twofold. The first aim was to compare the clinical features of supine-only OSA with REM-only OSA using strict functional definitions to assess gender bias, BMI, and overall OSA severity based upon AHI and minimum oxygen saturation (SpO2). The second aim was to assess the prevalence of supineonly OSA using a strict functional definition (supine:non-supine event ratio >2 plus a non-supine AHI <5 events/h) and a lenient, more traditional definition (supine:non-supine AHI ratio >2) in a population of patients with suspected OSA attending a university-based sleep clinic.
for sleep stage using standard criteria, as described previously [4]. Apnoeas were defined as an absence of air flow for at least 10 s. Hypopnoeas were defined as a discernable fall in respiratory effort or flow associated with either a P3% SpO2 desaturation or an arousal. Body position was assessed with a Compumedics positional sensor device with digital video correction. For inclusion in the study, subjects had to sleep for at least 4 h with at least 15 min of REM sleep and to spend at least 30 min in a supine position and 30 min in a non-supine position during sleep. Supine-only and REM-only OSA were defined as AHI <5 (i.e., strict definition). Supine-only OSA could involve any sleep stage, and REM-only OSA could involve any body position. Supine-only OSA was subdivided using a lenient definition (2:1 AHI ratio) and a strict definition (AHI < 5 events/h). Data are presented as mean ± standard deviation. Unpaired ttests were used for comparison between groups. Multivariate analysis was used to assess the relative and significant contributions of positional- and REM-related AHI towards the overall AHI using multiple variant analysis. P < 0.05 was taken to indicate significance.
2. Methods
3. Results
Consecutive subjects of either gender aged >18 years, diagnosed with OSA (AHI > 5 events/h) using laboratory-attended polysomnography at an Australian-Sleep-Association-accredited sleep laboratory within a large university based teaching hospital, were enrolled in the study. Patients with known cardiac, pulmonary or neurological disease were excluded. In total, 100 subjects were identified. Approval was obtained from the Alfred Ethics Committee and patients provided informed consent. Polysomnography (Compumedics, Abbottsford, Australia) included electro-encephalography (C3A2, C4A1, O3A2, O4A1), submental electromyography (EMG), electro-oculography, electrocardiography, chest and abdominal movement using inductance plethysmography, oronasal thermistor, nasal pressure, digital video observation of body position, SpO2 (3-s averaging time), and anterior tibialis EMG. All sleep studies were scored manually
The cohort was aged 53.6 ± 12.9 years with a BMI of 31.5 ± 8.6 kg/m2 and Epworth Sleepiness Scale (ESS) score of 8.9 ± 5.3. Sixty-nine percent of participants were male. Patients slept for 351 ± 6 min, with 40% (138 min) of total sleep time in a supine position and 13% (46 min) of total sleep time in REM sleep (Table 1). The patients slept with 1.7 ± 0.5 pillows. Typical polysomnograph examples of REM-only and supine-only OSA are shown in Fig. 1. The AHI values were 29.6 ± 23.2 events/h (overall), 43.2 ± 27 events/h (supine), 20.9 ± 23 events/h (non-supine), 33.5 ± 21 events/h (REM) and 28.4 ± 24.7 events/h (non-REM) (Table 1). The corresponding minimum SpO2 values were 88.9%, 88.2%, 90.4%, 88.3%, and 88.6%. The proportion of total sleep time spent in a supine position was significantly greater than the proportion of total sleep time spent in REM sleep (40% vs 13%). Body
Fig. 1. Polysomnographic summary graphs of two male patients of similar age (45 years) and body mass index (BMI): one with rapid eye movement (REM)-only obstructive sleep apnoea (OSA) and the other with supine-only OSA.
877
A. Gillman et al. / Sleep Medicine 13 (2012) 875–878 Table 1 Baseline demographics of entire group (n = 100).
Table 3 Comparison of lenient and strict definitions of supine-only obstructive sleep apnoea.
Age (years) Gender (M:F) Body mass index (kg/m2) Epworth Sleepiness Scale score Total sleep time (hours:minutes)
53.6 ± 12.9 69:31 31.5 ± 8.6 8.9 ± 5.3 5:51 ± 6
AHI (events/h) Total Supine Non-supine REM Non-REM
29.6 ± 23.2 43.2 ± 27.3 20.9 ± 22.9 33.5 ± 21.2 28.4 ± 24.7
Supine sleep time (% total sleep time) Number of pillows
39.6% 1.7 ± 0.5
AHI, apnoea–hypopnoea index; REM, rapid eye movement. Data are mean ± standard deviation.
Table 2 Comparison of rapid eye movement (REM)-only and supine-only obstructive sleep apnoea (OSA) using the strict definition (apnoea–hypopnoea index [AHI] < 5 events/ h).
Prevalence (%) Age (years) Males (%) Body mass index (kg/m2) Epworth Sleepiness Scale score Total AHI (events/h) Supine AHI (events/h) Non-supine AHI (events/h) Minimum oxygen satuation (%)
REM-only OSA
Supine-only OSA
P
10 44.3 ± 9.0 80 ± 4 28.1 ± 5.8 7.4 ± 4.7 7.2 ± 2.2 8.4 ± 4.5 5.5 ± 4.3 89 ± 1.6
23 50.2 ± 12.3 87 ± 3 30.2 ± 11.9 9.2 ± 6.1 12.6 ± 6.1 28 ± 16.4 3.1 ± 2.4 88 ± 1.3
<0.01 0.22 0.62 0.65 0.4 0.01 0.001 0.048 ns
ns, not significant. Data are mean ± standard deviation.
position and REM sleep explained the variance in overall AHI, whereas minimum SpO2 was explained by position but not REM sleep. The prevalence of supine-only OSA was approximately twofold greater than the prevalence of REM-only OSA (23% vs 10%) (Table 2). The overall AHI in the supine-only OSA group was significantly greater than that of the REM-only group (12.6 ± 6.1 vs 7.2 ± 2.2 events/h; P = 0.01). The supine-only and REM-only OSA patients were of similar age (50 ± 12 vs 44 ± 9 years; P = 0.22), similar weight (BMI 30 ± 12 vs 28 ± 6 kg/m2; P = 0.65), had similar symptoms of sleepiness (ESS score 9.2 ± 6.1 vs 7.4 ± 4.7; P = 0.4), and a similar proportion were male (71% vs 69%). There was a significant variance in the prevalence of supine-only OSA dependent upon the definition used for non-supine, despite all patients having a supine:non-supine ratio of >2. The prevalence varied from 23% (using a strict definition of non-supine AHI of <5 events/h) to 63% (using a more lenient definition where non-supine AHI was unstated) (Table 3). Within the supine-only OSA group, the lenient definition group had a significantly lower overall AHI and a trend to be younger with a lower BMI compared with the strictly defined non-supine OSA group, although there were no differences in the supine sleep time or the mean supine AHI compared with the non-supine OSA group. 4. Discussion This study examined the prevalence and clinical features of supine-only and REM-only OSA. There were several novel and clinically relevant findings. First, the supine sleep time was longer than the REM sleep time (40% vs 13%). Second, the prevalence of
Prevalence (%) Age (years) Males (%) Body mass index (kg/m2) Epworth Sleepiness Scale score Total AHI (events/h) Supine AHI (events/h) Non-supine AHI (events/h) Minimum oxygen saturation (%)
Lenient definition
Strict definition
63 53 71 31 9.8 23 43 11 83
23 50 87 30 9.1 12 28 3 88
P
ns ns ns ns <0.01 <0.01 <0.01 <0.01
AHI, apnoea–hypopnoea index; ns, not significant. Lenient definition: overall apnoea–hypopnoea index (AHI) > 5 events/h, and supine:non-supine AHI ratio > 2. Strict definition: overall AHI > 5 events/h and supine:non-supine AHI ratio > 2 plus non-supine AHI < 5 events/h.
supine-only OSA was greater than the prevalence of REM-only OSA. Third, the overall severity of OSA was greater in supine-only OSA than in REM-only OSA. Finally, when the various published definitions of supine-only OSA were used, the prevalence varied from 23% (strict definition) to 63% (lenient definition); these values are similar to those published in the literature, namely 27% [7] and 50–60% [5]. The prevalence of supine-only OSA in this study varied from 23% to 63% depending upon the definition used. Mador et al. [15] previously introduced a stricter definition of supine-only OSA with a non-supine AHI of <5 events/h. Based upon 326 patients studied with polysomnography in two centres (17% had split night, balance full night diagnostic) and body position by direct observation, they demonstrated that 27% of their cohort of OSA patients met this criterion for supine-only OSA, similar to the present value of 23%. The authors felt that split night studies did not provide sufficient time for a sufficient mixture of supine and non-supine sleep. The finding that body position is a more important determinant than sleep stage on overall AHI (and thus OSA severity) is novel. Multivariate analysis showed that body position had a more important effect than sleep stage on OSA severity as measured by total AHI and minimum SpO2. Supine-only OSA was more common than REM-only OSA and was associated with more severe markers of OSA (AHI) in an older and more obese population using multivariate analysis, although a trend of increased mean body weight was only observed in the supine-only OSA group (Table 2). Thus, greater attention should be paid to identifying supine-only OSA. The prevalence of REM-related OSA in this study (10%) was consistent with the results of Boujaoude et al. [16] who reported a prevalence of 13% using similar definitions. This study extends the work of Boujaoude et al. by comparing the REM-only and supine-only OSA groups. Importantly, those subjects with REM-only OSA in the present study had mild severity of OSA overall, as indicated by a mean AHI <10 events/h. These findings have important implications in clinical sleep medicine, namely monitoring with accurate diagnosis and management. Particular attention was undertaken in this study to use simultaneous digital video recordings of patients to verify body position during the sleep studies. However, since the development of the modern diagnostic sleep laboratory, most ‘‘quality assurance’’ attention in terms of polysomnography has been based around sleep stage monitoring, mainly for the exclusion of wakefulness and the identification of REM sleep, arousals and other sleep stages rather than monitoring body position. To the authors’ knowledge, there are no calibration details or guidelines from peer groups regarding body position sensors that are capable of accurate mea-
878
A. Gillman et al. / Sleep Medicine 13 (2012) 875–878
surement throughout the night. Importantly, the assessment of body position without video monitoring remains to be determined. This study found that supine-only OSA patients tended to be younger, less obese, and to have less severe OSA compared with patients with non-supine OSA [2]. Using the stricter definition of supine-only OSA may be considered important for the following reasons. First: a strict supine-only OSA definition identifies patients with no disease or mild disease in the non-supine position who would be suitable candidates for positional therapy [3,8,9]. Cartwright et al. [3] trained 10 patients with severe positional OSA (mean supine AHI 72 events/h, mean non-supine AHI 19 events/h) to sleep a further single night in the non-supine position using an alarmed motion sensor system. Polysomnography was used on both nights. The alarm resulted in a decrease in average supine sleep time from 51% to 2% of total sleep time, and a decrease in total AHI from 55 to 24 events/h. However, moderate persistant disease occurred, presumably due to the high non-supine AHI at baseline. In another short-term study comparing positional therapy with CPAP with 13 patients with mild-to-moderate supine-only OSA, Jokic et al. [8] demonstrated a significant improvement in total AHI from 17 to 10 events/h with positional therapy, and to three events/h with CPAP. Finally, a more recent randomized controlled four-week crossover study by Skinner et al. [9] compared positional therapy (using a thoracic antisupine band) with CPAP in 20 functional supine OSA patients (mean age 56 years, BMI 31 kg/m2, non-supine AHI < 10 events/h). Similar to Jokic et al. [8] and Cartwright et al. [7], they observed a decrease in overall AHI (based upon unattended cardiopulmonary and positional monitoring) from 23 to 12 events/h with positional therapy, and a decrease to five events/h with CPAP. Positional therapy has been shown to decrease systemic blood pressure (from 133/78 to 127/75 mm Hg) in patients with supine-related OSA [11], and similar levels have been seen with CPAP [13]. A second reason for accurate identification of supine-only OSA is related to the prescription of oral appliances to treat OSA. Marklund et al. [6] demonstrated that successful apnoea reduction with a mandibular advancement splint is related to a non-supine AHI < 10 events/h on diagnostic testing. A third and final reason relates to CPAP pressure requirements. Oksenberg et al. [10] examined the effect of body position and sleep stage on optimal nasal CPAP pressure in OSA patients. The optimal CPAP was higher in the supine position in both REM and non-REM sleep, with a cascading pressure sequence of supine REM > supine non-REM > non-supine REM and non-supine non-REM sleep. Patients with positional OSA are possibly more likely to respond to autotitrating CPAP devices; this hypothesis needs further clarification. In summary, this study demonstrated that supine-only OSA is common, occurring in 23–63% of a standard OSA population (depending upon the definition used); REM-only OSA is much less
common (10%). Supine-only OSA has a greater effect on the overall severity of OSA than REM-only OSA. Finally, accurate identification of supine-only OSA with monitoring is required if therapies such as positional therapy, mandibular advancement splints, or autotitrating CPAP devices are to be considered. 5. Conflict of Interest The ICMJE Uniform Disclosure Form for Potential Conflicts of Interest associated with this article can be viewed by clicking on the following link: http://dx.doi.org/10.1016/j.sleep.2012.01.016.
References [1] Findley L, Wilhoit S, Suratt P. Apnea duration and hypoxemia during REM sleep in patients with OSA. Chest 1985;8:432–6. [2] O’Connor C, Thornley K, Hanly P. Gender differences in the PSG features of OSA. Am J Respir Crit Care Med 2000;161:1465–72. [3] Cartwright RD. Effect of sleep position on sleep apnoea severity. Sleep 1984;7:110–4. [4] Szollosi I, Roebuck T, Thompson B, Naughton MT. Lateral sleeping position promotes ventilatory stability in central sleep apnea/Cheyne-Stokes respiration. Sleep 2006;29:1045–51. [5] Oksenberg A, Silverberg DS, Arons E, Radwan H. Positional vs non positional obstructive sleep apnoea patients: anthropomorphic, nocturnal polysomnographic and multiple sleep latency test data. Chest 1997;112:629–39. [6] Marklund M, Persson M, Franklin K. Treatment success with mandibular advancement device is related to supine-dependent sleep apnea. Chest 1998;114:1630–5. [7] Cartwright RD, Lloyd S, Lilie J, Kravitz H. Sleep position training for sleep apnea syndrome: a preliminary study. Sleep 1985;8:87–94. [8] Jokic R, Klimaszewski A, Crossley M, Sridharg G, Fitzpatrick MF. Positional treatment vs CPAP in patients with obstructive sleep apnoea syndrome. Chest 1999;115:771–81. [9] Skinner M, Kingshott R, Filsell S, et al. Efficacy of the tennis ball technique versus nCPAP in the management of position-dependent OSAS. Respirology 2008;13:708–15. [10] Oksenberg A, Silverberg D, Arons E, Radwan H. The sleep supine position has a major effect on optimal nasal continuous positive airway pressure. Chest 1999;116:1000–6. [11] Berger M, Oksenberg A, Silverberg DS, Arons E, Radwan H, Iaina A. Avoiding the supine position during sleep lowers 24 h blood pressure in obstructive sleep apnea (OSA) patients. J Hum Hypertens 1997;11:657–64. [12] Haba-Rubio J, Janssens J, Rochat T, Sforza E. Rapid eye movement-related disordered breathing. Chest 2005;128:3350–7. [13] Pepperell JC, Ramdassingh-Dow S, Crosthwaite N, Mullins R, Jenkinson C, Stradling JR, et al. Ambulatory blood pressure after therapeutic and subtherapeutic nasal continuous positive airway pressure for obstructive sleep apnoea: a randomised parallel trial. Lancet 2002;359:204–10. [14] Young T, Finn L, Peppard PE, Szklo-Coxe M, Austin D, Nieto FJ, et al. Sleepdisordered breathing and mortality: 18 year follow-up of the Wisconsin Sleep Cohort. Sleep 2008;31:1071–8. [15] Mador M, Kufel T, Magalang U, Rajesh SK, Watwe V, Grant BJB. Prevalence of positional sleep apnoea in patients undergoing polysomnography. Chest 2005;128:2130–7. [16] Boujaoude ZC, Abouzgheib WB, Aboujaoude RN, Dagher HN, Pratter MR, Akers SM, et al. The frequency of REM specific OSA. Chest 2003;124(4):74S.