Central sleep apnea syndrome (CSAS) is a clinical entity characterized by the temporary absence or diminution of respiratory drive coming from the respiratory center during sleep.1 Central sleep apnea (CSA) occurs in less than 5% of subjects admitted to sleep clinics.2 CSA could be primary (idiopathic) or secondary to medical conditions (e.g. heart failure, renal failure, cerebrovascular disease), high altitude, opioids, and drug use.3 Primary CSA is diagnosed by exclusion of any identifiable cardiac and neurologic cause in patients with CSA. Primary CSA seems to be driven by elevated chemosensitivity to PCO2. Circulation delay is defined as normal in these patients and is unlikely to contribute to CSA.4

Obstructive sleep apnea (OSA) is a more common and well-described form of sleep-disordered breathing (SDB) than CSA. One of the most striking features of OSA is that the respiratory events are particularly severe and frequent in the supine sleeping position.5 Indeed, supine-related OSA is a dominant phenotype of OSA with a prevalence of 20%–60% in the general population.6 Similar to OSA, in patients with medical conditions such as heart failure and stroke, the severity of CSA is increased when sleeping in the supine position compared with the lateral position.7,8 The possible reasons for supine-related worsening of central apneas in patients with medical conditions include upper airway and lung volume effects, and altering plant gain when in the supine position.9 Although the impact of sleeping position on patients with OSA and CSA with HF and stroke is well described, the effects of body position on primary CSA have not been commonly investigated. There are few reports in the literature of positional CSA in patients without cardiac history or congestive heart failure.10,11 Currently, phenotyping of SDB is widely discussed, but the lack of clinical studies considering primary CSA and positional dependence is evident.

In view of these findings, the purpose of the current study was to evaluate the impact of body position on primary central sleep apnea syndrome. Thereby, we also assessed the features of primary positional central sleep apnea (PCSA) and phenotypic approach to primary CSA.

Overall, 39 subjects who were diagnosed as having primary CSA were included in the study. A total of 61.5% of the subjects had the diagnosis of primary PCSA. 92.3% of the study population were males and the 7.7% were females. The mean age of the subjects was 45.5±10.3 years and the mean body mass index (BMI) was 30.2±3.9kg/m2. 17.9% of the subjects had hypertension (HT), 10.3% had diabetes mellitus (DM), and 25.6% had a history of septoplasty surgery. Only one subject had previously undergone tonsillectomy. Regarding sleep symptoms; 24 patients had the symptoms of snoring, witnessed apnea and daytime sleepiness, 11 patients had two of three symptoms, and four patients had only one symptom. The demographic characteristics of the study population are presented in Table 1.

There were no differences between the primary PCSA and non-PCSA groups regarding age, sex, BMI, ESS, and history of HT, DM, and septoplasty (Table 2). In terms of PSG parameters; AHI (P=.001), rapid eye movement (REM) AHI (P=.015), non-REM AHI (P=.001), oxygen desaturation index (P=.002), the time spent under 88% saturation during sleep (P=.003), and the index of obstructive apnea (P=.008), mixed apnea (P=.008) and central apnea (P=.004) was lower in the primary PCSA group than in the non-PCSA group. The duration of REM sleep was longer in the primary PCSA group compared with the non-PCSA group (Table 3). There was no difference between the two groups in terms of supine AHI, but there was a statistically significant difference in terms of non-supine AHI (P=.984, P<.001, respectively).

According to the PSG results, six patients (15.6%) had mild CSA, 14 (35.9%) had moderate CSA, and 19 patients (48.7%) had severe CSA. When we compared the mild-to-moderate CSA group with the severe CSA group, a statistically significant difference was found regarding the history of septoplasty surgery (P=.022); two (10%) patients had a history of septoplasty in the mild-to-moderate group compared with eight (42.1%) in the severe CSA group.

When the primary PCSA group was classified into two groups as supine-predominant and supine-isolated, 17 (71%) patients were in the supine-predominant group and 7 (29%) were in the supine-isolated group. The treatment methods applied to the subjects are illustrated in Table 4. Three of the seven patients who were recommended position therapy in the primary PCSA group could not adapt to the treatment. Therefore, PAP titration was also applied to these patients. Overall, 20 patients in the primary PCSA group were recommended PAP therapy, but three of those rejected PAP titration. Consequently 17% patients were applied position therapy and 71% received PAP titration (63% CPAP, 4% BPAP-ST, 4% ASV). In the non-PCSA group, PAP therapy was recommended to all patients, but one patient refused the therapy. The remaining 93% of patients in the non-PCSA group were responsive to PAP therapy (73% CPAP, 20% BPAP-ST). The mean CPAP pressure used for the responsive patients undergoing CPAP titration in the primary PCSA group and non-PCSA group was 8.00±2.44cm H2O and 8.36±1.91cm H2O, respectively.

In this study, we aimed to investigate the impact of body position on primary central sleep apnea syndrome. Our results demonstrated that primary PCSA was observed in 61.5% of patients with primary CSA. Additionally, primary PCSA was associated with milder disease severity compared with non-PCSA.

To our knowledge, this is the first study to evaluate the impact of body position on primary CSA syndrome and the features of primary PCSA. There are a limited number of studies on the relation between CSA and position in the literature. In these studies, the effect of body position in patients with CSA associated with hearth failure was evaluated.15,16 There are no prospective studies evaluating the relation between body position and primary CSA. Our results demonstrated that the primary PCSA ratio was high, similar to supine-related OSA. Pinna et al. reported that the severity of CSA decreased when the subjects were in the lateral position compared with the supine position in patients with congestive heart failure (CHF).15 Similar results were revealed in other studies.8,17 Several mechanisms have been proposed to explain the supine-related worsening of CSA in patients with CHF. Circulatory delay between the lungs and chemoreceptors was believed to play an important role in the pathogenesis of CSA in patients with CHF.4 However, the study conducted by Traversi et al. showed that the reduction of CSA severity from the supine position to the lateral position in patients with CHF was not due to the improvement in cardiac hemodynamics. They put forward that non-cardiac factors such as the change in functional residual capacity were likely to represent the main cause.16 Sleeping in the supine position can result in a reduction of both functional residual capacity and metabolic rate, which consequently enhances plant gain. Enhanced plant gain is defined as a large change in carbon dioxide levels relative to a small change in ventilation.11 The mechanism responsible for supine-related OSA is mainly related to the direct and indirect effects of gravity on upper airway collapsibility. Increased loop gain may also contribute to the pathogenesis of OSA.15 Reports on primary PCSA are rare as previously mentioned and the reason for positional dependency of primary CSA is not yet understood. The reduction in the functional residual capacity resulting in enhanced plant gain may also play a role in encouraging central apneas in patients with primary PCSA. Previous studies suggest that arousals play an important role in triggering and perpetuating central apneas in both HF and non-HF patients with CSA. Arousals can provoke a fall in PaCO2 below the apnea threshold that can result in central apnea.18 In this study there was no difference between primary PSCA and non-PCSA groups with regard to arousals index.

Primary PCSA is associated with milder disease severity, like positional OSA in this study. AHI, REM AHI, non-REM AHI, oxygen desaturation index, and the desaturation time spent under 88% during sleep were lower in patients with primary PCSA compared with those with non-PCSA. The duration of REM sleep was longer in patients with primary PCSA. Over the past decade, evidence for an association between REM sleep and neurologic and cardiovascular diseases has accumulated. Zhang et al. demonstrated that the proportion and duration of REM sleep were negatively associated with all-cause mortality.19 Primary PCSA is both associated with milder disease severity and longer REM sleep time compared with non-PCSA. These findings suggest that there may be a difference between primary PCSA and non-PCSA groups with regard to the risk of developing cardiometabolic complications. Hence, primary PCSA should be evaluated as a separate clinical entity.

The outstanding features of the patients with primary CSA in our study were their younger age, male predominance, and the presence of septoplasty history. Although there was no statistically significant difference, it was noteworthy that the presence of septoplasty history was higher in the primary PCSA group than in the non-PCSA group. Additionally, the history of septoplasty in patients with severe primary CSA was higher than in the mild-to-moderate group. There may be a link between nasal physiology and the development of CSA, which should be evaluated. Tanaka et al. proposed that after a switch to oral breathing during sleep, central respiratory events increase due to greater CO2 elimination during expiration.20 Nasal receptors responsive to air flow may be important in maintaining the rhythmicity of breathing and preventing airway collapse during sleep.21

Primary CSA has no proven standardized treatment. PAP therapy, use of acetazolamide, zolpidem, and triazolam are the treatment options for primary CSA; however, these drugs have limited supporting evidence.13 In this study, patients with primary PCSA were recommended PAP therapy as well as position treatment options. Position treatment was recommended principally to patients with supine isolated-PCSA. Four of these seven patients accepted position therapy as a treatment option. In patients with primary PCSA, 71% of the patients underwent PAP titration and 63% were responsive to CPAP. Unfortunately, the adherence of the patients with sleep apnea to the CPAP treatment was not sufficient.22 In clinical practice, positional therapy might be an alternative treatment in patients with primary PCSA. However, future research is required to validate this approach. PAP therapy was recommended to all patients in the non-PCSA group. The mean CPAP pressure was lower in the primary PCSA group than in the non-PCSA group.

Our study has some limitations that must be addressed. The number of patient population was too small to allow for subgroup analysis. However, it should be considered that the prevalence of primary CSA is low in the general population. Our hospital is a chest disease and chest surgery branch hospital. It is not a primary center for patients with serious cardiovascular diseases such as heart failure, renal failure, and cerebrovascular accidents. Therefore, rates of secondary CSA may be lower than those of primary CSA found in this study. Another major limitation of our study is that we could not measure the PC02 during sleep. We could only measure the PC02 values of the patients with a BMI 30 and above to rule out the obesity hypoventilation syndrome. PC02 values of 19 patients in whom BMI≥30kg/m2, was 38.2±1.20. None of the patients had the diagnosis of obesity hypoventilation syndrome. Complex sleep apnea syndrome (CompSAS) may be considered in patients with a history of septoplasty. The International Classification of Sleep Disorders (ICSD)-3 mentioned a case report of the development of central apnea following nasal surgery in a patient with known obstructive apnea.12 In this case, a PSG examination was performed 4 months after nasal surgery. Changes in carbon dioxide regulation have been implicated in the pathogenesis. It is proposed that central sleep apnea resolves in these patients over time, as seen in the vast majority of patients with CPAP-emergent central sleep apnea.23 Spontaneous resolution is reported in 50%–75% of PAP-emergent central apneas after several months of PAP therapy.24 Morgenthaler et al. showed that CompSAS resolved in about one-third to two-thirds of patients treated with CPAP after 90 days of therapy.25 In our study, history of septoplasty was obtained from the patient's own reports and was performed prior to admission to our clinic. None of the patients had undergone PSG examinations before surgery. There were 3.3±1.15 years between the nasal surgery and PSG examination. ICSD-3 also emphasized that treatment induced central apnea resolves with time, but the time course for this resolution is unclear. Longitudinal observation of the patients in our study could not be evaluated because it was not included in the study protocol.

It is important to reveal the relationship between primary CSA and body position. In this study, it was demonstrated that the rate of primary PCSA was high (61.5%) and primary PCSA was associated with milder disease severity. The classification of patients with primary CSA regarding positional dependency may be helpful in terms of developing clinical approaches and treatment recommendations. Further clinical studies are required to validate this approach.

The authors declare that they have no conflict of interest.