Effectiveness of Adaptive Servo Ventilation in the treatment of hypocapnic central sleep apnea of various etiologies

Abstract

Background

Central sleep apnea (CSA) occurs in clinical situations that induce hypocapnia and respiratory instability during sleep. This is true, not only in heart failure patients, but also in patients suffering from neurological diseases and idiopathic CSA. Adaptive Servo Ventilation (ASV) is frequently prescribed in France for the treatment of CSA, but only a few studies have evaluated ASV treatment with regards to long term effectiveness and compliance.

Methods

Retrospective chart review in two French centers of the outcome of 74 CSA patients treated by ASV with a mean follow up on ASV of 36 ± 18 months.

Results

Thirty-three of the 74 patients suffered from CSA related to heart failure (HF), whereas the 41 others exhibited CSA mainly associated with neurological disorders or idiopathic CSA. Mean ASV compliance was 5.2 ± 2.6 and 5.9 ± 2.9 h per night in cardiac failure and non-cardiac failure patients, respectively. All patients significantly improved their apnea + hypopnea index (from 47.4 ± 19.8 to 6.9 ± 9.3/h [p < 0.001]) and mean nocturnal SaO₂ (from 92.1 ± 2.6% to 93.6 ± 3.2% [p < 0.001]). The Epworth sleepiness scale score was reduced from 10.2 ± 5.2 to 6.5 ± 3.9 (p < 0.01) in compliant patients but not in non-compliant patients (less than 3 h per night). Moreover, compliant cardiac failure patients demonstrated a significant improvement in their NYHA score [p < 0.05]. Lastly, ASV significantly reduced chronic hyperventilation as assessed by blood gases.

Conclusion

Our findings suggest that ASV is well tolerated and effective for most patients with hypocapnic central sleep apnea and chronic hyperventilation.

Introduction

The pathophysiology and the prevalence of the various forms of central sleep apnea (CSA) vary greatly [1]. In clinical practice, two main categories of central sleep apnea can be distinguished by diurnal blood gases and chemosensitivity. Patients with blunted chemosensitivity exhibit daytime hypercapnia and then, during sleep, the removal of the wakefulness drive worsens hypercapnia leading to hypercapnic CSA. On the opposite end of the spectrum are hypocapnic CSA patients with a PaCO₂ close to the apneic threshold [2]. In these subjects, respiratory control system instability is characterized by elevated hypercapnic ventilatory responses leading to daytime hypocapnia. During sleep, any arousal will result in a ventilatory overshoot because of exaggerated ventilatory responses. PaCO₂ then decreases and crosses the apnea threshold leading to central respiratory events inducing micro-arousals and, thus, to the perpetuation of breathing instability. These are the important pathophysiological traits occurring in CSA associated with chronic cardiac failure (CHF) [3], but also with idiopathic CSA [4] and some neurological disorders [5].

Sleep breathing disorders occur commonly in CHF patients and seem to be associated with deterioration in cardiac function and potentially with increased mortality [6], [7], [8]. Approximately one third of CHF patients present with CSA and Cheyne–Stokes respiration (CSA/CSR), one third with obstructive sleep apnea, and the others are free of sleep breathing disorders [9]. Cardiac failure and the resulting pulmonary congestion activate lung vagal irritant receptors, thus leading to hyperventilation, hypocapnia, and CSA [3]. Continuous positive airway pressure (CPAP) has been proposed to suppress CSA/CSR and improve prognosis in CHF patients. Open studies or small randomized controlled trials suggest that CPAP use over 1–3 months alleviates CSA–CSR [10], [11], increases LVEF [12], and reduces adrenergic tone [13]. CPAP stabilizes ventilation by improving cardiovascular function and the added dead space elevates PaCO₂. The CANPAP trial, the first large scale randomized controlled trial treating CHF patients by CPAP, did not confirm any reduction in mortality [14]. Nevertheless, a post hoc analysis demonstrated a significant improvement in mortality in the subgroup of the CSA–CSR population effectively treated by CPAP when compared to CPAP-non-responders [15]. Adaptive Servo Ventilation (ASV) may be more effective on central sleep respiratory disturbances than CPAP and associated with a higher rate of long term compliance [16].

ASV delivered in the present study by AutoSet CS™² provides baseline ventilatory support in addition to a fixed end-expiratory pressure and a default back-up rate of 15 breaths/min. Depending on the setting, inspiratory pressure servo-adapts in order to maintain ventilation at 90% of a running 3-min reference period. The inspiratory pressure will increase or decrease in mirror of patient respiratory efforts to avoid ventilatory instability. Thus, as CPAP, ASV increases overnight PaCO₂ [17] and stabilizes ventilation by eliminating apneas and hypopneas [16], [17], [18], but may have a lesser impact on intra thoracic hemodynamics. ASV is the first line of treatment for sleep breathing disorder in CHF patients [16], [17], [18], but has also been suggested as potentially effective in other causes of CSA such as stroke or other neurological diseases, in patients chronically treated with opioids [19], and in patients suffering from idiopathic CSA breathing [20]. Apart from studies in CHF, previous reports are small case series and very few studies have evaluated ASV treatment in terms of feasibility, effectiveness, and compliance in clinical cohorts [21].

The aim of our study was to evaluate clinical outcomes and long term compliance in a real life, large cohort of patients using ASV for CHF-related CSA as well as hypocapnic CSA from other causes.