LAUSR

Sleep-disordered breathing (SDB) is a highly prevalent co-morbidity in patients with heart failure (HF) and can manifest as both obstructive (OSA) and central sleep apnoea (CSA).1 The latter shows a typical periodic breathing pattern (waxing and waning of tidal volume alternating with apnoeas or hypopnoeas) commonly referred to as Cheyne–Stokes respiration (CSR-CSA). Periodic breathing patterns have also been reported as highly frequent during daytime short-term laboratory recordings in HF patients.2,3 Previous studies suggest that both daytime periodic breathing (PB_day) and CSR-CSA share the same primary pathophysiological mechanism, namely an instability in the chemical control system that regulates breathing.4,5 Only few studies, however, have analysed the association between PB_day and CSR-CSA, leading to conflicting results.3,6 Accordingly, in this study, we aimed to (i) assess the relationship between PB_day and CSR-CSA, and (ii) determine whether the occurrence of PB_day can be used as a screening tool for the identification of patients with CSR-CSA. We studied 339 patients with HF and no current use of therapy for OSA/CSA, who were consecutively admitted to our Heart Failure Unit for assessment and therapy of HF. Patients were studied at pre-discharge in clinically stable condition (i.e. on oral therapy). The study was approved by our institutional Review Board and Ethics Committee. All patients underwent a 10 min, daytime, supine respiratory recording and an overnight polysomnography on the same or subsequent night. Respiratory studies were carried out in a quiet environment, in the morning, after a light breakfast avoiding coffee intake, as previously reported.2 PB_day was defined as a cyclic rise and fall in tidal volume alternating with apnoea or hypopnoea, with a greater than 25% variation in peak to through values and occurring in more than 75% of the 10 min recording2 (Figure 1). Standard polysomnography was performed unattended in the patient’s own bed using the polysomnograph Embla titaniumTM (Embla, Thornton, CO, USA) and was scored according to the American Academy of Sleep Medicine recommendations.7 The severity of SDB was quantified by (i) the apnoea–hypopnoea index (AHI), (ii) the mean and minimum value of related oxygen desaturations, and (iii) the total time spent with an oxygen saturation< 90% (T90). The AHI was used both as continuous and dichotomized (AHI ≥15 events/h) variable. Each patient was classified as having positional sleep apnoea if there was a> 50% reduction in the AHI between the supine and the lateral position.8 Between-group comparisons were carried out by the t-test for independent samples, the chi-square, or Mann–Whitney U test, when appropriate. A P-value of <0.05 was considered statistically significant and all tests were two-sided. Daytime periodic breathing was observed in 158 out of 339 patients (47%). Patients with periodic breathing were older (P= 0.005) and with more severe HF as defined by a more advanced New York Heart Association class (P= 0.016), higher levels of NT-proBNP (P= 0.016) and creatinine (P= 0.004). More than 90% of the patients were receiving betablockers and more than 80% of the patients were carrying an implanted device. No differences were found in drug or device therapies. During the night, the prevalence of moderate-to-severe SDB (AHI ≥15/h) was 55% and the prevalence of CSA-CSR ≥15 events/h was 36%. Table 1 describes the polysomnographic data in patients with and without PB_day. Total sleep time (TST) did not differ at all, but patients with PB_day had a higher percentage of TST spent in light sleep (NREM 1) (P= 0.05) and a lower percentage in NREM 3 sleep (P= 0.045). Most relevant, patients with PB_day showed a significantly higher AHI (P= 0.0005) that was exclusively due to a higher prevalence of central events (P= 0.0002). Obstructive events were almost identical in the two groups. Patients with PB_day had also a higher oxygen desaturation index (P= 0.002), a lower minimum oxygen saturation (P= 0.0004) and a higher T90 (P= 0.002). There was no relationship between PB_day and the proportion of positional sleep apnoea. CSR-CSA ≥15 events/h was found in 45% of patients with PB_day and in 28% of patients