Two types of sleep disordered breathing are common among patients with heart failure - obstructive sleep apnea-hypopnea (OSAH) and Cheyne-Stokes breathing (CSB). OSAH is characterized by reductions or cessations of airflow during sleep, despite ongoing respiratory effort. It is due to upper airway obstruction. CSB is characterized by cyclic crescendo-decrescendo respiratory effort and airflow during wakefulness or sleep, without upper airway obstruction. When the decrescendo effort is accompanied by apnea during sleep, it is considered a type of central sleep apnea syndrome (CSAS). OSAH and CSB are often referred to collectively as sleep disordered breathing (SDB). They frequently coexist and can be clinically difficult to distinguish from one another because there is overlap in pathophysiology and clinical manifestations. SDB in patients with heart failure is under-recognized but important because it may independently predict mortality due to heart failure and may contribute to disease progression.
While OSAH is more common than CSB in the general population, CSB may be more common than OSAH in patients with heart failure (Javarheri 1998, Corra 2006). Single-center retrospective and observational studies estimate that the prevalence of SDB may be as high as 50 percent among all patients with heart failure and as high as 70 percent among those with heart failure who are referred to a sleep laboratory (Javaheri 1995, Sin 1999). Such prevalence estimates are limited by referral bias, variable definitions of SDB, variable severities of heart failure, and variable optimization of the medical management of the heart failure. The pathogenesis of CSB is uncertain, but the favored hypothesis is based on the observation that patients who have heart failure with CSB tend to have lower arterial carbon dioxide tensions (PaCO2) than patients who have heart failure without CSB (Naughton 1993, Hanly 1993). The occurrence of apnea is central to SDB. Hypercapnia exists by the time the respiratory control center terminates the apnea. The hypercapnia then stimulates robust hyperpnea, which yields marked hypocapnia and allows the cycle of events to repeat. The net effect is oscillation of ventilation between apnea and hyperpnea. Elimination of the hypocapnia with inhaled CO2, continuous positive airway pressure (CPAP), or oxygen can markedly attenuate CSB (Steens 1994, Naughthon 1994). Both OSAH and CSB can impair systolic and diastolic cardiac function. Intermittent hypoxemia and arousals induce adrenergic surges that may lead to disease progression. In addition, extremely negative intrapleural pressures increase ventricular transmural wall stress and afterload (Malhotra 2003). Patients whose heart failure is complicated by SDB have a worse prognosis than patients without SDB (Wang 2007).
There are only very few data on SDB in patients with type-2 diabetes and virtually no single evidence in those with cachexia complicating the underlying syndrome of heart failure. Therefore WP06 aims to recruit 500 patients with CHF and obesity, cachexia, or normal body weight either having type-2 diabetes mellitus (50%) or not (1:1 sub-grouping) according to standard criteria outlined in WP03. All patients are being screened for SDB using a standardised questionnaire, an ambulatory sleep-apnoea-screening device, and a 24-hour ECG-monitoring recorder with the measurement of peripheral oxygen saturation. If possible and consented by the patient, SCCS will contribute to WP04, 07, 08, 09, and 10 by providing biomaterial, tissue samples, and patient data.
This work package addresses four main objectives:
This WP will be chaired by Stephan von Haehling (Charité). He is a senior researcher in the field of biomarker research, immune function, and endothelial dysfunction in chronic heart failure. His expertise is most valuable to the SICA-HF consortium. Due to the lack of data, the involvement in studies on sleep-associated breathing disorders and close collaborations with European partners the WP hypothesis was postulated and developed for a joint SICA-HF application. We will co-operate with the institution in Wroclaw, Hull, and the Charité. Patients for this work package will be recruited by the following partners of the SICA-HF consortium: SCCS (n=120), Wroclaw (n=50), Hull (n=100), Charité (n=50) and Bolnisnica Golnik (n=80). All patients with diagnosed heart failure are eligible for studies on SDB within SICA-HF. Patients will be screened for potential enrolment into this work package using the following criteria:
1. Heart failure with a) obesity, b) cachexia, c) type-2 diabetes mellitus as defined by WP03;
2. Written informed consent and understanding of the purposes of the study provided by WP06;
3. Both ambulatory and hospitalised patients are elegible, in hospitalised patients pre-discharge assessment will be used as baseline;
4. Assessment will be done by an experienced physician and a specialised heart failure nurse;
5. Baseline characteristics, blood samples, and clinical tests will be performed according to the criteria defined in WP03;
6. Follow-up for all patients is planned using time-schedules provided by WP03 (i.e. after 4-6 weeks, 16-18 months, and 28-30 months).
Assessment for SDB will be done by using two standardised, commercially available devices: ApneaLink (RESMED, Australia) and a 24-hour holter ECG-monitoring system together with the measurement of oxygen saturation (Cardioscan 12, MTMmultitechmed, USA) (Objective 1). Both devices will be applied to the patient in an ambulatory setting, i.e. the patient takes them home after detailed information provided by the attending physician, study nurse, and an information leaflet. With these devices, it is possible to differentiate between OSAH and CSB. Further we are able to monitor autonomic dysfunction and potential arrhythmias simultaneously (objective 2). An information leaflet on how to use the devices will be prepared and provided by WP06 based on current guidelines.
The following parameters will be obtained:
For SDB: Apnoea/Hypopnoea-Index (AHI), Apnoea Index (AI), Hypopnoea Index (HI), Oxygen Desaturation Index (ODI), mean oxygen saturation (O2Sm), basal oxygen saturation (O2Sb), number of breaths, number of apnoeas, number of hypopnoeas, snaring events, number of desaturations, number of cheyne stokes respiration (CSB);
For autonomic dysfunction and pro-arrhythmias: Heart frequency, number of heart beats, ventricular ectopies, ventricular tachycardia, supra-ventricular ectopies, supra-ventricular tachycardia, atrial fibrillation, atrial flutter, heart rate variability (time and frequency domain), QT-QTc-assessment, ST-deviation, vector cardiogram, sleep-apnoea algorithm based on heart rate variability, peripheral oxygen saturation;
Patients will be provided with the devices for 24-hour assessment (ECG) and over-night assessment (SDB), respectively. They will be advised to send back the devises on the next day by post (pre-addressed envelops will be provided).
To identify the potential shift from an obstructive to a more centrally-mediated SDB, follow-up of patients will be used to observe changes over time (objective 3). However, as this is our underlying hypothesis, no potential factor contributing to this hypothesis has been described yet neither in patients with CHF nor in those with the co-morbidities type-2 diabetes, obesity or cachexia. Therefore, routine biochemistry as defined by WP03 will be collected together with additional markers if this is clinically required in the opinion of the attending physician. A close collaboration with WP10 of the SICA-HF consortium is envisaged in order to identify novel, biomarkers that could markers to identify patients who shift from an obstructive to a more centrally-mediated SDB. Additionally, the center in Charité (n=50) aims to study a limited number of patients receiving treatment of their SDB (objective 4) in order to better understand the underlying mechanisms of this perturbation. For this purpose, 24-hour assessment (ECG) and over-night assessment (SDB) will be repeated during and after treatment.
Primary analysis: the exact Cochran-Armitage test will be used for trend. Secondary analysis for the primary endpoint: ordinal regression with demographic and baseline functional variables to identify and adjust for co-predictors of the primary outcome. Secondary endpoints: Cox regression for time to death, first hospitalisation and first worsening of CHF. Extended Kaplan-Meier estimates for joint analysis of the particular components of the secondary endpoint. Analysis of covariance for quantitative outcomes (functional parameters, biomarkers and quality of life).