Search

It is proven that music listening can have a therapeutic impact in many clinical fields. However, to assume a curative value, musical stimuli should have a therapeutic logic. This study aimed at assessing short-term effects of algorithmic music on cardiac autonomic nervous system activity. Twenty-two healthy subjects underwent a crossover study including random listening to relaxing and activating algorithmic music. Electrocardiogram (ECG) and non-invasive arterial blood pressure were continuously recorded and were later analyzed to measure Heart Rate (HR) mean, HR variability and baroreflex sensitivity (BRS). Statistical analysis was performed using a general linear model, testing for carryover, period and treatment effects. Relaxing tracks decreased HR and increased root mean square of successive squared differences of normal-to-normal (NN) intervals, proportion of interval differences of successive NN intervals greater than 50 ms, low-frequency (LF) and high-frequency (HF) power and BRS. Activating tracks caused almost no change or an opposite effect in the same variables. The difference between the effects of the two stimuli was statistically significant in all these variables. No difference was found in the standard deviation of normal-to-normal RR intervals, LFpower in normalized units and LFpower/HFpower variables. The study suggests that algorithmic relaxing music increases cardiac vagal modulation and tone. These results open interesting perspectives in various clinical areas.

It is still not known whether the oscillation in heart rate (HR) induced by sleep-disordered breathing (SDB) in patients with heart failure entails significant chronotropic effects. We hypothesised that since cyclical changes in ventilation and arterial blood gases during SDB affect HR through multiple and complexly interacting mechanisms characterised by large inter-subject variability, chronotropic effects may change from patient to patient. A total of 42 patients with moderate-to-severe chronic heart failure with systolic dysfunction underwent an in-hospital sleep study. Chronotropic effects of SDB were quantified by comparing the distribution of instantaneous HR during SDB with that during periods without SDB (noSDB) within the same night in each patient. Based on distribution changes from noSDB to SDB, 12, nine, 11, and 10 patients showed a significant tachycardic, bradycardic, tachycardic and bradycardic, and neither significant tachycardic nor significant bradycardic effect, respectively. Tachycardic and bradycardic effects were primarily due to an increase in the rate rather than in the magnitude of cyclical HR elevations and reductions, and were more prevalent and severe in patients with dominant obstructive and central events, respectively. The apnea-hypopnea index did not differ between groups. Conversely, the time spent with an oxygen saturation of

Background It has been hypothesized that pre-capillary pulmonary hypertension (PH) may trigger sleep disordered breathing (SDB). In patients with chronic thromboembolic PH (CTEPH), pulmonary endarterectomy (PEA) is potentially effective to improve PH. We assessed the pre- and post-operative prevalence of SDB in CTEPH patients submitted to PEA and the relationship between SDB and clinical, pulmonary and hemodynamic factors. Methods Unattended cardiorespiratory recording was performed the night before and one month after elective PEA in 50 patients. Results Before the intervention SDB prevalence (obstructive or central AHI ≥ 5/h) was 64%: 18 patients (66% female) had No-SDB, 22 (68% female) had dominant obstructive (dOSA), and 10 (20% female) had dominant central sleep apnea (dCSA). There were no differences in risk factors and the need for supplemental oxygen. Mean right atrial (mRAP) and pulmonary artery pressures (mPAP) showed a more compromised profile from No-SDB to dOSA and dCSA (mRAP: 5.5 ± 3.9 vs 7.0 ± 4.5 vs 9.7 ± 4.3 mm Hg (p = 0.054), mPAP: 39 ± 12 vs 48 ± 11 vs 51 ± 16 mm Hg (p = 0.0.47)). By contrast, cardiac index did not differ. At post-intervention, the prevalence of SDB was 68%: 16 patients had No-SDB, while 30 had dOSA and 4 dCSA, with no relationship with the relief from PH. Interestingly, 5 patients with previous CSA moved to the OSA group and 2 normalized. Conclusions Prevalence of SDB is high in patients with CTEPH even after resolution of PH. Our data support the hypothesis that pre-capillary PH may trigger CSA but not OSA, and suggest that OSA may play a role in the development of CTEPH.

Home monitoring is the most practical means of collecting sleep data in large-scale research investigations. Because the portion of recording time with poor-quality data is higher than in attended polysomnography, a quantitative assessment of the quality of each signal should be recommended. Currently, only qualitative or semi-quantitative assessments are carried out, likely because of the lack of computer-based applications to carry out this task efficiently. This paper presents an innovative computer-assisted procedure designed to perform a quantitative quality assessment of standard respiratory signals recorded by Type 2 and Type 3 portable sleep monitors. The proposed system allows to assess the quality (good versus bad) of consecutive 1-min segments of thoraco-abdominal movements, oronasal, nasal airflow and oxygen saturation through an automatic classifier. The performance of the classifier was evaluated in a sample of 30 unattended polysomnography recordings, comparing the computer output with the consensus of two expert scorers. The difference (computer versus scorers) in the percentage of good-quality segments was on average very small, ranging from -3.1% (abdominal movements) to 0.8% (nasal flow), with an average total classification accuracy from 90.2 (oronasal flow) to 94.9 (nasal flow), a Sensitivity from 0.93 (oronasal flow) to 0.98 (nasal flow), and a Specificity from 0.74 (nasal flow) to 0.86 (abdominal movements). In practical applications, the scorer can run a check-and-edit procedure, further improving the classification accuracy. Considering a sample of 270 unattended polysomnography recordings (recording time: 545 ± 44 min), the average time taken for the check-and-edit procedure of each recording was 6.9 ± 2.1 min for all respiratory signals.

Objective The interplay between arousals and respiratory events during Cheyne-Stokes respiration (CSR) with central sleep apnea (CSA) in heart failure (HF) patients is still not fully understood. We investigated the temporal relationship between arousals and CSR-CSA. Methods Episodes of CSR-CSA during sleep stages N1-N2 were analyzed in 22 HF patients with an apnea-hypopnea index ≥15/h, dominant CSA and central apnea index ≥5/h. For each CSR-CSA cycle (apnea + hyperpnea), we determined the onset (ARonset, relative to hyperpnea onset) and duration of detected arousals. Results Arousals (N = 2348) mostly occurred within the first half of the hyperpneic phase (42.6%, ARonset = 10.6 ± 2.1 s; duration = 10.6 ± 5.2 s) or close to hyperpnea onset (21.5%, ARonset = −0.1 ± 0.6 s; duration = 13.9 ± 5.4 s). Within-apnea arousals were less frequent (12.4%, ARonset = −16.0 ± 4.7 s; duration = 3.8 ± 1.4 s). The proportion of CSR-CSA cycles without any hyperpnea-related arousal was 27.5 ± 18.2%. Hyperpnea-related arousability (total number of hyperpneic arousals/total duration of hyperpneas) and apnea-related arousability were 63.4 ± 21.0/h and 23.8 ± 16.9/h, respectively (p Conclusion During CSR-CSA, a significant proportion of arousals occur at hyperpnea onset, indicating a low arousal threshold. Hyperpneic arousals are not essential for CSR-CSA. Arousability markedly increases during hyperpneas, likely due to the concurrent increase in chemoreceptor stimulation. Significance This study extends current knowledge on the interplay between sleep instability and respiratory events during CSR-CSA.

Aims Obstructive (OSA) and central sleep apnoea (CSA) are a common comorbidity in patients with heart failure. The purpose of this study was to assess and compare the impact of body position on the severity of sleep apnoea in these two groups of patients. Methods and results Standard polysomnography was performed in consecutive, clinically stable, optimally treated patients with moderate-to-severe heart failure and systolic dysfunction. Patients with an apnoea–hypopnoea index (AHI) ≥15/h (n = 120) were included in the study. The severity of sleep-disordered breathing was quantified by the AHI, the mean value of oxygen desaturations (O2desat) and the apnoea ratio. Data from the right and left positions were combined into a single lateral position. Positional sleep apnoea was defined as a >50% reduction in the AHI between the supine and the lateral position. Twenty-nine and 91 subjects had dominant OSA and CSA, respectively. The AHI markedly decreased from the supine to the lateral position in both groups [OSA: (median [q1,q3]) 50.3 [36.9, 67.6]/h vs. 10.4 [7.0, 18.5]/h, P

In patients with chronic thromboembolic PH (CTEPH), pulmonary endoarterectomy (PEA) provides effective relief from PH. We assess the pre- and post-operative prevalence of SDB in patients with CTEPH who had been submitted to PEA and the relationship between SDB and clinical and physiological factors. Cardiorespiratory recording was performed on the night before intervention and 3-months later in 50 patients (mean age 62.5±11.1 yrs) undergoing PEA. Patients were classified as central sleep apnea (CSA) if >70% of the events were central or as obstructive sleep apnea (OSA) if > 70% were obstructive. At pre-intervention, 14 patients (57% female) had No SDB, 26 (73% female) had OSA, and 10 (20% female) CSA. There were no differences in the history of venous thrombosis, atrial fibrillation, hypertension, diabetes, COPD and in the presence of oxygen therapy. Mean right atrial pressure (RAP) and mean pulmonary artery pressure (PAP) and TAPSE showed a trend toward a more compromised profile from NoSDB to OSA and CSA. RAP: 5.6±4.2 vs 6.7±4.3 vs 9.7±4.3 mmHg (p=0.07), PAP: 40±13 vs 47±12 vs 51±16 mmHg (p=0.11), TAPSE: 19±4 vs 16±3 vs 17±3 (p=0.06). Cardiac Index, PaO2 and PaCO2 were similar among the 3 groups. At post-intervention, the prevalence of SDB did not change: 12 patients had NoSDB, while 34 had OSA and 4 CSA. However, no trend was found in the relief from PH. PAP was 27±9 vs 25±7 vs 23±6 mmHg (p=0.73). Interestingly, 7 patients with previous CSA moved to the OSA group. Prevalence of SDB is high in patients with CTEPH even after resolution of PH. These data do not support the hypothesis that precapillary PH may trigger SBD, but rather suggest that SDB may be a causative factor leading to PH.

Background The pattern of oxygen saturation (SaO2) measured by pulse oximetry at the finger after apnea-induced hypoxemia has been shown to be a delayed and distorted version of the same measurement at the ear (i.e. in close proximity to carotid chemoreceptors). The impact of these different sites of recording on measures of hypoxemia during Cheyne-Stokes respiration (CSR) in heart failure (HF) patients is not known. Aims and objectives The aim of this study was to investigate the difference in SaO2 measured at the ear and at the finger during CSR in HF patients. Methods We studied 8 HF patients who developed CSR during a 40min laboratory recording of SaO2 by two identical pulse oximeters with finger and ear probe. The average SaO2, the mean value of oxygen desaturations and of minima and maxima and the delay between ear and finger signals were computed and compared by bias and 95% limits of agreement (LoA). Results Data are reported in Table as mean±SD. The delay between finger and ear signals was 20.6±7.2s. Conclusions This pilot study shows that, despite little, non-significant bias, the LoA were rather wide, indicating poor agreement between SaO2 measurements at the ear and finger, with clinically relevant differences in some patients. This may potentially lead to wrong assessment of the hypoxemic burden in some HF patients during CSR.

Background Arousals from sleep are thought to play an important role in sustaining Cheyne-Stokes respiration (CSR) in heart failure (HF) patients. Whether they represent a regular or occasional component of CSR and when they occur, however, have not yet been thoroughly investigated. Aims and objectives To determine the incidence and time of occurrence of arousals during CSR. Methods We studied 20 stable HF patients with reduced ejection fraction, an apnea-hypopnea index ≥15/h and dominant central sleep apnea. Arousal onset (ARon) was objectively determined using dedicated software. The analysis focused on all arousals at least partly concurrent with the hyperpneic phase of CS. Arousals were classified as “Early” (EA), if ARonoccurred before 3 s from hyperpnea onset; “Almost synchronous” (AS), if ARonoccurred between 3 s before hyperpnea onset and the end of the first post-apneic breath; “Late” (LA), if ARonoccurred after the first post-apneic breath. Results Incidence data (number of arousals /number of CSR cycles x 100) from the studied patients are given below (from 2048 CSR cycles): Conclusions Although, on average, the great majority of CSR cycles are associated with an arousal, there are large differences between patients both in the total incidence and in the time of occurrence of arousals. These findings suggest that attempts at stabilizing breathing by pharmacological manipulation of the arousal threshold should be patient specific.

Summary Fluctuations in sleep/wake state have been proposed as an important mechanism contributing to the development of oscillatory breathing patterns, including Cheyne–Stokes respiration in patients with heart failure. In order to properly assess the interactions between changes in state and changes in ventilatory parameters, a methodology capable of continuously and reliably detecting state transitions is needed. Traditional fixed-epoch analysis of polysomnographic recordings is not suitable for this purpose. Moreover, visual identification of changes in the dominant electroencephalogram activity at the transition from wakefulness to sleep and vice versa is often very subjective. We have therefore developed a hybrid approach – including both visual scoring and computer-based procedures – for continuous analysis of state transitions from polysomnographic recordings, specifically tailored for fluctuations between wakefulness and non-rapid eye movement-1 and -2 sleep. The overall analysis process comprises three major phases: (1) manual identification of relevant electroencephalogram/electrooculogram features and events, including a sample of unequivocal alpha and theta-delta activity; (2) automatic statistical discrimination of dominant electroencephalogram activity; and (3) state classification (wakefulness, non-rapid eye movement-1 and -2). The latter is carried out by merging information from visual scoring with the output of the discriminator. Validation has been carried out in 16 patients with heart failure during daytime Cheyne–Stokes respiration, using a training and testing set of electroencephalogram polysomnograms. The statistical discriminator correctly classified 99.1 ± 1.4% and 99.2 ± 1.1% of unequivocal alpha and theta-delta activity. This approach has therefore the potential to be used to reliably measure the incidence and location of sleep–wake transitions during abnormal breathing patterns, as well as their temporal relationship with major ventilatory events.

The physiological mechanisms responsible for periodic breathing (PB) in heart failure (HF) patients are still debated. A role for rhythmic shifts in the level of wakefulness has been suggested, but their existence has never been proven. In this study we investigated the existence of an oscillation in EEG activity during PB in these patients and assessed its relationship with the ventilatory oscillation. EEG activity was measured by the fractal dimension (FD) and by a spectral technique (weighted mean frequency, WMF) in 17 stable HF patients (mean age +/- SD: 57+/-10 yrs, NYHA class: 2.6 +/- 0.4, LVEF: 24 +/- 6%), with sustained PB during supine rest. The relationship between minute ventilation (MV) signal and FD and WMF was assessed by coherence analysis. Most patients (10/17) showed a well defined oscillation in FD and WMF at the frequency of PB closely linked (coherence0.7) with the oscillation of MV. In the remaining patients, neither FD nor WMF showed a clear oscillatory pattern synchronous with MV. Overall, the two EEG-derived parameters showed the same coherence with the ventilatory oscillation (mean coherence +/- SD: 0.65 +/- 0.25 vs 0.66 +/- 0.23, for FD and WMF respectively, p = 0.44). Our results provide evidence that during PB in HF patients, EEG activity often, but not always, fluctuates synchronously with the ventilatory oscillation. These fluctuations can be effectively detected by the fractal dimension, but classical spectral methods provide substantially the same information. Other mechanisms, particularly chemical instability in the respiratory control system, are likely to play a role in the genesis of PB.

AIMS: Nocturnal breathing disorders in the form of periodic breathing (PB) are very common in heart failure (HF) patients. There is an increasing interest in simple and affordable tools to screen patients and monitor these disorders at home on a long-term basis. We aimed to assess the pathophysiological and clinical relevance of a simplified method for monitoring of PB suitable to be self-managed by the patient at home. METHODS AND RESULTS: A night-time respiratory recording was performed in 397 optimally treated HF patients (age 60 +/- 11 years, NYHA class 2.4 +/- 0.6, left ventricular ejection fraction 29 +/- 7%) and the duration of PB (PB(Dur)) automatically computed. Patients were followed-up for 1 year and the prognostic value of PB(Dur) evaluated. In 45 patients, we assessed the association between PB(Dur) and severity of oxygen desaturations (number of desaturations >3%). Twenty six of the 397 patients died of cardiac causes. A PB(Dur) > or =2 h was significantly associated with an increased risk of cardiac death after adjustment for major clinical predictors [hazard ratio (95% CI): 3.5 (1.6-7.9), P = 0.002]. The correlation between PB(Dur) and severity of desaturations was 0.94 (P < 0.0001). CONCLUSION: Relevant pathophysiological and clinical information can be obtained from simplified monitoring of breathing disorders managed by the patient. These results provide new perspectives in the investigation of the clinical impact of abnormal breathing in HF patients.

ObjectivesThis study investigated the clinical correlates and prognostic value of depressed baroreceptor-heart rate reflex sensitivity (BRS) among patients with heart failure (HF), with and without beta-blockade.BackgroundAbnormalities in autonomic reflexes play an important role in the development and progression of HF. Few studies have assessed the effects of beta-blockers on BRS in HF.MethodsThe study population consisted of 103 stable HF patients, age (median [interquartile range]) 54 years (48 to 57 years), with New York Heart Association (NYHA) functional class ≥III in 22, and with a left ventricular ejection fraction (LVEF) of 30% (24% to 36%), treated with beta-blockers; and 144 untreated patients, age 55 years (48 to 60 years), with NYHA functional class ≥III in 47%, and an LVEF of 26% (21% to 30%). They underwent BRS testing (phenylephrine technique).ResultsIn both treated and untreated patients, a lower BRS was associated with a higher (≥III) NYHA functional class (p = 0.0002 and p < 0.0001, respectively); a more severe (≥2) mitral regurgitation (p = 0.007 and p = 0.0002), respectively; a lower LVEF (p = 0.0004 and p = 0.001, respectively), baseline RR interval (p = 0.0004 and p = 0.0002, respectively), and SDNN (p < 0.0001, p = 0.002, respectively); and a higher blood urea nitrogen (p = 0.004, p < 0.0001, respectively). Clinical variables explained only 43% of BRS variability among treated and 36% among untreated patients. During a median follow-up of 29 months, 17 of 103 patients and 55 of 144 patients, respectively, experienced a cardiac event. A depressed BRS (

Background and purpose Nocturnal cardiac arrhythmias occur in patients with obstructive sleep apnea (OSA), reportedly as a consequence of the autonomic effects of recurrent apnea with subsequent oxygen desaturation. We have investigated whether different patterns of OSA are associated with specific arrhythmia during sleep. Patients and methods Electrocardiographic (ECG) data recorded during polysomnography (PSG) were analysed in 257 consecutive OSA patients to determine the prevalence of cardiac rhythm disturbances, and to relate these to breathing pattern (normal, apnea/hypopnea, recovering ventilation, snoring) and oxygen saturation. Results Arrhythmias were found in 18.5% of patients. Patients with nocturnal bradyarrhythmia (BA) had higher values of ventilatory disturbance (apnea–hypopnea index [AHI] 58.8±36.8 vs 37.2±30.3, p =0.02), mean desaturation amplitude (8.9±4 vs 5.9±3.4%, p =0.03), and a lower SaO 2 nadir (69% vs 77%, p = 0.003) than those without arrhythmia. The prevalence of BA in patients with AHI⩾ 30/h was significantly higher than that observed in those with AHI χ 2 =5.61, p =0.01). In contrast, patients with tachyarrhythmia (TA) had no significant differences in AHI, mean desaturation amplitude or SaO 2 nadir than those without arrhythmia. No associations were found between arrhythmia and the presence of comorbidity or concomitant medical therapy, except for an association between tachyarrhythmia and chronic obstructive pulmonary disease (COPD) (odds ratio 2.53; 95% confidence intervals 1.1–5.8, p =0.03). Conclusions We conclude that while BA during sleep is associated with OSA severity, concomitant COPD or β 2 -treatment may play a role in the development of TA during sleep.

ObjectivesWe sought to assess applicability, clinical correlates, and prognostic value of the transfer function method for measuring baroreflex sensitivity (TF-BRS).BackgroundAbnormalities in autonomic reflexes play an important role in the development and progression of chronic heart failure (CHF). Simple and non-invasive techniques for clinical measurement of such reflexes are desirable.MethodsIn 317 stable CHF patients in sinus rhythm (median age [interquartile range]: 54 years [48 to 59 years], New York Heart Association [NYHA] functional class II to III: 88%, left ventricular ejection fraction [LVEF]: 27% [22% to 33%]) we recorded electrocardiograms and non-invasive arterial pressure during paced breathing to measure TF-BRS.ResultsOwing to a high number of ectopic beats, TF-BRS could be computed in 72% of the patients; TF-BRS was lower in NYHA functional class III to IV and mitral regurgitation 2 to 3 (p < 0.0005 for both). Correlation with LVEF and standard deviation of all normal-to-normal intervals was 0.18 and 0.31 (p < 0.001 for both). During a mean follow-up of 26 months, 23% of the patients experienced a cardiac event. A depressed TF-BRS (≤3.1 ms/mm Hg) was significantly associated with the outcome (hazard ratio 3.2, 95% confidence interval [CI] 1.7 to 6.0, p = 0.0003). Patients with a missing TF-BRS had a high event rate (36%). Combining this information with available TF-BRS measurements, a new prognostic index could be computed in 97% of the patients that significantly predicted the outcome after adjustment for clinical and functional variables (hazard ratio 2.5, 95% CI 1.3 to 4.6 p = 0.004).ConclusionsIn CHF patients in sinus rhythm, TF-BRS conveys relevant clinical and prognostic information, but its measurability is markedly affected by ectopic activity. Nevertheless, a TF-BRS–based risk index carrying significant and independent prognostic information can be computed in almost all patients.

In this paper, we describe a computer program (RESP-24) specifically devised to assess the prevalence and characteristics of breathing disorders in ambulant chronic heart failure patients during the overall 24 h period. The system works on a single channel respiratory signal (RS) recorded through a Holter-like portable device. In the pre-processing stage RESP-24 removes noise, baseline drift and motion artefacts from the RS using a non-linear filter, enhances respiratory frequency components through high-pass filtering and derives an instantaneous tidal volume (ITV) signal. The core processing is devoted to the identification and classification of the breathing pattern into periodic breathing (PB), normal breathing or non-classifiable breathing using a 60 s segmentation, and to the identification and estimation of apnea and hypopnea events. Sustained episodes of PB are detected by cross analysis of both the spectral content and time behavior of the ITV signal. User-friendly interactive facilities allow all the results of the automatic analysis procedure to be edited. The final report provides a set of standard and non-standard parameters quantifying breathing abnormalities during the 24 h period, the night-time and the day-time, including the apnea/hypopnea index, the apnea index, the total time spent in apnea or in hypopnea and the prevalence of non-apneic and apneic PB. The accuracy of these measurements was appraised on a data set of 14 recordings, by comparing them with those provided by a trained analyst. The mean and standard deviation of the error of the automatic procedure were below respectively 6 and 8% of the reference value for all parameters considered and the mean total classification accuracy was 92%. In most cases, the individual error was

Summary Fluctuations in sleep–wake state are thought to contribute to the respiratory instability of Cheyne–Stokes respiration in patients with heart failure by promoting the rhythmic occurrence of central apnea and ventilatory overshoot. There are no data, however, on the relationship between vigilance state and respiratory events. In this study we used a novel method to detect the occurrence of state transitions (time resolution: 0.25 s, minimum duration of state changes: 2 s) and to assess their time relationship with apnoeic events. We also evaluated whether end-apnoeic arousals are associated with a ventilatory overshoot. A polysomnographic, daytime laboratory recording (25 min) was performed during Cheyne–Stokes respiration in 16 patients with heart failure. Automatic state classification included wakefulness and non-rapid eye movement sleep stages 1–2. As a rule, wakefulness occurred during hyperpnoeic phases, and non-rapid eye movement sleep occurred during apnoeic events. Ninety-two percent of the observed central apneas (N = 272) were associated with a concurrent wakefulness non-rapid eye movement sleep wakefulness transition. The delay between wakefulness non-rapid eye movement sleep transitions and apnea onset was −0.3 [−3.1, 3.0] s [median (lower quartile, upper quartile); P = 0.99 testing the null hypothesis: median delay = 0], and the delay between non-rapid eye movement sleep wakefulness transitions and apnea termination was 0.2 [−0.5, 1.2] s (P = 0.7). A positive/negative delay indicates that the state transition occurred before/after the onset or termination of apnea. Non-rapid eye movement sleep wakefulness transitions synchronous with apnea termination were associated with a threefold increase in tidal volume and a twofold increase in ventilation (all P

Portable polysomnography (PSG) and cardiorespiratory polygraphy are increasingly being used in the assessment of sleep-disordered breathing (SDB) in heart failure patients. Scoring of SDB from cardiorespiratory polygraphy recordings is based only on respiratory signals, while electroencephalographic, electrooculographic and electromyographic channels are taken into account when using PSG recordings. The aim of this study was to assess the agreement between these two scoring methods. An overnight sleep study was performed in 67 heart failure patients using a standard portable polysomnograph. Each recording was scored twice, once using all acquired signals (PSG mode) and, after a median of 64 days, using only respiratory signals (cardiorespiratory mode). Agreement was assessed by Bland–Altman analysis and Cohen’s kappa. We found that (1) more respiratory events were detected using cardiorespiratory analysis [median (25th percentile, 75th percentile), 75 (39, 200) events] compared to analysis of portable PSG [69 (29, 173) events, p

Fluctuations in sleep/wake state have been proposed as an important mechanism contributing to the development of oscillatory breathing patterns, including Cheyne-Stokes respiration in patients with heart failure. In order to properly assess the interactions between changes in state and changes in ventilatory parameters, a methodology capable of continuously and reliably detecting state transitions is needed. Traditional fixed-epoch analysis of polysomnographic recordings is not suitable for this purpose. Moreover, visual identification of changes in the dominant electroencephalogram activity at the transition from wakefulness to sleep and vice versa is often very subjective. We have therefore developed a hybrid approach--including both visual scoring and computer-based procedures--for continuous analysis of state transitions from polysomnographic recordings, specifically tailored for fluctuations between wakefulness and non-rapid eye movement-1 and -2 sleep. The overall analysis process comprises three major phases: (1) manual identification of relevant electroencephalogram/electrooculogram features and events, including a sample of unequivocal alpha and theta-delta activity; (2) automatic statistical discrimination of dominant electroencephalogram activity; and (3) state classification (wakefulness, non-rapid eye movement-1 and -2). The latter is carried out by merging information from visual scoring with the output of the discriminator. Validation has been carried out in 16 patients with heart failure during daytime Cheyne-Stokes respiration, using a training and testing set of electroencephalogram polysomnograms. The statistical discriminator correctly classified 99.1 ± 1.4% and 99.2 ± 1.1% of unequivocal alpha and theta-delta activity. This approach has therefore the potential to be used to reliably measure the incidence and location of sleep-wake transitions during abnormal breathing patterns, as well as their temporal relationship with major ventilatory events.

OBJECTIVES No studies have compared the prognostic values of invasive (phenylephrine, Phe) and noninvasive (transfer function) assessments of baroreflex sensitivity (BRS). METHODS Three hundred and one heart failure patients [age: 53 ± 8 years, New York Heart Association class II-III: 88%, left-ventricular ejection fraction (LVEF): 28 ± 8%] underwent an 8 min ECG and arterial pressure recording, followed by Phe administration. RESULTS Phe-BRS and transfer function BRS (TF-BRS) could be measured in 89 and 72% of cases, respectively. The correlation and the 5-95th percentiles of the difference between the two methods were 0.61 (P

In this manuscript we present an overview of novel signal processing techniques developed by our group to reduce scoring time in the assessment of the severity of sleep-related breathing disorders in heart failure patients and to detect sleep/wake fluctuations during periodic breathing. Besides describing these methods, we present the results of validation experiments. Our work shows that novel signal processing techniques can reduce costs and resources needed to screen the patients and can provide relevant information for better understanding the role of wake/sleep transitions in the development and maintenance of breathing disorders.

Portable sleep apnea monitors are often used to screen for sleep-disordered breathing in chronic heart failure patients (CHF), but night-to-night repeatability of obtained measurements of nocturnal breathing disorders has not been fully assessed. Fifty-six stable, moderate-to-severe CHF patients [male, 87%; age, 57 ± 9 years; NYHA class, 2.6 ± 0.6; left ventricular ejection fraction (LVEF), 32% ± 9%] underwent an unattended in-hospital cardiorespiratory recording using a portable sleep apnea monitor during two consecutive nights. The apnea/hypopnea index (AHI), apnea index (AI), oxygen desaturation index (ODI), and periodic breathing (PB) duration were computed. Intra-subject night-to-night variability was assessed by the 95% limits of random variation (LoV). We also estimated the contribution of intra-rater variability to the overall intra-subject variability. Dichotomizing the AHI and PB duration according to conventional cutoffs of, respectively, ≥5 events per hour, ≥15 events per hour, and ≥120 min, the percentage of patients concordantly classified by the two measurements was finally computed. The 95% LoV were ±10.6, ±7.7, ±11.3 events per hour for AHI, AI and ODI, and ±63.2 min for PB duration, respectively. The contribution of intra-rater variability to total intra-subject variability was 1.7%, 1.4%, 2.5%, and 1.3% for AHI, AI, ODI, and PB duration, respectively. Most patients (85%, 82%, and 95% for AHI ≥ 5, AHI ≥ 15, and PB duration, respectively) were classified concordantly by the two measurements. In patients with heart failure, measurements of severity of sleep-disordered breathing derived from portable sleep apnea monitors show significant night-to-night intra-subject variation with a negligible contribution from intra-rater variability; however, using the same measurements for classification purposes, as commonly performed in clinical practice to screen patients for sleep-disordered breathing, very stable results are obtained.

In this study we investigated the existence and the nature of rhythmic changes in EEG associated with ventilatory oscillations in heart failure (HF) patients with periodic breathing (PB). Since nonlinear mechanisms are thought to be involved in the generation of EEG, we hypothesized that a mathematical approach based on nonlinear methods would provide relevant information on the association between EEG and ventilatory oscillations. We studied five patients who developed a sustained non-obstructive PB pattern during a 20 min laboratory recording. The time course of the fractal dimension of the EEG signal (HFD) was estimated dividing this signal into 2 s segments, with a 1.5 s overlap and computing for each EEG segment the fractal dimension using the Higuchi's algorithm. From the lung volume signal, an instantaneous minute ventilation (IMV) signal was also computed. The relationship between IMV and HFD was assessed by bivariate spectral analysis, computing the magnitude square coherence function (MSC). In four patients the value of the MSC was very high, ranging from 0.75 to 0.91, while in one patient the value was only 0.29. Our results suggest that in patients with PB, rhythmic changes in the EEG signal are very common and, when present, they are associated with ventilatory oscillations. We have also demonstrated that such oscillations can be detected very effectively by a technique based on nonlinear methods.

Periodic Breathing (PB, cyclic waxing and waning of tidal volume) is very common in heart failure patients. It is commonly thought that PB is caused by an instability in the closed-loop chemical control of ventilation. Some investigators, however, have suggested that it may also result from fluctuations in sleep/wake state, namely state instability. The aim of this study was i) to set up a methodology for semi-automated state fluctuation analysis and ii) to assess whether state instability is a necessary condition for the development of PB. We carried out daytime and nighttime standard polysomnographic recordings in 9 patients. During daytime, 86% of state transitions were synchronous with an apneic event. Synchronous state and ventilatory changes were also observed during nighttime, but during the deeper stages of sleep PB occurred without concurrent state transitions. We conclude that state instability is not necessary for the development of PB.

Background: Periodic Breathing (PB, waxing and waning of tidal volume in which hyperventilation alternates with periods of apnoea or hypopnoea), is common during sleep and wakefulness in patients with Heart Failure (HF) and may increase mortality. Aim: To assess the effect of short-term, day-time PB on prognosis, in HF patients. Methods: We prospectively studied 380 consecutive HF referrals who had a 10 min, supine day-time respiratory recording. We related PB (adjusted for known predictors) to total cardiac mortality, during a median follow-up of 41 months. Results: Day-time PB occurred in 145/380 patients who had more severe HF and more compromised left ventricular function (p

To assess the accuracy of spectral indices of arterial pressure variability and baroreflex sensitivity obtained from beat-by-beat noninvasive blood pressure recordings by the Finometer device, we compared these measures with those obtained from intra-arterial recordings. The performance of the Finometer was also compared to the traditional Finapres device. In 19 cardiac disease patients, including myocardial infarction, heart failure and cardiac transplant, we estimated the power of systolic and diastolic pressures in the VLF (0.01-0.04 Hz), LF (0.04-0.15 Hz) and HF (0.15-0.45 Hz) bands and computed absolute and percentage errors relative to intra-arterial brachial pressure. We also computed the characteristic frequency of each band (i.e. the barycentric frequency of spectral components identified in the band). The variability of systolic pressure in the VLF and LF bands was markedly overestimated by both the Finometer and Finapres (p0.01), with percentage median errors of respectively 130% and 103% (Finometer), and 134% and 78% (Finapres). The HF power was substantially unchanged using the Finometer and reduced using the Finapres (-28%, p0.05). The limits of agreement between noninvasive and invasive spectral measurements were wide. Linear system analysis showed that most (80%) of the power of noninvasive signals was linearly related to the power of the invasive signal. The characteristic frequency of each band was substantially preserved in both noninvasive signals. The results for diastolic pressure were similar, but the Finapres errors in the VLF and LF bands were lower. Baroreflex sensitivity was significantly underestimated by both devices (Finometer: -31%, Finapres: -24%). Despite previous studies having shown that brachial artery waveform reconstruction performed by the Finometer has improved the accuracy of blood pressure measurement compared to the Finapres device, measurement of blood pressure variability in cardiac disease patients provides worse results in most spectral parameters and a better accuracy only in the HF band of systolic pressure.

We assessed in a sample of 30 healthy subjects the changes brought about by paced breathing on cardiovascular variability (CV) signals (heart period and arterial blood pressure). In order to better interpret the results, we also measured standard ventilatory parameters, including end-tidal CO/sub 2/. We found that paced breathing induces a very mild hyperventilation, with no evident signs of mental stress. CV parameters did not change or changed slightly with no valuable physiological or practical implications. Hence paced breathing may be safely used to standardize non-invasive laboratory investigations on the autonomic control of the cardiovascular function.

In this study we appraised the accuracy of the Finometer in spectral analysis of systolic (SAP) and diastolic (DAP) arterial pressure. In 24 patients the blood pressure signal was simultaneously recorded invasively and by the Finometer and the power of SAP and DAP in the VLF (0.01/spl divide/0.04 Hz), LF (0.04/spl divide/0.15 Hz), and HF (0.15/spl divide/0.45 Hz) bands computed. To investigate the relationship between invasive and Finometer measurements, the percentage error was computed and bivariate spectral analysis performed. The error for SAP analysis, was (%): 139, 76, -32, 75 for respectively VLF, LF, HF and total power; DAP errors were much smaller. SAP variability is greatly enhanced by the Finometer, mainly due to the increase of power in the VLF and LF bands which overrides the reduction in the HF band. Noise and nonlinearities account for < 20% of the Finometer power changes, indicating that the differences are mainly due to amplification or attenuation effects.

Voluntary control of breathing (CB) has been proposed as a means to standardize short-term investigations on cardiovascular variability (CVV). In this study, we assessed the changes induced by CB on ventilatory and cardiovascular parameters in 17 chronic heart failure (CHF) patients. We recorded the instantaneous lung volume (ILV), R-R interval (RR), systolic arterial pressure (SAP) and end-tidal CO/sub 2/ (ETCO/sub 2/, 10 patients only) during 8 minutes of spontaneous breathing (SB) and 8 minutes of CB at 0.22 Hz. CB caused a 14% reduction in respiratory rate, a 57% increase in tidal volume, a 27% increase in minute ventilation and a 7% decrease in ETCO/sub 2/. A negligible change was observed in mean RR and SAP. No significant change was observed in the low-frequency (LF) power of RR and SAP and in baroreflex sensitivity, whereas the high-frequency (HF) power increased markedly. Hence, CB in CHF patients causes a mild hyperventilation but does not seem to influence autonomic cardiovascular control. Only respiratory-related oscillations of CVV signals are affected by CB.

Non-invasive baroreflex sensitivity (BRS) by the classical transfer function (TF) method very often cannot be assessed in chronic heart failure (CHF) patients, due to a lack of coherence between the heart period and the systolic pressure. On the grounds of a previous investigation, we devised a new BRS index based on the whole-band average (WBA) of the TF in the whole low-frequency band (WBA-BRS), which does not require a coherence check. In this study, we assessed the clinical relevance of this index by testing its prognostic value in a population of 149 CHF patients. During a 20/spl plusmn/ 2 months follow-up, 28 patients died or were urgently transplanted. The univariate relation of the WBA-BRS index with mortality was assessed by the Cox proportional hazards model, and survival functions were estimated using the Kaplan-Meier method. We also investigated the independent effect of WBA-BRS, adjusting for the NYHA (New York Heart Association) class, LVEF (left ventricular ejection fraction), peak V/sub O2/ and Na/sup +/. WBA-BRS showed a significant univariate association with the risk of cardiac events. When its predictive value was assessed in multivariate analysis, WBA-BRS displayed independent prognostic information (risk ratio=3.0, 95% confidence interval: 1.1-7.9, p=0.03). These preliminary data suggest that BRS assessed by a modified version of the classical TF method is a potential prognostic marker in CHF patients.

This study was conceived to analyze EEG state transitions and their relationship with ventilatory oscillations during daytime supine periodic breathing (PB) in heart failure patients. To this purpose, a semi-automated, frequency-based methodology was developed, and testing was carried out in 5 patients with PB. We found that apneas and hypopneas very often (86% on average) begin in synchrony with transitions from wakefulness to theta activity, and end in synchrony with the opposite transition. These results suggest that in these patients chemical instability secondary to the impaired hemodynamics coexists and interacts with state instability.

Baroreflex sensitivity (BRS) can be estimated noninvasively by computing the transfer function modulus (TF) between spontaneous oscillations of systolic arterial pressure (SAP) and RR interval within the LF band. In 21 healthy volunteers we tested the hypothesis that during standing the reliability of BRS-TF increases as a result of increased SAP power. We also assessed whether standing causes any change in BRS-TF. BRS-TF was computed averaging the TF over the whole LF band and only within -3 db points. We found that during standing the error of TF estimates markedly decreased compared to the supine posture. As this change was also accompanied by a decrease of BRS-TF, the reduction of the relative error was modest. Hence, standing is capable of reducing the uncertainty of TF estimates but this beneficial effect is partly offset by a simultaneous reduction of TF values. Of the two approaches for measuring BTS-TF, the average between -3 db points showed the highest reliability.