More than half of the patients with chronic heart failure (CHF) have associated sleep-disordered breathing (SDB), which consists of either obstructive sleep apnoea (OSA) or central sleep apnoea (CSA).1,2 SDB has deleterious effects on systolic and diastolic ventricular function. In patients with CHF, the presence of SDB results in a poorer prognosis;3,4 thus, its early recognition and adequate management, including continuous positive airway pressure (CPAP) and adaptive servo-ventilation (ASV) therapies, could improve patient outcomes.
Echocardiography is a versatile technique for the assessment of cardiac geometry and function, and its portability allows serial assessment of ventricular function at the point of care. Multiple recent studies have revealed that CPAP therapy in patients with CHF and OSA/CSA improves left ventricular ejection fraction (LVEF) (see Table 1).5–11 However, LVEF is not a sensitive parameter to detect subtle left ventricular (LV) mechanical changes induced by SDB. In contrast, two-dimensional (2D) speckle tracking echocardiography, a relatively new technology, has potential for the quantification of global and regional 2D strains.12 The latter technique has been reported, in several clinical scenarios, to be superior to LVEF in the detection of subclinical LV dysfunction.13,14 As far as we know, there currently is no study that directly addresses the clinical utility of 2D speckle tracking echocardiography in patients with SDB and heart failure. This article will discuss the potential application of 2D speckle tracking echocardiography on ventricular mechanics in patients with SDB.
Effect of Obstructive Sleep Apnoea on Ventricular Mechanics in Sleep-disordered Breathing
OSA has a negative impact on ventricular function through multiple mechanisms.15 Augmented negative intrathoracic pressure during obstructive apnoea increases LV intramural pressure and afterload. Associated increased venous return produces right ventricular (RV) distension compromising LV filling, resulting in decreased LV stroke volume and cardiac output. Both the increased myocardial oxygen demand due to the increase in LV transmural pressure and the reduction of myocardial oxygen delivery during apnoea can induce myocardial ischaemia, resulting in impaired cardiac contractility and diastolic relaxation. Intermittent hypoxia and hypercapnea also stimulate the sympathetic nervous activity and blood pressure. These detrimental effects result in systolic and diastolic dysfunction, but patients with OSA also often have associated co-morbidities – such as obesity, hypertension and coronary artery disease – that might contribute to ventricular dysfunction. Thus it is very difficult to prove that OSA actually causes cardiac dysfunction independently of these confounding risk factors.
The Mueller manoeuvre closely simulates the changes in intrathoracic pressure induced by OSA while simultaneously increasing sympathetic nervous activity. To determine the acute effect of augmented negative intrathoracic pressure on ventricular function, Koshino and colleagues performed 2D strain and strain rate measurements using velocity vector imaging during the Mueller manoeuvre in 24 healthy subjects.16 Compared with baseline, global LV longitudinal strain and early diastolic strain rate were significantly decreased during the Mueller manoeuvre, in conjunction with a significant reduction in LVEF. The authors also found that global RV longitudinal strain and systolic strain rate were transiently but significantly reduced during the Mueller manoeuvre. These results suggest that an abrupt augmentation of negative intrathoracic pressure causes transient LV and RV dysfunction. In OSA patients, repetitive episodes of obstructive apnoea over time might therefore be expected to produce cumulative deleterious effects on both LV and RV function.
Haruki and collegues investigated the effect of overnight sleeping on LV global and regional function in 32 patients with OSA and normal LVEF using 2D speckle tracking echocardiography. They also evaluated the effects of chronic CPAP therapy on LV function during overnight sleeping.17 Although no significant changes in LVEF, global radial and circumferential strain were noted during overnight sleeping, global longitudinal strain was significantly reduced after overnight sleeping compared with before sleeping (see Figure 1A). The reduction in longitudinal strain was more remarkable in the apical segments and was also associated with an increase in the post-systolic shortening index, a marker of yocardial ischaemia. Interestingly, the authors demonstrated that effective CPAP therapy for a period of three months not only improved the apnoea–hypopnoea index and minimal oxygen saturation, but also abolished the reduction in longitudinal strain noted after the overnight sleep (see Figure 1B). Thus intermittent hypoxia due to repetitive obstructive apnoea during overnight sleep may be not sufficient to induce an apparent reduction in LVEF the next morning in OSA patients. However, it is sufficient to induce subclinical LV systolic dysfunction when assessed by 2D speckle tracking echocardiography.
Regarding the independent effect of OSA on RV function, Tugcu et al. demonstrated that patients with OSA but without systemic and pulmonary arterial hypertension have subclinical RV dysfunction.18 The authors measured RV free wall strain and strain rate in patients with OSA and in age- and body mass index (BMI)-matched control subjects, reporting that, compared with control subjects, patients with OSA had a significant impairment in RV strain and strain rate. In this study, the apnoea–hypopnoea index was the only independent determinant of RV free wall strain.
2D speckle tracking analysis can be also used to demonstrate LV twist abnormalities in patients with OSA. Vitarelli and colleagues performed 2D speckle tracking analysis of the left ventricle in 42 patients with OSA and no co-morbidities and in 25 healthy subjects.19 In addition to a significant reduction of global longitudinal strain, LV twist was significantly enhanced in patients with severe OSA compared with control subjects due to a predominant increase in apical rotation. Multivariate regression analysis revealed that the apnoea–hypopnoea index and aortic stiffness were independent predictors of the increase in LV twist. Furthermore, effective chronic CPAP therapy resulted in a significant decrease in LV twist and aortic stiffness together with an increase in global longitudinal strain. Subendocardial dysfunction resulting from OSA could be one of the possible mechanisms inducing these abnormalities.20 Left ventricular twist originates from two counteracting forces generated from the right-handed subendocardial helical fibres and left-handed subepicardial helical fibres. Since subendocardial dysfunction decreases the counteracting forces generated in the subendocardium, this finding could be associated with an increase in LV twist. Due to longitudinal orientation of subendocardial fibres, longitudinal dysfunction as evidenced by the reported reduction in longitudinal strain might also be a reflection of subendocardial dysfunction.
These four studies clearly showed that OSA independently results in subclinical LV and RV dysfunction; and chronic CPAP therapy not only decreases the severity of OSA, but also ameliorates the sleep-induced longitudinal LV dysfunction.
Effect of Central Sleep Apnoea on Ventricular Mechanics in Sleep Disordered Breathing
CSA with Cheyne–Stokes respiration is another type of SDB, and its prevalence increases in parallel to the severity of CHF.1,2 Controversy exists regarding whether CSA is just a reflection of severely depressed ventricular function with elevated LV filling pressure, or whether it exerts independent pathophysiological effects in the failing heart. Although CSA may expose the failing heart to lesser stress than OSA because of the generation of less negative intrathoracic pressure and milder degrees of hypoxia, its cumulative adverse effects could also result in reduced mechanical and geometrical properties of the already failing heart. Several studies have demonstrated that positive pressure ventilation therapy improves LVEF in patients with CSA and heart failure.21–23 However, the mechanism by which ASV therapy improves cardiac function in CHF remains unclear. Recently, Haruki et al. reported an acute and chronic effect of ASV on left chamber geometry and function in patients with CSA and CHF.24 Although no significant changes in LV volumes and E/E’[K6] were noted, acute ASV therapy (30 minutes) induced a significant increase in stroke volume and cardiac output associated with a reduction in systemic vascular resistance. The same authors also investigated the impact of chronic ASV therapy (six months) on cardiac performance and geometry. In addition to the persistent increase in stroke volume and cardiac output, a significant reduction of LV and left atrial volumes and decreased severity of mitral regurgitation were observed in the ASV group (see Figure 2, which shows one representative case from that group) compared with the withdrawal group. These results suggest that ASV therapy results in a reduction in LV afterload, which is a crucial factor responsible for the improvement of cardiac output in the acute study. In addition, chronic ASV therapy results in sufficient unloading of the left ventricle to cause reverse remodelling of the LV as well as left atrium.
To investigate the acute and chronic impact of ASV on global LV strain, we performed a preliminary study in which we measured global LV strain using 2D speckle tracking analysis in 15 patients with heart failure and CSA who were studied during both acute and chronic ASV therapy. Patient demographics were the same as those in the study by Haruki et al.24 In the acute setting, no significant changes in global radial, circumferential and longitudinal strain were noted before and at 30 minutes after the initiation of ASV therapy (see Figure 3). In the chronic setting, all three global strain values measured at six months after the initiation of ASV therapy were significantly increased compared with the baseline values. Multivariate regression analysis revealed that the decrease in LV end-diastolic volume was an independent determinant of the increase in global longitudinal strain. The lack of a significant improvement in 2D strain in the acute ASV study could be theoretically explained by the short exposure duration to the ASV device, small sample size, and wide standard deviation of baseline 2D strain values. However, this preliminary study supports that chronic ASV therapy induces not only LV reversal remodelling but also an improvement in LV mechanical properties that could be objectively quantified using 2D speckle tracking echocardiography. Future studies are required to validate these initial observations regarding the beneficial effects of ASV therapy on LV mechanics in patients with CHF and CSA.
Conclusion
SDB has detrimental effects on cardiac function. 2D speckle tracking echocardiography is an established, non-invasive and relatively easy method of quantifying LV function in patients with CHF and SDB. This modality has potential for:
- the detection of subclinical LV dysfunction induced by SDB; and
- the evaluation of the effectiveness of positive pressure ventilation therapy in patients with SDB and heart failure.
Recent advances in ultrasound technology mean that not only 2D, but also 3D speckle tracking analysis is now possible. Therefore, the quantitative analysis of the cardiac function of patients with CHF and SDB with speckle tracking echocardiography is definitely an interesting field worthy of further investigation.