Why the study?
Does pulse pressure predict cardiovascular events in patients with HFpEF?
Does pulse pressure predict cardiovascular events in patients with HFpEF?
Answers are grounded in this paper and cite their sources.
This editorial highlights that while pulse pressure may serve as a prognostic marker reflecting arterial stiffness in HFpEF, current findings are exploratory and require confirmation with more rigorous methodologies.
This article refers to ‘Clinical significance of pulse pressure in patients with heart failure with preserved left ventricular ejection fraction’ by T. Tokitsu et al., published in this issue on pages 1353–1361. Patients with heart failure and preserved ejection fraction (HFpEF) have been shown to have an abnormal LV diastolic function with features including slow relaxation, decreased recoil, and increased passive myocardial stiffness.1-3 Over the past decades, several groups have investigated the specific alterations on the heart structure and function, by analysing myocardial tissue properties, with the finding of specific alterations ranging from intramyocardial signalling changes, cardiomyocyte hypertrophy, interstitial fibrosis, incomplete relaxation of myocardial strips, and increased cardiomyocyte stiffness, that could be framed in a paradigm considering HFpEF as a consequence of microvascular dysfunction driven by inflammation and co-morbidities.4 However, excluding possible alterations occurring beyond the aortic valve from this interpretative model of cardiovascular alterations in HFpEF patients might prevent a deeper insight into the pathophysiology of this condition. Indeed, an important element to be considered in this context, emerging from recent studies and novel methodological approaches, is the effect of blood pressure (BP) values and vascular haemodynamic parameters on cardiac pathophysiological changes. When considering the systemic circulation, we should avoid the oversimplified view of aorta and large vessels as passive conduits for conveying blood to the periphery. Actually, the elastic large arteries have the important task of dampening the pulsatile output of the left ventricle and of changing the rhythmic, intermittent, and discontinuous activity of the cardiac pump into a continuous support to flow, thus guaranteeing blood flow within the circulatory system also during the diastolic phase of the cardiac cycle. After ejection of stroke volume and closure of the aortic valves, a great quantity of blood remains ‘stored’ in the aorta and large arteries. The potential energy stored in the walls of the aorta in systole then turns into kinetic energy in diastole, pushing the stored blood into the bloodstream (the so-called ‘Windkessel phenomenon’). Thus the aorta behaves as a sort of ‘pump in diastole’,5 ensuring the achievement of a proper BP level during the diastolic phase of the cardiac cycle required to guarantee adequate flow to the periphery. When arterial elastic properties are altered, and an increased aortic stiffness is present, a failure of this aortic ‘diastolic pump’ could result in an adverse impact on cardiac function and structure. Alterations in the viscoelastic properties of the large arteries cause a reduction of the amount of stroke volume that is stored by the aorta during the systolic ejection time, while most of the blood ejected at each systole is ‘pushed’ directly towards the periphery of the vascular system. As a consequence, aortic systolic BP increases, diastolic BP decreases, and therefore pulse pressure (PP) increases. Thus, PP, which represents the fluctuation in pressure around its mean value during the cardiac cycle, not only is a unit element of BP variability, but is also a hallmark of arterial stiffness. Moreover, arterial stiffness causes an earlier superimposition of the forward (centrifugal) and backward (centripetal) pressure waves in the aorta, in the protomesosystolic phase of the cardiac cycle, leading to a further increase in systolic BP. This increase in systolic BP causes an increase in LV afterload, leads to a rise in cardiac workload, and consequently to an increase in LV mass and myocardial oxygen needs, setting the conditions for the development of LV failure. On the other hand, the reduction of diastolic BP in the aorta may cause a reduction in subendocardial blood flow supply,6 and alterations in subendocardial perfusion may occur during diastole also because of development of extravascular compressive forces through the systolic phase of the cardiac cycle. In the absence of haemodynamically significant coronary artery stenosis, the diastolic pressure in the coronary arteries is equal to the diastolic pressure in the ascending aorta. Ischaemic heart disease, with undamaged coronaries, may therefore also be a consequence of diastolic pump failure due to an impairment in aortic viscoelastic properties, causing a reduced ratio between subendocardial oxygen supply and demand. Emerging data appear to support the occurrence of a tight relationship between arterial stiffness, which can be measured as aortic pulse wave velocity, and the diastolic function of the heart, on top of the relationship between pulse wave velocity and the steady BP components.7 It is also reasonable to hypothesize that the reciprocal interactions between LV function and a stiffer aorta might also determine an adverse coupling between these systems in systole, thus contributing to some of the commonly observed features of HFpEF, such as the lability of BP, the increased cardiac energy demand, and the limited cardiovascular reserve (Figure 1).8 In this context, the peculiar pulsatile characteristic behaviour of BP deserves further attention. It has been reported that in middle-aged and elderly individuals, the prognostic power for cardiovascular events of PP is superior to that of systolic or diastolic BP,9 and that an increased PP identifies patients at the highest risk of developing heart failure.10 An additional support for the prognostic relevance of PP for cardiovascular diseases has been provided by Tokitsu et al.11 in the study published in the current issue of the journal. In a group of 512 patients affected by HFpEF, the authors reported an increased risk of cardiovascular and heart failure-related events among subjects with the lowest and highest values of PP; and an increased frequency of coronary-related events among subjects in the highest PP group. This paper offers interesting results, by exploring the relationship between PP (i.e. an indirect index of arterial stiffness) and cardiovascular prognosis, an issue for which there is still very limited evidence in the particular case of subjects with HFpEF. The study by Tokitsu et al., however, is not free from limitations. In particular, PP values were directly correlated with ankle–brachial pulse wave velocity (abPWV), which was the method used in this study to assess arterial stiffness. This method estimates the pulse wave transit time using an oscillometric technique based on cuffs placed at the upper arm and at the ankle level. At present, carotid–femoral PWV is considered to be the gold standard measurement for aortic stiffness, whereas abPWV offers information on arterial distensibility including in this assessment also upper and lower limb muscular arteries. In this way, abPWV focuses not only on elastic but also on muscular arteries, and therefore it is importantly influenced by sympathetic nervous system activity, thus being an imprecise indicator of aortic distensibility. Moreover, the approach of considering only the highest PP and abPWV values from measurements on each side used in the present study is unusual, and not in line with current recommendations for measurement of BP and aortic stiffness,12 which suggest considering the mean of at least two measurements; and even a third one if the difference in the first two measurements is significant. Given the relevant issue addressed by Tokitsu et al.,11 it would also have been interesting to determine not only peripheral PP levels but also central PP, i.e. the PP that reflects the actual impact of LV stroke volume on large arteries and which could thus result in observations of additional pathophysiological interest. A few additional critical aspects of this study should be considered. First, the small number of specific cardiovascular events distributed across quintiles of PP may not have been adequate to ensure the consistency of the results obtained, also because of the relatively small sample size and the uneven quintile distribution of the number of subjects. In particular, while 26 cardiovascular events were recorded (40%) in the group of patients with PP lower than 45 mmHg, in 19 patients they consisted of hospitalization for heart failure decompensation. Thus, if we consider only major cardiovascular outcome (cardiovascular death, coronary-related events, and ischaemic stroke), only seven events occurred in the group characterized by the lowest PP values (11%). Secondly, the data obtained in the group of subjects with PP <45 mmHg, who were characterized by higher NYHA functional class, more frequent use of diuretics (32%), greater presence of male patients (63.4%), and higher prevalence of atrial fibrillation (42%), would have probably required a more in-depth analysis. While it is known that cardiovascular risk is higher in patients with atrial fibrillation, assessment of EF and PWV is not easy and is often unreliable in these patients. Additionally, it is also important to consider that low BP levels may be the result of decreased cardiac performance or the presence of other co-morbidities. The results of the PARTAGE Study13 in fact showed that a lower cardiac performance led to lower systolic BP and PP levels despite higher aortic PWV in the very elderly. In conclusion, although the hypothesis being addressed by Tokitsu et al.11 is of great potential interest for a better understanding of the mechanisms underlying the pathophysiology of cardiovascular complications in subjects with HFpEF, the results of this study, given its methodological limitations, are still to be considered as explorative findings which deserve to be confirmed by studies of larger size and based on state of the art methodology. Conflict of interest: P.S. is consultant for DiaTecne srl, Milan, Italy.
No takes yet. Share an insight, caveat, or question.
Salvi et al. (2016) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: