Exercise Intolerance in Heart Failure with Preserved Ejection Fraction: Time to Scrutinize Diuretic Therapy?

Heart failure (HF) is the most prevalent and costly heart disease, causing annual expenditures that amount to billions.1 Approximately half of patients with a diagnosis of HF present with normal (preserved) ejection fraction (HFpEF). To date, no therapy has been found to increase life expectancy in this condition. HFpEF is primarily distinguished by diastolic dysfunction attributed to impaired atrial and ventricular relaxation, as well as cardiovascular stiffening. Commonly, HFpEF patients present with severe exercise intolerance, which is associated with substantial morbidity and mortality.1 In HFpEF, such exercise intolerance is further compounded by anaemia as determined by haemoglobin concentration ([Hb]),2 the prevalence (≥43%) of which exceeds that in patients with HF and reduced ejection fraction (HFrEF).2, 3 Anaemia entails a reduction in the oxygen (O2)-carrying capacity of the blood, limiting O2 delivery and thereby exercise tolerance. Nearly nine in 10 anaemic HFpEF patients are reported to exhibit anaemia on the basis of a deficit in total red blood cell volume (RBCV) (i.e. true anaemia), which is not yet routinely quantified in clinical settings.3 The real prevalence of anaemia in HF populations may be higher than estimated by usual [Hb] measurements because low RBCV may be masked in non-anaemic patients by concomitant plasma volume (PV) contraction. Lower than normal PV, but preserved [Hb], has been described by independent research teams in compensated HF patients treated with standard pharmacotherapy including diuretics in a dose-dependent manner.4-6 Of interest, preload reduction diuretic therapy for 1 week prompted decrements of up to 30% in cardiac output in HFpEF but not HFrEF patients, particularly in those with greater impairment of diastolic filling.7 In these individuals, concealed hypovolaemia plausibly constrains the pressure gradient for venous return to the right atrium. Subsequent limitations of diastolic filling, cardiac preload and stroke volume reserve may contribute, in addition to anaemia, to exercise intolerance. This delineates, as a matter of fact, the typical response to acute hypovolaemia in humans. The donation of 1 unit of blood (∼500 mL) immediately provokes 10% decrements in stroke volume, maximal cardiac output (Qmax) and O2 consumption (VO2max), a hallmark of exercise tolerance, which do not recover until blood volume is normalized.8 It should therefore be considered in principle that any factor leading to hypovolaemia, despite normal [Hb], may contribute to the pathophysiology of exercise intolerance in HFpEF. Since the mid-20th century, oral diuretics have represented the mainstay therapy for fluid congestion in HF. At present, prescription is dominated by loop diuretics, consisting of sulfonamide derivatives such as furosemide, bumetanide and torasemide.1 Loop diuretics promote diuresis by diminishing sodium reabsorption in the thick ascending loop of Henle through inhibition of the Na-K-2Cl symporter. The effectiveness of these drugs in relieving symptoms in acute decompensated HF patients is beyond dispute. However, patients often remain in long-term treatment with loop diuretics after congestion has been alleviated.4 This fact holds true for both HFpEF and HFrEF patients, notwithstanding that diuretics have not been subject to the careful scrutiny to which other chronically prescribed drugs in HF have been submitted.9 Indeed, diuretics were introduced prior to the advent of large clinical trials, when the pathophysiology of HF, and particularly of HFpEF, was less well understood. Strikingly, recent albeit accumulating evidence demonstrates up to three- and five-fold increments in cardiovascular death and hospitalization rates, respectively, in HF patients treated with loop diuretics, which are accentuated among those receiving higher daily doses.10, 11 The fundamental aim of diuretic therapy is to achieve and sustain euvolaemia. Ideally, diuretic doses must be adjusted over time according to changes in volume, particularly those affecting the intravascular compartment. Tailoring medication, including loop diuretics, according to remote information from implantable haemodynamic monitoring has been recently shown to halve hospitalization rates in HFpEF patients.12, 13 Nonetheless, quantitative assessment of blood volume is not performed in daily practice and the clinician must rely on indirect signs of congestion such as dyspnoea, elevated jugular venous pressure, pulmonary rales and wheezing.14 In an unprecedented study by van Kraaij et al.,14 3 months of (loop) diuretic withdrawal did not alter any congestion marker in 90% of HFpEF patients. An identical (10%) incidence rate was observed in a parallel HFpEF group in which diuretic therapy was continued.14 However, ventricular diastolic filling, orthostatic tolerance and estimated glomerular filtration rate improved only in the withdrawal group.14 Likewise, persistent decreases in renin–angiotensin system (RAS) activation were noted only in withdrawn HFpEF patients,15 a remarkable effect given the accredited impact of RAS blockers in reducing morbidity and mortality in HF. All these phenotypic modifications associated with diuretic withdrawal are consistent with improved blood volume status, reasonably approaching euvolaemia, in HFpEF patients. Hence, the chronic administration of diuretics may prompt a cascade of sound (haemodynamic-related) mechanisms leading to exercise intolerance in HFpEF, as illustrated in Figure 1. Parenthetically, PV and RBCV are commonly increased by endurance training (ET), the most effective intervention to enhance exercise tolerance in healthy individuals. Gains in VO2max induced by ET are abolished when increases in RBCV and total blood volume are negated by means of phlebotomy.16 This strongly suggests that improvements in exercise tolerance with ET are primarily determined by blood volume adaptations. In addition, older endurance athletes preserve youthful levels of PV and RBCV,17, 18 which indicates that intravascular volumes are not inexorably affected by advanced age. Yet, despite representing a promising strategy, ET prompts exiguous increases in VO2max in HFpEF patients,19 although the optimal modality and dose of exercise remain to be elucidated.20 The question arises as to whether long-term diuretic treatment could preclude any improvement in blood volume, central haemodynamics and thereby major increases in exercise tolerance in HFpEF.19 Nowadays, it is untenable to perform studies using diuretics as monotherapy in HF populations, given the benefits in hard endpoints attributed to other drugs. A viable alternative may be a randomized placebo-controlled study of long-term diuretic dose reduction, which so far has been strictly implemented only once in HFpEF patients.14 This trial might comprise two main arms involving diuretic dose reduction alone and in combination with ET in order to pinpoint the independent and interactive effects of each on blood volume and exercise tolerance. Furthermore, routine monitoring of signs of congestion should be complemented with quantitative measurement of intravascular volumes, and accurate tracing of basal and changes in volume status should be attempted. The proposed investigation would ascertain whether current pharmacotherapy for HFpEF comprising diuretics is optimal in terms of fluid homeostasis and exercise tolerance. It is our conviction that such a determination will set the basis, or at the very least will narrow the search, for much-needed effective therapies for HFpEF, which take us back to the fundamental motive for this article. Conflict of interest: none declared.