Walter J. Paulus

Heart failure with preserved ejection fraction: pathophysiology, diagnosis, and treatment

Half of patients with heart failure (HF) have a preserved left ventricular ejection fraction (HFpEF). Morbidity and mortality in HFpEF are similar to values observed in patients with HF and reduced EF, yet no effective treatment has been identified. While early research focused on the importance of diastolic dysfunction in the pathophysiology of HFpEF, recent studies have revealed that multiple non-diastolic abnormalities in cardiovascular function also contribute. Diagnosis of HFpEF is frequently challenging and relies upon careful clinical evaluation, echo-Doppler cardiography, and invasive haemodynamic assessment. In this review, the principal mechanisms, diagnostic approaches, and clinical trials are reviewed, along with a discussion of novel treatment strategies that are currently under investigation or hold promise for the future.

Myocardial Structure and Function Differ in Systolic and Diastolic Heart Failure

Background— To support the clinical distinction between systolic heart failure (SHF) and diastolic heart failure (DHF), left ventricular (LV) myocardial structure and function were compared in LV endomyocardial biopsy samples of patients with systolic and diastolic heart failure. Methods and Results— Patients hospitalized for worsening heart failure were classified as having SHF (n=22; LV ejection fraction (EF) 34±2%) or DHF (n=22; LVEF 62±2%). No patient had coronary artery disease or biopsy evidence of infiltrative or inflammatory myocardial disease. More DHF patients had a history of arterial hypertension and were obese. Biopsy samples were analyzed with histomorphometry and electron microscopy. Single cardiomyocytes were isolated from the samples, stretched to a sarcomere length of 2.2 &mgr;m to measure passive force (Fpassive), and activated with calcium-containing solutions to measure total force. Cardiomyocyte diameter was higher in DHF (20.3±0.6 versus 15.1±0.4 &mgr;m, P<0.001), but collagen volume fraction was equally elevated. Myofibrillar density was lower in SHF (36±2% versus 46±2%, P<0.001). Cardiomyocytes of DHF patients had higher Fpassive (7.1±0.6 versus 5.3±0.3 kN/m2; P<0.01), but their total force was comparable. After administration of protein kinase A to the cardiomyocytes, the drop in Fpassive was larger (P<0.01) in DHF than in SHF. Conclusions— LV myocardial structure and function differ in SHF and DHF because of distinct cardiomyocyte abnormalities. These findings support the clinical separation of heart failure patients into SHF and DHF phenotypes.

Diastolic Stiffness of the Failing Diabetic Heart Importance of Fibrosis, Advanced Glycation End Products, and Myocyte Resting Tension

Background— Excessive diastolic left ventricular stiffness is an important contributor to heart failure in patients with diabetes mellitus. Diabetes is presumed to increase stiffness through myocardial deposition of collagen and advanced glycation end products (AGEs). Cardiomyocyte resting tension also elevates stiffness, especially in heart failure with normal left ventricular ejection fraction (LVEF). The contribution to diastolic stiffness of fibrosis, AGEs, and cardiomyocyte resting tension was assessed in diabetic heart failure patients with normal or reduced LVEF. Methods and Results— Left ventricular endomyocardial biopsy samples were procured in 28 patients with normal LVEF and 36 patients with reduced LVEF, all without coronary artery disease. Sixteen patients with normal LVEF and 10 with reduced LVEF had diabetes mellitus. Biopsy samples were used for quantification of collagen and AGEs and for isolation of cardiomyocytes to measure resting tension. Diabetic heart failure patients had higher diastolic left ventricular stiffness irrespective of LVEF. Diabetes mellitus increased the myocardial collagen volume fraction only in patients with reduced LVEF (from 14.6±1.0% to 22.4±2.2%, P<0.001) and increased cardiomyocyte resting tension only in patients with normal LVEF (from 5.1±0.7 to 8.5±0.9 kN/m2, P=0.006). Diabetes increased myocardial AGE deposition in patients with reduced LVEF (from 8.8±2.5 to 24.1±3.8 score/mm2; P=0.005) and less so in patients with normal LVEF (from 8.2±2.5 to 15.7±2.7 score/mm2, P=NS). Conclusions— Mechanisms responsible for the increased diastolic stiffness of the diabetic heart differ in heart failure with reduced and normal LVEF: Fibrosis and AGEs are more important when LVEF is reduced, whereas cardiomyocyte resting tension is more important when LVEF is normal.

What Mechanisms Underlie Diastolic Dysfunction in Heart Failure

Abnormalities of diastolic function are common to virtually all forms of cardiac failure. However, their underlying mechanisms, precise role in the generation and phenotypic expression of heart failure, and value as specific therapeutic targets remain poorly understood. A growing proportion of heart failure patients, particularly among the elderly, have apparently preserved systolic function, and this is fueling interest for better understanding and treating diastolic abnormalities. Much of the attention in clinical and experimental studies has focused on relaxation and filling abnormalities of the heart, whereas chamber stiffness has been less well studied, particularly in humans. Nonetheless, new insights from basic and clinical research are helping define the regulators of diastolic dysfunction and illuminate novel targets for treatment. This review puts these developments into perspective with the major aim of highlighting current knowledge gaps and controversies.

Coronary flow reserve calculated from pressure measurements in humans. Validation with positron emission tomography.

BACKGROUND Experimental studies have shown that fractional flow reserve (defined as the ratio of maximal achievable flow in a stenotic area to normal maximal achievable flow) can be calculated from coronary pressure measurements only. The objectives of this study were to validate fractional flow reserve calculation in humans and to compare this information with that derived from quantitative coronary angiography. METHODS AND RESULTS Twenty-two patients with an isolated, discrete proximal or mid left anterior descending coronary artery stenosis and normal left ventricular function were studied. Relative myocardial flow reserve, defined as the ratio of absolute myocardial perfusion during maximal vasodilation in the stenotic area to the absolute myocardial perfusion during maximal vasodilation (adenosine 140 micrograms.kg-1 x min-1 intravenously during 4 minutes) in the contralateral normally perfused area, was assessed by 15O-labeled water and positron emission tomography (PET). Myocardial and coronary fractional flow reserve were calculated from mean aortic, distal coronary, and right atrial pressures recorded during maximal vasodilation. Distal coronary pressures were measured by an ultrathin, pressure-monitoring guide wire with minimal influence on the trans-stenotic pressure gradient. Minimal obstruction area, percent area stenosis, and calculated stenosis flow reserve were assessed by quantitative coronary angiography. There was no difference in heart rate, mean aortic pressure, or rate-pressure product during maximal vasodilation during PET and during catheterization. Percent area stenosis ranged from 40% to 94% (mean, 77 +/- 13%), myocardial fractional flow reserve from 0.36 to 0.98 (mean, 0.61 +/- 0.17), and relative flow reserve from 0.27 to 1.23 (mean, 0.60 +/- 0.26). A close correlation was found between relative flow reserve obtained by PET and both myocardial fractional flow reserve (r = .87) and coronary fractional flow reserve obtained by pressure recordings (r = .86). The correlations between relative flow reserve obtained by PET and stenosis measurements derived from quantitative coronary angiography were markedly weaker (minimal obstruction area, r = .66; percent area stenosis, r = -.70; and stenosis flow reserve, r = .68). CONCLUSIONS Fractional flow reserve derived from pressure measurements correlates more closely to relative flow reserve derived from PET than angiographic parameters. This validates in humans the use of fractional flow reserve as an index of the physiological consequences of a given coronary artery stenosis.

Cardiomyocyte Stiffness in Diastolic Heart Failure

Background—Heart failure with preserved left ventricular (LV) ejection fraction (EF) is increasingly recognized and usually referred to as diastolic heart failure (DHF). Its pathogenetic mechanism remains unclear, partly because of a lack of myocardial biopsy material. Endomyocardial biopsy samples obtained from DHF patients were therefore analyzed for collagen volume fraction (CVF) and sarcomeric protein composition and compared with control samples. Single cardiomyocytes were isolated from these biopsy samples to assess cellular contractile performance. Methods and Results—DHF patients (n=12) had an LVEF of 71±11%, an LV end-diastolic pressure (LVEDP) of 28±4 mm Hg, and no significant coronary artery stenoses. DHF patients had higher CVFs (7.5±4.0%, P<0.05) than did controls (n=8, 3.8±2.0%), and no conspicuous changes in sarcomeric protein composition were detected. Cardiomyocytes, mechanically isolated and treated with Triton X-100 to remove all membranes, were stretched to a sarcomere length of 2.2 &mgr;m and activated with solutions containing varying [Ca2+]. Compared with cardiomyocytes of controls, cardiomyocytes of DHF patients developed a similar total isometric force at maximal [Ca2+], but their resting tension (Fpassive) in the absence of Ca2+ was almost twice as high (6.6±3.0 versus 3.5±1.7 kN/m2, P<0.001). Fpassive and CVF combined yielded stronger correlations with LVEDP than did either alone. Administration of protein kinase A (PKA) to DHF cardiomyocytes lowered Fpassive to control values. Conclusions—DHF patients had stiffer cardiomyocytes, as evident from a higher Fpassive at the same sarcomere length. Together with CVF, Fpassive determined in vivo diastolic LV dysfunction. Correction of this high Fpassive by PKA suggests that reduced phosphorylation of sarcomeric proteins is involved in DHF.

Acute effects of nitric oxide on left ventricular relaxation and diastolic distensibility in humans. Assessment by bicoronary sodium nitroprusside infusion.

BACKGROUND In isolated mammalian cardiomyocytes, papillary muscle preparations, and ejecting hearts, nitric oxide (NO) or other cyclic GMP-elevating interventions increase diastolic cell length and reduce peak contractile performance by hastening onset of myocardial relaxation. In the present study, the effect of NO on left ventricular (LV) relaxation and diastolic distensibility was investigated in humans. METHODS AND RESULTS The NO donor substance sodium nitroprusside was infused during cardiac catheterization in the global coronary bed of the LV of patients (n = 13) investigated for chest pain who were without evidence of obstructive coronary artery or other cardiac disease. Sodium nitroprusside was infused intracoronarily at a dosage (< or = 4 micrograms/min) that was previously shown to be devoid of systemic effects when infused into the brachial artery to investigate the reactivity of the forearm vascular bed. The effect of this global intracoronary infusion of the NO donor sodium nitroprusside was assessed by sequential LV angiograms and tip-micromanometer pressure recordings. During global intracoronary nitroprusside infusion, there was a decrease in heart rate from 78 +/- 11 to 76 +/- 12 beats per minute (P < .05), in LV peak systolic pressure from 161 +/- 18 to 146 +/- 18 mm Hg (P < .001), and in time to onset of LV relaxation (interval from Q wave on the ECG to LV dP/dtmin) from 432 +/- 36 to 419 +/- 36 milliseconds (P < .01). In 7 patients in whom adequate sequential LV angiograms could be obtained, LV end-diastolic volume increased from 158 +/- 34 to 165 +/- 40 mL (P < .05), whereas LV end-diastolic pressure fell from 18 +/- 5 to 12 +/- 3 mm Hg (P < .02), and in 5 of these 7 patients, a downward shift of the diastolic LV pressure-volume relation was observed. In 5 patients, a right atrial infusion of sodium nitroprusside was performed either before (n = 2) or after the global intracoronary infusion. The decrease in LV peak systolic pressure observed during right atrial infusion was significantly smaller (P < .01) than during global intracoronary infusion. CONCLUSIONS The present study reveals reduced LV pressure development, an LV relaxation-hastening effect, and improved LV diastolic distensibility during global intracoronary infusion of the NO donor substance sodium nitroprusside. These effects appeared to be unrelated to systemic vasodilation or to pericardial constraint and could be explained by a direct myocardial effect of NO, probably through activation of guanylyl cyclase to increase cyclic GMP or through modification of other cellular proteins.

Myocardial Contractile Response to Nitric Oxide and cGMP

BACKGROUND Cardiac endothelium releases a number of factors that may modulate performance of underlying cardiac muscle. Nitric oxide (NO), which accounts for the biological activity of the vascular endothelium-derived relaxing factor and relaxes vascular smooth muscle by elevating intracellular cGMP, may be involved in this cardiac modulation. METHODS AND RESULTS We examined the myocardial contractile effects of the NO-releasing nitrovasodilators sodium nitroprusside (SNP), 3-morpholino-sydnonimine (SIN-1), and S-nitroso-N-acetyl-penicillamine (SNAP); of a cGMP analogue, 8-bromo-cGMP; and of the cGMP-phosphodiesterase inhibitor zaprinast in isolated cat papillary muscle. Modulation of these effects by endocardial endothelium (EE) and by cholinergic and adrenergic stimulation was also investigated. Concentration-response curves with addition of NO-releasing nitrovasodilators (SNP, SIN-1, SNAP) and 8-bromo-cGMP resulted in a biphasic inotropic response. Although administration of low concentrations induced a positive inotropic effect, higher concentrations induced a negative inotropic effect. Both NO-induced positive and negative inotropic effects were attenuated by methylene blue, suggesting a role for cGMP. The response to high concentrations of 8-bromo-cGMP was shifted to the right in muscles with damaged EE, whereas cholinergic stimulation shifted the curve leftward. Zaprinast caused a monophasic concentration-dependent positive inotropic effect; damaging the EE shifted the terminal portion of the curve upward. Concomitant cholinergic or adrenergic stimulation modified the response to zaprinast into a negative inotropic response. CONCLUSIONS NO and cGMP induced a concentration-dependent biphasic contractile response. The myocardial contractile effects of NO and cGMP were modulated by the status of EE and by concomitant cholinergic or adrenergic stimulation.

Ultrasound and Microbubble-Targeted Delivery of Macromolecules Is Regulated by Induction of Endocytosis and Pore Formation

Contrast microbubbles in combination with ultrasound (US) are promising vehicles for local drug and gene delivery. However, the exact mechanisms behind intracellular delivery of therapeutic compounds remain to be resolved. We hypothesized that endocytosis and pore formation are involved during US and microbubble targeted delivery (UMTD) of therapeutic compounds. Therefore, primary endothelial cells were subjected to UMTD of fluorescent dextrans (4.4 to 500 kDa) using 1 MHz pulsed US with 0.22-MPa peak-negative pressure, during 30 seconds. Fluorescence microscopy showed homogeneous distribution of 4.4- and 70-kDa dextrans through the cytosol, and localization of 155- and 500-kDa dextrans in distinct vesicles after UMTD. After ATP depletion, reduced uptake of 4.4-kDa dextran and no uptake of 500-kDa dextran was observed after UMTD. Independently inhibiting clathrin- and caveolae-mediated endocytosis, as well as macropinocytosis significantly decreased intracellular delivery of 4.4- to 500-kDa dextrans. Furthermore, 3D fluorescence microscopy demonstrated dextran vesicles (500 kDa) to colocalize with caveolin-1 and especially clathrin. Finally, after UMTD of dextran (500 kDa) into rat femoral artery endothelium in vivo, dextran molecules were again localized in vesicles that partially colocalized with caveolin-1 and clathrin. Together, these data indicated uptake of molecules via endocytosis after UMTD. In addition to triggering endocytosis, UMTD also evoked transient pore formation, as demonstrated by the influx of calcium ions and cellular release of preloaded dextrans after US and microbubble exposure. In conclusion, these data demonstrate that endocytosis is a key mechanism in UMTD besides transient pore formation, with the contribution of endocytosis being dependent on molecular size.

B-Type Natriuretic Peptide and Prognosis in Heart Failure Patients With Preserved and Reduced Ejection Fraction

OBJECTIVES This study sought to determine the prognostic value of B-type natriuretic peptide (BNP) in patients with heart failure with preserved ejection fraction (HFPEF), in comparison to data in HF patients with reduced left ventricular (LV) EF (≤40%). BACKGROUND Management of patients with HFPEF is difficult. BNP is a useful biomarker in patients with reduced LVEF, but data in HFPEF are scarce. METHODS In this study, 615 patients with mild to moderate HF (mean age 70 years, LVEF 33%) were followed for 18 months. BNP concentrations were measured at baseline and were related to the primary outcome, that is, a composite of all-cause mortality and HF hospitalization, and to mortality alone. The population was divided in quintiles, according to LVEF, and patients with reduced LVEF were compared with those with HFPEF. RESULTS There were 257 patients (42%) who had a primary endpoint and 171 (28%) who died. BNP levels were significantly higher in patients with reduced LVEF than in those with HFPEF (p < 0.001). BNP was a strong predictor of outcome, but LVEF was not. Importantly, if similar levels of BNP were compared across the whole spectrum of LVEF, and for different cutoff levels of LVEF, the associated risk of adverse outcome was similar in HFPEF patients as in those with reduced LVEF. CONCLUSIONS BNP levels are lower in patients with HFPEF than in patients with HF with reduced LVEF, but for a given BNP level, the prognosis in patients with HFPEF is as poor as in those with reduced LVEF.