Saito, Yoshihiko

Background: To describe the baseline characteristics of patients with heart failure and preserved left ventricular ejection fraction enrolled in the PARAGON-HF trial (Prospective Comparison of Angiotensin Receptor Neprilysin Inhibitor With Angiotensin Receptor Blocker Global Outcomes in HFpEF) comparing sacubitril/valsartan to valsartan in reducing morbidity and mortality. Methods and Results: We report key demographic, clinical, and laboratory findings, and baseline therapies, of 4822 patients randomized in PARAGON-HF, grouped by factors that influence criteria for study inclusion. We further compared baseline characteristics of patients enrolled in PARAGON-HF with those patients enrolled in other recent trials of heart failure with preserved ejection fraction (HFpEF). Among patients enrolled from various regions (16% Asia-Pacific, 37% Central Europe, 7% Latin America, 12% North America, 28% Western Europe), the mean age of patients enrolled in PARAGON-HF was 72.7±8.4 years, 52% of patients were female, and mean left ventricular ejection fraction was 57.5%, similar to other trials of HFpEF. Most patients were in New York Heart Association class II, and 38% had ≥1 hospitalizations for heart failure within the previous 9 months. Diabetes mellitus (43%) and chronic kidney disease (47%) were more prevalent than in previous trials of HFpEF. Many patients were prescribed angiotensin-converting enzyme inhibitors or angiotensin receptor blockers (85%), β-blockers (80%), calcium channel blockers (36%), and mineralocorticoid receptor antagonists (24%). As specified in the protocol, virtually all patients were on diuretics, had elevated plasma concentrations of N-terminal pro-B-type natriuretic peptide (median, 911 pg/mL; interquartile range, 464–1610), and structural heart disease. Conclusions: PARAGON-HF represents a contemporary group of patients with HFpEF with similar age and sex distribution compared with prior HFpEF trials but higher prevalence of comorbidities. These findings provide insights into the impact of inclusion criteria on, and regional variation in, HFpEF patient characteristics. CLINICAL TRIAL REGISTRATION: URL: https://www.clinicaltrials.gov . Unique identifier: NCT01920711.

Atrial and brain natriuretic peptides (ANP and BNP) are produced by the heart, and their plasma concentrations are increased in human chronic congestive heart failure. Although separate studies have suggested that circulating levels of the biologically active C-terminal ANP, the biologically inactive N-terminal ANP, and BNP may have diagnostic utility in the detection of left ventricular systolic dysfunction or left ventricular hypertrophy, no studies have directly assessed the relative value of these peptides prospectively. We therefore designed this study to compare the relative ability of the different natriuretic peptides to detect abnormal left ventricular systolic and diastolic function and left ventricular hypertrophy. Using a prospective study design, we investigated 94 patients referred for cardiac catheterization and 15 age-matched normal subjects. The diagnostic abilities of elevated plasma C-terminal ANP, N-terminal ANP-(1-30), and BNP concentrations to identify systolic dysfunction (ejection fraction <45%), diastolic dysfunction (time constant of left ventricular relaxation >55 milliseconds, left ventricular end-diastolic pressure >18 mm Hg), and left ventricular hypertrophy (left ventricular mass index >120 g/m 2 ) were objectively compared by receiver operating characteristic analysis. The areas under the receiver operating characteristic curve of BNP for detecting each of these abnormalities ranged from 0.715 to 0.908 and were significantly greater than those of C-terminal ANP or N-terminal ANP-(1-30). The sensitivity and specificity of an elevated plasma BNP, which we defined as greater than the mean+3 SD of the 15 age-matched normal subjects, were 0.83 and 0.77, respectively, for detecting ejection fraction less than 45%, 0.85 and 0.70 for detecting the time constant of left ventricular relaxation greater than 55 milliseconds, 0.63 and 0.76 for detecting left ventricular end-diastolic pressure greater than 18 mm Hg, and 0.81 and 0.85 for detecting left ventricular mass index greater than 120 g/m 2 . The use of BNP and one other peptide increased sensitivity (0.90 to 0.96), albeit with lower specificity (0.56 to 0.71). An elevated plasma BNP was a more powerful marker of left ventricular systolic dysfunction, left ventricular diastolic dysfunction, and left ventricular hypertrophy than C-terminal ANP or N-terminal ANP-(1-30) in this population of patients with suspected cardiac disease. Measurement of BNP alone or in combination with C-terminal ANP or N-terminal ANP-(1-30) has potential utility for the detection of altered left ventricular structure and function in a patient population at risk for cardiovascular disease.