[Skip to Content]
Access to paid content on this site is currently suspended due to excessive activity being detected from your IP address 34.236.170.48. Please contact the publisher to request reinstatement.
[Skip to Content Landing]
Figure 1.
Kaplan-Meier Plots for All-Cause Mortality by Systolic Blood Pressure (SBP) Level
Kaplan-Meier Plots for All-Cause Mortality by Systolic Blood Pressure (SBP) Level

Kaplan-Meier plot for all-cause mortality in 901 pairs of propensity score–matched patients with heart failure and left ventricular ejection fraction of 50% or greater, by SBP level less than 120 vs 120 mm Hg or greater. Hazard ratios for all-cause mortality at 1 month, 1 year, and overall were 2.07 (95% CI, 1.45-2.95; P < .001), 1.36 (95% CI, 1.16-1.59; P < .001), and 1.17 (95% CI, 1.05-1.30; P = .005), respectively.

Figure 2.
Restricted Cubic Spline Plots for All-Cause Mortality by Systolic Blood Pressure
Restricted Cubic Spline Plots for All-Cause Mortality by Systolic Blood Pressure

Hazard ratios and 95% confidence intervals for all-cause mortality by discharge systolic blood pressure level in 3915 patients with heart failure with preserved ejection fraction of 50% or greater according to restricted cubic spline regression models using 4 knots at blood pressures of 110 mm Hg, 120 mm Hg (reference), 140 mm Hg, and 150 mm Hg. Solid black lines indicate hazard ratios, and shaded areas indicate 95% CI. Plots on the left panel (A) are based on 3906 prematch patients (9 prematch patients had systolic blood pressure levels >200 mm Hg and were excluded), adjusting for propensity scores, and those on the right panel (B) are based on 1802 matched patients (none had systolic blood pressure levels >200 mm Hg) balanced on 58 baseline characteristics. Spline curves were truncated at a systolic blood pressure level of 200 mm Hg.

Figure 3.
Forest Plots for Subgroup Analyses of Mortality by Systolic Blood Pressure (SBP) Level
Forest Plots for Subgroup Analyses of Mortality by Systolic Blood Pressure (SBP) Level

Forest plots displaying hazard ratios and 95% confidence intervals for all-cause mortality in subgroups of propensity score–matched patients with heart failure and a left ventricular ejection fraction of 50% or greater by discharge SBP level less than 120 vs 120 mm Hg or greater. ACE indicates angiotensin-converting enzyme; ARB, angiotensin II receptor blocker.

Table 1.  
Baseline Characteristics by Discharge Systolic Blood Pressure (SBP) Level in Patients With Heart Failure With Left Ventricular Ejection Fraction of 50% or Greater
Baseline Characteristics by Discharge Systolic Blood Pressure (SBP) Level in Patients With Heart Failure With Left Ventricular Ejection Fraction of 50% or Greater
Table 2.  
Outcomes by Discharge Systolic Blood Pressure (SBP) Level in 1802 Propensity Score–Matched Patients With Heart Failure With Ejection Fraction of 50% or Greater
Outcomes by Discharge Systolic Blood Pressure (SBP) Level in 1802 Propensity Score–Matched Patients With Heart Failure With Ejection Fraction of 50% or Greater
1.
Kitzman  DW, Little  WC, Brubaker  PH,  et al.  Pathophysiological characterization of isolated diastolic heart failure in comparison to systolic heart failure.  JAMA. 2002;288(17):2144-2150.PubMedGoogle ScholarCrossref
2.
Redfield  MM, Jacobsen  SJ, Burnett  JC  Jr, Mahoney  DW, Bailey  KR, Rodeheffer  RJ.  Burden of systolic and diastolic ventricular dysfunction in the community: appreciating the scope of the heart failure epidemic.  JAMA. 2003;289(2):194-202.PubMedGoogle ScholarCrossref
3.
Yancy  CW, Jessup  M, Bozkurt  B,  et al; American College of Cardiology Foundation; American Heart Association Task Force on Practice Guidelines.  2013 ACCF/AHA guideline for the management of heart failure: a report of the American College of Cardiology Foundation/American Heart Association Task Force on Practice Guidelines.  J Am Coll Cardiol. 2013;62(16):e147-e239.PubMedGoogle ScholarCrossref
4.
Yancy  CW, Jessup  M, Bozkurt  B,  et al.  2017 ACC/AHA/HFSA focused update of the 2013 ACCF/AHA guideline for the management of heart failure: a report of the American College of Cardiology/American Heart Association Task Force on Clinical Practice Guidelines and the Heart Failure Society of America.  Circulation. 2017;136(6):e137-e161.PubMedGoogle ScholarCrossref
5.
Levy  D, Larson  MG, Vasan  RS, Kannel  WB, Ho  KK.  The progression from hypertension to congestive heart failure.  JAMA. 1996;275(20):1557-1562.PubMedGoogle ScholarCrossref
6.
Kostis  JB, Davis  BR, Cutler  J,  et al; SHEP Cooperative Research Group.  Prevention of heart failure by antihypertensive drug treatment in older persons with isolated systolic hypertension.  JAMA. 1997;278(3):212-216.PubMedGoogle ScholarCrossref
7.
Chobanian  AV, Bakris  GL, Black  HR,  et al; National Heart, Lung, and Blood Institute Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure; National High Blood Pressure Education Program Coordinating Committee.  The seventh report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure: the JNC 7 report.  JAMA. 2003;289(19):2560-2572.PubMedGoogle ScholarCrossref
8.
Williamson  JD, Supiano  MA, Applegate  WB,  et al; SPRINT Research Group.  Intensive vs standard blood pressure control and cardiovascular disease outcomes in adults aged ≥75 years: a randomized clinical trial.  JAMA. 2016;315(24):2673-2682.PubMedGoogle ScholarCrossref
9.
Gheorghiade  M, Abraham  WT, Albert  NM,  et al; OPTIMIZE-HF Investigators and Coordinators.  Systolic blood pressure at admission, clinical characteristics, and outcomes in patients hospitalized with acute heart failure.  JAMA. 2006;296(18):2217-2226.PubMedGoogle ScholarCrossref
10.
Desai  RV, Banach  M, Ahmed  MI,  et al.  Impact of baseline systolic blood pressure on long-term outcomes in patients with advanced chronic systolic heart failure (insights from the BEST trial).  Am J Cardiol. 2010;106(2):221-227.PubMedGoogle ScholarCrossref
11.
Banach  M, Bhatia  V, Feller  MA,  et al.  Relation of baseline systolic blood pressure and long-term outcomes in ambulatory patients with chronic mild to moderate heart failure.  Am J Cardiol. 2011;107(8):1208-1214.PubMedGoogle ScholarCrossref
12.
Fonarow  GC, Abraham  WT, Albert  NM,  et al.  Organized Program to Initiate Lifesaving Treatment in Hospitalized Patients With Heart Failure (OPTIMIZE-HF): rationale and design.  Am Heart J. 2004;148(1):43-51.PubMedGoogle ScholarCrossref
13.
Zhang  Y, Kilgore  ML, Arora  T,  et al.  Design and rationale of studies of neurohormonal blockade and outcomes in diastolic heart failure using OPTIMIZE-HF registry linked to Medicare data.  Int J Cardiol. 2013;166(1):230-235.PubMedGoogle ScholarCrossref
14.
Lam  PH, Dooley  DJ, Deedwania  P,  et al.  Heart rate and outcomes in hospitalized patients with heart failure with preserved ejection fraction.  J Am Coll Cardiol. 2017;70(15):1861-1871.PubMedGoogle ScholarCrossref
15.
Fonarow  GC, Stough  WG, Abraham  WT,  et al; OPTIMIZE-HF Investigators and Hospitals.  Characteristics, treatments, and outcomes of patients with preserved systolic function hospitalized for heart failure: a report from the OPTIMIZE-HF registry.  J Am Coll Cardiol. 2007;50(8):768-777.PubMedGoogle ScholarCrossref
16.
Ahmed  A, Husain  A, Love  TE,  et al.  Heart failure, chronic diuretic use, and increase in mortality and hospitalization: an observational study using propensity score methods.  Eur Heart J. 2006;27(12):1431-1439.PubMedGoogle ScholarCrossref
17.
Austin  PC.  Primer on statistical interpretation or methods report card on propensity-score matching in the cardiology literature from 2004 to 2006: a systematic review.  Circ Cardiovasc Qual Outcomes. 2008;1(1):62-67.PubMedGoogle ScholarCrossref
18.
Rosenbaum  PR, Rubin  DB.  The central role of propensity score in observational studies for causal effects.  Biometrika. 1983;70(1):41-55. doi:10.1093/biomet/70.1.41Google ScholarCrossref
19.
Rubin  DB.  Using propensity score to help design observational studies: application to the tobacco litigation.  Health Serv Outcomes Res Methodol. 2001; 2(3-4):169-188. doi:10.1023/A:1020363010465Google ScholarCrossref
20.
Ahmed  MI, White  M, Ekundayo  OJ,  et al.  A history of atrial fibrillation and outcomes in chronic advanced systolic heart failure: a propensity-matched study.  Eur Heart J. 2009;30(16):2029-2037.PubMedGoogle ScholarCrossref
21.
Wahle  C, Adamopoulos  C, Ekundayo  OJ, Mujib  M, Aronow  WS, Ahmed  A.  A propensity-matched study of outcomes of chronic heart failure (HF) in younger and older adults.  Arch Gerontol Geriatr. 2009;49(1):165-171.PubMedGoogle ScholarCrossref
22.
Yancy  CW, Jessup  M, Bozkurt  B,  et al.  2017 ACC/AHA/HFSA focused update of the 2013 ACCF/AHA guideline for the management of heart failure: a report of the American College of Cardiology/American Heart Association Task Force on Clinical Practice Guidelines and the Heart Failure Society of America.  J Am Coll Cardiol. 2017;70(6):776-803.PubMedGoogle ScholarCrossref
23.
Whelton  P, Carey  R, Aronow  W,  et al.  2017 ACC/AHA/AAPA/ABC/ACPM/AGS/APhA/ASH/ASPC/NMA/PCNA guideline for the prevention, detection, evaluation, and management of high blood pressure in adults.  J Am Coll Cardiol. 2017;2017.Google Scholar
24.
Bibbins-Domingo  K, Pletcher  MJ, Lin  F,  et al.  Racial differences in incident heart failure among young adults.  N Engl J Med. 2009;360(12):1179-1190.PubMedGoogle ScholarCrossref
25.
Laskey  WK, Wu  J, Schulte  PJ,  et al.  Association of arterial pulse pressure with long-term clinical outcomes in patients with heart failure.  JACC Heart Fail. 2016;4(1):42-49.PubMedGoogle ScholarCrossref
26.
Greene  HL.  Sudden arrhythmic cardiac death: mechanisms, resuscitation and classification: the Seattle perspective.  Am J Cardiol. 1990;65(4):4B-12B.PubMedGoogle ScholarCrossref
27.
Davies  DW, Kadar  D, Steward  DJ, Munro  IR.  A sudden death associated with the use of sodium nitroprusside for induction of hypotension during anaesthesia.  Can Anaesth Soc J. 1975;22(5):547-552.PubMedGoogle ScholarCrossref
28.
Franciosa  JA, Heckel  R.  Significance of hypotension preceding fatal ventricular tachyarrhythmias in post-coronary obstruction sudden death.  Am Heart J. 1981;101(4):421-427.PubMedGoogle ScholarCrossref
29.
Zile  MR, Gaasch  WH, Anand  IS,  et al; I-Preserve Investigators.  Mode of death in patients with heart failure and a preserved ejection fraction: results from the Irbesartan in Heart Failure With Preserved Ejection Fraction Study (I-Preserve) trial.  Circulation. 2010;121(12):1393-1405.PubMedGoogle ScholarCrossref
30.
Fonarow  GC, Adams  KF  Jr, Abraham  WT, Yancy  CW, Boscardin  WJ; ADHERE Scientific Advisory Committee, Study Group, and Investigators.  Risk stratification for in-hospital mortality in acutely decompensated heart failure: classification and regression tree analysis.  JAMA. 2005;293(5):572-580.PubMedGoogle ScholarCrossref
31.
Núñez  J, Núñez  E, Fonarow  GC,  et al.  Differential prognostic effect of systolic blood pressure on mortality according to left-ventricular function in patients with acute heart failure.  Eur J Heart Fail. 2010;12(1):38-44.PubMedGoogle ScholarCrossref
32.
Vidán  MT, Bueno  H, Wang  Y,  et al.  The relationship between systolic blood pressure on admission and mortality in older patients with heart failure.  Eur J Heart Fail. 2010;12(2):148-155.PubMedGoogle ScholarCrossref
33.
Buiciuc  O, Rusinaru  D, Lévy  F,  et al.  Low systolic blood pressure at admission predicts long-term mortality in heart failure with preserved ejection fraction.  J Card Fail. 2011;17(11):907-915.PubMedGoogle ScholarCrossref
34.
Cleland  JG, Tendera  M, Adamus  J, Freemantle  N, Polonski  L, Taylor  J; PEP-CHF Investigators.  The Perindopril In Elderly People With Chronic Heart Failure (PEP-CHF) Study.  Eur Heart J. 2006;27(19):2338-2345.PubMedGoogle ScholarCrossref
35.
Yusuf  S, Pfeffer  MA, Swedberg  K,  et al; CHARM Investigators and Committees.  Effects of candesartan in patients with chronic heart failure and preserved left-ventricular ejection fraction: the CHARM-preserved trial.  Lancet. 2003;362(9386):777-781.PubMedGoogle ScholarCrossref
36.
Kalantar-Zadeh  K, Block  G, Horwich  T, Fonarow  GC.  Reverse epidemiology of conventional cardiovascular risk factors in patients with chronic heart failure.  J Am Coll Cardiol. 2004;43(8):1439-1444.PubMedGoogle ScholarCrossref
37.
Khalid  U, Ather  S, Bavishi  C,  et al.  Pre-morbid body mass index and mortality after incident heart failure: the ARIC study.  J Am Coll Cardiol. 2014;64(25):2743-2749.PubMedGoogle ScholarCrossref
38.
Fonarow  GC, Abraham  WT, Albert  NM,  et al.  A smoker’s paradox in patients hospitalized for heart failure: findings from OPTIMIZE-HF.  Eur Heart J. 2008;29(16):1983-1991.PubMedGoogle ScholarCrossref
39.
Butler  J, Fonarow  GC, Zile  MR,  et al.  Developing therapies for heart failure with preserved ejection fraction: current state and future directions.  JACC Heart Fail. 2014;2(2):97-112.PubMedGoogle ScholarCrossref
40.
Clarke  R, Shipley  M, Lewington  S,  et al.  Underestimation of risk associations due to regression dilution in long-term follow-up of prospective studies.  Am J Epidemiol. 1999;150(4):341-353.PubMedGoogle ScholarCrossref
Original Investigation
April 2018

Systolic Blood Pressure and Outcomes in Patients With Heart Failure With Preserved Ejection Fraction

Author Affiliations
  • 1Veterans Affairs Medical Center, Washington, DC
  • 2Georgetown University, Washington, DC
  • 3MedStar Washington Hospital Center, Washington, DC
  • 4George Washington University, Washington, DC
  • 5University of Alabama at Birmingham, Birmingham
  • 6University of California-San Francisco, Fresno
  • 7Stony Brook University, Stony Brook, New York
  • 8University of Mississippi, Jackson
  • 9Westchester Medical Center, Valhalla, New York
  • 10New York Medical College, Valhalla
  • 11Northwestern University, Chicago, Illinois
  • 12Deputy Editor, JAMA Cardiology
  • 13Ahmanson-UCLA Cardiomyopathy Center, Division of Cardiology, University of California, Los Angeles
  • 14Associate Editor for Health Care Quality and Guidelines, JAMA Cardiology
JAMA Cardiol. 2018;3(4):288-297. doi:10.1001/jamacardio.2017.5365
Key Points

Question  How is systolic blood pressure associated with outcomes in patients with heart failure with preserved ejection fraction?

Findings  In a propensity score–matched observational study of hospitalized patients with HF and ejection fraction 50% or greater in the national Medicare-linked OPTIMIZE-HF registry, a discharge systolic blood pressure level of less than 120 mm Hg was associated with a significantly higher risk of 30-day, 1-year, and long-term all-cause mortality.

Meaning  A systolic blood pressure level of less than 120 mm Hg identifies patients with heart failure with preserved ejection fraction at higher risk for short- and long-term mortality and emphasizes the need for future prospective studies to evaluate optimal systolic blood pressure treatment goals in this patient population.

Abstract

Importance  Lower systolic blood pressure (SBP) levels are associated with poor outcomes in patients with heart failure. Less is known about this association in heart failure with preserved ejection fraction (HFpEF).

Objective  To determine the associations of SBP levels with mortality and other outcomes in HFpEF.

Design, Setting, and Participants  A propensity score–matched observational study of the Medicare-linked Organized Program to Initiate Lifesaving Treatment in Hospitalized Patients with Heart Failure (OPTIMIZE-HF) registry included 25 354 patients who were discharged alive; 8873 (35.0%) had an ejection fraction of at least 50%, and of these, 3915 (44.1%) had stable SBP levels (≤20 mm Hg admission to discharge variation). Data were collected from 259 hospitals in 48 states between March 1, 2003, and December 31, 2004. Data were analyzed from March 1, 2003, to December 31, 2008.

Exposure  Discharge SBP levels less than 120 mm Hg. A total of 1076 of 3915 (27.5%) had SBP levels less than 120 mm Hg, of whom 901 (83.7%) were matched by propensity scores with 901 patients with SBP levels of 120 mm Hg or greater who were balanced on 58 baseline characteristics.

Main Outcomes and Measures  Thirty-day, 1-year, and overall all-cause mortality and heart failure readmission through December 31, 2008.

Results  The 1802 matched patients had a mean (SD) age of 79 (10) years; 1147 (63.7%) were women, and 134 (7.4%) were African American. Thirty-day all-cause mortality occurred in 91 (10%) and 45 (5%) of matched patients with discharge SBP of less than 120 mm Hg vs 120 mm Hg or greater, respectively (hazard ratio [HR], 2.07; 95% CI, 1.45-2.95; P < .001). Systolic blood pressure level less than 120 mm Hg was also associated with a higher risk of mortality at 1 year (39% vs 31%; HR, 1.36; 95% CI, 1.16-1.59; P < .001) and during a median follow-up of 2.1 (overall 6) years (HR, 1.17; 95% CI, 1.05-1.30; P = .005). Systolic blood pressure level less than 120 mm Hg was associated with a higher risk of heart failure readmission at 30 days (HR, 1.47; 95% CI, 1.08-2.01; P = .02) but not at 1 or 6 years. Hazard ratios for the combined end point of heart failure readmission or all-cause mortality associated with SBP level less than 120 mm at 30 days, 1 year, and overall were 1.71 (95% CI, 1.34-2.18; P < .001), 1.21 (95% CI, 1.07-1.38; P = .004), and 1.12 (95% CI, 1.01-1.24; P = .03), respectively.

Conclusions and Relevance  Among hospitalized patients with HFpEF, an SBP level less than 120 mm Hg is significantly associated with poor outcomes. Future studies need to prospectively evaluate optimal SBP treatment goals in patients with HFpEF.

Introduction

Heart failure (HF) with preserved ejection fraction (HFpEF) is common, and patients with HFpEF have similar poor outcomes as those with HF with reduced ejection fraction (EF).1,2 The American College of Cardiology/American Heart Association/Heart Failure Society of America guideline for HF recommends that systolic blood pressure (SBP) should be controlled in patients with HFpEF,3 and its 2017 update4 makes a new recommendation for an optimal SBP target level of less than 130 mm Hg for patients with HFpEF and persistent hypertension. However, an optimal SBP target level is less clear for patients with HFpEF. Hypertension is a major risk factor for incident HF, and treatment of hypertension, particularly to an SBP target level of less than 120 mm Hg, has been demonstrated to substantially lower the risk of incident HF.5-8 However, once patients develop HF, a lower SBP level may have a paradoxical association with a higher risk of cardiovascular morbidity and mortality.9-11 Less is known about this association in patients with HFpEF, the examination of which is the objective of this study.

Methods
Data Source and Study Population

The current analyses are based on data from the Medicare-linked Organized Program to Initiate Lifesaving Treatment in Hospitalized Patients with Heart Failure (OPTIMIZE-HF) registry, the details of which have been published before.12-14 Briefly, medical records of 48 612 HF hospitalizations from 259 hospitals in 48 states were abstracted using a web-based information system. These patients were hospitalized between March 1, 2003, and December 31, 2004, and had an International Classification of Diseases, Ninth Revision, Clinical Modification code for HF as a principal discharge diagnosis. Data on demographics characteristics, medical history, symptoms, signs, admission and discharge medications, inpatient procedures, and short-term outcomes were collected.12,15 Long-term outcomes data were obtained by linking data from the OPTIMIZE-HF registry to the Medicare data up to December 31, 2008, using a probabilistic linking approach, which identified 26 376 unique patients. A total of 25 354 were discharged alive,13,15 of whom 8873 (35.0%) had HFpEF defined as EF of 50% or greater.3

The original OPTIMIZE-HF protocol was approved by the local or central institutional review boards. The current study is based on a deidentified copy of the Medicare-linked OPTIMZE-HF data approved by the Centers for Medicare and Medicaid Services and by the local institutional review board and Research and Development Committee. Written informed consent was not required owing to the retrospective nature of the study.

Assembly of a Cohort With Stable SBP

Data on SBP levels were collected from patients in the supine position from times closest to admission and discharge, and automated electronic data checks were used to prevent outlying SBP values.9 We excluded 85 patients who had SBP levels greater than 300 mm Hg or less than 60 mm Hg. We restricted our analysis to 3915 patients (44.1%) with stable inpatient SBP levels, defined as an admission to discharge variation of 20 mm Hg or less (median, –4; range, –20 to 20). Of the 3915 patients with stable SBP levels, 1076 patients (27.5%) had a discharge SBP level of less than 120 mm Hg (eFigure 1 in the Supplement). We chose the SBP cutoff of 120 because SBP level less than 120 mm Hg has been shown to be associated with poor outcomes in HF.9-11

Of the 4873 patients (54.9%) with a variable inpatient SBP level (admission to discharge variation >20 mm Hg), 4077 patients (83.7%) had a mean (median; range) decrease of –47 mm Hg (–41 mm Hg; –178 to –21 mm Hg), and 796 patients (14.5%) had a mean (median; range) increase of 35 mm Hg (32 mm Hg; 21 to 102 mm Hg). We chose to restrict our primary analysis to patients with stable inpatient SBP levels to minimize bias due to measurement errors and acute inpatient events affecting SBP. This also allowed us to avoid potential exposure misclassification. For example, a patient with a high admission SBP level (eg, 180 mm Hg) and an in-hospital decrease of 80 mm Hg would be misclassified in the lower (<120 mm Hg) discharge SBP group.

Assembly of a Balanced Cohort

To attenuate bias due to imbalances in baseline characteristics and to enhance our ability to draw inferences about association, we used propensity score matching to assemble a cohort in which patients with SBP levels less than 120 mm Hg vs 120 mm Hg or greater would be expected to be balanced on all measured baseline characteristics.16-19 We estimated propensity scores for discharge SBP level less than 120 mm Hg for each of the 3915 patients with a stable SBP using a nonparsimonious multivariable logistic regression model, in which SBP level less than 120 mm Hg was used as the dependent variable and the 58 baseline variables were used as covariates (eFigure 2 in the Supplement). We included diastolic blood pressure (DBP) in the model to attenuate any independent association of DBP level with outcomes. However, we also repeated the model after excluding DBP.

The propensity score for SBP level less than 120 mm Hg for a patient is that patient’s probability of having an SBP level less than 120 mm Hg given the 58 baseline characteristics used in the propensity score model (57 for the model without DBP). We then used a matching algorithm, described previously,20 to match 901 of 1076 patients (83.7%) with an SBP level of less than 120 mm Hg with 901 patients with an SBP level of 120 mm Hg or greater who had similar propensity scores, thus assembling a matched cohort of 1802 patients (eFigure 1A in the Supplement). Between-group postmatch balance was assessed by estimating absolute standardized differences for each of the 58 baseline characteristics and presented as a Love plot (eFigure 2 in the Supplement).21 An absolute standardized difference of 0% indicates no residual bias, and values less than 10% indicate inconsequential residual bias.

Assembly of Sensitivity Cohorts

To ensure that the findings of our primary cohort were not sensitive to methodological approaches described earlier, we assembled 4 separate sensitivity cohorts. First, we repeated the above process using 4873 patients (54.9%) without a stable SBP level, defined as an admission to discharge variation of at least 20 mm Hg. Of these, 4077 patients (83.7%) had a decrease in SBP level from admission to discharge and a median discharge SBP level of 122 mm Hg. Overall, 1736 patients (42.6%) had a SBP level less than 120 mm Hg, of which 1116 (64.3%) could be matched by propensity scores, thus assembling a matched cohort of 2232 patients (eFigure 1B in the Supplement). The 2232 matched patients had a median (range) decrease in SBP from admission to discharge of –40 mm Hg (–21 to –165 mm Hg) and were balanced on 58 baseline characteristics. We excluded the 796 patients with a variable increase in SBP because they would be expected to be characteristically (only 6% had and SBP level <120 mm Hg) and prognostically different from those with a decrease in SBP level.

We then assembled 2 other sensitivity cohorts using all patients regardless of admission to discharge SBP variations: 4582 matched patients with admission SBP level less than 120 mm Hg vs 120 mm Hg or greater (eFigure 1C in the Supplement) and 2958 matched patients with discharge SBP level less than 120 vs 120 mm Hg or greater (eFigure 1D in the Supplement). Finally, because of the recent guideline recommendations of a target SBP level of less than 130 mm Hg in patients with HFpEF and persistent hypertension,22,23 we assembled another sensitivity cohort in which lower SBP level was defined as less than 130 mm Hg using the prematching data of 3915 patients of our primary analysis. We matched 1114 of 1785 patients (62.4%) with an SBP level less than 130 mm Hg, thus assembling a propensity score–matched cohort of 2228 patients with discharge SBP levels less than 130 mm Hg vs 130 mm Hg or greater that were balanced on 58 baseline characteristics.

Outcomes Data

Our primary outcomes included all-cause mortality and HF readmission at 30 days, 1 year, and during overall median follow-up of 2.1 (overall 6) years up to December 31, 2008. Secondary outcomes included all-cause readmission and the 2 combined end points of all-cause readmission or all-cause mortality and HF readmission or all-cause mortality. All outcomes data were collected from Medicare 100% MedPAR File and 100% Beneficiary Summary File.13

Statistical Analyses

Between-group baseline characteristics were compared using Pearson χ2 and Wilcoxon rank-sum tests, as appropriate. All outcome analyses were conducted using matched data. Kaplan-Meier survival analysis was used to compare cumulative risk of all-cause mortality by discharge SBP level less than 120 vs 120 mm Hg or greater, censoring patients at readmission or study end, whichever came first. Cox proportional hazards models were used to estimate hazard ratios (HRs) and 95% confidence intervals for outcomes associated with SBP level less than 120 mm Hg using SBP level of 120 mm Hg or greater as reference. For the readmission model, patients were censored at death or study end, whichever came first. To assess for nonlinearity, we fitted restricted cubic spline models with 4 knots at SBP levels of 110 mm Hg, 120 mm Hg (reference), 140 mm Hg, and 150 mm Hg using both matched data and prematched data, adjusting for propensity scores. Formal sensitivity analyses were conducted to quantify the degree of hidden bias that could potentially explain any significant associations. Subgroup analyses were conducted to assess homogeneity of the association between discharge SBP less than 120 mm Hg and overall all-cause mortality in clinically relevant subgroups of patients in the primary matched cohort. All statistical analyses were conducted using SPSS statistical software version 24 (IBM SPSS) and SAS statistical software version 9.4 (SAS Institute Inc).

Results
Baseline Characteristics

The 1802 matched patients had a mean (SD) age of 79 (10) years and a mean (SD) EF of 59% (7%). A total of 1147 (63.7%) were women and 134 (7.4%) were African American. The mean (median; range) SBP level was 121 (120; 75-193) mm Hg, with only 13 matched patients having an SBP level less than 90 mm Hg (2 had <80 mm Hg). Before matching, patients with a discharge SBP level less than 120 mm Hg had a lower prevalence of hypertension and diabetes and a higher prevalence of atrial fibrillation, and fewer of these patients were taking angiotensin converting enzyme (ACE) inhibitors and β-blockers (Table 1). These and other measured baseline characteristics were balanced after matching; the absolute standardized differences for all 58 baseline characteristics were less than 10% (Table 1; eFigure 2 in the Supplement).

Outcomes in Patients With HF, EF 50% or Greater, and Stable SBP

Thirty-day all-cause mortality occurred in 91 (10%) and 45 (5%) of matched patients (n = 1802) with a discharge SBP level less than 120 vs 120 mm Hg or greater, respectively (HR, 2.07; 95% CI, 1.45-2.95; P < .001) ( Table 2). The association between an SBP level less than 120 mm Hg and higher risk of all-cause mortality persisted both at 1 year (39% vs 31%; HR, 1.36; 95% CI, 1.16-1.59; P < .001) (Table 2) and 6 years (75% vs 71%; HR, 1.17; 95% CI, 1.05-1.30; P = .005) (Table 2; Figure 1). Findings from our restricted cubic spline analysis demonstrate that there was no evidence of a nonlinear association between SBP and all-cause mortality (P > .20) (Figure 2). Findings from our subgroup analyses demonstrate that the association between SBP levels less than 120 mm Hg and overall all-cause mortality in our matched cohort was homogenous across various clinically relevant subgroups of patients, except those with a glomerular filtration rate less than 45 mL/min/1.73 m2 and those who received a discharge prescription for ACE inhibitors (Figure 3).

Of the 901 matched pairs, 125 pairs (13.9%) clearly had a shorter 30-day survival rate than their matched counterparts, and 85 of 125 pairs (68%) of those patients belonged to the lower SBP group (sign score test P < .001). A hidden covariate that is a near-perfect probability of 30-day all-cause mortality and not strongly associated with any of the 58 variables used in our propensity score model could potentially explain this association if it would also increase the odds of having an SBP level less than 120 mm Hg by about 56%. Patients with SBP levels less than 120 mm Hg also had shorter 1-year and overall survival than their matched counterparts, and these associations were also rather insensitive (sign score test P < .001) to bias by a hidden confounder. Associations of SBP level less than 120 mm Hg with other outcomes are displayed in Table 2.

When we excluded DBP from the logistic regression model for propensity score, we were able to assemble a matched balanced cohort of 2104 patients who had a mean (SD) age of 79 (11) years and a mean (SD) EF of 59% (7%); 1321 (62.8%) were women, and 160 (7.6%) were African American. Matched patients in the 2 SBP groups were balanced on all 57 baseline characteristics. Mean (SD) for DBP patients in the SBP level less than 120 vs 120 mm Hg or greater groups were 60 and 69 mm Hg, respectively, which is similar to the prematch DBP level values (Table 1). Respective SBP values were 108 and 138 mm Hg. Among the 2104 matched patients with imbalanced discharge DBP, a discharge SBP less than 120 mm Hg was associated with a higher risk of all-cause mortality at 30 days (113 [10.7%] vs 45 [4.3%]; HR, 2.59; 95% CI, 1.83-3.66; P < .001), 1 year (422 [40.1%] vs 332 [31.6%]; HR, 1.39; 95% CI, 1.20-1.60; P < .001), and 6 years (790 [75.1%] vs 737 [70.1%]; HR, 1.22; 95% CI, 1.10-1.35; P < .001). Systolic blood pressure level less than 120 mm Hg was also significantly associated with a higher risk of the combined end point of HF readmission or all-cause mortality at all 3 times.

Outcomes in Patients With HF, EF of 50% or Greater, and Variable SBP

The 2232 matched patients with an EF of 50% or greater and a variable SBP level (decrease of >20 mm Hg from admission to discharge) had a mean (SD) age of 79 (10) years and a mean (SD) EF of 59% (7%); 1573 (70.5%) were women, and 237 (10.6%) were African American. The 2 SBP groups were balanced on all 58 baseline characteristics. Before matching, 1285 (74%) of the patients in the lower SBP group had a history of hypertension (vs 670 [62%] in the stable SBP cohort) (Table 1). Overall, all-cause mortality occurred in 771 (69%) and 750 (67.2%) of patients with a discharge SBP level less than 120 vs 120 mm Hg or greater, respectively (HR, 1.10; 95% CI, 0.99-1.22; P = .07) (eTable in the Supplement). Associations with other outcomes are displayed in the eTable in the Supplement.

Outcomes in Patients With HF and EF 50% or Greater, Regardless of SBP Stability

The 4582 matched patients with an EF of 50% or greater regardless of SBP stability had a mean (SD) age of 79 (10) years and a mean (SD) EF of 59% (7%), and 3074 (67.1%) were women and 410 (8.9%) were African American. These patients were balanced on all 58 baseline characteristics. All-cause mortality over 6 years of follow-up occurred in 1637 (72%) and 1595 (69.6%) of patients with a discharge SBP level less than 120 vs 120 mm Hg or greater, respectively (HR, 1.09; 95% CI, 1.02-1.17; P = .01) (eTable in the Supplement). Associations with other outcomes are displayed in the eTable in the Supplement.

Association With Admission SBP Less Than 120 mm Hg

The 2958 matched patients with an EF of 50% or greater had a mean (SD) age of 79 (10) years, a mean (SD) EF of 59% (7%) and 1850 (62.5%) were women and 203 (6.9%) were African American. These patients were balanced on all 58 baseline characteristics. Six-year all-cause mortality occurred in 1128 (76.3%) and 1041 (70.4%) of patients with an admission SBP level less than 120 vs 120 mm Hg or greater, respectively (HR, 1.28; 95% CI, 1.18-1.40; P < .001). Associations with other outcomes are displayed in eTable in the Supplement.

Association With Discharge SBP Less Than 130 mm Hg

The 2228 matched patients had a mean (SD) age of 79 (10) years, a mean (SD) EF of 59% (7%); 1437 (64.5%) were women, and 181 (8.1%) were African American. The 2 SBP groups were balanced on all 58 baseline characteristics. Thirty-day all-cause mortality occurred in 79 (7.1%) and 54 (4.8%) of matched patients with a discharge SBP level less than 130 vs 130 mm Hg or greater, respectively (HR, 1.47; 95% CI, 1.04-2.08; P = .03). A discharge SBP level less than 130 mm Hg was also associated with a higher risk of all-cause mortality at 1 year (375 [33.7%] vs 318 [28.5%]; HR, 1.24; 95% CI, 1.07-1.44; P = .005) and 6 years (803 [72.1%] vs 778 [69.8%]; HR, 1.11; 95% CI, 1.01-1.22; P = .045). Systolic blood pressure level less than 130 mm Hg was also significantly associated with a higher risk of the combined end point of HF readmission or all-cause mortality at all 3 times.

Discussion

Findings from the current study demonstrate that among hospitalized patients with HFpEF, a discharge SBP level of less than 120 mm Hg was associated with a significantly higher risk of 30-day, 1-year, and overall all-cause mortality and that this association was essentially unchanged when lower SBP was defined as SBP levels less than 130 mm Hg. An SBP level less than 120 mm Hg was also associated with a significantly higher risk of the combined end points of HF readmission or mortality at all 3 times. These findings, taken together with those from multiple sensitivity cohorts, provide evidence of a consistent association between a lower SBP level and poor outcomes in patients with HFpEF.

Higher mortality associated with a lower SBP level in patients with HFpEF may reflect differences in cause, pathophysiology, and stage of disease. Hypertension often precedes the development of clinical HF and is the most common cause of morbidity in patients with HF.5,6,24 However, fewer (prematched) patients in the SBP level less than 120 mm Hg group in our study (Table 1) had a history of hypertension, suggesting that HFpEF in these patients may not have been associated with hypertension. A lower SBP level in patients with HFpEF may be due to SBP-lowering drugs; however, fewer patients in the lower SBP group were receiving these drugs. It is possible that a lower SBP level in HFpEF may also reflect a more advanced disease state and lower cardiac output. A higher pulse pressure has been shown to be associated with a higher risk of death in HFpEF (EF ≥ 50%).25 However, that is unlikely to explain our findings because patients in the higher SBP group had a lower mortality despite a higher pulse pressure.

A higher proportion of prematched patients with a lower SBP level were receiving loop diuretics, which may have contributed to the lower SBP level, greater reflex neurohormonal activation, and poor outcomes.16 While our propensity score matching achieved substantial between-group balance in all measured confounders, imbalances in their severity may remain and persist during follow-up. Furthermore, imbalances in unmeasured confounders may also in part explain the observed poor outcomes in the lower SBP group. Nearly half of the matched patients with lower SBP levels were receiving ACE inhibitors and β-blockers, which could have further lowered SBP levels during follow-up. Hypotension has been shown to be associated with sudden cardiac death.26-28 Thus, a higher incidence of sudden cardiac death in the lower SBP group may also in part explain the lack of association with readmission in that group.

The association between SBP level less than 120 mm Hg and HF readmission was modest and variable. If a higher proportion of patients in the lower SBP group had more advanced HF, then pump failure could in part explain the higher HF readmission in that group.29 However, lack of consistency of this association across various cohorts and timeframes suggest this association may not be intrinsic in nature. For example, among patients with stable SBP levels, a discharge SBP level less than 120 mm Hg had a significant association with 30-day HF readmission but not with 1-year and overall HF readmissions. In contrast, among those with variable SBP levels, a discharge SBP level less than 120 mm Hg had a significant association with HF readmission overall but not during shorter follow-up. Finally, an admission SBP level less than 120 mm Hg had a significant association with all outcomes but not with HF readmission.

Several prior studies have examined the association of a lower SBP level with outcomes in hospitalized older adults with HF.9,30-33 Most of these studies were small, and only 1 was dedicated to HFpEF.33 In contrast, our study is distinguished by a larger sample size, the use of a cohort with stable admission to discharge SBP, the use of an EF cutoff of 50% or greater to define HFpEF, the use of propensity score matching to assemble a balanced cohort, the use of subgroup analyses to demonstrate homogeneity, the use of multiple sensitivity analyses, and the use of formal sensitivity analyses to assess bias by a potential unmeasured confounder.

National guidelines for HF and hypertension recommend that SBP levels should be controlled in patients with HFpEF in general and to less than 130 mm Hg in patients with HF and persistent hypertension.22,23 These recommendations are extrapolated from populations without HF because direct evidence from patients with HFpEF is limited. The ACE inhibitors and angiotensin II receptor blockers have been shown to modestly lower SBP levels in patients with HFpEF, but this did not translate into better outcomes.34,35 Findings from observational studies suggest that a lower SBP level may be associated with poor outcomes in patients with HF.9-11 This poorly understood phenomenon of reverse epidemiology is not unique to SBP and has also been described for other traditional HF risk factors.36-38 Findings from the current study provide evidence that a lower SBP level is a marker of underlying pathophysiologic processes that is associated with poor outcomes in patients with HFpEF, an observation that may help design future trials testing new therapies in HFpEF.39 Future prospective randomized clinical trials also need to examine the effect of various SBP target levels on outcomes in patients with HFpEF.

Limitations

There are several limitations to our study. Because of the observational nature of our study, bias due to unmeasured confounders is possible. Findings from our sensitivity analyses suggest that the harmful association of SBP level less than 120 mm Hg and all-cause mortality is rather insensitive to a hidden bias. We had no data on postdischarge SBP level, and SBP crossover during follow-up may result in regression dilution and potential underestimation of the true association.40 Our database on hospitalized patients with HFpEF may not be generalizable to ambulatory patients with HFpEF because determinants of blood pressure as well as measurement of blood pressure in these 2 settings are different.

Conclusions

In hospitalized older patients with HFpEF, an SBP level less than 120 mm Hg is independently associated with a higher risk of all-cause mortality and the combined end points of readmissions or mortality. Future studies need to prospectively evaluate optimal SBP treatment goals in patients with HFpEF.

Back to top
Article Information

Corresponding Author: Ali Ahmed, MD, MPH, Center for Health and Aging, Veterans Affairs Medical Center, 50 Irving St NW, Ste 1D 129D, Washington, DC 20422 (ali.ahmed@va.gov).

Accepted for Publication: December 22, 2017.

Correction: This article was corrected on February 28, 2018, to fix an omission in the correspondence address.

Published Online: February 14, 2018. doi:10.1001/jamacardio.2017.5365

Author Contributions: Dr Ahmed had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Drs Tsimploulis and Lam contributed equally as lead authors. Drs Fonarow and Ahmed contributed equally as senior authors.

Study concept and design: Tsimploulis, Lam, Singh, Butler, Aronow, Yancy, Fonarow, Ahmed.

Acquisition, analysis, or interpretation of data: Tsimploulis, Lam, Arundel, Singh, Morgan, Faselis, Deedwania, Aronow, Fonarow, Ahmed.

Drafting of the manuscript: Tsimploulis, Lam, Arundel, Singh, Faselis, Butler, Aronow, Ahmed.

Critical revision of the manuscript for important intellectual content: Tsimploulis, Lam, Arundel, Singh, Morgan, Deedwania, Butler, Faselis, Aronow, Yancy, Fonarow, Ahmed

Statistical analysis: Tsimploulis, Morgan, Lam, Deedwania, Ahmed.

Obtained funding: Faselis, Fonarow, Ahmed.

Administrative, technical, or material support: Singh, Faselis, Fonarow, Ahmed.

Study supervision: Singh, Deedwania, Butler, Yancy, Fonarow, Ahmed.

Conflict of Interest Disclosures: All authors have completed and submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Dr Fonarow reports consulting with Amgen, Novartis, Medtronic, and St Jude Medical and was the principle investigator of the Organized Program to Initiate Lifesaving Treatment in Hospitalized Patients With Heart Failure. Dr Butler reports consulting with Amgen, Astra-Zeneca, Bayer, Boehringer-Ingelheim, Bristol Mayers Squibb, CVrX, Janssen, Luitpold, Medtronic, Novartis, Relypsa, Roche, Vifor, and ZS Pharma. No other disclosures were reported.

Funding/Support: Dr Ahmed was supported by the National Institutes of Health grants No. R01-HL085561, R01-HL085561-S, and R01-HL097047 from the National Heart, Lung, and Blood Institute. The Organized Program to Initiate Lifesaving Treatment in Hospitalized Patients with Heart Failure (OPTIMIZE-HF) was sponsored by GlaxoSmithKline.

Role of the Funder/Sponsor: The funders had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

Disclaimer: Dr Yancy is deputy editor and Dr Forarow is an associate editor of JAMA Cardiology. They were not involved in the evaluation or decision to accept this article for publication.

References
1.
Kitzman  DW, Little  WC, Brubaker  PH,  et al.  Pathophysiological characterization of isolated diastolic heart failure in comparison to systolic heart failure.  JAMA. 2002;288(17):2144-2150.PubMedGoogle ScholarCrossref
2.
Redfield  MM, Jacobsen  SJ, Burnett  JC  Jr, Mahoney  DW, Bailey  KR, Rodeheffer  RJ.  Burden of systolic and diastolic ventricular dysfunction in the community: appreciating the scope of the heart failure epidemic.  JAMA. 2003;289(2):194-202.PubMedGoogle ScholarCrossref
3.
Yancy  CW, Jessup  M, Bozkurt  B,  et al; American College of Cardiology Foundation; American Heart Association Task Force on Practice Guidelines.  2013 ACCF/AHA guideline for the management of heart failure: a report of the American College of Cardiology Foundation/American Heart Association Task Force on Practice Guidelines.  J Am Coll Cardiol. 2013;62(16):e147-e239.PubMedGoogle ScholarCrossref
4.
Yancy  CW, Jessup  M, Bozkurt  B,  et al.  2017 ACC/AHA/HFSA focused update of the 2013 ACCF/AHA guideline for the management of heart failure: a report of the American College of Cardiology/American Heart Association Task Force on Clinical Practice Guidelines and the Heart Failure Society of America.  Circulation. 2017;136(6):e137-e161.PubMedGoogle ScholarCrossref
5.
Levy  D, Larson  MG, Vasan  RS, Kannel  WB, Ho  KK.  The progression from hypertension to congestive heart failure.  JAMA. 1996;275(20):1557-1562.PubMedGoogle ScholarCrossref
6.
Kostis  JB, Davis  BR, Cutler  J,  et al; SHEP Cooperative Research Group.  Prevention of heart failure by antihypertensive drug treatment in older persons with isolated systolic hypertension.  JAMA. 1997;278(3):212-216.PubMedGoogle ScholarCrossref
7.
Chobanian  AV, Bakris  GL, Black  HR,  et al; National Heart, Lung, and Blood Institute Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure; National High Blood Pressure Education Program Coordinating Committee.  The seventh report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure: the JNC 7 report.  JAMA. 2003;289(19):2560-2572.PubMedGoogle ScholarCrossref
8.
Williamson  JD, Supiano  MA, Applegate  WB,  et al; SPRINT Research Group.  Intensive vs standard blood pressure control and cardiovascular disease outcomes in adults aged ≥75 years: a randomized clinical trial.  JAMA. 2016;315(24):2673-2682.PubMedGoogle ScholarCrossref
9.
Gheorghiade  M, Abraham  WT, Albert  NM,  et al; OPTIMIZE-HF Investigators and Coordinators.  Systolic blood pressure at admission, clinical characteristics, and outcomes in patients hospitalized with acute heart failure.  JAMA. 2006;296(18):2217-2226.PubMedGoogle ScholarCrossref
10.
Desai  RV, Banach  M, Ahmed  MI,  et al.  Impact of baseline systolic blood pressure on long-term outcomes in patients with advanced chronic systolic heart failure (insights from the BEST trial).  Am J Cardiol. 2010;106(2):221-227.PubMedGoogle ScholarCrossref
11.
Banach  M, Bhatia  V, Feller  MA,  et al.  Relation of baseline systolic blood pressure and long-term outcomes in ambulatory patients with chronic mild to moderate heart failure.  Am J Cardiol. 2011;107(8):1208-1214.PubMedGoogle ScholarCrossref
12.
Fonarow  GC, Abraham  WT, Albert  NM,  et al.  Organized Program to Initiate Lifesaving Treatment in Hospitalized Patients With Heart Failure (OPTIMIZE-HF): rationale and design.  Am Heart J. 2004;148(1):43-51.PubMedGoogle ScholarCrossref
13.
Zhang  Y, Kilgore  ML, Arora  T,  et al.  Design and rationale of studies of neurohormonal blockade and outcomes in diastolic heart failure using OPTIMIZE-HF registry linked to Medicare data.  Int J Cardiol. 2013;166(1):230-235.PubMedGoogle ScholarCrossref
14.
Lam  PH, Dooley  DJ, Deedwania  P,  et al.  Heart rate and outcomes in hospitalized patients with heart failure with preserved ejection fraction.  J Am Coll Cardiol. 2017;70(15):1861-1871.PubMedGoogle ScholarCrossref
15.
Fonarow  GC, Stough  WG, Abraham  WT,  et al; OPTIMIZE-HF Investigators and Hospitals.  Characteristics, treatments, and outcomes of patients with preserved systolic function hospitalized for heart failure: a report from the OPTIMIZE-HF registry.  J Am Coll Cardiol. 2007;50(8):768-777.PubMedGoogle ScholarCrossref
16.
Ahmed  A, Husain  A, Love  TE,  et al.  Heart failure, chronic diuretic use, and increase in mortality and hospitalization: an observational study using propensity score methods.  Eur Heart J. 2006;27(12):1431-1439.PubMedGoogle ScholarCrossref
17.
Austin  PC.  Primer on statistical interpretation or methods report card on propensity-score matching in the cardiology literature from 2004 to 2006: a systematic review.  Circ Cardiovasc Qual Outcomes. 2008;1(1):62-67.PubMedGoogle ScholarCrossref
18.
Rosenbaum  PR, Rubin  DB.  The central role of propensity score in observational studies for causal effects.  Biometrika. 1983;70(1):41-55. doi:10.1093/biomet/70.1.41Google ScholarCrossref
19.
Rubin  DB.  Using propensity score to help design observational studies: application to the tobacco litigation.  Health Serv Outcomes Res Methodol. 2001; 2(3-4):169-188. doi:10.1023/A:1020363010465Google ScholarCrossref
20.
Ahmed  MI, White  M, Ekundayo  OJ,  et al.  A history of atrial fibrillation and outcomes in chronic advanced systolic heart failure: a propensity-matched study.  Eur Heart J. 2009;30(16):2029-2037.PubMedGoogle ScholarCrossref
21.
Wahle  C, Adamopoulos  C, Ekundayo  OJ, Mujib  M, Aronow  WS, Ahmed  A.  A propensity-matched study of outcomes of chronic heart failure (HF) in younger and older adults.  Arch Gerontol Geriatr. 2009;49(1):165-171.PubMedGoogle ScholarCrossref
22.
Yancy  CW, Jessup  M, Bozkurt  B,  et al.  2017 ACC/AHA/HFSA focused update of the 2013 ACCF/AHA guideline for the management of heart failure: a report of the American College of Cardiology/American Heart Association Task Force on Clinical Practice Guidelines and the Heart Failure Society of America.  J Am Coll Cardiol. 2017;70(6):776-803.PubMedGoogle ScholarCrossref
23.
Whelton  P, Carey  R, Aronow  W,  et al.  2017 ACC/AHA/AAPA/ABC/ACPM/AGS/APhA/ASH/ASPC/NMA/PCNA guideline for the prevention, detection, evaluation, and management of high blood pressure in adults.  J Am Coll Cardiol. 2017;2017.Google Scholar
24.
Bibbins-Domingo  K, Pletcher  MJ, Lin  F,  et al.  Racial differences in incident heart failure among young adults.  N Engl J Med. 2009;360(12):1179-1190.PubMedGoogle ScholarCrossref
25.
Laskey  WK, Wu  J, Schulte  PJ,  et al.  Association of arterial pulse pressure with long-term clinical outcomes in patients with heart failure.  JACC Heart Fail. 2016;4(1):42-49.PubMedGoogle ScholarCrossref
26.
Greene  HL.  Sudden arrhythmic cardiac death: mechanisms, resuscitation and classification: the Seattle perspective.  Am J Cardiol. 1990;65(4):4B-12B.PubMedGoogle ScholarCrossref
27.
Davies  DW, Kadar  D, Steward  DJ, Munro  IR.  A sudden death associated with the use of sodium nitroprusside for induction of hypotension during anaesthesia.  Can Anaesth Soc J. 1975;22(5):547-552.PubMedGoogle ScholarCrossref
28.
Franciosa  JA, Heckel  R.  Significance of hypotension preceding fatal ventricular tachyarrhythmias in post-coronary obstruction sudden death.  Am Heart J. 1981;101(4):421-427.PubMedGoogle ScholarCrossref
29.
Zile  MR, Gaasch  WH, Anand  IS,  et al; I-Preserve Investigators.  Mode of death in patients with heart failure and a preserved ejection fraction: results from the Irbesartan in Heart Failure With Preserved Ejection Fraction Study (I-Preserve) trial.  Circulation. 2010;121(12):1393-1405.PubMedGoogle ScholarCrossref
30.
Fonarow  GC, Adams  KF  Jr, Abraham  WT, Yancy  CW, Boscardin  WJ; ADHERE Scientific Advisory Committee, Study Group, and Investigators.  Risk stratification for in-hospital mortality in acutely decompensated heart failure: classification and regression tree analysis.  JAMA. 2005;293(5):572-580.PubMedGoogle ScholarCrossref
31.
Núñez  J, Núñez  E, Fonarow  GC,  et al.  Differential prognostic effect of systolic blood pressure on mortality according to left-ventricular function in patients with acute heart failure.  Eur J Heart Fail. 2010;12(1):38-44.PubMedGoogle ScholarCrossref
32.
Vidán  MT, Bueno  H, Wang  Y,  et al.  The relationship between systolic blood pressure on admission and mortality in older patients with heart failure.  Eur J Heart Fail. 2010;12(2):148-155.PubMedGoogle ScholarCrossref
33.
Buiciuc  O, Rusinaru  D, Lévy  F,  et al.  Low systolic blood pressure at admission predicts long-term mortality in heart failure with preserved ejection fraction.  J Card Fail. 2011;17(11):907-915.PubMedGoogle ScholarCrossref
34.
Cleland  JG, Tendera  M, Adamus  J, Freemantle  N, Polonski  L, Taylor  J; PEP-CHF Investigators.  The Perindopril In Elderly People With Chronic Heart Failure (PEP-CHF) Study.  Eur Heart J. 2006;27(19):2338-2345.PubMedGoogle ScholarCrossref
35.
Yusuf  S, Pfeffer  MA, Swedberg  K,  et al; CHARM Investigators and Committees.  Effects of candesartan in patients with chronic heart failure and preserved left-ventricular ejection fraction: the CHARM-preserved trial.  Lancet. 2003;362(9386):777-781.PubMedGoogle ScholarCrossref
36.
Kalantar-Zadeh  K, Block  G, Horwich  T, Fonarow  GC.  Reverse epidemiology of conventional cardiovascular risk factors in patients with chronic heart failure.  J Am Coll Cardiol. 2004;43(8):1439-1444.PubMedGoogle ScholarCrossref
37.
Khalid  U, Ather  S, Bavishi  C,  et al.  Pre-morbid body mass index and mortality after incident heart failure: the ARIC study.  J Am Coll Cardiol. 2014;64(25):2743-2749.PubMedGoogle ScholarCrossref
38.
Fonarow  GC, Abraham  WT, Albert  NM,  et al.  A smoker’s paradox in patients hospitalized for heart failure: findings from OPTIMIZE-HF.  Eur Heart J. 2008;29(16):1983-1991.PubMedGoogle ScholarCrossref
39.
Butler  J, Fonarow  GC, Zile  MR,  et al.  Developing therapies for heart failure with preserved ejection fraction: current state and future directions.  JACC Heart Fail. 2014;2(2):97-112.PubMedGoogle ScholarCrossref
40.
Clarke  R, Shipley  M, Lewington  S,  et al.  Underestimation of risk associations due to regression dilution in long-term follow-up of prospective studies.  Am J Epidemiol. 1999;150(4):341-353.PubMedGoogle ScholarCrossref
×