Skip to main content
PMC home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2014 May 1.
Published in final edited form as: Am J Med. 2013 Mar 16;126(5):401–410. doi: 10.1016/j.amjmed.2013.01.004

Angiotensin-Converting Enzyme Inhibitors and Outcomes in Heart Failure and Preserved Ejection Fraction

Marjan Mujib 1, Kanan Patel 2, Gregg C Fonarow 3, Dalane W Kitzman 4, Yan Zhang 2, Inmaculada B Aban 2, O James Ekundayo 5, Thomas E Love 6, Meredith L Kilgore 2, Richard M Allman 2,7, Mihai Gheorghiade 8, Ali Ahmed 2,7
PMCID: PMC3656660  NIHMSID: NIHMS437546  PMID: 23510948

Abstract

BACKGROUND

The role of angiotensin-converting enzyme (ACE) inhibitors in patients with heart failure and preserved ejection fraction remains unclear.

METHODS

Of the 10,570 patients ≥65 years with heart failure and preserved ejection fraction (≥40%) in OPTIMIZE-HF (2003–2004) linked to Medicare (through December, 2008), 7304 were not receiving angiotensin receptor blockers and had no contraindications to ACE inhibitors. After excluding 3115 patients with pre-admission ACE inhibitor use, the remaining 4189 were eligible for new discharge prescriptions for ACE inhibitors, and 1706 received them. Propensity scores for the receipt of ACE inhibitors, calculated for each of the 4189 patients, were used to assemble a cohort of 1337 pairs of patients, balanced on 114 baseline characteristics.

RESULTS

Matched patients had a mean age of 81 years, mean ejection fraction of 55%, 64% were women and 9% African American. Initiation of ACE inhibitor therapy was associated with lower risk of the primary composite endpoint of all-cause mortality or heart failure hospitalization during 2.4 years of median follow-up (hazard ratio {HR}, 0.91; 95% confidence interval {CI}, 0.84–0.99; p=0.028), but not with individual endpoints of all-cause mortality (HR, 0.96; 95% CI, 0.88–1.05; p=0.373) or heart failure hospitalization (HR, 0.93; 95% CI, 0.83–1.05; p=0.257).

CONCLUSION

In hospitalized older patients with heart failure and preserved ejection fraction not receiving angiotensin receptor blockers, discharge initiation of ACE inhibitor therapy was associated with a modest improvement in the composite endpoint of total mortality or heart failure hospitalization, but had no association with individual endpoint components.

Keywords: ACE inhibitors, Heart Failure, Preserved Ejection Fraction

Nearly half of the estimated 6 million heart failure patients in the United States have diastolic heart failure or heart failure with preserved ejection fraction.1 Most of these patients are older adults and they are prognostically similar to those with systolic heart failure or heart failure with reduced ejection fraction.2,3 Angiotensin-converting enzyme (ACE) inhibitors reduce all-cause mortality in patients with heart failure and reduced ejection fraction.4-6 Although angiotensin receptor blockers did not reduce mortality in patients with heart failure and reduced ejection fraction, they improved outcomes,7,8 and are considered drugs of choice for these patients who cannot tolerate ACE inhibitors.9 However, despite evidence of similar neurohormonal activation in heart failure with preserved ejection fraction,10 there is no clear evidence of efficacy of renin-angiotensin system inhibition in these patients.

The lack of efficacy of angiotensin receptor blockers in patients with heart failure and preserved ejection fraction has now been well established in two large multicenter randomized controlled trials.11,12 The role of ACE inhibitors, on the other hand, is less clear. In the Perindopril in Elderly People with Chronic Heart Failure (PEP-CHF) trial, the only randomized controlled trial of ACE inhibitors in heart failure and preserved ejection fraction, 850 patients (mean age, 75 years) recruited from 8 European countries were randomized to receive perindopril or placebo, and during 2.1 years of median follow-up, perindopril had no effect on the primary endpoint of all-cause mortality or heart failure hospitalization (hazard ratio {HR}, 0.92; p=0.545) or all-cause mortality (HR, 1.09; p=0.665).13

The non-significant effect of perindopril was explained in part by the unexpected low (45%) event rates and loss of power (from 90% to 35%) in PEP-CHF and a substantial open-label perindopril use after the first year of follow-up, before which perindopril tended to reduce the risk the primary endpoint (HR, 0.69; p=0.055) and significantly reduced the risk of heart failure hospitalization (HR, 0.63; p=0.033).13 This early benefit of perindopril in PEP-CHF is similar to the early benefit of enalapril in patients with heart failure and reduced ejection fraction in the Studies of Left Ventricular Dysfunction (SOLVD) in which enalapril had no effect after second year of follow-up.5 These observations, taken together with the neurohormonal activation in heart failure with preserved ejection fraction,10 led us to hypothesize that ACE inhibitor use may be associated with improved outcomes in patients with heart failure and preserved ejection fraction, despite the definitive lack of efficacy of angiotensin receptor blockers in these patients. Therefore, the objective of the current study was to test this hypothesis in a propensity-matched (balanced)14,15 inception cohort (new users)16,17 of restricted (excluding those with contraindications to ACE inhibitors)18,19 patients with heart failure and preserved ejection fraction.

MATERIALS AND METHODS

Data Sources and Study Population

The Organized Program to Initiate Lifesaving Treatment in Hospitalized Patients With Heart Failure (OPTIMIZE-HF) is a national registry of hospitalized heart failure patients and has been well described in the literature.20-22 Briefly, charts from 48,612 hospitalizations due to heart failure occurring in 259 hospitals from 48 states between March 2003 and December 2004 were abstracted by trained staff.20 A primary discharge diagnosis of heart failure was determined based on the International Classification of Diseases, 9th Revision codes 428, 402.01, 402.11, 402.91, 404.01, 404.03, 404.11, and 404.91.22 Of the 48,612 hospitalizations, 20,839 were due to heart failure and preserved ejection fraction ≥40%. Extensive data on baseline demographics, medical history including admission and discharge medications including ACE inhibitors and angiotensin receptor blockers, hospital course, discharge disposition, and physician specialty were also collected.22 Data on contraindications to the use of ACE inhibitors were also collected from patients not receiving these drugs. Missing values for continuous variables were imputed based on values predicted by age, sex and race.

Because OPTIMIZE-HF did not collect data on long-term outcomes, we linked OPTIMIZE-HF to Medicare outcomes data up to December 31, 2008, obtained from the Centers for Medicare and Medicaid Services.23 Of the 20,839 heart failure hospitalizations due to heart failure and preserved ejection fraction, 13,270 could be linked to Medicare data. These events occurred in 11,997 unique patients, 10,889 of whom were 65 years or older,24,25 of whom 10,570 were discharged alive (Figure 1). Because angiotensin receptor blockers have not been shown to improve outcomes in heart failure and preserved ejection fraction,7,11,12,25 we excluded 1871 patients who received angiotensin receptor blockers. Of the remaining 8699 patients, 107 without data on discharge use of ACE inhibitors and another 1288 patients with contraindication to ACE inhibitors were excluded, leading to a final working sample size of 7304 patients who would be eligible for a discharge prescription for ACE inhibitors (Figure 1).

Figure 1.

Figure 1

Open in a new tab

Flow chart displaying assembly of the inception cohort of matched patients with heart failure and preserved ejection fraction. ACE = angiotensin-converting enzyme; OPTIMIZE-HF = Organized Program to Initiate Lifesaving Treatment in Hospitalized Patients With Heart Failure

Assembly of an Inception Cohort

Because prevalent drug use may result in selection bias and left censoring,16,17,26 we assembled an inception cohort of 4189 patients who were not receiving prior ACE inhibitor therapy and a discharge prescription for ACE inhibitors for these patients would be considered an initiation of therapy. Therefore, we excluded 3115 patients who received ACE inhibitors before hospital admission. Of the 4189 patients with no history of prior ACE inhibitor use or no contraindication to new ACE inhibitor therapy, 1706 (41%) received a new discharge prescription for ACE inhibitors (Figure 1).

Assembly of a Balanced Cohort

In well-designed randomized controlled trial, the probability of receiving a treatment is 50%, regardless of whether a patient is randomized to the treatment or the placebo group. However, treatment assignment in the real world is seldom random, and as such, these probabilities in non-randomized controlled trial studies would vary between 0 and 100%. These probabilities are often dictated by various measured and unmeasured patient and care characteristics. In real-world patients with heart failure, the probability of the receipt of an ACE inhibitor may be influenced by age, ejection fraction, blood pressure, serum potassium, serum creatinine, known adverse effects, and perceived or real contraindications. For example, 75 year-old patient with heart failure and preserved ejection fraction who have low blood pressure and high serum potassium will likely have a low probability of receiving ACE inhibitors, while 45 year-old patient with heart failure and reduced ejection fraction who have normal blood pressure and normal serum potassium will likely have a high probability of receiving these drugs. These probabilities or propensity scores for the receipt of ACE inhibitors are predicted by data and may be similar in two patients. However, it is possible that one of these two patients actually received ACE inhibitor while the other patient did not. These two patients could then be matched to assemble a pair of patients receiving and not receiving ACE inhibitors who had similar predicted probabilities of receiving ACE inhibitors. In a properly conducted propensity-matched study, patients receiving and not receiving a treatment, such as an ACE inhibitor, would be balanced on all measured baseline characteristics.14,27-31 Importantly, this balance can be achieved while remaining blinded to study outcomes, a key feature of randomized controlled trial.27

We used propensity scores for the receipt of ACE inhibitors to assemble our study cohort so that patients receiving and not receiving these drugs would be balanced on all measured baseline characteristics.31-33 We estimated propensity scores for each of the 4189 patients using a non-parsimonious multivariable logistic regression model.32,33 In the model, the receipt of ACE inhibitors was the dependent variable, and 114 baseline characteristics displayed in Figure 2 were used as covariates. Although propensity scores can be used in regression models or for stratification, matching by propensity scores allows assembly of cohorts in which baseline balance can be estimated and displayed in visually pleasant tabular forms. We used a greedy matching protocol to match 1337 (78%) of the 1706 patients receiving ACE inhibitors with 1337 patients who did not receive ACE inhibitors but had the same propensity or probability to receive them.34,35 The effectiveness of propensity score model was assessed by estimating absolute standardized differences,15,28,36 and presented as a Love plot.37-39 A difference of 0% indicates no residual bias and values <10% are considered inconsequential.

Figure 2.

Figure 2

Open in a new tab

Love plot displaying absolute standardized differences comparing 114 baseline characteristics between older patients with heart failure and preserved ejection fraction, receiving a new discharge prescription of angiotensin-converting enzyme inhibitors, before and after propensity score matching

(Hx = medical history, A = admission, D = discharge, H = in-hospital, PF = precipitating factor; *the total number of variables do not equal 114 as the 4 hospital regions were entered as a single categorical variable in the model)

Mortality and Hospitalization

The primary outcome of the current analysis was the composite endpoint of all-cause mortality or heart failure hospitalization.24 Secondary outcomes included all-cause mortality, heart failure and all-cause hospitalizations. Data on mortality and hospitalization were obtained from the 100% MedPAR File and 100% Beneficiary Summary File between January 1, 2002 and December 31, 2008.

Statistical Analysis

For descriptive analyses, we used Pearson's Chi-square and Wilcoxon rank-sum tests for the pre-match data, and McNemar's test and paired sample t-test for post-match comparisons, as appropriate. Because measured baseline characteristics are balanced in propensity-matched cohorts, we used bivariate Cox proportional hazard models to determine the associations of a new discharge prescription for ACE inhibitors (independent variable) with outcomes (dependent variable) among matched patients during 6 years of follow-up (median, and 25th and 75th percentiles, 2.4, 0.7 and 4.5 years, respectively). Log-minus-log survival plots were used to check proportional hazards assumptions. We conducted a formal sensitivity analysis to estimate the degree of hidden bias that could potentially explain away a significant association between ACE inhibitors and the primary composite outcome among our matched patients.40 Subgroup analyses were conducted to determine the homogeneity of association between the use of ACE inhibitors and the composite primary outcome. Because an older cohort with a long follow-up will ultimately have 100% mortality, estimation of number needed to treat using absolute risk difference may be less useful. Therefore, using a formula proposed for survival analyses, we estimated number needed to treat with ACE inhibitors to prevent one primary composite endpoint event.41 All statistical tests were two-tailed and 95% confidence intervals (CI) were constructed. Finally, we examined the association of ACE inhibitors with outcomes among pre-match patients using multivariable Cox regression models adjusting for (1) all 114 baseline characteristics used in the propensity model and (2) propensity scores. All data analyses were performed using SPSS for Windows version 18 (SPSS, Inc., 2009, Chicago, IL).

RESULTS

Baseline Characteristics

Matched patients (n=2674) had a mean (±SD) age of 81 (±8) years, mean (±SD) LVEF of 55% (±9), 63% were women and 9% were African American. Before matching, patients receiving a new prescription for ACE inhibitors were more likely to be symptomatic but had lower prevalence of comorbidities such as atrial fibrillation and chronic kidney disease. These and other pre-match imbalances were balanced after matching (Tables 1 and 2, and Figure 2). Absolute standardized differences for all 114 baseline characteristics between the two treatment groups were <10% (mostly <5%) suggesting substantial bias reduction.

Table 1.

Baseline patients and care characteristics of older patients with heart failure and preserved ejection fraction, by a new discharge prescription for angiotensin-converting enzyme (ACE) inhibitors

Variables Mean (SD) or n (%) Before propensity score matching
 After propensity score matching
Use of ACE Inhibitors
 P value Use of ACE Inhibitors
 P value
No (n=2483) Yes (n=1706) No (n=1337) Yes (n=1337)
Age (years) 81 (8) 81 (8) 0.713 81 (8) 81 (8) 0.793
Female 1611 (65) 1087 (64) 0.439 843 (63) 860 (64) 0.513
African American 184 (7) 174 (10) 0.001 127 (10) 116 (9) 0.503
Left ventricular ejection fraction (%) 56 (9) 54 (10) <0.001 55 (9) 55 (10) 0.870
Precipitating factors for hospital admission
    Ischemic heart disease 242 (10) 219 (13) 0.002 160 (12) 152 (11) 0.676
    Uncontrolled hypertension 166 (7) 185 (11) <0.001 107 (8) 124 (9) 0.252
    Worsening renal function 112 (5) 45 (3) 0.002 46 (3) 41 (3) 0.668
    Arrhythmia 395 (16) 248 (15) 0.226 201 (15) 205 (15) 0.872
    Non-adherence to diet 58 (2) 51 (3) 0.192 34 (3) 36 (3) 0.904
    Non-adherence to medications 88 (4) 108 (6) <0.001 54 (4) 72 (5) 0.117
Past Medical History
    No prior heart failure hospitalization 348 (14) 321 (19) <0.001 215 (16) 209 (16) 0.786
    Coronary artery disease 1047 (42) 695 (41) 0.357 551 (41) 556 (42) 0.875
    Hypertension 1722 (69) 1205 (71) 0.374 923 (69) 939 (70) 0.538
    Diabetes mellitus 842 (34) 549 (32) 0.243 438 (33) 440 (33) 0.968
    Atrial fibrillation 972 (39) 585 (34) 0.001 489 (37) 493 (37) 0.906
    Hyperlipidemia 662 (27) 515 (30) 0.013 385 (29) 394 (30) 0.729
    Chronic obstructive pulmonary disease 804 (32) 457 (27) <0.001 378 (28) 377 (28) 1.000
    Peripheral vascular disease 334 (14) 201 (12) 0.112 155 (12) 165 (12) 0.598
    Chronic kidney disease 1608 (65) 984 (58) <0.001 798 (60) 795 (60) 0.937
Admission symptoms & signs
    Dyspnea on exertion 1465 (59) 1109 (65) <0.001 846 (63) 837 (63) 0.747
    Fatigue 567 (23) 350 (21) 0.074 267 (20) 289 (22) 0.317
    Orthopnea 493 (20) 469 (28) <0.001 331 (25) 321 (24) 0.686
    Paroxysmal nocturnal dyspnea 243 (10) 249 (15) <0.001 168 (13) 168 (13) 1.000
    Dyspnea at rest 1048 (42) 755 (44) 0.188 582 (44) 582 (44) 1.000
    Chest pain 510 (21) 361 (21) 0.627 277 (21) 284 (21) 0.779
    Pulse (beats/minute) 85 (22) 85 (21) 0.390 85 (22) 84 (21) 0.630
    Systolic blood pressure (mm Hg) 144 (30) 151 (31) <0.001 148 (31) 148 (29) 0.807
    Diastolic blood pressure (mm Hg) 73 (17) 77 (19) <0.001 75 (18) 75 (18) 0.865
    Jugular venous pressure elevation 614 (25) 473 (28) 0.030 370 (28) 351 (26) 0.435
    Third heart sound 118 (5) 118 (7) 0.003 76 (6) 73 (6) 0.865
    Pulmonary rales 1485 (60) 1122 (66) <0.001 839 (63) 847 (63) 0.775
    Lower extremity edema 1621 (65) 1105 (65) 0.732 873 (65) 866 (65) 0.806
Laboratory values
    Serum sodium (mEq/L) 137 (10) 137 (11) 0.529 137 (11) 137 (10) 0.647
    Serum creatinine (mg/dL)* 1.20 (0.70) 1.10 (0.60) <0.001 1.20 (0.70) 1.20 (0.60) 0.980
    Serum hemoglobin (g/dL) 11.9 (2.1) 12.0 (2.1) 0.067 12.0 (2.0) 11.9 (2.1) 0.639
    Serum brain natriuretic peptide, (pg/mL)* 740 (690.82) 806 (773.18) 0.003 753 (717.50) 780 (744.68) 0.604
    Serum troponin elevationt† 359 (15) 272 (16) 0.187 194 (15) 206 (15) 0.546
Length of hospital stay 6 (5) 14 (354) 0.242 6 (5) 6 (4) 0.245
Hospital characteristics
    Bed size* 350 (221) 375 (200) <0.001 355 (212) 360 (207) 0.790
    Academic 985 (40) 784 (46) <0.001 604 (45) 578 (43) 0.334
    Interventional 1816 (73) 1334 (78) <0.001 1030 (77) 1031 (77) 1.000
    Transplant 363 (15) 222 (13) 0.141 201 (15) 175 (13) 0.161
Hospital location by region
    Midwest 702 (28) 560 (33) 416 (31) 419 (31)
    Northeast 420 (17) 274 (16) <0.001 232 (17) 224 (17) 0.776
    South 879 (35) 499 (29) 402 (30) 413 (31)
    West 482 (19) 373 (22) 287 (22) 281 (21)
Open in a new tab
*

Displayed as median (interquartile range)

†

Determined by local laboratories

Table 2.

Procedures and treatment in older patients with heart failure and preserved ejection fraction, by a new discharge prescription for angiotensin-converting enzyme (ACE) inhibitors

Variables Mean (SD) or n (%) Before propensity score matching
 After propensity score matching
Use of ACE Inhibitors
 P value Use of ACE Inhibitors
 P value
No (n=2483) Yes (n=1706) No (n=1337) Yes (n=1337)
Admission medication
    Beta-blockers 1168 (47) 745 (44) 0.031 634 (47) 621 (46) 0.639
    Aldosterone antagonists 104 (4) 47 (3) 0.014 47 (4) 44 (3) 0.828
    Angiotensin receptor blockers* – – – – – –
    Diuretics 1538 (62) 838 (49) <0.001 732 (55) 724 (54) 0.778
    Digoxin 453 (18) 276 (16) 0.083 223 (17) 231 (17) 0.717
    Hydralazine 73 (3) 23 (1) 0.001 20 (2) 20 (2) 1.000
    Nitrates 467 (19) 266 (16) 0.007 219 (16) 220 (17) 1.000
    Amlodipine 230 (9) 137 (8) 0.166 103 (8) 112 (8) 0.573
    Non-amlodipine calcium channel blockers 480 (19) 259 (15) 0.001 196 (15) 210 (16) 0.476
    Anti-arrhythmic drugs 233 (9) 113 (7) 0.001 101 (8) 95 (7) 0.711
    Warfarin 584 (24) 322 (19) <0.001 284 (21) 286 (21) 0.963
    Anti-platelet drugs 267 (11) 166 (10) 0.285 141 (11) 140 (11) 1.000
    Aspirin 869 (35) 590 (35) 0.782 479 (36) 478 (36) 1.000
    Statins 638 (26) 437 (26) 0.954 365 (27) 356 (27) 0.730
In-hospital treatment/procedure
    Dobutamine 20 (1) 21 (1) 0.169 12 (1) 9 (1) 0.664
    Dopamine 51 (2) 26 (2) 0.210 16 (1) 17 (1) 1.000
    Milrinone 9 (0.4) 8 (0.5) 0.594 6 (0.4) 6 (0.4) 1.000
    Nesiritide 190 (8) 114 (7) 0.235 88 (7) 93 (7) 0.762
    Right heart catheterization 48 (2) 53 (3) 0.015 35 (3) 38 (3) 0.813
    Coronary angiography 117 (5) 132 (8) <0.001 77 (6) 85 (6) 0.578
    Coronary artery bypass grafting 12 (0.5) 13 (0.8) 0.250 8 (0.6) 6 (0.4) 0.791
    Percutaneous coronary intervention 18 (1) 20 (1) 0.133 14 (1) 13 (1) 1.000
    Electrophysiological study 17 (1) 9 (1) 0.525 8 (1) 8 (1) 1.000
    Cardioversion 30 (1) 16 (1) 0.409 14 (1) 14 (1) 1.000
    Pacemaker-biventricular 9 (0.4) 12 (0.7) 0.125 8 (1) 10 (1) 0.815
    Dialysis 82 (3) 35 (2) 0.016 29 (2) 32 (2) 0.788
Discharge medication
    Beta-blockers 1318 (53) 1603 (62) <0.001 790 (59) 783 (59) 0.811
    Aldosterone antagonists 179 (7) 155 (9) 0.028 99 (7) 112 (8) 0.385
    Angiotensin receptor blockers* – – – – – –
    Diuretics 1932 (78) 1399 (82) 0.001 1079 (81) 1084 (81) 0.842
    Digoxin 543 (22) 394 (23) 0.349 290 (22) 304 (23) 0.553
    Hydralazine 94 (4) 27 (2) <0.001 27 (2) 26 (2) 1.000
    Nitrates 562 (23) 412 (24) 0.254 329 (25) 318 (24) 0.650
    Amlodipine 223 (9) 122 (7) 0.034 95 (7) 110 (8) 0.311
    Non-amlodipine calcium channel blockers 500 (20) 198 (12) <0.001 167 (13) 182 (14) 0.410
    Anti-arrhythmic drugs 283 (11) 149 (9) 0.005 137 (10) 124 (9) 0.434
    Warfarin 669 (27) 447 (26) 0.594 366 (27) 366 (27) 1.000
    Anti-platelet drugs 298 (12) 234 (14) 0.102 178 (13) 176 (13) 0.955
    Aspirin 996 (40) 850 (50) <0.001 616 (46) 605 (45) 0.687
    Statins 646 (26) 512 (30) 0.005 398 (30) 381 (29) 0.487
Open in a new tab
*

Patients receiving angiotensin receptor blockers on admission and during discharge were excluded

New Prescriptions for ACE Inhibitors and Outcomes

During 2.4 years of median follow-up, the primary composite endpoint of all-cause mortality or heart failure hospitalization occurred in 80% (1076/1337) and 83% (1112/1337) of matched patients with heart failure and preserved ejection fraction receiving and not receiving a new discharge prescription for ACE inhibitors, respectively, (hazard ratio {HR} when the use of ACE inhibitors was compared with their non-use, 0.91; 95% CI, 0.84–0.99; p=0.028; Figure 3 and Table 3). An estimated 71 (95% CI, 40–646) patients will need to be treated over a median 2.4 years of follow-up to prevent one primary composite endpoint event. The association between new ACE inhibitor use and the primary composite endpoint was homogeneous across various subgroups of patients (Figure 4). ACE inhibitor use had no significant association with individual endpoints components of all-cause mortality and hospitalization (Table 3).

Figure 3.

Figure 3

Open in a new tab

Kaplan-Meier plot for primary composite endpoint of all-cause mortality or heart failure hospitalization in a propensity-matched inception cohort of older patients with heart failure and preserved ejection fraction, receiving and not receiving a new discharge prescription for angiotensin-converting enzyme (ACE) inhibitors (HR = hazard ratio, CI = confidence interval)

Table 3.

Outcomes by a new discharge prescription for angiotensin-converting enzyme (ACE) inhibitors in a propensity-matched inception cohort of older patients with heart failure and preserved ejection fraction

Outcomes Events (%)
 Hazard ratio* (95% confidence interval) P value
No ACE Inhibitors (n=1337) ACE Inhibitors (n=1337)
Combined endpoint of all-cause mortality or heart failure hospitalization 1112 (83%) 1076 (80%) 0.91 (0.84–0.99) 0.028
All-cause mortality 951 (71%) 930 (70%) 0.96 (0.88–1.05) 0.373
Heart failure hospitalization 564 (42%) 558 (42%) 0.93 (0.83–1.05) 0.257
All-cause hospitalization 1155 (86%) 1165 (87%) 0.97 (0.89–1.05) 0.401
Open in a new tab
*

Hazard ratios comparing patients receiving ACE inhibitors versus those not receiving ACE inhibitors calculated using Cox regression model

Figure 4.

Figure 4

Open in a new tab

Association of a new discharge prescription of angiotensin-converting enzyme (ACE) inhibitors with primary composite endpoint of all-cause mortality or heart failure hospitalization in subgroups of propensity-matched inception cohort of older patients with heart failure and preserved ejection fraction

Among the 4189 pre-match patients, the primary composite endpoint occurred in 79% (1351/1706) and 84% (2079/2483) of patients receiving and not receiving a new discharge prescription for ACE inhibitors, respectively (HR, 0.84; 95% CI, 0.78–0.90; p<0.001). Multivariable-adjusted and propensity-adjusted HRs for primary composite endpoint associated with ACE inhibitor use were 0.93 (95% CI, 0.86–1.00; p=0.049) and 0.94 (95% CI, 0.87–1.01; p=0.098), respectively.

DISCUSSION

Findings from our study demonstrate that a new discharge prescription for ACE inhibitors was associated with a statistically significant modest 9% lower risk of the composite endpoint of all-cause mortality or heart failure hospitalization in a wide spectrum of propensity-matched older patients with heart failure and preserved ejection fraction who were balanced on over one hundred potential confounders. Similar multivariable-adjusted or propensity-adjusted associations were observed when traditional regression-based risk adjustment models were used in the pre-match cohort. However, ACE inhibitors had no significant association with individual endpoint components of all-cause mortality or heart failure hospitalization. Findings from the current rigorously-conducted propensity-matched inception cohort study based on nationally representative real-world patients provide evidence that the use of ACE inhibitors may be associated with a modest improvement in the long-term composite endpoint of total mortality or heart failure hospitalization in older patients with heart failure and preserved ejection fraction.

The 9% reduction in the composite endpoint in our study is substantially smaller than the 26% reduction in the same endpoint in younger systolic heart failure patients in the SOLVD trial.5 In the SOLVD trial, enalapril seemed to have a more robust effect on heart failure hospitalization than on mortality which in part may also explain the overall benefit of ACE inhibitors in heart failure patients with preserved ejection fraction. The effect of ACE inhibitors may also be mediated by their beneficial effect on aortic stenosis, the prevalence of which would be expected to be high in older heart failure patients with preserved ejection fraction. The inhibition of the renin-angiotensin system has been shown to be associated with improved outcomes in patients with aortic stenosis.42 The lack of significant association with all-cause mortality in our study may in part be explained by the different modes of death in heart failure patients with preserved versus reduced ejection fraction. Findings from major randomized controlled trial of ACE inhibitors in systolic heart failure suggest that these drugs had no significant effect on sudden cardiac death but had a robust effect on death due to pump failure.5,6 While sudden death accounts for between 40% and 50% of cardiovascular deaths in heart failure patients regardless of ejection fraction, death due to pump failure is less common in those with preserved ejection fraction, accounting for 24% of cardiovascular deaths (versus 41% in those with reduced ejection fraction).43 This may in part explain the lack of an effect of ACE inhibitors on mortality in patients with heart failure and preserved ejection.

Most randomized controlled trials of ACE inhibitors in heart failure excluded those with preserved ejection fraction. The overall direction and magnitude of the associations with primary endpoint observed in our study (9% reduction) are consistent with those from PEP-CHF (8% reduction).13 A recent propensity-matched study of ACE inhibitors or angiotensin receptor blockers based on the Swedish Heart Failure Registry reported mortality reduction in patients with heart failure and preserved ejection fraction.44 This association seems inflated as nearly 25% of patients in that study were receiving angiotensin receptor blockers, with proven lack of effect on mortality.11,12 In addition, in PEP-CHF, perindopril had no effect on all-cause mortality, not even during the first year of follow-up, when it reduced heart failure hospitalization, suggesting lack of efficacy on mortality.13 That study based on the Swedish Heart Failure Registry was also limited by biases due to lack of restriction to patients without contraindications,18,19 lack of exclusion of prevalent drug users,16,17 and incomplete matching,45 as over a quarter of 43 variables used in propensity matching were imbalanced after matching.44 Despite potential overestimation of the association in the Swedish Heart Failure Registry, findings from PEP-CHF and our study suggest that ACE inhibitor therapy may be associated with a very modest improvement in the long-term clinical outcomes in patients with heart failure and preserved ejection fraction. However, given the lack of benefit of angiotensin receptor blockers in those patients,7,11,12,25 these findings need to be interpreted with caution and be replicated in other restricted propensity-matched inception cohorts.

Our study has several limitations. Findings from our sensitivity analysis suggest that this association could be potentially explained away by a hidden covariate that would increase the odds for the receipt of ACE inhibitors by about 1%. However, to act as a confounder, an unmeasured covariate must be a near-perfect predictor of outcome and also not be strongly correlated with any of the 114 measured baseline covariates used in our study, which is unlikely. We were able to match nearly 80% of patients receiving ACE inhibitors, thus minimizing any effect on external validity. We had no data on names and doses for individual ACE inhibitors. We also had no data on the use of ACE inhibitors after discharge.46 Substantial crossover may result in regression dilution,47 and may potentially explain the modest associations observed in our study. Lack of data on aortic stenosis is another limitation. The clinical data for the analyses were collected from the medical record and depended upon the accuracy and completeness of the clinical documentation. Although this study is confined to fee-for-service Medicare patients and hospital participation in OPTIMIZE-HF was voluntary and limited to all those hospitals participating in a quality improvement registry and this may limit the generalizability of the results. However, Medicare-linked OPTIMIZE-HF patients have been shown to have similar characteristics and outcomes as heart failure patients in the general Medicare population.48

CONCLUSIONS

In hospitalized older patients with heart failure and preserved ejection fraction who were not receiving angiotensin receptor blockers, a new discharge prescription for ACE inhibitors was associated with a modest improvement in the composite endpoint of total mortality or heart failure hospitalization, but had no association with the individual components of mortality and heart failure hospitalization. Findings from this rigorously conducted propensity-matched inception cohort study need to be interpreted in the context of inconclusive findings from the PEP-CHF trial and proven lack of efficacy of angiotensin receptor blockers in these patients. Additional well-designed prospective studies are needed.

Acknowledgments

Funding/Support: The project described was supported by the grant R01-HL097047 from NHLBI/NIH (PI: Ahmed, A). Its contents are solely the responsibility of the authors and do not necessarily represent the official views of the NHLBI or NIH. Dr. Ahmed is also supported by a generous gift from Ms. Jean B. Morris of Birmingham, Alabama. Dr. Allman is supported by NIH/NCRR grant 5UL1 RR025777. OPTIMIZE-HF was funded by GlaxoSmithKline (PI: Fonarow, GC).

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Conflict of Interest: None

Authorship: AA conceived the study hypothesis and design in collaboration with coauthors. AA and MM wrote the first draft. AA, MM, KP performed statistical analyses in collaboration with IA, TL, and YZ. All authors interpreted the data, participated in critical revision of the paper for important intellectual content, and approved the final version of the article. IA, AA, MM, KP and YZ had full access to the data.

References

  • 1.Roger VL, Go AS, Lloyd-Jones DM, et al. Heart disease and stroke statistics--2011 update: a report from the American Heart Association. Circulation. 2011;123:e18–e209. doi: 10.1161/CIR.0b013e3182009701. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.Ahmed A, Perry GJ, Fleg JL, Love TE, Goff DC, Jr., Kitzman DW. Outcomes in ambulatory chronic systolic and diastolic heart failure: a propensity score analysis. Am Heart J. 2006;152:956–966. doi: 10.1016/j.ahj.2006.06.020. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.Fonarow GC, Stough WG, Abraham WT, et al. 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;50:768–777. doi: 10.1016/j.jacc.2007.04.064. [DOI] [PubMed] [Google Scholar]
  • 4.The CONSENSUS Trial Study Group Effects of enalapril on mortality in severe congestive heart failure. Results of the Cooperative North Scandinavian Enalapril Survival Study (CONSENSUS). N Engl J Med. 1987;316:1429–1435. doi: 10.1056/NEJM198706043162301. [DOI] [PubMed] [Google Scholar]
  • 5.The SOLVD Investigators Effect of enalapril on survival in patients with reduced left ventricular ejection fractions and congestive heart failure. N Engl J Med. 1991;325:293–302. doi: 10.1056/NEJM199108013250501. [DOI] [PubMed] [Google Scholar]
  • 6.Garg R, Yusuf S. Overview of randomized trials of angiotensin-converting enzyme inhibitors on mortality and morbidity in patients with heart failure. Collaborative Group on ACE Inhibitor Trials. JAMA. 1995;273:1450–1456. [PubMed] [Google Scholar]
  • 7.Cohn JN, Tognoni G, Valsartan Heart Failure Trial I A randomized trial of the angiotensin-receptor blocker valsartan in chronic heart failure. N Engl J Med. 2001;345:1667–1675. doi: 10.1056/NEJMoa010713. [DOI] [PubMed] [Google Scholar]
  • 8.Granger CB, McMurray JJ, Yusuf S, et al. Effects of candesartan in patients with chronic heart failure and reduced left-ventricular systolic function intolerant to angiotensin-converting-enzyme inhibitors: the CHARM-Alternative trial. Lancet. 2003;362:772–776. doi: 10.1016/S0140-6736(03)14284-5. [DOI] [PubMed] [Google Scholar]
  • 9.Hunt SA, Abraham WT, Chin MH, et al. 2009 focused update incorporated into the ACC/AHA 2005 Guidelines for the Diagnosis and Management of Heart Failure in Adults: a report of the American College of Cardiology Foundation/American Heart Association Task Force on Practice Guidelines: developed in collaboration with the International Society for Heart and Lung Transplantation. Circulation. 2009;119:e391–479. doi: 10.1161/CIRCULATIONAHA.109.192065. [DOI] [PubMed] [Google Scholar]
  • 10.Kitzman DW, Little WC, Brubaker PH, et al. Pathophysiological characterization of isolated diastolic heart failure in comparison to systolic heart failure. JAMA. 2002;288:2144–2150. doi: 10.1001/jama.288.17.2144. [DOI] [PubMed] [Google Scholar]
  • 11.Massie BM, Carson PE, McMurray JJ, et al. Irbesartan in patients with heart failure and preserved ejection fraction. N Engl J Med. 2008;359:2456–2467. doi: 10.1056/NEJMoa0805450. [DOI] [PubMed] [Google Scholar]
  • 12.Yusuf S, Pfeffer MA, Swedberg K, et al. Effects of candesartan in patients with chronic heart failure and preserved left-ventricular ejection fraction: the CHARM-Preserved Trial. Lancet. 2003;362:777–781. doi: 10.1016/S0140-6736(03)14285-7. [DOI] [PubMed] [Google Scholar]
  • 13.Cleland JG, Tendera M, Adamus J, Freemantle N, Polonski L, Taylor J. The perindopril in elderly people with chronic heart failure (PEP-CHF) study. Eur Heart J. 2006;27:2338–2345. doi: 10.1093/eurheartj/ehl250. [DOI] [PubMed] [Google Scholar]
  • 14.Rosenbaum PR, Rubin DB. The central role of propensity score in observational studies for causal effects. Biometrika. 1983;70:41–55. [Google Scholar]
  • 15.Rosenbaum PR, Rubin DR. Constructing a control group using multivariate matched sampling methods that incorporate the propensity score. The American Statistician. 1985;39:33–38. [Google Scholar]
  • 16.Ray WA. Evaluating medication effects outside of clinical trials: new-user designs. Am J Epidemiol. 2003;158:915–920. doi: 10.1093/aje/kwg231. [DOI] [PubMed] [Google Scholar]
  • 17.Danaei G, Tavakkoli M, Hernan MA. Bias in observational studies of prevalent users: lessons for comparative effectiveness research from a meta-analysis of statins. Am J Epidemiol. 2012;175:250–262. doi: 10.1093/aje/kwr301. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Feenstra H, Grobbee RE, in't Veld BA, Stricker BH. Confounding by contraindication in a nationwide cohort study of risk for death in patients taking ibopamine. Ann Intern Med. 2001;134:569–572. doi: 10.7326/0003-4819-134-7-200104030-00010. [DOI] [PubMed] [Google Scholar]
  • 19.Psaty BM, Siscovick DS. Minimizing bias due to confounding by indication in comparative effectiveness research: the importance of restriction. JAMA. 2010;304:897–898. doi: 10.1001/jama.2010.1205. [DOI] [PubMed] [Google Scholar]
  • 20.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;148:43–51. doi: 10.1016/j.ahj.2004.03.004. [DOI] [PubMed] [Google Scholar]
  • 21.Gheorghiade M, Abraham WT, Albert NM, et al. Systolic blood pressure at admission, clinical characteristics, and outcomes in patients hospitalized with acute heart failure. JAMA. 2006;296:2217–2226. doi: 10.1001/jama.296.18.2217. [DOI] [PubMed] [Google Scholar]
  • 22.Fonarow GC, Abraham WT, Albert NM, et al. Association between performance measures and clinical outcomes for patients hospitalized with heart failure. JAMA. 2007;297:61–70. doi: 10.1001/jama.297.1.61. [DOI] [PubMed] [Google Scholar]
  • 23.Hammill BG, Hernandez AF, Peterson ED, Fonarow GC, Schulman KA, Curtis LH. Linking inpatient clinical registry data to Medicare claims data using indirect identifiers. Am Heart J. 2009;157:995–1000. doi: 10.1016/j.ahj.2009.04.002. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.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. 2011 doi: 10.1016/j.ijcard.2011.10.089. [Epub ahead of print]. doi.org/10.1016/j.ijcard.2011.10.089. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Patel K, Fonarow GC, Kitzman DW, et al. Angiotensin receptor blockers and outcomes in real-world older patients with heart failure and preserved ejection fraction: a propensity-matched inception cohort clinical effectiveness study. Eur J Heart Fail. 2012;14:1179–1188. doi: 10.1093/eurjhf/hfs101. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Hernan MA, Hernandez-Diaz S, Robins JM. A structural approach to selection bias. Epidemiology. 2004;15:615–625. doi: 10.1097/01.ede.0000135174.63482.43. [DOI] [PubMed] [Google Scholar]
  • 27.Rubin DB. Using propensity score to help design observational studies: Application to the tobacco litigation. Health Services and Outcomes Research Methodology. 2001;2:169–188. [Google Scholar]
  • 28.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;1:62–67. doi: 10.1161/CIRCOUTCOMES.108.790634. [DOI] [PubMed] [Google Scholar]
  • 29.Heinze G, Juni P. An overview of the objectives of and the approaches to propensity score analyses. Eur Heart J. 2011;32:1704–1708. doi: 10.1093/eurheartj/ehr031. [DOI] [PubMed] [Google Scholar]
  • 30.Michels KB, Braunwald E. Estimating treatment effects from observational data: dissonant and resonant notes from the SYMPHONY trials. JAMA. 2002;287:3130–3132. doi: 10.1001/jama.287.23.3130. [DOI] [PubMed] [Google Scholar]
  • 31.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;27:1431–1439. doi: 10.1093/eurheartj/ehi890. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Ahmed A, Fonarow GC, Zhang Y, et al. Renin-angiotensin inhibition in systolic heart failure and chronic kidney disease. Am J Med. 2012;125:399–410. doi: 10.1016/j.amjmed.2011.10.013. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.Ahmed A, Rich MW, Zile M, et al. Renin-angiotensin inhibition in diastolic heart failure and chronic kidney disease. Am J Med. 2013 Feb; doi: 10.1016/j.amjmed.2012.06.031. In Press. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.Filippatos GS, Ahmed MI, Gladden JD, et al. Hyperuricaemia, chronic kidney disease, and outcomes in heart failure: potential mechanistic insights from epidemiological data. Eur Heart J. 2011;32:712–720. doi: 10.1093/eurheartj/ehq473. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Guichard JL, Desai RV, Ahmed MI, et al. Isolated diastolic hypotension and incident heart failure in older adults. Hypertension. 2011;58:895–901. doi: 10.1161/HYPERTENSIONAHA.111.178178. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36.Normand ST, Landrum MB, Guadagnoli E, et al. Validating recommendations for coronary angiography following acute myocardial infarction in the elderly: a matched analysis using propensity scores. J Clin Epidemiol. 2001;54:387–398. doi: 10.1016/s0895-4356(00)00321-8. [DOI] [PubMed] [Google Scholar]
  • 37.Aronow WS, Ahmed MI, Ekundayo OJ, Allman RM, Ahmed A. A propensity-matched study of the association of peripheral arterial disease with cardiovascular outcomes in community-dwelling older adults. Am J Cardiol. 2009;103:130–135. doi: 10.1016/j.amjcard.2008.08.037. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38.Ekundayo OJ, Dell'Italia LJ, Sanders PW, et al. Association between hyperuricemia and incident heart failure among older adults: a propensity-matched study. Int J Cardiol. 2010;142:279–287. doi: 10.1016/j.ijcard.2009.01.010. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39.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;49:165–171. doi: 10.1016/j.archger.2008.06.009. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40.Rosenbaum PR. Sensitivity to hidden bias. In: Rosenbaum PR, editor. Observational Studies. Vol. 1. Springer-Verlag; New York: 2002. pp. 105–170. [Google Scholar]
  • 41.Altman DG, Andersen PK. Calculating the number needed to treat for trials where the outcome is time to an event. BMJ. 1999;319:1492–1495. doi: 10.1136/bmj.319.7223.1492. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 42.Nadir MA, Wei L, Elder DH, et al. Impact of renin-angiotensin system blockade therapy on outcome in aortic stenosis. J Am Coll Cardiol. 2011;58:570–576. doi: 10.1016/j.jacc.2011.01.063. [DOI] [PubMed] [Google Scholar]
  • 43.Zile MR, Gaasch WH, Anand IS, et al. 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;121:1393–1405. doi: 10.1161/CIRCULATIONAHA.109.909614. [DOI] [PubMed] [Google Scholar]
  • 44.Lund LH, Benson L, Dahlstrom U, Edner M. Association between use of renin-angiotensin system antagonists and mortality in patients with heart failure and preserved ejection fraction. JAMA. 2012;308:2108–2117. doi: 10.1001/jama.2012.14785. [DOI] [PubMed] [Google Scholar]
  • 45.Rosenbaum PR, Rubin DB. The bias due to incomplete matching. Biometrics. 1985;41:103–116. [PubMed] [Google Scholar]
  • 46.Butler J, Arbogast PG, Daugherty J, Jain MK, Ray WA, Griffin MR. Outpatient utilization of angiotensin-converting enzyme inhibitors among heart failure patients after hospital discharge. J Am Coll Cardiol. 2004;43:2036–2043. doi: 10.1016/j.jacc.2004.01.041. [DOI] [PubMed] [Google Scholar]
  • 47.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;150:341–353. doi: 10.1093/oxfordjournals.aje.a010013. [DOI] [PubMed] [Google Scholar]
  • 48.Curtis LH, Greiner MA, Hammill BG, et al. Representativeness of a national heart failure quality-of-care registry: comparison of OPTIMIZE-HF and non-OPTIMIZE-HF Medicare patients. Circ Cardiovasc Qual Outcomes. 2009;2:377–384. doi: 10.1161/CIRCOUTCOMES.108.822692. [DOI] [PMC free article] [PubMed] [Google Scholar]

RESOURCES