Angiotensin-Converting Enzyme Inhibitors and Angiotensin Receptor Blockers in Patients With Congestive Heart Failure and Chronic Kidney Disease

Alan K. Berger, MD; Sue Duval, PhD; Connie Manske , MD; Gabriela Vazquez, MS; Cheryl Barber, MPH; Leslie Miller, MD; Russell V. Luepker, MD, MS

Am Heart J.  2007;19(5):128-137.  ©2007 Mosby, Inc.
Posted 06/29/2007

Abstract and Introduction


Background: Patients with coexistent heart failure and chronic kidney disease (CKD) have a poor prognosis, possibly related to the underuse of standard medical therapies-angiotensin-converting enzyme inhibitors (ACE-I) and angiotensin receptor blockers (ARB).
Methods: We performed a retrospective analysis of the Minnesota Heart Survey, identifying patients hospitalized in 2000 in the Minneapolis-St Paul metropolitan area with heart failure. The main outcome measure was the association of ACE-I and ARB use on 30-day and 1-year mortality, stratified by glomerular filtration rate (GFR).
Results: Compared to patients with heart failure with preserved renal function (GFR ≥90 mL/min), patients with severely impaired renal function (GFR <15 mL/min) were far less likely to receive ACE-I or ARB during hospitalization (52.0% vs 69.5%, P < .0001) or at discharge (50.5% vs 65.1%, P < .0001). Worsening renal function was associated with increased mortality, both at 30 days and at 1 year. The inhospital use of either an ACE-I or ARB was associated with significantly reduced 30-day mortality (OR 0.45, 95% CI 0.28-0.59) after adjusting for multiple risk factors. Similarly, the discharge prescription of either an ACE-I or ARB was associated with a significant reduction in adjusted 1-year mortality (OR 0.72, 95% CI 0.58-0.91). However, among patients on dialysis, there was no benefit of ACE-I or ARB on either 30-day or 1-year mortality.
Conclusions: Angiotensin-converting enzyme inhibitors and ARB are underused in patients with heart failure with chronic kidney disease. Given the reduction in 30-day and 1-year mortality, these medications should be considered in most patients with heart failure, independent of underlying renal function. Among patients on hemodialysis, further investigation is warranted.


Congestive heart failure (CHF) is an increasingly prevalent condition in the United States and now affects 2.2% of the population.[1] There is strong evidence indicating that the use of angiotensin-converting enzyme inhibitors (ACE-I) among patients hospitalized with CHF results in decreased mortality.[2,3,4,5,6,7,8] There are also data confirming angiotensin receptor blockers (ARB) can provide a similar reduction in mortality compared to ACE-I.[9,10,11] The data from these trials form the basis for the American College of Cardiology/American Heart Association guidelines for the management of patients with chronic heart failure.[12,13] Contemporary data suggest the use of ACE-I and ARB remains suboptimal even in patients who have no contraindications to either therapy.[14]

Renal dysfunction is an independent predictor of morbidity and mortality in the setting of CHF. The role of ACE-I and ARB in this group of patients is less well established for 2 major reasons. First, randomized clinical trials have typically excluded patients with severely impaired renal dysfunction. Second, physicians have been reluctant to initiate either medication in patients with renal impairment because of the fear of precipitating acute renal failure or hyperkalemia. Consequently, there is a knowledge gap regarding the potential benefits and risks of ACE-I in patients with CHF and associated chronic kidney disease (CKD).

The population-based Minnesota Heart Survey (MHS) is uniquely suited to address the inhospital and discharge use of CHF medical therapies. We hypothesized (1) ACE-I and ARB are underused in patients with CHF and coexistent CKD; (2) the inhospital use of ACE-I and ARB in patients with CHF and severe renal insufficiency is associated with a significant reduction in 30-day mortality; and (3) the discharge prescription of ACE-I and ARB in patients with CHF and severe renal insufficiency is associated with a significant reduction in 1-year mortality.

Design and Methods

The MHS is an ongoing surveillance project of CHF in all hospitalized patients 35 to 84 years of age residing in the Minneapolis-St Paul metropolitan area. The patient sample is based on an International Classification of Diseases, Ninth Revision (ICD-9) discharge diagnosis of CHF. The random selection of patient records is determined by a computer, and the medical records are abstracted by trained nurses. The extensive abstraction process includes demographic details, cardiac risk factors, medical history, clinical presentation, relevant laboratory and electrocardiographic data, medications, and inhospital complications.[15]

We identified patients in the MHS database with CHF using the Framingham criteria.[16,17] Major criteria for CHF include paroxysmal nocturnal dyspnea, orthopnea, engorged jugular veins, an S3 gallop, hepatomegaly, cardiomegaly, pulmonary edema, and a weight gain ≥4.5 kg. Minor criteria include dyspnea on exertion, cough, swollen extremities, tachycardia, and pleural effusion. The Framingham definition for CHF requires either ≥2 major criteria or the combination of 1 major criterion and ≥2 minor criteria. Both patients presenting to the hospital with CHF and those subsequently diagnosed with CHF during their hospitalization were included in the analysis.

The United States Renal Data System (USRDS) maintains a database on all patients with end-stage renal disease. We obtained permission from the USRDS to merge the MHS and USRDS data to identify patients who were on dialysis before hospitalization. All personal identifiers were removed before data analysis to ensure Health Insurance Portability and Accountabilty Act (HIPAA) compliance.

Glomerular filtration rate (GFR), a surrogate of renal function, was estimated using the Cockcroft-Gault equation. Serum creatinine was recorded at the time closest to the diagnosis of CHF. For most patients, this specimen was obtained at the time of hospital admission. A stratification scheme consistent with National Kidney Foundation Guidelines categorized renal function into 1 of 5 categories: stage 1 (≥90 mL/min), stage 2 (60-89 mL/min), stage 3 (30-59 mL/min), stage 4 (15-29 mL/min), and stage 5 (<15 mL/min).[18] Patients on dialysis before hospitalization were incorporated into the stage 5 group. Trends in characteristics across renal function strata were assessed using the Cochran-Armitage test for dichotomous variables and general linear models for continuous variables. The prescription rates-both inhospital and at discharge-of ACE-I, ARB, β-blockers, diuretics, aldosterone blockers, and loop diuretics were calculated for patients in each renal function stratum. The MHS project was able to identify whether a patient was treated with either hydralazine or a nitrate but did not obtain further detail. Characteristics associated with the use of ACE-I or ARB were assessed by logistic regression, and the individual renal function categories were forced into the model.

Mortality was measured at several time points-inhospital, 30 days, 6 months, and 1 year. All mortality data were presented after adjusting for multiple covariates. General linear models, stratified by renal function, were used to measure the association of inhospital ACE-I/ARB use with 30-day mortality and the association of discharge ACE-I/ARB use with 1-year mortality. Patients who died within 48 hours of admission were excluded from the 30-day mortality analyses because their condition may have precluded the early oral administration of either medication. Separate models were constructed to analyze GFR as both a 5-level ordinal variable and a continuous parameter. Covariables entered into the risk adjustment model included age, sex, body mass index (BMI), cardiovascular risk factors, documentation of prior coronary artery disease (angina, myocardial infarction, documented coronary atherosclerosis), prior coronary revascularization (percutaneous transluminal coronary angioplasty [PTCA], coronary artery bypass graft [CABG]), peripheral arterial disease, stroke, dementia, atrial fibrillation, use of medications in hospital and at discharge (β-blocker, hydralazine/nitrates, aldosterone blockers, and diuretics), systolic blood pressure (BP) and pulse at the time of diagnosis, serum potassium, ejection fraction (EF), inotropic agents, intraaortic balloon pump, and coronary revascularization (PTCA, CABG) during the index hospitalization. Ejection fraction data were abstracted both qualitatively and quantitatively where available and were dichotomized using an EF of 35% as a cut point. The selection of variables was based on clinical experience and known predictors of mortality. An interaction term was entered into both the 30-day and 1-year models to assess whether the potential benefit of ACE-I/ARB differed across varying levels of renal function. The reference group was patients with preserved renal function (GFR >90 mL/min) who were not treated with an ACE-I or ARB. A propensity score was developed using the aforementioned covariates to account for factors influencing patient allocation to ACE-I or ARB. The estimates were nearly identical to the results obtained from the general linear models. Consequently, we elected to report the results from the general linear model.

We were interested in determining whether ACE-I and ARB were used more frequently among patients on hemodialysis because the target organ could not be further jeopardized. We were also interested in knowing whether these medications offered a differential benefit among patients on hemodialysis compared to those not on hemodialysis. We therefore performed a post hoc subgroup analyses on the limited number of patients with stage 5 renal dysfunction, stratifying them into 2 groups based on whether they were on dialysis before hospitalization. The prescription of ACE-I and ARB were compared between patients on hemodialysis and those not on hemodialysis. The association of inhospital ACE-I and ARB use with 30-day age- and sex-adjusted mortality and the association of discharge prescription of these medications with 1-year age- and sex-adjusted mortality were similarly assessed by tabular methods. All statistical analyses were performed using SAS (version 9.1; SAS Institute, Cary, NC).

The MHS is a National Institutes of Health-supported study, and this study was approved by the institutional review boards of the participating institutions. No other external source of funding was used for this investigation.

Main Outcome Measures

The outcome measure was all-cause mortality. Using the admission date as the reference, mortality was measured at discharge, 30 days, 6 months, and 1 year. Inhospital mortality was identified directly through medical record abstraction; postdischarge mortality was supplied by MINNDEX (Minnesota Death Index, Minneapolis, MN), a software program that analyzes mortality tapes provided by the state of Minnesota.[19]

Main Results

We identified 4573 patients with an ICD-9 discharge diagnosis of CHF who were hospitalized in the Minneapolis-St Paul metropolitan area hospitals between January 1 and December 31, 2000. Figure 1 describes the derivation of the study sample. We were able to validate the diagnosis of CHF using the Framingham criteria in 2317 patients. Both a weight and serum creatinine were available in 2169 (93.6%) patients, permitting the calculation of a serum GFR. This cohort-mean age, 69.1 ± 11.0 years-constituted the inhospital population in whom the use of medical therapies was assessed. The GFR was ≥90 mL/min in 469 (21.7%), 60 to 89 mL/min in 546 (25.1%), 30 to 59 mL/min in 773 (35.6%), 15 to 29 mL/min in 238 (11.0%), and <15 mL/min in 143 (6.6%). Both clinical characteristics and hospital therapies varied by degree of renal dysfunction ( Table 1 ). Trends in characteristics were consistent from stage 1 to stage 4 kidney disease. However, patients with stage 5 kidney disease included numerous individuals on dialysis, and their characteristics were quite distinct from those with less severe renal insufficiency. Overall, patients with worsening renal function tended to be older and were more likely to be female. The prevalence of cardiac risk factors-hypertension and diabetes-increased with declining renal function, whereas the prevalence of smoking decreased. Atherosclerosis (coronary artery disease, peripheral arterial disease, and cerebrovascular disease) was also more prevalent among patients with poor renal function. The left ventricular EF (LVEF) was reduced among patients with worse renal function as well. Patients with worse renal function were less likely to undergo PTCA during their hospitalization.

Figure 1. 

Derivation of study sample from MHS cohort of CHF patients. International Classification of Diseases, Ninth Revision discharge diagnosis codes were used to identify patients with CHF from the MHS. The diagnosis of CHF was validated using the Framingham criteria. Inhospital and 30-day mortality were measured using the entire cohort. The 6-month and 1-year mortality were based on the subgroup of patients who survived the index hospitalization.


The inhospital use of ACE-I declined with worsening renal function with a steep cutoff among patients who had a GFR <30 mL/min (Figure 2A). Angiotensin-converting enzyme inhibitors were administered during hospitalization to 319 patients (68.0%) with a GFR >90 mL/min and to 157 patients (42.3%) with a GFR <30 mL/min. The increased use of ARB (11.3%) in patients with worse renal function partially offset the lower use of ACE-I in this group. The use of hydralazine and/or nitrates steadily increased as renal function worsened. Renal function did not appear to have any impact on the inhospital prescription of β-blockers ( Table 1 ).

Figure 2. 

A, Prescription of standard medical therapies during index hospitalization, age- and sex-adjusted, and stratified by renal function. This analysis includes all patients admitted to hospital with a validated diagnosis of CHF in whom GFR could be calculated. B, Prescription of standard medical therapies at hospital discharge, age- and sex-adjusted, and stratified by renal function. This analysis is restricted to patients with a validated diagnosis of CHF who survived the index hospitalization and in whom GFR could be calculated.


The discharge prescription of CHF therapies paralleled the inhospital use of these medications (Figure 2B). The discharge prescription rates were slightly reduced compared to the inhospital rates, presumably due to the development of either allergic reactions or adverse side effects during hospitalization. The use of ACE-I declined with worsening renal function, whereas the use of ARB increased with worsening renal function. The combined use of ACE-I or ARB declined as well, although the greatest difference was among patients with a GFR <30 mL/min. There was no significant difference in the discharge prescription of β-blockers among patients with different renal function (data not shown).

Several characteristics were significantly associated with the prescription of ACE-I or ARB during hospitalization ( Table 2 ). Male sex, as well as the presence of hypertension, diabetes, and hypercholesterolemia, were all strongly associated with the prescription of either of these medications. In addition, coronary artery disease, LVEF <35% and prior coronary revascularization were significantly associated with the inhospital prescription of these medications. The mean serum creatinine was significantly less among patients prescribed these therapies compared to those who were not prescribed these agents. Patients prescribed ACE-I or ARB during their hospitalization were more likely to receive other standard cardiovascular therapies including β-blockers, aldosterone blockers, and diuretics. Coronary revascularization-both PTCA and CABG-were also more likely to be performed among patients receiving ACE-I or ARB.

Glomerular filtration rate was strongly associated with mortality, whether measured during hospitalization, at 30 days, at 6 months, or at 1 year (Figure 3). The fully adjusted 30-day mortality was lower among patients who received ACE-I or ARB compared to those patients who received neither therapy ( Table 3 ). There was no interaction between severity of renal dysfunction and use of ACE-I or ARB. Among the 2286 patients (98.7%) who survived the initial 48 hours of hospitalization and in whom receipt of standard medical therapies may have arguably had a greater impact, the results were essentially unchanged.

Figure 3. 

Mortality for CHF patients stratified by GFR during index hospitalization. Crude mortality rates-inhospital, 30-day, 6-month, and 1-year are referenced from admission date of index hospitalization. The mortality data are fully adjusted for the covariates in the model.


We performed a post hoc subgroup analysis on patients (n = 143) with severe renal dysfunction (GFR <15 mL/min) and found 100 (69.9%) of these patients were on hemodialysis before hospitalization. A prior history of dialysis did not have a significant influence on the use of ACE-I or ARB, either inhospital (43.5% on dialysis vs 53.3% not on dialysis, P = nonsignificant) or at discharge (39.4% on dialysis vs 53.5% not on dialysis, P = nonsignificant). The P value interaction between inhospital use of ACE-I or ARB and dialysis on 30-day mortality was .10, and the P value for the interaction between discharge use of ACE-I or ARB and dialysis was .22. Among patients on dialysis, the use of either ACE-I or ARB was not associated with any significant reduction in mortality, either at 30 days or 1 year (Figure 4). In contrast, these medications appeared to be associated with a mortality reduction among patients not on dialysis. At 30 days, there was a 34% relative risk reduction in mortality (12.3% vs 18.5%, P = .05). At 1 year, there was a 73% relative reduction in mortality (11.1% vs 41.0%, P = .05).

Figure 4. 

Association of inhospital use of ACE-I and/or ARB with 30-day and 1-year adjusted mortality. The association of inhospital ACE-I and/or ARB prescription with 30-day and 1-year adjusted mortality is stratified by prior exposure to hemodialysis.



We confirmed renal dysfunction as a strong predictor of both short- and intermediate-term mortality among patients hospitalized with CHF. We documented a benefit of ACE-I and ARB on 30-day and 1-year mortality among patients with CHF, independent of the degree of severity of renal dysfunction.

In spite of the potential benefit of these medications, we found an inverse correlation between the degree of renal dysfunction and the use of both ACE-I and ARB. Interestingly, ACE-I and ARB did not appear to have an impact on mortality among dialysis patients.

The clinical significance of our findings stems from the high prevalence of renal dysfunction among patients with CHF. McAlister et al[20] recently documented a GFR ≤30 mL/min in 16% of CHF patients and a GFR 30 to 59 mL/min in 40% of patients. The association of renal dysfunction and increased cardiovascular morbidity and mortality is well documented in the literature.[16,17,21,22,23] The existence of both conditions frequently reflects atherosclerotic burden, and so, it should not be surprising that patients with chronic renal sufficiency suffer increased mortality. In the cardiac health study, renal dysfunction was an independent predictor for the development of cardiovascular disease, CHF, and cardiovascular mortality.[23] Hillege et al,[24] in a recent analysis of the CHARM trials, found impaired renal function to be a strong predictor of death, cardiovascular mortality, and rehospitalization for CHF.

Although multiple studies have addressed the poor prognosis of CHF patients with renal dysfunction, few investigators have addressed the impact of medical therapies among these patients. Randomized clinical trials of both ACE-I have tended to exclude patients with impaired renal function, as defined by a serum Cr >2.5 mg/dL. In a recent publication, McAlister et al[20] found ACE-I had a similar benefit on mortality at 2.5 years whether the patients had a GFR ≥60 mL/min (OR 0.28 [95% CI 0.11-0.70]) or a GFR <60 mL/min (OR 0.46 [95% CI 0.26-0.82]).

The use of ACE-I among patients with severe renal insufficiency is clearly not an established practice. In our study, approximately 50% of patients with a GFR <30 mL/min received either an ACE-I or ARB. Bakris and Weir[25] recently reviewed 12 randomized clinical trials of ACE-I in patients with baseline renal insufficiency (serum Cr >1.4 mg/dL). They found a strong association between acute increases in serum creatinine that stabilized within the first 2 months of ACE-I therapy and long-term preservation of renal function. The authors concluded ACE-I should only be withheld when the rise in creatinine exceeds 30% above baseline within the first 2 months of ACE-I initiation or hyperkalemia (serum potassium ≥5.6 mmol/L) develops. Philbin et al[26] studied 1076 hospital survivors identified from a consecutive series of CHF inpatients and found angiotensin-converting enzyme inhibitors were associated with a clinical benefit among patients with a serum Cr ≤1.9 mg/dL. Among patients with a serum Cr ≥2.0 mg/dL, there was no impact on mortality, rehospitalization, or quality of life. In a recently published randomized clinical trial of 224 nondiabetic patients with advanced renal insufficiency (serum Cr 3.1-5.0 mg/dL), Hou et al[27] found benazepril reduced both the level or proteinuria and the rate of decline in renal function.

The absence of a significant mortality benefit of ACE-I and ARB among dialysis patients is an interesting observation and, to the best of our knowledge, has not been previously reported. McCullough et al[28] found a benefit of ACE-I among patients with end-stage renal disease who were hospitalized in the coronary care unit with either an acute coronary syndrome or CHF. After adjusting for confounders, there was a 37% reduction in all-cause mortality among patients who received ACE-I (P = .014). In contrast, we found no benefit of ACE-I or ARB among patients with a GFR <15 mL/min who were on dialysis. Interestingly, we observed a 34% relative reduction in 30-day mortality and a 73% reduction in 1-year mortality among patients with a GFR <15 mL/min who were not on dialysis. The benefit of these agents among stage 5 nondialysis patients must be interpreted with caution due to the small sample size.

Congestive heart failure is a prevalent finding among dialysis patients and accounts for significant morbidity and mortality in this population.[29,30,31] The pathologic characteristics of dilated cardiomyopathy in dialysis patients-interstitial fibrosis and severe myocyte hypertrophy-are postulated to contribute to the dismal clinical prognosis.[32] These findings may explain why ACE-I and ARB may not achieve the same outcomes in patients with dialysis. Alternatively, the failure of these medications to provide a benefit may stem from the fact that the kidney is both intrinsically involved in the metabolism of the drug and serves as the target organ for these agents. Finally, our inability to identify a benefit may stem from selection bias or the relatively small number of patients with CHF and coexistent hemodialysis.

It should not be surprising that the CHF guidelines have been unable to achieve a consensus regarding the use of ACE-I/ARB in patients with severe renal dysfunction.[12,33,34] Most of the randomized clinical trials used serum Cr rather than estimated GFR and excluded patients with impaired renal function (serum Cr >2.0 mg/dL). In the studies that did permit a serum Cr as high as 3.0 mg/dL, few patients with severe renal dysfunction were actually enrolled. There is evidence from the CONSENSUS trial to support the use of ACE-I in patients with moderate renal insufficiency (GFR 30-60 mL/min). In the recently published American College of Cardiology/American Heart Association guidelines, GFR is not used to stratify risk.[12] The guidelines indicate ACE-I should be used with caution among patients with markedly increased levels of serum creatinine (>3 mg/dL).

Recent data from the ADHERE registry indicates ACE-I and ARB are underused among patients with CHF and coexistent renal dysfunction. The collaborators identified renal insufficiency (Cr ≥2.0 mg/dL) among 11798 (20.0%) of 58919 admissions of acute decompensated heart failure. Thirty-five percent of these patients were on an ACE-I before admission and 14% were on an ARB. An additional 9% of patients had an ACE-I added to their regimen at discharge, and 3% of patients had an ARB added to their regimen. These data emphasize the importance of reevaluating patients with CHF to determine whether they are appropriate candidates for ACE-I and/or an ARB.

Several factors presumably explain the reduced use of ACE-I among CHF patients with pre-existing CKD. First, ACE-I have the potential to precipitate hyperkalemia and increase the toxicity of other agents (ie, digoxin) in the setting of CKD.[26,33] Second, a significant increase in serum creatinine (>0.3 mg/dL) with the use of ACE-I is observed in 15% to 30% of CHF patients.[34] This is of greatest concern among patients with renal artery stenosis in whom the institution of an ACE-I may reduce GFR by reducing the efferent arteriolar pressure.[35] Finally, patients with CHF secondary to systolic dysfunction frequently have reduced systolic BP, and physicians often struggle with balancing the benefits of multiagent therapy against the adverse effects associated with systemic hypertension.

There were several limitations to our analysis. First, this was a retrospective study and was subject to both diagnosis misclassification as well as selection bias. We chose the Framingham definition of CHF to validate the ICD-9 diagnosis of CHF. To address the issue of selection bias, we developed risk adjustment models and included covariates associated with CHF and worse prognosis. Second, the MHS project is cross-sectional, sampling each patient at the time of hospitalization for CHF, and does not collect information post discharge other than death because of privacy regulations. Consequently, it is difficult to estimate patient compliance with long-term ACE-I and/or ARB use as well as which patients with CKD may have gone on to require hemodialysis post discharge. Third, as with any retrospective analysis, there may have been unmeasurable variables (eg, microalbuminuria) and unknown confounders that modified the association between the use of ACE-I/ARB and mortality. The incorporation of a propensity model had minimal impact on the estimates. Finally, there may have been a small number of Minnesota residents who moved during the year after hospital discharge for whom we could not identify long-term mortality.


Chronic kidney disease is highly prevalent among CHF patients, and the severity of CKD is a strong predictor of short- and intermediate-term mortality. Although ACE-I and ARB are not widely used in this population, our data suggests their administration may be associated with an improved survival, both at 30 days and 1 year. The initiation of these agents at low dose with careful monitoring of renal function and serum electrolytes should be considered in all patients with CHF, independent of renal function. The use of these agents in patients on hemodialysis clearly warrants further investigation.

Table 1. Comparison of Patients With CHF, Stratified by Renal Function

  Stage 1(≥90 mL/min) Stage 2 (60-89 mL/min) Stage 3 (30-59 mL/min) Stage 4 (15-30 mL/min) Stage 5 (<15 mL/min) P
N 469 546 773 238 143 
Age (mean ± SD) 59.8 ± 11.2 69.5 ± 9.3 73.7 ± 8.1 73.8 ± 8.8 68.0 ± 12.7 <.0001
Male (%) 63.5 55.7 48.1 42.0 49.0 <.0001
Cardiac risk factors
Hypertension (%) 66.1 68.7 71.3 81.1 84.6 <.0001
Diabetes (%) 39.0 36.5 35.1 39.9 48.3 .25
Hypercholesterolemia (%) 39.2 41.2 39.5 34.0 35.7 .17
Tobacco use (%) 31.8 20.9 14.2 11.8 18.2 <.0001
Prior conditions
Coronary artery disease (%) 44.1 55.9 57.6 64.3 60.8 <.0001
Prior PTCA (%) 13.7 14.5 15.5 13.0 14.7 .75
Prior CABG (%) 13.2 21.3 25.1 28.2 18.9 <.0001
Cerebrovascular disease (%) 12.4 18.9 23.4 26.1 27.3 <.0001
Peripheral arterial disease (%) 11.3 15.8 18.1 21.0 39.9 <.0001
Atrial fibrillation (%) 22.2 22.9 33.5 35.3 29.4 .0001
Dementia (%) 6.2 8.2 8.3 10.1 9.8 .06
Clinical presentation
BMI (kg/m2) 34.3 ± 9.0 28.8 ± 6.1 26.8 ± 5.7 25.9 ± 5.7 25.7 ± 5.4 <.0001
Systolic BP (mm Hg) 143 ± 30 140 ± 31 140 ± 35 134 ± 37 146 ± 40 .85
Pulse (beat/min) 97 ± 25 94 ± 25 93 ± 27 89 ± 23 89 ± 24 <.0001
Serum potassium (mEq/L) 4.0 ± 0.6 4.1 ± 0.6 4.3 ± 0.7 4.5 ± 0.8 4.8 ± 0.9 <.0001
K+ >4.5 (mEq/L) (%) 14.1 17.6 30.1 45.4 53.9 <.0001
Serum creatinine (mg/dL) 132 ± 43.9 74 ± 8.6 45 ± 8.4 23 ± 4.3 14 ± 12.3 <.0001
LVEF <35% (%) 33.9 35.9 42.6 45.4 34.3 .013
Inhospital therapies
β-Blockers (%) 51.2 49.6 54.3 46.6 49.0 .77
Aldosterone blockers (%) 22.0 18.9 21.6 14.3 5.6 .0003
Hyralazine/nitrates (%) 14.7 20.4 28.7 40.3 35.0 <.0001
Diuretics (%) 97.0 97.6 97.8 97.9 58.0 <.0001
Inotropic agents (%) 9.4 10.8 16.8 20.6 16.1 <.0001
Intraaortic balloon pump (%) 3.2 1.7 1.6 0.8 1.4 .03
Coronary angioplasty (%) 6.4 5.9 3.4 2.1 2.1 .0004
Coronary bypass surgery (%) 3.6 3.7 4.0 1.7 1.4 .17


Table 2. Characteristics Associated With the Inhospital Administration of ACE-I or ARB

  ACE-I or ARB administered in hospital ACE-I or ARB not administered in hospital P
N 1451 718 
Age (mean ± SD) 68.9 ± 11.2 69.7 ± 10.8 .14
Male (%) 55.0 48.3 .004
Cardiac risk factors
Hypertension (%) 74.6 65.2 <.0001
Diabetes (%) 40.0 33.4 .004
Hypercholesterolemia (%) 42.1 32.7 <.0001
Tobacco use (%) 20.0 18.9 .54
Prior conditions
Coronary artery disease (%) 59.2 47.1 <.0001
Prior PTCA (%) 16.3 10.9 .0007
Prior CABG (%) 24.1 16.0 <.0001
Cerebrovascular disease (%) 20.9 19.5 .45
Peripheral arterial disease (%) 17.1 19.2 .22
Atrial fibrillation (%) 27.4 30.2 .16
Dementia (%) 7.0 10.4 .005
Clinical presentation
BMI (kg/m2) 29.7 ± 7.0 29.3 ± 7.9 .20
Systolic BP (mm Hg) 144 ± 34 135 ± 32 <.0001
Pulse (beat/min) 93 ± 25 94 ± 27 .78
Serum potassium (mEq/L) 4.2 ± 0.7 4.2 ± 0.8 .77
K+ >4.5 (mEq/L) (%) 26.1 28.1 .30
Serum creatinine (mg/dL) 68.9 ± 42.1 62.0 ± 47.3 .0006
LVEF <35% (%) 47.0 22.1 <.0001
Inhospital therapies
β-Blockers (%) 56.0 41.6 <.0001
Aldosterone blockers (%) 23.9 9.4 <.0001
Hyralazine/nitrates (%) 26.5 22.8 .07
Diuretics (%) 96.5 91.9 <.0001
Inotropic agents (%) 13.9 14.3 .79
Intraaortic balloon pump (%) 1.9 1.7 .67
Coronary angioplasty (%) 5.5 2.2 .0005
Coronary bypass surgery (%) 3.9 2.2 .03


Table 3. Thirty-Day Adjusted Mortality for CHF Patients, Stratified by Inhospital Prescription of ACE-I or ARB

Renal function (GFR) Stage 1 (≥90 mL/min) Stage 2 (60-89 mL/min) Stage 3 (30-59 mL/min) Stage 4 (15-29 mL/min) Stage 5 (<15 mL/min)
All CHF hospitalizations?
n 469 546 773 238 143
ACE-I or ARB 6.1% 6.3% 5.4% 9.4% 11.9%
No ACE-I or ARB 11.3% 8.6% 14.0% 18.5% 22.8%
P .07 .37 .0001 .008 .03
Patients surviving initial 2 d
n 469 541 765 232 139
ACE-I or ARB 5.8% 5.8% 5.1% 8.4% 13.0%
No ACE-I or ARB 11.1% 6.0% 11.8% 15.8% 20.6%
P .05 .91 .001 .03 .13

?The analysis was performed for all patients admitted with CHF and the subgroup of patients who survived the initial 2 days of hospitalization. The P value for interaction between trend in use of medication and severity of renal disease was .32 for the entire cohort and .33 for patients surviving the initial 2 days of hospitalization.



  1. American Heart Association . Heart disease and stroke statistics-2005 update. Dallas: American Heart Association; 2005;.
  2. The SOLVD Investigattors . Effect of enalapril on mortality and the development of heart failure in asymptomatic patients with reduced left ventricular ejection fractions. N Engl J Med. 1992;327:685-691.
  3. 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.
  4. AIRE Study Investigators . Effect of ramipril on mortality and morbidity of survivors of acute myocardial infarction with clinical evidence of heart failure. The Acute Infarction Ramipril Efficacy (AIRE) Study Investigators. Lancet. 1993;342:821-828.
  5. Pfeffer MA, Braunwald E, Moye LA, et al.. Effect of captopril on mortality and morbidity in patients with left ventricular dysfunction after myocardial infarction. Results of the survival and ventricular enlargement trial. The SAVE Investigators. N Engl J Med. 1992;327:669-677.
  6. Rector TS, Johnson G, Dunkman WB, et al.. Evaluation by patients with heart failure of the effects of enalapril compared with hydralazine plus isosorbide dinitrate on quality of life. V-HeFT II. The V-HeFT VA Cooperative Studies Group. Circulation. 1993;87(6 Suppl):VI71-VI77.
  7. 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.
  8. 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.
  9. Pitt B, Segal R, Martinez FA, et al.. Randomised trial of losartan versus captopril in patients over 65 with heart failure (Evaluation of Losartan in the Elderly Study, ELITE). Lancet. 1997;349:747-752.
  10. Pitt B, Poole-Wilson PA, Segal R, et al.. Effect of losartan compared with captopril on mortality in patients with symptomatic heart failure: randomised trial-the Losartan Heart Failure Survival Study ELITE II. Lancet. 2000;355:1582-1587.
  11. Pfeffer MA, Swedberg K, Granger CB, et al.. Effects of candesartan on mortality and morbidity in patients with chronic heart failure: the CHARM-Overall programme. Lancet. 2003;362:759-766.
  12. Hunt SA, Abraham WT, Chin MH, et al.. ACC/AHA 2005 Guideline Update for the Diagnosis and Management of Chronic Heart Failure in the Adult: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Writing Committee to Update the 2001 Guidelines for the Evaluation and Management of Heart Failure): developed in collaboration with the American College of Chest Physicians and the International Society for Heart and Lung Transplantation: endorsed by the Heart Rhythm Society. Circulation. 2005;112:e154-e235.
  13. Bonow RO, Bennett S, Casey DE, et al.. ACC/AHA Clinical Performance Measures for Adults with Chronic Heart Failure: a report of the American College of Cardiology/American Heart Association Task Force on Performance Measures (writing committee to develop heart failure clinical performance measures): endorsed by the Heart Failure Society of America. Circulation. 2005;112:1853-1887.
  14. ADHERE Registry Collaborators. In: ADHERE Registry. 2nd Quarter 2003 Benchmark Report. Subset analysis: treatment of acute heart failure patients with renal insufficiency (Cr >2.0 mg/dL). Scios, Inc.; 2005;
  15. Shahar E, Lee S, Kim J, et al.. Hospitalized heart failure: rates and long-term mortality. J Card Fail. 2004;10:374-379.
  16. Abbott R, McGee D. The Framingham Study: an epidemiological investigation of cardiovascular disease, section 37. Bethesda (MD): National Heart, Lung, and Blood Institute; 1987;.
  17. Kannel WB, D'Agostino RB, Silbershatz H, et al.. Profile for estimating risk of heart failure. Arch Intern Med. 1999;159:1197-1204.
  18. K/DOQI Clinical Practice Guidelines on Chronic Kidney Disease Work Group and Evidence Review Team Membership . K/DOQI clinical practice guidelines for chronic kidney disease: evaluation, classification, and stratification. Am J Kidney Dis. 2002;39(2 Suppl 1):S1-S266.
  19. Sprafka J, Virnig B, Shahar E, et al.. Trends in diabetes prevalence among stroke patients and the effect of diabetes on stroke survival: the Minnesota Heart Survey. Diabet Med. 1994;11:678-684.
  20. McAlister F, Ezekowitz J, Tonelli M, et al.. Renal insufficiency and heart failure. Prognostic and therapeutic implications from a prospective cohort study. Circulation. 2004;109:1004-1009.
  21. Dries D, Exner D, Domanski M, et al.. The prognostic implications of renal insufficiency in asymptomatic and symptomatic patients with left ventricular systolic dysfunction. J Am Coll Cardiol. 2000;35:681-689.
  22. Shlipak MG. Pharmacotherapy for heart failure in patients with renal insufficiency. Ann Intern Med. 2003;138:917-924.
  23. Fried LF, Shlipak MG, Crump C, et al.. Renal insufficiency as a predictor of cardiovascular outcomes and mortality in elderly individuals. J Am Coll Cardiol. 2003;41:1364-1372.
  24. Hillege HL, Nitsch D, Pfeffer MA, et al.. Renal function as a predictor of outcome in a broad spectrum of patients with heart failure. Circulation. 2006;113:671-678.
  25. Bakris G, Weir M. Angiotensin-converting enzyme inhibitor-associated elevations in serum creatinine: is this a cause for concern?. Arch Intern Med. 2003;160:685-693.
  26. Philbin EF, Santella RN, Rocco TA. Angiotensin-converting enzyme inhibitor use in older patients with heart failure and renal dysfunction. J Am Geriatr Soc. 1999;47:302-308.
  27. Hou FF, Zhang X, Zhang GH, et al.. Efficacy and safety of benazepril for advanced chronic renal insufficiency. N Engl J Med. 2006;354:131-140.
  28. McCullough PA, Sandberg KR, Yee J, et al.. Mortality benefit of angiotensin-converting enzyme inhibitors after cardiac events in patients with end-stage renal disease. J Renin Angiotensin Aldosterone Syst. 2002;3:188-191.
  29. Trespalacios FC, Taylor AJ, Agodoa LY, et al.. Heart failure as a cause for hospitalization in chronic dialysis patients. Am J Kidney Dis. 2003;41:1267-1277.
  30. Harnett JD, Foley RN, Kent GM, et al.. Congestive heart failure in dialysis patients: prevalence, incidence, prognosis and risk factors. Kidney Int. 1995;47:884-890.
  31. Stack AG, Bloembergen WE. A cross-sectional study of the prevalence and clinical correlates of congestive heart failure among incident US dialysis patients. Am J Kidney Dis. 2001;38:992-1000.
  32. Aoki J, Ikari Y, Nakajima H, et al.. Clinical and pathologic characteristics of dilated cardiomyopathy in hemodialysis patients. Kidney Int. 2005;67:333-340.
  33. Steiner JF, Robbins LJ, Hammermeister KE, et al.. Incidence of digoxin toxicity in outpatients. West J Med. 1994;161:474-478.
  34. Giles TD, Katz R, Sullivan JM, et al.. Short- and long-acting angiotensin-converting enzyme inhibitors: a randomized trial of lisinopril versus captopril in the treatment of congestive heart failure. The Multicenter Lisinopril-Captopril Congestive Heart Failure Study Group. J Am Coll Cardiol. 1989;13:1240-1247.
  35. Burnier M, Waeber B, Nussberger J, et al.. Effect of angiotensin converting enzyme inhibition in renovascular hypertension. J Hypertens Suppl. 1989;7:S27-S31.
Funding Information

This study received funding/support from the National Institutes of Health and the National Heart, Lung and Blood Institute (RO1-HL60959), Bethesda, MD.

Reprint Address

Alan Keith Berger, MD, Division of Epidemiology and Community Health, University of Minnesota, 1300 South Second Street, Suite 300, Minneapolis, MN 55454.

Alan K. Berger, MD,ab Sue Duval, PhD,b Connie Manske , MDc, Gabriela Vazquez, MSb, Cheryl Barber, MPHb, Leslie Miller, MDa, Russell V. Luepker, MD, MSab

a Section of Cardiovascular Medicine, Department of Medicine, University of MN, Minneapolis, MN

b Division of Epidemiology and Community Health, University of MN, Minneapolis, MN

c Section of Nephrology, Department of Medicine, University of MN, Minneapolis, MN