P < .001 based on Cochran-Mantel-Haenszel row mean scores statistics to test for trend (association between aldosterone antagonist use and calendar years between first half of 2005 and second half of 2007, after controlling for site). F1 = 12.08. Error bars indicate 95% confidence intervals.
Customize your JAMA Network experience by selecting one or more topics from the list below.
Albert NM, Yancy CW, Liang L, et al. Use of Aldosterone Antagonists in Heart Failure. JAMA. 2009;302(15):1658–1665. doi:10.1001/jama.2009.1493
Context Aldosterone antagonists are recommended for patients with moderate to severe heart failure (HF) and systolic dysfunction. Prior studies suggest underuse of aldosterone antagonists in eligible patients as well as overuse in settings in which therapy may be harmful.
Objective To examine aldosterone antagonist prescription based on HF management guideline criteria, investigator-defined appropriateness criteria, and trends over time in patients hospitalized with heart failure.
Design, Setting, and Patients Observational analysis of 43 625 patients admitted with HF and discharged home from 241 hospitals participating in the Get With The Guidelines–HF quality improvement registry between January 2005 and December 2007.
Main Outcome Measures Prescription and predictors of use of aldosterone antagonists, based on guideline criteria.
Results Among 12 565 patients eligible for aldosterone antagonist therapy, 4087 (32.5%) received an aldosterone antagonist at discharge, and treatment increased modestly from 28% to 34% over the study period. There was also wide variation in aldosterone antagonist use among hospitals (0%-90.6%). Aldosterone antagonist use in eligible patients was associated with younger age (adjusted odds ratio [OR], 0.85; 95% confidence interval [CI], 0.82-0.88), African American race/ethnicity (adjusted OR, 1.17; 95% CI, 1.04-1.32), lower systolic blood pressure (adjusted OR, 0.94; 95% CI, 0.92-0.95), history of implantable cardioverter-defibrillator use (adjusted OR, 1.51; 95% CI, 1.34-1.69), depression (adjusted OR, 1.15; 95% CI, 1.01-1.30), alcohol use (adjusted OR, 1.23; 95% CI, 1.02-1.50), and pacemaker implantation (adjusted OR, 1.21; 95% CI, 1.06-1.38), and with having no history of renal insufficiency (adjusted OR, 0.85; 95% CI, 0.75-0.96). Applying serum creatinine and potassium appropriateness criteria, inappropriate and potentially inappropriate use of aldosterone antagonist therapy was low and did not change over the 3-year study period.
Conclusions Less than one-third of eligible patients hospitalized for HF and participating in a quality improvement registry received HF guideline–recommended aldosterone antagonist therapy. Use of aldosterone antagonist therapy among patients with documented contraindications was low.
Clinical trials have established the incremental benefits of aldosterone antagonist therapy in patients with heart failure (HF) and reduced left ventricular ejection fraction (LVEF),1-3 such that aldosterone antagonists were designated as class I, “useful and recommended,” within the American College of Cardiology/American Heart Association (ACC/AHA) Chronic HF Guidelines.1 Because aldosterone antagonists can cause hyperkalemia, guidelines promote safety by specifying that prior to use, serum creatinine levels should be 2.5 mg/dL or less in men and 2.0 mg/dL or less in women, and serum potassium levels should be less than 5.0 mEq/L.1 Adoption of aldosterone antagonists in HF has been mixed; early community studies in Canada and the United States found modest uptake but also increased use in individuals with contraindications and increased admissions for hyperkalemia.4,5
The Get With The Guidelines–HF (GWTG-HF) program is a national quality improvement program designed to promote adherence to guideline-based recommendations. It is unknown whether participation in a hospital-based quality improvement program may lead to greater frequency of use of aldosterone antagonist therapy for appropriate indications as well as lower use in situations of increased risk. Therefore, the purpose of this study was to examine contemporary aldosterone antagonist use among hospitalized patients with HF, temporal trends in use, and appropriateness of use.
GWTG-HF is an ongoing, prospective registry and quality improvement program initiated in January 2005 that has been previously described.6-8 The registry enrolls adults hospitalized with an episode of new or worsening HF as the primary reason for admission or with significant HF symptoms that developed during hospitalization for which HF was the primary discharge diagnosis. Participating institutions submit data on consecutive eligible patients in compliance with Joint Commission and Centers for Medicare and Medicaid standards.
Hospital personnel are trained, and standardized data definitions are used. Data from medical records are collected on patient demographics, medical history, laboratory values, pharmacologic and nonpharmacologic interventions, HF performance measures, and outcomes at discharge. Admissions staff, medical staff, or both record self-reported race/ethnicity, usually as the patient is registered. Patients are assigned to race/ethnicity categories using options defined by the electronic case report form. Prior studies have suggested differences in use of HF therapy by race/ethnicity.8
Data are submitted and assessed for completeness and quality using an Internet-based patient management tool (Get With The Guidelines Heart Failure Patient Management Tool; Outcome Sciences Inc, Cambridge, Massachusetts). The Internet-based system performs checks to ensure that the reported data are complete. In addition, data quality is monitored for completeness and accuracy. Only fully participating hospital sites and variables with a high degree of completeness (>95%) are used in analyses.
Participating institutions submit the GWTG-HF protocol for review and approval by their institutional review boards. Data are used at local sites for quality improvement and are granted a waiver of informed consent under the common rule. Outcome Sciences serves as the data coordinating center. The Duke Clinical Research Institute (Durham, North Carolina) serves as the data analysis center and has institutional review board approval for analyzing the aggregate deidentified data.
From January 2005 through December 2007, 66 218 patients admitted with HF were discharged from 242 hospitals participating in GWTG-HF and routinely collecting detailed clinical data. Of these, we excluded 1645 patients without a history of HF and 4081 with contraindications to aldosterone antagonist therapy (documented contraindications of allergy, hyperkalemia, renal dysfunction, or other [n = 2384] and history of chronic dialysis [n = 1697]). We also excluded 16 867 who died (n = 1965); left against medical advice (n = 413); were discharged to a skilled nursing facility (n = 10 462), rehabilitation center (n = 1047), hospice center (n = 1195), or another hospital (transfer, n = 1620); or who had missing information on discharge destination (n = 165). The final study population thus included 43 625 patients with HF treated at 241 hospitals who were discharged home without medical contraindications to aldosterone antagonist therapy.
The primary outcome measure was prescription of an aldosterone antagonist at discharge, as defined by the ACC/AHA management guidelines.1 Evidence-based guidelines for aldosterone antagonist prescription included 3 components: hospitalization for HF with LVEF of 35% or less, serum potassium level of 5.0 mEq/L or less, and serum creatinine level of 2.5 mg/dL or less (to convert to μmol/L, multiply by 88.4) in men and 2.0 mg/dL or less in women on admission during the current HF episode.
We assessed evidence-based guideline prescription of an aldosterone antagonist at discharge in hospitals with at least 10 eligible patients as well as the relationship of aldosterone antagonist prescription and quality-of-care measures. To assess appropriateness of aldosterone antagonist use, we examined use based on each of the 3 components of the ACC/AHA guidelines (LVEF, serum creatinine level, and serum potassium level); use in patients with LVEF of 35% or less who received angiotensin-converting enzyme (ACE) inhibitor, angiotensin receptor blocker (ARB), or β-blocker therapy; and use in patients with preserved LVEF (>40%) and history of hypertension.
We also assessed inappropriate use of aldosterone antagonists, guideline-specified inappropriate use of aldosterone antagonists, and potentially inappropriate use of aldosterone antagonists in patients who received therapy at discharge. Inappropriate use of aldosterone antagonists was defined as patients receiving therapy when they had contraindications documented in the medical record (hyperkalemia, allergy to aldosterone antagonist, circumstances in which serum electrolyte monitoring was not anticipated to be feasible, pregnancy, or other reasons), serum creatinine levels of 3.0 mg/dL or greater, or potassium levels of 6.0 mEq/L or greater.
Guideline-specified inappropriate use was defined as receiving an aldosterone antagonist when also receiving both ACE inhibitor and ARB therapy at discharge. Potentially inappropriate use of aldosterone antagonists was defined as patients receiving therapy when serum creatinine levels were 2.5 mg/dL or greater and less than 3 mg/dL, serum potassium levels were greater than 5.0 mEq/L and less than 6.0 mEq/L, or LVEF was normal (>40%) without history of hypertension.
Data were compared between patients prescribed and not prescribed an aldosterone antagonist, using χ2 tests for categorical variables and Wilcoxon rank sum 2-sample tests for continuous variables. Percentages and means with standard deviations were reported for categorical and normally distributed continuous variables, and medians with interquartile ranges were reported for nonnormally distributed continuous variables.
Cochran-Mantel-Haenzel general association statistics were used to test the association between aldosterone antagonist prescription and patient groups. Time was divided into six 6-month intervals. Cochran-Mantel-Haenzel row mean scores statistics were used to test the trend of aldosterone antagonist use over the 3-year study period. Cochran-Mantel-Haenzel tests were adjusted for within-hospital clustering. Multivariable logistic regression was used to identify important factors associated with aldosterone antagonist prescription in eligible patients. The generalized estimating equations approach was used to account for within-hospital clustering.9 Missing patient baseline factors were less than 7% and imputed. Missing data on hospital characteristics (2%) were excluded from multivariable modeling. Tests were 2-tailed, and P < .05 was considered statistically significant. Analyses were performed using SAS version 9.1.3 (SAS Institute Inc, Cary, North Carolina).
In 241 hospitals, 43 625 patients treated for HF and discharged to home did not have a documented contraindication to aldosterone antagonist therapy. In total, 12 565 patients (28.8%) from 201 hospitals met ACC/AHA management guidelines criteria, and 4087 eligible patients (32.5%) received an aldosterone antagonist at hospital discharge. Table 1 provides patient characteristics by prescription of aldosterone antagonist based on ACC/AHA guideline criteria. Table 2 provides hospital characteristics and compliance with Joint Commission and ACC/AHA performance measures based on aldosterone antagonist prescription.
Hospital use of aldosterone antagonist therapy among eligible patients ranged from 0% to 100%. When the analysis was confined to 140 hospitals that had at least 10 patients who met ACC/AHA management guidelines criteria, use of aldosterone antagonist therapy by hospital remained highly variable (median, 28.3%; interquartile range, 17.6%-39.8%; range, 0%-90.6%). In these 140 hospitals, the median number of eligible patients was 38 (interquartile range, 21-103).
Table 3 demonstrates independent patient and hospital characteristics associated with aldosterone antagonist prescription in patients meeting ACC/AHA management guidelines criteria. After adjustment, patients more likely to have an aldosterone antagonist prescription were younger, African American, had lower systolic blood pressure, and had a history of implantable cardioverter-defibrillator implantation, depression, alcohol use, and pacemaker implantation. Patients with renal insufficiency were less likely to receive a prescription for an aldosterone antagonist. In contrast, hospital characteristics did not remain associated with aldosterone antagonist prescription after adjustment for clinical factors.
In patients meeting ACC/AHA management guidelines criteria, aldosterone antagonist therapy prescription increased from 27.96% to 34.48% (P < .001) between January 2005 and December 2007 (Figure). To assess appropriateness of aldosterone antagonist prescription, different definitions based on guideline specifications were explored. Based on the 3 individual component criteria of the ACC/AHA management guidelines, aldosterone antagonist use in patients by LVEF criteria was 29.5%, by serum creatinine criteria was 32%, and by serum potassium criteria was 31.1%. For each component, aldosterone antagonist prescription increased significantly over the six 6-month intervals (P ≤ .002 for all).
Use of aldosterone antagonists in patients with LVEF of 35% or less who received ACE inhibitor, ARB, or β-blocker therapy at discharge was 30.1% and increased significantly over time (P < .001). Additionally, aldosterone antagonist use remained low in patients with HF and preserved systolic function and hypertension over 3 years; lowest use was between January and June 2005 (8.4%), and highest use was between January and June 2006 (10.8%) (P = .03).
Of 47 706 patients with HF who were discharged to home, 8610 received an aldosterone antagonist prescription. Inappropriate use, guideline-specified inappropriate use, and potentially inappropriate use of aldosterone antagonist therapy were evaluated among these patients (Table 4). Of patients who received an aldosterone antagonist, 269 (3.1%) met at least 1 criterion for inappropriate use: 0.5% had a documented contraindication in the medical record, 2.7% had a serum creatinine level of 3.0 mg/dL or greater, and 0.07% had a serum potassium level of 6.0 mEq/L or greater.
Of patients discharged with aldosterone antagonist therapy, 194 (2.3%) also received the ACC/AHA guideline–proscribed (class III recommendation) combination of ACE inhibitor and ARB therapy. Of 8610 patients discharged home with aldosterone antagonists, 640 (7.4%) met at least 1 criterion for potentially inappropriate use (Table 4). When evaluating trends over time in aldosterone antagonist therapy use by appropriateness criteria, there were no changes in prescription over time (P = .67 for inappropriate use, P = .42 for nonguideline inappropriate use of triple therapy, and P = .49 for potentially inappropriate use).
This is, to our knowledge, the first report of aldosterone antagonist use and appropriateness since the 2005 ACC/AHA guideline statements emphasized both appropriate indications and safety precautions.
This study had 4 main findings. First, only one-third of patients meeting current guideline criteria for aldosterone antagonist therapy and without documented contraindications were treated. Second, prescription of aldosterone antagonist at discharge varies widely among clinicians. Third, aldosterone antagonist use for appropriate indications is less common among patients who are elderly, white, have a lower systolic blood pressure, do not have an implantable cardioverter-defibrillator or pacemaker, do not have a history of alcohol use or depression, and have a history of renal insufficiency. Fourth, in contrast to prior reports, rates of inappropriate use are infrequent; only 0.5% of patients received therapy when they had a documented contraindication, and 2.7% received therapy when they had a higher than recommended creatinine level.
Aldosterone antagonist therapy increased modestly in the United States, from 21.3% in 2001 in Medicare beneficiaries after hospitalization for HF5 to 28% in the first half of 2005 and 34% in the second half of 2007 among adult patients hospitalized with HF. Similarly, investigators from the Registry to Improve the Use of Evidence-Based HF Therapies in the Outpatient Setting (IMPROVE-HF) found that 36% of patients in an ambulatory cardiology practice who met usage criteria and were without documented contraindications were prescribed an aldosterone antagonist.10 In the EuroHeart Failure Survey II (EHFS II) of 3580 hospitalized patients with acute HF from 133 centers in 30 European countries, in 2004-2005 the rate of aldosterone antagonist use at hospital discharge was 48%.11 Although prescription of aldosterone antagonists increased over time based on this GWTG-HF report, use was less than expected on the basis of clinical trial evidence demonstrating substantial reductions in all-cause mortality and all-cause hospitalization and also less than the European use rate.
Slow adoption of aldosterone antagonists in patients meeting HF management guideline criteria may be attributable to wording in the original version of the 2005 guidelines stating that aldosterone antagonist therapy “should be considered” rather than “is recommended.” The wording was subsequently corrected12,13 but may have influenced treatment patterns prior to the 2009 guidelines update.
Other factors that may have slowed adoption are (1) published hyperkalemia rates that may have been an overestimation of harm4 as well as other published reports of safety concerns14-19 and adequacy in laboratory monitoring,20,21 (2) poor understanding of using spironolactone as an aldosterone antagonist rather than as a potassium-sparing diuretic, (3) minimal pharmaceutical company–sponsored drug marketing and drug education to clinicians, (4) the adverse-effect profile of spironolactone, (5) evidence for use based primarily on a single randomized, multicenter trial initiated before β-blocker therapy was routinely used, (6) selection of other HF add-on drug therapies when patients remain symptomatic, and (7) clinicians' plans for initiating therapy after hospital discharge once ACE inhibitor or ARB and β-blocker therapies are optimized.
Slow adoption by hospitals could be attributable to lack of monitoring aldosterone antagonist use, since it is not a HF performance measure selected by The Joint Commission/Centers for Medicare and Medicaid Services or the ACC/AHA. Admitting and discharge order sets, algorithms, and other tools developed to aid in the provision of quality care may be focused on core performance measures. Other factors that may have slowed adoption by hospitals are physician specialty, concerns that hyperkalemia and an increased creatinine level would extend hospital length of stay, or lack of time to start aldosterone antagonist therapy owing to a short length of stay for HF. However, when aldosterone antagonists were assessed for use in patients with left ventricular systolic dysfunction by length of stay, prescription at discharge was not associated with hospital length of stay.22
Moreover, slow adoption may be attributable simply to nonadherence to evidence-based practice. In this study, use of aldosterone antagonists was significantly higher in patients also receiving other evidence-based HF therapies. When European cardiologists were asked to cite rationale for guideline nonadherence, 3 prominent factors were patient compliance, long and complexly written guidelines, and time needed by cardiologists to implement guidelines.23
In our study, aldosterone antagonist–treated patients were younger than those not treated. When pre- and post–Randomized Aldactone Evaluation Study (RALES) study populations were compared, patient age was unchanged over time in one report4 and was higher in another.5 There are possible implications of age on appropriateness of aldosterone antagonist use. Advanced age is a risk factor for hyperkalemia when drugs are used that interfere with the renin-angiotensin-aldosterone system,24 but it is unknown if less use of aldosterone antagonists in older patients in this study represents a disparity that needs to be overcome or a safety measure. After spironolactone initiation, hyperkalemia occurrence did not differ by age20 except in 1 single-center study report,15 and patients who received laboratory monitoring were not older than those who did not.21 Moreover, the mean age at spironolactone initiation was lower than mean age at the time of hyperkalemia (67 vs 73 years), suggesting that decreased creatinine clearance with age contributes to hyperkalemia.17 Thus, age may be a surrogate of renal function in patients receiving multiple therapies targeting the renin-angiotensin-aldosterone system. After drug dose stabilization, hyperkalemia monitoring may need to be based on patient age and baseline renal function rather than on a set schedule.
A high proportion of patients in GWTG-HF hospitals had normal potassium and creatinine levels compared with previous observational reports using similar data collection methods,5,14-17 reflecting clinician knowledge of safety recommendations for hyperkalemia and renal dysfunction. Appropriateness of aldosterone antagonist therapy may reflect the influence of evidence-based, guideline-driven therapy or participation in a performance improvement program; however, our analyses were not designed to determine either circumstance. It is plausible that well-refereed guideline statements inform clinical practice and when incorporated into a program designed to improve quality yield more appropriate application of evidence-based care. Ongoing monitoring of processes of care may influence discontinuation or nonuse of aldosterone antagonist therapy based on abnormal laboratory findings, because GWTG-HF provides opportunities to evaluate care outcomes and continually improve performance.
This study has several limitations. The data are dependent on the accuracy and completeness of data abstraction from medical chart review. Contraindications and intolerance of aldosterone antagonist and other core HF drug therapies were abstracted as recorded in the medical record. As such, a proportion of patients reported to be eligible for aldosterone antagonist therapy but not treated may have had undocumented contraindications or intolerances. The database does not track specific aldosterone antagonist agent or dosage, which may have influenced rates of hyperkalemia and renal dysfunction. Data on physician specialty or type of inpatient service were not collected. GWTG-HF hospitals are self-selected; our analysis could overrepresent high-performing hospitals or hospitals with cardiologists specializing in care of HF. Rationale for decisions regarding therapy utilization was not captured. Patients may have received an aldosterone antagonist for a non-HF condition; thus, our definitions for appropriate use may not have applied.
Because our data set does not include longitudinal follow-up, a portion of eligible patients may have been started on aldosterone antagonist therapy following discharge, underestimating true use. Furthermore, consequences of potential inappropriate use following discharge, safety issues, and benefits after in-hospital therapy initiation are unknown. In clinical practice, clinicians may use a definition of moderate-to-severe HF different from that used in this study, which would alter findings. Lastly, these findings may not apply to practices that differed in patient characteristics or care patterns from GWTG-HF hospitals.
Aldosterone antagonist usage in eligible patients with HF increased only modestly from 2005 through 2007 and remained at less than 35% penetration. Importantly, inappropriate use was low. These data confirm that in the context of a hospital-based performance improvement program, aldosterone antagonist therapy can be used according to guidelines with little inappropriate use. Given the substantial morbidity and mortality risk faced by patients hospitalized with HF and the established efficacy of aldosterone antagonist prescription in HF, a stronger uptake of aldosterone antagonist therapy indicated by evidence-based guidelines may be warranted.
Corresponding Author: Nancy M. Albert, PhD, RN, 9500 Euclid Ave, Mail Code J3-4, Cleveland, OH 44195 (firstname.lastname@example.org).
Author Contributions: Drs Albert and Fonarow had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.
Study concept and design: Albert, Yancy, Cannon, Fonarow.
Acquisition of data: Albert, Fonarow.
Analysis and interpretation of data: Albert, Liang, Zhao, Hernandez, Peterson, Fonarow.
Drafting of the manuscript: Albert, Fonarow.
Critical revision of the manuscript for important intellectual content: Albert, Yancy, Liang, Zhao, Hernandez, Peterson, Cannon, Fonarow.
Statistical analysis: Liang, Zhao, Peterson.
Obtained funding: Peterson, Cannon, Fonarow.
Administrative, technical, or material support: Albert, Hernandez, Cannon, Fonarow.
Study supervision: Albert, Yancy, Cannon, Fonarow.
Financial Disclosures: Dr Albert reported serving on the speakers bureau of GlaxoSmithKline, Medtronic, Nitromed, and Scios (since 2004; speakers bureau relationships with Nitromed, Medtronic, and Scios ended in 2007); serving as a consultant for Arca biopharma (2008); and currently serving as a consultant for Medtronic, Merck, and Impedance Cardiology Systems Inc. Dr Yancy reported receiving research grants from GlaxoSmithKline, Medtronic, NitroMed, and Scios (since 2004); previously serving as a consultant for or on the speakers bureau of Astella, AstraZeneca, GlaxoSmithKline, Medtronic, NitroMed, Novartis, Otsuka, and Scios; previously serving on the advisory board for CHF Solutions (all of these relationships expired prior to 2007); serving on the advisory board for ARCA Discovery and receiving additional research support from GlaxoSmithKline, Medtronic, and Thoratec Inc (all of which have expired or represent uncompensated work as of June 2008); currently serving on the US Food and Drug Administration cardiovascular device panel and study section for the National Institutes of Health; and holding an editorial position with the American Journal of Cardiology and serving on several other editorial boards. Dr Hernandez reported receiving research grants from GlaxoSmithKline, Johnson & Johnson (Scios Inc), Medtronic, Merck, and Roche Diagnostics; and serving on the speakers bureau for or receiving honoraria within the past 5 years from AstraZeneca, Medtronic, Novartis, Sanofi-Aventis, and Thoratec Corporation; Dr Hernandez has made detailed listings of financial disclosures available online at http://www.dcri.duke.edu/research/coi.jsp. Dr Peterson reported serving as the principal investigator for the American Heart Assocation Get With The Guidelines data analytic center at the Duke Clinical Research Institute; receiving a research grant from BMS/Sanofi; and previously serving on an advisory board for Pfizer and as a consultant for Eli Lilly; Dr Peterson has made detailed listings of financial disclosures available online at http://www.dcri.duke.edu/research/coi.jsp. Dr Cannon reported receiving research grants/support from Accumetrics, AstraZeneca, Bristol-Myers Squibb/Sanofi Partnership, GlaxoSmithKline, Intekrin, Merck, Merck/Schering-Plough Partnership, Novartis, and Takeda; and serving as a clinical advisor with equity in Automedics Medical Systems. Dr Fonarow reported receiving research grants or other research support from GlaxoSmithKline, Pfizer, and the National Institutes of Health; receiving honoraria from Amgen, AstraZeneca, Boston Scientific/Guidant, GlaxoSmithKline, Medtronic, Merck, Novartis, Pfizer, Schering-Plough, and St Jude Medical; serving as a consultant for ARCA, Boston Scientific/Guidant, DebioPharm, GlaxoSmithKline, Medtronic, Merck, Novartis, Pfizer, Relypsa, Scios, and St Jude Medical; and serving as chair of the American Heart Association Get With The Guidelines Steering Committee;. Dr Fonarow holds the Eliot Corday Chair of Cardiovascular Medicine at UCLA and is also supported by the Ahmanson Foundation (Los Angeles, California). No other authors reported disclosures.
Funding/Support: The Get With The Guidelines–Heart Failure program is sponsored by the American Heart Association and receives funding in part from GlaxoSmithKline and Medtronic.
Role of the Sponsors: GlaxoSmithKline and Medtronic had no role in the design and conduct of the study; the collection, management, analysis, and interpretation of the data; or the preparation, review, or approval of the manuscript.
Disclaimer: Dr Peterson, a contributing editor for JAMA, was not involved in the editorial review of or the decision to publish this article.