Adoption of Spironolactone Therapy for Older Patients With Heart Failure and Left Ventricular Systolic Dysfunction in the United States, 199
http://www.100md.com
《循环学杂志》
the Division of Cardiology, Department of Medicine, Denver Health Medical Center (F.A.M., E.P.H.)
the Division of Cardiology, Department of Medicine (F.A.M., E.P.H.)
the Division of Geriatric Medicine, Department of Medicine (F.A.M.)
University of Colorado Health Sciences Center, Denver, Colo
the Colorado Foundation for Medical Care (F.A.M., E.P.H., H.M.K.), Aurora, Colo
the Colorado Health Outcomes Program (F.A.M.), Aurora, Colo
the Section of General Internal Medicine, Department of Medicine (C.P.G.)
the Section of Cardiovascular Medicine, Department of Medicine (S.S.R., Y.W., J.M.F., H.M.K.), and the Section of Health Policy and Administration, Department of Epidemiology and Public Health (H.M.K.), Yale University School of Medicine, New Haven, Conn; the Center for Outcomes Research and Evaluation (H.M.K., C.G.), Yale–New Haven Hospital, New Haven, Conn; and the Section of Cardiology, Department of Medicine, West Haven Veteran’s Administration Medical Center (J.M.F.), West Haven, Conn.
Abstract
Background— Concerns have been raised about the appropriateness of spironolactone use in some patients with heart failure. We studied the adoption of spironolactone therapy after publication of the Randomized Aldactone Evaluation Study (RALES) in national cohorts of older patients hospitalized for heart failure.
Methods and Results— This is a study of serial cross-sectional samples of Medicare beneficiaries 65 years old discharged after hospitalization for the primary diagnosis of heart failure and with left ventricular systolic dysfunction. The first sample was discharged before (April 1998 to March 1999, n=9758) and the second sample after (July 2000 to June 2001, n=9468) publication of RALES in September 1999. We assessed spironolactone prescriptions at hospital discharge in patient groups defined by enrollment criteria for the trial. Using multivariable logistic regression, we identified factors independently associated with prescriptions not meeting these criteria. Spironolactone use increased >7-fold (3.0% to 21.3% P<0.0001) after RALES. Of patients meeting enrollment criteria, 24.1% received spironolactone, as compared with 17.4% of those not meeting the criteria. Of all prescriptions after RALES, 30.9% were provided to patients not meeting enrollment criteria. Spironolactone was prescribed to 22.8% of patients with a serum potassium value 5.0 mmol/L, to 14.1% with a serum creatinine value 2.5 mg/dL, and to 17.3% with severe renal dysfunction (estimated glomerular filtration rate <30 mL · min–1 · 1.73 m–2). In multivariable analyses, factors associated with prescriptions not meeting enrollment criteria included advanced age, noncardiovascular comorbidities, discharge to skilled nursing facilities, and care provided by physicians without board certification.
Conclusions— Spironolactone prescriptions increased markedly after the publication of RALES, and many treated patients were at risk for hyperkalemia. Simultaneously, many patients who might have benefited were not treated. These findings demonstrate the importance of balancing efforts to enhance use among appropriate patients and minimizing use in patients at risk for adverse events.
Key Words: heart failure ; aging ; aldosterone antagonists ; potassium
Introduction
Although the Randomized Aldactone Evaluation Study (RALES) demonstrated the benefits of aldosterone blockade with spironolactone in selected patients with severely symptomatic left ventricular systolic dysfunction (LVSD), this therapy has not been proven beneficial for all heart failure patients.1 Spironolactone may cause serious hyperkalemia, particularly in patients with marginal renal function, type IV renal tubular acidosis, or relatively high baseline potassium levels. Because of the potential risks of spironolactone therapy, concerns have been raised about the application of the trial results in case reports and series from single centers.2–9 More recently, a study from Canada reported simultaneous increases in spironolactone prescriptions and hospitalizations for hyperkalemia after the publication of RALES.10 Although that study could not determine whether that observation resulted from spironolactone use in patients at high risk for hyperkalemia in clinical practice, it suggests the possibility of inappropriate use or inadequate monitoring of spironolactone in the community. Because aldosterone blockade may improve survival in some patients with heart failure but pose a serious risk in others, the simultaneous assessment of appropriate and potentially inappropriate use of spironolactone is important.
Current clinical guidelines acknowledge this difficult balance, recommending that spironolactone be restricted to patients similar to those enrolled in RALES.11 Specifically, the guidelines recommend spironolactone for patients with LVSD; recent or current symptoms at rest despite the use of digoxin, diuretics, an angiotensin-converting enzyme (ACE) inhibitors, and (usually) a ;-blocker; a serum potassium value <5.0 mmol/L; and creatinine value <2.5 mg/dL before initiation of treatment.11 Because many community-based patients do not fit these criteria,12 widespread use of spironolactone without consideration of these factors could expose many patients to adverse outcomes.
The degree to which spironolactone is used for patients who might derive a benefit and is avoided among those at high risk for adverse events is not known. To address this issue, we assessed national patterns of spironolactone use before and after publication of the RALES results in September 1999 in a community-based, nationally representative sample of patients. This study was intended to assess the rate of adoption of spironolactone in the United States and the degree to which the selection of patients for treatment after publication of RALES conformed to study enrollment criteria.
Methods
National Heart Care Project
The National Heart Care (NHC) Project is an ongoing initiative funded by the Centers for Medicare and Medicaid Services designed to improve the quality of care for Medicare beneficiaries with heart failure.13 Fee-for-service Medicare beneficiaries hospitalized with a principal discharge diagnosis of heart failure (International Classification of Diseases, Ninth Revision, Clinical Modification codes 402.01, 402.11, 402.91, 404.01, 404.91, or 428) between April 1998 and March 1999 or July 2000 and June 2001, inclusive, were identified. These periods straddled the publication of RALES. According to a sampling strategy described previously,13,14 discharges were grouped by state and sorted by age, sex, race, and treating hospital, and up to 800 discharges were randomly selected from each state. The selected records underwent detailed review by trained data abstractors in central data abstraction centers. Patients with invalid social security numbers, those receiving long-term hemodialysis, those transferred to another hospital, or those who left against medical advice based on data from the administrative records or the chart abstraction were excluded. The NHC sample thus consisted of 78 882 records, of which 39 477 were from 1998 to 1999 and 39 405 from 2000 to 2001.
Study Sample
Because this was a study of the prescription of spironolactone at discharge, patients who died during hospitalization were excluded (n=4126). Because RALES enrolled only patients with LVSD, we further restricted the analysis to patients with an LV ejection fraction <40% or a qualitative description of moderate or severe systolic dysfunction, based on the evaluation closest to the discharge date (n=22 928). From this group, we excluded patients younger than 65 years because younger Medicare beneficiaries are not representative of the younger population as a whole. If a patient appeared more than once in the sample, 1 record was included by random selection. Patients who were transferred to another acute care hospital, left against medical advice, were receiving long-term renal dialysis, or with missing vital status were also excluded. A total of 3702 records of patients with LVSD were excluded by these selection criteria, resulting in a final study cohort of 19 226 patients (9758 in 1998 to 1999 and 9468 in 2000 to 2001).
Clinical Variables
The NHC database includes a wide range of variables, including demographic characteristics, medical history, characteristics on presentation, diagnostic testing, laboratory data, and a comprehensive list of the names of medications prescribed at hospital discharge. For the purposes of interpretability, drugs were assigned to classes. Combination drugs were assigned to all classes contained in the combination. The following drugs were assessed to determine spironolactone prescriptions (in alphabetical order): Aldactone (Pfizer), Aldactazide (Pfizer), hydrochlorothiazide/spironolactone, and spironolactone.
Spironolactone treatment was also assessed in subgroups defined by the primary selection criteria used for RALES.1 Patients meeting all of the following criteria were designated as conforming to these criteria: (1) discharge prescription for an ACE inhibitor or angiotensin receptor blocker or a documented reason for nonuse of ACE inhibitors; (2) discharge prescription for a diuretic (not including spironolactone); (3) a serum creatinine value closest to hospital discharge <2.5 mg/dL; and (4) a serum potassium value closest to hospital discharge of <5.0 mmol/L. All other patients were designated as not meeting RALES enrollment criteria. Because the serum creatinine value may overestimate renal function, particularly in elderly patients, an estimated glomerular filtration rate (GFR) was also calculated for all patients according to the abbreviated Modification of Diet in Renal Disease Study equation.15
Additional Data Sources
NHC data were linked with the American Medical Association Physician Masterfile16,17 by using the unique physician identification number of the attending physician, defined as the clinician primarily responsible for the patient’s care during hospitalization.18 Hospital characteristics were ascertained by linking with the 1998 and 2000 American Hospital Association Annual Surveys.19,20
Statistical Analysis
Patient, physician, and hospital characteristics, as well as documentation of spironolactone prescription at discharge, were compared between the 1998 to 1999 and 2000 to 2001 periods by 2 statistics. Correlates of spironolactone prescription were determined by characteristics of the patients (demographics, medical history, comorbidities, and admission variables), treating physician (board certification, cardiology specialty), and hospital (teaching status, cardiac care facilities, urban or rural location, ownership, and US census division). Differences in proportions were evaluated by 2 statistics.
To assess changes in use between 1998 to 1999 and 2000 to 2001 after adjustment for differences in patient, physician, and hospital characteristics, logistic regression was performed, with spironolactone prescription at discharge as the dependent variable. Models comprising statistically significant covariates (P<0.05) were constructed by backward selection. Subsequently, a variable indicating the sampling time frame was introduced into the model. The statistical significance of 2-way interactions between the time frame variable and other covariates included in the final model was tested to identify heterogeneity in the rate of increase in spironolactone use.
Finally, to identify characteristics associated with spironolactone prescription that did not conform to RALES enrollment criteria, a model in only those patients who were treated with spironolactone in 2000 to 2001 was constructed, with failure to conform to all RALES enrollment criteria as the dependent variable. Because the characteristics composing the criteria defined the dependent variable in this model, these variables were not candidate independent variables. All analyses used probability weights to account for the NHC sampling strategy, thus providing national level estimates. Statistical analyses were conducted with Stata 7.0 (Stata Corp) and SAS version 8.0 (SAS Institute, Inc).
Results
Patient Population
In both samples, 20% were at least 85 years old, the majority were male, and most had a history of hypertension and coronary artery disease (Table 1). Noncardiovascular comorbidities, including diabetes and chronic pulmonary disease, were also common. An elevated potassium level at hospital discharge (>5.0 mmol/L) was present in >5% of patients. Although the prevalence of serum creatinine >2.5 mg/dL was 8.3% in 1998 to 1999 and 9.5% in 2000 to 2001, the prevalence of a severely reduced estimated GFR (ie, <30 mL · min–1 · 1.73 m–2) was higher (15.2% and 16.4%, respectively). Approximately 42% of patients in 2000 to 2001 did not meet the appropriateness criteria for spironolactone treatment.
1998 to 1999 and 2000 to 2001, there was a decline in the proportion of patients treated at hospital discharge with ACE inhibitors or angiotensin receptor blockers, diuretics (not including spironolactone), and potassium replacements. The proportion treated with ;-blockers increased significantly between periods.
Changes in Spironolactone Prescription
Overall, spironolactone prescription rates in the cohort increased from 3.0% in 1998 to 1999 to 21.3% in 2000 to 2001 (P<0.001, Table 2). Although the increase in prescriptions between the 2 time periods was significant in all patient subgroups, the change was significantly greater in younger patients (P for interaction, 0.001); those with a lower admission blood pressure (P<0.001); those with angina (P<0.001); and those not admitted who were already being treated with spironolactone (P<0.001). Rates of increase did not differ significantly, however, by levels of discharge serum creatinine, estimated GFR, or discharge potassium supplementation.
In the population of patients discharged in 2000 to 2001, spironolactone was prescribed for 22.8% of patients with a serum potassium value >5.0 mmol/L, for 14.1% of patients with a serum creatinine value >2.5 mg/dL, and for 17.3% of patients with severe renal dysfunction, as defined by an estimated GFR (<30 mL · min–1 · 1.73 m–2; the Figure). Among patients not receiving an ACE inhibitor prescription, 17.2% received spironolactone, and among those receiving a prescription for potassium supplements, 17.1% were discharged with a spironolactone prescription. Among patients meeting study enrollment criteria after RALES, spironolactone was prescribed to 24.1% (a 21.2% absolute increase); among those not meeting these criteria, 17.4% received a discharge prescription for spironolactone (14.2% absolute increase). Of all spironolactone prescriptions in 2000 to 2001, 30.9% were provided to patients not meeting enrollment criteria.
Changes in proportions of patients receiving prescription for spironolactone at hospital discharge between 1998 to 1999 (before RALES) and 2000 to 2001 (after RALES) in all patients and patients stratified by serum potassium, creatinine, and estimated GFR.
Correlates of Spironolactone Prescription Not Meeting RALES Criteria
Among those patients receiving a prescription for spironolactone at discharge, prescriptions not conforming to RALES criteria were more common among older patients (compared with an age 65- to 74-year referent; age 75 to 84 years odds ratio [OR], 1.40; 95% confidence interval [CI], 1.14 to 1.72; P<0.001; age 85+ OR, 1.28; 95% CI, 0.96 to 1.70; P=0.09, Table 3) and were less common among women (OR, 0.80; 95% CI, 0.66 to 0.97; P=0.02). Prior history of heart failure, atrial fibrillation, and higher admission systolic blood pressures were associated with a lower OR of prescription not conforming to criteria, whereas noncardiovascular comorbidities, including chronic lung disease and anemia, were associated with higher ORs. Patients discharged to a skilled nursing facility had >2-fold higher odds of receiving spironolactone prescriptions not meeting criteria (OR, 2.31; 95% CI, 1.71 to 3.11; P<0.001). Patients cared for by nongeneralist physicians without cardiology consultation were more likely to conform to the RALES criteria (OR, 0.57; 95% CI, 0.39 to 0.84; P=0.005) compared with those cared for by a cardiologist (1.00 referent), as were those cared for by a board-certified physician (OR, 0.70; 95% CI, 0.54 to 0.90; P=0.005) compared with those cared for by a physician without board certification.
Discussion
In this study of 2 cross-sectional cohorts of older patients with heart failure and LVSD, prescription of spironolactone increased >7-fold after publication of RALES. Of the discharge prescriptions written after RALES, almost one third were provided to patients not fitting the study enrollment criteria, many of whom were at high risk for hyperkalemia. Slow adoption among patients who might benefit from spironolactone simultaneously with rapidly increasing use among those at higher risk for adverse consequences indicates that the integration of clinical trials results in practice may not maximize either effectiveness or safety. These findings demonstrate the complexity of adoption of information from new clinical trials, emphasizing the need for faster adoption among appropriate patients and simultaneous efforts to minimize use in patients who might suffer severe adverse events from inappropriate use.
RALES established the efficacy of spironolactone in reducing mortality and hospitalization in selected patients with heart failure and LVSD.1 In our study, a minority of patients who had serum potassium and creatinine levels within the ranges of RALES subjects and who were receiving other evidence-based therapy for heart failure were not treated with spironolactone at hospital discharge. This finding is concordant with other data demonstrating that standards of clinical care lag behind the evidence generated in clinical trials. The development of mechanisms to facilitate rapid and appropriate diffusion of trial results into patient care is likely to improve important patient outcomes.
RALES, however, excluded patients who were not receiving evidence-based heart failure therapy and those at potentially high risk for hyperkalemia. Several factors predispose some patients with heart failure to the risk of potentially life-threatening hyperkalemia with an agent that impairs aldosterone action. Both diabetes, which is associated with type IV renal tubular acidosis,21 and renal insufficiency are common in populations with heart failure.13,22–24 The addition of an aldosterone antagonist to the regimen of patients with underlying abnormalities of potassium excretion already being treated with ACE inhibitors, angiotensin receptor blockers, ;-blockers, or potassium supplements may increase the risk of hyperkalemia to varying degrees. Although RALES demonstrated significant mortality benefits in a patient population predominantly treated with ACE inhibitors, the trial did not enroll patients with severe renal insufficiency (defined as a serum creatinine value <2.5 mg/dL) or baseline hyperkalemia (potassium >5.0 mmol/L).1 Patients in the trial were also assessed frequently after the initiation of therapy for the development of hyperkalemia and worsening renal insufficiency, which may not be the case in usual clinical practice.4
Since the publication of RALES, concerns have been raised about the adoption of spironolactone outside the context of a carefully controlled clinical trial.2 Within a year of the release of RALES, a single-center case series of 25 patients treated with spironolactone were admitted with serious hyperkalemia. These patients were predominantly older, and none had a serum potassium value exceeding 4.8 mmol/L before admission.25 Several subsequent case series have described patients with hyperkalemia requiring hospitalization, some of whom died.5–9 Many patients in these series were elderly, in whom the serum creatinine value often overestimates true renal function. The results of our study raise significant concerns about the safety of current patterns of spironolactone prescription in community-based populations, particularly among older patients with noncardiovascular comorbidities.
A population-based study from Ontario, Canada, found that the use of spironolactone increased from 3.4% before the publication of RALES to 14.9% afterward.10 Simultaneously, hospitalization rates for hyperkalemia increased >4-fold, and associated mortality increased by >6-fold. The authors of that study speculated that hyperkalemia resulted in part from the inappropriate use of spironolactone but were unable to demonstrate whether the patterns they observed were due to misapplication of the RALES results. This is the first study in a nationally representative sample demonstrating widespread use of spironolactone in patients for whom there is no good evidence for benefit, many of whom are at high risk for hyperkalemia.
The finding of limited adoption of spironolactone use in patients like those enrolled in RALES with simultaneous widespread use in populations of patients at risk for hyperkalemia has important implications for quality improvement and patient safety initiatives in heart failure. Typically, efforts to improve patterns of pharmacological therapy in patients with heart failure have focused exclusively on either underuse (eg, ACE inhibitors or ;-blockers) or overuse (eg, type I antiarrhythmic agents or nonsteroidal antiinflammatory drugs).26–28 Given the patterns of care shown in this study, any efforts to increase the use of spironolactone in patients with heart failure should also include mechanisms to ensure that this drug is not used in populations in whom the risks may outweigh the benefits.
The demonstration that spironolactone is frequently prescribed to patients at high risk in a national sample of older patients with heart failure in conjunction with evidence of higher population rates of hyperkalemia associated with increasing spironolactone use10 suggest the need for action to change current patterns of use of aldosterone-blocking drugs. Although several mechanisms may be useful in reducing the potential harm to patients, it is possible that changing the labeling of aldosterone-blocking agents to reflect these concerns (eg, more stringent precautions) and efforts by industry and professional associations to inform practitioners of these hazards would have a positive influence. With the more recent publication of EPHESUS, a study of the selective aldosterone blocker eplerenone in patients with heart failure after myocardial infarction,29 the use of aldosterone blockade is likely to proliferate further, increasing the importance of ensuring that these agents are used only in those populations for whom the benefit outweighs the risk.
Certain issues should be considered in the interpretation of these results. First, it was not possible to determine the relation between the timing of initiation of spironolactone and the measurement of laboratory values. Thus, in some cases, serum potassium and creatinine levels may have reflected in part the results of treatment with spironolactone. Patients with elevated potassium levels and marked renal dysfunction, however, are at relatively higher risk for hyperkalemia if spironolactone is continued. Second, we used a relatively conservative definition for RALES enrollment criteria. The data did not include symptom severity by New York Heart Association (NYHA) classification or the frequency of subsequent clinical assessments. Because RALES studied only those patients with advanced symptoms (NYHA class III or IV) and provided very close follow-up for hyperkalemia, it is possible that our study overestimates the proportion of safe prescriptions. Third, the doses of spironolactone prescribed at discharge were not available in our dataset. Because evidence from case series indicates that the risk of hyperkalemia increases with higher spironolactone dose,6,25 such data might identify a larger number of patients at risk for adverse outcomes. Third, we were unable to assess changes in spironolactone use after hospital discharge. Because of the rapid adoption of spironolactone during the time period under study, prescriptions at the time of hospital discharge may have underestimated use in the outpatient setting. This also prohibits an accurate assessment of the relation between spironolactone use and outcomes such as mortality or readmission because of the potential for significant misclassification of exposure. Although we could not assess the impact of the use of spironolactone on outcomes, there is evidence that use of spironolactone in elderly patients without consideration of serum potassium and creatinine values can result in adverse outcomes, including death.7–9
In conclusion, the prescription of spironolactone in older patients with LSVD hospitalized with heart failure increased substantially in the United States after publication of RALES. More than one third of prescriptions were provided to patients who did not meet enrollment criteria of the trial, many of whom had characteristics that placed them at high risk for hyperkalemia. The means of ensuring use only among those patients with an acceptably low risk of adverse events will be important for optimizing the benefits and risks of aldosterone antagonists in the clinical care of populations of patients with heart failure.
Acknowledgments
Dr Masoudi is supported by NIH/NIA Research Career Award K08-AG01011. Dr Gross is supported by a Cancer Prevention, Control and Population Sciences Career Development Award (1K07CA-90402), the Claude D. Pepper Older Americans Independence Center at Yale (P30AG21342), and Paul Beeson career development award in Aging (K08 AG24842). Dr Foody is supported by NIH/NIA Research Career Award K08-AG20623 and NIA/Hartford Foundation Fellowship in Geriatrics. Saif Rathore is supported by NIH/National Institute of General Medical Sciences Medical Scientist Training Grant GM07205.
Disclosure
Dr Masoudi has received honoraria from Pfizer; Dr Foody has received honoraria from Pfizer. The content of this publication does not necessarily reflect the views or policies of the Department of Health and Human Services, nor does mention of trade names, commercial products, or organizations imply endorsement by the US government. The analyses on which this publication is based were performed under contract No. 500-02-CO01, entitled "Utilization and Quality Control Peer Review Organization for the State (commonwealth) of Colorado," sponsored by the Centers for Medicare and Medicaid Services (formerly Health Care Financing Administration), Department of Health and Human Services. The authors assume full responsibility for the accuracy and completeness of the ideas presented herein. This article is a direct result of the Health Care Quality Improvement Program initiated by the Centers for Medicare and Medicaid Services, which has encouraged identification of quality improvement projects derived from analysis of patterns of care, and therefore required no special funding on the part of this contractor. Ideas and contributions to the author concerning experience in engaging with issues presented are welcomed.
Footnotes
Guest Editor for this article was Clyde W. Yancy, MD.
References
Pitt B, Zannad F, Remme WJ, Cody R, Castaigne A, Perez A, Palensky J, Wittes J. The effect of spironolactone on morbidity and mortality in patients with severe heart failure: Randomized Aldactone Evaluation Study Investigators. N Engl J Med. 1999; 341: 709–717.
Geraci JM, Knowlton AA. Spironolactone for heart failure: spiraling out of control. Chest. 2000; 118: 1522–1523.
Berry C, McMurray J. Serious adverse events experienced by patients with chronic heart failure taking spironolactone. Heart. 2001; 85: e8–e9.
Bozkurt B, Agoston I, Knowlton AA. Complications of inappropriate use of spironolactone in heart failure: when an old medicine spirals out of new guidelines. J Am Coll Cardiol. 2003; 41: 211–214.
Svensson M, Gustafsson F, Galatius S, Hildebrandt PR, Atar D. Hyperkalaemia and impaired renal function in patients taking spironolactone for congestive heart failure: retrospective study. BMJ. 2003; 327: 1141–1142.
Wrenger E, Muller R, Moesenthin M, Welte T, Frolich JC, Neumann KH. Interaction of spironolactone with ACE inhibitors or angiotensin receptor blockers: analysis of 44 cases. BMJ. 2003; 327: 147–149.
Blaustein DA, Babu K, Reddy A, Schwenk MH, Avram MM. Estimation of glomerular filtration rate to prevent life-threatening hyperkalemia due to combined therapy with spironolactone and angiotensin-converting enzyme inhibition or angiotensin receptor blockade. Am J Cardiol. 2002; 90: 662–663.
Obialo CI, Ofili EO, Mirza T. Hyperkalemia in congestive heart failure patients aged 63 to 85 years with subclinical renal disease. Am J Cardiol. 2002; 90: 663–665.
Vanpee D, Swine CH. Elderly heart failure patients with drug-induced serious hyperkalemia. Aging Clin Exp Res. 2000; 12: 315–319.
Juurlink DN, Mamdani MM, Lee DS, Kopp A, Austin PC, Laupacis A, Redelmeier DA. Rates of hyperkalemia after publication of the Randomized Aldactone Evaluation Study. N Engl J Med. 2004; 351: 543–551.
Hunt SA, Baker DW, Chin MH, Cinquegrani MP, Feldman AM, Francis GS, Ganiats TG, Goldstein S, Gregoratos G, Jessup ML, Noble RJ, Packer M, Silver MA, Stevenson LW, Gibbons RJ, Antman EM, Alpert JS, Faxon DP, Fuster V, Jacobs AK, Hiratzka LF, Russell RO, Smith SC Jr. ACC/AHA guidelines for the evaluation and management of chronic heart failure in the adult: executive summary a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Committee to Revise the 1995 Guidelines for the Evaluation and Management of Heart Failure). Circulation. 2001; 104: 2996–3007.
Masoudi FA, Havranek EP, Wolfe P, Gross CP, Rathore SS, Steiner JF, Ordin DL, Krumholz HM. Most hospitalized older persons do not meet the enrollment criteria for clinical trials in heart failure. Am Heart J. 2003; 146: 250–257.
Havranek EP, Masoudi FA, Westfall KA, Wolfe P, Ordin DL, Krumholz HM. Spectrum of heart failure in older patients: results from the National Heart Failure project. Am Heart J. 2002; 143: 412–417.
Rathore SS, Foody JM, Wang Y, Smith GL, Herrin J, Masoudi FA, Wolfe P, Havranek EP, Ordin DL, Krumholz HM. Race, quality of care, and outcomes of elderly patients hospitalized with heart failure. JAMA. 2003; 289: 2517–2524.
National Kidney Foundation. K/DOQI clinical practice guidelines for chronic kidney disease: evaluation, classification, and stratification. Am J Kidney Dis. 2002; 39 (suppl 1): S1–S266.
Kenward K. The scope of the data available in the AMA’s Physician Masterfile. Am J Public Health. 1996; 86: 1481–1482.
Baldwin LM, Adamache W, Klabunde CN, Kenward K, Dahlman C, Warren L. Linking physician characteristics and Medicare claims data: issues in data availability, quality, and measurement. Med Care. 2002; 40 (suppl): IV-82–IV-95.
Iezzoni LI. Risk Adjustment for Measuring Health Care Outcomes, 2nd ed. Chicago: Health Administration Press; 1997.
American Hospital Association. The AHA Annual Survey Database: Fiscal Year 1998 Documentation. Chicago: Health Forum, AHA; 1999.
American Hospital Association. The AHA Annual Survey Database: Fiscal Year 2000 Documentation. Chicago: Health Forum, AHA; 2001.
Tan SY, Burton M. Hyporeninemic hypoaldosteronism: an overlooked cause of hyperkalemia. Arch Intern Med. 1981; 141: 30–33.
Nichols GA, Hillier TA, Erbey JR, Brown JB. Congestive heart failure in type 2 diabetes: prevalence, incidence, and risk factors. Diabetes Care. 2001; 24: 1614–1619.
Dries DL, Exner DV, Domanski MJ, Greenberg B, Stevenson LW. The prognostic implications of renal insufficiency in asymptomatic and symptomatic patients with left ventricular systolic dysfunction. J Am Coll Cardiol. 2000; 35: 681–689.
Hillege HL, Girbes AR, de Kam PJ, Boomsma F, de Zeeuw D, Charlesworth A, Hampton JR, van Veldhuisen DJ. Renal function, neurohormonal activation, and survival in patients with chronic heart failure. Circulation. 2000; 102: 203–210.
Schepkens H, Vanholder R, Billiouw JM, Lameire N. Life-threatening hyperkalemia during combined therapy with angiotensin-converting enzyme inhibitors and spironolactone: an analysis of 25 cases. Am J Med. 2001; 110: 438–441.
Jencks SF, Huff ED, Cuerdon T. Change in the quality of care delivered to Medicare beneficiaries, 1998–1999 to 2000–2001. JAMA. 2003; 289: 305–312.
Wenger NS, Shekelle PG. Assessing care of vulnerable elders: ACOVE project overview. Ann Intern Med. 2001; 135 (pt 2): 642–646.
Shekelle PG, MacLean CH, Morton SC, Wenger NS. ACOVE quality indicators. Ann Intern Med. 2001; 135 (pt 2): 653–667.
Pitt B, Remme W, Zannad F, Neaton J, Martinez F, Roniker B, Bittman R, Hurley S, Kleiman J, Gatlin M. Eplerenone, a selective aldosterone blocker, in patients with left ventricular dysfunction after myocardial infarction. N Engl J Med. 2003; 348: 1309–1321.(Frederick A. Masoudi, MD,)
the Division of Cardiology, Department of Medicine (F.A.M., E.P.H.)
the Division of Geriatric Medicine, Department of Medicine (F.A.M.)
University of Colorado Health Sciences Center, Denver, Colo
the Colorado Foundation for Medical Care (F.A.M., E.P.H., H.M.K.), Aurora, Colo
the Colorado Health Outcomes Program (F.A.M.), Aurora, Colo
the Section of General Internal Medicine, Department of Medicine (C.P.G.)
the Section of Cardiovascular Medicine, Department of Medicine (S.S.R., Y.W., J.M.F., H.M.K.), and the Section of Health Policy and Administration, Department of Epidemiology and Public Health (H.M.K.), Yale University School of Medicine, New Haven, Conn; the Center for Outcomes Research and Evaluation (H.M.K., C.G.), Yale–New Haven Hospital, New Haven, Conn; and the Section of Cardiology, Department of Medicine, West Haven Veteran’s Administration Medical Center (J.M.F.), West Haven, Conn.
Abstract
Background— Concerns have been raised about the appropriateness of spironolactone use in some patients with heart failure. We studied the adoption of spironolactone therapy after publication of the Randomized Aldactone Evaluation Study (RALES) in national cohorts of older patients hospitalized for heart failure.
Methods and Results— This is a study of serial cross-sectional samples of Medicare beneficiaries 65 years old discharged after hospitalization for the primary diagnosis of heart failure and with left ventricular systolic dysfunction. The first sample was discharged before (April 1998 to March 1999, n=9758) and the second sample after (July 2000 to June 2001, n=9468) publication of RALES in September 1999. We assessed spironolactone prescriptions at hospital discharge in patient groups defined by enrollment criteria for the trial. Using multivariable logistic regression, we identified factors independently associated with prescriptions not meeting these criteria. Spironolactone use increased >7-fold (3.0% to 21.3% P<0.0001) after RALES. Of patients meeting enrollment criteria, 24.1% received spironolactone, as compared with 17.4% of those not meeting the criteria. Of all prescriptions after RALES, 30.9% were provided to patients not meeting enrollment criteria. Spironolactone was prescribed to 22.8% of patients with a serum potassium value 5.0 mmol/L, to 14.1% with a serum creatinine value 2.5 mg/dL, and to 17.3% with severe renal dysfunction (estimated glomerular filtration rate <30 mL · min–1 · 1.73 m–2). In multivariable analyses, factors associated with prescriptions not meeting enrollment criteria included advanced age, noncardiovascular comorbidities, discharge to skilled nursing facilities, and care provided by physicians without board certification.
Conclusions— Spironolactone prescriptions increased markedly after the publication of RALES, and many treated patients were at risk for hyperkalemia. Simultaneously, many patients who might have benefited were not treated. These findings demonstrate the importance of balancing efforts to enhance use among appropriate patients and minimizing use in patients at risk for adverse events.
Key Words: heart failure ; aging ; aldosterone antagonists ; potassium
Introduction
Although the Randomized Aldactone Evaluation Study (RALES) demonstrated the benefits of aldosterone blockade with spironolactone in selected patients with severely symptomatic left ventricular systolic dysfunction (LVSD), this therapy has not been proven beneficial for all heart failure patients.1 Spironolactone may cause serious hyperkalemia, particularly in patients with marginal renal function, type IV renal tubular acidosis, or relatively high baseline potassium levels. Because of the potential risks of spironolactone therapy, concerns have been raised about the application of the trial results in case reports and series from single centers.2–9 More recently, a study from Canada reported simultaneous increases in spironolactone prescriptions and hospitalizations for hyperkalemia after the publication of RALES.10 Although that study could not determine whether that observation resulted from spironolactone use in patients at high risk for hyperkalemia in clinical practice, it suggests the possibility of inappropriate use or inadequate monitoring of spironolactone in the community. Because aldosterone blockade may improve survival in some patients with heart failure but pose a serious risk in others, the simultaneous assessment of appropriate and potentially inappropriate use of spironolactone is important.
Current clinical guidelines acknowledge this difficult balance, recommending that spironolactone be restricted to patients similar to those enrolled in RALES.11 Specifically, the guidelines recommend spironolactone for patients with LVSD; recent or current symptoms at rest despite the use of digoxin, diuretics, an angiotensin-converting enzyme (ACE) inhibitors, and (usually) a ;-blocker; a serum potassium value <5.0 mmol/L; and creatinine value <2.5 mg/dL before initiation of treatment.11 Because many community-based patients do not fit these criteria,12 widespread use of spironolactone without consideration of these factors could expose many patients to adverse outcomes.
The degree to which spironolactone is used for patients who might derive a benefit and is avoided among those at high risk for adverse events is not known. To address this issue, we assessed national patterns of spironolactone use before and after publication of the RALES results in September 1999 in a community-based, nationally representative sample of patients. This study was intended to assess the rate of adoption of spironolactone in the United States and the degree to which the selection of patients for treatment after publication of RALES conformed to study enrollment criteria.
Methods
National Heart Care Project
The National Heart Care (NHC) Project is an ongoing initiative funded by the Centers for Medicare and Medicaid Services designed to improve the quality of care for Medicare beneficiaries with heart failure.13 Fee-for-service Medicare beneficiaries hospitalized with a principal discharge diagnosis of heart failure (International Classification of Diseases, Ninth Revision, Clinical Modification codes 402.01, 402.11, 402.91, 404.01, 404.91, or 428) between April 1998 and March 1999 or July 2000 and June 2001, inclusive, were identified. These periods straddled the publication of RALES. According to a sampling strategy described previously,13,14 discharges were grouped by state and sorted by age, sex, race, and treating hospital, and up to 800 discharges were randomly selected from each state. The selected records underwent detailed review by trained data abstractors in central data abstraction centers. Patients with invalid social security numbers, those receiving long-term hemodialysis, those transferred to another hospital, or those who left against medical advice based on data from the administrative records or the chart abstraction were excluded. The NHC sample thus consisted of 78 882 records, of which 39 477 were from 1998 to 1999 and 39 405 from 2000 to 2001.
Study Sample
Because this was a study of the prescription of spironolactone at discharge, patients who died during hospitalization were excluded (n=4126). Because RALES enrolled only patients with LVSD, we further restricted the analysis to patients with an LV ejection fraction <40% or a qualitative description of moderate or severe systolic dysfunction, based on the evaluation closest to the discharge date (n=22 928). From this group, we excluded patients younger than 65 years because younger Medicare beneficiaries are not representative of the younger population as a whole. If a patient appeared more than once in the sample, 1 record was included by random selection. Patients who were transferred to another acute care hospital, left against medical advice, were receiving long-term renal dialysis, or with missing vital status were also excluded. A total of 3702 records of patients with LVSD were excluded by these selection criteria, resulting in a final study cohort of 19 226 patients (9758 in 1998 to 1999 and 9468 in 2000 to 2001).
Clinical Variables
The NHC database includes a wide range of variables, including demographic characteristics, medical history, characteristics on presentation, diagnostic testing, laboratory data, and a comprehensive list of the names of medications prescribed at hospital discharge. For the purposes of interpretability, drugs were assigned to classes. Combination drugs were assigned to all classes contained in the combination. The following drugs were assessed to determine spironolactone prescriptions (in alphabetical order): Aldactone (Pfizer), Aldactazide (Pfizer), hydrochlorothiazide/spironolactone, and spironolactone.
Spironolactone treatment was also assessed in subgroups defined by the primary selection criteria used for RALES.1 Patients meeting all of the following criteria were designated as conforming to these criteria: (1) discharge prescription for an ACE inhibitor or angiotensin receptor blocker or a documented reason for nonuse of ACE inhibitors; (2) discharge prescription for a diuretic (not including spironolactone); (3) a serum creatinine value closest to hospital discharge <2.5 mg/dL; and (4) a serum potassium value closest to hospital discharge of <5.0 mmol/L. All other patients were designated as not meeting RALES enrollment criteria. Because the serum creatinine value may overestimate renal function, particularly in elderly patients, an estimated glomerular filtration rate (GFR) was also calculated for all patients according to the abbreviated Modification of Diet in Renal Disease Study equation.15
Additional Data Sources
NHC data were linked with the American Medical Association Physician Masterfile16,17 by using the unique physician identification number of the attending physician, defined as the clinician primarily responsible for the patient’s care during hospitalization.18 Hospital characteristics were ascertained by linking with the 1998 and 2000 American Hospital Association Annual Surveys.19,20
Statistical Analysis
Patient, physician, and hospital characteristics, as well as documentation of spironolactone prescription at discharge, were compared between the 1998 to 1999 and 2000 to 2001 periods by 2 statistics. Correlates of spironolactone prescription were determined by characteristics of the patients (demographics, medical history, comorbidities, and admission variables), treating physician (board certification, cardiology specialty), and hospital (teaching status, cardiac care facilities, urban or rural location, ownership, and US census division). Differences in proportions were evaluated by 2 statistics.
To assess changes in use between 1998 to 1999 and 2000 to 2001 after adjustment for differences in patient, physician, and hospital characteristics, logistic regression was performed, with spironolactone prescription at discharge as the dependent variable. Models comprising statistically significant covariates (P<0.05) were constructed by backward selection. Subsequently, a variable indicating the sampling time frame was introduced into the model. The statistical significance of 2-way interactions between the time frame variable and other covariates included in the final model was tested to identify heterogeneity in the rate of increase in spironolactone use.
Finally, to identify characteristics associated with spironolactone prescription that did not conform to RALES enrollment criteria, a model in only those patients who were treated with spironolactone in 2000 to 2001 was constructed, with failure to conform to all RALES enrollment criteria as the dependent variable. Because the characteristics composing the criteria defined the dependent variable in this model, these variables were not candidate independent variables. All analyses used probability weights to account for the NHC sampling strategy, thus providing national level estimates. Statistical analyses were conducted with Stata 7.0 (Stata Corp) and SAS version 8.0 (SAS Institute, Inc).
Results
Patient Population
In both samples, 20% were at least 85 years old, the majority were male, and most had a history of hypertension and coronary artery disease (Table 1). Noncardiovascular comorbidities, including diabetes and chronic pulmonary disease, were also common. An elevated potassium level at hospital discharge (>5.0 mmol/L) was present in >5% of patients. Although the prevalence of serum creatinine >2.5 mg/dL was 8.3% in 1998 to 1999 and 9.5% in 2000 to 2001, the prevalence of a severely reduced estimated GFR (ie, <30 mL · min–1 · 1.73 m–2) was higher (15.2% and 16.4%, respectively). Approximately 42% of patients in 2000 to 2001 did not meet the appropriateness criteria for spironolactone treatment.
1998 to 1999 and 2000 to 2001, there was a decline in the proportion of patients treated at hospital discharge with ACE inhibitors or angiotensin receptor blockers, diuretics (not including spironolactone), and potassium replacements. The proportion treated with ;-blockers increased significantly between periods.
Changes in Spironolactone Prescription
Overall, spironolactone prescription rates in the cohort increased from 3.0% in 1998 to 1999 to 21.3% in 2000 to 2001 (P<0.001, Table 2). Although the increase in prescriptions between the 2 time periods was significant in all patient subgroups, the change was significantly greater in younger patients (P for interaction, 0.001); those with a lower admission blood pressure (P<0.001); those with angina (P<0.001); and those not admitted who were already being treated with spironolactone (P<0.001). Rates of increase did not differ significantly, however, by levels of discharge serum creatinine, estimated GFR, or discharge potassium supplementation.
In the population of patients discharged in 2000 to 2001, spironolactone was prescribed for 22.8% of patients with a serum potassium value >5.0 mmol/L, for 14.1% of patients with a serum creatinine value >2.5 mg/dL, and for 17.3% of patients with severe renal dysfunction, as defined by an estimated GFR (<30 mL · min–1 · 1.73 m–2; the Figure). Among patients not receiving an ACE inhibitor prescription, 17.2% received spironolactone, and among those receiving a prescription for potassium supplements, 17.1% were discharged with a spironolactone prescription. Among patients meeting study enrollment criteria after RALES, spironolactone was prescribed to 24.1% (a 21.2% absolute increase); among those not meeting these criteria, 17.4% received a discharge prescription for spironolactone (14.2% absolute increase). Of all spironolactone prescriptions in 2000 to 2001, 30.9% were provided to patients not meeting enrollment criteria.
Changes in proportions of patients receiving prescription for spironolactone at hospital discharge between 1998 to 1999 (before RALES) and 2000 to 2001 (after RALES) in all patients and patients stratified by serum potassium, creatinine, and estimated GFR.
Correlates of Spironolactone Prescription Not Meeting RALES Criteria
Among those patients receiving a prescription for spironolactone at discharge, prescriptions not conforming to RALES criteria were more common among older patients (compared with an age 65- to 74-year referent; age 75 to 84 years odds ratio [OR], 1.40; 95% confidence interval [CI], 1.14 to 1.72; P<0.001; age 85+ OR, 1.28; 95% CI, 0.96 to 1.70; P=0.09, Table 3) and were less common among women (OR, 0.80; 95% CI, 0.66 to 0.97; P=0.02). Prior history of heart failure, atrial fibrillation, and higher admission systolic blood pressures were associated with a lower OR of prescription not conforming to criteria, whereas noncardiovascular comorbidities, including chronic lung disease and anemia, were associated with higher ORs. Patients discharged to a skilled nursing facility had >2-fold higher odds of receiving spironolactone prescriptions not meeting criteria (OR, 2.31; 95% CI, 1.71 to 3.11; P<0.001). Patients cared for by nongeneralist physicians without cardiology consultation were more likely to conform to the RALES criteria (OR, 0.57; 95% CI, 0.39 to 0.84; P=0.005) compared with those cared for by a cardiologist (1.00 referent), as were those cared for by a board-certified physician (OR, 0.70; 95% CI, 0.54 to 0.90; P=0.005) compared with those cared for by a physician without board certification.
Discussion
In this study of 2 cross-sectional cohorts of older patients with heart failure and LVSD, prescription of spironolactone increased >7-fold after publication of RALES. Of the discharge prescriptions written after RALES, almost one third were provided to patients not fitting the study enrollment criteria, many of whom were at high risk for hyperkalemia. Slow adoption among patients who might benefit from spironolactone simultaneously with rapidly increasing use among those at higher risk for adverse consequences indicates that the integration of clinical trials results in practice may not maximize either effectiveness or safety. These findings demonstrate the complexity of adoption of information from new clinical trials, emphasizing the need for faster adoption among appropriate patients and simultaneous efforts to minimize use in patients who might suffer severe adverse events from inappropriate use.
RALES established the efficacy of spironolactone in reducing mortality and hospitalization in selected patients with heart failure and LVSD.1 In our study, a minority of patients who had serum potassium and creatinine levels within the ranges of RALES subjects and who were receiving other evidence-based therapy for heart failure were not treated with spironolactone at hospital discharge. This finding is concordant with other data demonstrating that standards of clinical care lag behind the evidence generated in clinical trials. The development of mechanisms to facilitate rapid and appropriate diffusion of trial results into patient care is likely to improve important patient outcomes.
RALES, however, excluded patients who were not receiving evidence-based heart failure therapy and those at potentially high risk for hyperkalemia. Several factors predispose some patients with heart failure to the risk of potentially life-threatening hyperkalemia with an agent that impairs aldosterone action. Both diabetes, which is associated with type IV renal tubular acidosis,21 and renal insufficiency are common in populations with heart failure.13,22–24 The addition of an aldosterone antagonist to the regimen of patients with underlying abnormalities of potassium excretion already being treated with ACE inhibitors, angiotensin receptor blockers, ;-blockers, or potassium supplements may increase the risk of hyperkalemia to varying degrees. Although RALES demonstrated significant mortality benefits in a patient population predominantly treated with ACE inhibitors, the trial did not enroll patients with severe renal insufficiency (defined as a serum creatinine value <2.5 mg/dL) or baseline hyperkalemia (potassium >5.0 mmol/L).1 Patients in the trial were also assessed frequently after the initiation of therapy for the development of hyperkalemia and worsening renal insufficiency, which may not be the case in usual clinical practice.4
Since the publication of RALES, concerns have been raised about the adoption of spironolactone outside the context of a carefully controlled clinical trial.2 Within a year of the release of RALES, a single-center case series of 25 patients treated with spironolactone were admitted with serious hyperkalemia. These patients were predominantly older, and none had a serum potassium value exceeding 4.8 mmol/L before admission.25 Several subsequent case series have described patients with hyperkalemia requiring hospitalization, some of whom died.5–9 Many patients in these series were elderly, in whom the serum creatinine value often overestimates true renal function. The results of our study raise significant concerns about the safety of current patterns of spironolactone prescription in community-based populations, particularly among older patients with noncardiovascular comorbidities.
A population-based study from Ontario, Canada, found that the use of spironolactone increased from 3.4% before the publication of RALES to 14.9% afterward.10 Simultaneously, hospitalization rates for hyperkalemia increased >4-fold, and associated mortality increased by >6-fold. The authors of that study speculated that hyperkalemia resulted in part from the inappropriate use of spironolactone but were unable to demonstrate whether the patterns they observed were due to misapplication of the RALES results. This is the first study in a nationally representative sample demonstrating widespread use of spironolactone in patients for whom there is no good evidence for benefit, many of whom are at high risk for hyperkalemia.
The finding of limited adoption of spironolactone use in patients like those enrolled in RALES with simultaneous widespread use in populations of patients at risk for hyperkalemia has important implications for quality improvement and patient safety initiatives in heart failure. Typically, efforts to improve patterns of pharmacological therapy in patients with heart failure have focused exclusively on either underuse (eg, ACE inhibitors or ;-blockers) or overuse (eg, type I antiarrhythmic agents or nonsteroidal antiinflammatory drugs).26–28 Given the patterns of care shown in this study, any efforts to increase the use of spironolactone in patients with heart failure should also include mechanisms to ensure that this drug is not used in populations in whom the risks may outweigh the benefits.
The demonstration that spironolactone is frequently prescribed to patients at high risk in a national sample of older patients with heart failure in conjunction with evidence of higher population rates of hyperkalemia associated with increasing spironolactone use10 suggest the need for action to change current patterns of use of aldosterone-blocking drugs. Although several mechanisms may be useful in reducing the potential harm to patients, it is possible that changing the labeling of aldosterone-blocking agents to reflect these concerns (eg, more stringent precautions) and efforts by industry and professional associations to inform practitioners of these hazards would have a positive influence. With the more recent publication of EPHESUS, a study of the selective aldosterone blocker eplerenone in patients with heart failure after myocardial infarction,29 the use of aldosterone blockade is likely to proliferate further, increasing the importance of ensuring that these agents are used only in those populations for whom the benefit outweighs the risk.
Certain issues should be considered in the interpretation of these results. First, it was not possible to determine the relation between the timing of initiation of spironolactone and the measurement of laboratory values. Thus, in some cases, serum potassium and creatinine levels may have reflected in part the results of treatment with spironolactone. Patients with elevated potassium levels and marked renal dysfunction, however, are at relatively higher risk for hyperkalemia if spironolactone is continued. Second, we used a relatively conservative definition for RALES enrollment criteria. The data did not include symptom severity by New York Heart Association (NYHA) classification or the frequency of subsequent clinical assessments. Because RALES studied only those patients with advanced symptoms (NYHA class III or IV) and provided very close follow-up for hyperkalemia, it is possible that our study overestimates the proportion of safe prescriptions. Third, the doses of spironolactone prescribed at discharge were not available in our dataset. Because evidence from case series indicates that the risk of hyperkalemia increases with higher spironolactone dose,6,25 such data might identify a larger number of patients at risk for adverse outcomes. Third, we were unable to assess changes in spironolactone use after hospital discharge. Because of the rapid adoption of spironolactone during the time period under study, prescriptions at the time of hospital discharge may have underestimated use in the outpatient setting. This also prohibits an accurate assessment of the relation between spironolactone use and outcomes such as mortality or readmission because of the potential for significant misclassification of exposure. Although we could not assess the impact of the use of spironolactone on outcomes, there is evidence that use of spironolactone in elderly patients without consideration of serum potassium and creatinine values can result in adverse outcomes, including death.7–9
In conclusion, the prescription of spironolactone in older patients with LSVD hospitalized with heart failure increased substantially in the United States after publication of RALES. More than one third of prescriptions were provided to patients who did not meet enrollment criteria of the trial, many of whom had characteristics that placed them at high risk for hyperkalemia. The means of ensuring use only among those patients with an acceptably low risk of adverse events will be important for optimizing the benefits and risks of aldosterone antagonists in the clinical care of populations of patients with heart failure.
Acknowledgments
Dr Masoudi is supported by NIH/NIA Research Career Award K08-AG01011. Dr Gross is supported by a Cancer Prevention, Control and Population Sciences Career Development Award (1K07CA-90402), the Claude D. Pepper Older Americans Independence Center at Yale (P30AG21342), and Paul Beeson career development award in Aging (K08 AG24842). Dr Foody is supported by NIH/NIA Research Career Award K08-AG20623 and NIA/Hartford Foundation Fellowship in Geriatrics. Saif Rathore is supported by NIH/National Institute of General Medical Sciences Medical Scientist Training Grant GM07205.
Disclosure
Dr Masoudi has received honoraria from Pfizer; Dr Foody has received honoraria from Pfizer. The content of this publication does not necessarily reflect the views or policies of the Department of Health and Human Services, nor does mention of trade names, commercial products, or organizations imply endorsement by the US government. The analyses on which this publication is based were performed under contract No. 500-02-CO01, entitled "Utilization and Quality Control Peer Review Organization for the State (commonwealth) of Colorado," sponsored by the Centers for Medicare and Medicaid Services (formerly Health Care Financing Administration), Department of Health and Human Services. The authors assume full responsibility for the accuracy and completeness of the ideas presented herein. This article is a direct result of the Health Care Quality Improvement Program initiated by the Centers for Medicare and Medicaid Services, which has encouraged identification of quality improvement projects derived from analysis of patterns of care, and therefore required no special funding on the part of this contractor. Ideas and contributions to the author concerning experience in engaging with issues presented are welcomed.
Footnotes
Guest Editor for this article was Clyde W. Yancy, MD.
References
Pitt B, Zannad F, Remme WJ, Cody R, Castaigne A, Perez A, Palensky J, Wittes J. The effect of spironolactone on morbidity and mortality in patients with severe heart failure: Randomized Aldactone Evaluation Study Investigators. N Engl J Med. 1999; 341: 709–717.
Geraci JM, Knowlton AA. Spironolactone for heart failure: spiraling out of control. Chest. 2000; 118: 1522–1523.
Berry C, McMurray J. Serious adverse events experienced by patients with chronic heart failure taking spironolactone. Heart. 2001; 85: e8–e9.
Bozkurt B, Agoston I, Knowlton AA. Complications of inappropriate use of spironolactone in heart failure: when an old medicine spirals out of new guidelines. J Am Coll Cardiol. 2003; 41: 211–214.
Svensson M, Gustafsson F, Galatius S, Hildebrandt PR, Atar D. Hyperkalaemia and impaired renal function in patients taking spironolactone for congestive heart failure: retrospective study. BMJ. 2003; 327: 1141–1142.
Wrenger E, Muller R, Moesenthin M, Welte T, Frolich JC, Neumann KH. Interaction of spironolactone with ACE inhibitors or angiotensin receptor blockers: analysis of 44 cases. BMJ. 2003; 327: 147–149.
Blaustein DA, Babu K, Reddy A, Schwenk MH, Avram MM. Estimation of glomerular filtration rate to prevent life-threatening hyperkalemia due to combined therapy with spironolactone and angiotensin-converting enzyme inhibition or angiotensin receptor blockade. Am J Cardiol. 2002; 90: 662–663.
Obialo CI, Ofili EO, Mirza T. Hyperkalemia in congestive heart failure patients aged 63 to 85 years with subclinical renal disease. Am J Cardiol. 2002; 90: 663–665.
Vanpee D, Swine CH. Elderly heart failure patients with drug-induced serious hyperkalemia. Aging Clin Exp Res. 2000; 12: 315–319.
Juurlink DN, Mamdani MM, Lee DS, Kopp A, Austin PC, Laupacis A, Redelmeier DA. Rates of hyperkalemia after publication of the Randomized Aldactone Evaluation Study. N Engl J Med. 2004; 351: 543–551.
Hunt SA, Baker DW, Chin MH, Cinquegrani MP, Feldman AM, Francis GS, Ganiats TG, Goldstein S, Gregoratos G, Jessup ML, Noble RJ, Packer M, Silver MA, Stevenson LW, Gibbons RJ, Antman EM, Alpert JS, Faxon DP, Fuster V, Jacobs AK, Hiratzka LF, Russell RO, Smith SC Jr. ACC/AHA guidelines for the evaluation and management of chronic heart failure in the adult: executive summary a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Committee to Revise the 1995 Guidelines for the Evaluation and Management of Heart Failure). Circulation. 2001; 104: 2996–3007.
Masoudi FA, Havranek EP, Wolfe P, Gross CP, Rathore SS, Steiner JF, Ordin DL, Krumholz HM. Most hospitalized older persons do not meet the enrollment criteria for clinical trials in heart failure. Am Heart J. 2003; 146: 250–257.
Havranek EP, Masoudi FA, Westfall KA, Wolfe P, Ordin DL, Krumholz HM. Spectrum of heart failure in older patients: results from the National Heart Failure project. Am Heart J. 2002; 143: 412–417.
Rathore SS, Foody JM, Wang Y, Smith GL, Herrin J, Masoudi FA, Wolfe P, Havranek EP, Ordin DL, Krumholz HM. Race, quality of care, and outcomes of elderly patients hospitalized with heart failure. JAMA. 2003; 289: 2517–2524.
National Kidney Foundation. K/DOQI clinical practice guidelines for chronic kidney disease: evaluation, classification, and stratification. Am J Kidney Dis. 2002; 39 (suppl 1): S1–S266.
Kenward K. The scope of the data available in the AMA’s Physician Masterfile. Am J Public Health. 1996; 86: 1481–1482.
Baldwin LM, Adamache W, Klabunde CN, Kenward K, Dahlman C, Warren L. Linking physician characteristics and Medicare claims data: issues in data availability, quality, and measurement. Med Care. 2002; 40 (suppl): IV-82–IV-95.
Iezzoni LI. Risk Adjustment for Measuring Health Care Outcomes, 2nd ed. Chicago: Health Administration Press; 1997.
American Hospital Association. The AHA Annual Survey Database: Fiscal Year 1998 Documentation. Chicago: Health Forum, AHA; 1999.
American Hospital Association. The AHA Annual Survey Database: Fiscal Year 2000 Documentation. Chicago: Health Forum, AHA; 2001.
Tan SY, Burton M. Hyporeninemic hypoaldosteronism: an overlooked cause of hyperkalemia. Arch Intern Med. 1981; 141: 30–33.
Nichols GA, Hillier TA, Erbey JR, Brown JB. Congestive heart failure in type 2 diabetes: prevalence, incidence, and risk factors. Diabetes Care. 2001; 24: 1614–1619.
Dries DL, Exner DV, Domanski MJ, Greenberg B, Stevenson LW. The prognostic implications of renal insufficiency in asymptomatic and symptomatic patients with left ventricular systolic dysfunction. J Am Coll Cardiol. 2000; 35: 681–689.
Hillege HL, Girbes AR, de Kam PJ, Boomsma F, de Zeeuw D, Charlesworth A, Hampton JR, van Veldhuisen DJ. Renal function, neurohormonal activation, and survival in patients with chronic heart failure. Circulation. 2000; 102: 203–210.
Schepkens H, Vanholder R, Billiouw JM, Lameire N. Life-threatening hyperkalemia during combined therapy with angiotensin-converting enzyme inhibitors and spironolactone: an analysis of 25 cases. Am J Med. 2001; 110: 438–441.
Jencks SF, Huff ED, Cuerdon T. Change in the quality of care delivered to Medicare beneficiaries, 1998–1999 to 2000–2001. JAMA. 2003; 289: 305–312.
Wenger NS, Shekelle PG. Assessing care of vulnerable elders: ACOVE project overview. Ann Intern Med. 2001; 135 (pt 2): 642–646.
Shekelle PG, MacLean CH, Morton SC, Wenger NS. ACOVE quality indicators. Ann Intern Med. 2001; 135 (pt 2): 653–667.
Pitt B, Remme W, Zannad F, Neaton J, Martinez F, Roniker B, Bittman R, Hurley S, Kleiman J, Gatlin M. Eplerenone, a selective aldosterone blocker, in patients with left ventricular dysfunction after myocardial infarction. N Engl J Med. 2003; 348: 1309–1321.(Frederick A. Masoudi, MD,)