Treatment of Heart Failure with Spironolactone — Trial and Tribulations
http://www.100md.com
《新英格兰医药杂志》
The application of evidence based on the results of clinical trials is a key aim of modern medical practice. Viewed from that perspective, the findings of Juurlink and colleagues in this issue of the Journal (pages 543–551) could be considered encouraging. The Randomized Aldactone Evaluation Study (RALES) showed that, when added to standard treatment (including an angiotensin-converting–enzyme [ACE] inhibitor), a low dose of the aldosterone antagonist spironolactone reduced the risk of death by 30 percent over an average follow-up period of two years among carefully selected patients with current or recent heart failure of New York Heart Association (NYHA) functional class IV.1 Among these closely monitored patients, there was a low incidence of serious adverse events, including renal dysfunction and hyperkalemia, in the spironolactone group. After the online release of these results by the Journal on July 19, 1999, there was, in Ontario, Canada, a rapid increase in the number of prescriptions of spironolactone written for older patients with heart failure who were already being treated with an ACE inhibitor. Unfortunately, Juurlink et al. also show that there was an equally brisk and striking increase in the number of hospital admissions (and subsequent deaths) related to hyperkalemia. How can we explain these findings, assuming that they reflect cause and effect, and why has the experience in clinical practice been so different from that in the clinical trial?
A number of causes are likely. One generic explanation is the clear difference between the patients in the RALES trial and those in the "real world." In part, these differences reflect the restrictive criteria for inclusion and exclusion that are common to all clinical trials. Inevitably, these conditions mean that physicians must extrapolate beyond trials to their own practice. When this extrapolation is appropriate and when it becomes inappropriate is a matter of educated clinical judgment. Another issue, specific to the risk of hyperkalemia associated with spironolactone, may be the recent and rapidly increasing use of beta-blockers in patients with heart failure. It appears that these factors have conspired to create a greater risk of hyperkalemia in association with spironolactone therapy in usual clinical practice than in the RALES trial. Concern had already been expressed about this possibility in multiple reports of small case series. On the basis of our understanding of potassium homeostasis, some explanations may be suggested.
The kidney retains a remarkable capacity to excrete potassium despite the progressive decline in the glomerular filtration rate that accompanies certain diseases and aging. This capacity is substantially attributable to distal tubular adaptation. For this adaptation to be compensatory and for adequate potassium excretion to occur, sodium delivery to the distal nephron must be maintained (along with sufficient urine volume), and aldosterone must be present to facilitate sodium–potassium exchange by the cells of the cortical collecting tubule (which must also remain responsive to aldosterone). A further reduction in the glomerular filtration rate, attenuation of the aldosterone-mediated distal tubular adaptation, or both will precipitate hyperkalemia in a vulnerable patient. Aldosterone production is decreased in the elderly and in patients receiving drugs that block the production or action of renin (beta-blockers) and angiotensin II (ACE inhibitors and angiotensin-receptor blockers). Prostanoids facilitate renin production, and nonsteroidal antiinflammatory agents and related drugs suppress renin release. Diabetes is also associated with hyporeninemic hypoaldosteronism. Heparin preparations impair aldosterone secretion. Other pharmacologic and nonpharmacologic perturbations of potassium homeostasis (e.g., the effects of potassium-containing drugs, potassium-sparing diuretics, and trimethoprim) and the glomerular filtration rate (e.g., volume depletion and hypotension) may also lead to hyperkalemia in predisposed patients relying on this tubular adaptation.
(Figure)
Cross Section of a Kidney Tubule.
Scanning electron micrograph by David Gregory and Debbie Marshall, by permission of the Wellcome Library, London.
The Ontario cohort study and the other reports of serious hyperkalemia in association with spironolactone therapy describe many of the scenarios that are likely to cause this problem, usually reflecting the interplay of several of the above factors. Affected patients have generally been much older than those in RALES. Often, their pretreatment creatinine concentration was higher than the threshold for exclusion from RALES (>2.5 mg per deciliter [221 μmol per liter]). In any case, creatinine concentrations below the upper limit used in RALES may be associated with a profoundly reduced glomerular filtration rate in frail elderly patients, especially women (e.g., a creatinine concentration of 2 mg per deciliter [177 μmol per liter] is equal to a glomerular filtration rate of about 26 ml per minute per 1.73 m2 of body-surface area in a 75-year-old white woman), emphasizing the importance of estimating the glomerular filtration rate rather than relying on measurements of creatinine. The Ontario cohort study and other series in which hyperkalemia was reported are notable for the high proportions of women included, as well as for the high proportions of patients with diabetes.
Furthermore, beta-blocker use was much more common in these clinical series than in RALES; not only do beta-blockers inhibit renin release, but nonselective agents also block adrenergic-mediated potassium uptake by peripheral tissues. Often, nonsteroidal antiinflammatory drugs and potassium supplements were coadministered. Disturbingly, in many series, the doses of spironolactone were much higher than those used in RALES, despite the dose-dependent risk of hyperkalemia observed in the RALES pilot study.2 Larger doses of spironolactone may also induce marked diuresis, volume depletion, and prerenal failure. Similarly, spironolactone was used in patients with NYHA class II heart failure and in patients with preserved left ventricular systolic function; such patients have less marked elevations in aldosterone. The dose of ACE inhibitors taken by patients at the time of enrollment in RALES was relatively low, and the RALES pilot study showed that a high dose of an ACE inhibitor was a predictor of hyperkalemia once spironolactone therapy was added.2
In summary, it seems that the excess hyperkalemia in clinical practice, as compared with RALES, may largely be explained by the use of higher doses of spironolactone and the treatment of patients who had a lower glomerular filtration rate or whose aldosterone-mediated, compensatory distal tubular potassium excretion was already attenuated.
Why did physicians use larger doses of spironolactone than those used in the RALES trial, especially given that these patients were more susceptible to hyperkalemia? Familiarity with a long-established treatment and lack of the usual educational activities related to the safe use of a new drug may have contributed. Similarly, the index of suspicion and the intensity of surveillance for adverse effects related to a new product would have been higher than those for an established drug.
What lessons can we learn from the post-RALES experience reported by Juurlink et al. and others? First, every effort should be made to define the inclusion criteria for clinical trials as broadly, and the exclusion criteria as narrowly, as possible, so that the findings are relevant to the greatest proportion of patients in clinical practice. Second, it behooves the sponsors and leaders of trials to educate physicians about the careful use of the study treatment. For their part, physicians who prescribe the treatment must fully familiarize themselves with the way in which it was used in the trial and with the contraindications, cautions, adverse effects, and drug interactions. Guidelines may need to do more than recommend which treatments should be used; practical guidance on how to use them is also needed. Finally, we need to make more linked data sets, such as those in Ontario, accessible to interested research groups and formal post-marketing surveillance programs.
It is, however, important to put these findings in perspective. Any medication may be a double-edged sword, but most medications, used appropriately, have much to offer patients and society. Over a two-year period in RALES, spironolactone, used at a low dose in carefully selected subjects who were closely monitored, postponed or prevented 11 premature deaths and 8 hospital admissions for cardiac causes for every 100 patients treated. To maximize these benefits, we have to accept that the modern treatment of heart failure necessitates complex, carefully instituted and monitored polypharmacy. This treatment is probably best provided within the framework of specialized, multidisciplinary programs of chronic-disease management.
Dr. McMurray reports having received grant support from Novartis and AstraZeneca and has been a consultant to those companies and to Pfizer and Takeda.
Source Information
From the Department of Cardiology, Western Infirmary, Glasgow, Scotland, United Kingdom.
References
Pitt B, Zannad F, Remme WJ, et al. The effect of spironolactone on morbidity and mortality in patients with severe heart failure. N Engl J Med 1999;341:709-717.
Effectiveness of spironolactone added to an angiotensin-converting enzyme inhibitor and a loop diuretic for severe chronic congestive heart failure (the Randomized Aldactone Evaluation Study [RALES]). Am J Cardiol 1996;78:902-907.(John J.V. McMurray, M.D.,)
A number of causes are likely. One generic explanation is the clear difference between the patients in the RALES trial and those in the "real world." In part, these differences reflect the restrictive criteria for inclusion and exclusion that are common to all clinical trials. Inevitably, these conditions mean that physicians must extrapolate beyond trials to their own practice. When this extrapolation is appropriate and when it becomes inappropriate is a matter of educated clinical judgment. Another issue, specific to the risk of hyperkalemia associated with spironolactone, may be the recent and rapidly increasing use of beta-blockers in patients with heart failure. It appears that these factors have conspired to create a greater risk of hyperkalemia in association with spironolactone therapy in usual clinical practice than in the RALES trial. Concern had already been expressed about this possibility in multiple reports of small case series. On the basis of our understanding of potassium homeostasis, some explanations may be suggested.
The kidney retains a remarkable capacity to excrete potassium despite the progressive decline in the glomerular filtration rate that accompanies certain diseases and aging. This capacity is substantially attributable to distal tubular adaptation. For this adaptation to be compensatory and for adequate potassium excretion to occur, sodium delivery to the distal nephron must be maintained (along with sufficient urine volume), and aldosterone must be present to facilitate sodium–potassium exchange by the cells of the cortical collecting tubule (which must also remain responsive to aldosterone). A further reduction in the glomerular filtration rate, attenuation of the aldosterone-mediated distal tubular adaptation, or both will precipitate hyperkalemia in a vulnerable patient. Aldosterone production is decreased in the elderly and in patients receiving drugs that block the production or action of renin (beta-blockers) and angiotensin II (ACE inhibitors and angiotensin-receptor blockers). Prostanoids facilitate renin production, and nonsteroidal antiinflammatory agents and related drugs suppress renin release. Diabetes is also associated with hyporeninemic hypoaldosteronism. Heparin preparations impair aldosterone secretion. Other pharmacologic and nonpharmacologic perturbations of potassium homeostasis (e.g., the effects of potassium-containing drugs, potassium-sparing diuretics, and trimethoprim) and the glomerular filtration rate (e.g., volume depletion and hypotension) may also lead to hyperkalemia in predisposed patients relying on this tubular adaptation.
(Figure)
Cross Section of a Kidney Tubule.
Scanning electron micrograph by David Gregory and Debbie Marshall, by permission of the Wellcome Library, London.
The Ontario cohort study and the other reports of serious hyperkalemia in association with spironolactone therapy describe many of the scenarios that are likely to cause this problem, usually reflecting the interplay of several of the above factors. Affected patients have generally been much older than those in RALES. Often, their pretreatment creatinine concentration was higher than the threshold for exclusion from RALES (>2.5 mg per deciliter [221 μmol per liter]). In any case, creatinine concentrations below the upper limit used in RALES may be associated with a profoundly reduced glomerular filtration rate in frail elderly patients, especially women (e.g., a creatinine concentration of 2 mg per deciliter [177 μmol per liter] is equal to a glomerular filtration rate of about 26 ml per minute per 1.73 m2 of body-surface area in a 75-year-old white woman), emphasizing the importance of estimating the glomerular filtration rate rather than relying on measurements of creatinine. The Ontario cohort study and other series in which hyperkalemia was reported are notable for the high proportions of women included, as well as for the high proportions of patients with diabetes.
Furthermore, beta-blocker use was much more common in these clinical series than in RALES; not only do beta-blockers inhibit renin release, but nonselective agents also block adrenergic-mediated potassium uptake by peripheral tissues. Often, nonsteroidal antiinflammatory drugs and potassium supplements were coadministered. Disturbingly, in many series, the doses of spironolactone were much higher than those used in RALES, despite the dose-dependent risk of hyperkalemia observed in the RALES pilot study.2 Larger doses of spironolactone may also induce marked diuresis, volume depletion, and prerenal failure. Similarly, spironolactone was used in patients with NYHA class II heart failure and in patients with preserved left ventricular systolic function; such patients have less marked elevations in aldosterone. The dose of ACE inhibitors taken by patients at the time of enrollment in RALES was relatively low, and the RALES pilot study showed that a high dose of an ACE inhibitor was a predictor of hyperkalemia once spironolactone therapy was added.2
In summary, it seems that the excess hyperkalemia in clinical practice, as compared with RALES, may largely be explained by the use of higher doses of spironolactone and the treatment of patients who had a lower glomerular filtration rate or whose aldosterone-mediated, compensatory distal tubular potassium excretion was already attenuated.
Why did physicians use larger doses of spironolactone than those used in the RALES trial, especially given that these patients were more susceptible to hyperkalemia? Familiarity with a long-established treatment and lack of the usual educational activities related to the safe use of a new drug may have contributed. Similarly, the index of suspicion and the intensity of surveillance for adverse effects related to a new product would have been higher than those for an established drug.
What lessons can we learn from the post-RALES experience reported by Juurlink et al. and others? First, every effort should be made to define the inclusion criteria for clinical trials as broadly, and the exclusion criteria as narrowly, as possible, so that the findings are relevant to the greatest proportion of patients in clinical practice. Second, it behooves the sponsors and leaders of trials to educate physicians about the careful use of the study treatment. For their part, physicians who prescribe the treatment must fully familiarize themselves with the way in which it was used in the trial and with the contraindications, cautions, adverse effects, and drug interactions. Guidelines may need to do more than recommend which treatments should be used; practical guidance on how to use them is also needed. Finally, we need to make more linked data sets, such as those in Ontario, accessible to interested research groups and formal post-marketing surveillance programs.
It is, however, important to put these findings in perspective. Any medication may be a double-edged sword, but most medications, used appropriately, have much to offer patients and society. Over a two-year period in RALES, spironolactone, used at a low dose in carefully selected subjects who were closely monitored, postponed or prevented 11 premature deaths and 8 hospital admissions for cardiac causes for every 100 patients treated. To maximize these benefits, we have to accept that the modern treatment of heart failure necessitates complex, carefully instituted and monitored polypharmacy. This treatment is probably best provided within the framework of specialized, multidisciplinary programs of chronic-disease management.
Dr. McMurray reports having received grant support from Novartis and AstraZeneca and has been a consultant to those companies and to Pfizer and Takeda.
Source Information
From the Department of Cardiology, Western Infirmary, Glasgow, Scotland, United Kingdom.
References
Pitt B, Zannad F, Remme WJ, et al. The effect of spironolactone on morbidity and mortality in patients with severe heart failure. N Engl J Med 1999;341:709-717.
Effectiveness of spironolactone added to an angiotensin-converting enzyme inhibitor and a loop diuretic for severe chronic congestive heart failure (the Randomized Aldactone Evaluation Study [RALES]). Am J Cardiol 1996;78:902-907.(John J.V. McMurray, M.D.,)